From 0889a9441d98af7951c5377647413d79c84c9efa Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 23 Sep 2006 22:11:07 +0000 Subject: CIFS: Use SEEK_END instead of hardcoded value Signed-off-by: Josef 'Jeff' Sipek Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index c3ef1c0..f5ba411 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -508,7 +508,7 @@ static ssize_t cifs_file_aio_write(struct kiocb *iocb, const char __user *buf, static loff_t cifs_llseek(struct file *file, loff_t offset, int origin) { /* origin == SEEK_END => we must revalidate the cached file length */ - if (origin == 2) { + if (origin == SEEK_END) { int retval = cifs_revalidate(file->f_dentry); if (retval < 0) return (loff_t)retval; -- cgit v0.10.2 From 1bd5bbcb6531776a8f73e2cc6287fc4dd542e1c7 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 03:35:57 +0000 Subject: [CIFS] Legacy time handling for Win9x and OS/2 part 1 Signed-off-by: Steve French diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index b35c55c..2fbc982 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -80,6 +80,9 @@ extern struct oplock_q_entry * AllocOplockQEntry(struct inode *, u16, extern void DeleteOplockQEntry(struct oplock_q_entry *); extern struct timespec cifs_NTtimeToUnix(u64 /* utc nanoseconds since 1601 */ ); extern u64 cifs_UnixTimeToNT(struct timespec); +extern __le64 cnvrtDosCifsTm(__u16 date, __u16 time); +extern struct timespec cnvrtDosUnixTm(__u16 date, __u16 time); + extern int cifs_get_inode_info(struct inode **pinode, const unsigned char *search_path, FILE_ALL_INFO * pfile_info, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 075d8fb..2851d6e 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2856,7 +2856,6 @@ qsec_out: return rc; } - /* Legacy Query Path Information call for lookup to old servers such as Win9x/WinME */ int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, @@ -2898,7 +2897,16 @@ QInfRetry: if (rc) { cFYI(1, ("Send error in QueryInfo = %d", rc)); } else if (pFinfo) { /* decode response */ + struct timespec ts; + __u32 time = le32_to_cpu(pSMBr->last_write_time); + /* BB FIXME - add time zone adjustment BB */ memset(pFinfo, 0, sizeof(FILE_ALL_INFO)); + ts.tv_nsec = 0; + ts.tv_sec = time; + /* decode time fields */ + pFinfo->ChangeTime = cifs_UnixTimeToNT(ts); + pFinfo->LastWriteTime = pFinfo->ChangeTime; + pFinfo->LastAccessTime = 0; pFinfo->AllocationSize = cpu_to_le64(le32_to_cpu(pSMBr->size)); pFinfo->EndOfFile = pFinfo->AllocationSize; diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index b88147c..06dbce3 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -432,8 +432,11 @@ int cifs_get_inode_info(struct inode **pinode, (pTcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE) & 0xFFFFFE00;*/ /* Linux can not store file creation time so ignore it */ - inode->i_atime = - cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastAccessTime)); + if(pfindData->LastAccessTime) + inode->i_atime = cifs_NTtimeToUnix + (le64_to_cpu(pfindData->LastAccessTime)); + else /* do not need to use current_fs_time - time not stored */ + inode->i_atime = CURRENT_TIME; inode->i_mtime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastWriteTime)); inode->i_ctime = diff --git a/fs/cifs/link.c b/fs/cifs/link.c index a57f5d6..0bee8b7 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -254,7 +254,11 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) tmpbuffer, len - 1, cifs_sb->local_nls); - else { + else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + cERROR(1,("SFU style symlinks not implemented yet")); + /* add open and read as in fs/cifs/inode.c */ + + } else { rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, GENERIC_READ, OPEN_REPARSE_POINT,&fid, &oplock, NULL, cifs_sb->local_nls, diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index ce87550..fa5124d 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -909,3 +909,54 @@ cifs_UnixTimeToNT(struct timespec t) /* Convert to 100ns intervals and then add the NTFS time offset. */ return (u64) t.tv_sec * 10000000 + t.tv_nsec/100 + NTFS_TIME_OFFSET; } + +static int total_days_of_prev_months[] = +{0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334}; + + +__le64 cnvrtDosCifsTm(__u16 date, __u16 time) +{ + return cpu_to_le64(cifs_UnixTimeToNT(cnvrtDosUnixTm(date, time))); +} +struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) +{ + __u8 dt[2]; + __u8 tm[2]; + struct timespec ts; + int sec,min, days, month, year; + struct timespec removeme; /* BB removeme BB */ +/* SMB_TIME * st = (SMB_TIME *)&time;*/ + + cFYI(1,("date %d time %d",date, time)); + + dt[0] = date & 0xFF; + dt[1] = (date & 0xFF00) >> 8; + tm[0] = time & 0xFF; + tm[1] = (time & 0xFF00) >> 8; + + sec = tm[0] & 0x1F; + sec = 2 * sec; + min = ((tm[0] >>5)&0xFF) + ((tm[1] & 0x7)<<3); + + sec += (min * 60); + sec += 60 * 60 * ((tm[1] >> 3) &0xFF) /* hours */; + days = (dt[0] & 0x1F) - 1; + month = ((dt[0] >> 5) & 0xFF) + ((dt[1] & 0x1) <<3); + if(month > 12) + cERROR(1,("illegal month %d in date", month)); + month -= 1; + days += total_days_of_prev_months[month]; + days += 3653; /* account for difference in days between 1980 and 1970 */ + year = (dt[1]>>1) & 0xFF; + days += year * 365; + days += (year/4); /* leap year */ + /* adjust for leap year where we are still before leap day */ + days -= ((year & 0x03) == 0) && (month < 2 ? 1 : 0); + sec += 24 * 60 * 60 * days; + + removeme = CURRENT_TIME; /* BB removeme BB */ + ts.tv_sec = sec; + + ts.tv_nsec = 0; + return ts; +} diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 9aeb58a..71e86c3 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -135,12 +135,19 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, tmp_inode->i_ctime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->ChangeTime)); } else { /* legacy, OS2 and DOS style */ +/* struct timespec ts;*/ FIND_FILE_STANDARD_INFO * pfindData = (FIND_FILE_STANDARD_INFO *)buf; +/* ts = cnvrtDosUnixTm( + le16_to_cpu(pfindData->LastWriteDate), + le16_to_cpu(pfindData->LastWriteTime));*/ attr = le16_to_cpu(pfindData->Attributes); allocation_size = le32_to_cpu(pfindData->AllocationSize); end_of_file = le32_to_cpu(pfindData->DataSize); + /* do not need to use current_fs_time helper function since + time not stored for this case so atime can not "go backwards" + by pulling newer older from disk when inode refrenshed */ tmp_inode->i_atime = CURRENT_TIME; /* tmp_inode->i_mtime = BB FIXME - add dos time handling tmp_inode->i_ctime = 0; BB FIXME */ -- cgit v0.10.2 From 2cd646a2d1d5e0e46aa4bb55b1847b0cb35bd855 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 19:43:08 +0000 Subject: [CIFS] Remove static and unused symbols Most cases of the ones found by Shaggy by "make namespacecheck" could be removed or made static Ack: Dave Kleikamp Signed-off-by: Steve French diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index d0776ac..5eff35d 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -31,8 +31,8 @@ struct cifs_sid { } __attribute__((packed)); /* everyone */ -extern const struct cifs_sid sid_everyone; +/* extern const struct cifs_sid sid_everyone;*/ /* group users */ -extern const struct cifs_sid sid_user; +/* extern const struct cifs_sid sid_user;*/ #endif /* _CIFSACL_H */ diff --git a/fs/cifs/cifsencrypt.h b/fs/cifs/cifsencrypt.h index 03e359b..152fa2d 100644 --- a/fs/cifs/cifsencrypt.h +++ b/fs/cifs/cifsencrypt.h @@ -27,8 +27,6 @@ extern void mdfour(unsigned char *out, unsigned char *in, int n); /* smbdes.c */ extern void E_P16(unsigned char *p14, unsigned char *p16); extern void E_P24(unsigned char *p21, unsigned char *c8, unsigned char *p24); -extern void D_P16(unsigned char *p14, unsigned char *in, unsigned char *out); -extern void E_old_pw_hash(unsigned char *, unsigned char *, unsigned char *); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index f5ba411..cd17d4b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -442,7 +442,7 @@ static int cifs_remount(struct super_block *sb, int *flags, char *data) return 0; } -struct super_operations cifs_super_ops = { +static struct super_operations cifs_super_ops = { .read_inode = cifs_read_inode, .put_super = cifs_put_super, .statfs = cifs_statfs, @@ -930,7 +930,7 @@ init_cifs(void) #ifdef CONFIG_PROC_FS cifs_proc_init(); #endif - INIT_LIST_HEAD(&GlobalServerList); /* BB not implemented yet */ +/* INIT_LIST_HEAD(&GlobalServerList);*/ /* BB not implemented yet */ INIT_LIST_HEAD(&GlobalSMBSessionList); INIT_LIST_HEAD(&GlobalTreeConnectionList); INIT_LIST_HEAD(&GlobalOplock_Q); @@ -958,6 +958,7 @@ init_cifs(void) GlobalCurrentXid = 0; GlobalTotalActiveXid = 0; GlobalMaxActiveXid = 0; + memset(Local_System_Name, 0, 15); rwlock_init(&GlobalSMBSeslock); spin_lock_init(&GlobalMid_Lock); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index bea875d..a243f779 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -36,7 +36,7 @@ extern const struct address_space_operations cifs_addr_ops; extern const struct address_space_operations cifs_addr_ops_smallbuf; /* Functions related to super block operations */ -extern struct super_operations cifs_super_ops; +/* extern struct super_operations cifs_super_ops;*/ extern void cifs_read_inode(struct inode *); extern void cifs_delete_inode(struct inode *); /* extern void cifs_write_inode(struct inode *); *//* BB not needed yet */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index b24006c..441f8d2 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -512,7 +512,8 @@ require use of the stronger protocol */ * This list helps improve performance and eliminate the messages indicating * that we had a communications error talking to the server in this list. */ -GLOBAL_EXTERN struct servers_not_supported *NotSuppList; /*@z4a */ +/* Feature not supported */ +/* GLOBAL_EXTERN struct servers_not_supported *NotSuppList; */ /* * The following is a hash table of all the users we know about. @@ -568,7 +569,6 @@ GLOBAL_EXTERN unsigned int lookupCacheEnabled; GLOBAL_EXTERN unsigned int extended_security; /* if on, session setup sent with more secure ntlmssp2 challenge/resp */ GLOBAL_EXTERN unsigned int sign_CIFS_PDUs; /* enable smb packet signing */ -GLOBAL_EXTERN unsigned int secFlags; GLOBAL_EXTERN unsigned int linuxExtEnabled;/*enable Linux/Unix CIFS extensions*/ GLOBAL_EXTERN unsigned int CIFSMaxBufSize; /* max size not including hdr */ GLOBAL_EXTERN unsigned int cifs_min_rcv; /* min size of big ntwrk buf pool */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 2fbc982..7dd2f48 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -50,11 +50,11 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, struct kvec *, int /* nvec to send */, int * /* type of buf returned */ , const int long_op); -extern int SendReceiveBlockingLock(const unsigned int /* xid */ , struct cifsTconInfo *, +extern int SendReceiveBlockingLock(const unsigned int /* xid */ , + struct cifsTconInfo *, struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , int * /* bytes returned */); -extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); extern int is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); extern int is_size_safe_to_change(struct cifsInodeInfo *); @@ -282,8 +282,6 @@ extern void sesInfoFree(struct cifsSesInfo *); extern struct cifsTconInfo *tconInfoAlloc(void); extern void tconInfoFree(struct cifsTconInfo *); -extern int cifs_reconnect(struct TCP_Server_Info *server); - extern int cifs_sign_smb(struct smb_hdr *, struct TCP_Server_Info *,__u32 *); extern int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *, __u32 *); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 2851d6e..dcd7087 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2773,9 +2773,11 @@ GetExtAttrOut: /* security id for everyone */ -const struct cifs_sid sid_everyone = {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; +const static struct cifs_sid sid_everyone = + {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; /* group users */ -const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; +const static struct cifs_sid sid_user = + {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; /* Convert CIFS ACL to POSIX form */ static int parse_sec_desc(struct cifs_sid * psec_desc, int acl_len) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 0e9ba0b..b3268e5 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -109,7 +109,7 @@ static int ipv6_connect(struct sockaddr_in6 *psin_server, * wake up waiters on reconnection? - (not needed currently) */ -int +static int cifs_reconnect(struct TCP_Server_Info *server) { int rc = 0; @@ -771,13 +771,17 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) separator[0] = ','; separator[1] = 0; - memset(vol->source_rfc1001_name,0x20,15); - for(i=0;i < strnlen(system_utsname.nodename,15);i++) { - /* does not have to be a perfect mapping since the field is - informational, only used for servers that do not support - port 445 and it can be overridden at mount time */ - vol->source_rfc1001_name[i] = - toupper(system_utsname.nodename[i]); + if(Local_System_Name[0] != 0) + memcpy(vol->source_rfc1001_name, Local_System_Name,15); + else { + memset(vol->source_rfc1001_name,0x20,15); + for(i=0;i < strnlen(system_utsname.nodename,15);i++) { + /* does not have to be perfect mapping since field is + informational, only used for servers that do not support + port 445 and it can be overridden at mount time */ + vol->source_rfc1001_name[i] = + toupper(system_utsname.nodename[i]); + } } vol->source_rfc1001_name[15] = 0; /* null target name indicates to use *SMBSERVR default called name diff --git a/fs/cifs/md5.c b/fs/cifs/md5.c index 7aa2349..273aa03 100644 --- a/fs/cifs/md5.c +++ b/fs/cifs/md5.c @@ -255,7 +255,7 @@ MD5Transform(__u32 buf[4], __u32 const in[16]) /*********************************************************************** the rfc 2104 version of hmac_md5 initialisation. ***********************************************************************/ -void +static void hmac_md5_init_rfc2104(unsigned char *key, int key_len, struct HMACMD5Context *ctx) { @@ -350,7 +350,7 @@ hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx) single function to calculate an HMAC MD5 digest from data. use the microsoft hmacmd5 init method because the key is 16 bytes. ************************************************************/ -void +static void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, unsigned char *digest) { diff --git a/fs/cifs/md5.h b/fs/cifs/md5.h index 00e1c53..f7d4f41 100644 --- a/fs/cifs/md5.h +++ b/fs/cifs/md5.h @@ -27,12 +27,12 @@ void MD5Final(unsigned char digest[16], struct MD5Context *context); /* The following definitions come from lib/hmacmd5.c */ -void hmac_md5_init_rfc2104(unsigned char *key, int key_len, - struct HMACMD5Context *ctx); +/* void hmac_md5_init_rfc2104(unsigned char *key, int key_len, + struct HMACMD5Context *ctx);*/ void hmac_md5_init_limK_to_64(const unsigned char *key, int key_len, struct HMACMD5Context *ctx); void hmac_md5_update(const unsigned char *text, int text_len, struct HMACMD5Context *ctx); void hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx); -void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, - unsigned char *digest); +/* void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, + unsigned char *digest);*/ diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 22c937e..ca6e9b1 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -389,7 +389,7 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , return; } -int +static int checkSMBhdr(struct smb_hdr *smb, __u16 mid) { /* Make sure that this really is an SMB, that it is a response, diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c index efaa044..2b193e4 100644 --- a/fs/cifs/smbdes.c +++ b/fs/cifs/smbdes.c @@ -364,14 +364,14 @@ E_P24(unsigned char *p21, unsigned char *c8, unsigned char *p24) smbhash(p24 + 16, c8, p21 + 14, 1); } -void +static void D_P16(unsigned char *p14, unsigned char *in, unsigned char *out) { smbhash(out, in, p14, 0); smbhash(out + 8, in + 8, p14 + 7, 0); } -void +static void E_old_pw_hash(unsigned char *p14, unsigned char *in, unsigned char *out) { smbhash(out, in, p14, 1); diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index f518c5e..c7e55a9 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -145,7 +145,7 @@ E_md4hash(const unsigned char *passwd, unsigned char *p16) } /* Does both the NT and LM owfs of a user's password */ -void +static void nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]) { char passwd[514]; @@ -223,7 +223,7 @@ SMBOWFencrypt(unsigned char passwd[16], unsigned char *c8, } /* Does the des encryption from the FIRST 8 BYTES of the NT or LM MD4 hash. */ -void +static void NTLMSSPOWFencrypt(unsigned char passwd[8], unsigned char *ntlmchalresp, unsigned char p24[24]) { -- cgit v0.10.2 From e33c74d06e2b46a5f187ec7f60248da774c84e72 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 20:35:48 +0000 Subject: [CIFS] Fix build break Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index cd17d4b..51e888f 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -63,6 +63,7 @@ extern struct task_struct * oplockThread; /* remove sparse warning */ struct task_struct * oplockThread = NULL; extern struct task_struct * dnotifyThread; /* remove sparse warning */ struct task_struct * dnotifyThread = NULL; +static struct super_operations cifs_super_ops; unsigned int CIFSMaxBufSize = CIFS_MAX_MSGSIZE; module_param(CIFSMaxBufSize, int, 0); MODULE_PARM_DESC(CIFSMaxBufSize,"Network buffer size (not including header). Default: 16384 Range: 8192 to 130048"); -- cgit v0.10.2 From 2eaf55862e8eb03999169d84f21eadffc88a36ce Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 20:41:48 +0000 Subject: [CIFS] Remove unused prototypes Signed-off-by: Steve French diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index c7e55a9..48314e5 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -51,11 +51,8 @@ void SMBencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24); void E_md4hash(const unsigned char *passwd, unsigned char *p16); -void nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]); static void SMBOWFencrypt(unsigned char passwd[16], unsigned char *c8, unsigned char p24[24]); -void NTLMSSPOWFencrypt(unsigned char passwd[8], - unsigned char *ntlmchalresp, unsigned char p24[24]); void SMBNTencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24); /* -- cgit v0.10.2 From e10847ed499cb86bf8ce12f3a686be8a98f8e140 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 20:49:01 +0000 Subject: [CIFS] More removing of unused functions Signed-off-by: Steve French diff --git a/fs/cifs/md5.c b/fs/cifs/md5.c index 273aa03..e6a2097 100644 --- a/fs/cifs/md5.c +++ b/fs/cifs/md5.c @@ -252,6 +252,7 @@ MD5Transform(__u32 buf[4], __u32 const in[16]) buf[3] += d; } +#if 0 /* currently unused */ /*********************************************************************** the rfc 2104 version of hmac_md5 initialisation. ***********************************************************************/ @@ -289,6 +290,7 @@ hmac_md5_init_rfc2104(unsigned char *key, int key_len, MD5Init(&ctx->ctx); MD5Update(&ctx->ctx, ctx->k_ipad, 64); } +#endif /*********************************************************************** the microsoft version of hmac_md5 initialisation. @@ -333,6 +335,7 @@ hmac_md5_update(const unsigned char *text, int text_len, /*********************************************************************** finish off hmac_md5 "inner" buffer and generate outer one. ***********************************************************************/ +#if 0 /* currently unused */ void hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx) { @@ -361,3 +364,4 @@ hmac_md5(unsigned char key[16], unsigned char *data, int data_len, } hmac_md5_final(digest, &ctx); } +#endif diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c index 2b193e4..7a1b2b9 100644 --- a/fs/cifs/smbdes.c +++ b/fs/cifs/smbdes.c @@ -364,6 +364,7 @@ E_P24(unsigned char *p21, unsigned char *c8, unsigned char *p24) smbhash(p24 + 16, c8, p21 + 14, 1); } +#if 0 /* currently unsued */ static void D_P16(unsigned char *p14, unsigned char *in, unsigned char *out) { @@ -377,7 +378,6 @@ E_old_pw_hash(unsigned char *p14, unsigned char *in, unsigned char *out) smbhash(out, in, p14, 1); smbhash(out + 8, in + 8, p14 + 7, 1); } -#if 0 /* these routines are currently unneeded, but may be needed later */ void diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index 48314e5..4b25ba9 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -141,6 +141,7 @@ E_md4hash(const unsigned char *passwd, unsigned char *p16) memset(wpwd,0,129 * 2); } +#if 0 /* currently unused */ /* Does both the NT and LM owfs of a user's password */ static void nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]) @@ -168,6 +169,7 @@ nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]) /* clear out local copy of user's password (just being paranoid). */ memset(passwd, '\0', sizeof (passwd)); } +#endif /* Does the NTLMv2 owfs of a user's password */ #if 0 /* function not needed yet - but will be soon */ @@ -220,6 +222,7 @@ SMBOWFencrypt(unsigned char passwd[16], unsigned char *c8, } /* Does the des encryption from the FIRST 8 BYTES of the NT or LM MD4 hash. */ +#if 0 /* currently unused */ static void NTLMSSPOWFencrypt(unsigned char passwd[8], unsigned char *ntlmchalresp, unsigned char p24[24]) @@ -232,6 +235,7 @@ NTLMSSPOWFencrypt(unsigned char passwd[8], E_P24(p21, ntlmchalresp, p24); } +#endif /* Does the NT MD4 hash then des encryption. */ -- cgit v0.10.2 From a3ab41f10e2f5087e515da358680c88dd61d4832 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 20:52:08 +0000 Subject: [CIFS] Fix build break ifdef in wrong place Signed-off-by: Steve French diff --git a/fs/cifs/md5.c b/fs/cifs/md5.c index e6a2097..ccebf9b 100644 --- a/fs/cifs/md5.c +++ b/fs/cifs/md5.c @@ -335,7 +335,6 @@ hmac_md5_update(const unsigned char *text, int text_len, /*********************************************************************** finish off hmac_md5 "inner" buffer and generate outer one. ***********************************************************************/ -#if 0 /* currently unused */ void hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx) { @@ -353,6 +352,7 @@ hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx) single function to calculate an HMAC MD5 digest from data. use the microsoft hmacmd5 init method because the key is 16 bytes. ************************************************************/ +#if 0 /* currently unused */ static void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, unsigned char *digest) -- cgit v0.10.2 From bf97d28711e2dc4dc947faa6477cd1b36b91a2da Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 28 Sep 2006 21:34:06 +0000 Subject: [CIFS] CIFS support for /proc//mountstats part 1 Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 51e888f..7ecfcbf 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -437,6 +437,14 @@ static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) return; } +#ifdef CONFIG_CIFS_STATS2 +static int cifs_show_stats(struct seq_file *s, struct vfsmount *mnt) +{ + /* BB FIXME */ + return 0; +} +#endif + static int cifs_remount(struct super_block *sb, int *flags, char *data) { *flags |= MS_NODIRATIME; @@ -456,6 +464,9 @@ static struct super_operations cifs_super_ops = { .show_options = cifs_show_options, .umount_begin = cifs_umount_begin, .remount_fs = cifs_remount, +#ifdef CONFIG_CIFS_STATS2 + cifs_show_stats, +#endif }; static int -- cgit v0.10.2 From 25ee4a98c662317a7973f3053567d4ec51857511 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Sep 2006 00:54:23 +0000 Subject: [CIFS] Handle legacy servers which return undefined time zone Signed-off-by: Guenter Kukkukk Signed-off-by: Steve French diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index dcd7087..9971859 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -447,6 +447,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) #ifdef CONFIG_CIFS_WEAK_PW_HASH } else if((pSMBr->hdr.WordCount == 13) && (pSMBr->DialectIndex == LANMAN_PROT)) { + int tmp, adjust; struct lanman_neg_rsp * rsp = (struct lanman_neg_rsp *)pSMBr; if((secFlags & CIFSSEC_MAY_LANMAN) || @@ -473,11 +474,36 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->capabilities = CAP_MPX_MODE; } server->timeZone = le16_to_cpu(rsp->ServerTimeZone); + tmp = le16_to_cpu(rsp->ServerTimeZone); + if (tmp == (int)0xffff) { + /* OS/2 often does not set timezone therefore + * we must use server time to calc time zone. + * Could deviate slightly from the right zone. Not easy + * to adjust, since timezones are not always a multiple + * of 60 (sometimes 30 minutes - are there smaller?) + */ + struct timespec ts, utc; + utc = CURRENT_TIME; + ts = cnvrtDosUnixTm(le16_to_cpu(rsp->SrvTime.Date), + le16_to_cpu(rsp->SrvTime.Time)); + cFYI(1,("SrvTime: %d sec since 1970 (utc: %d) diff: %d", + (int)ts.tv_sec, (int)utc.tv_sec, + (int)(utc.tv_sec - ts.tv_sec))); + tmp = (int)(utc.tv_sec - ts.tv_sec); + adjust = tmp < 0 ? -29 : 29; + tmp = ((tmp + adjust) / 60) * 60; + server->timeZone = tmp; + } else { + server->timeZone = tmp * 60; /* also in seconds */ + } + cFYI(1,("server->timeZone: %d seconds", server->timeZone)); + /* BB get server time for time conversions and add code to use it and timezone since this is not UTC */ - if (rsp->EncryptionKeyLength == cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { + if (rsp->EncryptionKeyLength == + cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { memcpy(server->cryptKey, rsp->EncryptionKey, CIFS_CRYPTO_KEY_SIZE); } else if (server->secMode & SECMODE_PW_ENCRYPT) { -- cgit v0.10.2 From 175ec9e11cf18f8373b32f7a33e75a4cf7ce25e3 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Sep 2006 01:07:38 +0000 Subject: [CIFS] Rename server time zone field Server time zone is not really a time zone, rather a time adjustement in seconds. CC: Guenter Kukkukk Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 7ecfcbf..e7641f9 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -465,7 +465,7 @@ static struct super_operations cifs_super_ops = { .umount_begin = cifs_umount_begin, .remount_fs = cifs_remount, #ifdef CONFIG_CIFS_STATS2 - cifs_show_stats, + .cifs_show_stats, #endif }; diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 441f8d2..98eb544 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -153,7 +153,7 @@ struct TCP_Server_Info { char sessid[4]; /* unique token id for this session */ /* (returned on Negotiate */ int capabilities; /* allow selective disabling of caps by smb sess */ - __u16 timeZone; + __u16 timeAdj; /* Adjust for difference in server time zone in sec */ __u16 CurrentMid; /* multiplex id - rotating counter */ char cryptKey[CIFS_CRYPTO_KEY_SIZE]; /* 16th byte of RFC1001 workstation name is always null */ diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 81df2bf..e5dd870 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -417,7 +417,10 @@ typedef struct lanman_neg_rsp { __le16 MaxNumberVcs; __le16 RawMode; __le32 SessionKey; - __le32 ServerTime; + struct { + __le16 Time; + __le16 Date; + } __attribute__((packed)) SrvTime; __le16 ServerTimeZone; __le16 EncryptionKeyLength; __le16 Reserved; diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 9971859..6e00458 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -473,7 +473,6 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->maxRw = 0;/* we do not need to use raw anyway */ server->capabilities = CAP_MPX_MODE; } - server->timeZone = le16_to_cpu(rsp->ServerTimeZone); tmp = le16_to_cpu(rsp->ServerTimeZone); if (tmp == (int)0xffff) { /* OS/2 often does not set timezone therefore @@ -492,11 +491,11 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) tmp = (int)(utc.tv_sec - ts.tv_sec); adjust = tmp < 0 ? -29 : 29; tmp = ((tmp + adjust) / 60) * 60; - server->timeZone = tmp; + server->timeAdj = tmp; } else { - server->timeZone = tmp * 60; /* also in seconds */ + server->timeAdj = tmp * 60; /* also in seconds */ } - cFYI(1,("server->timeZone: %d seconds", server->timeZone)); + cFYI(1,("server->timeAdj: %d seconds", server->timeAdj)); /* BB get server time for time conversions and add @@ -557,7 +556,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) cFYI(0, ("Max buf = %d", ses->server->maxBuf)); GETU32(ses->server->sessid) = le32_to_cpu(pSMBr->SessionKey); server->capabilities = le32_to_cpu(pSMBr->Capabilities); - server->timeZone = le16_to_cpu(pSMBr->ServerTimeZone); + server->timeAdj = le16_to_cpu(pSMBr->ServerTimeZone) * 60; if (pSMBr->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { memcpy(server->cryptKey, pSMBr->u.EncryptionKey, CIFS_CRYPTO_KEY_SIZE); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index b3268e5..083b2b2 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3320,15 +3320,16 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, if(linuxExtEnabled == 0) pSesInfo->capabilities &= (~CAP_UNIX); /* pSesInfo->sequence_number = 0;*/ - cFYI(1,("Security Mode: 0x%x Capabilities: 0x%x Time Zone: %d", + cFYI(1,("Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d", pSesInfo->server->secMode, pSesInfo->server->capabilities, - pSesInfo->server->timeZone)); + pSesInfo->server->timeAdj)); if(experimEnabled < 2) rc = CIFS_SessSetup(xid, pSesInfo, first_time, nls_info); else if (extended_security - && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) + && (pSesInfo->capabilities + & CAP_EXTENDED_SECURITY) && (pSesInfo->server->secType == NTLMSSP)) { rc = -EOPNOTSUPP; } else if (extended_security @@ -3342,7 +3343,7 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, if (!rc) { if(ntlmv2_flag) { char * v2_response; - cFYI(1,("Can use more secure NTLM version 2 password hash")); + cFYI(1,("more secure NTLM ver2 hash")); if(CalcNTLMv2_partial_mac_key(pSesInfo, nls_info)) { rc = -ENOMEM; -- cgit v0.10.2 From f46d3e11903e452924ef2996aa9aca2aae4427e2 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Sep 2006 01:08:55 +0000 Subject: [CIFS] Fix typo in name of new cifs_show_stats Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index e7641f9..ca53720 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -465,7 +465,7 @@ static struct super_operations cifs_super_ops = { .umount_begin = cifs_umount_begin, .remount_fs = cifs_remount, #ifdef CONFIG_CIFS_STATS2 - .cifs_show_stats, + .show_stats = cifs_show_stats, #endif }; -- cgit v0.10.2 From 9ac00b7d96045fa3ce573e0ad5cdc0350ad8e1d2 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Sep 2006 04:13:17 +0000 Subject: [CIFS] Do not send newer QFSInfo to legacy servers which can not support it Fix dialect negotiation to save off when we have negotiated lanman. This allows us to avoid sending some somewhat newer requests that the server can not handle and go directly to the older version (infolevel) of the same call. Make sure we try to negotiate a level which allows us to get the server OS (which we check so we can detect Win9x vs. other legacy servers and eventually work around the Win9x DOS time bug (they reverse date/time fields). Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index ca53720..d6d226a 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -199,10 +199,12 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) /* Only need to call the old QFSInfo if failed on newer one */ if(rc) - rc = CIFSSMBQFSInfo(xid, pTcon, buf); + if((pTcon->ses->flags & CIFS_SES_LANMAN) == 0) + rc = CIFSSMBQFSInfo(xid, pTcon, buf); /* not supported by OS2 */ - /* Old Windows servers do not support level 103, retry with level - one if old server failed the previous call */ + /* Some old Windows servers also do not support level 103, retry with + older level one if old server failed the previous call or we + bypassed it because we detected that this was an older LANMAN sess */ if(rc) rc = SMBOldQFSInfo(xid, pTcon, buf); /* diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 98eb544..597afdf 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -203,9 +203,14 @@ struct cifsSesInfo { char * domainName; char * password; }; -/* session flags */ +/* no more than one of the following three session flags may be set */ #define CIFS_SES_NT4 1 - +#define CIFS_SES_OS2 2 +#define CIFS_SES_W9X 4 +/* following flag is set for old servers such as OS2 (and Win95?) + which do not negotiate NTLM or POSIX dialects, but instead + negotiate one of the older LANMAN dialects */ +#define CIFS_SES_LANMAN 8 /* * there is one of these for each connection to a resource on a particular * session diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index e5dd870..5050542 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -26,7 +26,8 @@ #ifdef CONFIG_CIFS_WEAK_PW_HASH #define LANMAN_PROT 0 -#define CIFS_PROT 1 +#define LANMAN2_PROT 1 +#define CIFS_PROT 2 #else #define CIFS_PROT 0 #endif diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 6e00458..f2fa05b 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -46,6 +46,7 @@ static struct { } protocols[] = { #ifdef CONFIG_CIFS_WEAK_PW_HASH {LANMAN_PROT, "\2LM1.2X002"}, + {LANMAN2_PROT, "\2LANMAN2.1"}, #endif /* weak password hashing for legacy clients */ {CIFS_PROT, "\2NT LM 0.12"}, {POSIX_PROT, "\2POSIX 2"}, @@ -67,13 +68,13 @@ static struct { /* define the number of elements in the cifs dialect array */ #ifdef CONFIG_CIFS_POSIX #ifdef CONFIG_CIFS_WEAK_PW_HASH -#define CIFS_NUM_PROT 3 +#define CIFS_NUM_PROT 4 #else #define CIFS_NUM_PROT 2 #endif /* CIFS_WEAK_PW_HASH */ #else /* not posix */ #ifdef CONFIG_CIFS_WEAK_PW_HASH -#define CIFS_NUM_PROT 2 +#define CIFS_NUM_PROT 3 #else #define CIFS_NUM_PROT 1 #endif /* CONFIG_CIFS_WEAK_PW_HASH */ @@ -446,7 +447,8 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) goto neg_err_exit; #ifdef CONFIG_CIFS_WEAK_PW_HASH } else if((pSMBr->hdr.WordCount == 13) - && (pSMBr->DialectIndex == LANMAN_PROT)) { + && ((pSMBr->DialectIndex == LANMAN_PROT) + || (pSMBr->DialectIndex == LANMAN2_PROT))) { int tmp, adjust; struct lanman_neg_rsp * rsp = (struct lanman_neg_rsp *)pSMBr; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 083b2b2..c96f3ed 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3316,6 +3316,7 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, first_time = 1; } if (!rc) { + pSesInfo->flags = 0; pSesInfo->capabilities = pSesInfo->server->capabilities; if(linuxExtEnabled == 0) pSesInfo->capabilities &= (~CAP_UNIX); diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index d1705ab..e4c4e46 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -268,6 +268,10 @@ static int decode_ascii_ssetup(char ** pbcc_area, int bleft, struct cifsSesInfo ses->serverOS = kzalloc(len + 1, GFP_KERNEL); if(ses->serverOS) strncpy(ses->serverOS, bcc_ptr, len); + if(strncmp(ses->serverOS, "OS/2",4) == 0) { + cFYI(1,("OS/2 server")); + ses->flags |= CIFS_SES_OS2; + } bcc_ptr += len + 1; bleft -= len + 1; @@ -290,16 +294,11 @@ static int decode_ascii_ssetup(char ** pbcc_area, int bleft, struct cifsSesInfo if(len > bleft) return rc; - if(ses->serverDomain) - kfree(ses->serverDomain); - - ses->serverDomain = kzalloc(len + 1, GFP_KERNEL); - if(ses->serverOS) - strncpy(ses->serverOS, bcc_ptr, len); - - bcc_ptr += len + 1; - bleft -= len + 1; - + /* No domain field in LANMAN case. Domain is + returned by old servers in the SMB negprot response */ + /* BB For newer servers which do not support Unicode, + but thus do return domain here we could add parsing + for it later, but it is not very important */ cFYI(1,("ascii: bytes left %d",bleft)); return rc; @@ -366,6 +365,8 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, str_area = kmalloc(2000, GFP_KERNEL); bcc_ptr = str_area; + ses->flags &= ~CIFS_SES_LANMAN; + if(type == LANMAN) { #ifdef CONFIG_CIFS_WEAK_PW_HASH char lnm_session_key[CIFS_SESS_KEY_SIZE]; @@ -377,7 +378,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, /* and copy into bcc */ calc_lanman_hash(ses, lnm_session_key); - + ses->flags |= CIFS_SES_LANMAN; /* #ifdef CONFIG_CIFS_DEBUG2 cifs_dump_mem("cryptkey: ",ses->server->cryptKey, CIFS_SESS_KEY_SIZE); -- cgit v0.10.2 From de7ed55dbb2f2c44be669d56c4adf28cbffb26ce Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Sep 2006 13:25:52 +0000 Subject: [CIFS] Make use of newer QFSInfo dependent on capability bit instead of whether we negotiated legacy lanman dialect so we do not keep retrying for mount to WindowsME Signed-off-by: Steve French diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index d6d226a..4336436 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -199,7 +199,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) /* Only need to call the old QFSInfo if failed on newer one */ if(rc) - if((pTcon->ses->flags & CIFS_SES_LANMAN) == 0) + if(pTcon->ses->capabilities & CAP_NT_SMBS) rc = CIFSSMBQFSInfo(xid, pTcon, buf); /* not supported by OS2 */ /* Some old Windows servers also do not support level 103, retry with diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 597afdf..74d3ccb 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -153,7 +153,7 @@ struct TCP_Server_Info { char sessid[4]; /* unique token id for this session */ /* (returned on Negotiate */ int capabilities; /* allow selective disabling of caps by smb sess */ - __u16 timeAdj; /* Adjust for difference in server time zone in sec */ + int timeAdj; /* Adjust for difference in server time zone in sec */ __u16 CurrentMid; /* multiplex id - rotating counter */ char cryptKey[CIFS_CRYPTO_KEY_SIZE]; /* 16th byte of RFC1001 workstation name is always null */ -- cgit v0.10.2 From 18f75ca0dc0d5b6a2ec15d89d517b3c67e0f1c87 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 1 Oct 2006 03:13:01 +0000 Subject: [CIFS] Allow LANMAN21 support even in both POSIX non-POSIX path Signed-off-by: Guenter Kukkukk Signed-off-by: Steve French diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index f2fa05b..75f0603 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -59,6 +59,7 @@ static struct { } protocols[] = { #ifdef CONFIG_CIFS_WEAK_PW_HASH {LANMAN_PROT, "\2LM1.2X002"}, + {LANMAN2_PROT, "\2LANMAN2.1"}, #endif /* weak password hashing for legacy clients */ {CIFS_PROT, "\2NT LM 0.12"}, {BAD_PROT, "\2"} -- cgit v0.10.2 From 203cf2fc13a5db1fb202c294948fa9cb43bf69fa Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 1 Oct 2006 19:59:41 +0000 Subject: [CIFS] Fix readdir of large directories for backlevel servers (were not setting all of resume key) Signed-off-by: Steve French diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 71e86c3..b0e5db1 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -946,6 +946,7 @@ static int cifs_save_resume_key(const char *current_entry, filename = &pFindData->FileName[0]; /* one byte length, no name conversion */ len = (unsigned int)pFindData->FileNameLength; + cifsFile->srch_inf.resume_key = pFindData->ResumeKey; } else { cFYI(1,("Unknown findfirst level %d",level)); return -EINVAL; -- cgit v0.10.2 From b815f1e559e7cbdf3e561cf0c7cffc4a4a57a013 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 2 Oct 2006 05:53:29 +0000 Subject: [CIFS] Allow for 15 minute TZs (e.g. Nepal) and be more explicit about not setting time on close Signed-off-by: Guenter Kukkukk Signed-off-by: Steve French diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 5050542..6df9dad 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -409,6 +409,8 @@ typedef struct negotiate_req { /* Dialect index is 13 for LANMAN */ +#define MIN_TZ_ADJ (15 * 60) /* minimum grid for timezones in seconds */ + typedef struct lanman_neg_rsp { struct smb_hdr hdr; /* wct = 13 */ __le16 DialectIndex; @@ -678,7 +680,7 @@ typedef union smb_com_tree_disconnect { /* as an altetnative can use flag on typedef struct smb_com_close_req { struct smb_hdr hdr; /* wct = 3 */ __u16 FileID; - __u32 LastWriteTime; /* should be zero */ + __u32 LastWriteTime; /* should be zero or -1 */ __u16 ByteCount; /* 0 */ } __attribute__((packed)) CLOSE_REQ; diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 75f0603..8d30a5c 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -450,7 +450,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) } else if((pSMBr->hdr.WordCount == 13) && ((pSMBr->DialectIndex == LANMAN_PROT) || (pSMBr->DialectIndex == LANMAN2_PROT))) { - int tmp, adjust; + __s16 tmp; struct lanman_neg_rsp * rsp = (struct lanman_neg_rsp *)pSMBr; if((secFlags & CIFSSEC_MAY_LANMAN) || @@ -476,14 +476,16 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->maxRw = 0;/* we do not need to use raw anyway */ server->capabilities = CAP_MPX_MODE; } - tmp = le16_to_cpu(rsp->ServerTimeZone); - if (tmp == (int)0xffff) { + tmp = (__s16)le16_to_cpu(rsp->ServerTimeZone); + if (tmp == 0xffff) { /* OS/2 often does not set timezone therefore * we must use server time to calc time zone. - * Could deviate slightly from the right zone. Not easy - * to adjust, since timezones are not always a multiple - * of 60 (sometimes 30 minutes - are there smaller?) + * Could deviate slightly from the right zone. + * Smallest defined timezone difference is 15 minutes + * (i.e. Nepal). Rounding up/down is done to match + * this requirement. */ + int val, seconds, remain, result; struct timespec ts, utc; utc = CURRENT_TIME; ts = cnvrtDosUnixTm(le16_to_cpu(rsp->SrvTime.Date), @@ -491,12 +493,18 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) cFYI(1,("SrvTime: %d sec since 1970 (utc: %d) diff: %d", (int)ts.tv_sec, (int)utc.tv_sec, (int)(utc.tv_sec - ts.tv_sec))); - tmp = (int)(utc.tv_sec - ts.tv_sec); - adjust = tmp < 0 ? -29 : 29; - tmp = ((tmp + adjust) / 60) * 60; - server->timeAdj = tmp; + val = (int)(utc.tv_sec - ts.tv_sec); + seconds = val < 0 ? -val : val; + result = (seconds / IN_TZ_ADJ) * MIN_TZ_ADJ; + remain = seconds % MIN_TZ_ADJ; + if(remain >= (MIN_TZ_ADJ / 2)) + result += MIN_TZ_ADJ; + if(val < 0) + result = - result; + server->timeAdj = result; } else { - server->timeAdj = tmp * 60; /* also in seconds */ + server->timeAdj = (int)tmp; + server->timeAdj *= 60; /* also in seconds */ } cFYI(1,("server->timeAdj: %d seconds", server->timeAdj)); @@ -559,7 +567,8 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) cFYI(0, ("Max buf = %d", ses->server->maxBuf)); GETU32(ses->server->sessid) = le32_to_cpu(pSMBr->SessionKey); server->capabilities = le32_to_cpu(pSMBr->Capabilities); - server->timeAdj = le16_to_cpu(pSMBr->ServerTimeZone) * 60; + server->timeAdj = (int)(__s16)le16_to_cpu(pSMBr->ServerTimeZone); + server->timeAdj *= 60; if (pSMBr->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { memcpy(server->cryptKey, pSMBr->u.EncryptionKey, CIFS_CRYPTO_KEY_SIZE); @@ -1645,7 +1654,7 @@ CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, int smb_file_id) pSMBr = (CLOSE_RSP *)pSMB; /* BB removeme BB */ pSMB->FileID = (__u16) smb_file_id; - pSMB->LastWriteTime = 0; + pSMB->LastWriteTime = 0xFFFFFFFF; pSMB->ByteCount = 0; rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); -- cgit v0.10.2 From 947a50679570ef7a66e3e3107e95943a1cb14d08 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 2 Oct 2006 05:55:25 +0000 Subject: [CIFS] Fix typo Signed-off-by: Steve French diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 8d30a5c..005fb31 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -495,7 +495,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) (int)(utc.tv_sec - ts.tv_sec))); val = (int)(utc.tv_sec - ts.tv_sec); seconds = val < 0 ? -val : val; - result = (seconds / IN_TZ_ADJ) * MIN_TZ_ADJ; + result = (seconds / MIN_TZ_ADJ) * MIN_TZ_ADJ; remain = seconds % MIN_TZ_ADJ; if(remain >= (MIN_TZ_ADJ / 2)) result += MIN_TZ_ADJ; -- cgit v0.10.2 From 1a70d6529ad9f5978af846440f8a809784d6e813 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 2 Oct 2006 05:59:18 +0000 Subject: [CIFS] Fix compiler warning with previous patch Signed-off-by: Steve French diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 005fb31..79a01d3 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -477,7 +477,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->capabilities = CAP_MPX_MODE; } tmp = (__s16)le16_to_cpu(rsp->ServerTimeZone); - if (tmp == 0xffff) { + if (tmp == -1) { /* OS/2 often does not set timezone therefore * we must use server time to calc time zone. * Could deviate slightly from the right zone. -- cgit v0.10.2 From 485ae77dc7f484563707557ccf8c5d228980619f Mon Sep 17 00:00:00 2001 From: Sven Anders Date: Thu, 24 Aug 2006 17:11:50 +0200 Subject: [WATCHDOG] Winbond SMsC37B787 watchdog driver New watchdog driver for the Winbond SMsC37B787 chipset. Signed-off-by: Sven Anders Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index 89e46d6d..8d2ebc7 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -395,6 +395,26 @@ config CPU5_WDT To compile this driver as a module, choose M here: the module will be called cpu5wdt. +config SMSC37B787_WDT + tristate "Winbond SMsC37B787 Watchdog Timer" + depends on WATCHDOG && X86 + ---help--- + This is the driver for the hardware watchdog component on the + Winbond SMsC37B787 chipset as used on the NetRunner Mainboard + from Vision Systems and maybe others. + + This watchdog simply watches your kernel to make sure it doesn't + freeze, and if it does, it reboots your computer after a certain + amount of time. + + Usually a userspace daemon will notify the kernel WDT driver that + userspace is still alive, at regular intervals. + + To compile this driver as a module, choose M here: the + module will be called smsc37b787_wdt. + + Most people will say N. + config W83627HF_WDT tristate "W83627HF Watchdog Timer" depends on WATCHDOG && X86 diff --git a/drivers/char/watchdog/Makefile b/drivers/char/watchdog/Makefile index 7f70aba..630526f 100644 --- a/drivers/char/watchdog/Makefile +++ b/drivers/char/watchdog/Makefile @@ -53,6 +53,7 @@ obj-$(CONFIG_SCx200_WDT) += scx200_wdt.o obj-$(CONFIG_60XX_WDT) += sbc60xxwdt.o obj-$(CONFIG_SBC8360_WDT) += sbc8360.o obj-$(CONFIG_CPU5_WDT) += cpu5wdt.o +obj-$(CONFIG_SMSC37B787_WDT) += smsc37b787_wdt.o obj-$(CONFIG_W83627HF_WDT) += w83627hf_wdt.o obj-$(CONFIG_W83877F_WDT) += w83877f_wdt.o obj-$(CONFIG_W83977F_WDT) += w83977f_wdt.o diff --git a/drivers/char/watchdog/smsc37b787_wdt.c b/drivers/char/watchdog/smsc37b787_wdt.c new file mode 100644 index 0000000..47141c0 --- /dev/null +++ b/drivers/char/watchdog/smsc37b787_wdt.c @@ -0,0 +1,614 @@ +/* + * SMsC 37B787 Watchdog Timer driver for Linux 2.6.x.x + * + * Based on acquirewdt.c by Alan Cox + * and some other existing drivers + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * The authors do NOT admit liability nor provide warranty for + * any of this software. This material is provided "AS-IS" in + * the hope that it may be useful for others. + * + * (C) Copyright 2003-2006 Sven Anders + * + * History: + * 2003 - Created version 1.0 for Linux 2.4.x. + * 2006 - Ported to Linux 2.6, added nowayout and MAGICCLOSE + * features. Released version 1.1 + * + * Theory of operation: + * + * A Watchdog Timer (WDT) is a hardware circuit that can + * reset the computer system in case of a software fault. + * You probably knew that already. + * + * Usually a userspace daemon will notify the kernel WDT driver + * via the /dev/watchdog special device file that userspace is + * still alive, at regular intervals. When such a notification + * occurs, the driver will usually tell the hardware watchdog + * that everything is in order, and that the watchdog should wait + * for yet another little while to reset the system. + * If userspace fails (RAM error, kernel bug, whatever), the + * notifications cease to occur, and the hardware watchdog will + * reset the system (causing a reboot) after the timeout occurs. + * + * Create device with: + * mknod /dev/watchdog c 10 130 + * + * For an example userspace keep-alive daemon, see: + * Documentation/watchdog/watchdog.txt + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +/* enable support for minutes as units? */ +/* (does not always work correctly, so disabled by default!) */ +#define SMSC_SUPPORT_MINUTES +#undef SMSC_SUPPORT_MINUTES + +#define MAX_TIMEOUT 255 + +#define UNIT_SECOND 0 +#define UNIT_MINUTE 1 + +#define MODNAME "smsc37b787_wdt: " +#define VERSION "1.1" + +#define WATCHDOG_MINOR 130 + +#define IOPORT 0x3F0 +#define IOPORT_SIZE 2 +#define IODEV_NO 8 + +static int unit = UNIT_SECOND; /* timer's unit */ +static int timeout = 60; /* timeout value: default is 60 "units" */ +static int timer_enabled = 0; /* is the timer enabled? */ + +static char expect_close; /* is the close expected? */ + +static int nowayout = WATCHDOG_NOWAYOUT; + +/* -- Low level function ----------------------------------------*/ + +/* unlock the IO chip */ + +static inline void open_io_config(void) +{ + outb(0x55, IOPORT); + mdelay(1); + outb(0x55, IOPORT); +} + +/* lock the IO chip */ +static inline void close_io_config(void) +{ + outb(0xAA, IOPORT); +} + +/* select the IO device */ +static inline void select_io_device(unsigned char devno) +{ + outb(0x07, IOPORT); + outb(devno, IOPORT+1); +} + +/* write to the control register */ +static inline void write_io_cr(unsigned char reg, unsigned char data) +{ + outb(reg, IOPORT); + outb(data, IOPORT+1); +} + +/* read from the control register */ +static inline char read_io_cr(unsigned char reg) +{ + outb(reg, IOPORT); + return inb(IOPORT+1); +} + +/* -- Medium level functions ------------------------------------*/ + +static inline void gpio_bit12(unsigned char reg) +{ + // -- General Purpose I/O Bit 1.2 -- + // Bit 0, In/Out: 0 = Output, 1 = Input + // Bit 1, Polarity: 0 = No Invert, 1 = Invert + // Bit 2, Group Enable Intr.: 0 = Disable, 1 = Enable + // Bit 3/4, Function select: 00 = GPI/O, 01 = WDT, 10 = P17, + // 11 = Either Edge Triggered Intr. 2 + // Bit 5/6 (Reserved) + // Bit 7, Output Type: 0 = Push Pull Bit, 1 = Open Drain + write_io_cr(0xE2, reg); +} + +static inline void gpio_bit13(unsigned char reg) +{ + // -- General Purpose I/O Bit 1.3 -- + // Bit 0, In/Out: 0 = Output, 1 = Input + // Bit 1, Polarity: 0 = No Invert, 1 = Invert + // Bit 2, Group Enable Intr.: 0 = Disable, 1 = Enable + // Bit 3, Function select: 0 = GPI/O, 1 = LED + // Bit 4-6 (Reserved) + // Bit 7, Output Type: 0 = Push Pull Bit, 1 = Open Drain + write_io_cr(0xE3, reg); +} + +static inline void wdt_timer_units(unsigned char new_units) +{ + // -- Watchdog timer units -- + // Bit 0-6 (Reserved) + // Bit 7, WDT Time-out Value Units Select + // (0 = Minutes, 1 = Seconds) + write_io_cr(0xF1, new_units); +} + +static inline void wdt_timeout_value(unsigned char new_timeout) +{ + // -- Watchdog Timer Time-out Value -- + // Bit 0-7 Binary coded units (0=Disabled, 1..255) + write_io_cr(0xF2, new_timeout); +} + +static inline void wdt_timer_conf(unsigned char conf) +{ + // -- Watchdog timer configuration -- + // Bit 0 Joystick enable: 0* = No Reset, 1 = Reset WDT upon Gameport I/O + // Bit 1 Keyboard enable: 0* = No Reset, 1 = Reset WDT upon KBD Intr. + // Bit 2 Mouse enable: 0* = No Reset, 1 = Reset WDT upon Mouse Intr. + // Bit 3 Reset the timer + // (Wrong in SMsC documentation? Given as: PowerLED Timout Enabled) + // Bit 4-7 WDT Interrupt Mapping: (0000* = Disabled, + // 0001=IRQ1, 0010=(Invalid), 0011=IRQ3 to 1111=IRQ15) + write_io_cr(0xF3, conf); +} + +static inline void wdt_timer_ctrl(unsigned char reg) +{ + // -- Watchdog timer control -- + // Bit 0 Status Bit: 0 = Timer counting, 1 = Timeout occured + // Bit 1 Power LED Toggle: 0 = Disable Toggle, 1 = Toggle at 1 Hz + // Bit 2 Force Timeout: 1 = Forces WD timeout event (self-cleaning) + // Bit 3 P20 Force Timeout enabled: + // 0 = P20 activity does not generate the WD timeout event + // 1 = P20 Allows rising edge of P20, from the keyboard + // controller, to force the WD timeout event. + // Bit 4 (Reserved) + // -- Soft power management -- + // Bit 5 Stop Counter: 1 = Stop software power down counter + // set via register 0xB8, (self-cleaning) + // (Upon read: 0 = Counter running, 1 = Counter stopped) + // Bit 6 Restart Counter: 1 = Restart software power down counter + // set via register 0xB8, (self-cleaning) + // Bit 7 SPOFF: 1 = Force software power down (self-cleaning) + + write_io_cr(0xF4, reg); +} + +/* -- Higher level functions ------------------------------------*/ + +/* initialize watchdog */ + +static void wb_smsc_wdt_initialize(void) +{ + unsigned char old; + + open_io_config(); + select_io_device(IODEV_NO); + + // enable the watchdog + gpio_bit13(0x08); // Select pin 80 = LED not GPIO + gpio_bit12(0x0A); // Set pin 79 = WDT not GPIO/Output/Polarity=Invert + + // disable the timeout + wdt_timeout_value(0); + + // reset control register + wdt_timer_ctrl(0x00); + + // reset configuration register + wdt_timer_conf(0x00); + + // read old (timer units) register + old = read_io_cr(0xF1) & 0x7F; + if (unit == UNIT_SECOND) old |= 0x80; // set to seconds + + // set the watchdog timer units + wdt_timer_units(old); + + close_io_config(); +} + +/* shutdown the watchdog */ + +static void wb_smsc_wdt_shutdown(void) +{ + open_io_config(); + select_io_device(IODEV_NO); + + // disable the watchdog + gpio_bit13(0x09); + gpio_bit12(0x09); + + // reset watchdog config register + wdt_timer_conf(0x00); + + // reset watchdog control register + wdt_timer_ctrl(0x00); + + // disable timeout + wdt_timeout_value(0x00); + + close_io_config(); +} + +/* set timeout => enable watchdog */ + +static void wb_smsc_wdt_set_timeout(unsigned char new_timeout) +{ + open_io_config(); + select_io_device(IODEV_NO); + + // set Power LED to blink, if we enable the timeout + wdt_timer_ctrl((new_timeout == 0) ? 0x00 : 0x02); + + // set timeout value + wdt_timeout_value(new_timeout); + + close_io_config(); +} + +/* get timeout */ + +static unsigned char wb_smsc_wdt_get_timeout(void) +{ + unsigned char set_timeout; + + open_io_config(); + select_io_device(IODEV_NO); + set_timeout = read_io_cr(0xF2); + close_io_config(); + + return set_timeout; +} + +/* disable watchdog */ + +static void wb_smsc_wdt_disable(void) +{ + // set the timeout to 0 to disable the watchdog + wb_smsc_wdt_set_timeout(0); +} + +/* enable watchdog by setting the current timeout */ + +static void wb_smsc_wdt_enable(void) +{ + // set the current timeout... + wb_smsc_wdt_set_timeout(timeout); +} + +/* reset the timer */ + +static void wb_smsc_wdt_reset_timer(void) +{ + open_io_config(); + select_io_device(IODEV_NO); + + // reset the timer + wdt_timeout_value(timeout); + wdt_timer_conf(0x08); + + close_io_config(); +} + +/* return, if the watchdog is enabled (timeout is set...) */ + +static int wb_smsc_wdt_status(void) +{ + return (wb_smsc_wdt_get_timeout() == 0) ? 0 : WDIOF_KEEPALIVEPING; +} + + +/* -- File operations -------------------------------------------*/ + +/* open => enable watchdog and set initial timeout */ + +static int wb_smsc_wdt_open(struct inode *inode, struct file *file) +{ + /* /dev/watchdog can only be opened once */ + + if (timer_enabled) + return -EBUSY; + + if (nowayout) + __module_get(THIS_MODULE); + + /* Reload and activate timer */ + timer_enabled = 1; + wb_smsc_wdt_enable(); + + printk(KERN_INFO MODNAME "Watchdog enabled. Timeout set to %d %s.\n", timeout, (unit == UNIT_SECOND) ? "second(s)" : "minute(s)"); + + return nonseekable_open(inode, file); +} + +/* close => shut off the timer */ + +static int wb_smsc_wdt_release(struct inode *inode, struct file *file) +{ + /* Shut off the timer. */ + + if (expect_close == 42) { + wb_smsc_wdt_disable(); + printk(KERN_INFO MODNAME "Watchdog disabled, sleeping again...\n"); + } else { + printk(KERN_CRIT MODNAME "Unexpected close, not stopping watchdog!\n"); + wb_smsc_wdt_reset_timer(); + } + + timer_enabled = 0; + expect_close = 0; + return 0; +} + +/* write => update the timer to keep the machine alive */ + +static ssize_t wb_smsc_wdt_write(struct file *file, const char __user *data, + size_t len, loff_t *ppos) +{ + /* See if we got the magic character 'V' and reload the timer */ + if (len) { + if (!nowayout) { + size_t i; + + /* reset expect flag */ + expect_close = 0; + + /* scan to see whether or not we got the magic character */ + for (i = 0; i != len; i++) { + char c; + if (get_user(c, data+i)) + return -EFAULT; + if (c == 'V') + expect_close = 42; + } + } + + /* someone wrote to us, we should reload the timer */ + wb_smsc_wdt_reset_timer(); + } + return len; +} + +/* ioctl => control interface */ + +static int wb_smsc_wdt_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg) +{ + int new_timeout; + + union { + struct watchdog_info __user *ident; + int __user *i; + } uarg; + + static struct watchdog_info ident = { + .options = WDIOF_KEEPALIVEPING | + WDIOF_SETTIMEOUT | + WDIOF_MAGICCLOSE, + .firmware_version = 0, + .identity = "SMsC 37B787 Watchdog" + }; + + uarg.i = (int __user *)arg; + + switch (cmd) { + default: + return -ENOTTY; + + case WDIOC_GETSUPPORT: + return copy_to_user(uarg.ident, &ident, sizeof(ident)); + + case WDIOC_GETSTATUS: + return put_user(wb_smsc_wdt_status(), uarg.i); + + case WDIOC_GETBOOTSTATUS: + return put_user(0, uarg.i); + + case WDIOC_KEEPALIVE: + wb_smsc_wdt_reset_timer(); + return 0; + + case WDIOC_SETTIMEOUT: + if (get_user(new_timeout, uarg.i)) + return -EFAULT; + + // the API states this is given in secs + if (unit == UNIT_MINUTE) + new_timeout /= 60; + + if (new_timeout < 0 || new_timeout > MAX_TIMEOUT) + return -EINVAL; + + timeout = new_timeout; + wb_smsc_wdt_set_timeout(timeout); + + // fall through and return the new timeout... + + case WDIOC_GETTIMEOUT: + + new_timeout = timeout; + + if (unit == UNIT_MINUTE) + new_timeout *= 60; + + return put_user(new_timeout, uarg.i); + + case WDIOC_SETOPTIONS: + { + int options, retval = -EINVAL; + + if (get_user(options, uarg.i)) + return -EFAULT; + + if (options & WDIOS_DISABLECARD) { + wb_smsc_wdt_disable(); + retval = 0; + } + + if (options & WDIOS_ENABLECARD) { + wb_smsc_wdt_enable(); + retval = 0; + } + + return retval; + } + } +} + +/* -- Notifier funtions -----------------------------------------*/ + +static int wb_smsc_wdt_notify_sys(struct notifier_block *this, unsigned long code, void *unused) +{ + if (code == SYS_DOWN || code == SYS_HALT) + { + // set timeout to 0, to avoid possible race-condition + timeout = 0; + wb_smsc_wdt_disable(); + } + return NOTIFY_DONE; +} + +/* -- Module's structures ---------------------------------------*/ + +static struct file_operations wb_smsc_wdt_fops = +{ + .owner = THIS_MODULE, + .llseek = no_llseek, + .write = wb_smsc_wdt_write, + .ioctl = wb_smsc_wdt_ioctl, + .open = wb_smsc_wdt_open, + .release = wb_smsc_wdt_release +}; + +static struct notifier_block wb_smsc_wdt_notifier = +{ + .notifier_call = wb_smsc_wdt_notify_sys +}; + +static struct miscdevice wb_smsc_wdt_miscdev = +{ + .minor = WATCHDOG_MINOR, + .name = "watchdog", + .fops = &wb_smsc_wdt_fops, +}; + +/* -- Module init functions -------------------------------------*/ + +/* module's "constructor" */ + +static int __init wb_smsc_wdt_init(void) +{ + int ret; + + printk("SMsC 37B787 watchdog component driver " VERSION " initialising...\n"); + + if (!request_region(IOPORT, IOPORT_SIZE, "SMsC 37B787 watchdog")) { + printk(KERN_ERR MODNAME "Unable to register IO port %#x\n", IOPORT); + ret = -EBUSY; + goto out_pnp; + } + + ret = register_reboot_notifier(&wb_smsc_wdt_notifier); + if (ret) { + printk(KERN_ERR MODNAME "Unable to register reboot notifier err = %d\n", ret); + goto out_io; + } + + ret = misc_register(&wb_smsc_wdt_miscdev); + if (ret) { + printk(KERN_ERR MODNAME "Unable to register miscdev on minor %d\n", WATCHDOG_MINOR); + goto out_rbt; + } + + // init the watchdog timer + wb_smsc_wdt_initialize(); + + // set new maximum, if it's too big + if (timeout > MAX_TIMEOUT) + timeout = MAX_TIMEOUT; + + // output info + printk(KERN_INFO MODNAME "Timeout set to %d %s.\n", timeout, (unit == UNIT_SECOND) ? "second(s)" : "minute(s)"); + printk(KERN_INFO MODNAME "Watchdog initialized and sleeping (nowayout=%d)...\n", nowayout); + + // ret = 0 + +out_clean: + return ret; + +out_rbt: + unregister_reboot_notifier(&wb_smsc_wdt_notifier); + +out_io: + release_region(IOPORT, IOPORT_SIZE); + +out_pnp: + goto out_clean; +} + +/* module's "destructor" */ + +static void __exit wb_smsc_wdt_exit(void) +{ + /* Stop the timer before we leave */ + if (!nowayout) + { + wb_smsc_wdt_shutdown(); + printk(KERN_INFO MODNAME "Watchdog disabled.\n"); + } + + misc_deregister(&wb_smsc_wdt_miscdev); + unregister_reboot_notifier(&wb_smsc_wdt_notifier); + release_region(IOPORT, IOPORT_SIZE); + + printk("SMsC 37B787 watchdog component driver removed.\n"); +} + +module_init(wb_smsc_wdt_init); +module_exit(wb_smsc_wdt_exit); + +MODULE_AUTHOR("Sven Anders "); +MODULE_DESCRIPTION("Driver for SMsC 37B787 watchdog component (Version " VERSION ")"); +MODULE_LICENSE("GPL"); + +MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); + +#ifdef SMSC_SUPPORT_MINUTES +module_param(unit, int, 0); +MODULE_PARM_DESC(unit, "set unit to use, 0=seconds or 1=minutes, default is 0"); +#endif + +module_param(timeout, int, 60); +MODULE_PARM_DESC(timeout, "range is 1-255 units, default is 60"); + +module_param(nowayout, int, 0); +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CONFIG_WATCHDOG_NOWAYOUT)"); -- cgit v0.10.2 From 8386c8cfb2131b2a9caae3db6bf94292bbbe1caf Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sat, 2 Sep 2006 19:32:26 +0200 Subject: [WATCHDOG] Winbond SMsC37B787 - remove trailing whitespace Remove trailing whitespace. Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/smsc37b787_wdt.c b/drivers/char/watchdog/smsc37b787_wdt.c index 47141c0..1d01b30 100644 --- a/drivers/char/watchdog/smsc37b787_wdt.c +++ b/drivers/char/watchdog/smsc37b787_wdt.c @@ -8,7 +8,7 @@ * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. - * + * * The authors do NOT admit liability nor provide warranty for * any of this software. This material is provided "AS-IS" in * the hope that it may be useful for others. @@ -422,7 +422,7 @@ static int wb_smsc_wdt_ioctl(struct inode *inode, struct file *file, switch (cmd) { default: - return -ENOTTY; + return -ENOTTY; case WDIOC_GETSUPPORT: return copy_to_user(uarg.ident, &ident, sizeof(ident)); @@ -573,7 +573,7 @@ out_io: out_pnp: goto out_clean; -} +} /* module's "destructor" */ -- cgit v0.10.2 From aa1fd4d7c3b131026bf156da40fdf94bcbd705aa Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sat, 2 Sep 2006 20:53:19 +0200 Subject: [WATCHDOG] Winbond SMsC37B787 watchdog fixes * Added io spinlocking * Deleted WATCHDOG_MINOR (it's in the miscdevice include * Changed timer_enabled to use set_bit functions * WDIOC_GETSUPPORT should return -EFAULT or 0 * timeout should be correct before we initialize the watchdog * we should initialize the watchdog before we give access to userspace * Third parameter of module_param is not the default or initial value Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/smsc37b787_wdt.c b/drivers/char/watchdog/smsc37b787_wdt.c index 1d01b30..9f56913 100644 --- a/drivers/char/watchdog/smsc37b787_wdt.c +++ b/drivers/char/watchdog/smsc37b787_wdt.c @@ -54,6 +54,7 @@ #include #include #include +#include #include #include @@ -72,18 +73,18 @@ #define MODNAME "smsc37b787_wdt: " #define VERSION "1.1" -#define WATCHDOG_MINOR 130 - #define IOPORT 0x3F0 #define IOPORT_SIZE 2 #define IODEV_NO 8 static int unit = UNIT_SECOND; /* timer's unit */ static int timeout = 60; /* timeout value: default is 60 "units" */ -static int timer_enabled = 0; /* is the timer enabled? */ +static unsigned long timer_enabled = 0; /* is the timer enabled? */ static char expect_close; /* is the close expected? */ +static spinlock_t io_lock; /* to guard the watchdog from io races */ + static int nowayout = WATCHDOG_NOWAYOUT; /* -- Low level function ----------------------------------------*/ @@ -210,6 +211,7 @@ static void wb_smsc_wdt_initialize(void) { unsigned char old; + spin_lock(&io_lock); open_io_config(); select_io_device(IODEV_NO); @@ -234,12 +236,14 @@ static void wb_smsc_wdt_initialize(void) wdt_timer_units(old); close_io_config(); + spin_unlock(&io_lock); } /* shutdown the watchdog */ static void wb_smsc_wdt_shutdown(void) { + spin_lock(&io_lock); open_io_config(); select_io_device(IODEV_NO); @@ -257,12 +261,14 @@ static void wb_smsc_wdt_shutdown(void) wdt_timeout_value(0x00); close_io_config(); + spin_unlock(&io_lock); } /* set timeout => enable watchdog */ static void wb_smsc_wdt_set_timeout(unsigned char new_timeout) { + spin_lock(&io_lock); open_io_config(); select_io_device(IODEV_NO); @@ -273,6 +279,7 @@ static void wb_smsc_wdt_set_timeout(unsigned char new_timeout) wdt_timeout_value(new_timeout); close_io_config(); + spin_unlock(&io_lock); } /* get timeout */ @@ -281,10 +288,12 @@ static unsigned char wb_smsc_wdt_get_timeout(void) { unsigned char set_timeout; + spin_lock(&io_lock); open_io_config(); select_io_device(IODEV_NO); set_timeout = read_io_cr(0xF2); close_io_config(); + spin_unlock(&io_lock); return set_timeout; } @@ -309,6 +318,7 @@ static void wb_smsc_wdt_enable(void) static void wb_smsc_wdt_reset_timer(void) { + spin_lock(&io_lock); open_io_config(); select_io_device(IODEV_NO); @@ -317,6 +327,7 @@ static void wb_smsc_wdt_reset_timer(void) wdt_timer_conf(0x08); close_io_config(); + spin_unlock(&io_lock); } /* return, if the watchdog is enabled (timeout is set...) */ @@ -335,14 +346,13 @@ static int wb_smsc_wdt_open(struct inode *inode, struct file *file) { /* /dev/watchdog can only be opened once */ - if (timer_enabled) + if (test_and_set_bit(0, &timer_enabled)) return -EBUSY; if (nowayout) __module_get(THIS_MODULE); /* Reload and activate timer */ - timer_enabled = 1; wb_smsc_wdt_enable(); printk(KERN_INFO MODNAME "Watchdog enabled. Timeout set to %d %s.\n", timeout, (unit == UNIT_SECOND) ? "second(s)" : "minute(s)"); @@ -364,7 +374,7 @@ static int wb_smsc_wdt_release(struct inode *inode, struct file *file) wb_smsc_wdt_reset_timer(); } - timer_enabled = 0; + clear_bit(0, &timer_enabled); expect_close = 0; return 0; } @@ -425,7 +435,8 @@ static int wb_smsc_wdt_ioctl(struct inode *inode, struct file *file, return -ENOTTY; case WDIOC_GETSUPPORT: - return copy_to_user(uarg.ident, &ident, sizeof(ident)); + return copy_to_user(uarg.ident, &ident, + sizeof(ident)) ? -EFAULT : 0; case WDIOC_GETSTATUS: return put_user(wb_smsc_wdt_status(), uarg.i); @@ -506,12 +517,12 @@ static struct file_operations wb_smsc_wdt_fops = .write = wb_smsc_wdt_write, .ioctl = wb_smsc_wdt_ioctl, .open = wb_smsc_wdt_open, - .release = wb_smsc_wdt_release + .release = wb_smsc_wdt_release, }; static struct notifier_block wb_smsc_wdt_notifier = { - .notifier_call = wb_smsc_wdt_notify_sys + .notifier_call = wb_smsc_wdt_notify_sys, }; static struct miscdevice wb_smsc_wdt_miscdev = @@ -529,6 +540,8 @@ static int __init wb_smsc_wdt_init(void) { int ret; + spin_lock_init(&io_lock); + printk("SMsC 37B787 watchdog component driver " VERSION " initialising...\n"); if (!request_region(IOPORT, IOPORT_SIZE, "SMsC 37B787 watchdog")) { @@ -537,6 +550,13 @@ static int __init wb_smsc_wdt_init(void) goto out_pnp; } + // set new maximum, if it's too big + if (timeout > MAX_TIMEOUT) + timeout = MAX_TIMEOUT; + + // init the watchdog timer + wb_smsc_wdt_initialize(); + ret = register_reboot_notifier(&wb_smsc_wdt_notifier); if (ret) { printk(KERN_ERR MODNAME "Unable to register reboot notifier err = %d\n", ret); @@ -549,13 +569,6 @@ static int __init wb_smsc_wdt_init(void) goto out_rbt; } - // init the watchdog timer - wb_smsc_wdt_initialize(); - - // set new maximum, if it's too big - if (timeout > MAX_TIMEOUT) - timeout = MAX_TIMEOUT; - // output info printk(KERN_INFO MODNAME "Timeout set to %d %s.\n", timeout, (unit == UNIT_SECOND) ? "second(s)" : "minute(s)"); printk(KERN_INFO MODNAME "Watchdog initialized and sleeping (nowayout=%d)...\n", nowayout); @@ -607,7 +620,7 @@ module_param(unit, int, 0); MODULE_PARM_DESC(unit, "set unit to use, 0=seconds or 1=minutes, default is 0"); #endif -module_param(timeout, int, 60); +module_param(timeout, int, 0); MODULE_PARM_DESC(timeout, "range is 1-255 units, default is 60"); module_param(nowayout, int, 0); -- cgit v0.10.2 From f9a8c8913a95aed91bfa81f7d4043c6430423bf8 Mon Sep 17 00:00:00 2001 From: Marcus Junker Date: Thu, 24 Aug 2006 17:11:50 +0200 Subject: [WATCHDOG] w83697hf WDT driver New watchdog driver for the Winbond W83697HF chipset. Signed-off-by: Marcus Junker Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index 8d2ebc7..3e67e01 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -430,6 +430,19 @@ config W83627HF_WDT Most people will say N. +config W83697HF_WDT + tristate "W83697HF Watchdog Timer" + depends on WATCHDOG && X86 + ---help--- + This is the driver for the hardware watchdog on the W83697HF chipset + This watchdog simply watches your kernel to make sure it doesn't freeze, + and if it does, it reboots your computer after a certain amount of time. + + To compile this driver as a module, choose M here: the + module will be called w83697hf_wdt. + + Most people will say N. + config W83877F_WDT tristate "W83877F (EMACS) Watchdog Timer" depends on WATCHDOG && X86 diff --git a/drivers/char/watchdog/Makefile b/drivers/char/watchdog/Makefile index 630526f..ee34741 100644 --- a/drivers/char/watchdog/Makefile +++ b/drivers/char/watchdog/Makefile @@ -55,6 +55,7 @@ obj-$(CONFIG_SBC8360_WDT) += sbc8360.o obj-$(CONFIG_CPU5_WDT) += cpu5wdt.o obj-$(CONFIG_SMSC37B787_WDT) += smsc37b787_wdt.o obj-$(CONFIG_W83627HF_WDT) += w83627hf_wdt.o +obj-$(CONFIG_W83697HF_WDT) += w83697hf_wdt.o obj-$(CONFIG_W83877F_WDT) += w83877f_wdt.o obj-$(CONFIG_W83977F_WDT) += w83977f_wdt.o obj-$(CONFIG_MACHZ_WDT) += machzwd.o diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c new file mode 100644 index 0000000..ef6612e --- /dev/null +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -0,0 +1,367 @@ +/* + * w83697hf WDT driver + * + * (c) Copyright 2006 Marcus Junker + * + * Based on w83627hf_wdt.c advantechwdt.c which is based on wdt.c. + * Original copyright messages: + * + * (c) Copyright 2003 Pádraig Brady + * + * (c) Copyright 2000-2001 Marek Michalkiewicz + * + * (c) Copyright 1996 Alan Cox , All Rights Reserved. + * http://www.redhat.com + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Neither Marcus Junker nor ANDURAS AG admit liability nor provide + * warranty for any of this software. This material is provided + * "AS-IS" and at no charge. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#define WATCHDOG_NAME "w83697hf WDT" +#define PFX WATCHDOG_NAME ": " +#define WATCHDOG_TIMEOUT 60 /* 60 sec default timeout */ + +static unsigned long wdt_is_open; +static char expect_close; + +/* You must set this - there is no sane way to probe for this board. */ +static int wdt_io = 0x2E; +module_param(wdt_io, int, 0); +MODULE_PARM_DESC(wdt_io, "w83697hf WDT io port (default 0x2E)"); + +static int timeout = WATCHDOG_TIMEOUT; /* in seconds */ +module_param(timeout, int, 0); +MODULE_PARM_DESC(timeout, "Watchdog timeout in seconds. 1<= timeout <=63, default=" __MODULE_STRING(WATCHDOG_TIMEOUT) "."); + +static int nowayout = WATCHDOG_NOWAYOUT; +module_param(nowayout, int, 0); +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CONFIG_WATCHDOG_NOWAYOUT)"); + +/* + * Kernel methods. + */ + +#define WDT_EFER (wdt_io+0) /* Extended Function Enable Registers */ +#define WDT_EFIR (wdt_io+0) /* Extended Function Index Register (same as EFER) */ +#define WDT_EFDR (WDT_EFIR+1) /* Extended Function Data Register */ + +static void +w83697hf_select_wd_register(void) +{ + outb_p(0x87, WDT_EFER); /* Enter extended function mode */ + outb_p(0x87, WDT_EFER); /* Again according to manual */ + + outb_p(0x29, WDT_EFER); /* select CR29 */ + outb_p(0x20, WDT_EFDR); /* select WDTO */ + + outb_p(0x07, WDT_EFER); /* point to logical device number reg */ + outb_p(0x08, WDT_EFDR); /* select logical device 8 (GPIO2) */ + outb_p(0x30, WDT_EFER); /* select CR30 */ + outb_p(0x01, WDT_EFDR); /* set bit 0 to activate GPIO2 */ +} + +static void +w83697hf_unselect_wd_register(void) +{ + outb_p(0xAA, WDT_EFER); /* Leave extended function mode */ +} + +/* tyan motherboards seem to set F5 to 0x4C ? + * So explicitly init to appropriate value. */ +static void +w83697hf_init(void) +{ + unsigned char t; + + w83697hf_select_wd_register(); + + outb_p(0xF3, WDT_EFER); /* Select CRF3 */ + + t=inb_p(WDT_EFDR); /* read CRF6 */ + if (t != 0) { + printk (KERN_INFO PFX "Watchdog already running. Resetting timeout to %d sec\n", timeout); + outb_p(timeout, WDT_EFDR); /* Write back to CRF6 */ + } + outb_p(0xF4, WDT_EFER); /* Select CRF4 */ + t=inb_p(WDT_EFDR); /* read CRF4 */ + t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ + outb_p(t, WDT_EFDR); /* Write back to CRF5 */ + + w83697hf_unselect_wd_register(); +} + +static void +wdt_ctrl(int timeout) +{ + w83697hf_select_wd_register(); + + outb_p(0xF4, WDT_EFER); /* Select CRF4 */ + outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF4 */ + + w83697hf_unselect_wd_register(); +} + +static int +wdt_ping(void) +{ + wdt_ctrl(timeout); + return 0; +} + +static int +wdt_disable(void) +{ + wdt_ctrl(0); + return 0; +} + +static int +wdt_set_heartbeat(int t) +{ + if ((t < 1) || (t > 63)) + return -EINVAL; + + timeout = t; + return 0; +} + +static ssize_t +wdt_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) +{ + if (count) { + if (!nowayout) { + size_t i; + + expect_close = 0; + + for (i = 0; i != count; i++) { + char c; + if (get_user(c, buf+i)) + return -EFAULT; + if (c == 'V') + expect_close = 42; + } + } + wdt_ping(); + } + return count; +} + +static int +wdt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, + unsigned long arg) +{ + void __user *argp = (void __user *)arg; + int __user *p = argp; + int new_timeout; + static struct watchdog_info ident = { + .options = WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE, + .firmware_version = 1, + .identity = "W83697HF WDT", + }; + + switch (cmd) { + case WDIOC_GETSUPPORT: + if (copy_to_user(argp, &ident, sizeof(ident))) + return -EFAULT; + break; + + case WDIOC_GETSTATUS: + case WDIOC_GETBOOTSTATUS: + return put_user(0, p); + + case WDIOC_KEEPALIVE: + wdt_ping(); + break; + + case WDIOC_SETTIMEOUT: + if (get_user(new_timeout, p)) + return -EFAULT; + if (wdt_set_heartbeat(new_timeout)) + return -EINVAL; + wdt_ping(); + /* Fall */ + + case WDIOC_GETTIMEOUT: + return put_user(timeout, p); + + case WDIOC_SETOPTIONS: + { + int options, retval = -EINVAL; + + if (get_user(options, p)) + return -EFAULT; + + if (options & WDIOS_DISABLECARD) { + wdt_disable(); + retval = 0; + } + + if (options & WDIOS_ENABLECARD) { + wdt_ping(); + retval = 0; + } + + return retval; + } + + default: + return -ENOIOCTLCMD; + } + return 0; +} + +static int +wdt_open(struct inode *inode, struct file *file) +{ + if (test_and_set_bit(0, &wdt_is_open)) + return -EBUSY; + /* + * Activate + */ + + wdt_ping(); + return nonseekable_open(inode, file); +} + +static int +wdt_close(struct inode *inode, struct file *file) +{ + if (expect_close == 42) { + wdt_disable(); + } else { + printk(KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n"); + wdt_ping(); + } + expect_close = 0; + clear_bit(0, &wdt_is_open); + return 0; +} + +/* + * Notifier for system down + */ + +static int +wdt_notify_sys(struct notifier_block *this, unsigned long code, + void *unused) +{ + if (code == SYS_DOWN || code == SYS_HALT) { + /* Turn the WDT off */ + wdt_disable(); + } + return NOTIFY_DONE; +} + +/* + * Kernel Interfaces + */ + +static struct file_operations wdt_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .write = wdt_write, + .ioctl = wdt_ioctl, + .open = wdt_open, + .release = wdt_close, +}; + +static struct miscdevice wdt_miscdev = { + .minor = WATCHDOG_MINOR, + .name = "watchdog", + .fops = &wdt_fops, +}; + +/* + * The WDT needs to learn about soft shutdowns in order to + * turn the timebomb registers off. + */ + +static struct notifier_block wdt_notifier = { + .notifier_call = wdt_notify_sys, +}; + +static int __init +wdt_init(void) +{ + int ret; + + printk(KERN_INFO "WDT driver for the Winbond(TM) W83697HF Super I/O chip initialising.\n"); + + if (wdt_set_heartbeat(timeout)) { + wdt_set_heartbeat(WATCHDOG_TIMEOUT); + printk (KERN_INFO PFX "timeout value must be 1<=timeout<=63, using %d\n", + WATCHDOG_TIMEOUT); + } + + if (!request_region(wdt_io, 1, WATCHDOG_NAME)) { + printk (KERN_ERR PFX "I/O address 0x%04x already in use\n", + wdt_io); + ret = -EIO; + goto out; + } + + w83697hf_init(); + + ret = register_reboot_notifier(&wdt_notifier); + if (ret != 0) { + printk (KERN_ERR PFX "cannot register reboot notifier (err=%d)\n", + ret); + goto unreg_regions; + } + + ret = misc_register(&wdt_miscdev); + if (ret != 0) { + printk (KERN_ERR PFX "cannot register miscdev on minor=%d (err=%d)\n", + WATCHDOG_MINOR, ret); + goto unreg_reboot; + } + + printk (KERN_INFO PFX "initialized. timeout=%d sec (nowayout=%d)\n", + timeout, nowayout); + +out: + return ret; +unreg_reboot: + unregister_reboot_notifier(&wdt_notifier); +unreg_regions: + release_region(wdt_io, 1); + goto out; +} + +static void __exit +wdt_exit(void) +{ + misc_deregister(&wdt_miscdev); + unregister_reboot_notifier(&wdt_notifier); + release_region(wdt_io,1); +} + +module_init(wdt_init); +module_exit(wdt_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Marcus Junker "); +MODULE_DESCRIPTION("w83697hf WDT driver"); +MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); -- cgit v0.10.2 From e0845bf4e1df07e16fa39d96508a1ba4a480ce3e Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sat, 2 Sep 2006 17:59:54 +0200 Subject: [WATCHDOG] Kconfig clean-up * fix typo's according to spellings checker * Fix some leading and trailing spaces Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index 3e67e01..00b21db 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -13,7 +13,7 @@ config WATCHDOG subsequently opening the file and then failing to write to it for longer than 1 minute will result in rebooting the machine. This could be useful for a networked machine that needs to come back - online as fast as possible after a lock-up. There's both a watchdog + on-line as fast as possible after a lock-up. There's both a watchdog implementation entirely in software (which can sometimes fail to reboot the machine) and a driver for hardware watchdog boards, which are more robust and can also keep track of the temperature inside @@ -71,7 +71,7 @@ config 21285_WATCHDOG tristate "DC21285 watchdog" depends on WATCHDOG && FOOTBRIDGE help - The Intel Footbridge chip contains a builtin watchdog circuit. Say Y + The Intel Footbridge chip contains a built-in watchdog circuit. Say Y here if you wish to use this. Alternatively say M to compile the driver as a module, which will be called wdt285. @@ -273,7 +273,7 @@ config IBMASR depends on WATCHDOG && X86 help This is the driver for the IBM Automatic Server Restart watchdog - timer builtin into some eServer xSeries machines. + timer built-in into some eServer xSeries machines. To compile this driver as a module, choose M here: the module will be called ibmasr. @@ -431,17 +431,17 @@ config W83627HF_WDT Most people will say N. config W83697HF_WDT - tristate "W83697HF Watchdog Timer" - depends on WATCHDOG && X86 - ---help--- - This is the driver for the hardware watchdog on the W83697HF chipset - This watchdog simply watches your kernel to make sure it doesn't freeze, - and if it does, it reboots your computer after a certain amount of time. + tristate "W83697HF Watchdog Timer" + depends on WATCHDOG && X86 + ---help--- + This is the driver for the hardware watchdog on the W83697HF chipset + This watchdog simply watches your kernel to make sure it doesn't freeze, + and if it does, it reboots your computer after a certain amount of time. - To compile this driver as a module, choose M here: the - module will be called w83697hf_wdt. + To compile this driver as a module, choose M here: the + module will be called w83697hf_wdt. - Most people will say N. + Most people will say N. config W83877F_WDT tristate "W83877F (EMACS) Watchdog Timer" @@ -476,7 +476,7 @@ config MACHZ_WDT depends on WATCHDOG && X86 ---help--- If you are using a ZF Micro MachZ processor, say Y here, otherwise - N. This is the driver for the watchdog timer builtin on that + N. This is the driver for the watchdog timer built-in on that processor using ZF-Logic interface. This watchdog simply watches your kernel to make sure it doesn't freeze, and if it does, it reboots your computer after a certain amount of time. -- cgit v0.10.2 From ab9d441425559aa035ba6327f21e8922e8a13927 Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sat, 2 Sep 2006 18:50:20 +0200 Subject: [WATCHDOG] w836?7hf_wdt spinlock fixes. Add io spinlocks to prevent possible race conditions between start and stop operations that are issued from different child processes where the master process opened /dev/watchdog. Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83627hf_wdt.c b/drivers/char/watchdog/w83627hf_wdt.c index b4adc52..07d4bff 100644 --- a/drivers/char/watchdog/w83627hf_wdt.c +++ b/drivers/char/watchdog/w83627hf_wdt.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include @@ -44,6 +45,7 @@ static unsigned long wdt_is_open; static char expect_close; +static spinlock_t io_lock; /* You must set this - there is no sane way to probe for this board. */ static int wdt_io = 0x2E; @@ -110,12 +112,16 @@ w83627hf_init(void) static void wdt_ctrl(int timeout) { + spin_lock(&io_lock); + w83627hf_select_wd_register(); outb_p(0xF6, WDT_EFER); /* Select CRF6 */ outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF6 */ w83627hf_unselect_wd_register(); + + spin_unlock(&io_lock); } static int @@ -303,6 +309,8 @@ wdt_init(void) { int ret; + spin_lock_init(&io_lock); + printk(KERN_INFO "WDT driver for the Winbond(TM) W83627HF Super I/O chip initialising.\n"); if (wdt_set_heartbeat(timeout)) { diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index ef6612e..21e822e 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include @@ -44,6 +45,7 @@ static unsigned long wdt_is_open; static char expect_close; +static spinlock_t io_lock; /* You must set this - there is no sane way to probe for this board. */ static int wdt_io = 0x2E; @@ -114,12 +116,16 @@ w83697hf_init(void) static void wdt_ctrl(int timeout) { + spin_lock(&io_lock); + w83697hf_select_wd_register(); outb_p(0xF4, WDT_EFER); /* Select CRF4 */ outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF4 */ w83697hf_unselect_wd_register(); + + spin_unlock(&io_lock); } static int @@ -307,6 +313,8 @@ wdt_init(void) { int ret; + spin_lock_init(&io_lock); + printk(KERN_INFO "WDT driver for the Winbond(TM) W83697HF Super I/O chip initialising.\n"); if (wdt_set_heartbeat(timeout)) { -- cgit v0.10.2 From c310e2b950c949cfc14754baed877eadb1a26f6b Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sat, 2 Sep 2006 19:04:02 +0200 Subject: [WATCHDOG] Kconfig clean up fixed some more trailing spaces. Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index 00b21db..ecae59c 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -269,9 +269,9 @@ config IB700_WDT Most people will say N. config IBMASR - tristate "IBM Automatic Server Restart" - depends on WATCHDOG && X86 - help + tristate "IBM Automatic Server Restart" + depends on WATCHDOG && X86 + help This is the driver for the IBM Automatic Server Restart watchdog timer built-in into some eServer xSeries machines. @@ -505,7 +505,6 @@ config SBC_EPX_C3_WATCHDOG To compile this driver as a module, choose M here: the module will be called sbc_epx_c3. - # PowerPC Architecture config 8xx_WDT @@ -535,7 +534,7 @@ config WATCHDOG_RTAS help This driver adds watchdog support for the RTAS watchdog. - To compile this driver as a module, choose M here. The module + To compile this driver as a module, choose M here. The module will be called wdrtas. # MIPS Architecture -- cgit v0.10.2 From 196f29c8e8cd3352d26ed7bdf44f622e14adb931 Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Wed, 13 Sep 2006 21:27:29 +0200 Subject: [WATCHDOG] use ENOTTY instead of ENOIOCTLCMD in ioctl() Return ENOTTY instead of ENOIOCTLCMD in user-visible ioctl() results The watchdog drivers used to return ENOIOCTLCMD for bad ioctl() commands. ENOIOCTLCMD should not be visible by the user, so use ENOTTY instead. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck Acked-by: Alan Cox Signed-off-by: Andrew Morton diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 21e822e..32710a9 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -233,7 +233,7 @@ wdt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, } default: - return -ENOIOCTLCMD; + return -ENOTTY; } return 0; } -- cgit v0.10.2 From 8de6fc1e2023954ec21d4e84d002839afed4cad3 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 1 This is patch 1 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - the note concerning tyan motherboards has been copied from another driver, This doesn't apply here. - the comments concerning CRF6 are wrong as CRF3 is manipulated and CRF6 is never read nor written. - the comments concerning CRF5 are wrong as CRF4 is manipulated and CRF5 is never read nor written. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 32710a9..c31121e 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -89,8 +89,6 @@ w83697hf_unselect_wd_register(void) outb_p(0xAA, WDT_EFER); /* Leave extended function mode */ } -/* tyan motherboards seem to set F5 to 0x4C ? - * So explicitly init to appropriate value. */ static void w83697hf_init(void) { @@ -100,15 +98,15 @@ w83697hf_init(void) outb_p(0xF3, WDT_EFER); /* Select CRF3 */ - t=inb_p(WDT_EFDR); /* read CRF6 */ + t=inb_p(WDT_EFDR); /* read CRF3 */ if (t != 0) { printk (KERN_INFO PFX "Watchdog already running. Resetting timeout to %d sec\n", timeout); - outb_p(timeout, WDT_EFDR); /* Write back to CRF6 */ + outb_p(timeout, WDT_EFDR); /* Write back to CRF3 */ } outb_p(0xF4, WDT_EFER); /* Select CRF4 */ t=inb_p(WDT_EFDR); /* read CRF4 */ t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ - outb_p(t, WDT_EFDR); /* Write back to CRF5 */ + outb_p(t, WDT_EFDR); /* Write back to CRF4 */ w83697hf_unselect_wd_register(); } -- cgit v0.10.2 From b41a9f59d13a4c4c3f0e0b8d9ff15743607096a2 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 2 This is patch 2 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - wdt_io is 2 bytes long. We should do a request_region for 2 bytes instead of 1. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index c31121e..4f81943 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -321,7 +321,7 @@ wdt_init(void) WATCHDOG_TIMEOUT); } - if (!request_region(wdt_io, 1, WATCHDOG_NAME)) { + if (!request_region(wdt_io, 2, WATCHDOG_NAME)) { printk (KERN_ERR PFX "I/O address 0x%04x already in use\n", wdt_io); ret = -EIO; @@ -352,7 +352,7 @@ out: unreg_reboot: unregister_reboot_notifier(&wdt_notifier); unreg_regions: - release_region(wdt_io, 1); + release_region(wdt_io, 2); goto out; } @@ -361,7 +361,7 @@ wdt_exit(void) { misc_deregister(&wdt_miscdev); unregister_reboot_notifier(&wdt_notifier); - release_region(wdt_io,1); + release_region(wdt_io, 2); } module_init(wdt_init); -- cgit v0.10.2 From db16525e63f8cf554696045e0e360b81e2263279 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 3 This is patch 3 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Fix identation. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 4f81943..12bdcab 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -6,7 +6,7 @@ * Based on w83627hf_wdt.c advantechwdt.c which is based on wdt.c. * Original copyright messages: * - * (c) Copyright 2003 Pádraig Brady + * (c) Copyright 2003 Pádraig Brady * * (c) Copyright 2000-2001 Marek Michalkiewicz * @@ -96,16 +96,16 @@ w83697hf_init(void) w83697hf_select_wd_register(); - outb_p(0xF3, WDT_EFER); /* Select CRF3 */ + outb_p(0xF3, WDT_EFER); /* Select CRF3 */ t=inb_p(WDT_EFDR); /* read CRF3 */ if (t != 0) { printk (KERN_INFO PFX "Watchdog already running. Resetting timeout to %d sec\n", timeout); outb_p(timeout, WDT_EFDR); /* Write back to CRF3 */ } - outb_p(0xF4, WDT_EFER); /* Select CRF4 */ - t=inb_p(WDT_EFDR); /* read CRF4 */ - t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ + outb_p(0xF4, WDT_EFER); /* Select CRF4 */ + t=inb_p(WDT_EFDR); /* read CRF4 */ + t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ outb_p(t, WDT_EFDR); /* Write back to CRF4 */ w83697hf_unselect_wd_register(); @@ -187,51 +187,51 @@ wdt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, switch (cmd) { case WDIOC_GETSUPPORT: - if (copy_to_user(argp, &ident, sizeof(ident))) - return -EFAULT; - break; + if (copy_to_user(argp, &ident, sizeof(ident))) + return -EFAULT; + break; case WDIOC_GETSTATUS: case WDIOC_GETBOOTSTATUS: - return put_user(0, p); + return put_user(0, p); case WDIOC_KEEPALIVE: - wdt_ping(); - break; + wdt_ping(); + break; case WDIOC_SETTIMEOUT: - if (get_user(new_timeout, p)) - return -EFAULT; - if (wdt_set_heartbeat(new_timeout)) - return -EINVAL; - wdt_ping(); - /* Fall */ + if (get_user(new_timeout, p)) + return -EFAULT; + if (wdt_set_heartbeat(new_timeout)) + return -EINVAL; + wdt_ping(); + /* Fall */ case WDIOC_GETTIMEOUT: - return put_user(timeout, p); + return put_user(timeout, p); case WDIOC_SETOPTIONS: { - int options, retval = -EINVAL; + int options, retval = -EINVAL; - if (get_user(options, p)) - return -EFAULT; + if (get_user(options, p)) + return -EFAULT; - if (options & WDIOS_DISABLECARD) { - wdt_disable(); - retval = 0; - } + if (options & WDIOS_DISABLECARD) { + wdt_disable(); + retval = 0; + } - if (options & WDIOS_ENABLECARD) { - wdt_ping(); - retval = 0; - } + if (options & WDIOS_ENABLECARD) { + wdt_ping(); + retval = 0; + } - return retval; + return retval; } default: - return -ENOTTY; + return -ENOTTY; } return 0; } @@ -255,7 +255,7 @@ wdt_close(struct inode *inode, struct file *file) if (expect_close == 42) { wdt_disable(); } else { - printk(KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n"); + printk (KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n"); wdt_ping(); } expect_close = 0; @@ -313,7 +313,7 @@ wdt_init(void) spin_lock_init(&io_lock); - printk(KERN_INFO "WDT driver for the Winbond(TM) W83697HF Super I/O chip initialising.\n"); + printk (KERN_INFO "WDT driver for the Winbond(TM) W83697HF Super I/O chip initialising.\n"); if (wdt_set_heartbeat(timeout)) { wdt_set_heartbeat(WATCHDOG_TIMEOUT); -- cgit v0.10.2 From eb64419e397aaea55b2ef6904e86b6263a80acc7 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 4 This is patch 4 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - limits the watchdog timeout to 1-63 while this device accepts 1-255. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 12bdcab..94b1655 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -54,7 +54,7 @@ MODULE_PARM_DESC(wdt_io, "w83697hf WDT io port (default 0x2E)"); static int timeout = WATCHDOG_TIMEOUT; /* in seconds */ module_param(timeout, int, 0); -MODULE_PARM_DESC(timeout, "Watchdog timeout in seconds. 1<= timeout <=63, default=" __MODULE_STRING(WATCHDOG_TIMEOUT) "."); +MODULE_PARM_DESC(timeout, "Watchdog timeout in seconds. 1<= timeout <=255, default=" __MODULE_STRING(WATCHDOG_TIMEOUT) "."); static int nowayout = WATCHDOG_NOWAYOUT; module_param(nowayout, int, 0); @@ -143,7 +143,7 @@ wdt_disable(void) static int wdt_set_heartbeat(int t) { - if ((t < 1) || (t > 63)) + if ((t < 1) || (t > 255)) return -EINVAL; timeout = t; @@ -317,7 +317,7 @@ wdt_init(void) if (wdt_set_heartbeat(timeout)) { wdt_set_heartbeat(WATCHDOG_TIMEOUT); - printk (KERN_INFO PFX "timeout value must be 1<=timeout<=63, using %d\n", + printk (KERN_INFO PFX "timeout value must be 1<=timeout<=255, using %d\n", WATCHDOG_TIMEOUT); } -- cgit v0.10.2 From 44d7d3282baa4080b73adca31648e6ef1e191874 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 5 This is patch 5 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Rename the Extended Function Registers to the names used in the data-sheet. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 94b1655..c44f281 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -64,29 +64,29 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CON * Kernel methods. */ -#define WDT_EFER (wdt_io+0) /* Extended Function Enable Registers */ -#define WDT_EFIR (wdt_io+0) /* Extended Function Index Register (same as EFER) */ -#define WDT_EFDR (WDT_EFIR+1) /* Extended Function Data Register */ +#define W83697HF_EFER (wdt_io+0) /* Extended Function Enable Register */ +#define W83697HF_EFIR (wdt_io+0) /* Extended Function Index Register (same as EFER) */ +#define W83697HF_EFDR (wdt_io+1) /* Extended Function Data Register */ static void w83697hf_select_wd_register(void) { - outb_p(0x87, WDT_EFER); /* Enter extended function mode */ - outb_p(0x87, WDT_EFER); /* Again according to manual */ + outb_p(0x87, W83697HF_EFER); /* Enter extended function mode */ + outb_p(0x87, W83697HF_EFER); /* Again according to manual */ - outb_p(0x29, WDT_EFER); /* select CR29 */ - outb_p(0x20, WDT_EFDR); /* select WDTO */ + outb_p(0x29, W83697HF_EFER); /* select CR29 */ + outb_p(0x20, W83697HF_EFDR); /* select WDTO */ - outb_p(0x07, WDT_EFER); /* point to logical device number reg */ - outb_p(0x08, WDT_EFDR); /* select logical device 8 (GPIO2) */ - outb_p(0x30, WDT_EFER); /* select CR30 */ - outb_p(0x01, WDT_EFDR); /* set bit 0 to activate GPIO2 */ + outb_p(0x07, W83697HF_EFER); /* point to logical device number reg */ + outb_p(0x08, W83697HF_EFDR); /* select logical device 8 (GPIO2) */ + outb_p(0x30, W83697HF_EFER); /* select CR30 */ + outb_p(0x01, W83697HF_EFDR); /* set bit 0 to activate GPIO2 */ } static void w83697hf_unselect_wd_register(void) { - outb_p(0xAA, WDT_EFER); /* Leave extended function mode */ + outb_p(0xAA, W83697HF_EFER); /* Leave extended function mode */ } static void @@ -96,17 +96,17 @@ w83697hf_init(void) w83697hf_select_wd_register(); - outb_p(0xF3, WDT_EFER); /* Select CRF3 */ + outb_p(0xF3, W83697HF_EFER); /* Select CRF3 */ - t=inb_p(WDT_EFDR); /* read CRF3 */ + t=inb_p(W83697HF_EFDR); /* read CRF3 */ if (t != 0) { printk (KERN_INFO PFX "Watchdog already running. Resetting timeout to %d sec\n", timeout); - outb_p(timeout, WDT_EFDR); /* Write back to CRF3 */ + outb_p(timeout, W83697HF_EFDR); /* Write back to CRF3 */ } - outb_p(0xF4, WDT_EFER); /* Select CRF4 */ - t=inb_p(WDT_EFDR); /* read CRF4 */ - t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ - outb_p(t, WDT_EFDR); /* Write back to CRF4 */ + outb_p(0xF4, W83697HF_EFER); /* Select CRF4 */ + t=inb_p(W83697HF_EFDR); /* read CRF4 */ + t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ + outb_p(t, W83697HF_EFDR); /* Write back to CRF4 */ w83697hf_unselect_wd_register(); } @@ -118,8 +118,8 @@ wdt_ctrl(int timeout) w83697hf_select_wd_register(); - outb_p(0xF4, WDT_EFER); /* Select CRF4 */ - outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF4 */ + outb_p(0xF4, W83697HF_EFER); /* Select CRF4 */ + outb_p(timeout, W83697HF_EFDR); /* Write Timeout counter to CRF4 */ w83697hf_unselect_wd_register(); -- cgit v0.10.2 From de710d6871c7f569da007f1074710fadf1708c29 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 6 This is patch 6 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - The driver works for both the w83697hf and the w83697hg chipset's. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index c44f281..ad397f9 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -1,9 +1,10 @@ /* - * w83697hf WDT driver + * w83697hf/hg WDT driver * * (c) Copyright 2006 Marcus Junker * - * Based on w83627hf_wdt.c advantechwdt.c which is based on wdt.c. + * Based on w83627hf_wdt.c which is based on advantechwdt.c + * which is based on wdt.c. * Original copyright messages: * * (c) Copyright 2003 Pádraig Brady @@ -39,7 +40,7 @@ #include #include -#define WATCHDOG_NAME "w83697hf WDT" +#define WATCHDOG_NAME "w83697hf/hg WDT" #define PFX WATCHDOG_NAME ": " #define WATCHDOG_TIMEOUT 60 /* 60 sec default timeout */ @@ -313,7 +314,7 @@ wdt_init(void) spin_lock_init(&io_lock); - printk (KERN_INFO "WDT driver for the Winbond(TM) W83697HF Super I/O chip initialising.\n"); + printk (KERN_INFO PFX "WDT driver for W83697HF/HG initializing\n"); if (wdt_set_heartbeat(timeout)) { wdt_set_heartbeat(WATCHDOG_TIMEOUT); @@ -369,5 +370,5 @@ module_exit(wdt_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marcus Junker "); -MODULE_DESCRIPTION("w83697hf WDT driver"); +MODULE_DESCRIPTION("w83697hf/hg WDT driver"); MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); -- cgit v0.10.2 From f7be3328b6e8b09b3a910a93ef569cba162ea81d Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 7 This is patch 7 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - add w83697hf_unlock function to enter the chipsets extended function mode. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index ad397f9..b1f2257 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -69,11 +69,17 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CON #define W83697HF_EFIR (wdt_io+0) /* Extended Function Index Register (same as EFER) */ #define W83697HF_EFDR (wdt_io+1) /* Extended Function Data Register */ -static void -w83697hf_select_wd_register(void) +static inline void +w83697hf_unlock(void) { outb_p(0x87, W83697HF_EFER); /* Enter extended function mode */ outb_p(0x87, W83697HF_EFER); /* Again according to manual */ +} + +static void +w83697hf_select_wd_register(void) +{ + w83697hf_unlock(); outb_p(0x29, W83697HF_EFER); /* select CR29 */ outb_p(0x20, W83697HF_EFDR); /* select WDTO */ -- cgit v0.10.2 From fe851ebade80af9b58599c74d61718657b02cfd3 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 8 This is patch 8 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - add w83697hf_lock function to leave the chipsets extended function mode. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index b1f2257..6a357a8 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -76,6 +76,12 @@ w83697hf_unlock(void) outb_p(0x87, W83697HF_EFER); /* Again according to manual */ } +static inline void +w83697hf_lock(void) +{ + outb_p(0xAA, W83697HF_EFER); /* Leave extended function mode */ +} + static void w83697hf_select_wd_register(void) { @@ -93,7 +99,7 @@ w83697hf_select_wd_register(void) static void w83697hf_unselect_wd_register(void) { - outb_p(0xAA, W83697HF_EFER); /* Leave extended function mode */ + w83697hf_lock(); } static void -- cgit v0.10.2 From 0cd544763bacad14d0d15fb16d29999b450cb77f Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 9 This is patch 9 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - add w83697hf_get_reg() and w83697hf_set_reg() functions. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 6a357a8..f62f172 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -82,18 +82,34 @@ w83697hf_lock(void) outb_p(0xAA, W83697HF_EFER); /* Leave extended function mode */ } +/* + * The two functions w83697hf_get_reg() and w83697hf_set_reg() + * must be called with the device unlocked. + */ + +static unsigned char +w83697hf_get_reg(unsigned char reg) +{ + outb_p(reg, W83697HF_EFIR); + return inb_p(W83697HF_EFDR); +} + +static void +w83697hf_set_reg(unsigned char reg, unsigned char data) +{ + outb_p(reg, W83697HF_EFIR); + outb_p(data, W83697HF_EFDR); +} + static void w83697hf_select_wd_register(void) { w83697hf_unlock(); - outb_p(0x29, W83697HF_EFER); /* select CR29 */ - outb_p(0x20, W83697HF_EFDR); /* select WDTO */ + w83697hf_set_reg(0x29, 0x20); /* Set pin 119 to WDTO# mode (= CR29, WDT0) */ - outb_p(0x07, W83697HF_EFER); /* point to logical device number reg */ - outb_p(0x08, W83697HF_EFDR); /* select logical device 8 (GPIO2) */ - outb_p(0x30, W83697HF_EFER); /* select CR30 */ - outb_p(0x01, W83697HF_EFDR); /* set bit 0 to activate GPIO2 */ + w83697hf_set_reg(0x07, 0x08); /* Switch to logic device 8 (GPIO2) */ + w83697hf_set_reg(0x30, 0x01); /* Enable timer/activate GPIO2 via bit 0 */ } static void @@ -109,17 +125,14 @@ w83697hf_init(void) w83697hf_select_wd_register(); - outb_p(0xF3, W83697HF_EFER); /* Select CRF3 */ - - t=inb_p(W83697HF_EFDR); /* read CRF3 */ + t = w83697hf_get_reg(0xF3); /* Read CRF3 */ if (t != 0) { printk (KERN_INFO PFX "Watchdog already running. Resetting timeout to %d sec\n", timeout); - outb_p(timeout, W83697HF_EFDR); /* Write back to CRF3 */ + w83697hf_set_reg(0xF3, timeout); /* Write new timeout */ } - outb_p(0xF4, W83697HF_EFER); /* Select CRF4 */ - t=inb_p(W83697HF_EFDR); /* read CRF4 */ + t = w83697hf_get_reg(0xF4); /* Read CRF4 */ t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ - outb_p(t, W83697HF_EFDR); /* Write back to CRF4 */ + w83697hf_set_reg(0xF4, t); /* Write back to CRF4 */ w83697hf_unselect_wd_register(); } @@ -131,8 +144,7 @@ wdt_ctrl(int timeout) w83697hf_select_wd_register(); - outb_p(0xF4, W83697HF_EFER); /* Select CRF4 */ - outb_p(timeout, W83697HF_EFDR); /* Write Timeout counter to CRF4 */ + w83697hf_set_reg(0xF4, timeout); /* Write Timeout counter to CRF4 */ w83697hf_unselect_wd_register(); -- cgit v0.10.2 From c81b2996253a94278057f83a24dfa9053f0dee7a Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 10 This is patch 10 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - check whether the device is really present (we *can* probe for the device now). Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index f62f172..4e0bd4e 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -49,9 +49,9 @@ static char expect_close; static spinlock_t io_lock; /* You must set this - there is no sane way to probe for this board. */ -static int wdt_io = 0x2E; +static int wdt_io = 0x2e; module_param(wdt_io, int, 0); -MODULE_PARM_DESC(wdt_io, "w83697hf WDT io port (default 0x2E)"); +MODULE_PARM_DESC(wdt_io, "w83697hf/hg WDT io port (default 0x2e, 0 = autodetect)"); static int timeout = WATCHDOG_TIMEOUT; /* in seconds */ module_param(timeout, int, 0); @@ -331,28 +331,62 @@ static struct notifier_block wdt_notifier = { .notifier_call = wdt_notify_sys, }; +static int +w83697hf_check_wdt(void) +{ + if (!request_region(wdt_io, 2, WATCHDOG_NAME)) { + printk (KERN_ERR PFX "I/O address 0x%x already in use\n", wdt_io); + return -EIO; + } + + printk (KERN_DEBUG PFX "Looking for watchdog at address 0x%x\n", wdt_io); + w83697hf_unlock(); + if (w83697hf_get_reg(0x20) == 0x60) { + printk (KERN_INFO PFX "watchdog found at address 0x%x\n", wdt_io); + w83697hf_lock(); + return 0; + } + w83697hf_lock(); /* Reprotect in case it was a compatible device */ + + printk (KERN_INFO PFX "watchdog not found at address 0x%x\n", wdt_io); + release_region(wdt_io, 2); + return -EIO; +} + static int __init wdt_init(void) { - int ret; + int ret, autodetect; spin_lock_init(&io_lock); printk (KERN_INFO PFX "WDT driver for W83697HF/HG initializing\n"); + autodetect = wdt_io == 0; + if (autodetect) + wdt_io = 0x2e; + + if (!w83697hf_check_wdt()) + goto found; + + if (autodetect) { + wdt_io = 0x4e; + if (!w83697hf_check_wdt()) + goto found; + } + + printk (KERN_ERR PFX "No W83697HF/HG could be found\n"); + ret = -EIO; + goto out; + +found: + if (wdt_set_heartbeat(timeout)) { wdt_set_heartbeat(WATCHDOG_TIMEOUT); printk (KERN_INFO PFX "timeout value must be 1<=timeout<=255, using %d\n", WATCHDOG_TIMEOUT); } - if (!request_region(wdt_io, 2, WATCHDOG_NAME)) { - printk (KERN_ERR PFX "I/O address 0x%04x already in use\n", - wdt_io); - ret = -EIO; - goto out; - } - w83697hf_init(); ret = register_reboot_notifier(&wdt_notifier); -- cgit v0.10.2 From a7933e05d46f49385841d09028ee07fae2b383f2 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 11 This is patch 11 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Add w83697hf_select_wdt() and w83697hf_deselect_wdt() so that the start/stop/ping code can directly talk to the watchdog. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 4e0bd4e..b12f8b8 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -102,6 +102,19 @@ w83697hf_set_reg(unsigned char reg, unsigned char data) } static void +w83697hf_select_wdt(void) +{ + w83697hf_unlock(); + w83697hf_set_reg(0x07, 0x08); /* Switch to logic device 8 (GPIO2) */ +} + +static inline void +w83697hf_deselect_wdt(void) +{ + w83697hf_lock(); +} + +static void w83697hf_select_wd_register(void) { w83697hf_unlock(); @@ -142,11 +155,11 @@ wdt_ctrl(int timeout) { spin_lock(&io_lock); - w83697hf_select_wd_register(); + w83697hf_select_wdt(); w83697hf_set_reg(0xF4, timeout); /* Write Timeout counter to CRF4 */ - w83697hf_unselect_wd_register(); + w83697hf_deselect_wdt(); spin_unlock(&io_lock); } -- cgit v0.10.2 From d46ab596e251e35a7e27c95e4e4d01921f3e579e Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 12 This is patch 12 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Add w83697hf_write_timeout() to set the watchdog's timeout value. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index b12f8b8..b3dcc81 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -83,8 +83,8 @@ w83697hf_lock(void) } /* - * The two functions w83697hf_get_reg() and w83697hf_set_reg() - * must be called with the device unlocked. + * The three functions w83697hf_get_reg(), w83697hf_set_reg() and + * w83697hf_write_timeout() must be called with the device unlocked. */ static unsigned char @@ -102,6 +102,12 @@ w83697hf_set_reg(unsigned char reg, unsigned char data) } static void +w83697hf_write_timeout(int timeout) +{ + w83697hf_set_reg(0xF4, timeout); /* Write Timeout counter to CRF4 */ +} + +static void w83697hf_select_wdt(void) { w83697hf_unlock(); @@ -157,7 +163,7 @@ wdt_ctrl(int timeout) w83697hf_select_wdt(); - w83697hf_set_reg(0xF4, timeout); /* Write Timeout counter to CRF4 */ + w83697hf_write_timeout(timeout); w83697hf_deselect_wdt(); -- cgit v0.10.2 From 089d8139f4c19c2f4d6984323e9d8a6e77cc92f7 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 13 This is patch 13 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Remove wdt_ctrl (it has been replaced with the w83697hf_write_timeout() function) and redo/clean-up the start/stop/ping code. - Make sure that the watchdog is enabled or disabled When starting or stoping the device (with a call to w83697hf_set_reg(0x30, ?); ). Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index b3dcc81..2b3ce43 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -156,31 +156,44 @@ w83697hf_init(void) w83697hf_unselect_wd_register(); } -static void -wdt_ctrl(int timeout) +static int +wdt_ping(void) { spin_lock(&io_lock); - w83697hf_select_wdt(); w83697hf_write_timeout(timeout); w83697hf_deselect_wdt(); - spin_unlock(&io_lock); + return 0; } static int -wdt_ping(void) +wdt_enable(void) { - wdt_ctrl(timeout); + spin_lock(&io_lock); + w83697hf_select_wdt(); + + w83697hf_write_timeout(timeout); + w83697hf_set_reg(0x30, 1); /* Enable timer */ + + w83697hf_deselect_wdt(); + spin_unlock(&io_lock); return 0; } static int wdt_disable(void) { - wdt_ctrl(0); + spin_lock(&io_lock); + w83697hf_select_wdt(); + + w83697hf_set_reg(0x30, 0); /* Disable timer */ + w83697hf_write_timeout(0); + + w83697hf_deselect_wdt(); + spin_unlock(&io_lock); return 0; } @@ -267,7 +280,7 @@ wdt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, } if (options & WDIOS_ENABLECARD) { - wdt_ping(); + wdt_enable(); retval = 0; } @@ -289,7 +302,7 @@ wdt_open(struct inode *inode, struct file *file) * Activate */ - wdt_ping(); + wdt_enable(); return nonseekable_open(inode, file); } -- cgit v0.10.2 From fa69afd3c224252890cb30864dc648d1399dd9fe Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 14 This is patch 14 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Clean-up initialization code (part 1: remove w83697hf_select_wd_register() and w83697hf_unselect_wd_register() functions). - Make sure that the watchdog device is stopped as soon as we found it. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 2b3ce43..1ea43bf 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -121,28 +121,13 @@ w83697hf_deselect_wdt(void) } static void -w83697hf_select_wd_register(void) -{ - w83697hf_unlock(); - - w83697hf_set_reg(0x29, 0x20); /* Set pin 119 to WDTO# mode (= CR29, WDT0) */ - - w83697hf_set_reg(0x07, 0x08); /* Switch to logic device 8 (GPIO2) */ - w83697hf_set_reg(0x30, 0x01); /* Enable timer/activate GPIO2 via bit 0 */ -} - -static void -w83697hf_unselect_wd_register(void) -{ - w83697hf_lock(); -} - -static void w83697hf_init(void) { unsigned char t; - w83697hf_select_wd_register(); + w83697hf_select_wdt(); + + w83697hf_set_reg(0x29, 0x20); /* Set pin 119 to WDTO# mode (= CR29, WDT0) */ t = w83697hf_get_reg(0xF3); /* Read CRF3 */ if (t != 0) { @@ -153,7 +138,7 @@ w83697hf_init(void) t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ w83697hf_set_reg(0xF4, t); /* Write back to CRF4 */ - w83697hf_unselect_wd_register(); + w83697hf_deselect_wdt(); } static int @@ -412,6 +397,8 @@ wdt_init(void) goto out; found: + w83697hf_init(); + wdt_disable(); /* Disable watchdog until first use */ if (wdt_set_heartbeat(timeout)) { wdt_set_heartbeat(WATCHDOG_TIMEOUT); @@ -419,8 +406,6 @@ found: WATCHDOG_TIMEOUT); } - w83697hf_init(); - ret = register_reboot_notifier(&wdt_notifier); if (ret != 0) { printk (KERN_ERR PFX "cannot register reboot notifier (err=%d)\n", -- cgit v0.10.2 From b7b9868ba6f528d60e5869b4a6aad1fe49838b03 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 15 This is patch 15 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Clean-up initialization code - part 2: * the line reading "set second mode & disable keyboard ..." is plain wrong, the register being manipulated (CRF4) is the counter itself, not the control byte (CRF3) -- looks like it has been copied from another driver. * I think garbage is being written in CRF3 (the control word) as the timeout value is being stored in this register (such as 60 for 60 seconds). * We only want to set pin 119 to WDTO# mode and leave the rest of CR29 like it is. * Set count mode to seconds and not minutes. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 1ea43bf..78b6540 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -123,20 +123,18 @@ w83697hf_deselect_wdt(void) static void w83697hf_init(void) { - unsigned char t; + unsigned char bbuf; w83697hf_select_wdt(); - w83697hf_set_reg(0x29, 0x20); /* Set pin 119 to WDTO# mode (= CR29, WDT0) */ + bbuf = w83697hf_get_reg(0x29); + bbuf &= ~0x60; + bbuf |= 0x20; + w83697hf_set_reg(0x29, bbuf); /* Set pin 119 to WDTO# mode (= CR29, WDT0) */ - t = w83697hf_get_reg(0xF3); /* Read CRF3 */ - if (t != 0) { - printk (KERN_INFO PFX "Watchdog already running. Resetting timeout to %d sec\n", timeout); - w83697hf_set_reg(0xF3, timeout); /* Write new timeout */ - } - t = w83697hf_get_reg(0xF4); /* Read CRF4 */ - t&=~0x0C; /* set second mode & disable keyboard turning off watchdog */ - w83697hf_set_reg(0xF4, t); /* Write back to CRF4 */ + bbuf = w83697hf_get_reg(0xF3); + bbuf &= ~0x04; + w83697hf_set_reg(0xF3, bbuf); /* Count mode is seconds */ w83697hf_deselect_wdt(); } -- cgit v0.10.2 From 3fdee8db010d5cbf890ec49332ac4946f3f63720 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - patch 16 This is patch 16 in the series of patches that converts Marcus Junker's w83697hf watchdog driver to Samuel Tardieau's w83697hf/hg watchdog driver. This patch contains following changes: - Add copyright notice for Samuel Tardieu also. This is the last patch in this series. The original description for Samuel's driver was: driver for the Winbond W83697HF/W83697HG watchdog timer The Winbond SuperIO W83697HF/HG includes a watchdog that can count from 1 to 255 seconds (or minutes). This drivers allows the seconds mode to be used. It exposes a standard /dev/watchdog interface. This chip is currently being used on some motherboards designed by VIA. By default, the module looks for a chip at I/O port 0x2e. The chip can be configured to be at 0x4e on some motherboards, the address can be chosen using the wdt_io module parameter. Using 0 will try to autodetect the address. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 78b6540..21052de 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -1,6 +1,7 @@ /* * w83697hf/hg WDT driver * + * (c) Copyright 2006 Samuel Tardieu * (c) Copyright 2006 Marcus Junker * * Based on w83627hf_wdt.c which is based on advantechwdt.c @@ -442,6 +443,6 @@ module_init(wdt_init); module_exit(wdt_exit); MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Marcus Junker "); +MODULE_AUTHOR("Marcus Junker , Samuel Tardieu "); MODULE_DESCRIPTION("w83697hf/hg WDT driver"); MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); -- cgit v0.10.2 From e223f01a822e999b0aea2e720e12d8bb3532da70 Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Fri, 15 Sep 2006 17:59:07 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - autodetect patch Change the autodetect code so that it is more generic. Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/w83697hf_wdt.c b/drivers/char/watchdog/w83697hf_wdt.c index 21052de..7768b55 100644 --- a/drivers/char/watchdog/w83697hf_wdt.c +++ b/drivers/char/watchdog/w83697hf_wdt.c @@ -369,33 +369,35 @@ w83697hf_check_wdt(void) return -EIO; } +static int w83697hf_ioports[] = { 0x2e, 0x4e, 0x00 }; + static int __init wdt_init(void) { - int ret, autodetect; + int ret, i, found = 0; spin_lock_init(&io_lock); printk (KERN_INFO PFX "WDT driver for W83697HF/HG initializing\n"); - autodetect = wdt_io == 0; - if (autodetect) - wdt_io = 0x2e; - - if (!w83697hf_check_wdt()) - goto found; - - if (autodetect) { - wdt_io = 0x4e; + if (wdt_io == 0) { + /* we will autodetect the W83697HF/HG watchdog */ + for (i = 0; ((!found) && (w83697hf_ioports[i] != 0)); i++) { + wdt_io = w83697hf_ioports[i]; + if (!w83697hf_check_wdt()) + found++; + } + } else { if (!w83697hf_check_wdt()) - goto found; + found++; } - printk (KERN_ERR PFX "No W83697HF/HG could be found\n"); - ret = -EIO; - goto out; + if (!found) { + printk (KERN_ERR PFX "No W83697HF/HG could be found\n"); + ret = -EIO; + goto out; + } -found: w83697hf_init(); wdt_disable(); /* Disable watchdog until first use */ -- cgit v0.10.2 From ff02cfc76a5040ee125c597baa1cfc9874918ed2 Mon Sep 17 00:00:00 2001 From: Samuel Tardieu Date: Thu, 7 Sep 2006 11:57:00 +0200 Subject: [WATCHDOG] w83697hf/hg WDT driver - Kconfig patch Update Kconfig for the w83697hf/hg watchdog driver. Signed-off-by: Samuel Tardieu Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index ecae59c..847a260 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -431,12 +431,14 @@ config W83627HF_WDT Most people will say N. config W83697HF_WDT - tristate "W83697HF Watchdog Timer" + tristate "W83697HF/W83697HG Watchdog Timer" depends on WATCHDOG && X86 ---help--- - This is the driver for the hardware watchdog on the W83697HF chipset - This watchdog simply watches your kernel to make sure it doesn't freeze, - and if it does, it reboots your computer after a certain amount of time. + This is the driver for the hardware watchdog on the W83697HF/HG + chipset as used in Dedibox/VIA motherboards (and likely others). + This watchdog simply watches your kernel to make sure it doesn't + freeze, and if it does, it reboots your computer after a certain + amount of time. To compile this driver as a module, choose M here: the module will be called w83697hf_wdt. -- cgit v0.10.2 From 489b10c1f63fafcb89c330a0603694652068132a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 3 Oct 2006 16:39:12 -0700 Subject: [PATCH] sky2: incorrect length on receive packets The previous change to do fragmented receive (post 2.6.18) introduced a bug where packets are passed up with size set to the size of the receive buffer not the actual received data. IP silently trims this so it didn't show up right away. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 396e7df..6851515 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -1907,7 +1907,7 @@ static struct sk_buff *receive_copy(struct sky2_port *sky2, pci_dma_sync_single_for_device(sky2->hw->pdev, re->data_addr, length, PCI_DMA_FROMDEVICE); re->skb->ip_summed = CHECKSUM_NONE; - __skb_put(skb, length); + skb_put(skb, length); } return skb; } @@ -1970,7 +1970,7 @@ static struct sk_buff *receive_new(struct sky2_port *sky2, if (skb_shinfo(skb)->nr_frags) skb_put_frags(skb, hdr_space, length); else - skb_put(skb, hdr_space); + skb_put(skb, length); return skb; } -- cgit v0.10.2 From bbedefccc6b0da43cfaf785dac89c88bc59cb6ed Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 3 Oct 2006 12:24:23 -0500 Subject: [PATCH] ibmveth: Harden driver initilisation This patch has been floating around for a while now, Santi originally sent it in March: http://www.spinics.net/lists/netdev/msg00471.html After a kexec the ibmveth driver will fail when trying to register with the Hypervisor because the previous kernel has not unregistered. So if the registration fails, we unregister and then try again. We don't unconditionally unregister, because we don't want to disturb the regular code path for 99% of users. Signed-off-by: Michael Ellerman Acked-by: Anton Blanchard Signed-off-by: Santiago Leon Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 767203d..df3a59e 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -437,6 +437,31 @@ static void ibmveth_cleanup(struct ibmveth_adapter *adapter) &adapter->rx_buff_pool[i]); } +static int ibmveth_register_logical_lan(struct ibmveth_adapter *adapter, + union ibmveth_buf_desc rxq_desc, u64 mac_address) +{ + int rc, try_again = 1; + + /* After a kexec the adapter will still be open, so our attempt to + * open it will fail. So if we get a failure we free the adapter and + * try again, but only once. */ +retry: + rc = h_register_logical_lan(adapter->vdev->unit_address, + adapter->buffer_list_dma, rxq_desc.desc, + adapter->filter_list_dma, mac_address); + + if (rc != H_SUCCESS && try_again) { + do { + rc = h_free_logical_lan(adapter->vdev->unit_address); + } while (H_IS_LONG_BUSY(rc) || (rc == H_BUSY)); + + try_again = 0; + goto retry; + } + + return rc; +} + static int ibmveth_open(struct net_device *netdev) { struct ibmveth_adapter *adapter = netdev->priv; @@ -502,12 +527,7 @@ static int ibmveth_open(struct net_device *netdev) ibmveth_debug_printk("filter list @ 0x%p\n", adapter->filter_list_addr); ibmveth_debug_printk("receive q @ 0x%p\n", adapter->rx_queue.queue_addr); - - lpar_rc = h_register_logical_lan(adapter->vdev->unit_address, - adapter->buffer_list_dma, - rxq_desc.desc, - adapter->filter_list_dma, - mac_address); + lpar_rc = ibmveth_register_logical_lan(adapter, rxq_desc, mac_address); if(lpar_rc != H_SUCCESS) { ibmveth_error_printk("h_register_logical_lan failed with %ld\n", lpar_rc); -- cgit v0.10.2 From 6b4223748895ed5b200c8049231567ea399fc0c2 Mon Sep 17 00:00:00 2001 From: Santiago Leon Date: Tue, 3 Oct 2006 12:24:28 -0500 Subject: [PATCH] ibmveth: Add netpoll function This patch adds the net poll controller function to ibmveth to support netconsole and netdump. Signed-off-by: Santiago Leon Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index df3a59e..bd79d67 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -925,6 +925,14 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu) return -EINVAL; } +#ifdef CONFIG_NET_POLL_CONTROLLER +static void ibmveth_poll_controller(struct net_device *dev) +{ + ibmveth_replenish_task(dev->priv); + ibmveth_interrupt(dev->irq, dev, NULL); +} +#endif + static int __devinit ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id) { int rc, i; @@ -997,6 +1005,9 @@ static int __devinit ibmveth_probe(struct vio_dev *dev, const struct vio_device_ netdev->ethtool_ops = &netdev_ethtool_ops; netdev->change_mtu = ibmveth_change_mtu; SET_NETDEV_DEV(netdev, &dev->dev); +#ifdef CONFIG_NET_POLL_CONTROLLER + netdev->poll_controller = ibmveth_poll_controller; +#endif netdev->features |= NETIF_F_LLTX; spin_lock_init(&adapter->stats_lock); -- cgit v0.10.2 From 4347ef15f76dca33ae8da769d6900a468253bda2 Mon Sep 17 00:00:00 2001 From: Santiago Leon Date: Tue, 3 Oct 2006 12:24:34 -0500 Subject: [PATCH] ibmveth: kdump interrupt fix This patch fixes a race that panics the kernel when opening the device after a kdump. Without this patch there is a window where the hypervisor can send an interrupt before all the structures for the kdump ibmveth module are ready (because the hypervisor is not aware that the partition crashed and that the virtual driver is reloading). We close this window by disabling the interrupts before registering the adapter to the hypervisor. This patch depends on the "ibmveth: Harden driver initilisation" patch. Signed-off-by: Santiago Leon Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index bd79d67..2ec49d2 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -527,6 +527,8 @@ static int ibmveth_open(struct net_device *netdev) ibmveth_debug_printk("filter list @ 0x%p\n", adapter->filter_list_addr); ibmveth_debug_printk("receive q @ 0x%p\n", adapter->rx_queue.queue_addr); + h_vio_signal(adapter->vdev->unit_address, VIO_IRQ_DISABLE); + lpar_rc = ibmveth_register_logical_lan(adapter, rxq_desc, mac_address); if(lpar_rc != H_SUCCESS) { -- cgit v0.10.2 From 03a85d0907b2455c772b8fb179b0c07a66b00ddb Mon Sep 17 00:00:00 2001 From: Santiago Leon Date: Tue, 3 Oct 2006 12:24:39 -0500 Subject: [PATCH] ibmveth: rename proc entry name This patch changes the name of the proc file for each ibmveth adapter from the network device name to the slot number in the virtual bus. The proc file is created when the device is probed, so a change in the name of the device will not be reflected in the name of the proc file giving problems when identifying and removing the adapter. The slot number is a property that does not change through the life of the adapter so we use that instead. Signed-off-by: Santiago Leon Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 2ec49d2..6aff2bc 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -1165,7 +1165,9 @@ static void ibmveth_proc_register_adapter(struct ibmveth_adapter *adapter) { struct proc_dir_entry *entry; if (ibmveth_proc_dir) { - entry = create_proc_entry(adapter->netdev->name, S_IFREG, ibmveth_proc_dir); + char u_addr[10]; + sprintf(u_addr, "%x", adapter->vdev->unit_address); + entry = create_proc_entry(u_addr, S_IFREG, ibmveth_proc_dir); if (!entry) { ibmveth_error_printk("Cannot create adapter proc entry"); } else { @@ -1180,7 +1182,9 @@ static void ibmveth_proc_register_adapter(struct ibmveth_adapter *adapter) static void ibmveth_proc_unregister_adapter(struct ibmveth_adapter *adapter) { if (ibmveth_proc_dir) { - remove_proc_entry(adapter->netdev->name, ibmveth_proc_dir); + char u_addr[10]; + sprintf(u_addr, "%x", adapter->vdev->unit_address); + remove_proc_entry(u_addr, ibmveth_proc_dir); } } -- cgit v0.10.2 From 751ae21c6cd1493e3d0a4935b08fb298b9d89773 Mon Sep 17 00:00:00 2001 From: Santiago Leon Date: Tue, 3 Oct 2006 12:24:45 -0500 Subject: [PATCH] ibmveth: fix int rollover panic This patch fixes a nasty bug that has been sitting there since the very first versions of the driver, but is generating a panic because we changed the number of 2K buffers for 2.6.16. The consumer_index and producer_index are u32's that get incremented on every buffer emptied and replenished respectively. We use the {producer,consumer}_index mod'ed with the size of the pool to pick out an entry in the free_map. The problem happens when the u32 rolls over and the number of the buffers in the pool is not a perfect divisor of 2^32. i.e. if the number of 2K buffers is 0x300, before the consumer_index rolls over, our index to the free map = 0xffffffff mod 0x300 = 0xff. The next time a buffer is emptied, we want the index to the free map to be 0x100, but 0x0 mod 0x300 is 0x0. This patch assigns the mod'ed result back to the consumer and producer indexes so that they never roll over. The second chunk of the patch covers the unlikely case where the consumer_index has just been reset to 0x0 and the hypervisor is not able to accept that buffer. Signed-off-by: Santiago Leon Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 6aff2bc..16f3faa 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -213,6 +213,7 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc } free_index = pool->consumer_index++ % pool->size; + pool->consumer_index = free_index; index = pool->free_map[free_index]; ibmveth_assert(index != IBM_VETH_INVALID_MAP); @@ -238,7 +239,10 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc if(lpar_rc != H_SUCCESS) { pool->free_map[free_index] = index; pool->skbuff[index] = NULL; - pool->consumer_index--; + if (pool->consumer_index == 0) + pool->consumer_index = pool->size - 1; + else + pool->consumer_index--; dma_unmap_single(&adapter->vdev->dev, pool->dma_addr[index], pool->buff_size, DMA_FROM_DEVICE); @@ -326,6 +330,7 @@ static void ibmveth_remove_buffer_from_pool(struct ibmveth_adapter *adapter, u64 DMA_FROM_DEVICE); free_index = adapter->rx_buff_pool[pool].producer_index++ % adapter->rx_buff_pool[pool].size; + adapter->rx_buff_pool[pool].producer_index = free_index; adapter->rx_buff_pool[pool].free_map[free_index] = index; mb(); -- cgit v0.10.2 From 70fbf32703a9c4d3403663d1fc24fd8afc76d56f Mon Sep 17 00:00:00 2001 From: Maxime Bizon Date: Tue, 3 Oct 2006 10:27:10 -0700 Subject: [PATCH] mv643xx_eth: Fix ethtool stats Some stats reported by ethtool -S on mv643xx_eth device are cleared between each call. This patch fixes it. Signed-off-by: Maxime Bizon Signed-off-by: Dale Farnsworth Signed-off-by: Jeff Garzik diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index 7f8e5ad1..eccedf3 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -2156,7 +2156,7 @@ static void eth_update_mib_counters(struct mv643xx_private *mp) for (offset = ETH_MIB_BAD_OCTETS_RECEIVED; offset <= ETH_MIB_FRAMES_1024_TO_MAX_OCTETS; offset += 4) - *(u32 *)((char *)p + offset) = read_mib(mp, offset); + *(u32 *)((char *)p + offset) += read_mib(mp, offset); p->good_octets_sent += read_mib(mp, ETH_MIB_GOOD_OCTETS_SENT_LOW); p->good_octets_sent += @@ -2165,7 +2165,7 @@ static void eth_update_mib_counters(struct mv643xx_private *mp) for (offset = ETH_MIB_GOOD_FRAMES_SENT; offset <= ETH_MIB_LATE_COLLISION; offset += 4) - *(u32 *)((char *)p + offset) = read_mib(mp, offset); + *(u32 *)((char *)p + offset) += read_mib(mp, offset); } /* -- cgit v0.10.2 From 0a07bc645e818b88559d99f52ad45e35352e8228 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 19 Sep 2006 14:55:22 +0200 Subject: [PATCH] forcedeth: hardirq lockdep warning BUG: warning at kernel/lockdep.c:1816/trace_hardirqs_on() (Not tainted) Call Trace: show_trace dump_stack trace_hardirqs_on :forcedeth:nv_nic_irq_other handle_IRQ_event __do_IRQ do_IRQ ret_from_intr DWARF2 barf default_idle cpu_idle rest_init start_kernel _sinittext These 3 functions nv_nic_irq_tx(), nv_nic_irq_rx() and nv_nic_irq_other() are reachable from IRQ context and process context. Make use of the irq-save/restore spinlock variant. (Compile tested only, since I do not have the hardware) Signed-off-by: Peter Zijlstra Cc: Jeff Garzik Cc: Ingo Molnar Cc: Arjan van de Ven Cc: Dave Jones Cc: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c index eea1d66..35467e0 100644 --- a/drivers/net/forcedeth.c +++ b/drivers/net/forcedeth.c @@ -2497,6 +2497,7 @@ static irqreturn_t nv_nic_irq_tx(int foo, void *data, struct pt_regs *regs) u8 __iomem *base = get_hwbase(dev); u32 events; int i; + unsigned long flags; dprintk(KERN_DEBUG "%s: nv_nic_irq_tx\n", dev->name); @@ -2508,16 +2509,16 @@ static irqreturn_t nv_nic_irq_tx(int foo, void *data, struct pt_regs *regs) if (!(events & np->irqmask)) break; - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); nv_tx_done(dev); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); if (events & (NVREG_IRQ_TX_ERR)) { dprintk(KERN_DEBUG "%s: received irq with events 0x%x. Probably TX fail.\n", dev->name, events); } if (i > max_interrupt_work) { - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); /* disable interrupts on the nic */ writel(NVREG_IRQ_TX_ALL, base + NvRegIrqMask); pci_push(base); @@ -2527,7 +2528,7 @@ static irqreturn_t nv_nic_irq_tx(int foo, void *data, struct pt_regs *regs) mod_timer(&np->nic_poll, jiffies + POLL_WAIT); } printk(KERN_DEBUG "%s: too many iterations (%d) in nv_nic_irq_tx.\n", dev->name, i); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); break; } @@ -2601,6 +2602,7 @@ static irqreturn_t nv_nic_irq_rx(int foo, void *data, struct pt_regs *regs) u8 __iomem *base = get_hwbase(dev); u32 events; int i; + unsigned long flags; dprintk(KERN_DEBUG "%s: nv_nic_irq_rx\n", dev->name); @@ -2614,14 +2616,14 @@ static irqreturn_t nv_nic_irq_rx(int foo, void *data, struct pt_regs *regs) nv_rx_process(dev, dev->weight); if (nv_alloc_rx(dev)) { - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); if (!np->in_shutdown) mod_timer(&np->oom_kick, jiffies + OOM_REFILL); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); } if (i > max_interrupt_work) { - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); /* disable interrupts on the nic */ writel(NVREG_IRQ_RX_ALL, base + NvRegIrqMask); pci_push(base); @@ -2631,7 +2633,7 @@ static irqreturn_t nv_nic_irq_rx(int foo, void *data, struct pt_regs *regs) mod_timer(&np->nic_poll, jiffies + POLL_WAIT); } printk(KERN_DEBUG "%s: too many iterations (%d) in nv_nic_irq_rx.\n", dev->name, i); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); break; } } @@ -2648,6 +2650,7 @@ static irqreturn_t nv_nic_irq_other(int foo, void *data, struct pt_regs *regs) u8 __iomem *base = get_hwbase(dev); u32 events; int i; + unsigned long flags; dprintk(KERN_DEBUG "%s: nv_nic_irq_other\n", dev->name); @@ -2660,14 +2663,14 @@ static irqreturn_t nv_nic_irq_other(int foo, void *data, struct pt_regs *regs) break; if (events & NVREG_IRQ_LINK) { - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); nv_link_irq(dev); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); } if (np->need_linktimer && time_after(jiffies, np->link_timeout)) { - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); nv_linkchange(dev); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); np->link_timeout = jiffies + LINK_TIMEOUT; } if (events & (NVREG_IRQ_UNKNOWN)) { @@ -2675,7 +2678,7 @@ static irqreturn_t nv_nic_irq_other(int foo, void *data, struct pt_regs *regs) dev->name, events); } if (i > max_interrupt_work) { - spin_lock_irq(&np->lock); + spin_lock_irqsave(&np->lock, flags); /* disable interrupts on the nic */ writel(NVREG_IRQ_OTHER, base + NvRegIrqMask); pci_push(base); @@ -2685,7 +2688,7 @@ static irqreturn_t nv_nic_irq_other(int foo, void *data, struct pt_regs *regs) mod_timer(&np->nic_poll, jiffies + POLL_WAIT); } printk(KERN_DEBUG "%s: too many iterations (%d) in nv_nic_irq_other.\n", dev->name, i); - spin_unlock_irq(&np->lock); + spin_unlock_irqrestore(&np->lock, flags); break; } -- cgit v0.10.2 From 46767aeba58ca9357a2309765201bad38d8f5e9b Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 29 Sep 2006 18:26:47 +0100 Subject: [PATCH] libata: Don't believe bogus claims in the older PIO mode register Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index dce6565..ad8e2c6 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -870,7 +870,11 @@ static unsigned int ata_id_xfermask(const u16 *id) * the PIO timing number for the maximum. Turn it into * a mask. */ - pio_mask = (2 << (id[ATA_ID_OLD_PIO_MODES] & 0xFF)) - 1 ; + u8 mode = id[ATA_ID_OLD_PIO_MODES] & 0xFF; + if (mode < 5) /* Valid PIO range */ + pio_mask = (2 << mode) - 1; + else + pio_mask = 1; /* But wait.. there's more. Design your standards by * committee and you too can get a free iordy field to -- cgit v0.10.2 From 39984a9fad0c642182f426d7771332d46f222103 Mon Sep 17 00:00:00 2001 From: Karsten Keil Date: Fri, 29 Sep 2006 23:28:42 -0700 Subject: [PATCH] bonding: fix deadlock on high loads in bond_alb_monitor() In bond_alb_monitor the bond->curr_slave_lock write lock is taken and then dev_set_promiscuity maybe called which can take some time, depending on the network HW. If a network IRQ for this card come in the softirq handler maybe try to deliver more packets which end up in a request to the read lock of bond->curr_slave_lock -> deadlock. This issue was found by a test lab during network stress tests, this patch disable the softirq handler for this case and solved the issue. Signed-off-by: Karsten Keil Acked-by: Jay Vosburgh Signed-off-by: Jeff Garzik diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index e83bc82..3292316 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -1433,7 +1433,7 @@ void bond_alb_monitor(struct bonding *bond) * write lock to protect from other code that also * sets the promiscuity. */ - write_lock(&bond->curr_slave_lock); + write_lock_bh(&bond->curr_slave_lock); if (bond_info->primary_is_promisc && (++bond_info->rlb_promisc_timeout_counter >= RLB_PROMISC_TIMEOUT)) { @@ -1448,7 +1448,7 @@ void bond_alb_monitor(struct bonding *bond) bond_info->primary_is_promisc = 0; } - write_unlock(&bond->curr_slave_lock); + write_unlock_bh(&bond->curr_slave_lock); if (bond_info->rlb_rebalance) { bond_info->rlb_rebalance = 0; -- cgit v0.10.2 From 2f614fe04f4463ff22234133319067d7361f54e5 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Thu, 5 Oct 2006 07:10:38 -0400 Subject: [netdrvr] b44: handle excessive multicast groups If there are more than B44_MCAST_TABLE_SIZE groups in the dev->mc_list, it will only listen to the first B44_MCAST_TABLE_SIZE that it sees. This change makes the driver go into RXCONFIG_ALLMULTI mode if there are more than B44_MCAST_TABLE_SIZE groups being subscribed to, similar to other network drivers. Noticed by Bill Helfinstine Signed-off-by: Jeff Garzik diff --git a/drivers/net/b44.c b/drivers/net/b44.c index e891ea2..973b8eb 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -1706,14 +1706,15 @@ static void __b44_set_rx_mode(struct net_device *dev) __b44_set_mac_addr(bp); - if (dev->flags & IFF_ALLMULTI) + if ((dev->flags & IFF_ALLMULTI) || + (dev->mc_count > B44_MCAST_TABLE_SIZE)) val |= RXCONFIG_ALLMULTI; else i = __b44_load_mcast(bp, dev); - for (; i < 64; i++) { + for (; i < 64; i++) __b44_cam_write(bp, zero, i); - } + bw32(bp, B44_RXCONFIG, val); val = br32(bp, B44_CAM_CTRL); bw32(bp, B44_CAM_CTRL, val | CAM_CTRL_ENABLE); -- cgit v0.10.2 From 458821452642fd5dc2377b73cd1323fd4a9653e7 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Wed, 4 Oct 2006 13:21:45 +0900 Subject: sh: First step at generic timeofday support. At the moment we wrap GENERIC_TIME around our existing timer API. As boards start providing their own clocksources, they're able to select GENERIC_TIME accordingly and optimize out most of the timer API. Once the current timers have been reworked as proper clocksource drivers, the rest of the place holders for the timer API can go away and we can flip on GENERIC_TIME unconditionally. Signed-off-by: Paul Mundt diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index f6a0c44..6a461d4 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -45,6 +45,9 @@ config GENERIC_CALIBRATE_DELAY config GENERIC_IOMAP bool +config GENERIC_TIME + def_bool n + config ARCH_MAY_HAVE_PC_FDC bool @@ -357,6 +360,7 @@ config CPU_HAS_SR_RB endmenu menu "Timer support" +depends on !GENERIC_TIME config SH_TMU bool "TMU timer support" diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c index 450c68f..1fbb83c 100644 --- a/arch/sh/kernel/time.c +++ b/arch/sh/kernel/time.c @@ -47,6 +47,7 @@ unsigned long long __attribute__ ((weak)) sched_clock(void) return (unsigned long long)jiffies * (1000000000 / HZ); } +#ifndef CONFIG_GENERIC_TIME void do_gettimeofday(struct timeval *tv) { unsigned long seq; @@ -99,6 +100,7 @@ int do_settimeofday(struct timespec *tv) return 0; } EXPORT_SYMBOL(do_settimeofday); +#endif /* !CONFIG_GENERIC_TIME */ /* last time the RTC clock got updated */ static long last_rtc_update; diff --git a/include/asm-sh/timer.h b/include/asm-sh/timer.h index c7ab280..ebc78db 100644 --- a/include/asm-sh/timer.h +++ b/include/asm-sh/timer.h @@ -8,7 +8,9 @@ struct sys_timer_ops { int (*init)(void); int (*start)(void); int (*stop)(void); +#ifndef CONFIG_GENERIC_TIME unsigned long (*get_offset)(void); +#endif unsigned long (*get_frequency)(void); }; @@ -24,10 +26,12 @@ struct sys_timer { extern struct sys_timer tmu_timer; extern struct sys_timer *sys_timer; +#ifndef CONFIG_GENERIC_TIME static inline unsigned long get_timer_offset(void) { return sys_timer->ops->get_offset(); } +#endif static inline unsigned long get_timer_frequency(void) { -- cgit v0.10.2 From a700f3594d63a85af196ac64984f7375d903afad Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Wed, 4 Oct 2006 13:27:32 +0900 Subject: sh: Kill off timer_ops get_frequency(). We're not using this anywhere these days, kill it off. Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c index 205816f..badfedb 100644 --- a/arch/sh/kernel/timers/timer-tmu.c +++ b/arch/sh/kernel/timers/timer-tmu.c @@ -111,60 +111,6 @@ static struct irqaction tmu_irq = { .mask = CPU_MASK_NONE, }; -/* - * Hah! We'll see if this works (switching from usecs to nsecs). - */ -static unsigned long tmu_timer_get_frequency(void) -{ - u32 freq; - struct timespec ts1, ts2; - unsigned long diff_nsec; - unsigned long factor; - - /* Setup the timer: We don't want to generate interrupts, just - * have it count down at its natural rate. - */ - ctrl_outb(0, TMU_TSTR); -#if !defined(CONFIG_CPU_SUBTYPE_SH7300) && !defined(CONFIG_CPU_SUBTYPE_SH7760) - ctrl_outb(TMU_TOCR_INIT, TMU_TOCR); -#endif - ctrl_outw(TMU0_TCR_CALIB, TMU0_TCR); - ctrl_outl(0xffffffff, TMU0_TCOR); - ctrl_outl(0xffffffff, TMU0_TCNT); - - rtc_sh_get_time(&ts2); - - do { - rtc_sh_get_time(&ts1); - } while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec); - - /* actually start the timer */ - ctrl_outb(TMU_TSTR_INIT, TMU_TSTR); - - do { - rtc_sh_get_time(&ts2); - } while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec); - - freq = 0xffffffff - ctrl_inl(TMU0_TCNT); - if (ts2.tv_nsec < ts1.tv_nsec) { - ts2.tv_nsec += 1000000000; - ts2.tv_sec--; - } - - diff_nsec = (ts2.tv_sec - ts1.tv_sec) * 1000000000 + (ts2.tv_nsec - ts1.tv_nsec); - - /* this should work well if the RTC has a precision of n Hz, where - * n is an integer. I don't think we have to worry about the other - * cases. */ - factor = (1000000000 + diff_nsec/2) / diff_nsec; - - if (factor * diff_nsec > 1100000000 || - factor * diff_nsec < 900000000) - panic("weird RTC (diff_nsec %ld)", diff_nsec); - - return freq * factor; -} - static void tmu_clk_init(struct clk *clk) { u8 divisor = TMU0_TCR_INIT & 0x7; @@ -232,12 +178,12 @@ struct sys_timer_ops tmu_timer_ops = { .init = tmu_timer_init, .start = tmu_timer_start, .stop = tmu_timer_stop, - .get_frequency = tmu_timer_get_frequency, +#ifndef CONFIG_GENERIC_TIME .get_offset = tmu_timer_get_offset, +#endif }; struct sys_timer tmu_timer = { .name = "tmu", .ops = &tmu_timer_ops, }; - diff --git a/include/asm-sh/timer.h b/include/asm-sh/timer.h index ebc78db..341cb71 100644 --- a/include/asm-sh/timer.h +++ b/include/asm-sh/timer.h @@ -11,7 +11,6 @@ struct sys_timer_ops { #ifndef CONFIG_GENERIC_TIME unsigned long (*get_offset)(void); #endif - unsigned long (*get_frequency)(void); }; struct sys_timer { @@ -33,11 +32,6 @@ static inline unsigned long get_timer_offset(void) } #endif -static inline unsigned long get_timer_frequency(void) -{ - return sys_timer->ops->get_frequency(); -} - /* arch/sh/kernel/timers/timer.c */ struct sys_timer *get_sys_timer(void); -- cgit v0.10.2 From 35f3c5185b1e28e6591aa649db8bf4fa16f1a7f3 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 6 Oct 2006 15:31:16 +0900 Subject: sh: Updates for IRQ handler changes. Trivial fixes for build breakage introduced by IRQ handler changes. Signed-off-by: Paul Mundt diff --git a/arch/sh/boards/hp6xx/hp6xx_apm.c b/arch/sh/boards/hp6xx/hp6xx_apm.c index 75f91aa..2191791 100644 --- a/arch/sh/boards/hp6xx/hp6xx_apm.c +++ b/arch/sh/boards/hp6xx/hp6xx_apm.c @@ -83,7 +83,7 @@ static int hp6x0_apm_get_info(char *buf, char **start, off_t fpos, int length) return p - buf; } -static irqreturn_t hp6x0_apm_interrupt(int irq, void *dev, struct pt_regs *regs) +static irqreturn_t hp6x0_apm_interrupt(int irq, void *dev) { if (!apm_suspended) apm_queue_event(APM_USER_SUSPEND); @@ -96,7 +96,7 @@ static int __init hp6x0_apm_init(void) int ret; ret = request_irq(HP680_BTN_IRQ, hp6x0_apm_interrupt, - SA_INTERRUPT, MODNAME, 0); + IRQF_DISABLED, MODNAME, 0); if (unlikely(ret < 0)) { printk(KERN_ERR MODNAME ": IRQ %d request failed\n", HP680_BTN_IRQ); diff --git a/arch/sh/boards/landisk/landisk_pwb.c b/arch/sh/boards/landisk/landisk_pwb.c index 0b7bee1..e625249 100644 --- a/arch/sh/boards/landisk/landisk_pwb.c +++ b/arch/sh/boards/landisk/landisk_pwb.c @@ -135,7 +135,7 @@ static int swdrv_write(struct file *filp, const char *buff, size_t count, return count; } -static irqreturn_t sw_interrupt(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t sw_interrupt(int irq, void *dev_id) { landisk_btn = (0x0ff & (~ctrl_inb(PA_STATUS))); disable_irq(IRQ_BUTTON); diff --git a/arch/sh/boards/mpc1211/setup.c b/arch/sh/boards/mpc1211/setup.c index 01c10fa..7c3d1d3 100644 --- a/arch/sh/boards/mpc1211/setup.c +++ b/arch/sh/boards/mpc1211/setup.c @@ -69,7 +69,6 @@ static void __init pci_write_config(unsigned long busNo, static unsigned char m_irq_mask = 0xfb; static unsigned char s_irq_mask = 0xff; -volatile unsigned long irq_err_count; static void disable_mpc1211_irq(unsigned int irq) { @@ -118,7 +117,7 @@ static void mask_and_ack_mpc1211(unsigned int irq) if(irq < 8) { if(m_irq_mask & (1<channel[i].size - \ g2_dma->status[i].size) & 0x0fffffff) -static irqreturn_t g2_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t g2_dma_interrupt(int irq, void *dev_id) { int i; diff --git a/arch/sh/drivers/dma/dma-pvr2.c b/arch/sh/drivers/dma/dma-pvr2.c index c1b6bc2..838fad5 100644 --- a/arch/sh/drivers/dma/dma-pvr2.c +++ b/arch/sh/drivers/dma/dma-pvr2.c @@ -21,7 +21,7 @@ static unsigned int xfer_complete; static int count; -static irqreturn_t pvr2_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t pvr2_dma_interrupt(int irq, void *dev_id) { if (get_dma_residue(PVR2_CASCADE_CHAN)) { printk(KERN_WARNING "DMA: SH DMAC did not complete transfer " diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c index cbbe8bc..d8ece20 100644 --- a/arch/sh/drivers/dma/dma-sh.c +++ b/arch/sh/drivers/dma/dma-sh.c @@ -60,9 +60,9 @@ static inline unsigned int calc_xmit_shift(struct dma_channel *chan) * Besides that it needs to waken any waiting process, which should handle * setting up the next transfer. */ -static irqreturn_t dma_tei(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t dma_tei(int irq, void *dev_id) { - struct dma_channel *chan = (struct dma_channel *)dev_id; + struct dma_channel *chan = dev_id; u32 chcr; chcr = ctrl_inl(CHCR[chan->chan]); @@ -228,7 +228,7 @@ static inline int dmaor_reset(void) } #if defined(CONFIG_CPU_SH4) -static irqreturn_t dma_err(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t dma_err(int irq, void *dummy) { dmaor_reset(); disable_irq(irq); diff --git a/arch/sh/drivers/pci/pci-st40.c b/arch/sh/drivers/pci/pci-st40.c index 4ab5ea6..efecb3d 100644 --- a/arch/sh/drivers/pci/pci-st40.c +++ b/arch/sh/drivers/pci/pci-st40.c @@ -161,7 +161,7 @@ static char * pci_commands[16]={ "Memory Write-and-Invalidate" }; -static irqreturn_t st40_pci_irq(int irq, void *dev_instance, struct pt_regs *regs) +static irqreturn_t st40_pci_irq(int irq, void *dev_instance) { unsigned pci_int, pci_air, pci_cir, pci_aint; static int count=0; diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c index c7ebd6a..3b93682 100644 --- a/arch/sh/kernel/irq.c +++ b/arch/sh/kernel/irq.c @@ -17,6 +17,8 @@ #include #include +atomic_t irq_err_count; + /* * 'what should we do if we get a hw irq event on an illegal vector'. * each architecture has to answer this themselves, it doesn't deserve @@ -47,8 +49,10 @@ int show_interrupts(struct seq_file *p, void *v) if (!action) goto unlock; seq_printf(p, "%3d: ",i); - seq_printf(p, "%10u ", kstat_irqs(i)); - seq_printf(p, " %14s", irq_desc[i].chip->typename); + for_each_online_cpu(j) + seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); + seq_printf(p, " %14s", irq_desc[i].chip->name); + seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) @@ -56,7 +60,9 @@ int show_interrupts(struct seq_file *p, void *v) seq_putc(p, '\n'); unlock: spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } + } else if (i == NR_IRQS) + seq_printf(p, "Err: %10u\n", atomic_read(&irq_err_count)); + return 0; } #endif @@ -78,6 +84,7 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5, unsigned long r6, unsigned long r7, struct pt_regs regs) { + struct pt_regs *old_regs = set_irq_regs(®s); int irq = r4; #ifdef CONFIG_4KSTACKS union irq_ctx *curctx, *irqctx; @@ -139,7 +146,6 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5, __asm__ __volatile__ ( "mov %0, r4 \n" - "mov %1, r5 \n" "mov r15, r9 \n" "jsr @%2 \n" /* swith to the irq stack */ @@ -147,17 +153,18 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5, /* restore the stack (ring zero) */ "mov r9, r15 \n" : /* no outputs */ - : "r" (irq), "r" (®s), "r" (__do_IRQ), "r" (isp) + : "r" (irq), "r" (generic_handle_irq), "r" (isp) /* XXX: A somewhat excessive clobber list? -PFM */ : "memory", "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "t", "pr" ); } else #endif - __do_IRQ(irq, ®s); + generic_handle_irq(irq); irq_exit(); + set_irq_regs(old_regs); return 1; } diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c index 1fbb83c..57e708d 100644 --- a/arch/sh/kernel/time.c +++ b/arch/sh/kernel/time.c @@ -109,13 +109,14 @@ static long last_rtc_update; * handle_timer_tick() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick */ -void handle_timer_tick(struct pt_regs *regs) +void handle_timer_tick(void) { do_timer(1); #ifndef CONFIG_SMP - update_process_times(user_mode(regs)); + update_process_times(user_mode(get_irq_regs())); #endif - profile_tick(CPU_PROFILING, regs); + if (current->pid) + profile_tick(CPU_PROFILING); #ifdef CONFIG_HEARTBEAT if (sh_mv.mv_heartbeat != NULL) diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c index badfedb..2492701 100644 --- a/arch/sh/kernel/timers/timer-tmu.c +++ b/arch/sh/kernel/timers/timer-tmu.c @@ -80,8 +80,7 @@ static unsigned long tmu_timer_get_offset(void) return count; } -static irqreturn_t tmu_timer_interrupt(int irq, void *dev_id, - struct pt_regs *regs) +static irqreturn_t tmu_timer_interrupt(int irq, void *dummy) { unsigned long timer_status; @@ -98,7 +97,7 @@ static irqreturn_t tmu_timer_interrupt(int irq, void *dev_id, * locally disabled. -arca */ write_seqlock(&xtime_lock); - handle_timer_tick(regs); + handle_timer_tick(); write_sequnlock(&xtime_lock); return IRQ_HANDLED; diff --git a/drivers/rtc/rtc-sh.c b/drivers/rtc/rtc-sh.c index 8b6efcc..143302a 100644 --- a/drivers/rtc/rtc-sh.c +++ b/drivers/rtc/rtc-sh.c @@ -160,7 +160,7 @@ static int sh_rtc_open(struct device *dev) tmp |= RCR1_CIE; writeb(tmp, rtc->regbase + RCR1); - ret = request_irq(rtc->periodic_irq, sh_rtc_periodic, SA_INTERRUPT, + ret = request_irq(rtc->periodic_irq, sh_rtc_periodic, IRQF_DISABLED, "sh-rtc period", dev); if (unlikely(ret)) { dev_err(dev, "request period IRQ failed with %d, IRQ %d\n", @@ -168,7 +168,7 @@ static int sh_rtc_open(struct device *dev) return ret; } - ret = request_irq(rtc->carry_irq, sh_rtc_interrupt, SA_INTERRUPT, + ret = request_irq(rtc->carry_irq, sh_rtc_interrupt, IRQF_DISABLED, "sh-rtc carry", dev); if (unlikely(ret)) { dev_err(dev, "request carry IRQ failed with %d, IRQ %d\n", @@ -177,7 +177,7 @@ static int sh_rtc_open(struct device *dev) goto err_bad_carry; } - ret = request_irq(rtc->alarm_irq, sh_rtc_interrupt, SA_INTERRUPT, + ret = request_irq(rtc->alarm_irq, sh_rtc_interrupt, IRQF_DISABLED, "sh-rtc alarm", dev); if (unlikely(ret)) { dev_err(dev, "request alarm IRQ failed with %d, IRQ %d\n", diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c index 266aa32..cfcc3ca 100644 --- a/drivers/serial/sh-sci.c +++ b/drivers/serial/sh-sci.c @@ -808,7 +808,7 @@ static int sci_request_irq(struct sci_port *port) } if (request_irq(port->irqs[0], sci_mpxed_interrupt, - SA_INTERRUPT, "sci", port)) { + IRQF_DISABLED, "sci", port)) { printk(KERN_ERR "sci: Cannot allocate irq.\n"); return -ENODEV; } @@ -817,7 +817,7 @@ static int sci_request_irq(struct sci_port *port) if (!port->irqs[i]) continue; if (request_irq(port->irqs[i], handlers[i], - SA_INTERRUPT, desc[i], port)) { + IRQF_DISABLED, desc[i], port)) { printk(KERN_ERR "sci: Cannot allocate irq.\n"); return -ENODEV; } diff --git a/include/asm-sh/hw_irq.h b/include/asm-sh/hw_irq.h index fed2661..80ee1cd 100644 --- a/include/asm-sh/hw_irq.h +++ b/include/asm-sh/hw_irq.h @@ -1,4 +1,8 @@ #ifndef __ASM_SH_HW_IRQ_H #define __ASM_SH_HW_IRQ_H +#include + +extern atomic_t irq_err_count; + #endif /* __ASM_SH_HW_IRQ_H */ diff --git a/include/asm-sh/irq_regs.h b/include/asm-sh/irq_regs.h new file mode 100644 index 0000000..3dd9c0b --- /dev/null +++ b/include/asm-sh/irq_regs.h @@ -0,0 +1 @@ +#include diff --git a/include/asm-sh/timer.h b/include/asm-sh/timer.h index 341cb71..5df842b 100644 --- a/include/asm-sh/timer.h +++ b/include/asm-sh/timer.h @@ -36,7 +36,6 @@ static inline unsigned long get_timer_offset(void) struct sys_timer *get_sys_timer(void); /* arch/sh/kernel/time.c */ -void handle_timer_tick(struct pt_regs *); +void handle_timer_tick(void); #endif /* __ASM_SH_TIMER_H */ - -- cgit v0.10.2 From 257440b00ba42a96522255029aa9406ffb7e2f62 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 6 Oct 2006 15:33:00 +0900 Subject: sh: Convert r7780rp IRQ handler to IRQ chip. Simple conversion of the R7780RP IRQ handler to struct irq_chip. Signed-off-by: Paul Mundt diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c index 2d960e9..5519d3d 100644 --- a/arch/sh/boards/renesas/r7780rp/irq.c +++ b/arch/sh/boards/renesas/r7780rp/irq.c @@ -1,18 +1,16 @@ /* - * linux/arch/sh/boards/renesas/r7780rp/irq.c - * - * Copyright (C) 2000 Kazumoto Kojima - * * Renesas Solutions Highlander R7780RP-1 Support. * - * Modified for R7780RP-1 by - * Atom Create Engineering Co., Ltd. 2002. + * Copyright (C) 2002 Atom Create Engineering Co., Ltd. + * Copyright (C) 2006 Paul Mundt + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. */ #include #include #include -#include -#include #ifdef CONFIG_SH_R7780MP static int mask_pos[] = {12, 11, 9, 14, 15, 8, 13, 6, 5, 4, 3, 2, 0, 0, 1, 0}; @@ -20,71 +18,26 @@ static int mask_pos[] = {12, 11, 9, 14, 15, 8, 13, 6, 5, 4, 3, 2, 0, 0, 1, 0}; static int mask_pos[] = {15, 14, 13, 12, 11, 10, 9, 8, 7, 5, 6, 4, 0, 1, 2, 0}; #endif -static void enable_r7780rp_irq(unsigned int irq); -static void disable_r7780rp_irq(unsigned int irq); - -/* shutdown is same as "disable" */ -#define shutdown_r7780rp_irq disable_r7780rp_irq - -static void ack_r7780rp_irq(unsigned int irq); -static void end_r7780rp_irq(unsigned int irq); - -static unsigned int startup_r7780rp_irq(unsigned int irq) -{ - enable_r7780rp_irq(irq); - return 0; /* never anything pending */ -} - -static void disable_r7780rp_irq(unsigned int irq) -{ - unsigned short val; - unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]); - - /* Set the priority in IPR to 0 */ - val = ctrl_inw(IRLCNTR1); - val &= mask; - ctrl_outw(val, IRLCNTR1); -} - static void enable_r7780rp_irq(unsigned int irq) { - unsigned short val; - unsigned short value = (0x0001 << mask_pos[irq]); - /* Set priority in IPR back to original value */ - val = ctrl_inw(IRLCNTR1); - val |= value; - ctrl_outw(val, IRLCNTR1); -} - -static void ack_r7780rp_irq(unsigned int irq) -{ - disable_r7780rp_irq(irq); + ctrl_outw(ctrl_inw(IRLCNTR1) | (1 << mask_pos[irq]), IRLCNTR1); } -static void end_r7780rp_irq(unsigned int irq) +static void disable_r7780rp_irq(unsigned int irq) { - if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) - enable_r7780rp_irq(irq); + /* Set the priority in IPR to 0 */ + ctrl_outw(ctrl_inw(IRLCNTR1) & (0xffff ^ (1 << mask_pos[irq])), + IRLCNTR1); } -static struct hw_interrupt_type r7780rp_irq_type = { - .typename = "R7780RP-IRQ", - .startup = startup_r7780rp_irq, - .shutdown = shutdown_r7780rp_irq, - .enable = enable_r7780rp_irq, - .disable = disable_r7780rp_irq, - .ack = ack_r7780rp_irq, - .end = end_r7780rp_irq, +static struct irq_chip r7780rp_irq_chip __read_mostly = { + .name = "r7780rp", + .mask = disable_r7780rp_irq, + .unmask = enable_r7780rp_irq, + .mask_ack = disable_r7780rp_irq, }; -static void make_r7780rp_irq(unsigned int irq) -{ - disable_irq_nosync(irq); - irq_desc[irq].chip = &r7780rp_irq_type; - disable_r7780rp_irq(irq); -} - /* * Initialize IRQ setting */ @@ -92,24 +45,10 @@ void __init init_r7780rp_IRQ(void) { int i; - /* IRL0=PCI Slot #A - * IRL1=PCI Slot #B - * IRL2=PCI Slot #C - * IRL3=PCI Slot #D - * IRL4=CF Card - * IRL5=CF Card Insert - * IRL6=M66596 - * IRL7=SD Card - * IRL8=Touch Panel - * IRL9=SCI - * IRL10=Serial - * IRL11=Extention #A - * IRL11=Extention #B - * IRL12=Debug LAN - * IRL13=Push Switch - * IRL14=ZiggBee IO - */ - - for (i=0; i<15; i++) - make_r7780rp_irq(i); + for (i = 0; i < 15; i++) { + disable_irq_nosync(i); + set_irq_chip_and_handler(i, &r7780rp_irq_chip, + handle_level_irq); + disable_r7780rp_irq(i); + } } -- cgit v0.10.2 From 580410005daecd8e9f0e1baa9ddadbb7f706e7e2 Mon Sep 17 00:00:00 2001 From: Jamie Lenehan Date: Fri, 6 Oct 2006 15:36:15 +0900 Subject: sh: Fix pr_debug statements for sh4 Fix a problem uncovered by the recent change to always check the arguments to pr_debug. The sh7751 was using the wrong name for the PCI IO base address. Signed-off-by: Jamie Lenehan Signed-off-by: Paul Mundt diff --git a/arch/sh/drivers/pci/pci-sh7751.c b/arch/sh/drivers/pci/pci-sh7751.c index dbe8378..85e1ee2 100644 --- a/arch/sh/drivers/pci/pci-sh7751.c +++ b/arch/sh/drivers/pci/pci-sh7751.c @@ -155,7 +155,7 @@ int __init sh7751_pcic_init(struct sh4_pci_address_map *map) */ pr_debug("PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO, (64 << 10), - SH4_PCI_IO_BASE + PCIBIOS_MIN_IO); + SH7751_PCI_IO_BASE + PCIBIOS_MIN_IO); /* * XXX: For now, leave this board-specific. In the event we have other @@ -163,7 +163,7 @@ int __init sh7751_pcic_init(struct sh4_pci_address_map *map) */ #ifdef CONFIG_SH_BIGSUR bigsur_port_map(PCIBIOS_MIN_IO, (64 << 10), - SH4_PCI_IO_BASE + PCIBIOS_MIN_IO, 0); + SH7751_PCI_IO_BASE + PCIBIOS_MIN_IO, 0); #endif /* Make sure the MSB's of IO window are set to access PCI space -- cgit v0.10.2 From 525ccc452c79db41874c5edac3f67618a0997d6f Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 6 Oct 2006 17:35:48 +0900 Subject: sh: Convert INTC2 IRQ handler to irq_chip. More struct irq_chip conversions, this time the INTC2 handlers. Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/cpu/irq/intc2.c b/arch/sh/kernel/cpu/irq/intc2.c index e30e4b7..d4b2bb7 100644 --- a/arch/sh/kernel/cpu/irq/intc2.c +++ b/arch/sh/kernel/cpu/irq/intc2.c @@ -10,93 +10,32 @@ * These are the "new Hitachi style" interrupts, as present on the * Hitachi 7751, the STM ST40 STB1, SH7760, and SH7780. */ - #include #include #include #include #include -#include - -struct intc2_data { - unsigned char msk_offset; - unsigned char msk_shift; - - int (*clear_irq) (int); -}; - -static struct intc2_data intc2_data[NR_INTC2_IRQS]; - -static void enable_intc2_irq(unsigned int irq); -static void disable_intc2_irq(unsigned int irq); - -/* shutdown is same as "disable" */ -#define shutdown_intc2_irq disable_intc2_irq - -static void mask_and_ack_intc2(unsigned int); -static void end_intc2_irq(unsigned int irq); - -static unsigned int startup_intc2_irq(unsigned int irq) -{ - enable_intc2_irq(irq); - return 0; /* never anything pending */ -} - -static struct hw_interrupt_type intc2_irq_type = { - .typename = "INTC2-IRQ", - .startup = startup_intc2_irq, - .shutdown = shutdown_intc2_irq, - .enable = enable_intc2_irq, - .disable = disable_intc2_irq, - .ack = mask_and_ack_intc2, - .end = end_intc2_irq -}; static void disable_intc2_irq(unsigned int irq) { - int irq_offset = irq - INTC2_FIRST_IRQ; - int msk_shift, msk_offset; - - /* Sanity check */ - if (unlikely(irq_offset < 0 || irq_offset >= NR_INTC2_IRQS)) - return; - - msk_shift = intc2_data[irq_offset].msk_shift; - msk_offset = intc2_data[irq_offset].msk_offset; - - ctrl_outl(1 << msk_shift, - INTC2_BASE + INTC2_INTMSK_OFFSET + msk_offset); + struct intc2_data *p = get_irq_chip_data(irq); + ctrl_outl(1 << p->msk_shift, + INTC2_BASE + INTC2_INTMSK_OFFSET + p->msk_offset); } static void enable_intc2_irq(unsigned int irq) { - int irq_offset = irq - INTC2_FIRST_IRQ; - int msk_shift, msk_offset; - - /* Sanity check */ - if (unlikely(irq_offset < 0 || irq_offset >= NR_INTC2_IRQS)) - return; - - msk_shift = intc2_data[irq_offset].msk_shift; - msk_offset = intc2_data[irq_offset].msk_offset; - - ctrl_outl(1 << msk_shift, - INTC2_BASE + INTC2_INTMSKCLR_OFFSET + msk_offset); -} - -static void mask_and_ack_intc2(unsigned int irq) -{ - disable_intc2_irq(irq); + struct intc2_data *p = get_irq_chip_data(irq); + ctrl_outl(1 << p->msk_shift, + INTC2_BASE + INTC2_INTMSKCLR_OFFSET + p->msk_offset); } -static void end_intc2_irq(unsigned int irq) -{ - if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) - enable_intc2_irq(irq); - - if (unlikely(intc2_data[irq - INTC2_FIRST_IRQ].clear_irq)) - intc2_data[irq - INTC2_FIRST_IRQ].clear_irq(irq); -} +static struct irq_chip intc2_irq_chip = { + .typename = "intc2", + .mask = disable_intc2_irq, + .unmask = enable_intc2_irq, + .mask_ack = disable_intc2_irq, +}; /* * Setup an INTC2 style interrupt. @@ -108,46 +47,30 @@ static void end_intc2_irq(unsigned int irq) * | | | | * make_intc2_irq(84, 0, 16, 0, 13); */ -void make_intc2_irq(unsigned int irq, - unsigned int ipr_offset, unsigned int ipr_shift, - unsigned int msk_offset, unsigned int msk_shift, - unsigned int priority) +void make_intc2_irq(struct intc2_data *p) { - int irq_offset = irq - INTC2_FIRST_IRQ; unsigned int flags; unsigned long ipr; - if (unlikely(irq_offset < 0 || irq_offset >= NR_INTC2_IRQS)) - return; - - disable_irq_nosync(irq); - - /* Fill the data we need */ - intc2_data[irq_offset].msk_offset = msk_offset; - intc2_data[irq_offset].msk_shift = msk_shift; - intc2_data[irq_offset].clear_irq = NULL; + disable_irq_nosync(p->irq); /* Set the priority level */ local_irq_save(flags); - ipr = ctrl_inl(INTC2_BASE + INTC2_INTPRI_OFFSET + ipr_offset); - ipr &= ~(0xf << ipr_shift); - ipr |= priority << ipr_shift; - ctrl_outl(ipr, INTC2_BASE + INTC2_INTPRI_OFFSET + ipr_offset); + ipr = ctrl_inl(INTC2_BASE + INTC2_INTPRI_OFFSET + p->ipr_offset); + ipr &= ~(0xf << p->ipr_shift); + ipr |= p->priority << p->ipr_shift; + ctrl_outl(ipr, INTC2_BASE + INTC2_INTPRI_OFFSET + p->ipr_offset); local_irq_restore(flags); - irq_desc[irq].chip = &intc2_irq_type; + set_irq_chip_and_handler(p->irq, &intc2_irq_chip, handle_level_irq); + set_irq_chip_data(p->irq, p); - disable_intc2_irq(irq); + enable_intc2_irq(p->irq); } -static struct intc2_init { - unsigned short irq; - unsigned char ipr_offset, ipr_shift; - unsigned char msk_offset, msk_shift; - unsigned char priority; -} intc2_init_data[] __initdata = { +static struct intc2_data intc2_irq_table[] = { #if defined(CONFIG_CPU_SUBTYPE_ST40) {64, 0, 0, 0, 0, 13}, /* PCI serr */ {65, 0, 4, 0, 1, 13}, /* PCI err */ @@ -266,19 +189,6 @@ void __init init_IRQ_intc2(void) { int i; - for (i = 0; i < ARRAY_SIZE(intc2_init_data); i++) { - struct intc2_init *p = intc2_init_data + i; - make_intc2_irq(p->irq, p->ipr_offset, p->ipr_shift, - p-> msk_offset, p->msk_shift, p->priority); - } -} - -/* Adds a termination callback to the interrupt */ -void intc2_add_clear_irq(int irq, int (*fn)(int)) -{ - if (unlikely(irq < INTC2_FIRST_IRQ)) - return; - - intc2_data[irq - INTC2_FIRST_IRQ].clear_irq = fn; + for (i = 0; i < ARRAY_SIZE(intc2_irq_table); i++) + make_intc2_irq(intc2_irq_table + i); } - diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h index 0e5f365..28996f9 100644 --- a/include/asm-sh/irq.h +++ b/include/asm-sh/irq.h @@ -697,13 +697,15 @@ extern int ipr_irq_demux(int irq); #define INTC2_INTPRI_OFFSET 0x00 -void make_intc2_irq(unsigned int irq, - unsigned int ipr_offset, unsigned int ipr_shift, - unsigned int msk_offset, unsigned int msk_shift, - unsigned int priority); +struct intc2_data { + unsigned short irq; + unsigned char ipr_offset, ipr_shift; + unsigned char msk_offset, msk_shift; + unsigned char priority; +}; + +void make_intc2_irq(struct intc2_data *); void init_IRQ_intc2(void); -void intc2_add_clear_irq(int irq, int (*fn)(int)); - #endif extern int shmse_irq_demux(int irq); -- cgit v0.10.2 From 0f13804ae9d894c1fbd90bde38ae2aa0f01b0edd Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 6 Oct 2006 17:55:25 +0900 Subject: sh: Convert IPR-IRQ to IRQ chip. One more initial conversion.. Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c index f785822..8944abd 100644 --- a/arch/sh/kernel/cpu/irq/ipr.c +++ b/arch/sh/kernel/cpu/irq/ipr.c @@ -1,11 +1,10 @@ /* - * arch/sh/kernel/cpu/irq/ipr.c + * Interrupt handling for IPR-based IRQ. * * Copyright (C) 1999 Niibe Yutaka & Takeshi Yaegashi * Copyright (C) 2000 Kazumoto Kojima - * Copyright (C) 2003 Takashi Kusuda - * - * Interrupt handling for IPR-based IRQ. + * Copyright (C) 2003 Takashi Kusuda + * Copyright (C) 2006 Paul Mundt * * Supported system: * On-chip supporting modules (TMU, RTC, etc.). @@ -13,12 +12,13 @@ * Hitachi SolutionEngine external I/O: * MS7709SE01, MS7709ASE01, and MS7750SE01 * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. */ - #include #include #include - #include #include #include @@ -28,93 +28,45 @@ struct ipr_data { int shift; /* Shifts of the 16-bit data */ int priority; /* The priority */ }; -static struct ipr_data ipr_data[NR_IRQS]; - -static void enable_ipr_irq(unsigned int irq); -static void disable_ipr_irq(unsigned int irq); - -/* shutdown is same as "disable" */ -#define shutdown_ipr_irq disable_ipr_irq - -static void mask_and_ack_ipr(unsigned int); -static void end_ipr_irq(unsigned int irq); - -static unsigned int startup_ipr_irq(unsigned int irq) -{ - enable_ipr_irq(irq); - return 0; /* never anything pending */ -} - -static struct hw_interrupt_type ipr_irq_type = { - .typename = "IPR-IRQ", - .startup = startup_ipr_irq, - .shutdown = shutdown_ipr_irq, - .enable = enable_ipr_irq, - .disable = disable_ipr_irq, - .ack = mask_and_ack_ipr, - .end = end_ipr_irq -}; static void disable_ipr_irq(unsigned int irq) { - unsigned long val; - unsigned int addr = ipr_data[irq].addr; - unsigned short mask = 0xffff ^ (0x0f << ipr_data[irq].shift); - + struct ipr_data *p = get_irq_chip_data(irq); /* Set the priority in IPR to 0 */ - val = ctrl_inw(addr); - val &= mask; - ctrl_outw(val, addr); + ctrl_outw(ctrl_inw(p->addr) & (0xffff ^ (0xf << p->shift)), p->addr); } static void enable_ipr_irq(unsigned int irq) { - unsigned long val; - unsigned int addr = ipr_data[irq].addr; - int priority = ipr_data[irq].priority; - unsigned short value = (priority << ipr_data[irq].shift); - + struct ipr_data *p = get_irq_chip_data(irq); /* Set priority in IPR back to original value */ - val = ctrl_inw(addr); - val |= value; - ctrl_outw(val, addr); + ctrl_outw(ctrl_inw(p->addr) | (p->priority << p->shift), p->addr); } -static void mask_and_ack_ipr(unsigned int irq) -{ - disable_ipr_irq(irq); - -#if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \ - defined(CONFIG_CPU_SUBTYPE_SH7706) || \ - defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705) - /* This is needed when we use edge triggered setting */ - /* XXX: Is it really needed? */ - if (IRQ0_IRQ <= irq && irq <= IRQ5_IRQ) { - /* Clear external interrupt request */ - int a = ctrl_inb(INTC_IRR0); - a &= ~(1 << (irq - IRQ0_IRQ)); - ctrl_outb(a, INTC_IRR0); - } -#endif -} - -static void end_ipr_irq(unsigned int irq) -{ - if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) - enable_ipr_irq(irq); -} +static struct irq_chip ipr_irq_chip = { + .name = "ipr", + .mask = disable_ipr_irq, + .unmask = enable_ipr_irq, + .mask_ack = disable_ipr_irq, +}; void make_ipr_irq(unsigned int irq, unsigned int addr, int pos, int priority) { + struct ipr_data ipr_data; + disable_irq_nosync(irq); - ipr_data[irq].addr = addr; - ipr_data[irq].shift = pos*4; /* POSition (0-3) x 4 means shift */ - ipr_data[irq].priority = priority; - irq_desc[irq].chip = &ipr_irq_type; - disable_ipr_irq(irq); + ipr_data.addr = addr; + ipr_data.shift = pos*4; /* POSition (0-3) x 4 means shift */ + ipr_data.priority = priority; + + set_irq_chip_and_handler(irq, &ipr_irq_chip, handle_level_irq); + set_irq_chip_data(irq, &ipr_data); + + enable_ipr_irq(irq); } +/* XXX: This needs to die a horrible death.. */ void __init init_IRQ(void) { #ifndef CONFIG_CPU_SUBTYPE_SH7780 -- cgit v0.10.2 From 268f3be177ce93791da38facc34126b5038cd851 Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 6 Oct 2006 21:47:09 +0000 Subject: [CIFS] readdir (ffirst) enablement of accurate timestamps from legacy servers Signed-off-by: Steve French diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index fa5124d..3d86b31 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -918,45 +918,42 @@ __le64 cnvrtDosCifsTm(__u16 date, __u16 time) { return cpu_to_le64(cifs_UnixTimeToNT(cnvrtDosUnixTm(date, time))); } + struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) { - __u8 dt[2]; - __u8 tm[2]; struct timespec ts; - int sec,min, days, month, year; - struct timespec removeme; /* BB removeme BB */ -/* SMB_TIME * st = (SMB_TIME *)&time;*/ + int sec, min, days, month, year; + SMB_TIME * st = (SMB_TIME *)&time; + SMB_DATE * sd = (SMB_DATE *)&date; cFYI(1,("date %d time %d",date, time)); - dt[0] = date & 0xFF; - dt[1] = (date & 0xFF00) >> 8; - tm[0] = time & 0xFF; - tm[1] = (time & 0xFF00) >> 8; - - sec = tm[0] & 0x1F; - sec = 2 * sec; - min = ((tm[0] >>5)&0xFF) + ((tm[1] & 0x7)<<3); - + sec = 2 * st->TwoSeconds; + min = st->Minutes; + if((sec > 59) || (min > 59)) + cERROR(1,("illegal time min %d sec %d", min, sec)); sec += (min * 60); - sec += 60 * 60 * ((tm[1] >> 3) &0xFF) /* hours */; - days = (dt[0] & 0x1F) - 1; - month = ((dt[0] >> 5) & 0xFF) + ((dt[1] & 0x1) <<3); - if(month > 12) - cERROR(1,("illegal month %d in date", month)); + sec += 60 * 60 * st->Hours; + if(st->Hours > 24) + cERROR(1,("illegal hours %d",st->Hours)); + days = sd->Day; + month = sd->Month; + if((days > 31) || (month > 12)) + cERROR(1,("illegal date, month %d day: %d", month, days)); month -= 1; days += total_days_of_prev_months[month]; days += 3653; /* account for difference in days between 1980 and 1970 */ - year = (dt[1]>>1) & 0xFF; + year = sd->Year; days += year * 365; days += (year/4); /* leap year */ /* adjust for leap year where we are still before leap day */ days -= ((year & 0x03) == 0) && (month < 2 ? 1 : 0); sec += 24 * 60 * 60 * days; - removeme = CURRENT_TIME; /* BB removeme BB */ ts.tv_sec = sec; + /* cFYI(1,("sec after cnvrt dos to unix time %d",sec)); */ + ts.tv_nsec = 0; return ts; } diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index b0e5db1..81e7b2e 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -139,19 +139,19 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, FIND_FILE_STANDARD_INFO * pfindData = (FIND_FILE_STANDARD_INFO *)buf; -/* ts = cnvrtDosUnixTm( + tmp_inode->i_mtime = cnvrtDosUnixTm( le16_to_cpu(pfindData->LastWriteDate), - le16_to_cpu(pfindData->LastWriteTime));*/ + le16_to_cpu(pfindData->LastWriteTime)); + tmp_inode->i_atime = cnvrtDosUnixTm( + le16_to_cpu(pfindData->LastAccessDate), + le16_to_cpu(pfindData->LastAccessTime)); + tmp_inode->i_ctime = cnvrtDosUnixTm( + le16_to_cpu(pfindData->LastWriteDate), + le16_to_cpu(pfindData->LastWriteTime)); + attr = le16_to_cpu(pfindData->Attributes); allocation_size = le32_to_cpu(pfindData->AllocationSize); end_of_file = le32_to_cpu(pfindData->DataSize); - /* do not need to use current_fs_time helper function since - time not stored for this case so atime can not "go backwards" - by pulling newer older from disk when inode refrenshed */ - tmp_inode->i_atime = CURRENT_TIME; - /* tmp_inode->i_mtime = BB FIXME - add dos time handling - tmp_inode->i_ctime = 0; BB FIXME */ - } /* Linux can not store file creation time unfortunately so ignore it */ -- cgit v0.10.2 From e34477e9906acc137329b654a51fb7d4598813f7 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Fri, 6 Oct 2006 13:41:12 -0700 Subject: [WATCHDOG] ioremap balanced with iounmap for drivers/char/watchdog/s3c2410_wdt.c ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. Signed-off-by: Amol Lad Signed-off-by: Wim Van Sebroeck Signed-off-by: Andrew Morton diff --git a/drivers/char/watchdog/s3c2410_wdt.c b/drivers/char/watchdog/s3c2410_wdt.c index b36a04a..d54d0ef 100644 --- a/drivers/char/watchdog/s3c2410_wdt.c +++ b/drivers/char/watchdog/s3c2410_wdt.c @@ -381,18 +381,21 @@ static int s3c2410wdt_probe(struct platform_device *pdev) res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); if (res == NULL) { printk(KERN_INFO PFX "failed to get irq resource\n"); + iounmap(wdt_base); return -ENOENT; } ret = request_irq(res->start, s3c2410wdt_irq, 0, pdev->name, pdev); if (ret != 0) { printk(KERN_INFO PFX "failed to install irq (%d)\n", ret); + iounmap(wdt_base); return ret; } wdt_clock = clk_get(&pdev->dev, "watchdog"); if (wdt_clock == NULL) { printk(KERN_INFO PFX "failed to find watchdog clock source\n"); + iounmap(wdt_base); return -ENOENT; } @@ -416,6 +419,7 @@ static int s3c2410wdt_probe(struct platform_device *pdev) if (ret) { printk (KERN_ERR PFX "cannot register miscdev on minor=%d (%d)\n", WATCHDOG_MINOR, ret); + iounmap(wdt_base); return ret; } @@ -452,6 +456,7 @@ static int s3c2410wdt_remove(struct platform_device *dev) wdt_clock = NULL; } + iounmap(wdt_base); misc_deregister(&s3c2410wdt_miscdev); return 0; } -- cgit v0.10.2 From bcbf25bd0d4afb108a755e1c4e4e2d854a2869d7 Mon Sep 17 00:00:00 2001 From: "Arnaud Patard (Rtp)" Date: Wed, 4 Oct 2006 14:18:29 +0200 Subject: [WATCHDOG] add ich8 support to iTCO_wdt.c Add ICH8 support to the iTCO_wdt driver. Signed-off-by: Arnaud Patard Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/iTCO_wdt.c b/drivers/char/watchdog/iTCO_wdt.c index aaac94d..639d84f 100644 --- a/drivers/char/watchdog/iTCO_wdt.c +++ b/drivers/char/watchdog/iTCO_wdt.c @@ -85,6 +85,7 @@ enum iTCO_chipsets { TCO_ICH7, /* ICH7 & ICH7R */ TCO_ICH7M, /* ICH7-M */ TCO_ICH7MDH, /* ICH7-M DH */ + TCO_ICH8, /* ICH8 */ }; static struct { @@ -108,6 +109,7 @@ static struct { {"ICH7 or ICH7R", 2}, {"ICH7-M", 2}, {"ICH7-M DH", 2}, + {"ICH8 or ICH8R", 2}, {NULL,0} }; @@ -135,6 +137,7 @@ static struct pci_device_id iTCO_wdt_pci_tbl[] = { { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH7 }, { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH7M }, { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH7MDH }, + { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH8 }, { 0, }, /* End of list */ }; MODULE_DEVICE_TABLE (pci, iTCO_wdt_pci_tbl); -- cgit v0.10.2 From a8edd74e4404d011ab821d5bf35b27335d26f001 Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sun, 8 Oct 2006 21:05:21 +0200 Subject: [WATCHDOG] add ich8 support to iTCO_wdt.c (patch 2) Add ICH8 support to the iTCO_wdt driver. Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/iTCO_wdt.c b/drivers/char/watchdog/iTCO_wdt.c index 639d84f..505aae9 100644 --- a/drivers/char/watchdog/iTCO_wdt.c +++ b/drivers/char/watchdog/iTCO_wdt.c @@ -35,6 +35,10 @@ * 82801GDH (ICH7DH) : document number 307013-002, 307014-009, * 82801GBM (ICH7-M) : document number 307013-002, 307014-009, * 82801GHM (ICH7-M DH) : document number 307013-002, 307014-009, + * 82801HB (ICH8) : document number 313056-002, 313057-004, + * 82801HR (ICH8R) : document number 313056-002, 313057-004, + * 82801HH (ICH8DH) : document number 313056-002, 313057-004, + * 82801HO (ICH8DO) : document number 313056-002, 313057-004, * 6300ESB (6300ESB) : document number 300641-003 */ @@ -45,7 +49,7 @@ /* Module and version information */ #define DRV_NAME "iTCO_wdt" #define DRV_VERSION "1.00" -#define DRV_RELDATE "30-Jul-2006" +#define DRV_RELDATE "08-Oct-2006" #define PFX DRV_NAME ": " /* Includes */ @@ -85,7 +89,9 @@ enum iTCO_chipsets { TCO_ICH7, /* ICH7 & ICH7R */ TCO_ICH7M, /* ICH7-M */ TCO_ICH7MDH, /* ICH7-M DH */ - TCO_ICH8, /* ICH8 */ + TCO_ICH8, /* ICH8 & ICH8R */ + TCO_ICH8DH, /* ICH8DH */ + TCO_ICH8DO, /* ICH8DO */ }; static struct { @@ -110,6 +116,8 @@ static struct { {"ICH7-M", 2}, {"ICH7-M DH", 2}, {"ICH8 or ICH8R", 2}, + {"ICH8DH", 2}, + {"ICH8DO", 2}, {NULL,0} }; @@ -138,6 +146,8 @@ static struct pci_device_id iTCO_wdt_pci_tbl[] = { { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH7M }, { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH7MDH }, { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH8 }, + { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH8DH }, + { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TCO_ICH8DO }, { 0, }, /* End of list */ }; MODULE_DEVICE_TABLE (pci, iTCO_wdt_pci_tbl); -- cgit v0.10.2 From 73f5e28b336772c4b08ee82e5bf28ab872898ee1 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 9 Oct 2006 21:58:54 +0200 Subject: r8169: PCI ID for Corega Gigabit network card Fix for http://bugzilla.kernel.org/show_bug.cgi?id=7239. Signed-off-by: Andrew Morton Signed-off-by: Francois Romieu diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 4c47c5b..c7309e9 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -214,6 +214,7 @@ static struct pci_device_id rtl8169_pci_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8168), 0, 0, RTL_CFG_2 }, { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8169), 0, 0, RTL_CFG_0 }, { PCI_DEVICE(PCI_VENDOR_ID_DLINK, 0x4300), 0, 0, RTL_CFG_0 }, + { PCI_DEVICE(0x1259, 0xc107), 0, 0, RTL_CFG_0 }, { PCI_DEVICE(0x16ec, 0x0116), 0, 0, RTL_CFG_0 }, { PCI_VENDOR_ID_LINKSYS, 0x1032, PCI_ANY_ID, 0x0024, 0, 0, RTL_CFG_0 }, -- cgit v0.10.2 From 833abf7fe011c3dfe9bcab405856b73deab17062 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Tue, 10 Oct 2006 18:33:10 +0900 Subject: sh: Zero-out coherent buffer in consistent_alloc(). Be sure to zero out the buffer, this was causing occasional problems under heavier PCI tests. Signed-off-by: Paul Mundt diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c index c81e6b6..38c82d8 100644 --- a/arch/sh/mm/consistent.c +++ b/arch/sh/mm/consistent.c @@ -28,6 +28,7 @@ void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle) split_page(page, order); ret = page_address(page); + memset(ret, 0, size); *handle = virt_to_phys(ret); /* -- cgit v0.10.2 From 438dd926260f11ff01fc3441ac6dd4c412d20ea4 Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 11 Oct 2006 03:49:30 +0000 Subject: [CIFS] Fix leaps year calculation for years after 2100 Signed-off-by: Steve French diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 3d86b31..32562d1 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -946,6 +946,15 @@ struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) year = sd->Year; days += year * 365; days += (year/4); /* leap year */ + /* generalized leap year calculation is more complex, ie no leap year + for years/100 except for years/400, but since the maximum number for DOS + year is 2**7, the last year is 1980+127, which means we need only + consider 2 special case years, ie the years 2000 and 2100, and only + adjust for the lack of leap year for the year 2100, as 2000 was a + leap year (divisable by 400) */ + if(year >= 120) /* the year 2100 */ + days = days - 1; /* do not count leap year for the year 2100 */ + /* adjust for leap year where we are still before leap day */ days -= ((year & 0x03) == 0) && (month < 2 ? 1 : 0); sec += 24 * 60 * 60 * days; -- cgit v0.10.2 From 80fc9f532d8c05d4cb12d55660624ce53a378349 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 11 Oct 2006 01:43:58 -0400 Subject: Input: add missing exports to fix modular build Signed-off-by: Dmitry Torokhov diff --git a/drivers/char/random.c b/drivers/char/random.c index 07f47a0..eb6b13f 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -645,6 +645,7 @@ void add_input_randomness(unsigned int type, unsigned int code, add_timer_randomness(&input_timer_state, (type << 4) ^ code ^ (code >> 4) ^ value); } +EXPORT_SYMBOL_GPL(add_input_randomness); void add_interrupt_randomness(int irq) { diff --git a/lib/kobject.c b/lib/kobject.c index 1699eb9..7dd5c0e 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -119,6 +119,7 @@ char *kobject_get_path(struct kobject *kobj, gfp_t gfp_mask) return path; } +EXPORT_SYMBOL_GPL(kobject_get_path); /** * kobject_init - initialize object. -- cgit v0.10.2 From 817e6ba3623de9cdc66c6aba90eae30b5588ff11 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 11 Oct 2006 01:44:28 -0400 Subject: Input: i8042 - supress ACK/NAKs when blinking during panic This allows using SysRq and not fill logs with complaints from atkbd. Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c index 09b06e6..7e3141f 100644 --- a/drivers/input/serio/i8042.c +++ b/drivers/input/serio/i8042.c @@ -106,6 +106,7 @@ static unsigned char i8042_ctr; static unsigned char i8042_mux_present; static unsigned char i8042_kbd_irq_registered; static unsigned char i8042_aux_irq_registered; +static unsigned char i8042_suppress_kbd_ack; static struct platform_device *i8042_platform_device; static irqreturn_t i8042_interrupt(int irq, void *dev_id); @@ -316,7 +317,7 @@ static irqreturn_t i8042_interrupt(int irq, void *dev_id) unsigned char str, data; unsigned int dfl; unsigned int port_no; - int ret; + int ret = 1; spin_lock_irqsave(&i8042_lock, flags); str = i8042_read_status(); @@ -378,10 +379,16 @@ static irqreturn_t i8042_interrupt(int irq, void *dev_id) dfl & SERIO_PARITY ? ", bad parity" : "", dfl & SERIO_TIMEOUT ? ", timeout" : ""); + if (unlikely(i8042_suppress_kbd_ack)) + if (port_no == I8042_KBD_PORT_NO && + (data == 0xfa || data == 0xfe)) { + i8042_suppress_kbd_ack = 0; + goto out; + } + if (likely(port->exists)) serio_interrupt(port->serio, data, dfl); - ret = 1; out: return IRQ_RETVAL(ret); } @@ -842,11 +849,13 @@ static long i8042_panic_blink(long count) led ^= 0x01 | 0x04; while (i8042_read_status() & I8042_STR_IBF) DELAY; + i8042_suppress_kbd_ack = 1; i8042_write_data(0xed); /* set leds */ DELAY; while (i8042_read_status() & I8042_STR_IBF) DELAY; DELAY; + i8042_suppress_kbd_ack = 1; i8042_write_data(led); DELAY; last_blink = count; -- cgit v0.10.2 From 86255d9d0bede79140f4912482447963f00818c0 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 11 Oct 2006 01:44:46 -0400 Subject: Input: atkbd - supress "too many keys" error message Many users seems to be annoyed by this warning so kill the message and implement a counter exported as a sysfs attribute so we still know what is going on. Make atkbd use attribute groups while we are at it. Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c index b6ef9ea..cbb9366 100644 --- a/drivers/input/keyboard/atkbd.c +++ b/drivers/input/keyboard/atkbd.c @@ -221,6 +221,7 @@ struct atkbd { unsigned long xl_bit; unsigned int last; unsigned long time; + unsigned long err_count; struct work_struct event_work; struct mutex event_mutex; @@ -234,11 +235,13 @@ static ssize_t atkbd_attr_set_helper(struct device *dev, const char *buf, size_t #define ATKBD_DEFINE_ATTR(_name) \ static ssize_t atkbd_show_##_name(struct atkbd *, char *); \ static ssize_t atkbd_set_##_name(struct atkbd *, const char *, size_t); \ -static ssize_t atkbd_do_show_##_name(struct device *d, struct device_attribute *attr, char *b) \ +static ssize_t atkbd_do_show_##_name(struct device *d, \ + struct device_attribute *attr, char *b) \ { \ return atkbd_attr_show_helper(d, b, atkbd_show_##_name); \ } \ -static ssize_t atkbd_do_set_##_name(struct device *d, struct device_attribute *attr, const char *b, size_t s) \ +static ssize_t atkbd_do_set_##_name(struct device *d, \ + struct device_attribute *attr, const char *b, size_t s) \ { \ return atkbd_attr_set_helper(d, b, s, atkbd_set_##_name); \ } \ @@ -251,6 +254,32 @@ ATKBD_DEFINE_ATTR(set); ATKBD_DEFINE_ATTR(softrepeat); ATKBD_DEFINE_ATTR(softraw); +#define ATKBD_DEFINE_RO_ATTR(_name) \ +static ssize_t atkbd_show_##_name(struct atkbd *, char *); \ +static ssize_t atkbd_do_show_##_name(struct device *d, \ + struct device_attribute *attr, char *b) \ +{ \ + return atkbd_attr_show_helper(d, b, atkbd_show_##_name); \ +} \ +static struct device_attribute atkbd_attr_##_name = \ + __ATTR(_name, S_IRUGO, atkbd_do_show_##_name, NULL); + +ATKBD_DEFINE_RO_ATTR(err_count); + +static struct attribute *atkbd_attributes[] = { + &atkbd_attr_extra.attr, + &atkbd_attr_scroll.attr, + &atkbd_attr_set.attr, + &atkbd_attr_softrepeat.attr, + &atkbd_attr_softraw.attr, + &atkbd_attr_err_count.attr, + NULL +}; + +static struct attribute_group atkbd_attribute_group = { + .attrs = atkbd_attributes, +}; + static const unsigned int xl_table[] = { ATKBD_RET_BAT, ATKBD_RET_ERR, ATKBD_RET_ACK, ATKBD_RET_NAK, ATKBD_RET_HANJA, ATKBD_RET_HANGEUL, @@ -396,7 +425,10 @@ static irqreturn_t atkbd_interrupt(struct serio *serio, unsigned char data, add_release_event = 1; break; case ATKBD_RET_ERR: + atkbd->err_count++; +#ifdef ATKBD_DEBUG printk(KERN_DEBUG "atkbd.c: Keyboard on %s reports too many keys pressed.\n", serio->phys); +#endif goto out; } @@ -786,12 +818,7 @@ static void atkbd_disconnect(struct serio *serio) synchronize_sched(); /* Allow atkbd_interrupt()s to complete. */ flush_scheduled_work(); - device_remove_file(&serio->dev, &atkbd_attr_extra); - device_remove_file(&serio->dev, &atkbd_attr_scroll); - device_remove_file(&serio->dev, &atkbd_attr_set); - device_remove_file(&serio->dev, &atkbd_attr_softrepeat); - device_remove_file(&serio->dev, &atkbd_attr_softraw); - + sysfs_remove_group(&serio->dev.kobj, &atkbd_attribute_group); input_unregister_device(atkbd->dev); serio_close(serio); serio_set_drvdata(serio, NULL); @@ -961,11 +988,7 @@ static int atkbd_connect(struct serio *serio, struct serio_driver *drv) atkbd_set_keycode_table(atkbd); atkbd_set_device_attrs(atkbd); - device_create_file(&serio->dev, &atkbd_attr_extra); - device_create_file(&serio->dev, &atkbd_attr_scroll); - device_create_file(&serio->dev, &atkbd_attr_set); - device_create_file(&serio->dev, &atkbd_attr_softrepeat); - device_create_file(&serio->dev, &atkbd_attr_softraw); + sysfs_create_group(&serio->dev.kobj, &atkbd_attribute_group); atkbd_enable(atkbd); @@ -1259,6 +1282,11 @@ static ssize_t atkbd_set_softraw(struct atkbd *atkbd, const char *buf, size_t co return count; } +static ssize_t atkbd_show_err_count(struct atkbd *atkbd, char *buf) +{ + return sprintf(buf, "%lu\n", atkbd->err_count); +} + static int __init atkbd_init(void) { -- cgit v0.10.2 From 4dfbb9d8c6cbfc32faa5c71145bd2a43e1f8237c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Oct 2006 01:45:14 -0400 Subject: Lockdep: add lockdep_set_class_and_subclass() and lockdep_set_subclass() This annotation makes it possible to assign a subclass on lock init. This annotation is meant to reduce the _nested() annotations by assigning a default subclass. One could do without this annotation and rely on lockdep_set_class() exclusively, but that would require a manual stack of struct lock_class_key objects. Signed-off-by: Peter Zijlstra Signed-off-by: Dmitry Torokhov diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 1314ca0..14fec2a 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -202,7 +202,7 @@ extern int lockdep_internal(void); */ extern void lockdep_init_map(struct lockdep_map *lock, const char *name, - struct lock_class_key *key); + struct lock_class_key *key, int subclass); /* * Reinitialize a lock key - for cases where there is special locking or @@ -211,9 +211,14 @@ extern void lockdep_init_map(struct lockdep_map *lock, const char *name, * or they are too narrow (they suffer from a false class-split): */ #define lockdep_set_class(lock, key) \ - lockdep_init_map(&(lock)->dep_map, #key, key) + lockdep_init_map(&(lock)->dep_map, #key, key, 0) #define lockdep_set_class_and_name(lock, key, name) \ - lockdep_init_map(&(lock)->dep_map, name, key) + lockdep_init_map(&(lock)->dep_map, name, key, 0) +#define lockdep_set_class_and_subclass(lock, key, sub) \ + lockdep_init_map(&(lock)->dep_map, #key, key, sub) +#define lockdep_set_subclass(lock, sub) \ + lockdep_init_map(&(lock)->dep_map, #lock, \ + (lock)->dep_map.key, sub) /* * Acquire a lock. @@ -257,10 +262,12 @@ static inline int lockdep_internal(void) # define lock_release(l, n, i) do { } while (0) # define lockdep_init() do { } while (0) # define lockdep_info() do { } while (0) -# define lockdep_init_map(lock, name, key) do { (void)(key); } while (0) +# define lockdep_init_map(lock, name, key, sub) do { (void)(key); } while (0) # define lockdep_set_class(lock, key) do { (void)(key); } while (0) # define lockdep_set_class_and_name(lock, key, name) \ do { (void)(key); } while (0) +#define lockdep_set_class_and_subclass(lock, key, sub) \ + do { (void)(key); } while (0) # define INIT_LOCKDEP # define lockdep_reset() do { debug_locks = 1; } while (0) # define lockdep_free_key_range(start, size) do { } while (0) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 4c05534..ba7156a 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -1177,7 +1177,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) * itself, so actual lookup of the hash should be once per lock object. */ static inline struct lock_class * -register_lock_class(struct lockdep_map *lock, unsigned int subclass) +register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) { struct lockdep_subclass_key *key; struct list_head *hash_head; @@ -1249,7 +1249,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass) out_unlock_set: __raw_spin_unlock(&hash_lock); - if (!subclass) + if (!subclass || force) lock->class_cache = class; DEBUG_LOCKS_WARN_ON(class->subclass != subclass); @@ -1937,7 +1937,7 @@ void trace_softirqs_off(unsigned long ip) * Initialize a lock instance's lock-class mapping info: */ void lockdep_init_map(struct lockdep_map *lock, const char *name, - struct lock_class_key *key) + struct lock_class_key *key, int subclass) { if (unlikely(!debug_locks)) return; @@ -1957,6 +1957,8 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, lock->name = name; lock->key = key; lock->class_cache = NULL; + if (subclass) + register_lock_class(lock, subclass, 1); } EXPORT_SYMBOL_GPL(lockdep_init_map); @@ -1995,7 +1997,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, * Not cached yet or subclass? */ if (unlikely(!class)) { - class = register_lock_class(lock, subclass); + class = register_lock_class(lock, subclass, 0); if (!class) return 0; } diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index e3203c6..1865164 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -91,7 +91,7 @@ void debug_mutex_init(struct mutex *lock, const char *name, * Make sure we are not reinitializing a held lock: */ debug_check_no_locks_freed((void *)lock, sizeof(*lock)); - lockdep_init_map(&lock->dep_map, name, key); + lockdep_init_map(&lock->dep_map, name, key, 0); #endif lock->owner = NULL; lock->magic = lock; diff --git a/lib/rwsem-spinlock.c b/lib/rwsem-spinlock.c index db4fed7..c4cfd6c 100644 --- a/lib/rwsem-spinlock.c +++ b/lib/rwsem-spinlock.c @@ -28,7 +28,7 @@ void __init_rwsem(struct rw_semaphore *sem, const char *name, * Make sure we are not reinitializing a held semaphore: */ debug_check_no_locks_freed((void *)sem, sizeof(*sem)); - lockdep_init_map(&sem->dep_map, name, key); + lockdep_init_map(&sem->dep_map, name, key, 0); #endif sem->activity = 0; spin_lock_init(&sem->wait_lock); diff --git a/lib/rwsem.c b/lib/rwsem.c index 901d0e7..cdb4e3d 100644 --- a/lib/rwsem.c +++ b/lib/rwsem.c @@ -19,7 +19,7 @@ void __init_rwsem(struct rw_semaphore *sem, const char *name, * Make sure we are not reinitializing a held semaphore: */ debug_check_no_locks_freed((void *)sem, sizeof(*sem)); - lockdep_init_map(&sem->dep_map, name, key); + lockdep_init_map(&sem->dep_map, name, key, 0); #endif sem->count = RWSEM_UNLOCKED_VALUE; spin_lock_init(&sem->wait_lock); diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index dafaf1d..b6c4f89 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -20,7 +20,7 @@ void __spin_lock_init(spinlock_t *lock, const char *name, * Make sure we are not reinitializing a held lock: */ debug_check_no_locks_freed((void *)lock, sizeof(*lock)); - lockdep_init_map(&lock->dep_map, name, key); + lockdep_init_map(&lock->dep_map, name, key, 0); #endif lock->raw_lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; lock->magic = SPINLOCK_MAGIC; @@ -38,7 +38,7 @@ void __rwlock_init(rwlock_t *lock, const char *name, * Make sure we are not reinitializing a held lock: */ debug_check_no_locks_freed((void *)lock, sizeof(*lock)); - lockdep_init_map(&lock->dep_map, name, key); + lockdep_init_map(&lock->dep_map, name, key, 0); #endif lock->raw_lock = (raw_rwlock_t) __RAW_RW_LOCK_UNLOCKED; lock->magic = RWLOCK_MAGIC; diff --git a/net/core/sock.c b/net/core/sock.c index b77e155..d472db4 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -823,7 +823,7 @@ static void inline sock_lock_init(struct sock *sk) af_family_slock_key_strings[sk->sk_family]); lockdep_init_map(&sk->sk_lock.dep_map, af_family_key_strings[sk->sk_family], - af_family_keys + sk->sk_family); + af_family_keys + sk->sk_family, 0); } /** -- cgit v0.10.2 From 88aa0103e408616e433c209e80169ab8d6eda99e Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Wed, 11 Oct 2006 01:45:31 -0400 Subject: Input: serio - add lockdep annotations Signed-off-by: Jiri Kosina Acked-by: Peter Zijlstra Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index dcb16b5..e5b1b60 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -189,7 +189,7 @@ int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) return -1; } - mutex_lock_nested(&ps2dev->cmd_mutex, SINGLE_DEPTH_NESTING); + mutex_lock(&ps2dev->cmd_mutex); serio_pause_rx(ps2dev->serio); ps2dev->flags = command == PS2_CMD_GETID ? PS2_FLAG_WAITID : 0; @@ -296,6 +296,7 @@ EXPORT_SYMBOL(ps2_schedule_command); void ps2_init(struct ps2dev *ps2dev, struct serio *serio) { mutex_init(&ps2dev->cmd_mutex); + lockdep_set_subclass(&ps2dev->cmd_mutex, serio->depth); init_waitqueue_head(&ps2dev->wait); ps2dev->serio = serio; } diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index 960fae3..480fdc5 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -538,8 +538,12 @@ static void serio_init_port(struct serio *serio) "serio%ld", (long)atomic_inc_return(&serio_no) - 1); serio->dev.bus = &serio_bus; serio->dev.release = serio_release_port; - if (serio->parent) + if (serio->parent) { serio->dev.parent = &serio->parent->dev; + serio->depth = serio->parent->depth + 1; + } else + serio->depth = 0; + lockdep_set_subclass(&serio->lock, serio->depth); } /* diff --git a/include/linux/serio.h b/include/linux/serio.h index 3a697cc..b99c5ca 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -41,6 +41,7 @@ struct serio { void (*stop)(struct serio *); struct serio *parent, *child; + unsigned int depth; /* level of nesting in serio hierarchy */ struct serio_driver *drv; /* accessed from interrupt, must be protected by serio->lock and serio->sem */ struct mutex drv_mutex; /* protects serio->drv so attributes can pin driver */ -- cgit v0.10.2 From 12f417ee95bf98cd3e42d2a771f7c6d360159b9d Mon Sep 17 00:00:00 2001 From: Deepak Saxena Date: Tue, 10 Oct 2006 14:33:22 -0700 Subject: [PATCH] Update smc91x driver with ARM Versatile board info We need to specify a Versatile-specific SMC_IRQ_FLAGS value or the new generic IRQ layer will complain thusly: No IRQF_TRIGGER set_type function for IRQ 25 () Signed-off-by: Deepak Saxena Cc: Jeff Garzik Cc: Russell King Cc: Nicolas Pitre Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h index 636dbfcd..0c9f1e7 100644 --- a/drivers/net/smc91x.h +++ b/drivers/net/smc91x.h @@ -398,6 +398,24 @@ static inline void LPD7_SMC_outsw (unsigned char* a, int r, #define SMC_IRQ_FLAGS (0) +#elif defined(CONFIG_ARCH_VERSATILE) + +#define SMC_CAN_USE_8BIT 1 +#define SMC_CAN_USE_16BIT 1 +#define SMC_CAN_USE_32BIT 1 +#define SMC_NOWAIT 1 + +#define SMC_inb(a, r) readb((a) + (r)) +#define SMC_inw(a, r) readw((a) + (r)) +#define SMC_inl(a, r) readl((a) + (r)) +#define SMC_outb(v, a, r) writeb(v, (a) + (r)) +#define SMC_outw(v, a, r) writew(v, (a) + (r)) +#define SMC_outl(v, a, r) writel(v, (a) + (r)) +#define SMC_insl(a, r, p, l) readsl((a) + (r), p, l) +#define SMC_outsl(a, r, p, l) writesl((a) + (r), p, l) + +#define SMC_IRQ_FLAGS (0) + #else #define SMC_CAN_USE_8BIT 1 -- cgit v0.10.2 From 6f62768344e46520ae585f3e201c9d3e497b028f Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Tue, 10 Oct 2006 14:33:26 -0700 Subject: [PATCH] b44: fix eeprom endianess issue This fixes eeprom read on big-endian architectures. readw returns the data in CPU order. With cpu_to_le16 we convert it to little endian, because "ptr" is a pointer to a _byte_ arrray. See the cast above. A byte array is little endian. The bug is: Reading u16 values with readw, casting them into an u8 array and accessing this u8 array as an u8 (byte) array. The correct fix is to swap the CPU-ordering value returned by readw into little endian, as the u8 array is little endian. This compiles to nothing on little endian hardware (so it does not change b44 code on LE hardware), but _fixes_ code on BE hardware. Signed-off-by: Michael Buesch Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/b44.c b/drivers/net/b44.c index ebb726e..1ec2174 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -2056,7 +2056,7 @@ static int b44_read_eeprom(struct b44 *bp, u8 *data) u16 *ptr = (u16 *) data; for (i = 0; i < 128; i += 2) - ptr[i / 2] = readw(bp->regs + 4096 + i); + ptr[i / 2] = cpu_to_le16(readw(bp->regs + 4096 + i)); return 0; } -- cgit v0.10.2 From 5f77113c01d8a9f8193769d2ca73763047af39ef Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 10 Oct 2006 14:33:30 -0700 Subject: [PATCH] ibmveth irq fix drivers/net/ibmveth.c:939: error: too many arguments to function `ibmveth_interrupt' Cc: Jeff Garzik Cc: Anton Blanchard Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index bf414a9..2802db2 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -936,7 +936,7 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu) static void ibmveth_poll_controller(struct net_device *dev) { ibmveth_replenish_task(dev->priv); - ibmveth_interrupt(dev->irq, dev, NULL); + ibmveth_interrupt(dev->irq, dev); } #endif -- cgit v0.10.2 From 08093c8fd66ef7a8c2f887812cc98c54f5f12703 Mon Sep 17 00:00:00 2001 From: Jan-Bernd Themann Date: Thu, 5 Oct 2006 16:53:12 +0200 Subject: [PATCH] ehea: firmware (hvcall) interface changes This eHEA patch covers required changes related to Anton Blanchard's new hvcall interface. Signed-off-by: Jan-Bernd Themann Signed-off-by: Jeff Garzik diff --git a/drivers/net/ehea/ehea_phyp.c b/drivers/net/ehea/ehea_phyp.c index 4a85aca..0b51a8c 100644 --- a/drivers/net/ehea/ehea_phyp.c +++ b/drivers/net/ehea/ehea_phyp.c @@ -44,71 +44,99 @@ static inline u16 get_order_of_qentries(u16 queue_entries) #define H_ALL_RES_TYPE_MR 5 #define H_ALL_RES_TYPE_MW 6 -static long ehea_hcall_9arg_9ret(unsigned long opcode, - unsigned long arg1, unsigned long arg2, - unsigned long arg3, unsigned long arg4, - unsigned long arg5, unsigned long arg6, - unsigned long arg7, unsigned long arg8, - unsigned long arg9, unsigned long *out1, - unsigned long *out2,unsigned long *out3, - unsigned long *out4,unsigned long *out5, - unsigned long *out6,unsigned long *out7, - unsigned long *out8,unsigned long *out9) +static long ehea_plpar_hcall_norets(unsigned long opcode, + unsigned long arg1, + unsigned long arg2, + unsigned long arg3, + unsigned long arg4, + unsigned long arg5, + unsigned long arg6, + unsigned long arg7) { - long hret; + long ret; int i, sleep_msecs; for (i = 0; i < 5; i++) { - hret = plpar_hcall_9arg_9ret(opcode,arg1, arg2, arg3, arg4, - arg5, arg6, arg7, arg8, arg9, out1, - out2, out3, out4, out5, out6, out7, - out8, out9); - if (H_IS_LONG_BUSY(hret)) { - sleep_msecs = get_longbusy_msecs(hret); + ret = plpar_hcall_norets(opcode, arg1, arg2, arg3, arg4, + arg5, arg6, arg7); + + if (H_IS_LONG_BUSY(ret)) { + sleep_msecs = get_longbusy_msecs(ret); msleep_interruptible(sleep_msecs); continue; } - if (hret < H_SUCCESS) - ehea_error("op=%lx hret=%lx " - "i1=%lx i2=%lx i3=%lx i4=%lx i5=%lx i6=%lx " - "i7=%lx i8=%lx i9=%lx " - "o1=%lx o2=%lx o3=%lx o4=%lx o5=%lx o6=%lx " - "o7=%lx o8=%lx o9=%lx", - opcode, hret, arg1, arg2, arg3, arg4, arg5, - arg6, arg7, arg8, arg9, *out1, *out2, *out3, - *out4, *out5, *out6, *out7, *out8, *out9); - return hret; + if (ret < H_SUCCESS) + ehea_error("opcode=%lx ret=%lx" + " arg1=%lx arg2=%lx arg3=%lx arg4=%lx" + " arg5=%lx arg6=%lx arg7=%lx ", + opcode, ret, + arg1, arg2, arg3, arg4, arg5, + arg6, arg7); + + return ret; } + return H_BUSY; } -u64 ehea_h_query_ehea_qp(const u64 adapter_handle, const u8 qp_category, - const u64 qp_handle, const u64 sel_mask, void *cb_addr) +static long ehea_plpar_hcall9(unsigned long opcode, + unsigned long *outs, /* array of 9 outputs */ + unsigned long arg1, + unsigned long arg2, + unsigned long arg3, + unsigned long arg4, + unsigned long arg5, + unsigned long arg6, + unsigned long arg7, + unsigned long arg8, + unsigned long arg9) { - u64 dummy; + long ret; + int i, sleep_msecs; - if ((((u64)cb_addr) & (PAGE_SIZE - 1)) != 0) { - ehea_error("not on pageboundary"); - return H_PARAMETER; + for (i = 0; i < 5; i++) { + ret = plpar_hcall9(opcode, outs, + arg1, arg2, arg3, arg4, arg5, + arg6, arg7, arg8, arg9); + + if (H_IS_LONG_BUSY(ret)) { + sleep_msecs = get_longbusy_msecs(ret); + msleep_interruptible(sleep_msecs); + continue; + } + + if (ret < H_SUCCESS) + ehea_error("opcode=%lx ret=%lx" + " arg1=%lx arg2=%lx arg3=%lx arg4=%lx" + " arg5=%lx arg6=%lx arg7=%lx arg8=%lx" + " arg9=%lx" + " out1=%lx out2=%lx out3=%lx out4=%lx" + " out5=%lx out6=%lx out7=%lx out8=%lx" + " out9=%lx", + opcode, ret, + arg1, arg2, arg3, arg4, arg5, + arg6, arg7, arg8, arg9, + outs[0], outs[1], outs[2], outs[3], + outs[4], outs[5], outs[6], outs[7], + outs[8]); + + return ret; } - return ehea_hcall_9arg_9ret(H_QUERY_HEA_QP, - adapter_handle, /* R4 */ - qp_category, /* R5 */ - qp_handle, /* R6 */ - sel_mask, /* R7 */ - virt_to_abs(cb_addr), /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return H_BUSY; +} + +u64 ehea_h_query_ehea_qp(const u64 adapter_handle, const u8 qp_category, + const u64 qp_handle, const u64 sel_mask, void *cb_addr) +{ + return ehea_plpar_hcall_norets(H_QUERY_HEA_QP, + adapter_handle, /* R4 */ + qp_category, /* R5 */ + qp_handle, /* R6 */ + sel_mask, /* R7 */ + virt_to_abs(cb_addr), /* R8 */ + 0, 0); } /* input param R5 */ @@ -180,6 +208,7 @@ u64 ehea_h_alloc_resource_qp(const u64 adapter_handle, u64 *qp_handle, struct h_epas *h_epas) { u64 hret; + u64 outs[PLPAR_HCALL9_BUFSIZE]; u64 allocate_controls = EHEA_BMASK_SET(H_ALL_RES_QP_EQPO, init_attr->low_lat_rq1 ? 1 : 0) @@ -219,45 +248,29 @@ u64 ehea_h_alloc_resource_qp(const u64 adapter_handle, EHEA_BMASK_SET(H_ALL_RES_QP_TH_RQ2, init_attr->rq2_threshold) | EHEA_BMASK_SET(H_ALL_RES_QP_TH_RQ3, init_attr->rq3_threshold); - u64 r5_out = 0; - u64 r6_out = 0; - u64 r7_out = 0; - u64 r8_out = 0; - u64 r9_out = 0; - u64 g_la_user_out = 0; - u64 r11_out = 0; - u64 r12_out = 0; - - hret = ehea_hcall_9arg_9ret(H_ALLOC_HEA_RESOURCE, - adapter_handle, /* R4 */ - allocate_controls, /* R5 */ - init_attr->send_cq_handle, /* R6 */ - init_attr->recv_cq_handle, /* R7 */ - init_attr->aff_eq_handle, /* R8 */ - r9_reg, /* R9 */ - max_r10_reg, /* R10 */ - r11_in, /* R11 */ - threshold, /* R12 */ - qp_handle, /* R4 */ - &r5_out, /* R5 */ - &r6_out, /* R6 */ - &r7_out, /* R7 */ - &r8_out, /* R8 */ - &r9_out, /* R9 */ - &g_la_user_out, /* R10 */ - &r11_out, /* R11 */ - &r12_out); /* R12 */ - - init_attr->qp_nr = (u32)r5_out; + hret = ehea_plpar_hcall9(H_ALLOC_HEA_RESOURCE, + outs, + adapter_handle, /* R4 */ + allocate_controls, /* R5 */ + init_attr->send_cq_handle, /* R6 */ + init_attr->recv_cq_handle, /* R7 */ + init_attr->aff_eq_handle, /* R8 */ + r9_reg, /* R9 */ + max_r10_reg, /* R10 */ + r11_in, /* R11 */ + threshold); /* R12 */ + + *qp_handle = outs[0]; + init_attr->qp_nr = (u32)outs[1]; init_attr->act_nr_send_wqes = - (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_SWQE, r6_out); + (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_SWQE, outs[2]); init_attr->act_nr_rwqes_rq1 = - (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_R1WQE, r6_out); + (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_R1WQE, outs[2]); init_attr->act_nr_rwqes_rq2 = - (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_R2WQE, r6_out); + (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_R2WQE, outs[2]); init_attr->act_nr_rwqes_rq3 = - (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_R3WQE, r6_out); + (u16)EHEA_BMASK_GET(H_ALL_RES_QP_ACT_R3WQE, outs[2]); init_attr->act_wqe_size_enc_sq = init_attr->wqe_size_enc_sq; init_attr->act_wqe_size_enc_rq1 = init_attr->wqe_size_enc_rq1; @@ -265,25 +278,25 @@ u64 ehea_h_alloc_resource_qp(const u64 adapter_handle, init_attr->act_wqe_size_enc_rq3 = init_attr->wqe_size_enc_rq3; init_attr->nr_sq_pages = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_SQ, r8_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_SQ, outs[4]); init_attr->nr_rq1_pages = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_RQ1, r8_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_RQ1, outs[4]); init_attr->nr_rq2_pages = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_RQ2, r9_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_RQ2, outs[5]); init_attr->nr_rq3_pages = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_RQ3, r9_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_SIZE_RQ3, outs[5]); init_attr->liobn_sq = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_SQ, r11_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_SQ, outs[7]); init_attr->liobn_rq1 = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_RQ1, r11_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_RQ1, outs[7]); init_attr->liobn_rq2 = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_RQ2, r12_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_RQ2, outs[8]); init_attr->liobn_rq3 = - (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_RQ3, r12_out); + (u32)EHEA_BMASK_GET(H_ALL_RES_QP_LIOBN_RQ3, outs[8]); if (!hret) - hcp_epas_ctor(h_epas, g_la_user_out, g_la_user_out); + hcp_epas_ctor(h_epas, outs[6], outs[6]); return hret; } @@ -292,31 +305,24 @@ u64 ehea_h_alloc_resource_cq(const u64 adapter_handle, struct ehea_cq_attr *cq_attr, u64 *cq_handle, struct h_epas *epas) { - u64 hret, dummy, act_nr_of_cqes_out, act_pages_out; - u64 g_la_privileged_out, g_la_user_out; - - hret = ehea_hcall_9arg_9ret(H_ALLOC_HEA_RESOURCE, - adapter_handle, /* R4 */ - H_ALL_RES_TYPE_CQ, /* R5 */ - cq_attr->eq_handle, /* R6 */ - cq_attr->cq_token, /* R7 */ - cq_attr->max_nr_of_cqes, /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - cq_handle, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &act_nr_of_cqes_out, /* R7 */ - &act_pages_out, /* R8 */ - &g_la_privileged_out, /* R9 */ - &g_la_user_out, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ - - cq_attr->act_nr_of_cqes = act_nr_of_cqes_out; - cq_attr->nr_pages = act_pages_out; + u64 hret; + u64 outs[PLPAR_HCALL9_BUFSIZE]; + + hret = ehea_plpar_hcall9(H_ALLOC_HEA_RESOURCE, + outs, + adapter_handle, /* R4 */ + H_ALL_RES_TYPE_CQ, /* R5 */ + cq_attr->eq_handle, /* R6 */ + cq_attr->cq_token, /* R7 */ + cq_attr->max_nr_of_cqes, /* R8 */ + 0, 0, 0, 0); /* R9-R12 */ + + *cq_handle = outs[0]; + cq_attr->act_nr_of_cqes = outs[3]; + cq_attr->nr_pages = outs[4]; if (!hret) - hcp_epas_ctor(epas, g_la_privileged_out, g_la_user_out); + hcp_epas_ctor(epas, outs[5], outs[6]); return hret; } @@ -361,9 +367,8 @@ u64 ehea_h_alloc_resource_cq(const u64 adapter_handle, u64 ehea_h_alloc_resource_eq(const u64 adapter_handle, struct ehea_eq_attr *eq_attr, u64 *eq_handle) { - u64 hret, dummy, eq_liobn, allocate_controls; - u64 ist1_out, ist2_out, ist3_out, ist4_out; - u64 act_nr_of_eqes_out, act_pages_out; + u64 hret, allocate_controls; + u64 outs[PLPAR_HCALL9_BUFSIZE]; /* resource type */ allocate_controls = @@ -372,27 +377,20 @@ u64 ehea_h_alloc_resource_eq(const u64 adapter_handle, | EHEA_BMASK_SET(H_ALL_RES_EQ_INH_EQE_GEN, !eq_attr->eqe_gen) | EHEA_BMASK_SET(H_ALL_RES_EQ_NON_NEQ_ISN, 1); - hret = ehea_hcall_9arg_9ret(H_ALLOC_HEA_RESOURCE, - adapter_handle, /* R4 */ - allocate_controls, /* R5 */ - eq_attr->max_nr_of_eqes, /* R6 */ - 0, 0, 0, 0, 0, 0, /* R7-R10 */ - eq_handle, /* R4 */ - &dummy, /* R5 */ - &eq_liobn, /* R6 */ - &act_nr_of_eqes_out, /* R7 */ - &act_pages_out, /* R8 */ - &ist1_out, /* R9 */ - &ist2_out, /* R10 */ - &ist3_out, /* R11 */ - &ist4_out); /* R12 */ - - eq_attr->act_nr_of_eqes = act_nr_of_eqes_out; - eq_attr->nr_pages = act_pages_out; - eq_attr->ist1 = ist1_out; - eq_attr->ist2 = ist2_out; - eq_attr->ist3 = ist3_out; - eq_attr->ist4 = ist4_out; + hret = ehea_plpar_hcall9(H_ALLOC_HEA_RESOURCE, + outs, + adapter_handle, /* R4 */ + allocate_controls, /* R5 */ + eq_attr->max_nr_of_eqes, /* R6 */ + 0, 0, 0, 0, 0, 0); /* R7-R10 */ + + *eq_handle = outs[0]; + eq_attr->act_nr_of_eqes = outs[3]; + eq_attr->nr_pages = outs[4]; + eq_attr->ist1 = outs[5]; + eq_attr->ist2 = outs[6]; + eq_attr->ist3 = outs[7]; + eq_attr->ist4 = outs[8]; return hret; } @@ -402,31 +400,22 @@ u64 ehea_h_modify_ehea_qp(const u64 adapter_handle, const u8 cat, void *cb_addr, u64 *inv_attr_id, u64 *proc_mask, u16 *out_swr, u16 *out_rwr) { - u64 hret, dummy, act_out_swr, act_out_rwr; - - if ((((u64)cb_addr) & (PAGE_SIZE - 1)) != 0) { - ehea_error("not on page boundary"); - return H_PARAMETER; - } - - hret = ehea_hcall_9arg_9ret(H_MODIFY_HEA_QP, - adapter_handle, /* R4 */ - (u64) cat, /* R5 */ - qp_handle, /* R6 */ - sel_mask, /* R7 */ - virt_to_abs(cb_addr), /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - inv_attr_id, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &act_out_swr, /* R7 */ - &act_out_rwr, /* R8 */ - proc_mask, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ - *out_swr = act_out_swr; - *out_rwr = act_out_rwr; + u64 hret; + u64 outs[PLPAR_HCALL9_BUFSIZE]; + + hret = ehea_plpar_hcall9(H_MODIFY_HEA_QP, + outs, + adapter_handle, /* R4 */ + (u64) cat, /* R5 */ + qp_handle, /* R6 */ + sel_mask, /* R7 */ + virt_to_abs(cb_addr), /* R8 */ + 0, 0, 0, 0); /* R9-R12 */ + + *inv_attr_id = outs[0]; + *out_swr = outs[3]; + *out_rwr = outs[4]; + *proc_mask = outs[5]; return hret; } @@ -435,122 +424,81 @@ u64 ehea_h_register_rpage(const u64 adapter_handle, const u8 pagesize, const u8 queue_type, const u64 resource_handle, const u64 log_pageaddr, u64 count) { - u64 dummy, reg_control; + u64 reg_control; reg_control = EHEA_BMASK_SET(H_REG_RPAGE_PAGE_SIZE, pagesize) | EHEA_BMASK_SET(H_REG_RPAGE_QT, queue_type); - return ehea_hcall_9arg_9ret(H_REGISTER_HEA_RPAGES, - adapter_handle, /* R4 */ - reg_control, /* R5 */ - resource_handle, /* R6 */ - log_pageaddr, /* R7 */ - count, /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return ehea_plpar_hcall_norets(H_REGISTER_HEA_RPAGES, + adapter_handle, /* R4 */ + reg_control, /* R5 */ + resource_handle, /* R6 */ + log_pageaddr, /* R7 */ + count, /* R8 */ + 0, 0); /* R9-R10 */ } u64 ehea_h_register_smr(const u64 adapter_handle, const u64 orig_mr_handle, const u64 vaddr_in, const u32 access_ctrl, const u32 pd, struct ehea_mr *mr) { - u64 hret, dummy, lkey_out; - - hret = ehea_hcall_9arg_9ret(H_REGISTER_SMR, - adapter_handle , /* R4 */ - orig_mr_handle, /* R5 */ - vaddr_in, /* R6 */ - (((u64)access_ctrl) << 32ULL), /* R7 */ - pd, /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - &mr->handle, /* R4 */ - &dummy, /* R5 */ - &lkey_out, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ - mr->lkey = (u32)lkey_out; + u64 hret; + u64 outs[PLPAR_HCALL9_BUFSIZE]; + + hret = ehea_plpar_hcall9(H_REGISTER_SMR, + outs, + adapter_handle , /* R4 */ + orig_mr_handle, /* R5 */ + vaddr_in, /* R6 */ + (((u64)access_ctrl) << 32ULL), /* R7 */ + pd, /* R8 */ + 0, 0, 0, 0); /* R9-R12 */ + + mr->handle = outs[0]; + mr->lkey = (u32)outs[2]; return hret; } u64 ehea_h_disable_and_get_hea(const u64 adapter_handle, const u64 qp_handle) { - u64 hret, dummy, ladr_next_sq_wqe_out; - u64 ladr_next_rq1_wqe_out, ladr_next_rq2_wqe_out, ladr_next_rq3_wqe_out; - - hret = ehea_hcall_9arg_9ret(H_DISABLE_AND_GET_HEA, - adapter_handle, /* R4 */ - H_DISABLE_GET_EHEA_WQE_P, /* R5 */ - qp_handle, /* R6 */ - 0, 0, 0, 0, 0, 0, /* R7-R12 */ - &ladr_next_sq_wqe_out, /* R4 */ - &ladr_next_rq1_wqe_out, /* R5 */ - &ladr_next_rq2_wqe_out, /* R6 */ - &ladr_next_rq3_wqe_out, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ - return hret; + u64 outs[PLPAR_HCALL9_BUFSIZE]; + + return ehea_plpar_hcall9(H_DISABLE_AND_GET_HEA, + outs, + adapter_handle, /* R4 */ + H_DISABLE_GET_EHEA_WQE_P, /* R5 */ + qp_handle, /* R6 */ + 0, 0, 0, 0, 0, 0); /* R7-R12 */ } u64 ehea_h_free_resource(const u64 adapter_handle, const u64 res_handle) { - u64 dummy; - - return ehea_hcall_9arg_9ret(H_FREE_RESOURCE, - adapter_handle, /* R4 */ - res_handle, /* R5 */ - 0, 0, 0, 0, 0, 0, 0, /* R6-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return ehea_plpar_hcall_norets(H_FREE_RESOURCE, + adapter_handle, /* R4 */ + res_handle, /* R5 */ + 0, 0, 0, 0, 0); /* R6-R10 */ } u64 ehea_h_alloc_resource_mr(const u64 adapter_handle, const u64 vaddr, const u64 length, const u32 access_ctrl, const u32 pd, u64 *mr_handle, u32 *lkey) { - u64 hret, dummy, lkey_out; - - hret = ehea_hcall_9arg_9ret(H_ALLOC_HEA_RESOURCE, - adapter_handle, /* R4 */ - 5, /* R5 */ - vaddr, /* R6 */ - length, /* R7 */ - (((u64) access_ctrl) << 32ULL),/* R8 */ - pd, /* R9 */ - 0, 0, 0, /* R10-R12 */ - mr_handle, /* R4 */ - &dummy, /* R5 */ - &lkey_out, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ - *lkey = (u32) lkey_out; - + u64 hret; + u64 outs[PLPAR_HCALL9_BUFSIZE]; + + hret = ehea_plpar_hcall9(H_ALLOC_HEA_RESOURCE, + outs, + adapter_handle, /* R4 */ + 5, /* R5 */ + vaddr, /* R6 */ + length, /* R7 */ + (((u64) access_ctrl) << 32ULL), /* R8 */ + pd, /* R9 */ + 0, 0, 0); /* R10-R12 */ + + *mr_handle = outs[0]; + *lkey = (u32)outs[2]; return hret; } @@ -570,23 +518,14 @@ u64 ehea_h_register_rpage_mr(const u64 adapter_handle, const u64 mr_handle, u64 ehea_h_query_ehea(const u64 adapter_handle, void *cb_addr) { - u64 hret, dummy, cb_logaddr; + u64 hret, cb_logaddr; cb_logaddr = virt_to_abs(cb_addr); - hret = ehea_hcall_9arg_9ret(H_QUERY_HEA, - adapter_handle, /* R4 */ - cb_logaddr, /* R5 */ - 0, 0, 0, 0, 0, 0, 0, /* R6-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + hret = ehea_plpar_hcall_norets(H_QUERY_HEA, + adapter_handle, /* R4 */ + cb_logaddr, /* R5 */ + 0, 0, 0, 0, 0); /* R6-R10 */ #ifdef DEBUG ehea_dmp(cb_addr, sizeof(struct hcp_query_ehea), "hcp_query_ehea"); #endif @@ -597,36 +536,28 @@ u64 ehea_h_query_ehea_port(const u64 adapter_handle, const u16 port_num, const u8 cb_cat, const u64 select_mask, void *cb_addr) { - u64 port_info, dummy; + u64 port_info; u64 cb_logaddr = virt_to_abs(cb_addr); u64 arr_index = 0; port_info = EHEA_BMASK_SET(H_MEHEAPORT_CAT, cb_cat) | EHEA_BMASK_SET(H_MEHEAPORT_PN, port_num); - return ehea_hcall_9arg_9ret(H_QUERY_HEA_PORT, - adapter_handle, /* R4 */ - port_info, /* R5 */ - select_mask, /* R6 */ - arr_index, /* R7 */ - cb_logaddr, /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return ehea_plpar_hcall_norets(H_QUERY_HEA_PORT, + adapter_handle, /* R4 */ + port_info, /* R5 */ + select_mask, /* R6 */ + arr_index, /* R7 */ + cb_logaddr, /* R8 */ + 0, 0); /* R9-R10 */ } u64 ehea_h_modify_ehea_port(const u64 adapter_handle, const u16 port_num, const u8 cb_cat, const u64 select_mask, void *cb_addr) { - u64 port_info, dummy, inv_attr_ident, proc_mask; + u64 outs[PLPAR_HCALL9_BUFSIZE]; + u64 port_info; u64 arr_index = 0; u64 cb_logaddr = virt_to_abs(cb_addr); @@ -635,29 +566,21 @@ u64 ehea_h_modify_ehea_port(const u64 adapter_handle, const u16 port_num, #ifdef DEBUG ehea_dump(cb_addr, sizeof(struct hcp_ehea_port_cb0), "Before HCALL"); #endif - return ehea_hcall_9arg_9ret(H_MODIFY_HEA_PORT, - adapter_handle, /* R4 */ - port_info, /* R5 */ - select_mask, /* R6 */ - arr_index, /* R7 */ - cb_logaddr, /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - &inv_attr_ident, /* R4 */ - &proc_mask, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return ehea_plpar_hcall9(H_MODIFY_HEA_PORT, + outs, + adapter_handle, /* R4 */ + port_info, /* R5 */ + select_mask, /* R6 */ + arr_index, /* R7 */ + cb_logaddr, /* R8 */ + 0, 0, 0, 0); /* R9-R12 */ } u64 ehea_h_reg_dereg_bcmc(const u64 adapter_handle, const u16 port_num, const u8 reg_type, const u64 mc_mac_addr, const u16 vlan_id, const u32 hcall_id) { - u64 r5_port_num, r6_reg_type, r7_mc_mac_addr, r8_vlan_id, dummy; + u64 r5_port_num, r6_reg_type, r7_mc_mac_addr, r8_vlan_id; u64 mac_addr = mc_mac_addr >> 16; r5_port_num = EHEA_BMASK_SET(H_REGBCMC_PN, port_num); @@ -665,41 +588,21 @@ u64 ehea_h_reg_dereg_bcmc(const u64 adapter_handle, const u16 port_num, r7_mc_mac_addr = EHEA_BMASK_SET(H_REGBCMC_MACADDR, mac_addr); r8_vlan_id = EHEA_BMASK_SET(H_REGBCMC_VLANID, vlan_id); - return ehea_hcall_9arg_9ret(hcall_id, - adapter_handle, /* R4 */ - r5_port_num, /* R5 */ - r6_reg_type, /* R6 */ - r7_mc_mac_addr, /* R7 */ - r8_vlan_id, /* R8 */ - 0, 0, 0, 0, /* R9-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return ehea_plpar_hcall_norets(hcall_id, + adapter_handle, /* R4 */ + r5_port_num, /* R5 */ + r6_reg_type, /* R6 */ + r7_mc_mac_addr, /* R7 */ + r8_vlan_id, /* R8 */ + 0, 0); /* R9-R12 */ } u64 ehea_h_reset_events(const u64 adapter_handle, const u64 neq_handle, const u64 event_mask) { - u64 dummy; - - return ehea_hcall_9arg_9ret(H_RESET_EVENTS, - adapter_handle, /* R4 */ - neq_handle, /* R5 */ - event_mask, /* R6 */ - 0, 0, 0, 0, 0, 0, /* R7-R12 */ - &dummy, /* R4 */ - &dummy, /* R5 */ - &dummy, /* R6 */ - &dummy, /* R7 */ - &dummy, /* R8 */ - &dummy, /* R9 */ - &dummy, /* R10 */ - &dummy, /* R11 */ - &dummy); /* R12 */ + return ehea_plpar_hcall_norets(H_RESET_EVENTS, + adapter_handle, /* R4 */ + neq_handle, /* R5 */ + event_mask, /* R6 */ + 0, 0, 0, 0); /* R7-R12 */ } -- cgit v0.10.2 From bff0a55f34e62970203c4af9c8ef4dc7d73e2f96 Mon Sep 17 00:00:00 2001 From: Jan-Bernd Themann Date: Thu, 5 Oct 2006 16:53:14 +0200 Subject: [PATCH] ehea: fix port state notification, default queue sizes This patch includes a bug fix for the port state notification and fixes the default queue sizes. Signed-off-by: Jan-Bernd Themann Signed-off-by: Jeff Garzik diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h index 23b451a..b40724f 100644 --- a/drivers/net/ehea/ehea.h +++ b/drivers/net/ehea/ehea.h @@ -39,7 +39,7 @@ #include #define DRV_NAME "ehea" -#define DRV_VERSION "EHEA_0028" +#define DRV_VERSION "EHEA_0034" #define EHEA_MSG_DEFAULT (NETIF_MSG_LINK | NETIF_MSG_TIMER \ | NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) @@ -50,6 +50,7 @@ #define EHEA_MAX_ENTRIES_SQ 32767 #define EHEA_MIN_ENTRIES_QP 127 +#define EHEA_SMALL_QUEUES #define EHEA_NUM_TX_QP 1 #ifdef EHEA_SMALL_QUEUES @@ -59,11 +60,11 @@ #define EHEA_DEF_ENTRIES_RQ2 1023 #define EHEA_DEF_ENTRIES_RQ3 1023 #else -#define EHEA_MAX_CQE_COUNT 32000 -#define EHEA_DEF_ENTRIES_SQ 16000 -#define EHEA_DEF_ENTRIES_RQ1 32080 -#define EHEA_DEF_ENTRIES_RQ2 4020 -#define EHEA_DEF_ENTRIES_RQ3 4020 +#define EHEA_MAX_CQE_COUNT 4080 +#define EHEA_DEF_ENTRIES_SQ 4080 +#define EHEA_DEF_ENTRIES_RQ1 8160 +#define EHEA_DEF_ENTRIES_RQ2 2040 +#define EHEA_DEF_ENTRIES_RQ3 2040 #endif #define EHEA_MAX_ENTRIES_EQ 20 diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index c6b3177..eb7d44d 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -766,7 +766,7 @@ static void ehea_parse_eqe(struct ehea_adapter *adapter, u64 eqe) if (EHEA_BMASK_GET(NEQE_PORT_UP, eqe)) { if (!netif_carrier_ok(port->netdev)) { ret = ehea_sense_port_attr( - adapter->port[portnum]); + port); if (ret) { ehea_error("failed resensing port " "attributes"); @@ -818,7 +818,7 @@ static void ehea_parse_eqe(struct ehea_adapter *adapter, u64 eqe) netif_stop_queue(port->netdev); break; default: - ehea_error("unknown event code %x", ec); + ehea_error("unknown event code %x, eqe=0x%lX", ec, eqe); break; } } @@ -1841,7 +1841,7 @@ static int ehea_start_xmit(struct sk_buff *skb, struct net_device *dev) if (netif_msg_tx_queued(port)) { ehea_info("post swqe on QP %d", pr->qp->init_attr.qp_nr); - ehea_dump(swqe, sizeof(*swqe), "swqe"); + ehea_dump(swqe, 512, "swqe"); } ehea_post_swqe(pr->qp, swqe); -- cgit v0.10.2 From 90f10841180e9b7938f63db69e90dacb7d21bbe5 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 15:56:04 -0500 Subject: [PATCH] powerpc/cell spidernet ethtool -i version number info. This patch adds version information as reported by ethtool -i to the Spidernet driver. From: James K Lewis Signed-off-by: James K Lewis Signed-off-by: Linas Vepstas Acked-by: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 46a0090..96dafb0 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -55,6 +55,7 @@ MODULE_AUTHOR("Utz Bacher and Jens Osterkamp " \ ""); MODULE_DESCRIPTION("Spider Southbridge Gigabit Ethernet driver"); MODULE_LICENSE("GPL"); +MODULE_VERSION(VERSION); static int rx_descriptors = SPIDER_NET_RX_DESCRIPTORS_DEFAULT; static int tx_descriptors = SPIDER_NET_TX_DESCRIPTORS_DEFAULT; @@ -2252,6 +2253,8 @@ static struct pci_driver spider_net_driver = { */ static int __init spider_net_init(void) { + printk(KERN_INFO "Spidernet version %s.\n", VERSION); + if (rx_descriptors < SPIDER_NET_RX_DESCRIPTORS_MIN) { rx_descriptors = SPIDER_NET_RX_DESCRIPTORS_MIN; pr_info("adjusting rx descriptors to %i.\n", rx_descriptors); diff --git a/drivers/net/spider_net.h b/drivers/net/spider_net.h index a59deda..6193ea8 100644 --- a/drivers/net/spider_net.h +++ b/drivers/net/spider_net.h @@ -24,6 +24,8 @@ #ifndef _SPIDER_NET_H #define _SPIDER_NET_H +#define VERSION "1.1 A" + #include "sungem_phy.h" extern int spider_net_stop(struct net_device *netdev); diff --git a/drivers/net/spider_net_ethtool.c b/drivers/net/spider_net_ethtool.c index 589e436..fda74f7 100644 --- a/drivers/net/spider_net_ethtool.c +++ b/drivers/net/spider_net_ethtool.c @@ -76,7 +76,7 @@ spider_net_ethtool_get_drvinfo(struct net_device *netdev, /* clear and fill out info */ memset(drvinfo, 0, sizeof(struct ethtool_drvinfo)); strncpy(drvinfo->driver, spider_net_driver_name, 32); - strncpy(drvinfo->version, "0.1", 32); + strncpy(drvinfo->version, VERSION, 32); strcpy(drvinfo->fw_version, "no information"); strncpy(drvinfo->bus_info, pci_name(card->pdev), 32); } -- cgit v0.10.2 From a02d601dd59d08a77563499cc05b48603df8f4a4 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 15:57:26 -0500 Subject: [PATCH] powerpc/cell spidernet burst alignment patch. This patch increases the Burst Address alignment from 64 to 1024 in the Spidernet driver. This improves transmit performance for large packets. From: James K Lewis Signed-off-by: James K Lewis Signed-off-by: Linas Vepstas Acked-by: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.h b/drivers/net/spider_net.h index 6193ea8..b2e3570 100644 --- a/drivers/net/spider_net.h +++ b/drivers/net/spider_net.h @@ -211,7 +211,7 @@ extern char spider_net_driver_name[]; #define SPIDER_NET_DMA_RX_FEND_VALUE 0x00030003 /* to set TX_DMA_EN */ #define SPIDER_NET_TX_DMA_EN 0x80000000 -#define SPIDER_NET_GDTDCEIDIS 0x00000002 +#define SPIDER_NET_GDTDCEIDIS 0x00000302 #define SPIDER_NET_DMA_TX_VALUE SPIDER_NET_TX_DMA_EN | \ SPIDER_NET_GDTDCEIDIS #define SPIDER_NET_DMA_TX_FEND_VALUE 0x00030003 -- cgit v0.10.2 From e2874f2e8c3695953b9ec26d396d678a7128ee64 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 15:59:02 -0500 Subject: [PATCH] Spidernet module parm permissions The module param permsissions should bw read-only, not writable. From: James K Lewis Signed-off-by: James K Lewis Signed-off-by: Linas Vepstas Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 96dafb0..ff409a1 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -60,8 +60,8 @@ MODULE_VERSION(VERSION); static int rx_descriptors = SPIDER_NET_RX_DESCRIPTORS_DEFAULT; static int tx_descriptors = SPIDER_NET_TX_DESCRIPTORS_DEFAULT; -module_param(rx_descriptors, int, 0644); -module_param(tx_descriptors, int, 0644); +module_param(rx_descriptors, int, 0444); +module_param(tx_descriptors, int, 0444); MODULE_PARM_DESC(rx_descriptors, "number of descriptors used " \ "in rx chains"); -- cgit v0.10.2 From c3fee4c55915c42b3278b65c91f9be8cee13426e Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:00:04 -0500 Subject: [PATCH] powerpc/cell spidernet force-end fix Bugfix: when cleaning up the transmit queue upon device close, be sure to walk the entire queue. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index ff409a1..52bf1b2 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -699,6 +699,8 @@ spider_net_release_tx_descr(struct spider_net_card *card) /* unmap the skb */ skb = descr->skb; + if (!skb) + return; pci_unmap_single(card->pdev, descr->buf_addr, skb->len, PCI_DMA_TODEVICE); dev_kfree_skb_any(skb); @@ -751,7 +753,8 @@ spider_net_release_tx_chain(struct spider_net_card *card, int brutal) default: card->netdev_stats.tx_dropped++; - return 1; + if (!brutal) + return 1; } spider_net_release_tx_descr(card); } -- cgit v0.10.2 From 808999c9a4749dc67c39bf52f712d0c27aa00e67 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:01:00 -0500 Subject: [PATCH] powerpc/cell spidernet zlen min packet length Polite device drivers pad short packets to 60 bytes, so that mean-spirited users don't accidentally DOS some other OS that can't handle short packets. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 52bf1b2..f5f7357 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -648,18 +648,26 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, { struct spider_net_descr *descr = card->tx_chain.head; dma_addr_t buf; + int length; - buf = pci_map_single(card->pdev, skb->data, skb->len, PCI_DMA_TODEVICE); + length = skb->len; + if (length < ETH_ZLEN) { + if (skb_pad(skb, ETH_ZLEN-length)) + return 0; + length = ETH_ZLEN; + } + + buf = pci_map_single(card->pdev, skb->data, length, PCI_DMA_TODEVICE); if (pci_dma_mapping_error(buf)) { if (netif_msg_tx_err(card) && net_ratelimit()) pr_err("could not iommu-map packet (%p, %i). " - "Dropping packet\n", skb->data, skb->len); + "Dropping packet\n", skb->data, length); card->spider_stats.tx_iommu_map_error++; return -ENOMEM; } descr->buf_addr = buf; - descr->buf_size = skb->len; + descr->buf_size = length; descr->next_descr_addr = 0; descr->skb = skb; descr->data_status = 0; @@ -693,6 +701,7 @@ spider_net_release_tx_descr(struct spider_net_card *card) { struct spider_net_descr *descr = card->tx_chain.tail; struct sk_buff *skb; + unsigned int len; card->tx_chain.tail = card->tx_chain.tail->next; descr->dmac_cmd_status |= SPIDER_NET_DESCR_NOT_IN_USE; @@ -701,7 +710,8 @@ spider_net_release_tx_descr(struct spider_net_card *card) skb = descr->skb; if (!skb) return; - pci_unmap_single(card->pdev, descr->buf_addr, skb->len, + len = skb->len < ETH_ZLEN ? ETH_ZLEN : skb->len; + pci_unmap_single(card->pdev, descr->buf_addr, len, PCI_DMA_TODEVICE); dev_kfree_skb_any(skb); } -- cgit v0.10.2 From 917a5b8e648f420105158023ae0317eb0e77a2d5 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:01:51 -0500 Subject: [PATCH] powerpc/cell spidernet add missing netdev watchdog Set the netdev watchdog timer. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index f5f7357..0aed693 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -686,6 +686,7 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, descr->prev->next_descr_addr = descr->bus_addr; + card->netdev->trans_start = jiffies; /* set netdev watchdog timer */ return 0; } -- cgit v0.10.2 From ded8028a0b61075d841c33a412da5c869140d7aa Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:02:54 -0500 Subject: [PATCH] Spidernet fix register field definitions This patch fixes the names of a few fields in the DMA control register. There is no functional change. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 0aed693..2f54cdd 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1614,7 +1614,7 @@ spider_net_enable_card(struct spider_net_card *card) SPIDER_NET_INT2_MASK_VALUE); spider_net_write_reg(card, SPIDER_NET_GDTDMACCNTR, - SPIDER_NET_GDTDCEIDIS); + SPIDER_NET_GDTBSTA | SPIDER_NET_GDTDCEIDIS); } /** diff --git a/drivers/net/spider_net.h b/drivers/net/spider_net.h index b2e3570..80f4d27 100644 --- a/drivers/net/spider_net.h +++ b/drivers/net/spider_net.h @@ -191,7 +191,9 @@ extern char spider_net_driver_name[]; #define SPIDER_NET_MACMODE_VALUE 0x00000001 #define SPIDER_NET_BURSTLMT_VALUE 0x00000200 /* about 16 us */ -/* 1(0) enable r/tx dma +/* DMAC control register GDMACCNTR + * + * 1(0) enable r/tx dma * 0000000 fixed to 0 * * 000000 fixed to 0 @@ -200,6 +202,7 @@ extern char spider_net_driver_name[]; * * 000000 fixed to 0 * 00 burst alignment: 128 bytes + * 11 burst alignment: 1024 bytes * * 00000 fixed to 0 * 0 descr writeback size 32 bytes @@ -210,10 +213,13 @@ extern char spider_net_driver_name[]; #define SPIDER_NET_DMA_RX_VALUE 0x80000000 #define SPIDER_NET_DMA_RX_FEND_VALUE 0x00030003 /* to set TX_DMA_EN */ -#define SPIDER_NET_TX_DMA_EN 0x80000000 -#define SPIDER_NET_GDTDCEIDIS 0x00000302 -#define SPIDER_NET_DMA_TX_VALUE SPIDER_NET_TX_DMA_EN | \ - SPIDER_NET_GDTDCEIDIS +#define SPIDER_NET_TX_DMA_EN 0x80000000 +#define SPIDER_NET_GDTBSTA 0x00000300 +#define SPIDER_NET_GDTDCEIDIS 0x00000002 +#define SPIDER_NET_DMA_TX_VALUE SPIDER_NET_TX_DMA_EN | \ + SPIDER_NET_GDTBSTA | \ + SPIDER_NET_GDTDCEIDIS + #define SPIDER_NET_DMA_TX_FEND_VALUE 0x00030003 /* SPIDER_NET_UA_DESCR_VALUE is OR'ed with the unicast address */ -- cgit v0.10.2 From 313ef4b76c96ef427a7613d89df550aa5d02bf21 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:04:00 -0500 Subject: [PATCH] Spidernet stop queue when queue is full. This patch adds a call to netif_stop_queue() when there is no more room for more packets on the transmit queue. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 2f54cdd..05bdd0b 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -823,39 +823,25 @@ spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) struct spider_net_descr_chain *chain = &card->tx_chain; struct spider_net_descr *descr = chain->head; unsigned long flags; - int result; spin_lock_irqsave(&chain->lock, flags); spider_net_release_tx_chain(card, 0); - if (chain->head->next == chain->tail->prev) { - card->netdev_stats.tx_dropped++; - result = NETDEV_TX_LOCKED; - goto out; - } + if ((chain->head->next == chain->tail->prev) || + (spider_net_get_descr_status(descr) != SPIDER_NET_DESCR_NOT_IN_USE) || + (spider_net_prepare_tx_descr(card, skb) != 0)) { - if (spider_net_get_descr_status(descr) != SPIDER_NET_DESCR_NOT_IN_USE) { card->netdev_stats.tx_dropped++; - result = NETDEV_TX_LOCKED; - goto out; + spin_unlock_irqrestore(&chain->lock, flags); + netif_stop_queue(netdev); + return NETDEV_TX_BUSY; } - if (spider_net_prepare_tx_descr(card, skb) != 0) { - card->netdev_stats.tx_dropped++; - result = NETDEV_TX_BUSY; - goto out; - } - - result = NETDEV_TX_OK; - spider_net_kick_tx_dma(card); card->tx_chain.head = card->tx_chain.head->next; - -out: spin_unlock_irqrestore(&chain->lock, flags); - netif_wake_queue(netdev); - return result; + return NETDEV_TX_OK; } /** @@ -874,9 +860,10 @@ spider_net_cleanup_tx_ring(struct spider_net_card *card) spin_lock_irqsave(&card->tx_chain.lock, flags); if ((spider_net_release_tx_chain(card, 0) != 0) && - (card->netdev->flags & IFF_UP)) + (card->netdev->flags & IFF_UP)) { spider_net_kick_tx_dma(card); - + netif_wake_queue(card->netdev); + } spin_unlock_irqrestore(&card->tx_chain.lock, flags); } -- cgit v0.10.2 From 43932d938d5a193bf9602b0ac8aa6783ba78b1aa Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:05:00 -0500 Subject: [PATCH] powerpc/cell spidernet bogus rx interrupt bit The current receive interrupt mask sets a bogus bit that doesn't even belong to the definition of this register. Remove it. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.h b/drivers/net/spider_net.h index 80f4d27..6c9d7ce 100644 --- a/drivers/net/spider_net.h +++ b/drivers/net/spider_net.h @@ -332,9 +332,8 @@ enum spider_net_int2_status { (1 << SPIDER_NET_GDTDCEINT) | \ (1 << SPIDER_NET_GDTFDCINT) ) -/* we rely on flagged descriptor interrupts*/ -#define SPIDER_NET_RXINT ( (1 << SPIDER_NET_GDAFDCINT) | \ - (1 << SPIDER_NET_GRMFLLINT) ) +/* We rely on flagged descriptor interrupts */ +#define SPIDER_NET_RXINT ( (1 << SPIDER_NET_GDAFDCINT) ) #define SPIDER_NET_ERRINT ( 0xffffffff & \ (~SPIDER_NET_TXINT) & \ -- cgit v0.10.2 From 37aad7500bf7064bf150ea1f234303f4173f7b24 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:06:53 -0500 Subject: [PATCH] powerpc/cell spidernet fix error interrupt print The print message associated with the descriptor chain end interrupt prints a bogs value. Fix that. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 05bdd0b..9d2ed04 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1356,7 +1356,7 @@ spider_net_handle_error_irq(struct spider_net_card *card, u32 status_reg) if (netif_msg_intr(card)) pr_err("got descriptor chain end interrupt, " "restarting DMAC %c.\n", - 'D'+i-SPIDER_NET_GDDDCEINT); + 'D'-(i-SPIDER_NET_GDDDCEINT)/3); spider_net_refill_rx_chain(card); spider_net_enable_rxdmac(card); show_error = 0; -- cgit v0.10.2 From 98b9040c747e50fe02ad616c9d5fee9aa4017cd1 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:08:42 -0500 Subject: [PATCH] powerpc/cell spidernet stop error printing patch. Turn off mis-interpretation of the queue-empty interrupt status bit as an error. Signed-off-by: Linas Vepstas Signed-off-by: James K Lewis Acked-by: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 9d2ed04..0eeff1a 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1245,12 +1245,15 @@ spider_net_handle_error_irq(struct spider_net_card *card, u32 status_reg) case SPIDER_NET_PHYINT: case SPIDER_NET_GMAC2INT: case SPIDER_NET_GMAC1INT: - case SPIDER_NET_GIPSINT: case SPIDER_NET_GFIFOINT: case SPIDER_NET_DMACINT: case SPIDER_NET_GSYSINT: break; */ + case SPIDER_NET_GIPSINT: + show_error = 0; + break; + case SPIDER_NET_GPWOPCMPINT: /* PHY write operation completed */ show_error = 0; @@ -1309,9 +1312,10 @@ spider_net_handle_error_irq(struct spider_net_card *card, u32 status_reg) case SPIDER_NET_GDTDCEINT: /* chain end. If a descriptor should be sent, kick off * tx dma - if (card->tx_chain.tail == card->tx_chain.head) + if (card->tx_chain.tail != card->tx_chain.head) spider_net_kick_tx_dma(card); - show_error = 0; */ + */ + show_error = 0; break; /* case SPIDER_NET_G1TMCNTINT: not used. print a message */ @@ -1425,8 +1429,9 @@ spider_net_handle_error_irq(struct spider_net_card *card, u32 status_reg) } if ((show_error) && (netif_msg_intr(card))) - pr_err("Got error interrupt, GHIINT0STS = 0x%08x, " + pr_err("Got error interrupt on %s, GHIINT0STS = 0x%08x, " "GHIINT1STS = 0x%08x, GHIINT2STS = 0x%08x\n", + card->netdev->name, status_reg, error_reg1, error_reg2); /* clear interrupt sources */ -- cgit v0.10.2 From b21606a773faffc2b3ec326325c433bdf37ecbdf Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:09:40 -0500 Subject: [PATCH] powerpc/cell spidernet incorrect offset Bugfix -- the rx chain is in memory after the tx chain -- the offset being used was wrong, resulting in memory corruption when the size of the rx and tx rings weren't exactly the same. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 0eeff1a..d779a0b 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1628,8 +1628,10 @@ spider_net_open(struct net_device *netdev) if (spider_net_init_chain(card, &card->tx_chain, card->descr, PCI_DMA_TODEVICE, card->tx_desc)) goto alloc_tx_failed; + + /* rx_chain is after tx_chain, so offset is descr + tx_count */ if (spider_net_init_chain(card, &card->rx_chain, - card->descr + card->rx_desc, + card->descr + card->tx_desc, PCI_DMA_FROMDEVICE, card->rx_desc)) goto alloc_rx_failed; -- cgit v0.10.2 From 204e5fa17c7ba45a89989f8da6dfe8e54d64b79b Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:11:33 -0500 Subject: [PATCH] powerpc/cell spidernet low watermark patch. Implement basic low-watermark support for the transmit queue. Hardware low-watermarks allow a properly configured kernel to continously stream data to a device and not have to handle any interrupts at all in doing so. Correct zero-interrupt operation can be actually observed for this driver, when the socket buffer is made large enough. The basic idea of a low-watermark interrupt is as follows. The device driver queues up a bunch of packets for the hardware to transmit, and then kicks the hardware to get it started. As the hardware drains the queue of pending, untransmitted packets, the device driver will want to know when the queue is almost empty, so that it can queue some more packets. If the queue drains down to the low waterark, then an interrupt will be generated. However, if the kernel/driver continues to add enough packets to keep the queue partially filled, no interrupt will actually be generated, and the hardware can continue streaming packets indefinitely in this mode. The impelmentation is done by setting the DESCR_TXDESFLG flag in one of the packets. When the hardware sees this flag, it will interrupt the device driver. Because this flag is on a fixed packet, rather than at fixed location in the queue, the code below needs to move the flag as more packets are queued up. This implementation attempts to keep the flag at about 1/4 from "empty". Signed-off-by: Linas Vepstas Signed-off-by: James K Lewis Acked-by: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index d779a0b..96b5d00 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -684,6 +684,7 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, break; } + /* Chain the bus address, so that the DMA engine finds this descr. */ descr->prev->next_descr_addr = descr->bus_addr; card->netdev->trans_start = jiffies; /* set netdev watchdog timer */ @@ -717,6 +718,41 @@ spider_net_release_tx_descr(struct spider_net_card *card) dev_kfree_skb_any(skb); } +static void +spider_net_set_low_watermark(struct spider_net_card *card) +{ + int status; + int cnt=0; + int i; + struct spider_net_descr *descr = card->tx_chain.tail; + + /* Measure the length of the queue. */ + while (descr != card->tx_chain.head) { + status = descr->dmac_cmd_status & SPIDER_NET_DESCR_NOT_IN_USE; + if (status == SPIDER_NET_DESCR_NOT_IN_USE) + break; + descr = descr->next; + cnt++; + } + + /* If TX queue is short, don't even bother with interrupts */ + if (cnt < card->tx_desc/4) + return; + + /* Set low-watermark 3/4th's of the way into the queue. */ + descr = card->tx_chain.tail; + cnt = (cnt*3)/4; + for (i=0;inext; + + /* Set the new watermark, clear the old watermark */ + descr->dmac_cmd_status |= SPIDER_NET_DESCR_TXDESFLG; + if (card->low_watermark && card->low_watermark != descr) + card->low_watermark->dmac_cmd_status = + card->low_watermark->dmac_cmd_status & ~SPIDER_NET_DESCR_TXDESFLG; + card->low_watermark = descr; +} + /** * spider_net_release_tx_chain - processes sent tx descriptors * @card: adapter structure @@ -838,6 +874,7 @@ spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) return NETDEV_TX_BUSY; } + spider_net_set_low_watermark(card); spider_net_kick_tx_dma(card); card->tx_chain.head = card->tx_chain.head->next; spin_unlock_irqrestore(&chain->lock, flags); @@ -1467,6 +1504,10 @@ spider_net_interrupt(int irq, void *ptr) spider_net_rx_irq_off(card); netif_rx_schedule(netdev); } + if (status_reg & SPIDER_NET_TXINT ) { + spider_net_cleanup_tx_ring(card); + netif_wake_queue(netdev); + } if (status_reg & SPIDER_NET_ERRINT ) spider_net_handle_error_irq(card, status_reg); @@ -1629,6 +1670,8 @@ spider_net_open(struct net_device *netdev) PCI_DMA_TODEVICE, card->tx_desc)) goto alloc_tx_failed; + card->low_watermark = NULL; + /* rx_chain is after tx_chain, so offset is descr + tx_count */ if (spider_net_init_chain(card, &card->rx_chain, card->descr + card->tx_desc, diff --git a/drivers/net/spider_net.h b/drivers/net/spider_net.h index 6c9d7ce..1f5c9dc 100644 --- a/drivers/net/spider_net.h +++ b/drivers/net/spider_net.h @@ -49,7 +49,7 @@ extern char spider_net_driver_name[]; #define SPIDER_NET_TX_DESCRIPTORS_MIN 16 #define SPIDER_NET_TX_DESCRIPTORS_MAX 512 -#define SPIDER_NET_TX_TIMER 20 +#define SPIDER_NET_TX_TIMER (HZ/5) #define SPIDER_NET_RX_CSUM_DEFAULT 1 @@ -328,9 +328,7 @@ enum spider_net_int2_status { SPIDER_NET_GRISPDNGINT }; -#define SPIDER_NET_TXINT ( (1 << SPIDER_NET_GTTEDINT) | \ - (1 << SPIDER_NET_GDTDCEINT) | \ - (1 << SPIDER_NET_GDTFDCINT) ) +#define SPIDER_NET_TXINT ( (1 << SPIDER_NET_GDTFDCINT) ) /* We rely on flagged descriptor interrupts */ #define SPIDER_NET_RXINT ( (1 << SPIDER_NET_GDAFDCINT) ) @@ -356,6 +354,7 @@ enum spider_net_int2_status { #define SPIDER_NET_DESCR_FORCE_END 0x50000000 /* used in rx and tx */ #define SPIDER_NET_DESCR_CARDOWNED 0xA0000000 /* used in rx and tx */ #define SPIDER_NET_DESCR_NOT_IN_USE 0xF0000000 +#define SPIDER_NET_DESCR_TXDESFLG 0x00800000 struct spider_net_descr { /* as defined by the hardware */ @@ -440,6 +439,7 @@ struct spider_net_card { struct spider_net_descr_chain tx_chain; struct spider_net_descr_chain rx_chain; + struct spider_net_descr *low_watermark; struct net_device_stats netdev_stats; -- cgit v0.10.2 From 68a8c609b3071c2441fa64f584d15311f2c10e61 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:13:05 -0500 Subject: [PATCH] powerpc/cell spidernet NAPI polling info. This patch moves transmit queue cleanup code out of the interrupt context, and into the NAPI polling routine. Signed-off-by: Linas Vepstas Acked-by: Arnd Bergmann Cc: James K Lewis Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 96b5d00..e429abc 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -715,7 +715,7 @@ spider_net_release_tx_descr(struct spider_net_card *card) len = skb->len < ETH_ZLEN ? ETH_ZLEN : skb->len; pci_unmap_single(card->pdev, descr->buf_addr, len, PCI_DMA_TODEVICE); - dev_kfree_skb_any(skb); + dev_kfree_skb(skb); } static void @@ -885,9 +885,10 @@ spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) * spider_net_cleanup_tx_ring - cleans up the TX ring * @card: card structure * - * spider_net_cleanup_tx_ring is called by the tx_timer (as we don't use - * interrupts to cleanup our TX ring) and returns sent packets to the stack - * by freeing them + * spider_net_cleanup_tx_ring is called by either the tx_timer + * or from the NAPI polling routine. + * This routine releases resources associted with transmitted + * packets, including updating the queue tail pointer. */ static void spider_net_cleanup_tx_ring(struct spider_net_card *card) @@ -1092,6 +1093,7 @@ spider_net_poll(struct net_device *netdev, int *budget) int packets_to_do, packets_done = 0; int no_more_packets = 0; + spider_net_cleanup_tx_ring(card); packets_to_do = min(*budget, netdev->quota); while (packets_to_do) { @@ -1504,10 +1506,8 @@ spider_net_interrupt(int irq, void *ptr) spider_net_rx_irq_off(card); netif_rx_schedule(netdev); } - if (status_reg & SPIDER_NET_TXINT ) { - spider_net_cleanup_tx_ring(card); - netif_wake_queue(netdev); - } + if (status_reg & SPIDER_NET_TXINT) + netif_rx_schedule(netdev); if (status_reg & SPIDER_NET_ERRINT ) spider_net_handle_error_irq(card, status_reg); -- cgit v0.10.2 From 9cc7bf7edf50a8a6b456b337aff97fe780ae369b Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:14:29 -0500 Subject: [PATCH] powerpc/cell spidernet refine locking The transmit side of the spider ethernet driver currently places locks around some very large chunks of code. This results in a fair amount of lock contention is some cases. This patch makes the locks much more fine-grained, protecting only the cirtical sections. One lock is used to protect three locations: the queue head and tail pointers, and the queue low-watermark location. Signed-off-by: Linas Vepstas Cc: Arnd Bergmann Cc: James K Lewis Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index e429abc..f8d7d0d 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -646,8 +646,9 @@ static int spider_net_prepare_tx_descr(struct spider_net_card *card, struct sk_buff *skb) { - struct spider_net_descr *descr = card->tx_chain.head; + struct spider_net_descr *descr; dma_addr_t buf; + unsigned long flags; int length; length = skb->len; @@ -666,6 +667,10 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, return -ENOMEM; } + spin_lock_irqsave(&card->tx_chain.lock, flags); + descr = card->tx_chain.head; + card->tx_chain.head = descr->next; + descr->buf_addr = buf; descr->buf_size = length; descr->next_descr_addr = 0; @@ -674,6 +679,8 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, descr->dmac_cmd_status = SPIDER_NET_DESCR_CARDOWNED | SPIDER_NET_DMAC_NOCS; + spin_unlock_irqrestore(&card->tx_chain.lock, flags); + if (skb->protocol == htons(ETH_P_IP)) switch (skb->nh.iph->protocol) { case IPPROTO_TCP: @@ -691,42 +698,17 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, return 0; } -/** - * spider_net_release_tx_descr - processes a used tx descriptor - * @card: card structure - * @descr: descriptor to release - * - * releases a used tx descriptor (unmapping, freeing of skb) - */ -static inline void -spider_net_release_tx_descr(struct spider_net_card *card) -{ - struct spider_net_descr *descr = card->tx_chain.tail; - struct sk_buff *skb; - unsigned int len; - - card->tx_chain.tail = card->tx_chain.tail->next; - descr->dmac_cmd_status |= SPIDER_NET_DESCR_NOT_IN_USE; - - /* unmap the skb */ - skb = descr->skb; - if (!skb) - return; - len = skb->len < ETH_ZLEN ? ETH_ZLEN : skb->len; - pci_unmap_single(card->pdev, descr->buf_addr, len, - PCI_DMA_TODEVICE); - dev_kfree_skb(skb); -} - static void spider_net_set_low_watermark(struct spider_net_card *card) { + unsigned long flags; int status; int cnt=0; int i; struct spider_net_descr *descr = card->tx_chain.tail; - /* Measure the length of the queue. */ + /* Measure the length of the queue. Measurement does not + * need to be precise -- does not need a lock. */ while (descr != card->tx_chain.head) { status = descr->dmac_cmd_status & SPIDER_NET_DESCR_NOT_IN_USE; if (status == SPIDER_NET_DESCR_NOT_IN_USE) @@ -746,11 +728,13 @@ spider_net_set_low_watermark(struct spider_net_card *card) descr = descr->next; /* Set the new watermark, clear the old watermark */ + spin_lock_irqsave(&card->tx_chain.lock, flags); descr->dmac_cmd_status |= SPIDER_NET_DESCR_TXDESFLG; if (card->low_watermark && card->low_watermark != descr) card->low_watermark->dmac_cmd_status = card->low_watermark->dmac_cmd_status & ~SPIDER_NET_DESCR_TXDESFLG; card->low_watermark = descr; + spin_unlock_irqrestore(&card->tx_chain.lock, flags); } /** @@ -769,21 +753,31 @@ static int spider_net_release_tx_chain(struct spider_net_card *card, int brutal) { struct spider_net_descr_chain *chain = &card->tx_chain; + struct spider_net_descr *descr; + struct sk_buff *skb; + u32 buf_addr; + unsigned long flags; int status; spider_net_read_reg(card, SPIDER_NET_GDTDMACCNTR); while (chain->tail != chain->head) { - status = spider_net_get_descr_status(chain->tail); + spin_lock_irqsave(&chain->lock, flags); + descr = chain->tail; + + status = spider_net_get_descr_status(descr); switch (status) { case SPIDER_NET_DESCR_COMPLETE: card->netdev_stats.tx_packets++; - card->netdev_stats.tx_bytes += chain->tail->skb->len; + card->netdev_stats.tx_bytes += descr->skb->len; break; case SPIDER_NET_DESCR_CARDOWNED: - if (!brutal) + if (!brutal) { + spin_unlock_irqrestore(&chain->lock, flags); return 1; + } + /* fallthrough, if we release the descriptors * brutally (then we don't care about * SPIDER_NET_DESCR_CARDOWNED) */ @@ -800,12 +794,25 @@ spider_net_release_tx_chain(struct spider_net_card *card, int brutal) default: card->netdev_stats.tx_dropped++; - if (!brutal) + if (!brutal) { + spin_unlock_irqrestore(&chain->lock, flags); return 1; + } } - spider_net_release_tx_descr(card); - } + chain->tail = descr->next; + descr->dmac_cmd_status |= SPIDER_NET_DESCR_NOT_IN_USE; + skb = descr->skb; + buf_addr = descr->buf_addr; + spin_unlock_irqrestore(&chain->lock, flags); + + /* unmap the skb */ + if (skb) { + int len = skb->len < ETH_ZLEN ? ETH_ZLEN : skb->len; + pci_unmap_single(card->pdev, buf_addr, len, PCI_DMA_TODEVICE); + dev_kfree_skb(skb); + } + } return 0; } @@ -857,27 +864,19 @@ spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) { struct spider_net_card *card = netdev_priv(netdev); struct spider_net_descr_chain *chain = &card->tx_chain; - struct spider_net_descr *descr = chain->head; - unsigned long flags; - - spin_lock_irqsave(&chain->lock, flags); spider_net_release_tx_chain(card, 0); if ((chain->head->next == chain->tail->prev) || - (spider_net_get_descr_status(descr) != SPIDER_NET_DESCR_NOT_IN_USE) || (spider_net_prepare_tx_descr(card, skb) != 0)) { card->netdev_stats.tx_dropped++; - spin_unlock_irqrestore(&chain->lock, flags); netif_stop_queue(netdev); return NETDEV_TX_BUSY; } spider_net_set_low_watermark(card); spider_net_kick_tx_dma(card); - card->tx_chain.head = card->tx_chain.head->next; - spin_unlock_irqrestore(&chain->lock, flags); return NETDEV_TX_OK; } @@ -893,16 +892,11 @@ spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) static void spider_net_cleanup_tx_ring(struct spider_net_card *card) { - unsigned long flags; - - spin_lock_irqsave(&card->tx_chain.lock, flags); - if ((spider_net_release_tx_chain(card, 0) != 0) && (card->netdev->flags & IFF_UP)) { spider_net_kick_tx_dma(card); netif_wake_queue(card->netdev); } - spin_unlock_irqrestore(&card->tx_chain.lock, flags); } /** @@ -1930,10 +1924,7 @@ spider_net_stop(struct net_device *netdev) spider_net_disable_rxdmac(card); /* release chains */ - if (spin_trylock(&card->tx_chain.lock)) { - spider_net_release_tx_chain(card, 1); - spin_unlock(&card->tx_chain.lock); - } + spider_net_release_tx_chain(card, 1); spider_net_free_chain(card, &card->tx_chain); spider_net_free_chain(card, &card->rx_chain); -- cgit v0.10.2 From 499eea18722e43f0ee15e11ac16ffcbd10b70b24 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:15:29 -0500 Subject: [PATCH] powerpc/cell spidernet Remove a dummy register read that is not needed. This reduces CPU usage notably during transmit. Signed-off-by: Linas Vepstas Cc: Arnd Bergmann Cc: James K Lewis Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index f8d7d0d..6bb8f1f 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -759,8 +759,6 @@ spider_net_release_tx_chain(struct spider_net_card *card, int brutal) unsigned long flags; int status; - spider_net_read_reg(card, SPIDER_NET_GDTDMACCNTR); - while (chain->tail != chain->head) { spin_lock_irqsave(&chain->lock, flags); descr = chain->tail; -- cgit v0.10.2 From a664ccf430547696951bf3949f5a2de5079ece5a Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:18:18 -0500 Subject: [PATCH] powerpc/cell spidernet reduce DMA kicking The current code attempts to start the TX dma every time a packet is queued. This is too conservative, and wastes CPU time. This patch changes behaviour to call the kick-dma function less often, only when the tx queue is at risk of emptying. This reduces cpu usage, improves performance. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 6bb8f1f..caa11c6 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -698,7 +698,7 @@ spider_net_prepare_tx_descr(struct spider_net_card *card, return 0; } -static void +static int spider_net_set_low_watermark(struct spider_net_card *card) { unsigned long flags; @@ -719,7 +719,7 @@ spider_net_set_low_watermark(struct spider_net_card *card) /* If TX queue is short, don't even bother with interrupts */ if (cnt < card->tx_desc/4) - return; + return cnt; /* Set low-watermark 3/4th's of the way into the queue. */ descr = card->tx_chain.tail; @@ -735,6 +735,7 @@ spider_net_set_low_watermark(struct spider_net_card *card) card->low_watermark->dmac_cmd_status & ~SPIDER_NET_DESCR_TXDESFLG; card->low_watermark = descr; spin_unlock_irqrestore(&card->tx_chain.lock, flags); + return cnt; } /** @@ -819,8 +820,12 @@ spider_net_release_tx_chain(struct spider_net_card *card, int brutal) * @card: card structure * @descr: descriptor address to enable TX processing at * - * spider_net_kick_tx_dma writes the current tx chain head as start address - * of the tx descriptor chain and enables the transmission DMA engine + * This routine will start the transmit DMA running if + * it is not already running. This routine ned only be + * called when queueing a new packet to an empty tx queue. + * Writes the current tx chain head as start address + * of the tx descriptor chain and enables the transmission + * DMA engine. */ static inline void spider_net_kick_tx_dma(struct spider_net_card *card) @@ -860,6 +865,7 @@ out: static int spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) { + int cnt; struct spider_net_card *card = netdev_priv(netdev); struct spider_net_descr_chain *chain = &card->tx_chain; @@ -873,8 +879,9 @@ spider_net_xmit(struct sk_buff *skb, struct net_device *netdev) return NETDEV_TX_BUSY; } - spider_net_set_low_watermark(card); - spider_net_kick_tx_dma(card); + cnt = spider_net_set_low_watermark(card); + if (cnt < 5) + spider_net_kick_tx_dma(card); return NETDEV_TX_OK; } -- cgit v0.10.2 From 66c097165cf6d4196e798145fb33c768164fb361 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:19:34 -0500 Subject: [PATCH] powerpc/cell spidernet variable name change Cosmetic patch: give the variable holding the numer of descriptors a more descriptive name, so to avoid confusion. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index caa11c6..88e7e41 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -718,7 +718,7 @@ spider_net_set_low_watermark(struct spider_net_card *card) } /* If TX queue is short, don't even bother with interrupts */ - if (cnt < card->tx_desc/4) + if (cnt < card->num_tx_desc/4) return cnt; /* Set low-watermark 3/4th's of the way into the queue. */ @@ -1666,15 +1666,15 @@ spider_net_open(struct net_device *netdev) result = -ENOMEM; if (spider_net_init_chain(card, &card->tx_chain, card->descr, - PCI_DMA_TODEVICE, card->tx_desc)) + PCI_DMA_TODEVICE, card->num_tx_desc)) goto alloc_tx_failed; card->low_watermark = NULL; /* rx_chain is after tx_chain, so offset is descr + tx_count */ if (spider_net_init_chain(card, &card->rx_chain, - card->descr + card->tx_desc, - PCI_DMA_FROMDEVICE, card->rx_desc)) + card->descr + card->num_tx_desc, + PCI_DMA_FROMDEVICE, card->num_rx_desc)) goto alloc_rx_failed; /* allocate rx skbs */ @@ -2060,8 +2060,8 @@ spider_net_setup_netdev(struct spider_net_card *card) card->options.rx_csum = SPIDER_NET_RX_CSUM_DEFAULT; - card->tx_desc = tx_descriptors; - card->rx_desc = rx_descriptors; + card->num_tx_desc = tx_descriptors; + card->num_rx_desc = rx_descriptors; spider_net_setup_netdev_ops(netdev); diff --git a/drivers/net/spider_net.h b/drivers/net/spider_net.h index 1f5c9dc..b3b4611 100644 --- a/drivers/net/spider_net.h +++ b/drivers/net/spider_net.h @@ -455,8 +455,8 @@ struct spider_net_card { /* for ethtool */ int msg_enable; - int rx_desc; - int tx_desc; + int num_rx_desc; + int num_tx_desc; struct spider_net_extra_stats spider_stats; struct spider_net_descr descr[0]; diff --git a/drivers/net/spider_net_ethtool.c b/drivers/net/spider_net_ethtool.c index fda74f7..91b9951 100644 --- a/drivers/net/spider_net_ethtool.c +++ b/drivers/net/spider_net_ethtool.c @@ -158,9 +158,9 @@ spider_net_ethtool_get_ringparam(struct net_device *netdev, struct spider_net_card *card = netdev->priv; ering->tx_max_pending = SPIDER_NET_TX_DESCRIPTORS_MAX; - ering->tx_pending = card->tx_desc; + ering->tx_pending = card->num_tx_desc; ering->rx_max_pending = SPIDER_NET_RX_DESCRIPTORS_MAX; - ering->rx_pending = card->rx_desc; + ering->rx_pending = card->num_rx_desc; } static int spider_net_get_stats_count(struct net_device *netdev) -- cgit v0.10.2 From 348bc2a6e306dc3e875cee3389e1405963ace617 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:21:10 -0500 Subject: [PATCH] powerpc/cell spidernet DMA direction fix The ring buffer descriptors are DMA-accessed bidirectionally, but are not declared in this way. Fix this. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 88e7e41..a26dd1c 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -301,7 +301,7 @@ static int spider_net_init_chain(struct spider_net_card *card, struct spider_net_descr_chain *chain, struct spider_net_descr *start_descr, - int direction, int no) + int no) { int i; struct spider_net_descr *descr; @@ -316,7 +316,7 @@ spider_net_init_chain(struct spider_net_card *card, buf = pci_map_single(card->pdev, descr, SPIDER_NET_DESCR_SIZE, - direction); + PCI_DMA_BIDIRECTIONAL); if (pci_dma_mapping_error(buf)) goto iommu_error; @@ -330,11 +330,6 @@ spider_net_init_chain(struct spider_net_card *card, (descr-1)->next = start_descr; start_descr->prev = descr-1; - descr = start_descr; - if (direction == PCI_DMA_FROMDEVICE) - for (i=0; i < no; i++, descr++) - descr->next_descr_addr = descr->next->bus_addr; - spin_lock_init(&chain->lock); chain->head = start_descr; chain->tail = start_descr; @@ -347,7 +342,7 @@ iommu_error: if (descr->bus_addr) pci_unmap_single(card->pdev, descr->bus_addr, SPIDER_NET_DESCR_SIZE, - direction); + PCI_DMA_BIDIRECTIONAL); return -ENOMEM; } @@ -368,7 +363,7 @@ spider_net_free_rx_chain_contents(struct spider_net_card *card) dev_kfree_skb(descr->skb); pci_unmap_single(card->pdev, descr->buf_addr, SPIDER_NET_MAX_FRAME, - PCI_DMA_FROMDEVICE); + PCI_DMA_BIDIRECTIONAL); } descr = descr->next; } @@ -1662,21 +1657,26 @@ int spider_net_open(struct net_device *netdev) { struct spider_net_card *card = netdev_priv(netdev); - int result; + struct spider_net_descr *descr; + int i, result; result = -ENOMEM; if (spider_net_init_chain(card, &card->tx_chain, card->descr, - PCI_DMA_TODEVICE, card->num_tx_desc)) + card->num_tx_desc)) goto alloc_tx_failed; card->low_watermark = NULL; /* rx_chain is after tx_chain, so offset is descr + tx_count */ if (spider_net_init_chain(card, &card->rx_chain, - card->descr + card->num_tx_desc, - PCI_DMA_FROMDEVICE, card->num_rx_desc)) + card->descr + card->num_tx_desc, + card->num_rx_desc)) goto alloc_rx_failed; + descr = card->rx_chain.head; + for (i=0; i < card->num_rx_desc; i++, descr++) + descr->next_descr_addr = descr->next->bus_addr; + /* allocate rx skbs */ if (spider_net_alloc_rx_skbs(card)) goto alloc_skbs_failed; -- cgit v0.10.2 From 647519100105fb4ddfe6455e820093999c08c4be Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 10 Oct 2006 16:22:29 -0500 Subject: [PATCH] powerpc/cell spidernet release all descrs Bugfix: rx descriptor release function fails to visit the last entry while walking receive descriptor ring. Signed-off-by: Linas Vepstas Cc: James K Lewis Cc: Arnd Bergmann Signed-off-by: Jeff Garzik diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index a26dd1c..418138d 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -358,7 +358,7 @@ spider_net_free_rx_chain_contents(struct spider_net_card *card) struct spider_net_descr *descr; descr = card->rx_chain.head; - while (descr->next != card->rx_chain.head) { + do { if (descr->skb) { dev_kfree_skb(descr->skb); pci_unmap_single(card->pdev, descr->buf_addr, @@ -366,7 +366,7 @@ spider_net_free_rx_chain_contents(struct spider_net_card *card) PCI_DMA_BIDIRECTIONAL); } descr = descr->next; - } + } while (descr != card->rx_chain.head); } /** -- cgit v0.10.2 From a1bc9b875be597cdf147db2748ba7ddc6b0f0fbe Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 5 Oct 2006 15:49:50 -0700 Subject: [PATCH] skge: fix stuck irq when fiber down The PHY interrupt from the internal fiber is getting stuck on when the link is down. Add code to handle the transition and mask it. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/skge.c b/drivers/net/skge.c index a4a58e4..5776493 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -884,6 +884,29 @@ static void skge_link_down(struct skge_port *skge) printk(KERN_INFO PFX "%s: Link is down.\n", skge->netdev->name); } + +static void xm_link_down(struct skge_hw *hw, int port) +{ + struct net_device *dev = hw->dev[port]; + struct skge_port *skge = netdev_priv(dev); + u16 cmd, msk; + + if (hw->phy_type == SK_PHY_XMAC) { + msk = xm_read16(hw, port, XM_IMSK); + msk |= XM_IS_INP_ASS | XM_IS_LIPA_RC | XM_IS_RX_PAGE | XM_IS_AND; + xm_write16(hw, port, XM_IMSK, msk); + } + + cmd = xm_read16(hw, port, XM_MMU_CMD); + cmd &= ~(XM_MMU_ENA_RX | XM_MMU_ENA_TX); + xm_write16(hw, port, XM_MMU_CMD, cmd); + /* dummy read to ensure writing */ + (void) xm_read16(hw, port, XM_MMU_CMD); + + if (netif_carrier_ok(dev)) + skge_link_down(skge); +} + static int __xm_phy_read(struct skge_hw *hw, int port, u16 reg, u16 *val) { int i; @@ -1008,14 +1031,7 @@ static void bcom_check_link(struct skge_hw *hw, int port) status = xm_phy_read(hw, port, PHY_BCOM_STAT); if ((status & PHY_ST_LSYNC) == 0) { - u16 cmd = xm_read16(hw, port, XM_MMU_CMD); - cmd &= ~(XM_MMU_ENA_RX | XM_MMU_ENA_TX); - xm_write16(hw, port, XM_MMU_CMD, cmd); - /* dummy read to ensure writing */ - (void) xm_read16(hw, port, XM_MMU_CMD); - - if (netif_carrier_ok(dev)) - skge_link_down(skge); + xm_link_down(hw, port); return; } @@ -1235,14 +1251,7 @@ static void xm_check_link(struct net_device *dev) status = xm_phy_read(hw, port, PHY_XMAC_STAT); if ((status & PHY_ST_LSYNC) == 0) { - u16 cmd = xm_read16(hw, port, XM_MMU_CMD); - cmd &= ~(XM_MMU_ENA_RX | XM_MMU_ENA_TX); - xm_write16(hw, port, XM_MMU_CMD, cmd); - /* dummy read to ensure writing */ - (void) xm_read16(hw, port, XM_MMU_CMD); - - if (netif_carrier_ok(dev)) - skge_link_down(skge); + xm_link_down(hw, port); return; } @@ -1568,6 +1577,10 @@ static void genesis_mac_intr(struct skge_hw *hw, int port) printk(KERN_DEBUG PFX "%s: mac interrupt status 0x%x\n", skge->netdev->name, status); + if (hw->phy_type == SK_PHY_XMAC && + (status & (XM_IS_INP_ASS | XM_IS_LIPA_RC))) + xm_link_down(hw, port); + if (status & XM_IS_TXF_UR) { xm_write32(hw, port, XM_MODE, XM_MD_FTF); ++skge->net_stats.tx_fifo_errors; @@ -1582,7 +1595,7 @@ static void genesis_link_up(struct skge_port *skge) { struct skge_hw *hw = skge->hw; int port = skge->port; - u16 cmd; + u16 cmd, msk; u32 mode; cmd = xm_read16(hw, port, XM_MMU_CMD); @@ -1631,7 +1644,11 @@ static void genesis_link_up(struct skge_port *skge) } xm_write32(hw, port, XM_MODE, mode); - xm_write16(hw, port, XM_IMSK, XM_DEF_MSK); + msk = XM_DEF_MSK; + if (hw->phy_type != SK_PHY_XMAC) + msk |= XM_IS_INP_ASS; /* disable GP0 interrupt bit */ + + xm_write16(hw, port, XM_IMSK, msk); xm_read16(hw, port, XM_ISRC); /* get MMU Command Reg. */ diff --git a/drivers/net/skge.h b/drivers/net/skge.h index d0b47d4..9cc955c 100644 --- a/drivers/net/skge.h +++ b/drivers/net/skge.h @@ -2195,7 +2195,8 @@ enum { XM_IS_RX_COMP = 1<<0, /* Bit 0: Frame Rx Complete */ }; -#define XM_DEF_MSK (~(XM_IS_RXC_OV | XM_IS_TXC_OV | XM_IS_RXF_OV | XM_IS_TXF_UR)) +#define XM_DEF_MSK (~(XM_IS_INP_ASS | XM_IS_LIPA_RC | \ + XM_IS_RXF_OV | XM_IS_TXF_UR)) /* XM_HW_CFG 16 bit r/w Hardware Config Register */ -- cgit v0.10.2 From 4b67be999ed5bfb1bfe4cc502d37d59b4f6b6b7f Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 5 Oct 2006 15:49:51 -0700 Subject: [PATCH] skge: pause mapping for fiber Do correct mapping of pause and duplex when using 1000BaseX fiber versions of the board. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/skge.c b/drivers/net/skge.c index 5776493..c19f49b 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -197,8 +197,8 @@ static u32 skge_supported_modes(const struct skge_hw *hw) else if (hw->chip_id == CHIP_ID_YUKON) supported &= ~SUPPORTED_1000baseT_Half; } else - supported = SUPPORTED_1000baseT_Full | SUPPORTED_FIBRE - | SUPPORTED_Autoneg; + supported = SUPPORTED_1000baseT_Full | SUPPORTED_1000baseT_Half + | SUPPORTED_FIBRE | SUPPORTED_Autoneg; return supported; } @@ -1018,6 +1018,14 @@ static const u16 phy_pause_map[] = { [FLOW_MODE_REM_SEND] = PHY_AN_PAUSE_CAP | PHY_AN_PAUSE_ASYM, }; +/* special defines for FIBER (88E1011S only) */ +static const u16 fiber_pause_map[] = { + [FLOW_MODE_NONE] = PHY_X_P_NO_PAUSE, + [FLOW_MODE_LOC_SEND] = PHY_X_P_ASYM_MD, + [FLOW_MODE_SYMMETRIC] = PHY_X_P_SYM_MD, + [FLOW_MODE_REM_SEND] = PHY_X_P_BOTH_MD, +}; + /* Check status of Broadcom phy link */ static void bcom_check_link(struct skge_hw *hw, int port) @@ -1207,17 +1215,7 @@ static void xm_phy_init(struct skge_port *skge) if (skge->advertising & ADVERTISED_1000baseT_Full) ctrl |= PHY_X_AN_FD; - switch(skge->flow_control) { - case FLOW_MODE_NONE: - ctrl |= PHY_X_P_NO_PAUSE; - break; - case FLOW_MODE_LOC_SEND: - ctrl |= PHY_X_P_ASYM_MD; - break; - case FLOW_MODE_SYMMETRIC: - ctrl |= PHY_X_P_BOTH_MD; - break; - } + ctrl |= fiber_pause_map[skge->flow_control]; xm_phy_write(hw, port, PHY_XMAC_AUNE_ADV, ctrl); @@ -1796,11 +1794,17 @@ static void yukon_init(struct skge_hw *hw, int port) adv |= PHY_M_AN_10_FD; if (skge->advertising & ADVERTISED_10baseT_Half) adv |= PHY_M_AN_10_HD; - } else /* special defines for FIBER (88E1011S only) */ - adv |= PHY_M_AN_1000X_AHD | PHY_M_AN_1000X_AFD; - /* Set Flow-control capabilities */ - adv |= phy_pause_map[skge->flow_control]; + /* Set Flow-control capabilities */ + adv |= phy_pause_map[skge->flow_control]; + } else { + if (skge->advertising & ADVERTISED_1000baseT_Full) + adv |= PHY_M_AN_1000X_AFD; + if (skge->advertising & ADVERTISED_1000baseT_Half) + adv |= PHY_M_AN_1000X_AHD; + + adv |= fiber_pause_map[skge->flow_control]; + } /* Restart Auto-negotiation */ ctrl |= PHY_CT_ANE | PHY_CT_RE_CFG; -- cgit v0.10.2 From 5d5c8e03786691d0d083142b922edce8609c0fd5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 5 Oct 2006 15:49:52 -0700 Subject: [PATCH] skge: better flow control negotiation Do flow control negotiation properly. Don't let auto negotiation status limit renegotiation. Separate desired pause values from the result of auto negotiation. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/skge.c b/drivers/net/skge.c index c19f49b..d844a83 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -487,31 +487,37 @@ static void skge_get_pauseparam(struct net_device *dev, { struct skge_port *skge = netdev_priv(dev); - ecmd->tx_pause = (skge->flow_control == FLOW_MODE_LOC_SEND) - || (skge->flow_control == FLOW_MODE_SYMMETRIC); - ecmd->rx_pause = (skge->flow_control == FLOW_MODE_REM_SEND) - || (skge->flow_control == FLOW_MODE_SYMMETRIC); + ecmd->rx_pause = (skge->flow_control == FLOW_MODE_SYMMETRIC) + || (skge->flow_control == FLOW_MODE_SYM_OR_REM); + ecmd->tx_pause = ecmd->rx_pause || (skge->flow_control == FLOW_MODE_LOC_SEND); - ecmd->autoneg = skge->autoneg; + ecmd->autoneg = ecmd->rx_pause || ecmd->tx_pause; } static int skge_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *ecmd) { struct skge_port *skge = netdev_priv(dev); + struct ethtool_pauseparam old; - skge->autoneg = ecmd->autoneg; - if (ecmd->rx_pause && ecmd->tx_pause) - skge->flow_control = FLOW_MODE_SYMMETRIC; - else if (ecmd->rx_pause && !ecmd->tx_pause) - skge->flow_control = FLOW_MODE_REM_SEND; - else if (!ecmd->rx_pause && ecmd->tx_pause) - skge->flow_control = FLOW_MODE_LOC_SEND; - else - skge->flow_control = FLOW_MODE_NONE; + skge_get_pauseparam(dev, &old); + + if (ecmd->autoneg != old.autoneg) + skge->flow_control = ecmd->autoneg ? FLOW_MODE_NONE : FLOW_MODE_SYMMETRIC; + else { + if (ecmd->rx_pause && ecmd->tx_pause) + skge->flow_control = FLOW_MODE_SYMMETRIC; + else if (ecmd->rx_pause && !ecmd->tx_pause) + skge->flow_control = FLOW_MODE_SYM_OR_REM; + else if (!ecmd->rx_pause && ecmd->tx_pause) + skge->flow_control = FLOW_MODE_LOC_SEND; + else + skge->flow_control = FLOW_MODE_NONE; + } if (netif_running(dev)) skge_phy_reset(skge); + return 0; } @@ -854,6 +860,23 @@ static int skge_rx_fill(struct net_device *dev) return 0; } +static const char *skge_pause(enum pause_status status) +{ + switch(status) { + case FLOW_STAT_NONE: + return "none"; + case FLOW_STAT_REM_SEND: + return "rx only"; + case FLOW_STAT_LOC_SEND: + return "tx_only"; + case FLOW_STAT_SYMMETRIC: /* Both station may send PAUSE */ + return "both"; + default: + return "indeterminated"; + } +} + + static void skge_link_up(struct skge_port *skge) { skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), @@ -862,16 +885,13 @@ static void skge_link_up(struct skge_port *skge) netif_carrier_on(skge->netdev); netif_wake_queue(skge->netdev); - if (netif_msg_link(skge)) + if (netif_msg_link(skge)) { printk(KERN_INFO PFX "%s: Link is up at %d Mbps, %s duplex, flow control %s\n", skge->netdev->name, skge->speed, skge->duplex == DUPLEX_FULL ? "full" : "half", - (skge->flow_control == FLOW_MODE_NONE) ? "none" : - (skge->flow_control == FLOW_MODE_LOC_SEND) ? "tx only" : - (skge->flow_control == FLOW_MODE_REM_SEND) ? "rx only" : - (skge->flow_control == FLOW_MODE_SYMMETRIC) ? "tx and rx" : - "unknown"); + skge_pause(skge->flow_status)); + } } static void skge_link_down(struct skge_port *skge) @@ -1015,7 +1035,7 @@ static const u16 phy_pause_map[] = { [FLOW_MODE_NONE] = 0, [FLOW_MODE_LOC_SEND] = PHY_AN_PAUSE_ASYM, [FLOW_MODE_SYMMETRIC] = PHY_AN_PAUSE_CAP, - [FLOW_MODE_REM_SEND] = PHY_AN_PAUSE_CAP | PHY_AN_PAUSE_ASYM, + [FLOW_MODE_SYM_OR_REM] = PHY_AN_PAUSE_CAP | PHY_AN_PAUSE_ASYM, }; /* special defines for FIBER (88E1011S only) */ @@ -1023,7 +1043,7 @@ static const u16 fiber_pause_map[] = { [FLOW_MODE_NONE] = PHY_X_P_NO_PAUSE, [FLOW_MODE_LOC_SEND] = PHY_X_P_ASYM_MD, [FLOW_MODE_SYMMETRIC] = PHY_X_P_SYM_MD, - [FLOW_MODE_REM_SEND] = PHY_X_P_BOTH_MD, + [FLOW_MODE_SYM_OR_REM] = PHY_X_P_BOTH_MD, }; @@ -1072,20 +1092,19 @@ static void bcom_check_link(struct skge_hw *hw, int port) return; } - /* We are using IEEE 802.3z/D5.0 Table 37-4 */ switch (aux & PHY_B_AS_PAUSE_MSK) { case PHY_B_AS_PAUSE_MSK: - skge->flow_control = FLOW_MODE_SYMMETRIC; + skge->flow_status = FLOW_STAT_SYMMETRIC; break; case PHY_B_AS_PRR: - skge->flow_control = FLOW_MODE_REM_SEND; + skge->flow_status = FLOW_STAT_REM_SEND; break; case PHY_B_AS_PRT: - skge->flow_control = FLOW_MODE_LOC_SEND; + skge->flow_status = FLOW_STAT_LOC_SEND; break; default: - skge->flow_control = FLOW_MODE_NONE; + skge->flow_status = FLOW_STAT_NONE; } skge->speed = SPEED_1000; } @@ -1283,15 +1302,20 @@ static void xm_check_link(struct net_device *dev) } /* We are using IEEE 802.3z/D5.0 Table 37-4 */ - if (lpa & PHY_X_P_SYM_MD) - skge->flow_control = FLOW_MODE_SYMMETRIC; - else if ((lpa & PHY_X_RS_PAUSE) == PHY_X_P_ASYM_MD) - skge->flow_control = FLOW_MODE_REM_SEND; - else if ((lpa & PHY_X_RS_PAUSE) == PHY_X_P_BOTH_MD) - skge->flow_control = FLOW_MODE_LOC_SEND; + if ((skge->flow_control == FLOW_MODE_SYMMETRIC || + skge->flow_control == FLOW_MODE_SYM_OR_REM) && + (lpa & PHY_X_P_SYM_MD)) + skge->flow_status = FLOW_STAT_SYMMETRIC; + else if (skge->flow_control == FLOW_MODE_SYM_OR_REM && + (lpa & PHY_X_RS_PAUSE) == PHY_X_P_ASYM_MD) + /* Enable PAUSE receive, disable PAUSE transmit */ + skge->flow_status = FLOW_STAT_REM_SEND; + else if (skge->flow_control == FLOW_MODE_LOC_SEND && + (lpa & PHY_X_RS_PAUSE) == PHY_X_P_BOTH_MD) + /* Disable PAUSE receive, enable PAUSE transmit */ + skge->flow_status = FLOW_STAT_LOC_SEND; else - skge->flow_control = FLOW_MODE_NONE; - + skge->flow_status = FLOW_STAT_NONE; skge->speed = SPEED_1000; } @@ -1602,8 +1626,8 @@ static void genesis_link_up(struct skge_port *skge) * enabling pause frame reception is required for 1000BT * because the XMAC is not reset if the link is going down */ - if (skge->flow_control == FLOW_MODE_NONE || - skge->flow_control == FLOW_MODE_LOC_SEND) + if (skge->flow_status == FLOW_STAT_NONE || + skge->flow_status == FLOW_STAT_LOC_SEND) /* Disable Pause Frame Reception */ cmd |= XM_MMU_IGN_PF; else @@ -1613,8 +1637,8 @@ static void genesis_link_up(struct skge_port *skge) xm_write16(hw, port, XM_MMU_CMD, cmd); mode = xm_read32(hw, port, XM_MODE); - if (skge->flow_control == FLOW_MODE_SYMMETRIC || - skge->flow_control == FLOW_MODE_LOC_SEND) { + if (skge->flow_status== FLOW_STAT_SYMMETRIC || + skge->flow_status == FLOW_STAT_LOC_SEND) { /* * Configure Pause Frame Generation * Use internal and external Pause Frame Generation. @@ -1938,6 +1962,11 @@ static void yukon_mac_init(struct skge_hw *hw, int port) case FLOW_MODE_LOC_SEND: /* disable Rx flow-control */ reg |= GM_GPCR_FC_RX_DIS | GM_GPCR_AU_FCT_DIS; + break; + case FLOW_MODE_SYMMETRIC: + case FLOW_MODE_SYM_OR_REM: + /* enable Tx & Rx flow-control */ + break; } gma_write16(hw, port, GM_GP_CTRL, reg); @@ -2132,13 +2161,11 @@ static void yukon_link_down(struct skge_port *skge) ctrl &= ~(GM_GPCR_RX_ENA | GM_GPCR_TX_ENA); gma_write16(hw, port, GM_GP_CTRL, ctrl); - if (skge->flow_control == FLOW_MODE_REM_SEND) { + if (skge->flow_status == FLOW_STAT_REM_SEND) { + ctrl = gm_phy_read(hw, port, PHY_MARV_AUNE_ADV); + ctrl |= PHY_M_AN_ASP; /* restore Asymmetric Pause bit */ - gm_phy_write(hw, port, PHY_MARV_AUNE_ADV, - gm_phy_read(hw, port, - PHY_MARV_AUNE_ADV) - | PHY_M_AN_ASP); - + gm_phy_write(hw, port, PHY_MARV_AUNE_ADV, ctrl); } yukon_reset(hw, port); @@ -2185,19 +2212,19 @@ static void yukon_phy_intr(struct skge_port *skge) /* We are using IEEE 802.3z/D5.0 Table 37-4 */ switch (phystat & PHY_M_PS_PAUSE_MSK) { case PHY_M_PS_PAUSE_MSK: - skge->flow_control = FLOW_MODE_SYMMETRIC; + skge->flow_status = FLOW_STAT_SYMMETRIC; break; case PHY_M_PS_RX_P_EN: - skge->flow_control = FLOW_MODE_REM_SEND; + skge->flow_status = FLOW_STAT_REM_SEND; break; case PHY_M_PS_TX_P_EN: - skge->flow_control = FLOW_MODE_LOC_SEND; + skge->flow_status = FLOW_STAT_LOC_SEND; break; default: - skge->flow_control = FLOW_MODE_NONE; + skge->flow_status = FLOW_STAT_NONE; } - if (skge->flow_control == FLOW_MODE_NONE || + if (skge->flow_status == FLOW_STAT_NONE || (skge->speed < SPEED_1000 && skge->duplex == DUPLEX_HALF)) skge_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_OFF); else @@ -3420,7 +3447,7 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port, /* Auto speed and flow control */ skge->autoneg = AUTONEG_ENABLE; - skge->flow_control = FLOW_MODE_SYMMETRIC; + skge->flow_control = FLOW_MODE_SYM_OR_REM; skge->duplex = -1; skge->speed = -1; skge->advertising = skge_supported_modes(hw); diff --git a/drivers/net/skge.h b/drivers/net/skge.h index 9cc955c..537c0aa 100644 --- a/drivers/net/skge.h +++ b/drivers/net/skge.h @@ -2427,13 +2427,24 @@ struct skge_hw { struct mutex phy_mutex; }; -enum { - FLOW_MODE_NONE = 0, /* No Flow-Control */ - FLOW_MODE_LOC_SEND = 1, /* Local station sends PAUSE */ - FLOW_MODE_REM_SEND = 2, /* Symmetric or just remote */ +enum pause_control { + FLOW_MODE_NONE = 1, /* No Flow-Control */ + FLOW_MODE_LOC_SEND = 2, /* Local station sends PAUSE */ FLOW_MODE_SYMMETRIC = 3, /* Both stations may send PAUSE */ + FLOW_MODE_SYM_OR_REM = 4, /* Both stations may send PAUSE or + * just the remote station may send PAUSE + */ +}; + +enum pause_status { + FLOW_STAT_INDETERMINATED=0, /* indeterminated */ + FLOW_STAT_NONE, /* No Flow Control */ + FLOW_STAT_REM_SEND, /* Remote Station sends PAUSE */ + FLOW_STAT_LOC_SEND, /* Local station sends PAUSE */ + FLOW_STAT_SYMMETRIC, /* Both station may send PAUSE */ }; + struct skge_port { u32 msg_enable; struct skge_hw *hw; @@ -2446,9 +2457,10 @@ struct skge_port { struct net_device_stats net_stats; struct work_struct link_thread; + enum pause_control flow_control; + enum pause_status flow_status; u8 rx_csum; u8 blink_on; - u8 flow_control; u8 wol; u8 autoneg; /* AUTONEG_ENABLE, AUTONEG_DISABLE */ u8 duplex; /* DUPLEX_HALF, DUPLEX_FULL */ -- cgit v0.10.2 From 370de6cdc2ed27c41b313c031e4258ffa32272bc Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 5 Oct 2006 15:49:53 -0700 Subject: [PATCH] skge: version 1.9 Want to be able to track downstream impact of fiber related fixes. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/skge.c b/drivers/net/skge.c index d844a83..e7e4149 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -43,7 +43,7 @@ #include "skge.h" #define DRV_NAME "skge" -#define DRV_VERSION "1.8" +#define DRV_VERSION "1.9" #define PFX DRV_NAME " " #define DEFAULT_TX_RING_SIZE 128 -- cgit v0.10.2 From 7bd656d12119708b37414bf909ab2995473da818 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 9 Oct 2006 14:40:38 -0700 Subject: [PATCH] sky2: revert pci express extensions The pci express error handling extensions don't work unless PCI access is via mmconfig. Otherwise, all accesses to pci config registers greater than 256 fail. Since the sky2 driver has other ways of getting to PCI config space, it works around this short coming, but the pci_find_ext_capablity doesn't work. This backs out commit 91aeb3edbcf4e6ed72d138ac8c22fd68e6d717c3 Go back to hardcoding, since we know where the error registers are anyway. Fixes http://bugzilla.kernel.org/show_bug.cgi?id=7222 Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 5a5289b..1f91f30 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -2220,8 +2220,7 @@ static void sky2_hw_intr(struct sky2_hw *hw) /* PCI-Express uncorrectable Error occurred */ u32 pex_err; - pex_err = sky2_pci_read32(hw, - hw->err_cap + PCI_ERR_UNCOR_STATUS); + pex_err = sky2_pci_read32(hw, PEX_UNC_ERR_STAT); if (net_ratelimit()) printk(KERN_ERR PFX "%s: pci express error (0x%x)\n", @@ -2229,20 +2228,15 @@ static void sky2_hw_intr(struct sky2_hw *hw) /* clear the interrupt */ sky2_write32(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON); - sky2_pci_write32(hw, - hw->err_cap + PCI_ERR_UNCOR_STATUS, - 0xffffffffUL); + sky2_pci_write32(hw, PEX_UNC_ERR_STAT, + 0xffffffffUL); sky2_write32(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF); - - /* In case of fatal error mask off to keep from getting stuck */ - if (pex_err & (PCI_ERR_UNC_POISON_TLP | PCI_ERR_UNC_FCP - | PCI_ERR_UNC_DLP)) { + if (pex_err & PEX_FATAL_ERRORS) { u32 hwmsk = sky2_read32(hw, B0_HWE_IMSK); hwmsk &= ~Y2_IS_PCI_EXP; sky2_write32(hw, B0_HWE_IMSK, hwmsk); } - } if (status & Y2_HWE_L1_MASK) @@ -2423,7 +2417,6 @@ static int sky2_reset(struct sky2_hw *hw) u16 status; u8 t8; int i; - u32 msk; sky2_write8(hw, B0_CTST, CS_RST_CLR); @@ -2464,13 +2457,9 @@ static int sky2_reset(struct sky2_hw *hw) sky2_write8(hw, B0_CTST, CS_MRST_CLR); /* clear any PEX errors */ - if (pci_find_capability(hw->pdev, PCI_CAP_ID_EXP)) { - hw->err_cap = pci_find_ext_capability(hw->pdev, PCI_EXT_CAP_ID_ERR); - if (hw->err_cap) - sky2_pci_write32(hw, - hw->err_cap + PCI_ERR_UNCOR_STATUS, - 0xffffffffUL); - } + if (pci_find_capability(hw->pdev, PCI_CAP_ID_EXP)) + sky2_pci_write32(hw, PEX_UNC_ERR_STAT, 0xffffffffUL); + hw->pmd_type = sky2_read8(hw, B2_PMD_TYP); hw->ports = 1; @@ -2527,10 +2516,7 @@ static int sky2_reset(struct sky2_hw *hw) sky2_write8(hw, RAM_BUFFER(i, B3_RI_RTO_XS2), SK_RI_TO_53); } - msk = Y2_HWE_ALL_MASK; - if (!hw->err_cap) - msk &= ~Y2_IS_PCI_EXP; - sky2_write32(hw, B0_HWE_IMSK, msk); + sky2_write32(hw, B0_HWE_IMSK, Y2_HWE_ALL_MASK); for (i = 0; i < hw->ports; i++) sky2_gmac_reset(hw, i); diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index f66109a..43d2acc 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -6,15 +6,24 @@ #define ETH_JUMBO_MTU 9000 /* Maximum MTU supported */ -/* PCI device specific config registers */ +/* PCI config registers */ enum { PCI_DEV_REG1 = 0x40, PCI_DEV_REG2 = 0x44, + PCI_DEV_STATUS = 0x7c, PCI_DEV_REG3 = 0x80, PCI_DEV_REG4 = 0x84, PCI_DEV_REG5 = 0x88, }; +enum { + PEX_DEV_CAP = 0xe4, + PEX_DEV_CTRL = 0xe8, + PEX_DEV_STA = 0xea, + PEX_LNK_STAT = 0xf2, + PEX_UNC_ERR_STAT= 0x104, +}; + /* Yukon-2 */ enum pci_dev_reg_1 { PCI_Y2_PIG_ENA = 1<<31, /* Enable Plug-in-Go (YUKON-2) */ @@ -63,6 +72,39 @@ enum pci_dev_reg_4 { PCI_STATUS_REC_MASTER_ABORT | \ PCI_STATUS_REC_TARGET_ABORT | \ PCI_STATUS_PARITY) + +enum pex_dev_ctrl { + PEX_DC_MAX_RRS_MSK = 7<<12, /* Bit 14..12: Max. Read Request Size */ + PEX_DC_EN_NO_SNOOP = 1<<11,/* Enable No Snoop */ + PEX_DC_EN_AUX_POW = 1<<10,/* Enable AUX Power */ + PEX_DC_EN_PHANTOM = 1<<9, /* Enable Phantom Functions */ + PEX_DC_EN_EXT_TAG = 1<<8, /* Enable Extended Tag Field */ + PEX_DC_MAX_PLS_MSK = 7<<5, /* Bit 7.. 5: Max. Payload Size Mask */ + PEX_DC_EN_REL_ORD = 1<<4, /* Enable Relaxed Ordering */ + PEX_DC_EN_UNS_RQ_RP = 1<<3, /* Enable Unsupported Request Reporting */ + PEX_DC_EN_FAT_ER_RP = 1<<2, /* Enable Fatal Error Reporting */ + PEX_DC_EN_NFA_ER_RP = 1<<1, /* Enable Non-Fatal Error Reporting */ + PEX_DC_EN_COR_ER_RP = 1<<0, /* Enable Correctable Error Reporting */ +}; +#define PEX_DC_MAX_RD_RQ_SIZE(x) (((x)<<12) & PEX_DC_MAX_RRS_MSK) + +/* PEX_UNC_ERR_STAT PEX Uncorrectable Errors Status Register (Yukon-2) */ +enum pex_err { + PEX_UNSUP_REQ = 1<<20, /* Unsupported Request Error */ + + PEX_MALFOR_TLP = 1<<18, /* Malformed TLP */ + + PEX_UNEXP_COMP = 1<<16, /* Unexpected Completion */ + + PEX_COMP_TO = 1<<14, /* Completion Timeout */ + PEX_FLOW_CTRL_P = 1<<13, /* Flow Control Protocol Error */ + PEX_POIS_TLP = 1<<12, /* Poisoned TLP */ + + PEX_DATA_LINK_P = 1<<4, /* Data Link Protocol Error */ + PEX_FATAL_ERRORS= (PEX_MALFOR_TLP | PEX_FLOW_CTRL_P | PEX_DATA_LINK_P), +}; + + enum csr_regs { B0_RAP = 0x0000, B0_CTST = 0x0004, @@ -1836,7 +1878,6 @@ struct sky2_hw { struct net_device *dev[2]; int pm_cap; - int err_cap; u8 chip_id; u8 chip_rev; u8 pmd_type; -- cgit v0.10.2 From 6e532cfe49b6e961e1260642a44959b645e9ab54 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 9 Oct 2006 15:49:27 -0700 Subject: [PATCH] sky2: set lower pause threshold to prevent overrun Adjust the pause threshold on slower systems to keep from getting overrun. Since FIFO is 2K bytes, don't send XON pause until there is space for a full frame. Signed-off-by: Jeff Garzik diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 1f91f30..c10e7f5 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -683,7 +683,7 @@ static void sky2_mac_init(struct sky2_hw *hw, unsigned port) sky2_write16(hw, SK_REG(port, TX_GMF_CTRL_T), GMF_OPER_ON); if (hw->chip_id == CHIP_ID_YUKON_EC_U) { - sky2_write8(hw, SK_REG(port, RX_GMF_LP_THR), 768/8); + sky2_write8(hw, SK_REG(port, RX_GMF_LP_THR), 512/8); sky2_write8(hw, SK_REG(port, RX_GMF_UP_THR), 1024/8); if (hw->dev[port]->mtu > ETH_DATA_LEN) { /* set Tx GMAC FIFO Almost Empty Threshold */ -- cgit v0.10.2 From 4a1d2d81fa327d095a0a8a1f961bace5b0a2f7da Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Fri, 6 Oct 2006 12:12:34 -0600 Subject: [PATCH] Fix section mismatch in de2104x.c WARNING: drivers/net/tulip/de2104x.o - Section mismatch: reference to .init.text:de_init_one from .data.rel.local after 'de_driver' (at offset 0x20) WARNING: drivers/net/tulip/de2104x.o - Section mismatch: reference to .exit.text:de_remove_one from .data.rel.local after 'de_driver' (at offset 0x28) Signed-off-by: Helge Deller Signed-off-by: Kyle McMartin Signed-off-by: Matthew Wilcox Signed-off-by: Jeff Garzik diff --git a/drivers/net/tulip/de2104x.c b/drivers/net/tulip/de2104x.c index 2cfd963..f6b3a94 100644 --- a/drivers/net/tulip/de2104x.c +++ b/drivers/net/tulip/de2104x.c @@ -1730,7 +1730,7 @@ static void __init de21040_get_media_info(struct de_private *de) } /* Note: this routine returns extra data bits for size detection. */ -static unsigned __init tulip_read_eeprom(void __iomem *regs, int location, int addr_len) +static unsigned __devinit tulip_read_eeprom(void __iomem *regs, int location, int addr_len) { int i; unsigned retval = 0; @@ -1926,7 +1926,7 @@ bad_srom: goto fill_defaults; } -static int __init de_init_one (struct pci_dev *pdev, +static int __devinit de_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) { struct net_device *dev; @@ -2082,7 +2082,7 @@ err_out_free: return rc; } -static void __exit de_remove_one (struct pci_dev *pdev) +static void __devexit de_remove_one (struct pci_dev *pdev) { struct net_device *dev = pci_get_drvdata(pdev); struct de_private *de = dev->priv; @@ -2164,7 +2164,7 @@ static struct pci_driver de_driver = { .name = DRV_NAME, .id_table = de_pci_tbl, .probe = de_init_one, - .remove = __exit_p(de_remove_one), + .remove = __devexit_p(de_remove_one), #ifdef CONFIG_PM .suspend = de_suspend, .resume = de_resume, -- cgit v0.10.2 From bbe1fe7ea3438f8c4447dbcd46a126581ed2ed41 Mon Sep 17 00:00:00 2001 From: Eran Tromer Date: Tue, 10 Oct 2006 14:29:25 -0700 Subject: [PATCH] libata: return sense data in HDIO_DRIVE_CMD ioctl Make the HDIO_DRIVE_CMD ioctl in libata (ATA command pass through) return a few ATA registers to userspace, following the same convention as the drivers/ide implementation of the same ioctl. This is needed to support ATA commands like CHECK POWER MODE, which return information in nsectors. This fixes "hdparm -C" on SATA drives. Forcing the sense data read via the cc flag causes spurious check conditions, so we filter these out (following the ATA command pass-through specification T10/04-262r7). Signed-off-by: Eran Tromer Acked-by: Tejun Heo Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index b0d0cc4..7af2a4b 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -164,10 +164,10 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) { int rc = 0; u8 scsi_cmd[MAX_COMMAND_SIZE]; - u8 args[4], *argbuf = NULL; + u8 args[4], *argbuf = NULL, *sensebuf = NULL; int argsize = 0; - struct scsi_sense_hdr sshdr; enum dma_data_direction data_dir; + int cmd_result; if (arg == NULL) return -EINVAL; @@ -175,6 +175,10 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) if (copy_from_user(args, arg, sizeof(args))) return -EFAULT; + sensebuf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_NOIO); + if (!sensebuf) + return -ENOMEM; + memset(scsi_cmd, 0, sizeof(scsi_cmd)); if (args[3]) { @@ -191,7 +195,7 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) data_dir = DMA_FROM_DEVICE; } else { scsi_cmd[1] = (3 << 1); /* Non-data */ - /* scsi_cmd[2] is already 0 -- no off.line, cc, or data xfer */ + scsi_cmd[2] = 0x20; /* cc but no off.line or data xfer */ data_dir = DMA_NONE; } @@ -210,18 +214,46 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) /* Good values for timeout and retries? Values below from scsi_ioctl_send_command() for default case... */ - if (scsi_execute_req(scsidev, scsi_cmd, data_dir, argbuf, argsize, - &sshdr, (10*HZ), 5)) { + cmd_result = scsi_execute(scsidev, scsi_cmd, data_dir, argbuf, argsize, + sensebuf, (10*HZ), 5, 0); + + if (driver_byte(cmd_result) == DRIVER_SENSE) {/* sense data available */ + u8 *desc = sensebuf + 8; + cmd_result &= ~(0xFF<<24); /* DRIVER_SENSE is not an error */ + + /* If we set cc then ATA pass-through will cause a + * check condition even if no error. Filter that. */ + if (cmd_result & SAM_STAT_CHECK_CONDITION) { + struct scsi_sense_hdr sshdr; + scsi_normalize_sense(sensebuf, SCSI_SENSE_BUFFERSIZE, + &sshdr); + if (sshdr.sense_key==0 && + sshdr.asc==0 && sshdr.ascq==0) + cmd_result &= ~SAM_STAT_CHECK_CONDITION; + } + + /* Send userspace a few ATA registers (same as drivers/ide) */ + if (sensebuf[0] == 0x72 && /* format is "descriptor" */ + desc[0] == 0x09 ) { /* code is "ATA Descriptor" */ + args[0] = desc[13]; /* status */ + args[1] = desc[3]; /* error */ + args[2] = desc[5]; /* sector count (0:7) */ + if (copy_to_user(arg, args, sizeof(args))) + rc = -EFAULT; + } + } + + + if (cmd_result) { rc = -EIO; goto error; } - /* Need code to retrieve data from check condition? */ - if ((argbuf) && copy_to_user(arg + sizeof(args), argbuf, argsize)) rc = -EFAULT; error: + kfree(sensebuf); kfree(argbuf); return rc; } -- cgit v0.10.2 From a83068bbaca39197dca26287c16186baee615f0a Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 04:46:52 -0400 Subject: [libata] sata_promise: add PCI ID Noticed by Steve Brown Signed-off-by: Jeff Garzik diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c index d636ede..1eb0d63 100644 --- a/drivers/ata/sata_promise.c +++ b/drivers/ata/sata_promise.c @@ -260,6 +260,7 @@ static const struct pci_device_id pdc_ata_pci_tbl[] = { #if 0 { PCI_VDEVICE(PROMISE, 0x3570), board_20771 }, #endif + { PCI_VDEVICE(PROMISE, 0x3577), board_20771 }, { } /* terminate list */ }; -- cgit v0.10.2 From 53e36ada37cb8b01cfbf674580a79edc0bb764c7 Mon Sep 17 00:00:00 2001 From: Peter Korsgaard Date: Mon, 9 Oct 2006 09:53:09 +0200 Subject: [PATCH] pata-qdi: fix le32 in data_xfer The following tiny patch fixes a typo in qdi_data_xfer (le32 instead of le16). Signed-off-by: Peter Korsgaard Signed-off-by: Jeff Garzik diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c index 7977f47..2c3cc0c 100644 --- a/drivers/ata/pata_qdi.c +++ b/drivers/ata/pata_qdi.c @@ -141,7 +141,7 @@ static void qdi_data_xfer(struct ata_device *adev, unsigned char *buf, unsigned memcpy(&pad, buf + buflen - slop, slop); outl(le32_to_cpu(pad), ap->ioaddr.data_addr); } else { - pad = cpu_to_le16(inl(ap->ioaddr.data_addr)); + pad = cpu_to_le32(inl(ap->ioaddr.data_addr)); memcpy(buf + buflen - slop, &pad, slop); } } -- cgit v0.10.2 From 9765d262b8230b735c4b2815b041c09a00833cf1 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 11 Oct 2006 22:29:51 +1000 Subject: [CRYPTO] api: fix crypto_alloc_base() return value This patch makes crypto_alloc_base() return proper return value. - If kzalloc() failure happens within __crypto_alloc_tfm(), crypto_alloc_base() returns NULL. But crypto_alloc_base() is supposed to return error code as pointer. So this patch makes it return -ENOMEM in that case. - crypto_alloc_base() is suppose to return -EINTR, if it is interrupted by signal. But it may not return -EINTR. Signed-off-by: Akinobu Mita Signed-off-by: Herbert Xu diff --git a/crypto/api.c b/crypto/api.c index 2e84d4b..4fb7fa4 100644 --- a/crypto/api.c +++ b/crypto/api.c @@ -331,7 +331,7 @@ struct crypto_tfm *__crypto_alloc_tfm(struct crypto_alg *alg, u32 flags) tfm_size = sizeof(*tfm) + crypto_ctxsize(alg, flags); tfm = kzalloc(tfm_size, GFP_KERNEL); if (tfm == NULL) - goto out; + goto out_err; tfm->__crt_alg = alg; @@ -355,6 +355,7 @@ cra_init_failed: crypto_exit_ops(tfm); out_free_tfm: kfree(tfm); +out_err: tfm = ERR_PTR(err); out: return tfm; @@ -414,14 +415,14 @@ struct crypto_tfm *crypto_alloc_base(const char *alg_name, u32 type, u32 mask) struct crypto_alg *alg; alg = crypto_alg_mod_lookup(alg_name, type, mask); - err = PTR_ERR(alg); - tfm = ERR_PTR(err); - if (IS_ERR(alg)) + if (IS_ERR(alg)) { + err = PTR_ERR(alg); goto err; + } tfm = __crypto_alloc_tfm(alg, 0); if (!IS_ERR(tfm)) - break; + return tfm; crypto_mod_put(alg); err = PTR_ERR(tfm); @@ -433,9 +434,9 @@ err: err = -EINTR; break; } - }; + } - return tfm; + return ERR_PTR(err); } EXPORT_SYMBOL_GPL(crypto_alloc_base); -- cgit v0.10.2 From 9d0a57cbdb4976f382eb1c03baee338e467b6592 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 11 Oct 2006 15:31:26 +0200 Subject: [S390] irq change improvements. Remove the last few places where a pointer to pt_regs gets passed. Also make sure we call set_irq_regs() before irq_enter() and after irq_exit(). This doesn't fix anything but makes sure s390 looks the same like all other architectures. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 2b1e6c9..45c9fa7 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -109,7 +109,7 @@ static LIST_HEAD(appldata_ops_list); * * schedule work and reschedule timer */ -static void appldata_timer_function(unsigned long data, struct pt_regs *regs) +static void appldata_timer_function(unsigned long data) { P_DEBUG(" -= Timer =-\n"); P_DEBUG("CPU: %i, expire_count: %i\n", smp_processor_id(), diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c index c49ab8c..4faf96f 100644 --- a/arch/s390/kernel/s390_ext.c +++ b/arch/s390/kernel/s390_ext.c @@ -117,8 +117,8 @@ void do_extint(struct pt_regs *regs, unsigned short code) int index; struct pt_regs *old_regs; - irq_enter(); old_regs = set_irq_regs(regs); + irq_enter(); asm volatile ("mc 0,0"); if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer) /** @@ -134,8 +134,8 @@ void do_extint(struct pt_regs *regs, unsigned short code) p->handler(code); } } - set_irq_regs(old_regs); irq_exit(); + set_irq_regs(old_regs); } EXPORT_SYMBOL(register_external_interrupt); diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index 1d7d393..21baaf5 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -209,11 +209,11 @@ static void list_add_sorted(struct vtimer_list *timer, struct list_head *head) * Do the callback functions of expired vtimer events. * Called from within the interrupt handler. */ -static void do_callbacks(struct list_head *cb_list, struct pt_regs *regs) +static void do_callbacks(struct list_head *cb_list) { struct vtimer_queue *vt_list; struct vtimer_list *event, *tmp; - void (*fn)(unsigned long, struct pt_regs*); + void (*fn)(unsigned long); unsigned long data; if (list_empty(cb_list)) @@ -224,7 +224,7 @@ static void do_callbacks(struct list_head *cb_list, struct pt_regs *regs) list_for_each_entry_safe(event, tmp, cb_list, entry) { fn = event->function; data = event->data; - fn(data, regs); + fn(data); if (!event->interval) /* delete one shot timer */ @@ -275,7 +275,7 @@ static void do_cpu_timer_interrupt(__u16 error_code) list_move_tail(&event->entry, &cb_list); } spin_unlock(&vt_list->lock); - do_callbacks(&cb_list, get_irq_regs()); + do_callbacks(&cb_list); /* next event is first in list */ spin_lock(&vt_list->lock); diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index f18b162..8936e46 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -609,8 +609,8 @@ do_IRQ (struct pt_regs *regs) struct irb *irb; struct pt_regs *old_regs; - irq_enter (); old_regs = set_irq_regs(regs); + irq_enter(); asm volatile ("mc 0,0"); if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer) /** @@ -655,8 +655,8 @@ do_IRQ (struct pt_regs *regs) * out of the sie which costs more cycles than it saves. */ } while (!MACHINE_IS_VM && tpi (NULL) != 0); + irq_exit(); set_irq_regs(old_regs); - irq_exit (); } #ifdef CONFIG_CCW_CONSOLE diff --git a/include/asm-s390/timer.h b/include/asm-s390/timer.h index fcd6c25..30e5cbe 100644 --- a/include/asm-s390/timer.h +++ b/include/asm-s390/timer.h @@ -26,7 +26,7 @@ struct vtimer_list { spinlock_t lock; unsigned long magic; - void (*function)(unsigned long, struct pt_regs*); + void (*function)(unsigned long); unsigned long data; }; -- cgit v0.10.2 From 08983787d2ccab64cb790965ba89621d96cc22c1 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 11 Oct 2006 15:31:30 +0200 Subject: [S390] cio: add missing KERN_INFO printk header. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index b676202..c36d8b6 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -885,7 +885,8 @@ ccw_device_w4sense(struct ccw_device *cdev, enum dev_event dev_event) /* Basic sense hasn't started. Try again. */ ccw_device_do_sense(cdev, irb); else { - printk("Huh? %s(%s): unsolicited interrupt...\n", + printk(KERN_INFO "Huh? %s(%s): unsolicited " + "interrupt...\n", __FUNCTION__, cdev->dev.bus_id); if (cdev->handler) cdev->handler (cdev, 0, irb); -- cgit v0.10.2 From 715d854bc215bbcca35097176d674c3ac58a085c Mon Sep 17 00:00:00 2001 From: Melissa Howland Date: Wed, 11 Oct 2006 15:31:34 +0200 Subject: [S390] monwriter kzalloc size. Fix length on kzalloc for data buffer so as to not overwrite unallocated storage. Signed-off-by: Melissa Howland Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/char/monwriter.c b/drivers/s390/char/monwriter.c index 4362ff2..abd02ed 100644 --- a/drivers/s390/char/monwriter.c +++ b/drivers/s390/char/monwriter.c @@ -110,7 +110,7 @@ static int monwrite_new_hdr(struct mon_private *monpriv) monbuf = kzalloc(sizeof(struct mon_buf), GFP_KERNEL); if (!monbuf) return -ENOMEM; - monbuf->data = kzalloc(monbuf->hdr.datalen, + monbuf->data = kzalloc(monhdr->datalen, GFP_KERNEL | GFP_DMA); if (!monbuf->data) { kfree(monbuf); -- cgit v0.10.2 From 789642680518b28e7dc13f96061460a8238ec622 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 11 Oct 2006 15:31:38 +0200 Subject: [S390] cio: Use ccw_dev_id and subchannel_id in ccw_device_private Use the proper structures to identify device and subchannel. Change get_disc_ccwdev_by_devno() to get_disc_ccwdev_by_dev_id(). Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/css.h b/drivers/s390/cio/css.h index 8aabb4a..15bd1e2 100644 --- a/drivers/s390/cio/css.h +++ b/drivers/s390/cio/css.h @@ -76,9 +76,8 @@ struct ccw_device_private { int state; /* device state */ atomic_t onoff; unsigned long registered; - __u16 devno; /* device number */ - __u16 sch_no; /* subchannel number */ - __u8 ssid; /* subchannel set id */ + struct ccw_dev_id dev_id; /* device id */ + struct subchannel_id schid; /* subchannel number */ __u8 imask; /* lpm mask for SNID/SID/SPGID */ int iretry; /* retry counter SNID/SID/SPGID */ struct { diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index 6889456..7646a99 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -552,21 +552,19 @@ ccw_device_register(struct ccw_device *cdev) } struct match_data { - unsigned int devno; - unsigned int ssid; + struct ccw_dev_id dev_id; struct ccw_device * sibling; }; static int match_devno(struct device * dev, void * data) { - struct match_data * d = (struct match_data *)data; + struct match_data * d = data; struct ccw_device * cdev; cdev = to_ccwdev(dev); if ((cdev->private->state == DEV_STATE_DISCONNECTED) && - (cdev->private->devno == d->devno) && - (cdev->private->ssid == d->ssid) && + ccw_dev_id_is_equal(&cdev->private->dev_id, &d->dev_id) && (cdev != d->sibling)) { cdev->private->state = DEV_STATE_NOT_OPER; return 1; @@ -574,15 +572,13 @@ match_devno(struct device * dev, void * data) return 0; } -static struct ccw_device * -get_disc_ccwdev_by_devno(unsigned int devno, unsigned int ssid, - struct ccw_device *sibling) +static struct ccw_device * get_disc_ccwdev_by_dev_id(struct ccw_dev_id *dev_id, + struct ccw_device *sibling) { struct device *dev; struct match_data data; - data.devno = devno; - data.ssid = ssid; + data.dev_id = *dev_id; data.sibling = sibling; dev = bus_find_device(&ccw_bus_type, NULL, &data, match_devno); @@ -618,7 +614,7 @@ ccw_device_do_unreg_rereg(void *data) cdev = (struct ccw_device *)data; sch = to_subchannel(cdev->dev.parent); - if (cdev->private->devno != sch->schib.pmcw.dev) { + if (cdev->private->dev_id.devno != sch->schib.pmcw.dev) { /* * The device number has changed. This is usually only when * a device has been detached under VM and then re-appeared @@ -633,10 +629,12 @@ ccw_device_do_unreg_rereg(void *data) * get possibly sick... */ struct ccw_device *other_cdev; + struct ccw_dev_id dev_id; need_rename = 1; - other_cdev = get_disc_ccwdev_by_devno(sch->schib.pmcw.dev, - sch->schid.ssid, cdev); + dev_id.devno = sch->schib.pmcw.dev; + dev_id.ssid = sch->schid.ssid; + other_cdev = get_disc_ccwdev_by_dev_id(&dev_id, cdev); if (other_cdev) { struct subchannel *other_sch; @@ -652,7 +650,7 @@ ccw_device_do_unreg_rereg(void *data) } /* Update ssd info here. */ css_get_ssd_info(sch); - cdev->private->devno = sch->schib.pmcw.dev; + cdev->private->dev_id.devno = sch->schib.pmcw.dev; } else need_rename = 0; device_remove_files(&cdev->dev); @@ -792,9 +790,9 @@ io_subchannel_recog(struct ccw_device *cdev, struct subchannel *sch) /* Init private data. */ priv = cdev->private; - priv->devno = sch->schib.pmcw.dev; - priv->ssid = sch->schid.ssid; - priv->sch_no = sch->schid.sch_no; + priv->dev_id.devno = sch->schib.pmcw.dev; + priv->dev_id.ssid = sch->schid.ssid; + priv->schid = sch->schid; priv->state = DEV_STATE_NOT_OPER; INIT_LIST_HEAD(&priv->cmb_list); init_waitqueue_head(&priv->wait_q); diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index c36d8b6..392eb33 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -183,7 +183,7 @@ ccw_device_handle_oper(struct ccw_device *cdev) cdev->id.cu_model != cdev->private->senseid.cu_model || cdev->id.dev_type != cdev->private->senseid.dev_type || cdev->id.dev_model != cdev->private->senseid.dev_model || - cdev->private->devno != sch->schib.pmcw.dev) { + cdev->private->dev_id.devno != sch->schib.pmcw.dev) { PREPARE_WORK(&cdev->private->kick_work, ccw_device_do_unreg_rereg, (void *)cdev); queue_work(ccw_device_work, &cdev->private->kick_work); @@ -255,7 +255,7 @@ ccw_device_recog_done(struct ccw_device *cdev, int state) case DEV_STATE_NOT_OPER: CIO_DEBUG(KERN_WARNING, 2, "SenseID : unknown device %04x on subchannel " - "0.%x.%04x\n", cdev->private->devno, + "0.%x.%04x\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no); break; case DEV_STATE_OFFLINE: @@ -282,14 +282,15 @@ ccw_device_recog_done(struct ccw_device *cdev, int state) CIO_DEBUG(KERN_INFO, 2, "SenseID : device 0.%x.%04x reports: " "CU Type/Mod = %04X/%02X, Dev Type/Mod = " "%04X/%02X\n", - cdev->private->ssid, cdev->private->devno, + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno, cdev->id.cu_type, cdev->id.cu_model, cdev->id.dev_type, cdev->id.dev_model); break; case DEV_STATE_BOXED: CIO_DEBUG(KERN_WARNING, 2, "SenseID : boxed device %04x on subchannel " - "0.%x.%04x\n", cdev->private->devno, + "0.%x.%04x\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no); break; } @@ -363,7 +364,7 @@ ccw_device_done(struct ccw_device *cdev, int state) if (state == DEV_STATE_BOXED) CIO_DEBUG(KERN_WARNING, 2, "Boxed device %04x on subchannel %04x\n", - cdev->private->devno, sch->schid.sch_no); + cdev->private->dev_id.devno, sch->schid.sch_no); if (cdev->private->flags.donotify) { cdev->private->flags.donotify = 0; @@ -412,7 +413,8 @@ static void __ccw_device_get_common_pgid(struct ccw_device *cdev) /* PGID mismatch, can't pathgroup. */ CIO_MSG_EVENT(0, "SNID - pgid mismatch for device " "0.%x.%04x, can't pathgroup\n", - cdev->private->ssid, cdev->private->devno); + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno); cdev->private->options.pgroup = 0; return; } diff --git a/drivers/s390/cio/device_id.c b/drivers/s390/cio/device_id.c index 1398367..a74785b 100644 --- a/drivers/s390/cio/device_id.c +++ b/drivers/s390/cio/device_id.c @@ -251,7 +251,7 @@ ccw_device_check_sense_id(struct ccw_device *cdev) */ CIO_MSG_EVENT(2, "SenseID : device %04x on Subchannel " "0.%x.%04x reports cmd reject\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no); return -EOPNOTSUPP; } @@ -259,7 +259,8 @@ ccw_device_check_sense_id(struct ccw_device *cdev) CIO_MSG_EVENT(2, "SenseID : UC on dev 0.%x.%04x, " "lpum %02X, cnt %02d, sns :" " %02X%02X%02X%02X %02X%02X%02X%02X ...\n", - cdev->private->ssid, cdev->private->devno, + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno, irb->esw.esw0.sublog.lpum, irb->esw.esw0.erw.scnt, irb->ecw[0], irb->ecw[1], @@ -274,14 +275,15 @@ ccw_device_check_sense_id(struct ccw_device *cdev) CIO_MSG_EVENT(2, "SenseID : path %02X for device %04x " "on subchannel 0.%x.%04x is " "'not operational'\n", sch->orb.lpm, - cdev->private->devno, sch->schid.ssid, - sch->schid.sch_no); + cdev->private->dev_id.devno, + sch->schid.ssid, sch->schid.sch_no); return -EACCES; } /* Hmm, whatever happened, try again. */ CIO_MSG_EVENT(2, "SenseID : start_IO() for device %04x on " "subchannel 0.%x.%04x returns status %02X%02X\n", - cdev->private->devno, sch->schid.ssid, sch->schid.sch_no, + cdev->private->dev_id.devno, sch->schid.ssid, + sch->schid.sch_no, irb->scsw.dstat, irb->scsw.cstat); return -EAGAIN; } @@ -330,7 +332,7 @@ ccw_device_sense_id_irq(struct ccw_device *cdev, enum dev_event dev_event) /* fall through. */ default: /* Sense ID failed. Try asking VM. */ if (MACHINE_IS_VM) { - VM_virtual_device_info (cdev->private->devno, + VM_virtual_device_info (cdev->private->dev_id.devno, &cdev->private->senseid); if (cdev->private->senseid.cu_type != 0xFFFF) { /* Got the device information from VM. */ diff --git a/drivers/s390/cio/device_ops.c b/drivers/s390/cio/device_ops.c index 84b9b18..9621993 100644 --- a/drivers/s390/cio/device_ops.c +++ b/drivers/s390/cio/device_ops.c @@ -592,13 +592,13 @@ ccw_device_get_chp_desc(struct ccw_device *cdev, int chp_no) int _ccw_device_get_subchannel_number(struct ccw_device *cdev) { - return cdev->private->sch_no; + return cdev->private->schid.sch_no; } int _ccw_device_get_device_number(struct ccw_device *cdev) { - return cdev->private->devno; + return cdev->private->dev_id.devno; } diff --git a/drivers/s390/cio/device_pgid.c b/drivers/s390/cio/device_pgid.c index 84917b3..2975ce8 100644 --- a/drivers/s390/cio/device_pgid.c +++ b/drivers/s390/cio/device_pgid.c @@ -79,7 +79,8 @@ __ccw_device_sense_pgid_start(struct ccw_device *cdev) CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel " "0.%x.%04x, lpm %02X, became 'not " "operational'\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, + sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); } @@ -135,7 +136,8 @@ __ccw_device_check_sense_pgid(struct ccw_device *cdev) CIO_MSG_EVENT(2, "SNID - device 0.%x.%04x, unit check, " "lpum %02X, cnt %02d, sns : " "%02X%02X%02X%02X %02X%02X%02X%02X ...\n", - cdev->private->ssid, cdev->private->devno, + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno, irb->esw.esw0.sublog.lpum, irb->esw.esw0.erw.scnt, irb->ecw[0], irb->ecw[1], @@ -147,7 +149,7 @@ __ccw_device_check_sense_pgid(struct ccw_device *cdev) if (irb->scsw.cc == 3) { CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel 0.%x.%04x," " lpm %02X, became 'not operational'\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, sch->orb.lpm); return -EACCES; } @@ -155,7 +157,7 @@ __ccw_device_check_sense_pgid(struct ccw_device *cdev) if (cdev->private->pgid[i].inf.ps.state2 == SNID_STATE2_RESVD_ELSE) { CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel 0.%x.%04x " "is reserved by someone else\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no); return -EUSERS; } @@ -261,7 +263,7 @@ __ccw_device_do_pgid(struct ccw_device *cdev, __u8 func) /* PGID command failed on this path. */ CIO_MSG_EVENT(2, "SPID - Device %04x on Subchannel " "0.%x.%04x, lpm %02X, became 'not operational'\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); return ret; } @@ -301,7 +303,7 @@ static int __ccw_device_do_nop(struct ccw_device *cdev) /* nop command failed on this path. */ CIO_MSG_EVENT(2, "NOP - Device %04x on Subchannel " "0.%x.%04x, lpm %02X, became 'not operational'\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); return ret; } @@ -328,8 +330,9 @@ __ccw_device_check_pgid(struct ccw_device *cdev) CIO_MSG_EVENT(2, "SPID - device 0.%x.%04x, unit check, " "cnt %02d, " "sns : %02X%02X%02X%02X %02X%02X%02X%02X ...\n", - cdev->private->ssid, - cdev->private->devno, irb->esw.esw0.erw.scnt, + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno, + irb->esw.esw0.erw.scnt, irb->ecw[0], irb->ecw[1], irb->ecw[2], irb->ecw[3], irb->ecw[4], irb->ecw[5], @@ -339,7 +342,7 @@ __ccw_device_check_pgid(struct ccw_device *cdev) if (irb->scsw.cc == 3) { CIO_MSG_EVENT(2, "SPID - Device %04x on Subchannel 0.%x.%04x," " lpm %02X, became 'not operational'\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); return -EACCES; } @@ -362,7 +365,7 @@ static int __ccw_device_check_nop(struct ccw_device *cdev) if (irb->scsw.cc == 3) { CIO_MSG_EVENT(2, "NOP - Device %04x on Subchannel 0.%x.%04x," " lpm %02X, became 'not operational'\n", - cdev->private->devno, sch->schid.ssid, + cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); return -EACCES; } diff --git a/drivers/s390/cio/device_status.c b/drivers/s390/cio/device_status.c index caf148d..3f7cbce 100644 --- a/drivers/s390/cio/device_status.c +++ b/drivers/s390/cio/device_status.c @@ -32,19 +32,18 @@ ccw_device_msg_control_check(struct ccw_device *cdev, struct irb *irb) SCHN_STAT_CHN_CTRL_CHK | SCHN_STAT_INTF_CTRL_CHK))) return; - CIO_MSG_EVENT(0, "Channel-Check or Interface-Control-Check " "received" " ... device %04x on subchannel 0.%x.%04x, dev_stat " ": %02X sch_stat : %02X\n", - cdev->private->devno, cdev->private->ssid, - cdev->private->sch_no, + cdev->private->dev_id.devno, cdev->private->schid.ssid, + cdev->private->schid.sch_no, irb->scsw.dstat, irb->scsw.cstat); if (irb->scsw.cc != 3) { char dbf_text[15]; - sprintf(dbf_text, "chk%x", cdev->private->sch_no); + sprintf(dbf_text, "chk%x", cdev->private->schid.sch_no); CIO_TRACE_EVENT(0, dbf_text); CIO_HEX_EVENT(0, irb, sizeof (struct irb)); } diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c index cde822d..0648ce5 100644 --- a/drivers/s390/cio/qdio.c +++ b/drivers/s390/cio/qdio.c @@ -1741,7 +1741,7 @@ qdio_fill_qs(struct qdio_irq *irq_ptr, struct ccw_device *cdev, void *ptr; int available; - sprintf(dbf_text,"qfqs%4x",cdev->private->sch_no); + sprintf(dbf_text,"qfqs%4x",cdev->private->schid.sch_no); QDIO_DBF_TEXT0(0,setup,dbf_text); for (i=0;iinput_qs[i]; @@ -2924,7 +2924,7 @@ qdio_establish_handle_irq(struct ccw_device *cdev, int cstat, int dstat) irq_ptr = cdev->private->qdio_data; - sprintf(dbf_text,"qehi%4x",cdev->private->sch_no); + sprintf(dbf_text,"qehi%4x",cdev->private->schid.sch_no); QDIO_DBF_TEXT0(0,setup,dbf_text); QDIO_DBF_TEXT0(0,trace,dbf_text); @@ -2943,7 +2943,7 @@ qdio_initialize(struct qdio_initialize *init_data) int rc; char dbf_text[15]; - sprintf(dbf_text,"qini%4x",init_data->cdev->private->sch_no); + sprintf(dbf_text,"qini%4x",init_data->cdev->private->schid.sch_no); QDIO_DBF_TEXT0(0,setup,dbf_text); QDIO_DBF_TEXT0(0,trace,dbf_text); @@ -2964,7 +2964,7 @@ qdio_allocate(struct qdio_initialize *init_data) struct qdio_irq *irq_ptr; char dbf_text[15]; - sprintf(dbf_text,"qalc%4x",init_data->cdev->private->sch_no); + sprintf(dbf_text,"qalc%4x",init_data->cdev->private->schid.sch_no); QDIO_DBF_TEXT0(0,setup,dbf_text); QDIO_DBF_TEXT0(0,trace,dbf_text); if ( (init_data->no_input_qs>QDIO_MAX_QUEUES_PER_IRQ) || @@ -3187,7 +3187,7 @@ qdio_establish(struct qdio_initialize *init_data) tiqdio_set_delay_target(irq_ptr,TIQDIO_DELAY_TARGET); } - sprintf(dbf_text,"qest%4x",cdev->private->sch_no); + sprintf(dbf_text,"qest%4x",cdev->private->schid.sch_no); QDIO_DBF_TEXT0(0,setup,dbf_text); QDIO_DBF_TEXT0(0,trace,dbf_text); diff --git a/include/asm-s390/cio.h b/include/asm-s390/cio.h index da063cd..81287d8 100644 --- a/include/asm-s390/cio.h +++ b/include/asm-s390/cio.h @@ -275,6 +275,12 @@ struct ccw_dev_id { u16 devno; }; +static inline int ccw_dev_id_is_equal(struct ccw_dev_id *dev_id1, + struct ccw_dev_id *dev_id2) +{ + return !memcmp(dev_id1, dev_id2, sizeof(struct ccw_dev_id)); +} + extern int diag210(struct diag210 *addr); extern void wait_cons_dev(void); -- cgit v0.10.2 From e7769b48a0216d7262fe2ba59b1b3697be462cbb Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 11 Oct 2006 15:31:41 +0200 Subject: [S390] cio: Remove grace period for vary off chpid. The grace period handling introduced needless complexity. It didn't help the dasd driver (which can handle terminated I/O just well), and it doesn't help for long running channel programs (which won't complete during the grace period anyway). Terminating I/O using a path that just disappeared immediately is much more consistent with what the user expects. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c index 07c7f19..eabe018 100644 --- a/drivers/s390/cio/chsc.c +++ b/drivers/s390/cio/chsc.c @@ -707,8 +707,7 @@ chp_process_crw(int chpid, int on) return chp_add(chpid); } -static inline int -__check_for_io_and_kill(struct subchannel *sch, int index) +static inline int check_for_io_on_path(struct subchannel *sch, int index) { int cc; @@ -718,10 +717,8 @@ __check_for_io_and_kill(struct subchannel *sch, int index) cc = stsch(sch->schid, &sch->schib); if (cc) return 0; - if (sch->schib.scsw.actl && sch->schib.pmcw.lpum == (0x80 >> index)) { - device_set_waiting(sch); + if (sch->schib.scsw.actl && sch->schib.pmcw.lpum == (0x80 >> index)) return 1; - } return 0; } @@ -750,12 +747,10 @@ __s390_subchannel_vary_chpid(struct subchannel *sch, __u8 chpid, int on) } else { sch->opm &= ~(0x80 >> chp); sch->lpm &= ~(0x80 >> chp); - /* - * Give running I/O a grace period in which it - * can successfully terminate, even using the - * just varied off path. Then kill it. - */ - if (!__check_for_io_and_kill(sch, chp) && !sch->lpm) { + if (check_for_io_on_path(sch, chp)) + /* Path verification is done after killing. */ + device_kill_io(sch); + else if (!sch->lpm) { if (css_enqueue_subchannel_slow(sch->schid)) { css_clear_subchannel_slow_list(); need_rescan = 1; diff --git a/drivers/s390/cio/css.h b/drivers/s390/cio/css.h index 15bd1e2..4c2ff83 100644 --- a/drivers/s390/cio/css.h +++ b/drivers/s390/cio/css.h @@ -170,7 +170,7 @@ void device_trigger_reprobe(struct subchannel *); /* Helper functions for vary on/off. */ int device_is_online(struct subchannel *); -void device_set_waiting(struct subchannel *); +void device_kill_io(struct subchannel *); /* Machine check helper function. */ void device_kill_pending_timer(struct subchannel *); diff --git a/drivers/s390/cio/device.h b/drivers/s390/cio/device.h index 00be9a5..c6140cc 100644 --- a/drivers/s390/cio/device.h +++ b/drivers/s390/cio/device.h @@ -21,7 +21,6 @@ enum dev_state { /* states to wait for i/o completion before doing something */ DEV_STATE_CLEAR_VERIFY, DEV_STATE_TIMEOUT_KILL, - DEV_STATE_WAIT4IO, DEV_STATE_QUIESCE, /* special states for devices gone not operational */ DEV_STATE_DISCONNECTED, diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index 392eb33..44e4a53 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -59,18 +59,6 @@ device_set_disconnected(struct subchannel *sch) cdev->private->state = DEV_STATE_DISCONNECTED; } -void -device_set_waiting(struct subchannel *sch) -{ - struct ccw_device *cdev; - - if (!sch->dev.driver_data) - return; - cdev = sch->dev.driver_data; - ccw_device_set_timeout(cdev, 10*HZ); - cdev->private->state = DEV_STATE_WAIT4IO; -} - /* * Timeout function. It just triggers a DEV_EVENT_TIMEOUT. */ @@ -947,7 +935,7 @@ ccw_device_killing_irq(struct ccw_device *cdev, enum dev_event dev_event) cdev->private->state = DEV_STATE_ONLINE; if (cdev->handler) cdev->handler(cdev, cdev->private->intparm, - ERR_PTR(-ETIMEDOUT)); + ERR_PTR(-EIO)); if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, ccw_device_nopath_notify, (void *)cdev); @@ -984,51 +972,15 @@ ccw_device_killing_timeout(struct ccw_device *cdev, enum dev_event dev_event) cdev->private->state = DEV_STATE_ONLINE; if (cdev->handler) cdev->handler(cdev, cdev->private->intparm, - ERR_PTR(-ETIMEDOUT)); + ERR_PTR(-EIO)); } -static void -ccw_device_wait4io_irq(struct ccw_device *cdev, enum dev_event dev_event) -{ - struct irb *irb; - struct subchannel *sch; - - irb = (struct irb *) __LC_IRB; - /* - * Accumulate status and find out if a basic sense is needed. - * This is fine since we have already adapted the lpm. - */ - ccw_device_accumulate_irb(cdev, irb); - if (cdev->private->flags.dosense) { - if (ccw_device_do_sense(cdev, irb) == 0) { - cdev->private->state = DEV_STATE_W4SENSE; - } - return; - } - - /* Iff device is idle, reset timeout. */ - sch = to_subchannel(cdev->dev.parent); - if (!stsch(sch->schid, &sch->schib)) - if (sch->schib.scsw.actl == 0) - ccw_device_set_timeout(cdev, 0); - /* Call the handler. */ - ccw_device_call_handler(cdev); - if (!sch->lpm) { - PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); - queue_work(ccw_device_notify_work, &cdev->private->kick_work); - } else if (cdev->private->flags.doverify) - ccw_device_online_verify(cdev, 0); -} - -static void -ccw_device_wait4io_timeout(struct ccw_device *cdev, enum dev_event dev_event) +void device_kill_io(struct subchannel *sch) { int ret; - struct subchannel *sch; + struct ccw_device *cdev; - sch = to_subchannel(cdev->dev.parent); - ccw_device_set_timeout(cdev, 0); + cdev = sch->dev.driver_data; ret = ccw_device_cancel_halt_clear(cdev); if (ret == -EBUSY) { ccw_device_set_timeout(cdev, 3*HZ); @@ -1047,12 +999,12 @@ ccw_device_wait4io_timeout(struct ccw_device *cdev, enum dev_event dev_event) } if (cdev->handler) cdev->handler(cdev, cdev->private->intparm, - ERR_PTR(-ETIMEDOUT)); + ERR_PTR(-EIO)); if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, ccw_device_nopath_notify, (void *)cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); - } else if (cdev->private->flags.doverify) + } else /* Start delayed path verification. */ ccw_device_online_verify(cdev, 0); } @@ -1289,12 +1241,6 @@ fsm_func_t *dev_jumptable[NR_DEV_STATES][NR_DEV_EVENTS] = { [DEV_EVENT_TIMEOUT] = ccw_device_killing_timeout, [DEV_EVENT_VERIFY] = ccw_device_nop, //FIXME }, - [DEV_STATE_WAIT4IO] = { - [DEV_EVENT_NOTOPER] = ccw_device_online_notoper, - [DEV_EVENT_INTERRUPT] = ccw_device_wait4io_irq, - [DEV_EVENT_TIMEOUT] = ccw_device_wait4io_timeout, - [DEV_EVENT_VERIFY] = ccw_device_delay_verify, - }, [DEV_STATE_QUIESCE] = { [DEV_EVENT_NOTOPER] = ccw_device_quiesce_done, [DEV_EVENT_INTERRUPT] = ccw_device_quiesce_done, diff --git a/drivers/s390/cio/device_ops.c b/drivers/s390/cio/device_ops.c index 9621993..b39c1fa 100644 --- a/drivers/s390/cio/device_ops.c +++ b/drivers/s390/cio/device_ops.c @@ -50,7 +50,6 @@ ccw_device_clear(struct ccw_device *cdev, unsigned long intparm) if (cdev->private->state == DEV_STATE_NOT_OPER) return -ENODEV; if (cdev->private->state != DEV_STATE_ONLINE && - cdev->private->state != DEV_STATE_WAIT4IO && cdev->private->state != DEV_STATE_W4SENSE) return -EINVAL; sch = to_subchannel(cdev->dev.parent); @@ -155,7 +154,6 @@ ccw_device_halt(struct ccw_device *cdev, unsigned long intparm) if (cdev->private->state == DEV_STATE_NOT_OPER) return -ENODEV; if (cdev->private->state != DEV_STATE_ONLINE && - cdev->private->state != DEV_STATE_WAIT4IO && cdev->private->state != DEV_STATE_W4SENSE) return -EINVAL; sch = to_subchannel(cdev->dev.parent); -- cgit v0.10.2 From 12975aef62836e9f3e179afaaded8045f8a25ac4 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 11 Oct 2006 15:31:47 +0200 Subject: [S390] cio: remove casts from/to (void *). Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c index eabe018..2d78f0f 100644 --- a/drivers/s390/cio/chsc.c +++ b/drivers/s390/cio/chsc.c @@ -370,7 +370,7 @@ __s390_process_res_acc(struct subchannel_id schid, void *data) struct res_acc_data *res_data; struct subchannel *sch; - res_data = (struct res_acc_data *)data; + res_data = data; sch = get_subchannel_by_schid(schid); if (!sch) /* Check if a subchannel is newly available. */ @@ -444,7 +444,7 @@ __get_chpid_from_lir(void *data) u32 isinfo[28]; } *lir; - lir = (struct lir*) data; + lir = data; if (!(lir->iq&0x80)) /* NULL link incident record */ return -EINVAL; @@ -628,7 +628,7 @@ __chp_add(struct subchannel_id schid, void *data) struct channel_path *chp; struct subchannel *sch; - chp = (struct channel_path *)data; + chp = data; sch = get_subchannel_by_schid(schid); if (!sch) /* Check if the subchannel is now available. */ diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c index 7086a74..a2dee5b 100644 --- a/drivers/s390/cio/css.c +++ b/drivers/s390/cio/css.c @@ -177,7 +177,7 @@ get_subchannel_by_schid(struct subchannel_id schid) struct device *dev; dev = bus_find_device(&css_bus_type, NULL, - (void *)&schid, check_subchannel); + &schid, check_subchannel); return dev ? to_subchannel(dev) : NULL; } diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index 7646a99..94bdd4d 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -591,7 +591,7 @@ ccw_device_add_changed(void *data) struct ccw_device *cdev; - cdev = (struct ccw_device *)data; + cdev = data; if (device_add(&cdev->dev)) { put_device(&cdev->dev); return; @@ -612,7 +612,7 @@ ccw_device_do_unreg_rereg(void *data) struct subchannel *sch; int need_rename; - cdev = (struct ccw_device *)data; + cdev = data; sch = to_subchannel(cdev->dev.parent); if (cdev->private->dev_id.devno != sch->schib.pmcw.dev) { /* @@ -660,7 +660,7 @@ ccw_device_do_unreg_rereg(void *data) snprintf (cdev->dev.bus_id, BUS_ID_SIZE, "0.%x.%04x", sch->schid.ssid, sch->schib.pmcw.dev); PREPARE_WORK(&cdev->private->kick_work, - ccw_device_add_changed, (void *)cdev); + ccw_device_add_changed, cdev); queue_work(ccw_device_work, &cdev->private->kick_work); } @@ -685,7 +685,7 @@ io_subchannel_register(void *data) int ret; unsigned long flags; - cdev = (struct ccw_device *) data; + cdev = data; sch = to_subchannel(cdev->dev.parent); if (klist_node_attached(&cdev->dev.knode_parent)) { @@ -757,7 +757,7 @@ io_subchannel_recog_done(struct ccw_device *cdev) break; sch = to_subchannel(cdev->dev.parent); PREPARE_WORK(&cdev->private->kick_work, - ccw_device_call_sch_unregister, (void *) cdev); + ccw_device_call_sch_unregister, cdev); queue_work(slow_path_wq, &cdev->private->kick_work); if (atomic_dec_and_test(&ccw_device_init_count)) wake_up(&ccw_device_init_wq); @@ -772,7 +772,7 @@ io_subchannel_recog_done(struct ccw_device *cdev) if (!get_device(&cdev->dev)) break; PREPARE_WORK(&cdev->private->kick_work, - io_subchannel_register, (void *) cdev); + io_subchannel_register, cdev); queue_work(slow_path_wq, &cdev->private->kick_work); break; } @@ -910,7 +910,7 @@ io_subchannel_remove (struct subchannel *sch) */ if (get_device(&cdev->dev)) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_unregister, (void *) cdev); + ccw_device_unregister, cdev); queue_work(ccw_device_work, &cdev->private->kick_work); } return 0; @@ -1053,7 +1053,7 @@ __ccwdev_check_busid(struct device *dev, void *id) { char *bus_id; - bus_id = (char *)id; + bus_id = id; return (strncmp(bus_id, dev->bus_id, BUS_ID_SIZE) == 0); } diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index 44e4a53..fcaf28d 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -173,7 +173,7 @@ ccw_device_handle_oper(struct ccw_device *cdev) cdev->id.dev_model != cdev->private->senseid.dev_model || cdev->private->dev_id.devno != sch->schib.pmcw.dev) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_do_unreg_rereg, (void *)cdev); + ccw_device_do_unreg_rereg, cdev); queue_work(ccw_device_work, &cdev->private->kick_work); return 0; } @@ -314,13 +314,13 @@ ccw_device_oper_notify(void *data) struct subchannel *sch; int ret; - cdev = (struct ccw_device *)data; + cdev = data; sch = to_subchannel(cdev->dev.parent); ret = (sch->driver && sch->driver->notify) ? sch->driver->notify(&sch->dev, CIO_OPER) : 0; if (!ret) /* Driver doesn't want device back. */ - ccw_device_do_unreg_rereg((void *)cdev); + ccw_device_do_unreg_rereg(cdev); else { /* Reenable channel measurements, if needed. */ cmf_reenable(cdev); @@ -357,7 +357,7 @@ ccw_device_done(struct ccw_device *cdev, int state) if (cdev->private->flags.donotify) { cdev->private->flags.donotify = 0; PREPARE_WORK(&cdev->private->kick_work, ccw_device_oper_notify, - (void *)cdev); + cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); } wake_up(&cdev->private->wait_q); @@ -513,7 +513,7 @@ ccw_device_nopath_notify(void *data) struct subchannel *sch; int ret; - cdev = (struct ccw_device *)data; + cdev = data; sch = to_subchannel(cdev->dev.parent); /* Extra sanity. */ if (sch->lpm) @@ -527,7 +527,7 @@ ccw_device_nopath_notify(void *data) if (get_device(&cdev->dev)) { PREPARE_WORK(&cdev->private->kick_work, ccw_device_call_sch_unregister, - (void *)cdev); + cdev); queue_work(ccw_device_work, &cdev->private->kick_work); } else @@ -582,7 +582,7 @@ ccw_device_verify_done(struct ccw_device *cdev, int err) break; default: PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); + ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); ccw_device_done(cdev, DEV_STATE_NOT_OPER); break; @@ -713,7 +713,7 @@ ccw_device_offline_notoper(struct ccw_device *cdev, enum dev_event dev_event) sch = to_subchannel(cdev->dev.parent); if (get_device(&cdev->dev)) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_call_sch_unregister, (void *)cdev); + ccw_device_call_sch_unregister, cdev); queue_work(ccw_device_work, &cdev->private->kick_work); } wake_up(&cdev->private->wait_q); @@ -744,7 +744,7 @@ ccw_device_online_notoper(struct ccw_device *cdev, enum dev_event dev_event) } if (get_device(&cdev->dev)) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_call_sch_unregister, (void *)cdev); + ccw_device_call_sch_unregister, cdev); queue_work(ccw_device_work, &cdev->private->kick_work); } wake_up(&cdev->private->wait_q); @@ -849,7 +849,7 @@ ccw_device_online_timeout(struct ccw_device *cdev, enum dev_event dev_event) sch = to_subchannel(cdev->dev.parent); if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); + ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); } else @@ -938,7 +938,7 @@ ccw_device_killing_irq(struct ccw_device *cdev, enum dev_event dev_event) ERR_PTR(-EIO)); if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); + ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); } else if (cdev->private->flags.doverify) /* Start delayed path verification. */ @@ -961,7 +961,7 @@ ccw_device_killing_timeout(struct ccw_device *cdev, enum dev_event dev_event) sch = to_subchannel(cdev->dev.parent); if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); + ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); } else @@ -990,7 +990,7 @@ void device_kill_io(struct subchannel *sch) if (ret == -ENODEV) { if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); + ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); } else @@ -1002,7 +1002,7 @@ void device_kill_io(struct subchannel *sch) ERR_PTR(-EIO)); if (!sch->lpm) { PREPARE_WORK(&cdev->private->kick_work, - ccw_device_nopath_notify, (void *)cdev); + ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); } else /* Start delayed path verification. */ -- cgit v0.10.2 From 75e9de18f079a51fa987ef0703112d5bc125fdb7 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Wed, 11 Oct 2006 15:31:52 +0200 Subject: [S390] stacktrace bug. The latest kernel 2.6.19-rc1 triggers a bug in the s390 specific stack trace code when compiled with gcc 3.4. This patch fixes the latest lock dependency validator code (2.6.19-rc1) on s390 gcc 3.4. The variable sp was fixed to r15 (which is the stack pointer in the s390 abi) and assigned new values to r15. Therefore, gcc 3.4 assigns a new value to r15 and does not restore it on exit (r15 is supposed to be call save) - the kernel stack is broken. Avoid trouble by not assigning any new value to sp (r15). Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c index d9428a0..0d14a47 100644 --- a/arch/s390/kernel/stacktrace.c +++ b/arch/s390/kernel/stacktrace.c @@ -62,27 +62,26 @@ static inline unsigned long save_context_stack(struct stack_trace *trace, void save_stack_trace(struct stack_trace *trace, struct task_struct *task) { register unsigned long sp asm ("15"); - unsigned long orig_sp; + unsigned long orig_sp, new_sp; - sp &= PSW_ADDR_INSN; - orig_sp = sp; + orig_sp = sp & PSW_ADDR_INSN; - sp = save_context_stack(trace, &trace->skip, sp, + new_sp = save_context_stack(trace, &trace->skip, orig_sp, S390_lowcore.panic_stack - PAGE_SIZE, S390_lowcore.panic_stack); - if ((sp != orig_sp) && !trace->all_contexts) + if ((new_sp != orig_sp) && !trace->all_contexts) return; - sp = save_context_stack(trace, &trace->skip, sp, + new_sp = save_context_stack(trace, &trace->skip, new_sp, S390_lowcore.async_stack - ASYNC_SIZE, S390_lowcore.async_stack); - if ((sp != orig_sp) && !trace->all_contexts) + if ((new_sp != orig_sp) && !trace->all_contexts) return; if (task) - save_context_stack(trace, &trace->skip, sp, + save_context_stack(trace, &trace->skip, new_sp, (unsigned long) task_stack_page(task), (unsigned long) task_stack_page(task) + THREAD_SIZE); else - save_context_stack(trace, &trace->skip, sp, + save_context_stack(trace, &trace->skip, new_sp, S390_lowcore.thread_info, S390_lowcore.thread_info + THREAD_SIZE); return; -- cgit v0.10.2 From 7e491092e442b3f8c0d90d470b398fdb74703ec7 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:20:35 -0700 Subject: [PATCH] revert "nvidiafb: use generic ddc reading" Olaf reports that this gave him a black screen. Cc: Olaf Hering Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index daaa486..7a43020 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -701,7 +701,6 @@ config FB_NVIDIA depends on FB && PCI select I2C_ALGOBIT if FB_NVIDIA_I2C select I2C if FB_NVIDIA_I2C - select FB_DDC if FB_NVIDIA_I2C select FB_MODE_HELPERS select FB_CFB_FILLRECT select FB_CFB_COPYAREA diff --git a/drivers/video/nvidia/nv_i2c.c b/drivers/video/nvidia/nv_i2c.c index e48de3c..19eef3a 100644 --- a/drivers/video/nvidia/nv_i2c.c +++ b/drivers/video/nvidia/nv_i2c.c @@ -160,12 +160,51 @@ void nvidia_delete_i2c_busses(struct nvidia_par *par) } +static u8 *nvidia_do_probe_i2c_edid(struct nvidia_i2c_chan *chan) +{ + u8 start = 0x0; + struct i2c_msg msgs[] = { + { + .addr = 0x50, + .len = 1, + .buf = &start, + }, { + .addr = 0x50, + .flags = I2C_M_RD, + .len = EDID_LENGTH, + }, + }; + u8 *buf; + + if (!chan->par) + return NULL; + + buf = kmalloc(EDID_LENGTH, GFP_KERNEL); + if (!buf) { + dev_warn(&chan->par->pci_dev->dev, "Out of memory!\n"); + return NULL; + } + msgs[1].buf = buf; + + if (i2c_transfer(&chan->adapter, msgs, 2) == 2) + return buf; + dev_dbg(&chan->par->pci_dev->dev, "Unable to read EDID block.\n"); + kfree(buf); + return NULL; +} + int nvidia_probe_i2c_connector(struct fb_info *info, int conn, u8 **out_edid) { struct nvidia_par *par = info->par; - u8 *edid; - - edid = fb_ddc_read(&par->chan[conn - 1].adapter); + u8 *edid = NULL; + int i; + + for (i = 0; i < 3; i++) { + /* Do the real work */ + edid = nvidia_do_probe_i2c_edid(&par->chan[conn - 1]); + if (edid) + break; + } if (!edid && conn == 1) { /* try to get from firmware */ -- cgit v0.10.2 From 41716c7c21b15e7ecf14f0caf1eef3980707fb74 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Wed, 11 Oct 2006 01:20:37 -0700 Subject: [PATCH] null dereference in fs/jbd/journal.c Since commit d1807793e1e7e502e3dc047115e9dbc3b50e4534 we dereference a NULL pointer. Coverity id #1432. We set journal to NULL, and use it directly afterwards. Signed-off-by: Eric Sesterhenn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index c518dd8..b85c686 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -725,6 +725,7 @@ journal_t * journal_init_dev(struct block_device *bdev, __FUNCTION__); kfree(journal); journal = NULL; + goto out; } journal->j_dev = bdev; journal->j_fs_dev = fs_dev; @@ -735,7 +736,7 @@ journal_t * journal_init_dev(struct block_device *bdev, J_ASSERT(bh != NULL); journal->j_sb_buffer = bh; journal->j_superblock = (journal_superblock_t *)bh->b_data; - +out: return journal; } -- cgit v0.10.2 From 4e0fadfcf62e252d2b14de0e0927eb2830c0c28c Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Wed, 11 Oct 2006 01:20:37 -0700 Subject: [PATCH] IRQ: Fix AVR32 breakage Make the necessary changes to AVR32 required by the irq regs stuff. Signed-off-by: Haavard Skinnemoen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c index 3e56b9f..5a247ba 100644 --- a/arch/avr32/kernel/time.c +++ b/arch/avr32/kernel/time.c @@ -124,15 +124,15 @@ unsigned long long sched_clock(void) * * In UP mode, it is invoked from the (global) timer_interrupt. */ -static void local_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) +static void local_timer_interrupt(int irq, void *dev_id) { if (current->pid) - profile_tick(CPU_PROFILING, regs); - update_process_times(user_mode(regs)); + profile_tick(CPU_PROFILING); + update_process_times(user_mode(get_irq_regs())); } static irqreturn_t -timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) +timer_interrupt(int irq, void *dev_id) { unsigned int count; @@ -157,7 +157,7 @@ timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) * * SMP is not supported yet. */ - local_timer_interrupt(irq, dev_id, regs); + local_timer_interrupt(irq, dev_id); return IRQ_HANDLED; } diff --git a/arch/avr32/mach-at32ap/extint.c b/arch/avr32/mach-at32ap/extint.c index 7da9c5f..4dff1f9 100644 --- a/arch/avr32/mach-at32ap/extint.c +++ b/arch/avr32/mach-at32ap/extint.c @@ -102,8 +102,7 @@ struct irq_chip eim_chip = { .set_type = eim_set_irq_type, }; -static void demux_eim_irq(unsigned int irq, struct irq_desc *desc, - struct pt_regs *regs) +static void demux_eim_irq(unsigned int irq, struct irq_desc *desc) { struct at32_sm *sm = desc->handler_data; struct irq_desc *ext_desc; @@ -121,7 +120,7 @@ static void demux_eim_irq(unsigned int irq, struct irq_desc *desc, ext_irq = i + sm->eim_first_irq; ext_desc = irq_desc + ext_irq; - ext_desc->handle_irq(ext_irq, ext_desc, regs); + ext_desc->handle_irq(ext_irq, ext_desc); } spin_unlock(&sm->lock); diff --git a/arch/avr32/mach-at32ap/intc.c b/arch/avr32/mach-at32ap/intc.c index 74f8c9f..eb87a18 100644 --- a/arch/avr32/mach-at32ap/intc.c +++ b/arch/avr32/mach-at32ap/intc.c @@ -52,16 +52,19 @@ static struct intc intc0 = { asmlinkage void do_IRQ(int level, struct pt_regs *regs) { struct irq_desc *desc; + struct pt_regs *old_regs; unsigned int irq; unsigned long status_reg; local_irq_disable(); + old_regs = set_irq_regs(regs); + irq_enter(); irq = intc_readl(&intc0, INTCAUSE0 - 4 * level); desc = irq_desc + irq; - desc->handle_irq(irq, desc, regs); + desc->handle_irq(irq, desc); /* * Clear all interrupt level masks so that we may handle @@ -75,6 +78,8 @@ asmlinkage void do_IRQ(int level, struct pt_regs *regs) sysreg_write(SR, status_reg); irq_exit(); + + set_irq_regs(old_regs); } void __init init_IRQ(void) diff --git a/include/asm-avr32/irq_regs.h b/include/asm-avr32/irq_regs.h new file mode 100644 index 0000000..3dd9c0b --- /dev/null +++ b/include/asm-avr32/irq_regs.h @@ -0,0 +1 @@ +#include -- cgit v0.10.2 From 6391af174ad75f72e92043c1dd8302660a2fec58 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 11 Oct 2006 01:20:39 -0700 Subject: [PATCH] mm: use symbolic names instead of indices for zone initialisation Arch-independent zone-sizing is using indices instead of symbolic names to offset within an array related to zones (max_zone_pfns). The unintended impact is that ZONE_DMA and ZONE_NORMAL is initialised on powerpc instead of ZONE_DMA and ZONE_HIGHMEM when CONFIG_HIGHMEM is set. As a result, the the machine fails to boot but will boot with CONFIG_HIGHMEM turned off. The following patch properly initialises the max_zone_pfns[] array and uses symbolic names instead of indices in each architecture using arch-independent zone-sizing. Two users have successfully booted their powerpcs with it (one an ibook G4). It has also been boot tested on x86, x86_64, ppc64 and ia64. Please merge for 2.6.19-rc2. Credit to Benjamin Herrenschmidt for identifying the bug and rolling the first fix. Additional credit to Johannes Berg and Andreas Schwab for reporting the problem and testing on powerpc. Signed-off-by: Mel Gorman Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 000cf03..519e63c 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -1083,16 +1083,15 @@ static unsigned long __init setup_memory(void) void __init zone_sizes_init(void) { + unsigned long max_zone_pfns[MAX_NR_ZONES]; + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); + max_zone_pfns[ZONE_DMA] = + virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT; + max_zone_pfns[ZONE_NORMAL] = max_low_pfn; #ifdef CONFIG_HIGHMEM - unsigned long max_zone_pfns[MAX_NR_ZONES] = { - virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT, - max_low_pfn, - highend_pfn}; + max_zone_pfns[ZONE_HIGHMEM] = highend_pfn; add_active_range(0, 0, highend_pfn); #else - unsigned long max_zone_pfns[MAX_NR_ZONES] = { - virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT, - max_low_pfn}; add_active_range(0, 0, max_low_pfn); #endif diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c index 455597d..ddbdb03 100644 --- a/arch/i386/mm/discontig.c +++ b/arch/i386/mm/discontig.c @@ -356,11 +356,12 @@ void __init numa_kva_reserve(void) void __init zone_sizes_init(void) { int nid; - unsigned long max_zone_pfns[MAX_NR_ZONES] = { - virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT, - max_low_pfn, - highend_pfn - }; + unsigned long max_zone_pfns[MAX_NR_ZONES]; + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); + max_zone_pfns[ZONE_DMA] = + virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT; + max_zone_pfns[ZONE_NORMAL] = max_low_pfn; + max_zone_pfns[ZONE_HIGHMEM] = highend_pfn; /* If SRAT has not registered memory, register it now */ if (find_max_pfn_with_active_regions() == 0) { diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index daf977f..82deaa3 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -233,6 +233,7 @@ paging_init (void) efi_memmap_walk(count_pages, &num_physpages); max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT; + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); max_zone_pfns[ZONE_DMA] = max_dma; max_zone_pfns[ZONE_NORMAL] = max_low_pfn; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index d497b6b..96722cb 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -709,6 +709,7 @@ void __init paging_init(void) max_pfn = mem_data[node].max_pfn; } + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); max_zone_pfns[ZONE_DMA] = max_dma; max_zone_pfns[ZONE_NORMAL] = max_pfn; free_area_init_nodes(max_zone_pfns); diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 16fe027..d1c0758 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -307,11 +307,12 @@ void __init paging_init(void) top_of_ram, total_ram); printk(KERN_DEBUG "Memory hole size: %ldMB\n", (top_of_ram - total_ram) >> 20); + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); #ifdef CONFIG_HIGHMEM - max_zone_pfns[0] = total_lowmem >> PAGE_SHIFT; - max_zone_pfns[1] = top_of_ram >> PAGE_SHIFT; + max_zone_pfns[ZONE_DMA] = total_lowmem >> PAGE_SHIFT; + max_zone_pfns[ZONE_HIGHMEM] = top_of_ram >> PAGE_SHIFT; #else - max_zone_pfns[0] = top_of_ram >> PAGE_SHIFT; + max_zone_pfns[ZONE_DMA] = top_of_ram >> PAGE_SHIFT; #endif free_area_init_nodes(max_zone_pfns); } diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index 43c2720..9da01dc 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -617,9 +617,9 @@ void __init do_init_bootmem(void) void __init paging_init(void) { - unsigned long max_zone_pfns[MAX_NR_ZONES] = { - lmb_end_of_DRAM() >> PAGE_SHIFT - }; + unsigned long max_zone_pfns[MAX_NR_ZONES]; + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); + max_zone_pfns[ZONE_DMA] = lmb_end_of_DRAM() >> PAGE_SHIFT; free_area_init_nodes(max_zone_pfns); } diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c index 4102000..c374e53 100644 --- a/arch/ppc/mm/init.c +++ b/arch/ppc/mm/init.c @@ -374,11 +374,12 @@ void __init paging_init(void) end_pfn = start_pfn + (total_memory >> PAGE_SHIFT); add_active_range(0, start_pfn, end_pfn); + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); #ifdef CONFIG_HIGHMEM - max_zone_pfns[0] = total_lowmem >> PAGE_SHIFT; - max_zone_pfns[1] = total_memory >> PAGE_SHIFT; + max_zone_pfns[ZONE_DMA] = total_lowmem >> PAGE_SHIFT; + max_zone_pfns[ZONE_HIGHMEM] = total_memory >> PAGE_SHIFT; #else - max_zone_pfns[0] = total_memory >> PAGE_SHIFT; + max_zone_pfns[ZONE_DMA] = total_memory >> PAGE_SHIFT; #endif /* CONFIG_HIGHMEM */ free_area_init_nodes(max_zone_pfns); } diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 19c7252..971dc11 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -406,9 +406,12 @@ void __cpuinit zap_low_mappings(int cpu) #ifndef CONFIG_NUMA void __init paging_init(void) { - unsigned long max_zone_pfns[MAX_NR_ZONES] = {MAX_DMA_PFN, - MAX_DMA32_PFN, - end_pfn}; + unsigned long max_zone_pfns[MAX_NR_ZONES]; + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); + max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; + max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN; + max_zone_pfns[ZONE_NORMAL] = end_pfn; + memory_present(0, 0, end_pfn); sparse_init(); free_area_init_nodes(max_zone_pfns); diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c index 829a008b..2ee2e00 100644 --- a/arch/x86_64/mm/numa.c +++ b/arch/x86_64/mm/numa.c @@ -338,9 +338,11 @@ static void __init arch_sparse_init(void) void __init paging_init(void) { int i; - unsigned long max_zone_pfns[MAX_NR_ZONES] = { MAX_DMA_PFN, - MAX_DMA32_PFN, - end_pfn}; + unsigned long max_zone_pfns[MAX_NR_ZONES]; + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); + max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; + max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN; + max_zone_pfns[ZONE_NORMAL] = end_pfn; arch_sparse_init(); -- cgit v0.10.2 From b888132b0ff29ca0733589a594c243ed612438eb Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 11 Oct 2006 01:20:40 -0700 Subject: [PATCH] mm: remove memmap_zone_idx() memmap_zone_idx() is not used anymore. It was required by an earlier version of account-for-memmap-and-optionally-the-kernel-image-as-holes.patch but not any more. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index a8c003e..8246e83 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2294,19 +2294,6 @@ unsigned long __init zone_absent_pages_in_node(int nid, return __absent_pages_in_range(nid, zone_start_pfn, zone_end_pfn); } -/* Return the zone index a PFN is in */ -int memmap_zone_idx(struct page *lmem_map) -{ - int i; - unsigned long phys_addr = virt_to_phys(lmem_map); - unsigned long pfn = phys_addr >> PAGE_SHIFT; - - for (i = 0; i < MAX_NR_ZONES; i++) - if (pfn < arch_zone_highest_possible_pfn[i]) - break; - - return i; -} #else static inline unsigned long zone_spanned_pages_in_node(int nid, unsigned long zone_type, @@ -2325,10 +2312,6 @@ static inline unsigned long zone_absent_pages_in_node(int nid, return zholes_size[zone_type]; } -static inline int memmap_zone_idx(struct page *lmem_map) -{ - return MAX_NR_ZONES; -} #endif static void __init calculate_node_totalpages(struct pglist_data *pgdat, -- cgit v0.10.2 From 1db11ea65eaf1e7f6a70805fe52bc0aef53f68ef Mon Sep 17 00:00:00 2001 From: Luca Tettamanti Date: Wed, 11 Oct 2006 01:20:41 -0700 Subject: [PATCH] Fix menuconfig build failure due to missing stdbool.h scripts/kconfig/lxdialog/util.c fails to build because it uses true/false without including stdbool.h: kronos:~/src/linux-2.6$ make O=../linux-build-git menuconfig GEN /home/kronos/src/linux-build/Makefile HOSTCC scripts/kconfig/lxdialog/util.o /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c: In function 'set_classic_theme': /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:68: error: 'true' undeclared (first use in this function) /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:68: error: (Each undeclared identifier is reported only once /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:68: error: for each function it appears in.) /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:70: error: 'false' undeclared (first use in this function) /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c: In function 'set_blackbg_theme': /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:101: error: 'true' undeclared (first use in this function) /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:102: error: 'false' undeclared (first use in this function) /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c: In function 'set_bluetitle_theme': /home/kronos/src/linux-2.6/scripts/kconfig/lxdialog/util.c:144: error: 'true' undeclared (first use in this function) make[2]: *** [scripts/kconfig/lxdialog/util.o] Error 1 make[1]: *** [menuconfig] Error 2 make: *** [menuconfig] Error 2 Add to dialog.h to fix the breakage. Signed-off-by: Luca Tettamanti Cc: Sam Ravnborg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/kconfig/lxdialog/dialog.h b/scripts/kconfig/lxdialog/dialog.h index 8dea47f..fd695e1 100644 --- a/scripts/kconfig/lxdialog/dialog.h +++ b/scripts/kconfig/lxdialog/dialog.h @@ -24,6 +24,7 @@ #include #include #include +#include #ifdef __sun__ #define CURS_MACROS -- cgit v0.10.2 From c37e108d156101dcde7ec7033eabe7abe83366bc Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Wed, 11 Oct 2006 01:20:43 -0700 Subject: [PATCH] use struct irq_chip instead of struct hw_interrupt_type hw_interrupt_type is deprecated in favour of struct irq_chip. [mingo@elte.hu: do x86_64 too] Acked-by: Thomas Gleixner Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index cd082c3..27bceaf 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2594,7 +2594,7 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) } #endif -static struct hw_interrupt_type ht_irq_chip = { +static struct irq_chip ht_irq_chip = { .name = "PCI-HT", .mask = mask_ht_irq, .unmask = unmask_ht_irq, diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 771bcf7..c3cdcab 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1897,7 +1897,7 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) } #endif -static struct hw_interrupt_type ht_irq_chip = { +static struct irq_chip ht_irq_chip = { .name = "PCI-HT", .mask = mask_ht_irq, .unmask = unmask_ht_irq, -- cgit v0.10.2 From dea20a3fbdd08e5ae2a0b33d2577c794a3764a11 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 11 Oct 2006 01:20:44 -0700 Subject: [PATCH] Disable DETECT_SOFTLOCKUP for s390 We got several false bug reports because of enabled CONFIG_DETECT_SOFTLOCKUP. Disable soft lockup detection on s390, since it doesn't work on a virtualized architecture. Cc: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 756a908..2d7cb0d 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -71,7 +71,7 @@ config LOG_BUF_SHIFT config DETECT_SOFTLOCKUP bool "Detect Soft Lockups" - depends on DEBUG_KERNEL + depends on DEBUG_KERNEL && !S390 default y help Say Y here to enable the kernel to detect "soft lockups", -- cgit v0.10.2 From 97c7801cd5b0bb6a38c16108a496235474dc6310 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 11 Oct 2006 01:20:45 -0700 Subject: [PATCH] swsusp: Use suspend_console Add suspend_console() and resume_console() to the suspend-to-disk code paths so that the users of netconsole can use swsusp with it. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/power/disk.c b/kernel/power/disk.c index d722349..d3a158a 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include "power.h" @@ -119,8 +120,10 @@ int pm_suspend_disk(void) if (error) return error; + suspend_console(); error = device_suspend(PMSG_FREEZE); if (error) { + resume_console(); printk("Some devices failed to suspend\n"); unprepare_processes(); return error; @@ -133,6 +136,7 @@ int pm_suspend_disk(void) if (in_suspend) { device_resume(); + resume_console(); pr_debug("PM: writing image.\n"); error = swsusp_write(); if (!error) @@ -148,6 +152,7 @@ int pm_suspend_disk(void) swsusp_free(); Done: device_resume(); + resume_console(); unprepare_processes(); return error; } @@ -212,7 +217,9 @@ static int software_resume(void) pr_debug("PM: Preparing devices for restore.\n"); + suspend_console(); if ((error = device_suspend(PMSG_PRETHAW))) { + resume_console(); printk("Some devices failed to suspend\n"); swsusp_free(); goto Thaw; @@ -224,6 +231,7 @@ static int software_resume(void) swsusp_resume(); pr_debug("PM: Restore failed, recovering.n"); device_resume(); + resume_console(); Thaw: unprepare_processes(); Done: diff --git a/kernel/power/user.c b/kernel/power/user.c index 93b5dd2..d991d3b 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -173,12 +174,14 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, /* Free memory before shutting down devices. */ error = swsusp_shrink_memory(); if (!error) { + suspend_console(); error = device_suspend(PMSG_FREEZE); if (!error) { in_suspend = 1; error = swsusp_suspend(); device_resume(); } + resume_console(); } up(&pm_sem); if (!error) @@ -196,11 +199,13 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, snapshot_free_unused_memory(&data->handle); down(&pm_sem); pm_prepare_console(); + suspend_console(); error = device_suspend(PMSG_PRETHAW); if (!error) { error = swsusp_resume(); device_resume(); } + resume_console(); pm_restore_console(); up(&pm_sem); break; @@ -289,6 +294,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, } /* Put devices to sleep */ + suspend_console(); error = device_suspend(PMSG_SUSPEND); if (error) { printk(KERN_ERR "Failed to suspend some devices.\n"); @@ -299,7 +305,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, /* Wake up devices */ device_resume(); } - + resume_console(); if (pm_ops->finish) pm_ops->finish(PM_SUSPEND_MEM); -- cgit v0.10.2 From 502717f4e112b18d9c37753a32f675bec9f2838b Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Wed, 11 Oct 2006 01:20:46 -0700 Subject: [PATCH] hugetlb: fix linked list corruption in unmap_hugepage_range() commit fe1668ae5bf0145014c71797febd9ad5670d5d05 causes kernel to oops with libhugetlbfs test suite. The problem is that hugetlb pages can be shared by multiple mappings. Multiple threads can fight over page->lru in the unmap path and bad things happen. We now serialize __unmap_hugepage_range to void concurrent linked list manipulation. Such serialization is also needed for shared page table page on hugetlb area. This patch will fixed the bug and also serve as a prepatch for shared page table. Signed-off-by: Ken Chen Cc: Hugh Dickins Cc: David Gibson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 5e03b2f..4ee3f00 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -293,7 +293,7 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) if (h_vm_pgoff >= h_pgoff) v_offset = 0; - unmap_hugepage_range(vma, + __unmap_hugepage_range(vma, vma->vm_start + v_offset, vma->vm_end); } } diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index c25a38d..5081d27 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -17,6 +17,7 @@ int hugetlb_sysctl_handler(struct ctl_table *, int, struct file *, void __user * int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *); int follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *, struct page **, struct vm_area_struct **, unsigned long *, int *, int); void unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long); +void __unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long); int hugetlb_prefault(struct address_space *, struct vm_area_struct *); int hugetlb_report_meminfo(char *); int hugetlb_report_node_meminfo(int, char *); diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 1d709ff..2dbec90 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -356,8 +356,8 @@ nomem: return -ENOMEM; } -void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, - unsigned long end) +void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) { struct mm_struct *mm = vma->vm_mm; unsigned long address; @@ -398,6 +398,24 @@ void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, } } +void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + /* + * It is undesirable to test vma->vm_file as it should be non-null + * for valid hugetlb area. However, vm_file will be NULL in the error + * cleanup path of do_mmap_pgoff. When hugetlbfs ->mmap method fails, + * do_mmap_pgoff() nullifies vma->vm_file before calling this function + * to clean up. Since no pte has actually been setup, it is safe to + * do nothing in this case. + */ + if (vma->vm_file) { + spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); + __unmap_hugepage_range(vma, start, end); + spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); + } +} + static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *ptep, pte_t pte) { -- cgit v0.10.2 From ac27a0ec112a089f1a5102bc8dffc79c8c815571 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Wed, 11 Oct 2006 01:20:50 -0700 Subject: [PATCH] ext4: initial copy of files from ext3 Start of the ext4 patch series. See Documentation/filesystems/ext4.txt for details. This is a simple copy of the files in fs/ext3 to fs/ext4 and /usr/incude/linux/ext3* to /usr/include/ex4* Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile new file mode 100644 index 0000000..704cd44 --- /dev/null +++ b/fs/ext4/Makefile @@ -0,0 +1,12 @@ +# +# Makefile for the linux ext3-filesystem routines. +# + +obj-$(CONFIG_EXT3_FS) += ext3.o + +ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ + ioctl.o namei.o super.o symlink.o hash.o resize.o + +ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o +ext3-$(CONFIG_EXT3_FS_POSIX_ACL) += acl.o +ext3-$(CONFIG_EXT3_FS_SECURITY) += xattr_security.o diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c new file mode 100644 index 0000000..1e5038d --- /dev/null +++ b/fs/ext4/acl.c @@ -0,0 +1,551 @@ +/* + * linux/fs/ext3/acl.c + * + * Copyright (C) 2001-2003 Andreas Gruenbacher, + */ + +#include +#include +#include +#include +#include +#include +#include +#include "xattr.h" +#include "acl.h" + +/* + * Convert from filesystem to in-memory representation. + */ +static struct posix_acl * +ext3_acl_from_disk(const void *value, size_t size) +{ + const char *end = (char *)value + size; + int n, count; + struct posix_acl *acl; + + if (!value) + return NULL; + if (size < sizeof(ext3_acl_header)) + return ERR_PTR(-EINVAL); + if (((ext3_acl_header *)value)->a_version != + cpu_to_le32(EXT3_ACL_VERSION)) + return ERR_PTR(-EINVAL); + value = (char *)value + sizeof(ext3_acl_header); + count = ext3_acl_count(size); + if (count < 0) + return ERR_PTR(-EINVAL); + if (count == 0) + return NULL; + acl = posix_acl_alloc(count, GFP_KERNEL); + if (!acl) + return ERR_PTR(-ENOMEM); + for (n=0; n < count; n++) { + ext3_acl_entry *entry = + (ext3_acl_entry *)value; + if ((char *)value + sizeof(ext3_acl_entry_short) > end) + goto fail; + acl->a_entries[n].e_tag = le16_to_cpu(entry->e_tag); + acl->a_entries[n].e_perm = le16_to_cpu(entry->e_perm); + switch(acl->a_entries[n].e_tag) { + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + value = (char *)value + + sizeof(ext3_acl_entry_short); + acl->a_entries[n].e_id = ACL_UNDEFINED_ID; + break; + + case ACL_USER: + case ACL_GROUP: + value = (char *)value + sizeof(ext3_acl_entry); + if ((char *)value > end) + goto fail; + acl->a_entries[n].e_id = + le32_to_cpu(entry->e_id); + break; + + default: + goto fail; + } + } + if (value != end) + goto fail; + return acl; + +fail: + posix_acl_release(acl); + return ERR_PTR(-EINVAL); +} + +/* + * Convert from in-memory to filesystem representation. + */ +static void * +ext3_acl_to_disk(const struct posix_acl *acl, size_t *size) +{ + ext3_acl_header *ext_acl; + char *e; + size_t n; + + *size = ext3_acl_size(acl->a_count); + ext_acl = kmalloc(sizeof(ext3_acl_header) + acl->a_count * + sizeof(ext3_acl_entry), GFP_KERNEL); + if (!ext_acl) + return ERR_PTR(-ENOMEM); + ext_acl->a_version = cpu_to_le32(EXT3_ACL_VERSION); + e = (char *)ext_acl + sizeof(ext3_acl_header); + for (n=0; n < acl->a_count; n++) { + ext3_acl_entry *entry = (ext3_acl_entry *)e; + entry->e_tag = cpu_to_le16(acl->a_entries[n].e_tag); + entry->e_perm = cpu_to_le16(acl->a_entries[n].e_perm); + switch(acl->a_entries[n].e_tag) { + case ACL_USER: + case ACL_GROUP: + entry->e_id = + cpu_to_le32(acl->a_entries[n].e_id); + e += sizeof(ext3_acl_entry); + break; + + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + e += sizeof(ext3_acl_entry_short); + break; + + default: + goto fail; + } + } + return (char *)ext_acl; + +fail: + kfree(ext_acl); + return ERR_PTR(-EINVAL); +} + +static inline struct posix_acl * +ext3_iget_acl(struct inode *inode, struct posix_acl **i_acl) +{ + struct posix_acl *acl = EXT3_ACL_NOT_CACHED; + + spin_lock(&inode->i_lock); + if (*i_acl != EXT3_ACL_NOT_CACHED) + acl = posix_acl_dup(*i_acl); + spin_unlock(&inode->i_lock); + + return acl; +} + +static inline void +ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl, + struct posix_acl *acl) +{ + spin_lock(&inode->i_lock); + if (*i_acl != EXT3_ACL_NOT_CACHED) + posix_acl_release(*i_acl); + *i_acl = posix_acl_dup(acl); + spin_unlock(&inode->i_lock); +} + +/* + * Inode operation get_posix_acl(). + * + * inode->i_mutex: don't care + */ +static struct posix_acl * +ext3_get_acl(struct inode *inode, int type) +{ + struct ext3_inode_info *ei = EXT3_I(inode); + int name_index; + char *value = NULL; + struct posix_acl *acl; + int retval; + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return NULL; + + switch(type) { + case ACL_TYPE_ACCESS: + acl = ext3_iget_acl(inode, &ei->i_acl); + if (acl != EXT3_ACL_NOT_CACHED) + return acl; + name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS; + break; + + case ACL_TYPE_DEFAULT: + acl = ext3_iget_acl(inode, &ei->i_default_acl); + if (acl != EXT3_ACL_NOT_CACHED) + return acl; + name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT; + break; + + default: + return ERR_PTR(-EINVAL); + } + retval = ext3_xattr_get(inode, name_index, "", NULL, 0); + if (retval > 0) { + value = kmalloc(retval, GFP_KERNEL); + if (!value) + return ERR_PTR(-ENOMEM); + retval = ext3_xattr_get(inode, name_index, "", value, retval); + } + if (retval > 0) + acl = ext3_acl_from_disk(value, retval); + else if (retval == -ENODATA || retval == -ENOSYS) + acl = NULL; + else + acl = ERR_PTR(retval); + kfree(value); + + if (!IS_ERR(acl)) { + switch(type) { + case ACL_TYPE_ACCESS: + ext3_iset_acl(inode, &ei->i_acl, acl); + break; + + case ACL_TYPE_DEFAULT: + ext3_iset_acl(inode, &ei->i_default_acl, acl); + break; + } + } + return acl; +} + +/* + * Set the access or default ACL of an inode. + * + * inode->i_mutex: down unless called from ext3_new_inode + */ +static int +ext3_set_acl(handle_t *handle, struct inode *inode, int type, + struct posix_acl *acl) +{ + struct ext3_inode_info *ei = EXT3_I(inode); + int name_index; + void *value = NULL; + size_t size = 0; + int error; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + + switch(type) { + case ACL_TYPE_ACCESS: + name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS; + if (acl) { + mode_t mode = inode->i_mode; + error = posix_acl_equiv_mode(acl, &mode); + if (error < 0) + return error; + else { + inode->i_mode = mode; + ext3_mark_inode_dirty(handle, inode); + if (error == 0) + acl = NULL; + } + } + break; + + case ACL_TYPE_DEFAULT: + name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT; + if (!S_ISDIR(inode->i_mode)) + return acl ? -EACCES : 0; + break; + + default: + return -EINVAL; + } + if (acl) { + value = ext3_acl_to_disk(acl, &size); + if (IS_ERR(value)) + return (int)PTR_ERR(value); + } + + error = ext3_xattr_set_handle(handle, inode, name_index, "", + value, size, 0); + + kfree(value); + if (!error) { + switch(type) { + case ACL_TYPE_ACCESS: + ext3_iset_acl(inode, &ei->i_acl, acl); + break; + + case ACL_TYPE_DEFAULT: + ext3_iset_acl(inode, &ei->i_default_acl, acl); + break; + } + } + return error; +} + +static int +ext3_check_acl(struct inode *inode, int mask) +{ + struct posix_acl *acl = ext3_get_acl(inode, ACL_TYPE_ACCESS); + + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl) { + int error = posix_acl_permission(inode, acl, mask); + posix_acl_release(acl); + return error; + } + + return -EAGAIN; +} + +int +ext3_permission(struct inode *inode, int mask, struct nameidata *nd) +{ + return generic_permission(inode, mask, ext3_check_acl); +} + +/* + * Initialize the ACLs of a new inode. Called from ext3_new_inode. + * + * dir->i_mutex: down + * inode->i_mutex: up (access to inode is still exclusive) + */ +int +ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) +{ + struct posix_acl *acl = NULL; + int error = 0; + + if (!S_ISLNK(inode->i_mode)) { + if (test_opt(dir->i_sb, POSIX_ACL)) { + acl = ext3_get_acl(dir, ACL_TYPE_DEFAULT); + if (IS_ERR(acl)) + return PTR_ERR(acl); + } + if (!acl) + inode->i_mode &= ~current->fs->umask; + } + if (test_opt(inode->i_sb, POSIX_ACL) && acl) { + struct posix_acl *clone; + mode_t mode; + + if (S_ISDIR(inode->i_mode)) { + error = ext3_set_acl(handle, inode, + ACL_TYPE_DEFAULT, acl); + if (error) + goto cleanup; + } + clone = posix_acl_clone(acl, GFP_KERNEL); + error = -ENOMEM; + if (!clone) + goto cleanup; + + mode = inode->i_mode; + error = posix_acl_create_masq(clone, &mode); + if (error >= 0) { + inode->i_mode = mode; + if (error > 0) { + /* This is an extended ACL */ + error = ext3_set_acl(handle, inode, + ACL_TYPE_ACCESS, clone); + } + } + posix_acl_release(clone); + } +cleanup: + posix_acl_release(acl); + return error; +} + +/* + * Does chmod for an inode that may have an Access Control List. The + * inode->i_mode field must be updated to the desired value by the caller + * before calling this function. + * Returns 0 on success, or a negative error number. + * + * We change the ACL rather than storing some ACL entries in the file + * mode permission bits (which would be more efficient), because that + * would break once additional permissions (like ACL_APPEND, ACL_DELETE + * for directories) are added. There are no more bits available in the + * file mode. + * + * inode->i_mutex: down + */ +int +ext3_acl_chmod(struct inode *inode) +{ + struct posix_acl *acl, *clone; + int error; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + if (!test_opt(inode->i_sb, POSIX_ACL)) + return 0; + acl = ext3_get_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(acl) || !acl) + return PTR_ERR(acl); + clone = posix_acl_clone(acl, GFP_KERNEL); + posix_acl_release(acl); + if (!clone) + return -ENOMEM; + error = posix_acl_chmod_masq(clone, inode->i_mode); + if (!error) { + handle_t *handle; + int retries = 0; + + retry: + handle = ext3_journal_start(inode, + EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + ext3_std_error(inode->i_sb, error); + goto out; + } + error = ext3_set_acl(handle, inode, ACL_TYPE_ACCESS, clone); + ext3_journal_stop(handle); + if (error == -ENOSPC && + ext3_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + } +out: + posix_acl_release(clone); + return error; +} + +/* + * Extended attribute handlers + */ +static size_t +ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len, + const char *name, size_t name_len) +{ + const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS); + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return 0; + if (list && size <= list_len) + memcpy(list, POSIX_ACL_XATTR_ACCESS, size); + return size; +} + +static size_t +ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len, + const char *name, size_t name_len) +{ + const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT); + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return 0; + if (list && size <= list_len) + memcpy(list, POSIX_ACL_XATTR_DEFAULT, size); + return size; +} + +static int +ext3_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size) +{ + struct posix_acl *acl; + int error; + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return -EOPNOTSUPP; + + acl = ext3_get_acl(inode, type); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl == NULL) + return -ENODATA; + error = posix_acl_to_xattr(acl, buffer, size); + posix_acl_release(acl); + + return error; +} + +static int +ext3_xattr_get_acl_access(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext3_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size); +} + +static int +ext3_xattr_get_acl_default(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext3_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size); +} + +static int +ext3_xattr_set_acl(struct inode *inode, int type, const void *value, + size_t size) +{ + handle_t *handle; + struct posix_acl *acl; + int error, retries = 0; + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return -EOPNOTSUPP; + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EPERM; + + if (value) { + acl = posix_acl_from_xattr(value, size); + if (IS_ERR(acl)) + return PTR_ERR(acl); + else if (acl) { + error = posix_acl_valid(acl); + if (error) + goto release_and_out; + } + } else + acl = NULL; + +retry: + handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + error = ext3_set_acl(handle, inode, type, acl); + ext3_journal_stop(handle); + if (error == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + +release_and_out: + posix_acl_release(acl); + return error; +} + +static int +ext3_xattr_set_acl_access(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext3_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size); +} + +static int +ext3_xattr_set_acl_default(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext3_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size); +} + +struct xattr_handler ext3_xattr_acl_access_handler = { + .prefix = POSIX_ACL_XATTR_ACCESS, + .list = ext3_xattr_list_acl_access, + .get = ext3_xattr_get_acl_access, + .set = ext3_xattr_set_acl_access, +}; + +struct xattr_handler ext3_xattr_acl_default_handler = { + .prefix = POSIX_ACL_XATTR_DEFAULT, + .list = ext3_xattr_list_acl_default, + .get = ext3_xattr_get_acl_default, + .set = ext3_xattr_set_acl_default, +}; diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h new file mode 100644 index 0000000..0d1e627 --- /dev/null +++ b/fs/ext4/acl.h @@ -0,0 +1,81 @@ +/* + File: fs/ext3/acl.h + + (C) 2001 Andreas Gruenbacher, +*/ + +#include + +#define EXT3_ACL_VERSION 0x0001 + +typedef struct { + __le16 e_tag; + __le16 e_perm; + __le32 e_id; +} ext3_acl_entry; + +typedef struct { + __le16 e_tag; + __le16 e_perm; +} ext3_acl_entry_short; + +typedef struct { + __le32 a_version; +} ext3_acl_header; + +static inline size_t ext3_acl_size(int count) +{ + if (count <= 4) { + return sizeof(ext3_acl_header) + + count * sizeof(ext3_acl_entry_short); + } else { + return sizeof(ext3_acl_header) + + 4 * sizeof(ext3_acl_entry_short) + + (count - 4) * sizeof(ext3_acl_entry); + } +} + +static inline int ext3_acl_count(size_t size) +{ + ssize_t s; + size -= sizeof(ext3_acl_header); + s = size - 4 * sizeof(ext3_acl_entry_short); + if (s < 0) { + if (size % sizeof(ext3_acl_entry_short)) + return -1; + return size / sizeof(ext3_acl_entry_short); + } else { + if (s % sizeof(ext3_acl_entry)) + return -1; + return s / sizeof(ext3_acl_entry) + 4; + } +} + +#ifdef CONFIG_EXT3_FS_POSIX_ACL + +/* Value for inode->u.ext3_i.i_acl and inode->u.ext3_i.i_default_acl + if the ACL has not been cached */ +#define EXT3_ACL_NOT_CACHED ((void *)-1) + +/* acl.c */ +extern int ext3_permission (struct inode *, int, struct nameidata *); +extern int ext3_acl_chmod (struct inode *); +extern int ext3_init_acl (handle_t *, struct inode *, struct inode *); + +#else /* CONFIG_EXT3_FS_POSIX_ACL */ +#include +#define ext3_permission NULL + +static inline int +ext3_acl_chmod(struct inode *inode) +{ + return 0; +} + +static inline int +ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) +{ + return 0; +} +#endif /* CONFIG_EXT3_FS_POSIX_ACL */ + diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c new file mode 100644 index 0000000..b41a7d7 --- /dev/null +++ b/fs/ext4/balloc.c @@ -0,0 +1,1818 @@ +/* + * linux/fs/ext3/balloc.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * Enhanced block allocation by Stephen Tweedie (sct@redhat.com), 1993 + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * balloc.c contains the blocks allocation and deallocation routines + */ + +/* + * The free blocks are managed by bitmaps. A file system contains several + * blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap + * block for inodes, N blocks for the inode table and data blocks. + * + * The file system contains group descriptors which are located after the + * super block. Each descriptor contains the number of the bitmap block and + * the free blocks count in the block. The descriptors are loaded in memory + * when a file system is mounted (see ext3_read_super). + */ + + +#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) + +/** + * ext3_get_group_desc() -- load group descriptor from disk + * @sb: super block + * @block_group: given block group + * @bh: pointer to the buffer head to store the block + * group descriptor + */ +struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, + unsigned int block_group, + struct buffer_head ** bh) +{ + unsigned long group_desc; + unsigned long offset; + struct ext3_group_desc * desc; + struct ext3_sb_info *sbi = EXT3_SB(sb); + + if (block_group >= sbi->s_groups_count) { + ext3_error (sb, "ext3_get_group_desc", + "block_group >= groups_count - " + "block_group = %d, groups_count = %lu", + block_group, sbi->s_groups_count); + + return NULL; + } + smp_rmb(); + + group_desc = block_group >> EXT3_DESC_PER_BLOCK_BITS(sb); + offset = block_group & (EXT3_DESC_PER_BLOCK(sb) - 1); + if (!sbi->s_group_desc[group_desc]) { + ext3_error (sb, "ext3_get_group_desc", + "Group descriptor not loaded - " + "block_group = %d, group_desc = %lu, desc = %lu", + block_group, group_desc, offset); + return NULL; + } + + desc = (struct ext3_group_desc *) sbi->s_group_desc[group_desc]->b_data; + if (bh) + *bh = sbi->s_group_desc[group_desc]; + return desc + offset; +} + +/** + * read_block_bitmap() + * @sb: super block + * @block_group: given block group + * + * Read the bitmap for a given block_group, reading into the specified + * slot in the superblock's bitmap cache. + * + * Return buffer_head on success or NULL in case of failure. + */ +static struct buffer_head * +read_block_bitmap(struct super_block *sb, unsigned int block_group) +{ + struct ext3_group_desc * desc; + struct buffer_head * bh = NULL; + + desc = ext3_get_group_desc (sb, block_group, NULL); + if (!desc) + goto error_out; + bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); + if (!bh) + ext3_error (sb, "read_block_bitmap", + "Cannot read block bitmap - " + "block_group = %d, block_bitmap = %u", + block_group, le32_to_cpu(desc->bg_block_bitmap)); +error_out: + return bh; +} +/* + * The reservation window structure operations + * -------------------------------------------- + * Operations include: + * dump, find, add, remove, is_empty, find_next_reservable_window, etc. + * + * We use a red-black tree to represent per-filesystem reservation + * windows. + * + */ + +/** + * __rsv_window_dump() -- Dump the filesystem block allocation reservation map + * @rb_root: root of per-filesystem reservation rb tree + * @verbose: verbose mode + * @fn: function which wishes to dump the reservation map + * + * If verbose is turned on, it will print the whole block reservation + * windows(start, end). Otherwise, it will only print out the "bad" windows, + * those windows that overlap with their immediate neighbors. + */ +#if 1 +static void __rsv_window_dump(struct rb_root *root, int verbose, + const char *fn) +{ + struct rb_node *n; + struct ext3_reserve_window_node *rsv, *prev; + int bad; + +restart: + n = rb_first(root); + bad = 0; + prev = NULL; + + printk("Block Allocation Reservation Windows Map (%s):\n", fn); + while (n) { + rsv = list_entry(n, struct ext3_reserve_window_node, rsv_node); + if (verbose) + printk("reservation window 0x%p " + "start: %lu, end: %lu\n", + rsv, rsv->rsv_start, rsv->rsv_end); + if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) { + printk("Bad reservation %p (start >= end)\n", + rsv); + bad = 1; + } + if (prev && prev->rsv_end >= rsv->rsv_start) { + printk("Bad reservation %p (prev->end >= start)\n", + rsv); + bad = 1; + } + if (bad) { + if (!verbose) { + printk("Restarting reservation walk in verbose mode\n"); + verbose = 1; + goto restart; + } + } + n = rb_next(n); + prev = rsv; + } + printk("Window map complete.\n"); + if (bad) + BUG(); +} +#define rsv_window_dump(root, verbose) \ + __rsv_window_dump((root), (verbose), __FUNCTION__) +#else +#define rsv_window_dump(root, verbose) do {} while (0) +#endif + +/** + * goal_in_my_reservation() + * @rsv: inode's reservation window + * @grp_goal: given goal block relative to the allocation block group + * @group: the current allocation block group + * @sb: filesystem super block + * + * Test if the given goal block (group relative) is within the file's + * own block reservation window range. + * + * If the reservation window is outside the goal allocation group, return 0; + * grp_goal (given goal block) could be -1, which means no specific + * goal block. In this case, always return 1. + * If the goal block is within the reservation window, return 1; + * otherwise, return 0; + */ +static int +goal_in_my_reservation(struct ext3_reserve_window *rsv, ext3_grpblk_t grp_goal, + unsigned int group, struct super_block * sb) +{ + ext3_fsblk_t group_first_block, group_last_block; + + group_first_block = ext3_group_first_block_no(sb, group); + group_last_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + + if ((rsv->_rsv_start > group_last_block) || + (rsv->_rsv_end < group_first_block)) + return 0; + if ((grp_goal >= 0) && ((grp_goal + group_first_block < rsv->_rsv_start) + || (grp_goal + group_first_block > rsv->_rsv_end))) + return 0; + return 1; +} + +/** + * search_reserve_window() + * @rb_root: root of reservation tree + * @goal: target allocation block + * + * Find the reserved window which includes the goal, or the previous one + * if the goal is not in any window. + * Returns NULL if there are no windows or if all windows start after the goal. + */ +static struct ext3_reserve_window_node * +search_reserve_window(struct rb_root *root, ext3_fsblk_t goal) +{ + struct rb_node *n = root->rb_node; + struct ext3_reserve_window_node *rsv; + + if (!n) + return NULL; + + do { + rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); + + if (goal < rsv->rsv_start) + n = n->rb_left; + else if (goal > rsv->rsv_end) + n = n->rb_right; + else + return rsv; + } while (n); + /* + * We've fallen off the end of the tree: the goal wasn't inside + * any particular node. OK, the previous node must be to one + * side of the interval containing the goal. If it's the RHS, + * we need to back up one. + */ + if (rsv->rsv_start > goal) { + n = rb_prev(&rsv->rsv_node); + rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); + } + return rsv; +} + +/** + * ext3_rsv_window_add() -- Insert a window to the block reservation rb tree. + * @sb: super block + * @rsv: reservation window to add + * + * Must be called with rsv_lock hold. + */ +void ext3_rsv_window_add(struct super_block *sb, + struct ext3_reserve_window_node *rsv) +{ + struct rb_root *root = &EXT3_SB(sb)->s_rsv_window_root; + struct rb_node *node = &rsv->rsv_node; + ext3_fsblk_t start = rsv->rsv_start; + + struct rb_node ** p = &root->rb_node; + struct rb_node * parent = NULL; + struct ext3_reserve_window_node *this; + + while (*p) + { + parent = *p; + this = rb_entry(parent, struct ext3_reserve_window_node, rsv_node); + + if (start < this->rsv_start) + p = &(*p)->rb_left; + else if (start > this->rsv_end) + p = &(*p)->rb_right; + else { + rsv_window_dump(root, 1); + BUG(); + } + } + + rb_link_node(node, parent, p); + rb_insert_color(node, root); +} + +/** + * ext3_rsv_window_remove() -- unlink a window from the reservation rb tree + * @sb: super block + * @rsv: reservation window to remove + * + * Mark the block reservation window as not allocated, and unlink it + * from the filesystem reservation window rb tree. Must be called with + * rsv_lock hold. + */ +static void rsv_window_remove(struct super_block *sb, + struct ext3_reserve_window_node *rsv) +{ + rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_alloc_hit = 0; + rb_erase(&rsv->rsv_node, &EXT3_SB(sb)->s_rsv_window_root); +} + +/* + * rsv_is_empty() -- Check if the reservation window is allocated. + * @rsv: given reservation window to check + * + * returns 1 if the end block is EXT3_RESERVE_WINDOW_NOT_ALLOCATED. + */ +static inline int rsv_is_empty(struct ext3_reserve_window *rsv) +{ + /* a valid reservation end block could not be 0 */ + return rsv->_rsv_end == EXT3_RESERVE_WINDOW_NOT_ALLOCATED; +} + +/** + * ext3_init_block_alloc_info() + * @inode: file inode structure + * + * Allocate and initialize the reservation window structure, and + * link the window to the ext3 inode structure at last + * + * The reservation window structure is only dynamically allocated + * and linked to ext3 inode the first time the open file + * needs a new block. So, before every ext3_new_block(s) call, for + * regular files, we should check whether the reservation window + * structure exists or not. In the latter case, this function is called. + * Fail to do so will result in block reservation being turned off for that + * open file. + * + * This function is called from ext3_get_blocks_handle(), also called + * when setting the reservation window size through ioctl before the file + * is open for write (needs block allocation). + * + * Needs truncate_mutex protection prior to call this function. + */ +void ext3_init_block_alloc_info(struct inode *inode) +{ + struct ext3_inode_info *ei = EXT3_I(inode); + struct ext3_block_alloc_info *block_i = ei->i_block_alloc_info; + struct super_block *sb = inode->i_sb; + + block_i = kmalloc(sizeof(*block_i), GFP_NOFS); + if (block_i) { + struct ext3_reserve_window_node *rsv = &block_i->rsv_window_node; + + rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + + /* + * if filesystem is mounted with NORESERVATION, the goal + * reservation window size is set to zero to indicate + * block reservation is off + */ + if (!test_opt(sb, RESERVATION)) + rsv->rsv_goal_size = 0; + else + rsv->rsv_goal_size = EXT3_DEFAULT_RESERVE_BLOCKS; + rsv->rsv_alloc_hit = 0; + block_i->last_alloc_logical_block = 0; + block_i->last_alloc_physical_block = 0; + } + ei->i_block_alloc_info = block_i; +} + +/** + * ext3_discard_reservation() + * @inode: inode + * + * Discard(free) block reservation window on last file close, or truncate + * or at last iput(). + * + * It is being called in three cases: + * ext3_release_file(): last writer close the file + * ext3_clear_inode(): last iput(), when nobody link to this file. + * ext3_truncate(): when the block indirect map is about to change. + * + */ +void ext3_discard_reservation(struct inode *inode) +{ + struct ext3_inode_info *ei = EXT3_I(inode); + struct ext3_block_alloc_info *block_i = ei->i_block_alloc_info; + struct ext3_reserve_window_node *rsv; + spinlock_t *rsv_lock = &EXT3_SB(inode->i_sb)->s_rsv_window_lock; + + if (!block_i) + return; + + rsv = &block_i->rsv_window_node; + if (!rsv_is_empty(&rsv->rsv_window)) { + spin_lock(rsv_lock); + if (!rsv_is_empty(&rsv->rsv_window)) + rsv_window_remove(inode->i_sb, rsv); + spin_unlock(rsv_lock); + } +} + +/** + * ext3_free_blocks_sb() -- Free given blocks and update quota + * @handle: handle to this transaction + * @sb: super block + * @block: start physcial block to free + * @count: number of blocks to free + * @pdquot_freed_blocks: pointer to quota + */ +void ext3_free_blocks_sb(handle_t *handle, struct super_block *sb, + ext3_fsblk_t block, unsigned long count, + unsigned long *pdquot_freed_blocks) +{ + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *gd_bh; + unsigned long block_group; + ext3_grpblk_t bit; + unsigned long i; + unsigned long overflow; + struct ext3_group_desc * desc; + struct ext3_super_block * es; + struct ext3_sb_info *sbi; + int err = 0, ret; + ext3_grpblk_t group_freed; + + *pdquot_freed_blocks = 0; + sbi = EXT3_SB(sb); + es = sbi->s_es; + if (block < le32_to_cpu(es->s_first_data_block) || + block + count < block || + block + count > le32_to_cpu(es->s_blocks_count)) { + ext3_error (sb, "ext3_free_blocks", + "Freeing blocks not in datazone - " + "block = "E3FSBLK", count = %lu", block, count); + goto error_return; + } + + ext3_debug ("freeing block(s) %lu-%lu\n", block, block + count - 1); + +do_more: + overflow = 0; + block_group = (block - le32_to_cpu(es->s_first_data_block)) / + EXT3_BLOCKS_PER_GROUP(sb); + bit = (block - le32_to_cpu(es->s_first_data_block)) % + EXT3_BLOCKS_PER_GROUP(sb); + /* + * Check to see if we are freeing blocks across a group + * boundary. + */ + if (bit + count > EXT3_BLOCKS_PER_GROUP(sb)) { + overflow = bit + count - EXT3_BLOCKS_PER_GROUP(sb); + count -= overflow; + } + brelse(bitmap_bh); + bitmap_bh = read_block_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + desc = ext3_get_group_desc (sb, block_group, &gd_bh); + if (!desc) + goto error_return; + + if (in_range (le32_to_cpu(desc->bg_block_bitmap), block, count) || + in_range (le32_to_cpu(desc->bg_inode_bitmap), block, count) || + in_range (block, le32_to_cpu(desc->bg_inode_table), + sbi->s_itb_per_group) || + in_range (block + count - 1, le32_to_cpu(desc->bg_inode_table), + sbi->s_itb_per_group)) + ext3_error (sb, "ext3_free_blocks", + "Freeing blocks in system zones - " + "Block = "E3FSBLK", count = %lu", + block, count); + + /* + * We are about to start releasing blocks in the bitmap, + * so we need undo access. + */ + /* @@@ check errors */ + BUFFER_TRACE(bitmap_bh, "getting undo access"); + err = ext3_journal_get_undo_access(handle, bitmap_bh); + if (err) + goto error_return; + + /* + * We are about to modify some metadata. Call the journal APIs + * to unshare ->b_data if a currently-committing transaction is + * using it + */ + BUFFER_TRACE(gd_bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, gd_bh); + if (err) + goto error_return; + + jbd_lock_bh_state(bitmap_bh); + + for (i = 0, group_freed = 0; i < count; i++) { + /* + * An HJ special. This is expensive... + */ +#ifdef CONFIG_JBD_DEBUG + jbd_unlock_bh_state(bitmap_bh); + { + struct buffer_head *debug_bh; + debug_bh = sb_find_get_block(sb, block + i); + if (debug_bh) { + BUFFER_TRACE(debug_bh, "Deleted!"); + if (!bh2jh(bitmap_bh)->b_committed_data) + BUFFER_TRACE(debug_bh, + "No commited data in bitmap"); + BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap"); + __brelse(debug_bh); + } + } + jbd_lock_bh_state(bitmap_bh); +#endif + if (need_resched()) { + jbd_unlock_bh_state(bitmap_bh); + cond_resched(); + jbd_lock_bh_state(bitmap_bh); + } + /* @@@ This prevents newly-allocated data from being + * freed and then reallocated within the same + * transaction. + * + * Ideally we would want to allow that to happen, but to + * do so requires making journal_forget() capable of + * revoking the queued write of a data block, which + * implies blocking on the journal lock. *forget() + * cannot block due to truncate races. + * + * Eventually we can fix this by making journal_forget() + * return a status indicating whether or not it was able + * to revoke the buffer. On successful revoke, it is + * safe not to set the allocation bit in the committed + * bitmap, because we know that there is no outstanding + * activity on the buffer any more and so it is safe to + * reallocate it. + */ + BUFFER_TRACE(bitmap_bh, "set in b_committed_data"); + J_ASSERT_BH(bitmap_bh, + bh2jh(bitmap_bh)->b_committed_data != NULL); + ext3_set_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, + bh2jh(bitmap_bh)->b_committed_data); + + /* + * We clear the bit in the bitmap after setting the committed + * data bit, because this is the reverse order to that which + * the allocator uses. + */ + BUFFER_TRACE(bitmap_bh, "clear bit"); + if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group), + bit + i, bitmap_bh->b_data)) { + jbd_unlock_bh_state(bitmap_bh); + ext3_error(sb, __FUNCTION__, + "bit already cleared for block "E3FSBLK, + block + i); + jbd_lock_bh_state(bitmap_bh); + BUFFER_TRACE(bitmap_bh, "bit already cleared"); + } else { + group_freed++; + } + } + jbd_unlock_bh_state(bitmap_bh); + + spin_lock(sb_bgl_lock(sbi, block_group)); + desc->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) + + group_freed); + spin_unlock(sb_bgl_lock(sbi, block_group)); + percpu_counter_mod(&sbi->s_freeblocks_counter, count); + + /* We dirtied the bitmap block */ + BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); + err = ext3_journal_dirty_metadata(handle, bitmap_bh); + + /* And the group descriptor block */ + BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); + ret = ext3_journal_dirty_metadata(handle, gd_bh); + if (!err) err = ret; + *pdquot_freed_blocks += group_freed; + + if (overflow && !err) { + block += count; + count = overflow; + goto do_more; + } + sb->s_dirt = 1; +error_return: + brelse(bitmap_bh); + ext3_std_error(sb, err); + return; +} + +/** + * ext3_free_blocks() -- Free given blocks and update quota + * @handle: handle for this transaction + * @inode: inode + * @block: start physical block to free + * @count: number of blocks to count + */ +void ext3_free_blocks(handle_t *handle, struct inode *inode, + ext3_fsblk_t block, unsigned long count) +{ + struct super_block * sb; + unsigned long dquot_freed_blocks; + + sb = inode->i_sb; + if (!sb) { + printk ("ext3_free_blocks: nonexistent device"); + return; + } + ext3_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks); + if (dquot_freed_blocks) + DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); + return; +} + +/** + * ext3_test_allocatable() + * @nr: given allocation block group + * @bh: bufferhead contains the bitmap of the given block group + * + * For ext3 allocations, we must not reuse any blocks which are + * allocated in the bitmap buffer's "last committed data" copy. This + * prevents deletes from freeing up the page for reuse until we have + * committed the delete transaction. + * + * If we didn't do this, then deleting something and reallocating it as + * data would allow the old block to be overwritten before the + * transaction committed (because we force data to disk before commit). + * This would lead to corruption if we crashed between overwriting the + * data and committing the delete. + * + * @@@ We may want to make this allocation behaviour conditional on + * data-writes at some point, and disable it for metadata allocations or + * sync-data inodes. + */ +static int ext3_test_allocatable(ext3_grpblk_t nr, struct buffer_head *bh) +{ + int ret; + struct journal_head *jh = bh2jh(bh); + + if (ext3_test_bit(nr, bh->b_data)) + return 0; + + jbd_lock_bh_state(bh); + if (!jh->b_committed_data) + ret = 1; + else + ret = !ext3_test_bit(nr, jh->b_committed_data); + jbd_unlock_bh_state(bh); + return ret; +} + +/** + * bitmap_search_next_usable_block() + * @start: the starting block (group relative) of the search + * @bh: bufferhead contains the block group bitmap + * @maxblocks: the ending block (group relative) of the reservation + * + * The bitmap search --- search forward alternately through the actual + * bitmap on disk and the last-committed copy in journal, until we find a + * bit free in both bitmaps. + */ +static ext3_grpblk_t +bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, + ext3_grpblk_t maxblocks) +{ + ext3_grpblk_t next; + struct journal_head *jh = bh2jh(bh); + + while (start < maxblocks) { + next = ext3_find_next_zero_bit(bh->b_data, maxblocks, start); + if (next >= maxblocks) + return -1; + if (ext3_test_allocatable(next, bh)) + return next; + jbd_lock_bh_state(bh); + if (jh->b_committed_data) + start = ext3_find_next_zero_bit(jh->b_committed_data, + maxblocks, next); + jbd_unlock_bh_state(bh); + } + return -1; +} + +/** + * find_next_usable_block() + * @start: the starting block (group relative) to find next + * allocatable block in bitmap. + * @bh: bufferhead contains the block group bitmap + * @maxblocks: the ending block (group relative) for the search + * + * Find an allocatable block in a bitmap. We honor both the bitmap and + * its last-committed copy (if that exists), and perform the "most + * appropriate allocation" algorithm of looking for a free block near + * the initial goal; then for a free byte somewhere in the bitmap; then + * for any free bit in the bitmap. + */ +static ext3_grpblk_t +find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, + ext3_grpblk_t maxblocks) +{ + ext3_grpblk_t here, next; + char *p, *r; + + if (start > 0) { + /* + * The goal was occupied; search forward for a free + * block within the next XX blocks. + * + * end_goal is more or less random, but it has to be + * less than EXT3_BLOCKS_PER_GROUP. Aligning up to the + * next 64-bit boundary is simple.. + */ + ext3_grpblk_t end_goal = (start + 63) & ~63; + if (end_goal > maxblocks) + end_goal = maxblocks; + here = ext3_find_next_zero_bit(bh->b_data, end_goal, start); + if (here < end_goal && ext3_test_allocatable(here, bh)) + return here; + ext3_debug("Bit not found near goal\n"); + } + + here = start; + if (here < 0) + here = 0; + + p = ((char *)bh->b_data) + (here >> 3); + r = memscan(p, 0, (maxblocks - here + 7) >> 3); + next = (r - ((char *)bh->b_data)) << 3; + + if (next < maxblocks && next >= start && ext3_test_allocatable(next, bh)) + return next; + + /* + * The bitmap search --- search forward alternately through the actual + * bitmap and the last-committed copy until we find a bit free in + * both + */ + here = bitmap_search_next_usable_block(here, bh, maxblocks); + return here; +} + +/** + * claim_block() + * @block: the free block (group relative) to allocate + * @bh: the bufferhead containts the block group bitmap + * + * We think we can allocate this block in this bitmap. Try to set the bit. + * If that succeeds then check that nobody has allocated and then freed the + * block since we saw that is was not marked in b_committed_data. If it _was_ + * allocated and freed then clear the bit in the bitmap again and return + * zero (failure). + */ +static inline int +claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh) +{ + struct journal_head *jh = bh2jh(bh); + int ret; + + if (ext3_set_bit_atomic(lock, block, bh->b_data)) + return 0; + jbd_lock_bh_state(bh); + if (jh->b_committed_data && ext3_test_bit(block,jh->b_committed_data)) { + ext3_clear_bit_atomic(lock, block, bh->b_data); + ret = 0; + } else { + ret = 1; + } + jbd_unlock_bh_state(bh); + return ret; +} + +/** + * ext3_try_to_allocate() + * @sb: superblock + * @handle: handle to this transaction + * @group: given allocation block group + * @bitmap_bh: bufferhead holds the block bitmap + * @grp_goal: given target block within the group + * @count: target number of blocks to allocate + * @my_rsv: reservation window + * + * Attempt to allocate blocks within a give range. Set the range of allocation + * first, then find the first free bit(s) from the bitmap (within the range), + * and at last, allocate the blocks by claiming the found free bit as allocated. + * + * To set the range of this allocation: + * if there is a reservation window, only try to allocate block(s) from the + * file's own reservation window; + * Otherwise, the allocation range starts from the give goal block, ends at + * the block group's last block. + * + * If we failed to allocate the desired block then we may end up crossing to a + * new bitmap. In that case we must release write access to the old one via + * ext3_journal_release_buffer(), else we'll run out of credits. + */ +static ext3_grpblk_t +ext3_try_to_allocate(struct super_block *sb, handle_t *handle, int group, + struct buffer_head *bitmap_bh, ext3_grpblk_t grp_goal, + unsigned long *count, struct ext3_reserve_window *my_rsv) +{ + ext3_fsblk_t group_first_block; + ext3_grpblk_t start, end; + unsigned long num = 0; + + /* we do allocation within the reservation window if we have a window */ + if (my_rsv) { + group_first_block = ext3_group_first_block_no(sb, group); + if (my_rsv->_rsv_start >= group_first_block) + start = my_rsv->_rsv_start - group_first_block; + else + /* reservation window cross group boundary */ + start = 0; + end = my_rsv->_rsv_end - group_first_block + 1; + if (end > EXT3_BLOCKS_PER_GROUP(sb)) + /* reservation window crosses group boundary */ + end = EXT3_BLOCKS_PER_GROUP(sb); + if ((start <= grp_goal) && (grp_goal < end)) + start = grp_goal; + else + grp_goal = -1; + } else { + if (grp_goal > 0) + start = grp_goal; + else + start = 0; + end = EXT3_BLOCKS_PER_GROUP(sb); + } + + BUG_ON(start > EXT3_BLOCKS_PER_GROUP(sb)); + +repeat: + if (grp_goal < 0 || !ext3_test_allocatable(grp_goal, bitmap_bh)) { + grp_goal = find_next_usable_block(start, bitmap_bh, end); + if (grp_goal < 0) + goto fail_access; + if (!my_rsv) { + int i; + + for (i = 0; i < 7 && grp_goal > start && + ext3_test_allocatable(grp_goal - 1, + bitmap_bh); + i++, grp_goal--) + ; + } + } + start = grp_goal; + + if (!claim_block(sb_bgl_lock(EXT3_SB(sb), group), + grp_goal, bitmap_bh)) { + /* + * The block was allocated by another thread, or it was + * allocated and then freed by another thread + */ + start++; + grp_goal++; + if (start >= end) + goto fail_access; + goto repeat; + } + num++; + grp_goal++; + while (num < *count && grp_goal < end + && ext3_test_allocatable(grp_goal, bitmap_bh) + && claim_block(sb_bgl_lock(EXT3_SB(sb), group), + grp_goal, bitmap_bh)) { + num++; + grp_goal++; + } + *count = num; + return grp_goal - num; +fail_access: + *count = num; + return -1; +} + +/** + * find_next_reservable_window(): + * find a reservable space within the given range. + * It does not allocate the reservation window for now: + * alloc_new_reservation() will do the work later. + * + * @search_head: the head of the searching list; + * This is not necessarily the list head of the whole filesystem + * + * We have both head and start_block to assist the search + * for the reservable space. The list starts from head, + * but we will shift to the place where start_block is, + * then start from there, when looking for a reservable space. + * + * @size: the target new reservation window size + * + * @group_first_block: the first block we consider to start + * the real search from + * + * @last_block: + * the maximum block number that our goal reservable space + * could start from. This is normally the last block in this + * group. The search will end when we found the start of next + * possible reservable space is out of this boundary. + * This could handle the cross boundary reservation window + * request. + * + * basically we search from the given range, rather than the whole + * reservation double linked list, (start_block, last_block) + * to find a free region that is of my size and has not + * been reserved. + * + */ +static int find_next_reservable_window( + struct ext3_reserve_window_node *search_head, + struct ext3_reserve_window_node *my_rsv, + struct super_block * sb, + ext3_fsblk_t start_block, + ext3_fsblk_t last_block) +{ + struct rb_node *next; + struct ext3_reserve_window_node *rsv, *prev; + ext3_fsblk_t cur; + int size = my_rsv->rsv_goal_size; + + /* TODO: make the start of the reservation window byte-aligned */ + /* cur = *start_block & ~7;*/ + cur = start_block; + rsv = search_head; + if (!rsv) + return -1; + + while (1) { + if (cur <= rsv->rsv_end) + cur = rsv->rsv_end + 1; + + /* TODO? + * in the case we could not find a reservable space + * that is what is expected, during the re-search, we could + * remember what's the largest reservable space we could have + * and return that one. + * + * For now it will fail if we could not find the reservable + * space with expected-size (or more)... + */ + if (cur > last_block) + return -1; /* fail */ + + prev = rsv; + next = rb_next(&rsv->rsv_node); + rsv = list_entry(next,struct ext3_reserve_window_node,rsv_node); + + /* + * Reached the last reservation, we can just append to the + * previous one. + */ + if (!next) + break; + + if (cur + size <= rsv->rsv_start) { + /* + * Found a reserveable space big enough. We could + * have a reservation across the group boundary here + */ + break; + } + } + /* + * we come here either : + * when we reach the end of the whole list, + * and there is empty reservable space after last entry in the list. + * append it to the end of the list. + * + * or we found one reservable space in the middle of the list, + * return the reservation window that we could append to. + * succeed. + */ + + if ((prev != my_rsv) && (!rsv_is_empty(&my_rsv->rsv_window))) + rsv_window_remove(sb, my_rsv); + + /* + * Let's book the whole avaliable window for now. We will check the + * disk bitmap later and then, if there are free blocks then we adjust + * the window size if it's larger than requested. + * Otherwise, we will remove this node from the tree next time + * call find_next_reservable_window. + */ + my_rsv->rsv_start = cur; + my_rsv->rsv_end = cur + size - 1; + my_rsv->rsv_alloc_hit = 0; + + if (prev != my_rsv) + ext3_rsv_window_add(sb, my_rsv); + + return 0; +} + +/** + * alloc_new_reservation()--allocate a new reservation window + * + * To make a new reservation, we search part of the filesystem + * reservation list (the list that inside the group). We try to + * allocate a new reservation window near the allocation goal, + * or the beginning of the group, if there is no goal. + * + * We first find a reservable space after the goal, then from + * there, we check the bitmap for the first free block after + * it. If there is no free block until the end of group, then the + * whole group is full, we failed. Otherwise, check if the free + * block is inside the expected reservable space, if so, we + * succeed. + * If the first free block is outside the reservable space, then + * start from the first free block, we search for next available + * space, and go on. + * + * on succeed, a new reservation will be found and inserted into the list + * It contains at least one free block, and it does not overlap with other + * reservation windows. + * + * failed: we failed to find a reservation window in this group + * + * @rsv: the reservation + * + * @grp_goal: The goal (group-relative). It is where the search for a + * free reservable space should start from. + * if we have a grp_goal(grp_goal >0 ), then start from there, + * no grp_goal(grp_goal = -1), we start from the first block + * of the group. + * + * @sb: the super block + * @group: the group we are trying to allocate in + * @bitmap_bh: the block group block bitmap + * + */ +static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv, + ext3_grpblk_t grp_goal, struct super_block *sb, + unsigned int group, struct buffer_head *bitmap_bh) +{ + struct ext3_reserve_window_node *search_head; + ext3_fsblk_t group_first_block, group_end_block, start_block; + ext3_grpblk_t first_free_block; + struct rb_root *fs_rsv_root = &EXT3_SB(sb)->s_rsv_window_root; + unsigned long size; + int ret; + spinlock_t *rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock; + + group_first_block = ext3_group_first_block_no(sb, group); + group_end_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + + if (grp_goal < 0) + start_block = group_first_block; + else + start_block = grp_goal + group_first_block; + + size = my_rsv->rsv_goal_size; + + if (!rsv_is_empty(&my_rsv->rsv_window)) { + /* + * if the old reservation is cross group boundary + * and if the goal is inside the old reservation window, + * we will come here when we just failed to allocate from + * the first part of the window. We still have another part + * that belongs to the next group. In this case, there is no + * point to discard our window and try to allocate a new one + * in this group(which will fail). we should + * keep the reservation window, just simply move on. + * + * Maybe we could shift the start block of the reservation + * window to the first block of next group. + */ + + if ((my_rsv->rsv_start <= group_end_block) && + (my_rsv->rsv_end > group_end_block) && + (start_block >= my_rsv->rsv_start)) + return -1; + + if ((my_rsv->rsv_alloc_hit > + (my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) { + /* + * if the previously allocation hit ratio is + * greater than 1/2, then we double the size of + * the reservation window the next time, + * otherwise we keep the same size window + */ + size = size * 2; + if (size > EXT3_MAX_RESERVE_BLOCKS) + size = EXT3_MAX_RESERVE_BLOCKS; + my_rsv->rsv_goal_size= size; + } + } + + spin_lock(rsv_lock); + /* + * shift the search start to the window near the goal block + */ + search_head = search_reserve_window(fs_rsv_root, start_block); + + /* + * find_next_reservable_window() simply finds a reservable window + * inside the given range(start_block, group_end_block). + * + * To make sure the reservation window has a free bit inside it, we + * need to check the bitmap after we found a reservable window. + */ +retry: + ret = find_next_reservable_window(search_head, my_rsv, sb, + start_block, group_end_block); + + if (ret == -1) { + if (!rsv_is_empty(&my_rsv->rsv_window)) + rsv_window_remove(sb, my_rsv); + spin_unlock(rsv_lock); + return -1; + } + + /* + * On success, find_next_reservable_window() returns the + * reservation window where there is a reservable space after it. + * Before we reserve this reservable space, we need + * to make sure there is at least a free block inside this region. + * + * searching the first free bit on the block bitmap and copy of + * last committed bitmap alternatively, until we found a allocatable + * block. Search start from the start block of the reservable space + * we just found. + */ + spin_unlock(rsv_lock); + first_free_block = bitmap_search_next_usable_block( + my_rsv->rsv_start - group_first_block, + bitmap_bh, group_end_block - group_first_block + 1); + + if (first_free_block < 0) { + /* + * no free block left on the bitmap, no point + * to reserve the space. return failed. + */ + spin_lock(rsv_lock); + if (!rsv_is_empty(&my_rsv->rsv_window)) + rsv_window_remove(sb, my_rsv); + spin_unlock(rsv_lock); + return -1; /* failed */ + } + + start_block = first_free_block + group_first_block; + /* + * check if the first free block is within the + * free space we just reserved + */ + if (start_block >= my_rsv->rsv_start && start_block < my_rsv->rsv_end) + return 0; /* success */ + /* + * if the first free bit we found is out of the reservable space + * continue search for next reservable space, + * start from where the free block is, + * we also shift the list head to where we stopped last time + */ + search_head = my_rsv; + spin_lock(rsv_lock); + goto retry; +} + +/** + * try_to_extend_reservation() + * @my_rsv: given reservation window + * @sb: super block + * @size: the delta to extend + * + * Attempt to expand the reservation window large enough to have + * required number of free blocks + * + * Since ext3_try_to_allocate() will always allocate blocks within + * the reservation window range, if the window size is too small, + * multiple blocks allocation has to stop at the end of the reservation + * window. To make this more efficient, given the total number of + * blocks needed and the current size of the window, we try to + * expand the reservation window size if necessary on a best-effort + * basis before ext3_new_blocks() tries to allocate blocks, + */ +static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv, + struct super_block *sb, int size) +{ + struct ext3_reserve_window_node *next_rsv; + struct rb_node *next; + spinlock_t *rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock; + + if (!spin_trylock(rsv_lock)) + return; + + next = rb_next(&my_rsv->rsv_node); + + if (!next) + my_rsv->rsv_end += size; + else { + next_rsv = list_entry(next, struct ext3_reserve_window_node, rsv_node); + + if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) + my_rsv->rsv_end += size; + else + my_rsv->rsv_end = next_rsv->rsv_start - 1; + } + spin_unlock(rsv_lock); +} + +/** + * ext3_try_to_allocate_with_rsv() + * @sb: superblock + * @handle: handle to this transaction + * @group: given allocation block group + * @bitmap_bh: bufferhead holds the block bitmap + * @grp_goal: given target block within the group + * @count: target number of blocks to allocate + * @my_rsv: reservation window + * @errp: pointer to store the error code + * + * This is the main function used to allocate a new block and its reservation + * window. + * + * Each time when a new block allocation is need, first try to allocate from + * its own reservation. If it does not have a reservation window, instead of + * looking for a free bit on bitmap first, then look up the reservation list to + * see if it is inside somebody else's reservation window, we try to allocate a + * reservation window for it starting from the goal first. Then do the block + * allocation within the reservation window. + * + * This will avoid keeping on searching the reservation list again and + * again when somebody is looking for a free block (without + * reservation), and there are lots of free blocks, but they are all + * being reserved. + * + * We use a red-black tree for the per-filesystem reservation list. + * + */ +static ext3_grpblk_t +ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, + unsigned int group, struct buffer_head *bitmap_bh, + ext3_grpblk_t grp_goal, + struct ext3_reserve_window_node * my_rsv, + unsigned long *count, int *errp) +{ + ext3_fsblk_t group_first_block, group_last_block; + ext3_grpblk_t ret = 0; + int fatal; + unsigned long num = *count; + + *errp = 0; + + /* + * Make sure we use undo access for the bitmap, because it is critical + * that we do the frozen_data COW on bitmap buffers in all cases even + * if the buffer is in BJ_Forget state in the committing transaction. + */ + BUFFER_TRACE(bitmap_bh, "get undo access for new block"); + fatal = ext3_journal_get_undo_access(handle, bitmap_bh); + if (fatal) { + *errp = fatal; + return -1; + } + + /* + * we don't deal with reservation when + * filesystem is mounted without reservation + * or the file is not a regular file + * or last attempt to allocate a block with reservation turned on failed + */ + if (my_rsv == NULL ) { + ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, + grp_goal, count, NULL); + goto out; + } + /* + * grp_goal is a group relative block number (if there is a goal) + * 0 < grp_goal < EXT3_BLOCKS_PER_GROUP(sb) + * first block is a filesystem wide block number + * first block is the block number of the first block in this group + */ + group_first_block = ext3_group_first_block_no(sb, group); + group_last_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + + /* + * Basically we will allocate a new block from inode's reservation + * window. + * + * We need to allocate a new reservation window, if: + * a) inode does not have a reservation window; or + * b) last attempt to allocate a block from existing reservation + * failed; or + * c) we come here with a goal and with a reservation window + * + * We do not need to allocate a new reservation window if we come here + * at the beginning with a goal and the goal is inside the window, or + * we don't have a goal but already have a reservation window. + * then we could go to allocate from the reservation window directly. + */ + while (1) { + if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) || + !goal_in_my_reservation(&my_rsv->rsv_window, + grp_goal, group, sb)) { + if (my_rsv->rsv_goal_size < *count) + my_rsv->rsv_goal_size = *count; + ret = alloc_new_reservation(my_rsv, grp_goal, sb, + group, bitmap_bh); + if (ret < 0) + break; /* failed */ + + if (!goal_in_my_reservation(&my_rsv->rsv_window, + grp_goal, group, sb)) + grp_goal = -1; + } else if (grp_goal > 0 && + (my_rsv->rsv_end-grp_goal+1) < *count) + try_to_extend_reservation(my_rsv, sb, + *count-my_rsv->rsv_end + grp_goal - 1); + + if ((my_rsv->rsv_start > group_last_block) || + (my_rsv->rsv_end < group_first_block)) { + rsv_window_dump(&EXT3_SB(sb)->s_rsv_window_root, 1); + BUG(); + } + ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, + grp_goal, &num, &my_rsv->rsv_window); + if (ret >= 0) { + my_rsv->rsv_alloc_hit += num; + *count = num; + break; /* succeed */ + } + num = *count; + } +out: + if (ret >= 0) { + BUFFER_TRACE(bitmap_bh, "journal_dirty_metadata for " + "bitmap block"); + fatal = ext3_journal_dirty_metadata(handle, bitmap_bh); + if (fatal) { + *errp = fatal; + return -1; + } + return ret; + } + + BUFFER_TRACE(bitmap_bh, "journal_release_buffer"); + ext3_journal_release_buffer(handle, bitmap_bh); + return ret; +} + +/** + * ext3_has_free_blocks() + * @sbi: in-core super block structure. + * + * Check if filesystem has at least 1 free block available for allocation. + */ +static int ext3_has_free_blocks(struct ext3_sb_info *sbi) +{ + ext3_fsblk_t free_blocks, root_blocks; + + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + root_blocks = le32_to_cpu(sbi->s_es->s_r_blocks_count); + if (free_blocks < root_blocks + 1 && !capable(CAP_SYS_RESOURCE) && + sbi->s_resuid != current->fsuid && + (sbi->s_resgid == 0 || !in_group_p (sbi->s_resgid))) { + return 0; + } + return 1; +} + +/** + * ext3_should_retry_alloc() + * @sb: super block + * @retries number of attemps has been made + * + * ext3_should_retry_alloc() is called when ENOSPC is returned, and if + * it is profitable to retry the operation, this function will wait + * for the current or commiting transaction to complete, and then + * return TRUE. + * + * if the total number of retries exceed three times, return FALSE. + */ +int ext3_should_retry_alloc(struct super_block *sb, int *retries) +{ + if (!ext3_has_free_blocks(EXT3_SB(sb)) || (*retries)++ > 3) + return 0; + + jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); + + return journal_force_commit_nested(EXT3_SB(sb)->s_journal); +} + +/** + * ext3_new_blocks() -- core block(s) allocation function + * @handle: handle to this transaction + * @inode: file inode + * @goal: given target block(filesystem wide) + * @count: target number of blocks to allocate + * @errp: error code + * + * ext3_new_blocks uses a goal block to assist allocation. It tries to + * allocate block(s) from the block group contains the goal block first. If that + * fails, it will try to allocate block(s) from other block groups without + * any specific goal block. + * + */ +ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode, + ext3_fsblk_t goal, unsigned long *count, int *errp) +{ + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *gdp_bh; + int group_no; + int goal_group; + ext3_grpblk_t grp_target_blk; /* blockgroup relative goal block */ + ext3_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ + ext3_fsblk_t ret_block; /* filesyetem-wide allocated block */ + int bgi; /* blockgroup iteration index */ + int fatal = 0, err; + int performed_allocation = 0; + ext3_grpblk_t free_blocks; /* number of free blocks in a group */ + struct super_block *sb; + struct ext3_group_desc *gdp; + struct ext3_super_block *es; + struct ext3_sb_info *sbi; + struct ext3_reserve_window_node *my_rsv = NULL; + struct ext3_block_alloc_info *block_i; + unsigned short windowsz = 0; +#ifdef EXT3FS_DEBUG + static int goal_hits, goal_attempts; +#endif + unsigned long ngroups; + unsigned long num = *count; + + *errp = -ENOSPC; + sb = inode->i_sb; + if (!sb) { + printk("ext3_new_block: nonexistent device"); + return 0; + } + + /* + * Check quota for allocation of this block. + */ + if (DQUOT_ALLOC_BLOCK(inode, num)) { + *errp = -EDQUOT; + return 0; + } + + sbi = EXT3_SB(sb); + es = EXT3_SB(sb)->s_es; + ext3_debug("goal=%lu.\n", goal); + /* + * Allocate a block from reservation only when + * filesystem is mounted with reservation(default,-o reservation), and + * it's a regular file, and + * the desired window size is greater than 0 (One could use ioctl + * command EXT3_IOC_SETRSVSZ to set the window size to 0 to turn off + * reservation on that particular file) + */ + block_i = EXT3_I(inode)->i_block_alloc_info; + if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0)) + my_rsv = &block_i->rsv_window_node; + + if (!ext3_has_free_blocks(sbi)) { + *errp = -ENOSPC; + goto out; + } + + /* + * First, test whether the goal block is free. + */ + if (goal < le32_to_cpu(es->s_first_data_block) || + goal >= le32_to_cpu(es->s_blocks_count)) + goal = le32_to_cpu(es->s_first_data_block); + group_no = (goal - le32_to_cpu(es->s_first_data_block)) / + EXT3_BLOCKS_PER_GROUP(sb); + goal_group = group_no; +retry_alloc: + gdp = ext3_get_group_desc(sb, group_no, &gdp_bh); + if (!gdp) + goto io_error; + + free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); + /* + * if there is not enough free blocks to make a new resevation + * turn off reservation for this allocation + */ + if (my_rsv && (free_blocks < windowsz) + && (rsv_is_empty(&my_rsv->rsv_window))) + my_rsv = NULL; + + if (free_blocks > 0) { + grp_target_blk = ((goal - le32_to_cpu(es->s_first_data_block)) % + EXT3_BLOCKS_PER_GROUP(sb)); + bitmap_bh = read_block_bitmap(sb, group_no); + if (!bitmap_bh) + goto io_error; + grp_alloc_blk = ext3_try_to_allocate_with_rsv(sb, handle, + group_no, bitmap_bh, grp_target_blk, + my_rsv, &num, &fatal); + if (fatal) + goto out; + if (grp_alloc_blk >= 0) + goto allocated; + } + + ngroups = EXT3_SB(sb)->s_groups_count; + smp_rmb(); + + /* + * Now search the rest of the groups. We assume that + * i and gdp correctly point to the last group visited. + */ + for (bgi = 0; bgi < ngroups; bgi++) { + group_no++; + if (group_no >= ngroups) + group_no = 0; + gdp = ext3_get_group_desc(sb, group_no, &gdp_bh); + if (!gdp) { + *errp = -EIO; + goto out; + } + free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); + /* + * skip this group if the number of + * free blocks is less than half of the reservation + * window size. + */ + if (free_blocks <= (windowsz/2)) + continue; + + brelse(bitmap_bh); + bitmap_bh = read_block_bitmap(sb, group_no); + if (!bitmap_bh) + goto io_error; + /* + * try to allocate block(s) from this group, without a goal(-1). + */ + grp_alloc_blk = ext3_try_to_allocate_with_rsv(sb, handle, + group_no, bitmap_bh, -1, my_rsv, + &num, &fatal); + if (fatal) + goto out; + if (grp_alloc_blk >= 0) + goto allocated; + } + /* + * We may end up a bogus ealier ENOSPC error due to + * filesystem is "full" of reservations, but + * there maybe indeed free blocks avaliable on disk + * In this case, we just forget about the reservations + * just do block allocation as without reservations. + */ + if (my_rsv) { + my_rsv = NULL; + group_no = goal_group; + goto retry_alloc; + } + /* No space left on the device */ + *errp = -ENOSPC; + goto out; + +allocated: + + ext3_debug("using block group %d(%d)\n", + group_no, gdp->bg_free_blocks_count); + + BUFFER_TRACE(gdp_bh, "get_write_access"); + fatal = ext3_journal_get_write_access(handle, gdp_bh); + if (fatal) + goto out; + + ret_block = grp_alloc_blk + ext3_group_first_block_no(sb, group_no); + + if (in_range(le32_to_cpu(gdp->bg_block_bitmap), ret_block, num) || + in_range(le32_to_cpu(gdp->bg_inode_bitmap), ret_block, num) || + in_range(ret_block, le32_to_cpu(gdp->bg_inode_table), + EXT3_SB(sb)->s_itb_per_group) || + in_range(ret_block + num - 1, le32_to_cpu(gdp->bg_inode_table), + EXT3_SB(sb)->s_itb_per_group)) + ext3_error(sb, "ext3_new_block", + "Allocating block in system zone - " + "blocks from "E3FSBLK", length %lu", + ret_block, num); + + performed_allocation = 1; + +#ifdef CONFIG_JBD_DEBUG + { + struct buffer_head *debug_bh; + + /* Record bitmap buffer state in the newly allocated block */ + debug_bh = sb_find_get_block(sb, ret_block); + if (debug_bh) { + BUFFER_TRACE(debug_bh, "state when allocated"); + BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap state"); + brelse(debug_bh); + } + } + jbd_lock_bh_state(bitmap_bh); + spin_lock(sb_bgl_lock(sbi, group_no)); + if (buffer_jbd(bitmap_bh) && bh2jh(bitmap_bh)->b_committed_data) { + int i; + + for (i = 0; i < num; i++) { + if (ext3_test_bit(grp_alloc_blk+i, + bh2jh(bitmap_bh)->b_committed_data)) { + printk("%s: block was unexpectedly set in " + "b_committed_data\n", __FUNCTION__); + } + } + } + ext3_debug("found bit %d\n", grp_alloc_blk); + spin_unlock(sb_bgl_lock(sbi, group_no)); + jbd_unlock_bh_state(bitmap_bh); +#endif + + if (ret_block + num - 1 >= le32_to_cpu(es->s_blocks_count)) { + ext3_error(sb, "ext3_new_block", + "block("E3FSBLK") >= blocks count(%d) - " + "block_group = %d, es == %p ", ret_block, + le32_to_cpu(es->s_blocks_count), group_no, es); + goto out; + } + + /* + * It is up to the caller to add the new buffer to a journal + * list of some description. We don't know in advance whether + * the caller wants to use it as metadata or data. + */ + ext3_debug("allocating block %lu. Goal hits %d of %d.\n", + ret_block, goal_hits, goal_attempts); + + spin_lock(sb_bgl_lock(sbi, group_no)); + gdp->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num); + spin_unlock(sb_bgl_lock(sbi, group_no)); + percpu_counter_mod(&sbi->s_freeblocks_counter, -num); + + BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor"); + err = ext3_journal_dirty_metadata(handle, gdp_bh); + if (!fatal) + fatal = err; + + sb->s_dirt = 1; + if (fatal) + goto out; + + *errp = 0; + brelse(bitmap_bh); + DQUOT_FREE_BLOCK(inode, *count-num); + *count = num; + return ret_block; + +io_error: + *errp = -EIO; +out: + if (fatal) { + *errp = fatal; + ext3_std_error(sb, fatal); + } + /* + * Undo the block allocation + */ + if (!performed_allocation) + DQUOT_FREE_BLOCK(inode, *count); + brelse(bitmap_bh); + return 0; +} + +ext3_fsblk_t ext3_new_block(handle_t *handle, struct inode *inode, + ext3_fsblk_t goal, int *errp) +{ + unsigned long count = 1; + + return ext3_new_blocks(handle, inode, goal, &count, errp); +} + +/** + * ext3_count_free_blocks() -- count filesystem free blocks + * @sb: superblock + * + * Adds up the number of free blocks from each block group. + */ +ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb) +{ + ext3_fsblk_t desc_count; + struct ext3_group_desc *gdp; + int i; + unsigned long ngroups = EXT3_SB(sb)->s_groups_count; +#ifdef EXT3FS_DEBUG + struct ext3_super_block *es; + ext3_fsblk_t bitmap_count; + unsigned long x; + struct buffer_head *bitmap_bh = NULL; + + es = EXT3_SB(sb)->s_es; + desc_count = 0; + bitmap_count = 0; + gdp = NULL; + + smp_rmb(); + for (i = 0; i < ngroups; i++) { + gdp = ext3_get_group_desc(sb, i, NULL); + if (!gdp) + continue; + desc_count += le16_to_cpu(gdp->bg_free_blocks_count); + brelse(bitmap_bh); + bitmap_bh = read_block_bitmap(sb, i); + if (bitmap_bh == NULL) + continue; + + x = ext3_count_free(bitmap_bh, sb->s_blocksize); + printk("group %d: stored = %d, counted = %lu\n", + i, le16_to_cpu(gdp->bg_free_blocks_count), x); + bitmap_count += x; + } + brelse(bitmap_bh); + printk("ext3_count_free_blocks: stored = "E3FSBLK + ", computed = "E3FSBLK", "E3FSBLK"\n", + le32_to_cpu(es->s_free_blocks_count), + desc_count, bitmap_count); + return bitmap_count; +#else + desc_count = 0; + smp_rmb(); + for (i = 0; i < ngroups; i++) { + gdp = ext3_get_group_desc(sb, i, NULL); + if (!gdp) + continue; + desc_count += le16_to_cpu(gdp->bg_free_blocks_count); + } + + return desc_count; +#endif +} + +static inline int +block_in_use(ext3_fsblk_t block, struct super_block *sb, unsigned char *map) +{ + return ext3_test_bit ((block - + le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block)) % + EXT3_BLOCKS_PER_GROUP(sb), map); +} + +static inline int test_root(int a, int b) +{ + int num = b; + + while (a > num) + num *= b; + return num == a; +} + +static int ext3_group_sparse(int group) +{ + if (group <= 1) + return 1; + if (!(group & 1)) + return 0; + return (test_root(group, 7) || test_root(group, 5) || + test_root(group, 3)); +} + +/** + * ext3_bg_has_super - number of blocks used by the superblock in group + * @sb: superblock for filesystem + * @group: group number to check + * + * Return the number of blocks used by the superblock (primary or backup) + * in this group. Currently this will be only 0 or 1. + */ +int ext3_bg_has_super(struct super_block *sb, int group) +{ + if (EXT3_HAS_RO_COMPAT_FEATURE(sb, + EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER) && + !ext3_group_sparse(group)) + return 0; + return 1; +} + +static unsigned long ext3_bg_num_gdb_meta(struct super_block *sb, int group) +{ + unsigned long metagroup = group / EXT3_DESC_PER_BLOCK(sb); + unsigned long first = metagroup * EXT3_DESC_PER_BLOCK(sb); + unsigned long last = first + EXT3_DESC_PER_BLOCK(sb) - 1; + + if (group == first || group == first + 1 || group == last) + return 1; + return 0; +} + +static unsigned long ext3_bg_num_gdb_nometa(struct super_block *sb, int group) +{ + if (EXT3_HAS_RO_COMPAT_FEATURE(sb, + EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER) && + !ext3_group_sparse(group)) + return 0; + return EXT3_SB(sb)->s_gdb_count; +} + +/** + * ext3_bg_num_gdb - number of blocks used by the group table in group + * @sb: superblock for filesystem + * @group: group number to check + * + * Return the number of blocks used by the group descriptor table + * (primary or backup) in this group. In the future there may be a + * different number of descriptor blocks in each group. + */ +unsigned long ext3_bg_num_gdb(struct super_block *sb, int group) +{ + unsigned long first_meta_bg = + le32_to_cpu(EXT3_SB(sb)->s_es->s_first_meta_bg); + unsigned long metagroup = group / EXT3_DESC_PER_BLOCK(sb); + + if (!EXT3_HAS_INCOMPAT_FEATURE(sb,EXT3_FEATURE_INCOMPAT_META_BG) || + metagroup < first_meta_bg) + return ext3_bg_num_gdb_nometa(sb,group); + + return ext3_bg_num_gdb_meta(sb,group); + +} diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c new file mode 100644 index 0000000..b9176ee --- /dev/null +++ b/fs/ext4/bitmap.c @@ -0,0 +1,32 @@ +/* + * linux/fs/ext3/bitmap.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + */ + +#include +#include +#include + +#ifdef EXT3FS_DEBUG + +static int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; + +unsigned long ext3_count_free (struct buffer_head * map, unsigned int numchars) +{ + unsigned int i; + unsigned long sum = 0; + + if (!map) + return (0); + for (i = 0; i < numchars; i++) + sum += nibblemap[map->b_data[i] & 0xf] + + nibblemap[(map->b_data[i] >> 4) & 0xf]; + return (sum); +} + +#endif /* EXT3FS_DEBUG */ + diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c new file mode 100644 index 0000000..d0b54f3 --- /dev/null +++ b/fs/ext4/dir.c @@ -0,0 +1,518 @@ +/* + * linux/fs/ext3/dir.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/dir.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * ext3 directory handling functions + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * + * Hash Tree Directory indexing (c) 2001 Daniel Phillips + * + */ + +#include +#include +#include +#include +#include +#include +#include + +static unsigned char ext3_filetype_table[] = { + DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK +}; + +static int ext3_readdir(struct file *, void *, filldir_t); +static int ext3_dx_readdir(struct file * filp, + void * dirent, filldir_t filldir); +static int ext3_release_dir (struct inode * inode, + struct file * filp); + +const struct file_operations ext3_dir_operations = { + .llseek = generic_file_llseek, + .read = generic_read_dir, + .readdir = ext3_readdir, /* we take BKL. needed?*/ + .ioctl = ext3_ioctl, /* BKL held */ +#ifdef CONFIG_COMPAT + .compat_ioctl = ext3_compat_ioctl, +#endif + .fsync = ext3_sync_file, /* BKL held */ +#ifdef CONFIG_EXT3_INDEX + .release = ext3_release_dir, +#endif +}; + + +static unsigned char get_dtype(struct super_block *sb, int filetype) +{ + if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_FILETYPE) || + (filetype >= EXT3_FT_MAX)) + return DT_UNKNOWN; + + return (ext3_filetype_table[filetype]); +} + + +int ext3_check_dir_entry (const char * function, struct inode * dir, + struct ext3_dir_entry_2 * de, + struct buffer_head * bh, + unsigned long offset) +{ + const char * error_msg = NULL; + const int rlen = le16_to_cpu(de->rec_len); + + if (rlen < EXT3_DIR_REC_LEN(1)) + error_msg = "rec_len is smaller than minimal"; + else if (rlen % 4 != 0) + error_msg = "rec_len % 4 != 0"; + else if (rlen < EXT3_DIR_REC_LEN(de->name_len)) + error_msg = "rec_len is too small for name_len"; + else if (((char *) de - bh->b_data) + rlen > dir->i_sb->s_blocksize) + error_msg = "directory entry across blocks"; + else if (le32_to_cpu(de->inode) > + le32_to_cpu(EXT3_SB(dir->i_sb)->s_es->s_inodes_count)) + error_msg = "inode out of bounds"; + + if (error_msg != NULL) + ext3_error (dir->i_sb, function, + "bad entry in directory #%lu: %s - " + "offset=%lu, inode=%lu, rec_len=%d, name_len=%d", + dir->i_ino, error_msg, offset, + (unsigned long) le32_to_cpu(de->inode), + rlen, de->name_len); + return error_msg == NULL ? 1 : 0; +} + +static int ext3_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + int error = 0; + unsigned long offset; + int i, stored; + struct ext3_dir_entry_2 *de; + struct super_block *sb; + int err; + struct inode *inode = filp->f_dentry->d_inode; + int ret = 0; + + sb = inode->i_sb; + +#ifdef CONFIG_EXT3_INDEX + if (EXT3_HAS_COMPAT_FEATURE(inode->i_sb, + EXT3_FEATURE_COMPAT_DIR_INDEX) && + ((EXT3_I(inode)->i_flags & EXT3_INDEX_FL) || + ((inode->i_size >> sb->s_blocksize_bits) == 1))) { + err = ext3_dx_readdir(filp, dirent, filldir); + if (err != ERR_BAD_DX_DIR) { + ret = err; + goto out; + } + /* + * We don't set the inode dirty flag since it's not + * critical that it get flushed back to the disk. + */ + EXT3_I(filp->f_dentry->d_inode)->i_flags &= ~EXT3_INDEX_FL; + } +#endif + stored = 0; + offset = filp->f_pos & (sb->s_blocksize - 1); + + while (!error && !stored && filp->f_pos < inode->i_size) { + unsigned long blk = filp->f_pos >> EXT3_BLOCK_SIZE_BITS(sb); + struct buffer_head map_bh; + struct buffer_head *bh = NULL; + + map_bh.b_state = 0; + err = ext3_get_blocks_handle(NULL, inode, blk, 1, + &map_bh, 0, 0); + if (err > 0) { + page_cache_readahead(sb->s_bdev->bd_inode->i_mapping, + &filp->f_ra, + filp, + map_bh.b_blocknr >> + (PAGE_CACHE_SHIFT - inode->i_blkbits), + 1); + bh = ext3_bread(NULL, inode, blk, 0, &err); + } + + /* + * We ignore I/O errors on directories so users have a chance + * of recovering data when there's a bad sector + */ + if (!bh) { + ext3_error (sb, "ext3_readdir", + "directory #%lu contains a hole at offset %lu", + inode->i_ino, (unsigned long)filp->f_pos); + filp->f_pos += sb->s_blocksize - offset; + continue; + } + +revalidate: + /* If the dir block has changed since the last call to + * readdir(2), then we might be pointing to an invalid + * dirent right now. Scan from the start of the block + * to make sure. */ + if (filp->f_version != inode->i_version) { + for (i = 0; i < sb->s_blocksize && i < offset; ) { + de = (struct ext3_dir_entry_2 *) + (bh->b_data + i); + /* It's too expensive to do a full + * dirent test each time round this + * loop, but we do have to test at + * least that it is non-zero. A + * failure will be detected in the + * dirent test below. */ + if (le16_to_cpu(de->rec_len) < + EXT3_DIR_REC_LEN(1)) + break; + i += le16_to_cpu(de->rec_len); + } + offset = i; + filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1)) + | offset; + filp->f_version = inode->i_version; + } + + while (!error && filp->f_pos < inode->i_size + && offset < sb->s_blocksize) { + de = (struct ext3_dir_entry_2 *) (bh->b_data + offset); + if (!ext3_check_dir_entry ("ext3_readdir", inode, de, + bh, offset)) { + /* On error, skip the f_pos to the + next block. */ + filp->f_pos = (filp->f_pos | + (sb->s_blocksize - 1)) + 1; + brelse (bh); + ret = stored; + goto out; + } + offset += le16_to_cpu(de->rec_len); + if (le32_to_cpu(de->inode)) { + /* We might block in the next section + * if the data destination is + * currently swapped out. So, use a + * version stamp to detect whether or + * not the directory has been modified + * during the copy operation. + */ + unsigned long version = filp->f_version; + + error = filldir(dirent, de->name, + de->name_len, + filp->f_pos, + le32_to_cpu(de->inode), + get_dtype(sb, de->file_type)); + if (error) + break; + if (version != filp->f_version) + goto revalidate; + stored ++; + } + filp->f_pos += le16_to_cpu(de->rec_len); + } + offset = 0; + brelse (bh); + } +out: + return ret; +} + +#ifdef CONFIG_EXT3_INDEX +/* + * These functions convert from the major/minor hash to an f_pos + * value. + * + * Currently we only use major hash numer. This is unfortunate, but + * on 32-bit machines, the same VFS interface is used for lseek and + * llseek, so if we use the 64 bit offset, then the 32-bit versions of + * lseek/telldir/seekdir will blow out spectacularly, and from within + * the ext2 low-level routine, we don't know if we're being called by + * a 64-bit version of the system call or the 32-bit version of the + * system call. Worse yet, NFSv2 only allows for a 32-bit readdir + * cookie. Sigh. + */ +#define hash2pos(major, minor) (major >> 1) +#define pos2maj_hash(pos) ((pos << 1) & 0xffffffff) +#define pos2min_hash(pos) (0) + +/* + * This structure holds the nodes of the red-black tree used to store + * the directory entry in hash order. + */ +struct fname { + __u32 hash; + __u32 minor_hash; + struct rb_node rb_hash; + struct fname *next; + __u32 inode; + __u8 name_len; + __u8 file_type; + char name[0]; +}; + +/* + * This functoin implements a non-recursive way of freeing all of the + * nodes in the red-black tree. + */ +static void free_rb_tree_fname(struct rb_root *root) +{ + struct rb_node *n = root->rb_node; + struct rb_node *parent; + struct fname *fname; + + while (n) { + /* Do the node's children first */ + if ((n)->rb_left) { + n = n->rb_left; + continue; + } + if (n->rb_right) { + n = n->rb_right; + continue; + } + /* + * The node has no children; free it, and then zero + * out parent's link to it. Finally go to the + * beginning of the loop and try to free the parent + * node. + */ + parent = rb_parent(n); + fname = rb_entry(n, struct fname, rb_hash); + while (fname) { + struct fname * old = fname; + fname = fname->next; + kfree (old); + } + if (!parent) + root->rb_node = NULL; + else if (parent->rb_left == n) + parent->rb_left = NULL; + else if (parent->rb_right == n) + parent->rb_right = NULL; + n = parent; + } + root->rb_node = NULL; +} + + +static struct dir_private_info *create_dir_info(loff_t pos) +{ + struct dir_private_info *p; + + p = kmalloc(sizeof(struct dir_private_info), GFP_KERNEL); + if (!p) + return NULL; + p->root.rb_node = NULL; + p->curr_node = NULL; + p->extra_fname = NULL; + p->last_pos = 0; + p->curr_hash = pos2maj_hash(pos); + p->curr_minor_hash = pos2min_hash(pos); + p->next_hash = 0; + return p; +} + +void ext3_htree_free_dir_info(struct dir_private_info *p) +{ + free_rb_tree_fname(&p->root); + kfree(p); +} + +/* + * Given a directory entry, enter it into the fname rb tree. + */ +int ext3_htree_store_dirent(struct file *dir_file, __u32 hash, + __u32 minor_hash, + struct ext3_dir_entry_2 *dirent) +{ + struct rb_node **p, *parent = NULL; + struct fname * fname, *new_fn; + struct dir_private_info *info; + int len; + + info = (struct dir_private_info *) dir_file->private_data; + p = &info->root.rb_node; + + /* Create and allocate the fname structure */ + len = sizeof(struct fname) + dirent->name_len + 1; + new_fn = kzalloc(len, GFP_KERNEL); + if (!new_fn) + return -ENOMEM; + new_fn->hash = hash; + new_fn->minor_hash = minor_hash; + new_fn->inode = le32_to_cpu(dirent->inode); + new_fn->name_len = dirent->name_len; + new_fn->file_type = dirent->file_type; + memcpy(new_fn->name, dirent->name, dirent->name_len); + new_fn->name[dirent->name_len] = 0; + + while (*p) { + parent = *p; + fname = rb_entry(parent, struct fname, rb_hash); + + /* + * If the hash and minor hash match up, then we put + * them on a linked list. This rarely happens... + */ + if ((new_fn->hash == fname->hash) && + (new_fn->minor_hash == fname->minor_hash)) { + new_fn->next = fname->next; + fname->next = new_fn; + return 0; + } + + if (new_fn->hash < fname->hash) + p = &(*p)->rb_left; + else if (new_fn->hash > fname->hash) + p = &(*p)->rb_right; + else if (new_fn->minor_hash < fname->minor_hash) + p = &(*p)->rb_left; + else /* if (new_fn->minor_hash > fname->minor_hash) */ + p = &(*p)->rb_right; + } + + rb_link_node(&new_fn->rb_hash, parent, p); + rb_insert_color(&new_fn->rb_hash, &info->root); + return 0; +} + + + +/* + * This is a helper function for ext3_dx_readdir. It calls filldir + * for all entres on the fname linked list. (Normally there is only + * one entry on the linked list, unless there are 62 bit hash collisions.) + */ +static int call_filldir(struct file * filp, void * dirent, + filldir_t filldir, struct fname *fname) +{ + struct dir_private_info *info = filp->private_data; + loff_t curr_pos; + struct inode *inode = filp->f_dentry->d_inode; + struct super_block * sb; + int error; + + sb = inode->i_sb; + + if (!fname) { + printk("call_filldir: called with null fname?!?\n"); + return 0; + } + curr_pos = hash2pos(fname->hash, fname->minor_hash); + while (fname) { + error = filldir(dirent, fname->name, + fname->name_len, curr_pos, + fname->inode, + get_dtype(sb, fname->file_type)); + if (error) { + filp->f_pos = curr_pos; + info->extra_fname = fname->next; + return error; + } + fname = fname->next; + } + return 0; +} + +static int ext3_dx_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + struct dir_private_info *info = filp->private_data; + struct inode *inode = filp->f_dentry->d_inode; + struct fname *fname; + int ret; + + if (!info) { + info = create_dir_info(filp->f_pos); + if (!info) + return -ENOMEM; + filp->private_data = info; + } + + if (filp->f_pos == EXT3_HTREE_EOF) + return 0; /* EOF */ + + /* Some one has messed with f_pos; reset the world */ + if (info->last_pos != filp->f_pos) { + free_rb_tree_fname(&info->root); + info->curr_node = NULL; + info->extra_fname = NULL; + info->curr_hash = pos2maj_hash(filp->f_pos); + info->curr_minor_hash = pos2min_hash(filp->f_pos); + } + + /* + * If there are any leftover names on the hash collision + * chain, return them first. + */ + if (info->extra_fname && + call_filldir(filp, dirent, filldir, info->extra_fname)) + goto finished; + + if (!info->curr_node) + info->curr_node = rb_first(&info->root); + + while (1) { + /* + * Fill the rbtree if we have no more entries, + * or the inode has changed since we last read in the + * cached entries. + */ + if ((!info->curr_node) || + (filp->f_version != inode->i_version)) { + info->curr_node = NULL; + free_rb_tree_fname(&info->root); + filp->f_version = inode->i_version; + ret = ext3_htree_fill_tree(filp, info->curr_hash, + info->curr_minor_hash, + &info->next_hash); + if (ret < 0) + return ret; + if (ret == 0) { + filp->f_pos = EXT3_HTREE_EOF; + break; + } + info->curr_node = rb_first(&info->root); + } + + fname = rb_entry(info->curr_node, struct fname, rb_hash); + info->curr_hash = fname->hash; + info->curr_minor_hash = fname->minor_hash; + if (call_filldir(filp, dirent, filldir, fname)) + break; + + info->curr_node = rb_next(info->curr_node); + if (!info->curr_node) { + if (info->next_hash == ~0) { + filp->f_pos = EXT3_HTREE_EOF; + break; + } + info->curr_hash = info->next_hash; + info->curr_minor_hash = 0; + } + } +finished: + info->last_pos = filp->f_pos; + return 0; +} + +static int ext3_release_dir (struct inode * inode, struct file * filp) +{ + if (filp->private_data) + ext3_htree_free_dir_info(filp->private_data); + + return 0; +} + +#endif diff --git a/fs/ext4/file.c b/fs/ext4/file.c new file mode 100644 index 0000000..e96c388 --- /dev/null +++ b/fs/ext4/file.c @@ -0,0 +1,139 @@ +/* + * linux/fs/ext3/file.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/file.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * ext3 fs regular file handling primitives + * + * 64-bit file support on 64-bit platforms by Jakub Jelinek + * (jj@sunsite.ms.mff.cuni.cz) + */ + +#include +#include +#include +#include +#include +#include "xattr.h" +#include "acl.h" + +/* + * Called when an inode is released. Note that this is different + * from ext3_file_open: open gets called at every open, but release + * gets called only when /all/ the files are closed. + */ +static int ext3_release_file (struct inode * inode, struct file * filp) +{ + /* if we are the last writer on the inode, drop the block reservation */ + if ((filp->f_mode & FMODE_WRITE) && + (atomic_read(&inode->i_writecount) == 1)) + { + mutex_lock(&EXT3_I(inode)->truncate_mutex); + ext3_discard_reservation(inode); + mutex_unlock(&EXT3_I(inode)->truncate_mutex); + } + if (is_dx(inode) && filp->private_data) + ext3_htree_free_dir_info(filp->private_data); + + return 0; +} + +static ssize_t +ext3_file_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_dentry->d_inode; + ssize_t ret; + int err; + + ret = generic_file_aio_write(iocb, iov, nr_segs, pos); + + /* + * Skip flushing if there was an error, or if nothing was written. + */ + if (ret <= 0) + return ret; + + /* + * If the inode is IS_SYNC, or is O_SYNC and we are doing data + * journalling then we need to make sure that we force the transaction + * to disk to keep all metadata uptodate synchronously. + */ + if (file->f_flags & O_SYNC) { + /* + * If we are non-data-journaled, then the dirty data has + * already been flushed to backing store by generic_osync_inode, + * and the inode has been flushed too if there have been any + * modifications other than mere timestamp updates. + * + * Open question --- do we care about flushing timestamps too + * if the inode is IS_SYNC? + */ + if (!ext3_should_journal_data(inode)) + return ret; + + goto force_commit; + } + + /* + * So we know that there has been no forced data flush. If the inode + * is marked IS_SYNC, we need to force one ourselves. + */ + if (!IS_SYNC(inode)) + return ret; + + /* + * Open question #2 --- should we force data to disk here too? If we + * don't, the only impact is that data=writeback filesystems won't + * flush data to disk automatically on IS_SYNC, only metadata (but + * historically, that is what ext2 has done.) + */ + +force_commit: + err = ext3_force_commit(inode->i_sb); + if (err) + return err; + return ret; +} + +const struct file_operations ext3_file_operations = { + .llseek = generic_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = generic_file_aio_read, + .aio_write = ext3_file_write, + .ioctl = ext3_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = ext3_compat_ioctl, +#endif + .mmap = generic_file_mmap, + .open = generic_file_open, + .release = ext3_release_file, + .fsync = ext3_sync_file, + .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, +}; + +struct inode_operations ext3_file_inode_operations = { + .truncate = ext3_truncate, + .setattr = ext3_setattr, +#ifdef CONFIG_EXT3_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext3_listxattr, + .removexattr = generic_removexattr, +#endif + .permission = ext3_permission, +}; + diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c new file mode 100644 index 0000000..dd1fd3c --- /dev/null +++ b/fs/ext4/fsync.c @@ -0,0 +1,88 @@ +/* + * linux/fs/ext3/fsync.c + * + * Copyright (C) 1993 Stephen Tweedie (sct@redhat.com) + * from + * Copyright (C) 1992 Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * from + * linux/fs/minix/truncate.c Copyright (C) 1991, 1992 Linus Torvalds + * + * ext3fs fsync primitive + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * + * Removed unnecessary code duplication for little endian machines + * and excessive __inline__s. + * Andi Kleen, 1997 + * + * Major simplications and cleanup - we only need to do the metadata, because + * we can depend on generic_block_fdatasync() to sync the data blocks. + */ + +#include +#include +#include +#include +#include +#include +#include + +/* + * akpm: A new design for ext3_sync_file(). + * + * This is only called from sys_fsync(), sys_fdatasync() and sys_msync(). + * There cannot be a transaction open by this task. + * Another task could have dirtied this inode. Its data can be in any + * state in the journalling system. + * + * What we do is just kick off a commit and wait on it. This will snapshot the + * inode to disk. + */ + +int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync) +{ + struct inode *inode = dentry->d_inode; + int ret = 0; + + J_ASSERT(ext3_journal_current_handle() == 0); + + /* + * data=writeback: + * The caller's filemap_fdatawrite()/wait will sync the data. + * sync_inode() will sync the metadata + * + * data=ordered: + * The caller's filemap_fdatawrite() will write the data and + * sync_inode() will write the inode if it is dirty. Then the caller's + * filemap_fdatawait() will wait on the pages. + * + * data=journal: + * filemap_fdatawrite won't do anything (the buffers are clean). + * ext3_force_commit will write the file data into the journal and + * will wait on that. + * filemap_fdatawait() will encounter a ton of newly-dirtied pages + * (they were dirtied by commit). But that's OK - the blocks are + * safe in-journal, which is all fsync() needs to ensure. + */ + if (ext3_should_journal_data(inode)) { + ret = ext3_force_commit(inode->i_sb); + goto out; + } + + /* + * The VFS has written the file data. If the inode is unaltered + * then we need not start a commit. + */ + if (inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC)) { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0, /* sys_fsync did this */ + }; + ret = sync_inode(inode, &wbc); + } +out: + return ret; +} diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c new file mode 100644 index 0000000..deeb27b --- /dev/null +++ b/fs/ext4/hash.c @@ -0,0 +1,152 @@ +/* + * linux/fs/ext3/hash.c + * + * Copyright (C) 2002 by Theodore Ts'o + * + * This file is released under the GPL v2. + * + * This file may be redistributed under the terms of the GNU Public + * License. + */ + +#include +#include +#include +#include +#include + +#define DELTA 0x9E3779B9 + +static void TEA_transform(__u32 buf[4], __u32 const in[]) +{ + __u32 sum = 0; + __u32 b0 = buf[0], b1 = buf[1]; + __u32 a = in[0], b = in[1], c = in[2], d = in[3]; + int n = 16; + + do { + sum += DELTA; + b0 += ((b1 << 4)+a) ^ (b1+sum) ^ ((b1 >> 5)+b); + b1 += ((b0 << 4)+c) ^ (b0+sum) ^ ((b0 >> 5)+d); + } while(--n); + + buf[0] += b0; + buf[1] += b1; +} + + +/* The old legacy hash */ +static __u32 dx_hack_hash (const char *name, int len) +{ + __u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; + while (len--) { + __u32 hash = hash1 + (hash0 ^ (*name++ * 7152373)); + + if (hash & 0x80000000) hash -= 0x7fffffff; + hash1 = hash0; + hash0 = hash; + } + return (hash0 << 1); +} + +static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) +{ + __u32 pad, val; + int i; + + pad = (__u32)len | ((__u32)len << 8); + pad |= pad << 16; + + val = pad; + if (len > num*4) + len = num * 4; + for (i=0; i < len; i++) { + if ((i % 4) == 0) + val = pad; + val = msg[i] + (val << 8); + if ((i % 4) == 3) { + *buf++ = val; + val = pad; + num--; + } + } + if (--num >= 0) + *buf++ = val; + while (--num >= 0) + *buf++ = pad; +} + +/* + * Returns the hash of a filename. If len is 0 and name is NULL, then + * this function can be used to test whether or not a hash version is + * supported. + * + * The seed is an 4 longword (32 bits) "secret" which can be used to + * uniquify a hash. If the seed is all zero's, then some default seed + * may be used. + * + * A particular hash version specifies whether or not the seed is + * represented, and whether or not the returned hash is 32 bits or 64 + * bits. 32 bit hashes will return 0 for the minor hash. + */ +int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) +{ + __u32 hash; + __u32 minor_hash = 0; + const char *p; + int i; + __u32 in[8], buf[4]; + + /* Initialize the default seed for the hash checksum functions */ + buf[0] = 0x67452301; + buf[1] = 0xefcdab89; + buf[2] = 0x98badcfe; + buf[3] = 0x10325476; + + /* Check to see if the seed is all zero's */ + if (hinfo->seed) { + for (i=0; i < 4; i++) { + if (hinfo->seed[i]) + break; + } + if (i < 4) + memcpy(buf, hinfo->seed, sizeof(buf)); + } + + switch (hinfo->hash_version) { + case DX_HASH_LEGACY: + hash = dx_hack_hash(name, len); + break; + case DX_HASH_HALF_MD4: + p = name; + while (len > 0) { + str2hashbuf(p, len, in, 8); + half_md4_transform(buf, in); + len -= 32; + p += 32; + } + minor_hash = buf[2]; + hash = buf[1]; + break; + case DX_HASH_TEA: + p = name; + while (len > 0) { + str2hashbuf(p, len, in, 4); + TEA_transform(buf, in); + len -= 16; + p += 16; + } + hash = buf[0]; + minor_hash = buf[1]; + break; + default: + hinfo->hash = 0; + return -1; + } + hash = hash & ~1; + if (hash == (EXT3_HTREE_EOF << 1)) + hash = (EXT3_HTREE_EOF-1) << 1; + hinfo->hash = hash; + hinfo->minor_hash = minor_hash; + return 0; +} diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c new file mode 100644 index 0000000..e45dbd6 --- /dev/null +++ b/fs/ext4/ialloc.c @@ -0,0 +1,758 @@ +/* + * linux/fs/ext3/ialloc.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * BSD ufs-inspired inode and directory allocation by + * Stephen Tweedie (sct@redhat.com), 1993 + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "xattr.h" +#include "acl.h" + +/* + * ialloc.c contains the inodes allocation and deallocation routines + */ + +/* + * The free inodes are managed by bitmaps. A file system contains several + * blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap + * block for inodes, N blocks for the inode table and data blocks. + * + * The file system contains group descriptors which are located after the + * super block. Each descriptor contains the number of the bitmap block and + * the free blocks count in the block. + */ + + +/* + * Read the inode allocation bitmap for a given block_group, reading + * into the specified slot in the superblock's bitmap cache. + * + * Return buffer_head of bitmap on success or NULL. + */ +static struct buffer_head * +read_inode_bitmap(struct super_block * sb, unsigned long block_group) +{ + struct ext3_group_desc *desc; + struct buffer_head *bh = NULL; + + desc = ext3_get_group_desc(sb, block_group, NULL); + if (!desc) + goto error_out; + + bh = sb_bread(sb, le32_to_cpu(desc->bg_inode_bitmap)); + if (!bh) + ext3_error(sb, "read_inode_bitmap", + "Cannot read inode bitmap - " + "block_group = %lu, inode_bitmap = %u", + block_group, le32_to_cpu(desc->bg_inode_bitmap)); +error_out: + return bh; +} + +/* + * NOTE! When we get the inode, we're the only people + * that have access to it, and as such there are no + * race conditions we have to worry about. The inode + * is not on the hash-lists, and it cannot be reached + * through the filesystem because the directory entry + * has been deleted earlier. + * + * HOWEVER: we must make sure that we get no aliases, + * which means that we have to call "clear_inode()" + * _before_ we mark the inode not in use in the inode + * bitmaps. Otherwise a newly created file might use + * the same inode number (not actually the same pointer + * though), and then we'd have two inodes sharing the + * same inode number and space on the harddisk. + */ +void ext3_free_inode (handle_t *handle, struct inode * inode) +{ + struct super_block * sb = inode->i_sb; + int is_directory; + unsigned long ino; + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *bh2; + unsigned long block_group; + unsigned long bit; + struct ext3_group_desc * gdp; + struct ext3_super_block * es; + struct ext3_sb_info *sbi; + int fatal = 0, err; + + if (atomic_read(&inode->i_count) > 1) { + printk ("ext3_free_inode: inode has count=%d\n", + atomic_read(&inode->i_count)); + return; + } + if (inode->i_nlink) { + printk ("ext3_free_inode: inode has nlink=%d\n", + inode->i_nlink); + return; + } + if (!sb) { + printk("ext3_free_inode: inode on nonexistent device\n"); + return; + } + sbi = EXT3_SB(sb); + + ino = inode->i_ino; + ext3_debug ("freeing inode %lu\n", ino); + + /* + * Note: we must free any quota before locking the superblock, + * as writing the quota to disk may need the lock as well. + */ + DQUOT_INIT(inode); + ext3_xattr_delete_inode(handle, inode); + DQUOT_FREE_INODE(inode); + DQUOT_DROP(inode); + + is_directory = S_ISDIR(inode->i_mode); + + /* Do this BEFORE marking the inode not in use or returning an error */ + clear_inode (inode); + + es = EXT3_SB(sb)->s_es; + if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { + ext3_error (sb, "ext3_free_inode", + "reserved or nonexistent inode %lu", ino); + goto error_return; + } + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); + bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb); + bitmap_bh = read_inode_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + + BUFFER_TRACE(bitmap_bh, "get_write_access"); + fatal = ext3_journal_get_write_access(handle, bitmap_bh); + if (fatal) + goto error_return; + + /* Ok, now we can actually update the inode bitmaps.. */ + if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group), + bit, bitmap_bh->b_data)) + ext3_error (sb, "ext3_free_inode", + "bit already cleared for inode %lu", ino); + else { + gdp = ext3_get_group_desc (sb, block_group, &bh2); + + BUFFER_TRACE(bh2, "get_write_access"); + fatal = ext3_journal_get_write_access(handle, bh2); + if (fatal) goto error_return; + + if (gdp) { + spin_lock(sb_bgl_lock(sbi, block_group)); + gdp->bg_free_inodes_count = cpu_to_le16( + le16_to_cpu(gdp->bg_free_inodes_count) + 1); + if (is_directory) + gdp->bg_used_dirs_count = cpu_to_le16( + le16_to_cpu(gdp->bg_used_dirs_count) - 1); + spin_unlock(sb_bgl_lock(sbi, block_group)); + percpu_counter_inc(&sbi->s_freeinodes_counter); + if (is_directory) + percpu_counter_dec(&sbi->s_dirs_counter); + + } + BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, bh2); + if (!fatal) fatal = err; + } + BUFFER_TRACE(bitmap_bh, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, bitmap_bh); + if (!fatal) + fatal = err; + sb->s_dirt = 1; +error_return: + brelse(bitmap_bh); + ext3_std_error(sb, fatal); +} + +/* + * There are two policies for allocating an inode. If the new inode is + * a directory, then a forward search is made for a block group with both + * free space and a low directory-to-inode ratio; if that fails, then of + * the groups with above-average free space, that group with the fewest + * directories already is chosen. + * + * For other inodes, search forward from the parent directory\'s block + * group to find a free inode. + */ +static int find_group_dir(struct super_block *sb, struct inode *parent) +{ + int ngroups = EXT3_SB(sb)->s_groups_count; + unsigned int freei, avefreei; + struct ext3_group_desc *desc, *best_desc = NULL; + struct buffer_head *bh; + int group, best_group = -1; + + freei = percpu_counter_read_positive(&EXT3_SB(sb)->s_freeinodes_counter); + avefreei = freei / ngroups; + + for (group = 0; group < ngroups; group++) { + desc = ext3_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) + continue; + if (!best_desc || + (le16_to_cpu(desc->bg_free_blocks_count) > + le16_to_cpu(best_desc->bg_free_blocks_count))) { + best_group = group; + best_desc = desc; + } + } + return best_group; +} + +/* + * Orlov's allocator for directories. + * + * We always try to spread first-level directories. + * + * If there are blockgroups with both free inodes and free blocks counts + * not worse than average we return one with smallest directory count. + * Otherwise we simply return a random group. + * + * For the rest rules look so: + * + * It's OK to put directory into a group unless + * it has too many directories already (max_dirs) or + * it has too few free inodes left (min_inodes) or + * it has too few free blocks left (min_blocks) or + * it's already running too large debt (max_debt). + * Parent's group is prefered, if it doesn't satisfy these + * conditions we search cyclically through the rest. If none + * of the groups look good we just look for a group with more + * free inodes than average (starting at parent's group). + * + * Debt is incremented each time we allocate a directory and decremented + * when we allocate an inode, within 0--255. + */ + +#define INODE_COST 64 +#define BLOCK_COST 256 + +static int find_group_orlov(struct super_block *sb, struct inode *parent) +{ + int parent_group = EXT3_I(parent)->i_block_group; + struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext3_super_block *es = sbi->s_es; + int ngroups = sbi->s_groups_count; + int inodes_per_group = EXT3_INODES_PER_GROUP(sb); + unsigned int freei, avefreei; + ext3_fsblk_t freeb, avefreeb; + ext3_fsblk_t blocks_per_dir; + unsigned int ndirs; + int max_debt, max_dirs, min_inodes; + ext3_grpblk_t min_blocks; + int group = -1, i; + struct ext3_group_desc *desc; + struct buffer_head *bh; + + freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter); + avefreei = freei / ngroups; + freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + avefreeb = freeb / ngroups; + ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter); + + if ((parent == sb->s_root->d_inode) || + (EXT3_I(parent)->i_flags & EXT3_TOPDIR_FL)) { + int best_ndir = inodes_per_group; + int best_group = -1; + + get_random_bytes(&group, sizeof(group)); + parent_group = (unsigned)group % ngroups; + for (i = 0; i < ngroups; i++) { + group = (parent_group + i) % ngroups; + desc = ext3_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) + continue; + if (le16_to_cpu(desc->bg_free_blocks_count) < avefreeb) + continue; + best_group = group; + best_ndir = le16_to_cpu(desc->bg_used_dirs_count); + } + if (best_group >= 0) + return best_group; + goto fallback; + } + + blocks_per_dir = (le32_to_cpu(es->s_blocks_count) - freeb) / ndirs; + + max_dirs = ndirs / ngroups + inodes_per_group / 16; + min_inodes = avefreei - inodes_per_group / 4; + min_blocks = avefreeb - EXT3_BLOCKS_PER_GROUP(sb) / 4; + + max_debt = EXT3_BLOCKS_PER_GROUP(sb) / max(blocks_per_dir, (ext3_fsblk_t)BLOCK_COST); + if (max_debt * INODE_COST > inodes_per_group) + max_debt = inodes_per_group / INODE_COST; + if (max_debt > 255) + max_debt = 255; + if (max_debt == 0) + max_debt = 1; + + for (i = 0; i < ngroups; i++) { + group = (parent_group + i) % ngroups; + desc = ext3_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) < min_inodes) + continue; + if (le16_to_cpu(desc->bg_free_blocks_count) < min_blocks) + continue; + return group; + } + +fallback: + for (i = 0; i < ngroups; i++) { + group = (parent_group + i) % ngroups; + desc = ext3_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei) + return group; + } + + if (avefreei) { + /* + * The free-inodes counter is approximate, and for really small + * filesystems the above test can fail to find any blockgroups + */ + avefreei = 0; + goto fallback; + } + + return -1; +} + +static int find_group_other(struct super_block *sb, struct inode *parent) +{ + int parent_group = EXT3_I(parent)->i_block_group; + int ngroups = EXT3_SB(sb)->s_groups_count; + struct ext3_group_desc *desc; + struct buffer_head *bh; + int group, i; + + /* + * Try to place the inode in its parent directory + */ + group = parent_group; + desc = ext3_get_group_desc (sb, group, &bh); + if (desc && le16_to_cpu(desc->bg_free_inodes_count) && + le16_to_cpu(desc->bg_free_blocks_count)) + return group; + + /* + * We're going to place this inode in a different blockgroup from its + * parent. We want to cause files in a common directory to all land in + * the same blockgroup. But we want files which are in a different + * directory which shares a blockgroup with our parent to land in a + * different blockgroup. + * + * So add our directory's i_ino into the starting point for the hash. + */ + group = (group + parent->i_ino) % ngroups; + + /* + * Use a quadratic hash to find a group with a free inode and some free + * blocks. + */ + for (i = 1; i < ngroups; i <<= 1) { + group += i; + if (group >= ngroups) + group -= ngroups; + desc = ext3_get_group_desc (sb, group, &bh); + if (desc && le16_to_cpu(desc->bg_free_inodes_count) && + le16_to_cpu(desc->bg_free_blocks_count)) + return group; + } + + /* + * That failed: try linear search for a free inode, even if that group + * has no free blocks. + */ + group = parent_group; + for (i = 0; i < ngroups; i++) { + if (++group >= ngroups) + group = 0; + desc = ext3_get_group_desc (sb, group, &bh); + if (desc && le16_to_cpu(desc->bg_free_inodes_count)) + return group; + } + + return -1; +} + +/* + * There are two policies for allocating an inode. If the new inode is + * a directory, then a forward search is made for a block group with both + * free space and a low directory-to-inode ratio; if that fails, then of + * the groups with above-average free space, that group with the fewest + * directories already is chosen. + * + * For other inodes, search forward from the parent directory's block + * group to find a free inode. + */ +struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) +{ + struct super_block *sb; + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *bh2; + int group; + unsigned long ino = 0; + struct inode * inode; + struct ext3_group_desc * gdp = NULL; + struct ext3_super_block * es; + struct ext3_inode_info *ei; + struct ext3_sb_info *sbi; + int err = 0; + struct inode *ret; + int i; + + /* Cannot create files in a deleted directory */ + if (!dir || !dir->i_nlink) + return ERR_PTR(-EPERM); + + sb = dir->i_sb; + inode = new_inode(sb); + if (!inode) + return ERR_PTR(-ENOMEM); + ei = EXT3_I(inode); + + sbi = EXT3_SB(sb); + es = sbi->s_es; + if (S_ISDIR(mode)) { + if (test_opt (sb, OLDALLOC)) + group = find_group_dir(sb, dir); + else + group = find_group_orlov(sb, dir); + } else + group = find_group_other(sb, dir); + + err = -ENOSPC; + if (group == -1) + goto out; + + for (i = 0; i < sbi->s_groups_count; i++) { + err = -EIO; + + gdp = ext3_get_group_desc(sb, group, &bh2); + if (!gdp) + goto fail; + + brelse(bitmap_bh); + bitmap_bh = read_inode_bitmap(sb, group); + if (!bitmap_bh) + goto fail; + + ino = 0; + +repeat_in_this_group: + ino = ext3_find_next_zero_bit((unsigned long *) + bitmap_bh->b_data, EXT3_INODES_PER_GROUP(sb), ino); + if (ino < EXT3_INODES_PER_GROUP(sb)) { + + BUFFER_TRACE(bitmap_bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, bitmap_bh); + if (err) + goto fail; + + if (!ext3_set_bit_atomic(sb_bgl_lock(sbi, group), + ino, bitmap_bh->b_data)) { + /* we won it */ + BUFFER_TRACE(bitmap_bh, + "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, + bitmap_bh); + if (err) + goto fail; + goto got; + } + /* we lost it */ + journal_release_buffer(handle, bitmap_bh); + + if (++ino < EXT3_INODES_PER_GROUP(sb)) + goto repeat_in_this_group; + } + + /* + * This case is possible in concurrent environment. It is very + * rare. We cannot repeat the find_group_xxx() call because + * that will simply return the same blockgroup, because the + * group descriptor metadata has not yet been updated. + * So we just go onto the next blockgroup. + */ + if (++group == sbi->s_groups_count) + group = 0; + } + err = -ENOSPC; + goto out; + +got: + ino += group * EXT3_INODES_PER_GROUP(sb) + 1; + if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { + ext3_error (sb, "ext3_new_inode", + "reserved inode or inode > inodes count - " + "block_group = %d, inode=%lu", group, ino); + err = -EIO; + goto fail; + } + + BUFFER_TRACE(bh2, "get_write_access"); + err = ext3_journal_get_write_access(handle, bh2); + if (err) goto fail; + spin_lock(sb_bgl_lock(sbi, group)); + gdp->bg_free_inodes_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1); + if (S_ISDIR(mode)) { + gdp->bg_used_dirs_count = + cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); + } + spin_unlock(sb_bgl_lock(sbi, group)); + BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, bh2); + if (err) goto fail; + + percpu_counter_dec(&sbi->s_freeinodes_counter); + if (S_ISDIR(mode)) + percpu_counter_inc(&sbi->s_dirs_counter); + sb->s_dirt = 1; + + inode->i_uid = current->fsuid; + if (test_opt (sb, GRPID)) + inode->i_gid = dir->i_gid; + else if (dir->i_mode & S_ISGID) { + inode->i_gid = dir->i_gid; + if (S_ISDIR(mode)) + mode |= S_ISGID; + } else + inode->i_gid = current->fsgid; + inode->i_mode = mode; + + inode->i_ino = ino; + /* This is the optimal IO size (for stat), not the fs block size */ + inode->i_blocks = 0; + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; + + memset(ei->i_data, 0, sizeof(ei->i_data)); + ei->i_dir_start_lookup = 0; + ei->i_disksize = 0; + + ei->i_flags = EXT3_I(dir)->i_flags & ~EXT3_INDEX_FL; + if (S_ISLNK(mode)) + ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL); + /* dirsync only applies to directories */ + if (!S_ISDIR(mode)) + ei->i_flags &= ~EXT3_DIRSYNC_FL; +#ifdef EXT3_FRAGMENTS + ei->i_faddr = 0; + ei->i_frag_no = 0; + ei->i_frag_size = 0; +#endif + ei->i_file_acl = 0; + ei->i_dir_acl = 0; + ei->i_dtime = 0; + ei->i_block_alloc_info = NULL; + ei->i_block_group = group; + + ext3_set_inode_flags(inode); + if (IS_DIRSYNC(inode)) + handle->h_sync = 1; + insert_inode_hash(inode); + spin_lock(&sbi->s_next_gen_lock); + inode->i_generation = sbi->s_next_generation++; + spin_unlock(&sbi->s_next_gen_lock); + + ei->i_state = EXT3_STATE_NEW; + ei->i_extra_isize = + (EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) ? + sizeof(struct ext3_inode) - EXT3_GOOD_OLD_INODE_SIZE : 0; + + ret = inode; + if(DQUOT_ALLOC_INODE(inode)) { + err = -EDQUOT; + goto fail_drop; + } + + err = ext3_init_acl(handle, inode, dir); + if (err) + goto fail_free_drop; + + err = ext3_init_security(handle,inode, dir); + if (err) + goto fail_free_drop; + + err = ext3_mark_inode_dirty(handle, inode); + if (err) { + ext3_std_error(sb, err); + goto fail_free_drop; + } + + ext3_debug("allocating inode %lu\n", inode->i_ino); + goto really_out; +fail: + ext3_std_error(sb, err); +out: + iput(inode); + ret = ERR_PTR(err); +really_out: + brelse(bitmap_bh); + return ret; + +fail_free_drop: + DQUOT_FREE_INODE(inode); + +fail_drop: + DQUOT_DROP(inode); + inode->i_flags |= S_NOQUOTA; + inode->i_nlink = 0; + iput(inode); + brelse(bitmap_bh); + return ERR_PTR(err); +} + +/* Verify that we are loading a valid orphan from disk */ +struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) +{ + unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count); + unsigned long block_group; + int bit; + struct buffer_head *bitmap_bh = NULL; + struct inode *inode = NULL; + + /* Error cases - e2fsck has already cleaned up for us */ + if (ino > max_ino) { + ext3_warning(sb, __FUNCTION__, + "bad orphan ino %lu! e2fsck was run?", ino); + goto out; + } + + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); + bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb); + bitmap_bh = read_inode_bitmap(sb, block_group); + if (!bitmap_bh) { + ext3_warning(sb, __FUNCTION__, + "inode bitmap error for orphan %lu", ino); + goto out; + } + + /* Having the inode bit set should be a 100% indicator that this + * is a valid orphan (no e2fsck run on fs). Orphans also include + * inodes that were being truncated, so we can't check i_nlink==0. + */ + if (!ext3_test_bit(bit, bitmap_bh->b_data) || + !(inode = iget(sb, ino)) || is_bad_inode(inode) || + NEXT_ORPHAN(inode) > max_ino) { + ext3_warning(sb, __FUNCTION__, + "bad orphan inode %lu! e2fsck was run?", ino); + printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n", + bit, (unsigned long long)bitmap_bh->b_blocknr, + ext3_test_bit(bit, bitmap_bh->b_data)); + printk(KERN_NOTICE "inode=%p\n", inode); + if (inode) { + printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", + is_bad_inode(inode)); + printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", + NEXT_ORPHAN(inode)); + printk(KERN_NOTICE "max_ino=%lu\n", max_ino); + } + /* Avoid freeing blocks if we got a bad deleted inode */ + if (inode && inode->i_nlink == 0) + inode->i_blocks = 0; + iput(inode); + inode = NULL; + } +out: + brelse(bitmap_bh); + return inode; +} + +unsigned long ext3_count_free_inodes (struct super_block * sb) +{ + unsigned long desc_count; + struct ext3_group_desc *gdp; + int i; +#ifdef EXT3FS_DEBUG + struct ext3_super_block *es; + unsigned long bitmap_count, x; + struct buffer_head *bitmap_bh = NULL; + + es = EXT3_SB(sb)->s_es; + desc_count = 0; + bitmap_count = 0; + gdp = NULL; + for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { + gdp = ext3_get_group_desc (sb, i, NULL); + if (!gdp) + continue; + desc_count += le16_to_cpu(gdp->bg_free_inodes_count); + brelse(bitmap_bh); + bitmap_bh = read_inode_bitmap(sb, i); + if (!bitmap_bh) + continue; + + x = ext3_count_free(bitmap_bh, EXT3_INODES_PER_GROUP(sb) / 8); + printk("group %d: stored = %d, counted = %lu\n", + i, le16_to_cpu(gdp->bg_free_inodes_count), x); + bitmap_count += x; + } + brelse(bitmap_bh); + printk("ext3_count_free_inodes: stored = %u, computed = %lu, %lu\n", + le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); + return desc_count; +#else + desc_count = 0; + for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { + gdp = ext3_get_group_desc (sb, i, NULL); + if (!gdp) + continue; + desc_count += le16_to_cpu(gdp->bg_free_inodes_count); + cond_resched(); + } + return desc_count; +#endif +} + +/* Called at mount-time, super-block is locked */ +unsigned long ext3_count_dirs (struct super_block * sb) +{ + unsigned long count = 0; + int i; + + for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { + struct ext3_group_desc *gdp = ext3_get_group_desc (sb, i, NULL); + if (!gdp) + continue; + count += le16_to_cpu(gdp->bg_used_dirs_count); + } + return count; +} + diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c new file mode 100644 index 0000000..03ba5bc --- /dev/null +++ b/fs/ext4/inode.c @@ -0,0 +1,3219 @@ +/* + * linux/fs/ext3/inode.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/inode.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * Goal-directed block allocation by Stephen Tweedie + * (sct@redhat.com), 1993, 1998 + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * 64-bit file support on 64-bit platforms by Jakub Jelinek + * (jj@sunsite.ms.mff.cuni.cz) + * + * Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000 + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "xattr.h" +#include "acl.h" + +static int ext3_writepage_trans_blocks(struct inode *inode); + +/* + * Test whether an inode is a fast symlink. + */ +static int ext3_inode_is_fast_symlink(struct inode *inode) +{ + int ea_blocks = EXT3_I(inode)->i_file_acl ? + (inode->i_sb->s_blocksize >> 9) : 0; + + return (S_ISLNK(inode->i_mode) && inode->i_blocks - ea_blocks == 0); +} + +/* + * The ext3 forget function must perform a revoke if we are freeing data + * which has been journaled. Metadata (eg. indirect blocks) must be + * revoked in all cases. + * + * "bh" may be NULL: a metadata block may have been freed from memory + * but there may still be a record of it in the journal, and that record + * still needs to be revoked. + */ +int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, + struct buffer_head *bh, ext3_fsblk_t blocknr) +{ + int err; + + might_sleep(); + + BUFFER_TRACE(bh, "enter"); + + jbd_debug(4, "forgetting bh %p: is_metadata = %d, mode %o, " + "data mode %lx\n", + bh, is_metadata, inode->i_mode, + test_opt(inode->i_sb, DATA_FLAGS)); + + /* Never use the revoke function if we are doing full data + * journaling: there is no need to, and a V1 superblock won't + * support it. Otherwise, only skip the revoke on un-journaled + * data blocks. */ + + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA || + (!is_metadata && !ext3_should_journal_data(inode))) { + if (bh) { + BUFFER_TRACE(bh, "call journal_forget"); + return ext3_journal_forget(handle, bh); + } + return 0; + } + + /* + * data!=journal && (is_metadata || should_journal_data(inode)) + */ + BUFFER_TRACE(bh, "call ext3_journal_revoke"); + err = ext3_journal_revoke(handle, blocknr, bh); + if (err) + ext3_abort(inode->i_sb, __FUNCTION__, + "error %d when attempting revoke", err); + BUFFER_TRACE(bh, "exit"); + return err; +} + +/* + * Work out how many blocks we need to proceed with the next chunk of a + * truncate transaction. + */ +static unsigned long blocks_for_truncate(struct inode *inode) +{ + unsigned long needed; + + needed = inode->i_blocks >> (inode->i_sb->s_blocksize_bits - 9); + + /* Give ourselves just enough room to cope with inodes in which + * i_blocks is corrupt: we've seen disk corruptions in the past + * which resulted in random data in an inode which looked enough + * like a regular file for ext3 to try to delete it. Things + * will go a bit crazy if that happens, but at least we should + * try not to panic the whole kernel. */ + if (needed < 2) + needed = 2; + + /* But we need to bound the transaction so we don't overflow the + * journal. */ + if (needed > EXT3_MAX_TRANS_DATA) + needed = EXT3_MAX_TRANS_DATA; + + return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed; +} + +/* + * Truncate transactions can be complex and absolutely huge. So we need to + * be able to restart the transaction at a conventient checkpoint to make + * sure we don't overflow the journal. + * + * start_transaction gets us a new handle for a truncate transaction, + * and extend_transaction tries to extend the existing one a bit. If + * extend fails, we need to propagate the failure up and restart the + * transaction in the top-level truncate loop. --sct + */ +static handle_t *start_transaction(struct inode *inode) +{ + handle_t *result; + + result = ext3_journal_start(inode, blocks_for_truncate(inode)); + if (!IS_ERR(result)) + return result; + + ext3_std_error(inode->i_sb, PTR_ERR(result)); + return result; +} + +/* + * Try to extend this transaction for the purposes of truncation. + * + * Returns 0 if we managed to create more room. If we can't create more + * room, and the transaction must be restarted we return 1. + */ +static int try_to_extend_transaction(handle_t *handle, struct inode *inode) +{ + if (handle->h_buffer_credits > EXT3_RESERVE_TRANS_BLOCKS) + return 0; + if (!ext3_journal_extend(handle, blocks_for_truncate(inode))) + return 0; + return 1; +} + +/* + * Restart the transaction associated with *handle. This does a commit, + * so before we call here everything must be consistently dirtied against + * this transaction. + */ +static int ext3_journal_test_restart(handle_t *handle, struct inode *inode) +{ + jbd_debug(2, "restarting handle %p\n", handle); + return ext3_journal_restart(handle, blocks_for_truncate(inode)); +} + +/* + * Called at the last iput() if i_nlink is zero. + */ +void ext3_delete_inode (struct inode * inode) +{ + handle_t *handle; + + truncate_inode_pages(&inode->i_data, 0); + + if (is_bad_inode(inode)) + goto no_delete; + + handle = start_transaction(inode); + if (IS_ERR(handle)) { + /* + * If we're going to skip the normal cleanup, we still need to + * make sure that the in-core orphan linked list is properly + * cleaned up. + */ + ext3_orphan_del(NULL, inode); + goto no_delete; + } + + if (IS_SYNC(inode)) + handle->h_sync = 1; + inode->i_size = 0; + if (inode->i_blocks) + ext3_truncate(inode); + /* + * Kill off the orphan record which ext3_truncate created. + * AKPM: I think this can be inside the above `if'. + * Note that ext3_orphan_del() has to be able to cope with the + * deletion of a non-existent orphan - this is because we don't + * know if ext3_truncate() actually created an orphan record. + * (Well, we could do this if we need to, but heck - it works) + */ + ext3_orphan_del(handle, inode); + EXT3_I(inode)->i_dtime = get_seconds(); + + /* + * One subtle ordering requirement: if anything has gone wrong + * (transaction abort, IO errors, whatever), then we can still + * do these next steps (the fs will already have been marked as + * having errors), but we can't free the inode if the mark_dirty + * fails. + */ + if (ext3_mark_inode_dirty(handle, inode)) + /* If that failed, just do the required in-core inode clear. */ + clear_inode(inode); + else + ext3_free_inode(handle, inode); + ext3_journal_stop(handle); + return; +no_delete: + clear_inode(inode); /* We must guarantee clearing of inode... */ +} + +typedef struct { + __le32 *p; + __le32 key; + struct buffer_head *bh; +} Indirect; + +static inline void add_chain(Indirect *p, struct buffer_head *bh, __le32 *v) +{ + p->key = *(p->p = v); + p->bh = bh; +} + +static int verify_chain(Indirect *from, Indirect *to) +{ + while (from <= to && from->key == *from->p) + from++; + return (from > to); +} + +/** + * ext3_block_to_path - parse the block number into array of offsets + * @inode: inode in question (we are only interested in its superblock) + * @i_block: block number to be parsed + * @offsets: array to store the offsets in + * @boundary: set this non-zero if the referred-to block is likely to be + * followed (on disk) by an indirect block. + * + * To store the locations of file's data ext3 uses a data structure common + * for UNIX filesystems - tree of pointers anchored in the inode, with + * data blocks at leaves and indirect blocks in intermediate nodes. + * This function translates the block number into path in that tree - + * return value is the path length and @offsets[n] is the offset of + * pointer to (n+1)th node in the nth one. If @block is out of range + * (negative or too large) warning is printed and zero returned. + * + * Note: function doesn't find node addresses, so no IO is needed. All + * we need to know is the capacity of indirect blocks (taken from the + * inode->i_sb). + */ + +/* + * Portability note: the last comparison (check that we fit into triple + * indirect block) is spelled differently, because otherwise on an + * architecture with 32-bit longs and 8Kb pages we might get into trouble + * if our filesystem had 8Kb blocks. We might use long long, but that would + * kill us on x86. Oh, well, at least the sign propagation does not matter - + * i_block would have to be negative in the very beginning, so we would not + * get there at all. + */ + +static int ext3_block_to_path(struct inode *inode, + long i_block, int offsets[4], int *boundary) +{ + int ptrs = EXT3_ADDR_PER_BLOCK(inode->i_sb); + int ptrs_bits = EXT3_ADDR_PER_BLOCK_BITS(inode->i_sb); + const long direct_blocks = EXT3_NDIR_BLOCKS, + indirect_blocks = ptrs, + double_blocks = (1 << (ptrs_bits * 2)); + int n = 0; + int final = 0; + + if (i_block < 0) { + ext3_warning (inode->i_sb, "ext3_block_to_path", "block < 0"); + } else if (i_block < direct_blocks) { + offsets[n++] = i_block; + final = direct_blocks; + } else if ( (i_block -= direct_blocks) < indirect_blocks) { + offsets[n++] = EXT3_IND_BLOCK; + offsets[n++] = i_block; + final = ptrs; + } else if ((i_block -= indirect_blocks) < double_blocks) { + offsets[n++] = EXT3_DIND_BLOCK; + offsets[n++] = i_block >> ptrs_bits; + offsets[n++] = i_block & (ptrs - 1); + final = ptrs; + } else if (((i_block -= double_blocks) >> (ptrs_bits * 2)) < ptrs) { + offsets[n++] = EXT3_TIND_BLOCK; + offsets[n++] = i_block >> (ptrs_bits * 2); + offsets[n++] = (i_block >> ptrs_bits) & (ptrs - 1); + offsets[n++] = i_block & (ptrs - 1); + final = ptrs; + } else { + ext3_warning(inode->i_sb, "ext3_block_to_path", "block > big"); + } + if (boundary) + *boundary = final - 1 - (i_block & (ptrs - 1)); + return n; +} + +/** + * ext3_get_branch - read the chain of indirect blocks leading to data + * @inode: inode in question + * @depth: depth of the chain (1 - direct pointer, etc.) + * @offsets: offsets of pointers in inode/indirect blocks + * @chain: place to store the result + * @err: here we store the error value + * + * Function fills the array of triples and returns %NULL + * if everything went OK or the pointer to the last filled triple + * (incomplete one) otherwise. Upon the return chain[i].key contains + * the number of (i+1)-th block in the chain (as it is stored in memory, + * i.e. little-endian 32-bit), chain[i].p contains the address of that + * number (it points into struct inode for i==0 and into the bh->b_data + * for i>0) and chain[i].bh points to the buffer_head of i-th indirect + * block for i>0 and NULL for i==0. In other words, it holds the block + * numbers of the chain, addresses they were taken from (and where we can + * verify that chain did not change) and buffer_heads hosting these + * numbers. + * + * Function stops when it stumbles upon zero pointer (absent block) + * (pointer to last triple returned, *@err == 0) + * or when it gets an IO error reading an indirect block + * (ditto, *@err == -EIO) + * or when it notices that chain had been changed while it was reading + * (ditto, *@err == -EAGAIN) + * or when it reads all @depth-1 indirect blocks successfully and finds + * the whole chain, all way to the data (returns %NULL, *err == 0). + */ +static Indirect *ext3_get_branch(struct inode *inode, int depth, int *offsets, + Indirect chain[4], int *err) +{ + struct super_block *sb = inode->i_sb; + Indirect *p = chain; + struct buffer_head *bh; + + *err = 0; + /* i_data is not going away, no lock needed */ + add_chain (chain, NULL, EXT3_I(inode)->i_data + *offsets); + if (!p->key) + goto no_block; + while (--depth) { + bh = sb_bread(sb, le32_to_cpu(p->key)); + if (!bh) + goto failure; + /* Reader: pointers */ + if (!verify_chain(chain, p)) + goto changed; + add_chain(++p, bh, (__le32*)bh->b_data + *++offsets); + /* Reader: end */ + if (!p->key) + goto no_block; + } + return NULL; + +changed: + brelse(bh); + *err = -EAGAIN; + goto no_block; +failure: + *err = -EIO; +no_block: + return p; +} + +/** + * ext3_find_near - find a place for allocation with sufficient locality + * @inode: owner + * @ind: descriptor of indirect block. + * + * This function returns the prefered place for block allocation. + * It is used when heuristic for sequential allocation fails. + * Rules are: + * + if there is a block to the left of our position - allocate near it. + * + if pointer will live in indirect block - allocate near that block. + * + if pointer will live in inode - allocate in the same + * cylinder group. + * + * In the latter case we colour the starting block by the callers PID to + * prevent it from clashing with concurrent allocations for a different inode + * in the same block group. The PID is used here so that functionally related + * files will be close-by on-disk. + * + * Caller must make sure that @ind is valid and will stay that way. + */ +static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind) +{ + struct ext3_inode_info *ei = EXT3_I(inode); + __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; + __le32 *p; + ext3_fsblk_t bg_start; + ext3_grpblk_t colour; + + /* Try to find previous block */ + for (p = ind->p - 1; p >= start; p--) { + if (*p) + return le32_to_cpu(*p); + } + + /* No such thing, so let's try location of indirect block */ + if (ind->bh) + return ind->bh->b_blocknr; + + /* + * It is going to be referred to from the inode itself? OK, just put it + * into the same cylinder group then. + */ + bg_start = ext3_group_first_block_no(inode->i_sb, ei->i_block_group); + colour = (current->pid % 16) * + (EXT3_BLOCKS_PER_GROUP(inode->i_sb) / 16); + return bg_start + colour; +} + +/** + * ext3_find_goal - find a prefered place for allocation. + * @inode: owner + * @block: block we want + * @chain: chain of indirect blocks + * @partial: pointer to the last triple within a chain + * @goal: place to store the result. + * + * Normally this function find the prefered place for block allocation, + * stores it in *@goal and returns zero. + */ + +static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, + Indirect chain[4], Indirect *partial) +{ + struct ext3_block_alloc_info *block_i; + + block_i = EXT3_I(inode)->i_block_alloc_info; + + /* + * try the heuristic for sequential allocation, + * failing that at least try to get decent locality. + */ + if (block_i && (block == block_i->last_alloc_logical_block + 1) + && (block_i->last_alloc_physical_block != 0)) { + return block_i->last_alloc_physical_block + 1; + } + + return ext3_find_near(inode, partial); +} + +/** + * ext3_blks_to_allocate: Look up the block map and count the number + * of direct blocks need to be allocated for the given branch. + * + * @branch: chain of indirect blocks + * @k: number of blocks need for indirect blocks + * @blks: number of data blocks to be mapped. + * @blocks_to_boundary: the offset in the indirect block + * + * return the total number of blocks to be allocate, including the + * direct and indirect blocks. + */ +static int ext3_blks_to_allocate(Indirect *branch, int k, unsigned long blks, + int blocks_to_boundary) +{ + unsigned long count = 0; + + /* + * Simple case, [t,d]Indirect block(s) has not allocated yet + * then it's clear blocks on that path have not allocated + */ + if (k > 0) { + /* right now we don't handle cross boundary allocation */ + if (blks < blocks_to_boundary + 1) + count += blks; + else + count += blocks_to_boundary + 1; + return count; + } + + count++; + while (count < blks && count <= blocks_to_boundary && + le32_to_cpu(*(branch[0].p + count)) == 0) { + count++; + } + return count; +} + +/** + * ext3_alloc_blocks: multiple allocate blocks needed for a branch + * @indirect_blks: the number of blocks need to allocate for indirect + * blocks + * + * @new_blocks: on return it will store the new block numbers for + * the indirect blocks(if needed) and the first direct block, + * @blks: on return it will store the total number of allocated + * direct blocks + */ +static int ext3_alloc_blocks(handle_t *handle, struct inode *inode, + ext3_fsblk_t goal, int indirect_blks, int blks, + ext3_fsblk_t new_blocks[4], int *err) +{ + int target, i; + unsigned long count = 0; + int index = 0; + ext3_fsblk_t current_block = 0; + int ret = 0; + + /* + * Here we try to allocate the requested multiple blocks at once, + * on a best-effort basis. + * To build a branch, we should allocate blocks for + * the indirect blocks(if not allocated yet), and at least + * the first direct block of this branch. That's the + * minimum number of blocks need to allocate(required) + */ + target = blks + indirect_blks; + + while (1) { + count = target; + /* allocating blocks for indirect blocks and direct blocks */ + current_block = ext3_new_blocks(handle,inode,goal,&count,err); + if (*err) + goto failed_out; + + target -= count; + /* allocate blocks for indirect blocks */ + while (index < indirect_blks && count) { + new_blocks[index++] = current_block++; + count--; + } + + if (count > 0) + break; + } + + /* save the new block number for the first direct block */ + new_blocks[index] = current_block; + + /* total number of blocks allocated for direct blocks */ + ret = count; + *err = 0; + return ret; +failed_out: + for (i = 0; i key). Upon the exit we have the same + * picture as after the successful ext3_get_block(), except that in one + * place chain is disconnected - *branch->p is still zero (we did not + * set the last link), but branch->key contains the number that should + * be placed into *branch->p to fill that gap. + * + * If allocation fails we free all blocks we've allocated (and forget + * their buffer_heads) and return the error value the from failed + * ext3_alloc_block() (normally -ENOSPC). Otherwise we set the chain + * as described above and return 0. + */ +static int ext3_alloc_branch(handle_t *handle, struct inode *inode, + int indirect_blks, int *blks, ext3_fsblk_t goal, + int *offsets, Indirect *branch) +{ + int blocksize = inode->i_sb->s_blocksize; + int i, n = 0; + int err = 0; + struct buffer_head *bh; + int num; + ext3_fsblk_t new_blocks[4]; + ext3_fsblk_t current_block; + + num = ext3_alloc_blocks(handle, inode, goal, indirect_blks, + *blks, new_blocks, &err); + if (err) + return err; + + branch[0].key = cpu_to_le32(new_blocks[0]); + /* + * metadata blocks and data blocks are allocated. + */ + for (n = 1; n <= indirect_blks; n++) { + /* + * Get buffer_head for parent block, zero it out + * and set the pointer to new one, then send + * parent to disk. + */ + bh = sb_getblk(inode->i_sb, new_blocks[n-1]); + branch[n].bh = bh; + lock_buffer(bh); + BUFFER_TRACE(bh, "call get_create_access"); + err = ext3_journal_get_create_access(handle, bh); + if (err) { + unlock_buffer(bh); + brelse(bh); + goto failed; + } + + memset(bh->b_data, 0, blocksize); + branch[n].p = (__le32 *) bh->b_data + offsets[n]; + branch[n].key = cpu_to_le32(new_blocks[n]); + *branch[n].p = branch[n].key; + if ( n == indirect_blks) { + current_block = new_blocks[n]; + /* + * End of chain, update the last new metablock of + * the chain to point to the new allocated + * data blocks numbers + */ + for (i=1; i < num; i++) + *(branch[n].p + i) = cpu_to_le32(++current_block); + } + BUFFER_TRACE(bh, "marking uptodate"); + set_buffer_uptodate(bh); + unlock_buffer(bh); + + BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, bh); + if (err) + goto failed; + } + *blks = num; + return err; +failed: + /* Allocation failed, free what we already allocated */ + for (i = 1; i <= n ; i++) { + BUFFER_TRACE(branch[i].bh, "call journal_forget"); + ext3_journal_forget(handle, branch[i].bh); + } + for (i = 0; i i_blocks, etc.). In case of success we end up with the full + * chain to new block and return 0. + */ +static int ext3_splice_branch(handle_t *handle, struct inode *inode, + long block, Indirect *where, int num, int blks) +{ + int i; + int err = 0; + struct ext3_block_alloc_info *block_i; + ext3_fsblk_t current_block; + + block_i = EXT3_I(inode)->i_block_alloc_info; + /* + * If we're splicing into a [td]indirect block (as opposed to the + * inode) then we need to get write access to the [td]indirect block + * before the splice. + */ + if (where->bh) { + BUFFER_TRACE(where->bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, where->bh); + if (err) + goto err_out; + } + /* That's it */ + + *where->p = where->key; + + /* + * Update the host buffer_head or inode to point to more just allocated + * direct blocks blocks + */ + if (num == 0 && blks > 1) { + current_block = le32_to_cpu(where->key) + 1; + for (i = 1; i < blks; i++) + *(where->p + i ) = cpu_to_le32(current_block++); + } + + /* + * update the most recently allocated logical & physical block + * in i_block_alloc_info, to assist find the proper goal block for next + * allocation + */ + if (block_i) { + block_i->last_alloc_logical_block = block + blks - 1; + block_i->last_alloc_physical_block = + le32_to_cpu(where[num].key) + blks - 1; + } + + /* We are done with atomic stuff, now do the rest of housekeeping */ + + inode->i_ctime = CURRENT_TIME_SEC; + ext3_mark_inode_dirty(handle, inode); + + /* had we spliced it onto indirect block? */ + if (where->bh) { + /* + * If we spliced it onto an indirect block, we haven't + * altered the inode. Note however that if it is being spliced + * onto an indirect block at the very end of the file (the + * file is growing) then we *will* alter the inode to reflect + * the new i_size. But that is not done here - it is done in + * generic_commit_write->__mark_inode_dirty->ext3_dirty_inode. + */ + jbd_debug(5, "splicing indirect only\n"); + BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, where->bh); + if (err) + goto err_out; + } else { + /* + * OK, we spliced it into the inode itself on a direct block. + * Inode was dirtied above. + */ + jbd_debug(5, "splicing direct\n"); + } + return err; + +err_out: + for (i = 1; i <= num; i++) { + BUFFER_TRACE(where[i].bh, "call journal_forget"); + ext3_journal_forget(handle, where[i].bh); + ext3_free_blocks(handle,inode,le32_to_cpu(where[i-1].key),1); + } + ext3_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks); + + return err; +} + +/* + * Allocation strategy is simple: if we have to allocate something, we will + * have to go the whole way to leaf. So let's do it before attaching anything + * to tree, set linkage between the newborn blocks, write them if sync is + * required, recheck the path, free and repeat if check fails, otherwise + * set the last missing link (that will protect us from any truncate-generated + * removals - all blocks on the path are immune now) and possibly force the + * write on the parent block. + * That has a nice additional property: no special recovery from the failed + * allocations is needed - we simply release blocks and do not touch anything + * reachable from inode. + * + * `handle' can be NULL if create == 0. + * + * The BKL may not be held on entry here. Be sure to take it early. + * return > 0, # of blocks mapped or allocated. + * return = 0, if plain lookup failed. + * return < 0, error case. + */ +int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, + sector_t iblock, unsigned long maxblocks, + struct buffer_head *bh_result, + int create, int extend_disksize) +{ + int err = -EIO; + int offsets[4]; + Indirect chain[4]; + Indirect *partial; + ext3_fsblk_t goal; + int indirect_blks; + int blocks_to_boundary = 0; + int depth; + struct ext3_inode_info *ei = EXT3_I(inode); + int count = 0; + ext3_fsblk_t first_block = 0; + + + J_ASSERT(handle != NULL || create == 0); + depth = ext3_block_to_path(inode,iblock,offsets,&blocks_to_boundary); + + if (depth == 0) + goto out; + + partial = ext3_get_branch(inode, depth, offsets, chain, &err); + + /* Simplest case - block found, no allocation needed */ + if (!partial) { + first_block = le32_to_cpu(chain[depth - 1].key); + clear_buffer_new(bh_result); + count++; + /*map more blocks*/ + while (count < maxblocks && count <= blocks_to_boundary) { + ext3_fsblk_t blk; + + if (!verify_chain(chain, partial)) { + /* + * Indirect block might be removed by + * truncate while we were reading it. + * Handling of that case: forget what we've + * got now. Flag the err as EAGAIN, so it + * will reread. + */ + err = -EAGAIN; + count = 0; + break; + } + blk = le32_to_cpu(*(chain[depth-1].p + count)); + + if (blk == first_block + count) + count++; + else + break; + } + if (err != -EAGAIN) + goto got_it; + } + + /* Next simple case - plain lookup or failed read of indirect block */ + if (!create || err == -EIO) + goto cleanup; + + mutex_lock(&ei->truncate_mutex); + + /* + * If the indirect block is missing while we are reading + * the chain(ext3_get_branch() returns -EAGAIN err), or + * if the chain has been changed after we grab the semaphore, + * (either because another process truncated this branch, or + * another get_block allocated this branch) re-grab the chain to see if + * the request block has been allocated or not. + * + * Since we already block the truncate/other get_block + * at this point, we will have the current copy of the chain when we + * splice the branch into the tree. + */ + if (err == -EAGAIN || !verify_chain(chain, partial)) { + while (partial > chain) { + brelse(partial->bh); + partial--; + } + partial = ext3_get_branch(inode, depth, offsets, chain, &err); + if (!partial) { + count++; + mutex_unlock(&ei->truncate_mutex); + if (err) + goto cleanup; + clear_buffer_new(bh_result); + goto got_it; + } + } + + /* + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary + */ + if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) + ext3_init_block_alloc_info(inode); + + goal = ext3_find_goal(inode, iblock, chain, partial); + + /* the number of blocks need to allocate for [d,t]indirect blocks */ + indirect_blks = (chain + depth) - partial - 1; + + /* + * Next look up the indirect map to count the totoal number of + * direct blocks to allocate for this branch. + */ + count = ext3_blks_to_allocate(partial, indirect_blks, + maxblocks, blocks_to_boundary); + /* + * Block out ext3_truncate while we alter the tree + */ + err = ext3_alloc_branch(handle, inode, indirect_blks, &count, goal, + offsets + (partial - chain), partial); + + /* + * The ext3_splice_branch call will free and forget any buffers + * on the new chain if there is a failure, but that risks using + * up transaction credits, especially for bitmaps where the + * credits cannot be returned. Can we handle this somehow? We + * may need to return -EAGAIN upwards in the worst case. --sct + */ + if (!err) + err = ext3_splice_branch(handle, inode, iblock, + partial, indirect_blks, count); + /* + * i_disksize growing is protected by truncate_mutex. Don't forget to + * protect it if you're about to implement concurrent + * ext3_get_block() -bzzz + */ + if (!err && extend_disksize && inode->i_size > ei->i_disksize) + ei->i_disksize = inode->i_size; + mutex_unlock(&ei->truncate_mutex); + if (err) + goto cleanup; + + set_buffer_new(bh_result); +got_it: + map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); + if (count > blocks_to_boundary) + set_buffer_boundary(bh_result); + err = count; + /* Clean up and exit */ + partial = chain + depth - 1; /* the whole chain */ +cleanup: + while (partial > chain) { + BUFFER_TRACE(partial->bh, "call brelse"); + brelse(partial->bh); + partial--; + } + BUFFER_TRACE(bh_result, "returned"); +out: + return err; +} + +#define DIO_CREDITS (EXT3_RESERVE_TRANS_BLOCKS + 32) + +static int ext3_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + handle_t *handle = journal_current_handle(); + int ret = 0; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + + if (!create) + goto get_block; /* A read */ + + if (max_blocks == 1) + goto get_block; /* A single block get */ + + if (handle->h_transaction->t_state == T_LOCKED) { + /* + * Huge direct-io writes can hold off commits for long + * periods of time. Let this commit run. + */ + ext3_journal_stop(handle); + handle = ext3_journal_start(inode, DIO_CREDITS); + if (IS_ERR(handle)) + ret = PTR_ERR(handle); + goto get_block; + } + + if (handle->h_buffer_credits <= EXT3_RESERVE_TRANS_BLOCKS) { + /* + * Getting low on buffer credits... + */ + ret = ext3_journal_extend(handle, DIO_CREDITS); + if (ret > 0) { + /* + * Couldn't extend the transaction. Start a new one. + */ + ret = ext3_journal_restart(handle, DIO_CREDITS); + } + } + +get_block: + if (ret == 0) { + ret = ext3_get_blocks_handle(handle, inode, iblock, + max_blocks, bh_result, create, 0); + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + } + return ret; +} + +/* + * `handle' can be NULL if create is zero + */ +struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, + long block, int create, int *errp) +{ + struct buffer_head dummy; + int fatal = 0, err; + + J_ASSERT(handle != NULL || create == 0); + + dummy.b_state = 0; + dummy.b_blocknr = -1000; + buffer_trace_init(&dummy.b_history); + err = ext3_get_blocks_handle(handle, inode, block, 1, + &dummy, create, 1); + /* + * ext3_get_blocks_handle() returns number of blocks + * mapped. 0 in case of a HOLE. + */ + if (err > 0) { + if (err > 1) + WARN_ON(1); + err = 0; + } + *errp = err; + if (!err && buffer_mapped(&dummy)) { + struct buffer_head *bh; + bh = sb_getblk(inode->i_sb, dummy.b_blocknr); + if (!bh) { + *errp = -EIO; + goto err; + } + if (buffer_new(&dummy)) { + J_ASSERT(create != 0); + J_ASSERT(handle != 0); + + /* + * Now that we do not always journal data, we should + * keep in mind whether this should always journal the + * new buffer as metadata. For now, regular file + * writes use ext3_get_block instead, so it's not a + * problem. + */ + lock_buffer(bh); + BUFFER_TRACE(bh, "call get_create_access"); + fatal = ext3_journal_get_create_access(handle, bh); + if (!fatal && !buffer_uptodate(bh)) { + memset(bh->b_data,0,inode->i_sb->s_blocksize); + set_buffer_uptodate(bh); + } + unlock_buffer(bh); + BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, bh); + if (!fatal) + fatal = err; + } else { + BUFFER_TRACE(bh, "not a new buffer"); + } + if (fatal) { + *errp = fatal; + brelse(bh); + bh = NULL; + } + return bh; + } +err: + return NULL; +} + +struct buffer_head *ext3_bread(handle_t *handle, struct inode *inode, + int block, int create, int *err) +{ + struct buffer_head * bh; + + bh = ext3_getblk(handle, inode, block, create, err); + if (!bh) + return bh; + if (buffer_uptodate(bh)) + return bh; + ll_rw_block(READ_META, 1, &bh); + wait_on_buffer(bh); + if (buffer_uptodate(bh)) + return bh; + put_bh(bh); + *err = -EIO; + return NULL; +} + +static int walk_page_buffers( handle_t *handle, + struct buffer_head *head, + unsigned from, + unsigned to, + int *partial, + int (*fn)( handle_t *handle, + struct buffer_head *bh)) +{ + struct buffer_head *bh; + unsigned block_start, block_end; + unsigned blocksize = head->b_size; + int err, ret = 0; + struct buffer_head *next; + + for ( bh = head, block_start = 0; + ret == 0 && (bh != head || !block_start); + block_start = block_end, bh = next) + { + next = bh->b_this_page; + block_end = block_start + blocksize; + if (block_end <= from || block_start >= to) { + if (partial && !buffer_uptodate(bh)) + *partial = 1; + continue; + } + err = (*fn)(handle, bh); + if (!ret) + ret = err; + } + return ret; +} + +/* + * To preserve ordering, it is essential that the hole instantiation and + * the data write be encapsulated in a single transaction. We cannot + * close off a transaction and start a new one between the ext3_get_block() + * and the commit_write(). So doing the journal_start at the start of + * prepare_write() is the right place. + * + * Also, this function can nest inside ext3_writepage() -> + * block_write_full_page(). In that case, we *know* that ext3_writepage() + * has generated enough buffer credits to do the whole page. So we won't + * block on the journal in that case, which is good, because the caller may + * be PF_MEMALLOC. + * + * By accident, ext3 can be reentered when a transaction is open via + * quota file writes. If we were to commit the transaction while thus + * reentered, there can be a deadlock - we would be holding a quota + * lock, and the commit would never complete if another thread had a + * transaction open and was blocking on the quota lock - a ranking + * violation. + * + * So what we do is to rely on the fact that journal_stop/journal_start + * will _not_ run commit under these circumstances because handle->h_ref + * is elevated. We'll still have enough credits for the tiny quotafile + * write. + */ +static int do_journal_get_write_access(handle_t *handle, + struct buffer_head *bh) +{ + if (!buffer_mapped(bh) || buffer_freed(bh)) + return 0; + return ext3_journal_get_write_access(handle, bh); +} + +static int ext3_prepare_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + struct inode *inode = page->mapping->host; + int ret, needed_blocks = ext3_writepage_trans_blocks(inode); + handle_t *handle; + int retries = 0; + +retry: + handle = ext3_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) + ret = nobh_prepare_write(page, from, to, ext3_get_block); + else + ret = block_prepare_write(page, from, to, ext3_get_block); + if (ret) + goto prepare_write_failed; + + if (ext3_should_journal_data(inode)) { + ret = walk_page_buffers(handle, page_buffers(page), + from, to, NULL, do_journal_get_write_access); + } +prepare_write_failed: + if (ret) + ext3_journal_stop(handle); + if (ret == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries)) + goto retry; +out: + return ret; +} + +int ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh) +{ + int err = journal_dirty_data(handle, bh); + if (err) + ext3_journal_abort_handle(__FUNCTION__, __FUNCTION__, + bh, handle,err); + return err; +} + +/* For commit_write() in data=journal mode */ +static int commit_write_fn(handle_t *handle, struct buffer_head *bh) +{ + if (!buffer_mapped(bh) || buffer_freed(bh)) + return 0; + set_buffer_uptodate(bh); + return ext3_journal_dirty_metadata(handle, bh); +} + +/* + * We need to pick up the new inode size which generic_commit_write gave us + * `file' can be NULL - eg, when called from page_symlink(). + * + * ext3 never places buffers on inode->i_mapping->private_list. metadata + * buffers are managed internally. + */ +static int ext3_ordered_commit_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + handle_t *handle = ext3_journal_current_handle(); + struct inode *inode = page->mapping->host; + int ret = 0, ret2; + + ret = walk_page_buffers(handle, page_buffers(page), + from, to, NULL, ext3_journal_dirty_data); + + if (ret == 0) { + /* + * generic_commit_write() will run mark_inode_dirty() if i_size + * changes. So let's piggyback the i_disksize mark_inode_dirty + * into that. + */ + loff_t new_i_size; + + new_i_size = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; + if (new_i_size > EXT3_I(inode)->i_disksize) + EXT3_I(inode)->i_disksize = new_i_size; + ret = generic_commit_write(file, page, from, to); + } + ret2 = ext3_journal_stop(handle); + if (!ret) + ret = ret2; + return ret; +} + +static int ext3_writeback_commit_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + handle_t *handle = ext3_journal_current_handle(); + struct inode *inode = page->mapping->host; + int ret = 0, ret2; + loff_t new_i_size; + + new_i_size = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; + if (new_i_size > EXT3_I(inode)->i_disksize) + EXT3_I(inode)->i_disksize = new_i_size; + + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) + ret = nobh_commit_write(file, page, from, to); + else + ret = generic_commit_write(file, page, from, to); + + ret2 = ext3_journal_stop(handle); + if (!ret) + ret = ret2; + return ret; +} + +static int ext3_journalled_commit_write(struct file *file, + struct page *page, unsigned from, unsigned to) +{ + handle_t *handle = ext3_journal_current_handle(); + struct inode *inode = page->mapping->host; + int ret = 0, ret2; + int partial = 0; + loff_t pos; + + /* + * Here we duplicate the generic_commit_write() functionality + */ + pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; + + ret = walk_page_buffers(handle, page_buffers(page), from, + to, &partial, commit_write_fn); + if (!partial) + SetPageUptodate(page); + if (pos > inode->i_size) + i_size_write(inode, pos); + EXT3_I(inode)->i_state |= EXT3_STATE_JDATA; + if (inode->i_size > EXT3_I(inode)->i_disksize) { + EXT3_I(inode)->i_disksize = inode->i_size; + ret2 = ext3_mark_inode_dirty(handle, inode); + if (!ret) + ret = ret2; + } + ret2 = ext3_journal_stop(handle); + if (!ret) + ret = ret2; + return ret; +} + +/* + * bmap() is special. It gets used by applications such as lilo and by + * the swapper to find the on-disk block of a specific piece of data. + * + * Naturally, this is dangerous if the block concerned is still in the + * journal. If somebody makes a swapfile on an ext3 data-journaling + * filesystem and enables swap, then they may get a nasty shock when the + * data getting swapped to that swapfile suddenly gets overwritten by + * the original zero's written out previously to the journal and + * awaiting writeback in the kernel's buffer cache. + * + * So, if we see any bmap calls here on a modified, data-journaled file, + * take extra steps to flush any blocks which might be in the cache. + */ +static sector_t ext3_bmap(struct address_space *mapping, sector_t block) +{ + struct inode *inode = mapping->host; + journal_t *journal; + int err; + + if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) { + /* + * This is a REALLY heavyweight approach, but the use of + * bmap on dirty files is expected to be extremely rare: + * only if we run lilo or swapon on a freshly made file + * do we expect this to happen. + * + * (bmap requires CAP_SYS_RAWIO so this does not + * represent an unprivileged user DOS attack --- we'd be + * in trouble if mortal users could trigger this path at + * will.) + * + * NB. EXT3_STATE_JDATA is not set on files other than + * regular files. If somebody wants to bmap a directory + * or symlink and gets confused because the buffer + * hasn't yet been flushed to disk, they deserve + * everything they get. + */ + + EXT3_I(inode)->i_state &= ~EXT3_STATE_JDATA; + journal = EXT3_JOURNAL(inode); + journal_lock_updates(journal); + err = journal_flush(journal); + journal_unlock_updates(journal); + + if (err) + return 0; + } + + return generic_block_bmap(mapping,block,ext3_get_block); +} + +static int bget_one(handle_t *handle, struct buffer_head *bh) +{ + get_bh(bh); + return 0; +} + +static int bput_one(handle_t *handle, struct buffer_head *bh) +{ + put_bh(bh); + return 0; +} + +static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) +{ + if (buffer_mapped(bh)) + return ext3_journal_dirty_data(handle, bh); + return 0; +} + +/* + * Note that we always start a transaction even if we're not journalling + * data. This is to preserve ordering: any hole instantiation within + * __block_write_full_page -> ext3_get_block() should be journalled + * along with the data so we don't crash and then get metadata which + * refers to old data. + * + * In all journalling modes block_write_full_page() will start the I/O. + * + * Problem: + * + * ext3_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() -> + * ext3_writepage() + * + * Similar for: + * + * ext3_file_write() -> generic_file_write() -> __alloc_pages() -> ... + * + * Same applies to ext3_get_block(). We will deadlock on various things like + * lock_journal and i_truncate_mutex. + * + * Setting PF_MEMALLOC here doesn't work - too many internal memory + * allocations fail. + * + * 16May01: If we're reentered then journal_current_handle() will be + * non-zero. We simply *return*. + * + * 1 July 2001: @@@ FIXME: + * In journalled data mode, a data buffer may be metadata against the + * current transaction. But the same file is part of a shared mapping + * and someone does a writepage() on it. + * + * We will move the buffer onto the async_data list, but *after* it has + * been dirtied. So there's a small window where we have dirty data on + * BJ_Metadata. + * + * Note that this only applies to the last partial page in the file. The + * bit which block_write_full_page() uses prepare/commit for. (That's + * broken code anyway: it's wrong for msync()). + * + * It's a rare case: affects the final partial page, for journalled data + * where the file is subject to bith write() and writepage() in the same + * transction. To fix it we'll need a custom block_write_full_page(). + * We'll probably need that anyway for journalling writepage() output. + * + * We don't honour synchronous mounts for writepage(). That would be + * disastrous. Any write() or metadata operation will sync the fs for + * us. + * + * AKPM2: if all the page's buffers are mapped to disk and !data=journal, + * we don't need to open a transaction here. + */ +static int ext3_ordered_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + struct buffer_head *page_bufs; + handle_t *handle = NULL; + int ret = 0; + int err; + + J_ASSERT(PageLocked(page)); + + /* + * We give up here if we're reentered, because it might be for a + * different filesystem. + */ + if (ext3_journal_current_handle()) + goto out_fail; + + handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); + + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_fail; + } + + if (!page_has_buffers(page)) { + create_empty_buffers(page, inode->i_sb->s_blocksize, + (1 << BH_Dirty)|(1 << BH_Uptodate)); + } + page_bufs = page_buffers(page); + walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, bget_one); + + ret = block_write_full_page(page, ext3_get_block, wbc); + + /* + * The page can become unlocked at any point now, and + * truncate can then come in and change things. So we + * can't touch *page from now on. But *page_bufs is + * safe due to elevated refcount. + */ + + /* + * And attach them to the current transaction. But only if + * block_write_full_page() succeeded. Otherwise they are unmapped, + * and generally junk. + */ + if (ret == 0) { + err = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, + NULL, journal_dirty_data_fn); + if (!ret) + ret = err; + } + walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, bput_one); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; + +out_fail: + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return ret; +} + +static int ext3_writeback_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + handle_t *handle = NULL; + int ret = 0; + int err; + + if (ext3_journal_current_handle()) + goto out_fail; + + handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_fail; + } + + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) + ret = nobh_writepage(page, ext3_get_block, wbc); + else + ret = block_write_full_page(page, ext3_get_block, wbc); + + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; + +out_fail: + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return ret; +} + +static int ext3_journalled_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + handle_t *handle = NULL; + int ret = 0; + int err; + + if (ext3_journal_current_handle()) + goto no_write; + + handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto no_write; + } + + if (!page_has_buffers(page) || PageChecked(page)) { + /* + * It's mmapped pagecache. Add buffers and journal it. There + * doesn't seem much point in redirtying the page here. + */ + ClearPageChecked(page); + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, + ext3_get_block); + if (ret != 0) { + ext3_journal_stop(handle); + goto out_unlock; + } + ret = walk_page_buffers(handle, page_buffers(page), 0, + PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); + + err = walk_page_buffers(handle, page_buffers(page), 0, + PAGE_CACHE_SIZE, NULL, commit_write_fn); + if (ret == 0) + ret = err; + EXT3_I(inode)->i_state |= EXT3_STATE_JDATA; + unlock_page(page); + } else { + /* + * It may be a page full of checkpoint-mode buffers. We don't + * really know unless we go poke around in the buffer_heads. + * But block_write_full_page will do the right thing. + */ + ret = block_write_full_page(page, ext3_get_block, wbc); + } + err = ext3_journal_stop(handle); + if (!ret) + ret = err; +out: + return ret; + +no_write: + redirty_page_for_writepage(wbc, page); +out_unlock: + unlock_page(page); + goto out; +} + +static int ext3_readpage(struct file *file, struct page *page) +{ + return mpage_readpage(page, ext3_get_block); +} + +static int +ext3_readpages(struct file *file, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) +{ + return mpage_readpages(mapping, pages, nr_pages, ext3_get_block); +} + +static void ext3_invalidatepage(struct page *page, unsigned long offset) +{ + journal_t *journal = EXT3_JOURNAL(page->mapping->host); + + /* + * If it's a full truncate we just forget about the pending dirtying + */ + if (offset == 0) + ClearPageChecked(page); + + journal_invalidatepage(journal, page, offset); +} + +static int ext3_releasepage(struct page *page, gfp_t wait) +{ + journal_t *journal = EXT3_JOURNAL(page->mapping->host); + + WARN_ON(PageChecked(page)); + if (!page_has_buffers(page)) + return 0; + return journal_try_to_free_buffers(journal, page, wait); +} + +/* + * If the O_DIRECT write will extend the file then add this inode to the + * orphan list. So recovery will truncate it back to the original size + * if the machine crashes during the write. + * + * If the O_DIRECT write is intantiating holes inside i_size and the machine + * crashes then stale disk data _may_ be exposed inside the file. + */ +static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, + const struct iovec *iov, loff_t offset, + unsigned long nr_segs) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + struct ext3_inode_info *ei = EXT3_I(inode); + handle_t *handle = NULL; + ssize_t ret; + int orphan = 0; + size_t count = iov_length(iov, nr_segs); + + if (rw == WRITE) { + loff_t final_size = offset + count; + + handle = ext3_journal_start(inode, DIO_CREDITS); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + if (final_size > inode->i_size) { + ret = ext3_orphan_add(handle, inode); + if (ret) + goto out_stop; + orphan = 1; + ei->i_disksize = inode->i_size; + } + } + + ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext3_get_block, NULL); + + /* + * Reacquire the handle: ext3_get_block() can restart the transaction + */ + handle = journal_current_handle(); + +out_stop: + if (handle) { + int err; + + if (orphan && inode->i_nlink) + ext3_orphan_del(handle, inode); + if (orphan && ret > 0) { + loff_t end = offset + ret; + if (end > inode->i_size) { + ei->i_disksize = end; + i_size_write(inode, end); + /* + * We're going to return a positive `ret' + * here due to non-zero-length I/O, so there's + * no way of reporting error returns from + * ext3_mark_inode_dirty() to userspace. So + * ignore it. + */ + ext3_mark_inode_dirty(handle, inode); + } + } + err = ext3_journal_stop(handle); + if (ret == 0) + ret = err; + } +out: + return ret; +} + +/* + * Pages can be marked dirty completely asynchronously from ext3's journalling + * activity. By filemap_sync_pte(), try_to_unmap_one(), etc. We cannot do + * much here because ->set_page_dirty is called under VFS locks. The page is + * not necessarily locked. + * + * We cannot just dirty the page and leave attached buffers clean, because the + * buffers' dirty state is "definitive". We cannot just set the buffers dirty + * or jbddirty because all the journalling code will explode. + * + * So what we do is to mark the page "pending dirty" and next time writepage + * is called, propagate that into the buffers appropriately. + */ +static int ext3_journalled_set_page_dirty(struct page *page) +{ + SetPageChecked(page); + return __set_page_dirty_nobuffers(page); +} + +static const struct address_space_operations ext3_ordered_aops = { + .readpage = ext3_readpage, + .readpages = ext3_readpages, + .writepage = ext3_ordered_writepage, + .sync_page = block_sync_page, + .prepare_write = ext3_prepare_write, + .commit_write = ext3_ordered_commit_write, + .bmap = ext3_bmap, + .invalidatepage = ext3_invalidatepage, + .releasepage = ext3_releasepage, + .direct_IO = ext3_direct_IO, + .migratepage = buffer_migrate_page, +}; + +static const struct address_space_operations ext3_writeback_aops = { + .readpage = ext3_readpage, + .readpages = ext3_readpages, + .writepage = ext3_writeback_writepage, + .sync_page = block_sync_page, + .prepare_write = ext3_prepare_write, + .commit_write = ext3_writeback_commit_write, + .bmap = ext3_bmap, + .invalidatepage = ext3_invalidatepage, + .releasepage = ext3_releasepage, + .direct_IO = ext3_direct_IO, + .migratepage = buffer_migrate_page, +}; + +static const struct address_space_operations ext3_journalled_aops = { + .readpage = ext3_readpage, + .readpages = ext3_readpages, + .writepage = ext3_journalled_writepage, + .sync_page = block_sync_page, + .prepare_write = ext3_prepare_write, + .commit_write = ext3_journalled_commit_write, + .set_page_dirty = ext3_journalled_set_page_dirty, + .bmap = ext3_bmap, + .invalidatepage = ext3_invalidatepage, + .releasepage = ext3_releasepage, +}; + +void ext3_set_aops(struct inode *inode) +{ + if (ext3_should_order_data(inode)) + inode->i_mapping->a_ops = &ext3_ordered_aops; + else if (ext3_should_writeback_data(inode)) + inode->i_mapping->a_ops = &ext3_writeback_aops; + else + inode->i_mapping->a_ops = &ext3_journalled_aops; +} + +/* + * ext3_block_truncate_page() zeroes out a mapping from file offset `from' + * up to the end of the block which corresponds to `from'. + * This required during truncate. We need to physically zero the tail end + * of that block so it doesn't yield old data if the file is later grown. + */ +static int ext3_block_truncate_page(handle_t *handle, struct page *page, + struct address_space *mapping, loff_t from) +{ + ext3_fsblk_t index = from >> PAGE_CACHE_SHIFT; + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned blocksize, iblock, length, pos; + struct inode *inode = mapping->host; + struct buffer_head *bh; + int err = 0; + void *kaddr; + + blocksize = inode->i_sb->s_blocksize; + length = blocksize - (offset & (blocksize - 1)); + iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); + + /* + * For "nobh" option, we can only work if we don't need to + * read-in the page - otherwise we create buffers to do the IO. + */ + if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && + ext3_should_writeback_data(inode) && PageUptodate(page)) { + kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + offset, 0, length); + flush_dcache_page(page); + kunmap_atomic(kaddr, KM_USER0); + set_page_dirty(page); + goto unlock; + } + + if (!page_has_buffers(page)) + create_empty_buffers(page, blocksize, 0); + + /* Find the buffer that contains "offset" */ + bh = page_buffers(page); + pos = blocksize; + while (offset >= pos) { + bh = bh->b_this_page; + iblock++; + pos += blocksize; + } + + err = 0; + if (buffer_freed(bh)) { + BUFFER_TRACE(bh, "freed: skip"); + goto unlock; + } + + if (!buffer_mapped(bh)) { + BUFFER_TRACE(bh, "unmapped"); + ext3_get_block(inode, iblock, bh, 0); + /* unmapped? It's a hole - nothing to do */ + if (!buffer_mapped(bh)) { + BUFFER_TRACE(bh, "still unmapped"); + goto unlock; + } + } + + /* Ok, it's mapped. Make sure it's up-to-date */ + if (PageUptodate(page)) + set_buffer_uptodate(bh); + + if (!buffer_uptodate(bh)) { + err = -EIO; + ll_rw_block(READ, 1, &bh); + wait_on_buffer(bh); + /* Uhhuh. Read error. Complain and punt. */ + if (!buffer_uptodate(bh)) + goto unlock; + } + + if (ext3_should_journal_data(inode)) { + BUFFER_TRACE(bh, "get write access"); + err = ext3_journal_get_write_access(handle, bh); + if (err) + goto unlock; + } + + kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + offset, 0, length); + flush_dcache_page(page); + kunmap_atomic(kaddr, KM_USER0); + + BUFFER_TRACE(bh, "zeroed end of block"); + + err = 0; + if (ext3_should_journal_data(inode)) { + err = ext3_journal_dirty_metadata(handle, bh); + } else { + if (ext3_should_order_data(inode)) + err = ext3_journal_dirty_data(handle, bh); + mark_buffer_dirty(bh); + } + +unlock: + unlock_page(page); + page_cache_release(page); + return err; +} + +/* + * Probably it should be a library function... search for first non-zero word + * or memcmp with zero_page, whatever is better for particular architecture. + * Linus? + */ +static inline int all_zeroes(__le32 *p, __le32 *q) +{ + while (p < q) + if (*p++) + return 0; + return 1; +} + +/** + * ext3_find_shared - find the indirect blocks for partial truncation. + * @inode: inode in question + * @depth: depth of the affected branch + * @offsets: offsets of pointers in that branch (see ext3_block_to_path) + * @chain: place to store the pointers to partial indirect blocks + * @top: place to the (detached) top of branch + * + * This is a helper function used by ext3_truncate(). + * + * When we do truncate() we may have to clean the ends of several + * indirect blocks but leave the blocks themselves alive. Block is + * partially truncated if some data below the new i_size is refered + * from it (and it is on the path to the first completely truncated + * data block, indeed). We have to free the top of that path along + * with everything to the right of the path. Since no allocation + * past the truncation point is possible until ext3_truncate() + * finishes, we may safely do the latter, but top of branch may + * require special attention - pageout below the truncation point + * might try to populate it. + * + * We atomically detach the top of branch from the tree, store the + * block number of its root in *@top, pointers to buffer_heads of + * partially truncated blocks - in @chain[].bh and pointers to + * their last elements that should not be removed - in + * @chain[].p. Return value is the pointer to last filled element + * of @chain. + * + * The work left to caller to do the actual freeing of subtrees: + * a) free the subtree starting from *@top + * b) free the subtrees whose roots are stored in + * (@chain[i].p+1 .. end of @chain[i].bh->b_data) + * c) free the subtrees growing from the inode past the @chain[0]. + * (no partially truncated stuff there). */ + +static Indirect *ext3_find_shared(struct inode *inode, int depth, + int offsets[4], Indirect chain[4], __le32 *top) +{ + Indirect *partial, *p; + int k, err; + + *top = 0; + /* Make k index the deepest non-null offest + 1 */ + for (k = depth; k > 1 && !offsets[k-1]; k--) + ; + partial = ext3_get_branch(inode, k, offsets, chain, &err); + /* Writer: pointers */ + if (!partial) + partial = chain + k-1; + /* + * If the branch acquired continuation since we've looked at it - + * fine, it should all survive and (new) top doesn't belong to us. + */ + if (!partial->key && *partial->p) + /* Writer: end */ + goto no_top; + for (p=partial; p>chain && all_zeroes((__le32*)p->bh->b_data,p->p); p--) + ; + /* + * OK, we've found the last block that must survive. The rest of our + * branch should be detached before unlocking. However, if that rest + * of branch is all ours and does not grow immediately from the inode + * it's easier to cheat and just decrement partial->p. + */ + if (p == chain + k - 1 && p > chain) { + p->p--; + } else { + *top = *p->p; + /* Nope, don't do this in ext3. Must leave the tree intact */ +#if 0 + *p->p = 0; +#endif + } + /* Writer: end */ + + while(partial > p) { + brelse(partial->bh); + partial--; + } +no_top: + return partial; +} + +/* + * Zero a number of block pointers in either an inode or an indirect block. + * If we restart the transaction we must again get write access to the + * indirect block for further modification. + * + * We release `count' blocks on disk, but (last - first) may be greater + * than `count' because there can be holes in there. + */ +static void ext3_clear_blocks(handle_t *handle, struct inode *inode, + struct buffer_head *bh, ext3_fsblk_t block_to_free, + unsigned long count, __le32 *first, __le32 *last) +{ + __le32 *p; + if (try_to_extend_transaction(handle, inode)) { + if (bh) { + BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, bh); + } + ext3_mark_inode_dirty(handle, inode); + ext3_journal_test_restart(handle, inode); + if (bh) { + BUFFER_TRACE(bh, "retaking write access"); + ext3_journal_get_write_access(handle, bh); + } + } + + /* + * Any buffers which are on the journal will be in memory. We find + * them on the hash table so journal_revoke() will run journal_forget() + * on them. We've already detached each block from the file, so + * bforget() in journal_forget() should be safe. + * + * AKPM: turn on bforget in journal_forget()!!! + */ + for (p = first; p < last; p++) { + u32 nr = le32_to_cpu(*p); + if (nr) { + struct buffer_head *bh; + + *p = 0; + bh = sb_find_get_block(inode->i_sb, nr); + ext3_forget(handle, 0, inode, bh, nr); + } + } + + ext3_free_blocks(handle, inode, block_to_free, count); +} + +/** + * ext3_free_data - free a list of data blocks + * @handle: handle for this transaction + * @inode: inode we are dealing with + * @this_bh: indirect buffer_head which contains *@first and *@last + * @first: array of block numbers + * @last: points immediately past the end of array + * + * We are freeing all blocks refered from that array (numbers are stored as + * little-endian 32-bit) and updating @inode->i_blocks appropriately. + * + * We accumulate contiguous runs of blocks to free. Conveniently, if these + * blocks are contiguous then releasing them at one time will only affect one + * or two bitmap blocks (+ group descriptor(s) and superblock) and we won't + * actually use a lot of journal space. + * + * @this_bh will be %NULL if @first and @last point into the inode's direct + * block pointers. + */ +static void ext3_free_data(handle_t *handle, struct inode *inode, + struct buffer_head *this_bh, + __le32 *first, __le32 *last) +{ + ext3_fsblk_t block_to_free = 0; /* Starting block # of a run */ + unsigned long count = 0; /* Number of blocks in the run */ + __le32 *block_to_free_p = NULL; /* Pointer into inode/ind + corresponding to + block_to_free */ + ext3_fsblk_t nr; /* Current block # */ + __le32 *p; /* Pointer into inode/ind + for current block */ + int err; + + if (this_bh) { /* For indirect block */ + BUFFER_TRACE(this_bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, this_bh); + /* Important: if we can't update the indirect pointers + * to the blocks, we can't free them. */ + if (err) + return; + } + + for (p = first; p < last; p++) { + nr = le32_to_cpu(*p); + if (nr) { + /* accumulate blocks to free if they're contiguous */ + if (count == 0) { + block_to_free = nr; + block_to_free_p = p; + count = 1; + } else if (nr == block_to_free + count) { + count++; + } else { + ext3_clear_blocks(handle, inode, this_bh, + block_to_free, + count, block_to_free_p, p); + block_to_free = nr; + block_to_free_p = p; + count = 1; + } + } + } + + if (count > 0) + ext3_clear_blocks(handle, inode, this_bh, block_to_free, + count, block_to_free_p, p); + + if (this_bh) { + BUFFER_TRACE(this_bh, "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, this_bh); + } +} + +/** + * ext3_free_branches - free an array of branches + * @handle: JBD handle for this transaction + * @inode: inode we are dealing with + * @parent_bh: the buffer_head which contains *@first and *@last + * @first: array of block numbers + * @last: pointer immediately past the end of array + * @depth: depth of the branches to free + * + * We are freeing all blocks refered from these branches (numbers are + * stored as little-endian 32-bit) and updating @inode->i_blocks + * appropriately. + */ +static void ext3_free_branches(handle_t *handle, struct inode *inode, + struct buffer_head *parent_bh, + __le32 *first, __le32 *last, int depth) +{ + ext3_fsblk_t nr; + __le32 *p; + + if (is_handle_aborted(handle)) + return; + + if (depth--) { + struct buffer_head *bh; + int addr_per_block = EXT3_ADDR_PER_BLOCK(inode->i_sb); + p = last; + while (--p >= first) { + nr = le32_to_cpu(*p); + if (!nr) + continue; /* A hole */ + + /* Go read the buffer for the next level down */ + bh = sb_bread(inode->i_sb, nr); + + /* + * A read failure? Report error and clear slot + * (should be rare). + */ + if (!bh) { + ext3_error(inode->i_sb, "ext3_free_branches", + "Read failure, inode=%lu, block="E3FSBLK, + inode->i_ino, nr); + continue; + } + + /* This zaps the entire block. Bottom up. */ + BUFFER_TRACE(bh, "free child branches"); + ext3_free_branches(handle, inode, bh, + (__le32*)bh->b_data, + (__le32*)bh->b_data + addr_per_block, + depth); + + /* + * We've probably journalled the indirect block several + * times during the truncate. But it's no longer + * needed and we now drop it from the transaction via + * journal_revoke(). + * + * That's easy if it's exclusively part of this + * transaction. But if it's part of the committing + * transaction then journal_forget() will simply + * brelse() it. That means that if the underlying + * block is reallocated in ext3_get_block(), + * unmap_underlying_metadata() will find this block + * and will try to get rid of it. damn, damn. + * + * If this block has already been committed to the + * journal, a revoke record will be written. And + * revoke records must be emitted *before* clearing + * this block's bit in the bitmaps. + */ + ext3_forget(handle, 1, inode, bh, bh->b_blocknr); + + /* + * Everything below this this pointer has been + * released. Now let this top-of-subtree go. + * + * We want the freeing of this indirect block to be + * atomic in the journal with the updating of the + * bitmap block which owns it. So make some room in + * the journal. + * + * We zero the parent pointer *after* freeing its + * pointee in the bitmaps, so if extend_transaction() + * for some reason fails to put the bitmap changes and + * the release into the same transaction, recovery + * will merely complain about releasing a free block, + * rather than leaking blocks. + */ + if (is_handle_aborted(handle)) + return; + if (try_to_extend_transaction(handle, inode)) { + ext3_mark_inode_dirty(handle, inode); + ext3_journal_test_restart(handle, inode); + } + + ext3_free_blocks(handle, inode, nr, 1); + + if (parent_bh) { + /* + * The block which we have just freed is + * pointed to by an indirect block: journal it + */ + BUFFER_TRACE(parent_bh, "get_write_access"); + if (!ext3_journal_get_write_access(handle, + parent_bh)){ + *p = 0; + BUFFER_TRACE(parent_bh, + "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, + parent_bh); + } + } + } + } else { + /* We have reached the bottom of the tree. */ + BUFFER_TRACE(parent_bh, "free data blocks"); + ext3_free_data(handle, inode, parent_bh, first, last); + } +} + +/* + * ext3_truncate() + * + * We block out ext3_get_block() block instantiations across the entire + * transaction, and VFS/VM ensures that ext3_truncate() cannot run + * simultaneously on behalf of the same inode. + * + * As we work through the truncate and commmit bits of it to the journal there + * is one core, guiding principle: the file's tree must always be consistent on + * disk. We must be able to restart the truncate after a crash. + * + * The file's tree may be transiently inconsistent in memory (although it + * probably isn't), but whenever we close off and commit a journal transaction, + * the contents of (the filesystem + the journal) must be consistent and + * restartable. It's pretty simple, really: bottom up, right to left (although + * left-to-right works OK too). + * + * Note that at recovery time, journal replay occurs *before* the restart of + * truncate against the orphan inode list. + * + * The committed inode has the new, desired i_size (which is the same as + * i_disksize in this case). After a crash, ext3_orphan_cleanup() will see + * that this inode's truncate did not complete and it will again call + * ext3_truncate() to have another go. So there will be instantiated blocks + * to the right of the truncation point in a crashed ext3 filesystem. But + * that's fine - as long as they are linked from the inode, the post-crash + * ext3_truncate() run will find them and release them. + */ +void ext3_truncate(struct inode *inode) +{ + handle_t *handle; + struct ext3_inode_info *ei = EXT3_I(inode); + __le32 *i_data = ei->i_data; + int addr_per_block = EXT3_ADDR_PER_BLOCK(inode->i_sb); + struct address_space *mapping = inode->i_mapping; + int offsets[4]; + Indirect chain[4]; + Indirect *partial; + __le32 nr = 0; + int n; + long last_block; + unsigned blocksize = inode->i_sb->s_blocksize; + struct page *page; + + if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode))) + return; + if (ext3_inode_is_fast_symlink(inode)) + return; + if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) + return; + + /* + * We have to lock the EOF page here, because lock_page() nests + * outside journal_start(). + */ + if ((inode->i_size & (blocksize - 1)) == 0) { + /* Block boundary? Nothing to do */ + page = NULL; + } else { + page = grab_cache_page(mapping, + inode->i_size >> PAGE_CACHE_SHIFT); + if (!page) + return; + } + + handle = start_transaction(inode); + if (IS_ERR(handle)) { + if (page) { + clear_highpage(page); + flush_dcache_page(page); + unlock_page(page); + page_cache_release(page); + } + return; /* AKPM: return what? */ + } + + last_block = (inode->i_size + blocksize-1) + >> EXT3_BLOCK_SIZE_BITS(inode->i_sb); + + if (page) + ext3_block_truncate_page(handle, page, mapping, inode->i_size); + + n = ext3_block_to_path(inode, last_block, offsets, NULL); + if (n == 0) + goto out_stop; /* error */ + + /* + * OK. This truncate is going to happen. We add the inode to the + * orphan list, so that if this truncate spans multiple transactions, + * and we crash, we will resume the truncate when the filesystem + * recovers. It also marks the inode dirty, to catch the new size. + * + * Implication: the file must always be in a sane, consistent + * truncatable state while each transaction commits. + */ + if (ext3_orphan_add(handle, inode)) + goto out_stop; + + /* + * The orphan list entry will now protect us from any crash which + * occurs before the truncate completes, so it is now safe to propagate + * the new, shorter inode size (held for now in i_size) into the + * on-disk inode. We do this via i_disksize, which is the value which + * ext3 *really* writes onto the disk inode. + */ + ei->i_disksize = inode->i_size; + + /* + * From here we block out all ext3_get_block() callers who want to + * modify the block allocation tree. + */ + mutex_lock(&ei->truncate_mutex); + + if (n == 1) { /* direct blocks */ + ext3_free_data(handle, inode, NULL, i_data+offsets[0], + i_data + EXT3_NDIR_BLOCKS); + goto do_indirects; + } + + partial = ext3_find_shared(inode, n, offsets, chain, &nr); + /* Kill the top of shared branch (not detached) */ + if (nr) { + if (partial == chain) { + /* Shared branch grows from the inode */ + ext3_free_branches(handle, inode, NULL, + &nr, &nr+1, (chain+n-1) - partial); + *partial->p = 0; + /* + * We mark the inode dirty prior to restart, + * and prior to stop. No need for it here. + */ + } else { + /* Shared branch grows from an indirect block */ + BUFFER_TRACE(partial->bh, "get_write_access"); + ext3_free_branches(handle, inode, partial->bh, + partial->p, + partial->p+1, (chain+n-1) - partial); + } + } + /* Clear the ends of indirect blocks on the shared branch */ + while (partial > chain) { + ext3_free_branches(handle, inode, partial->bh, partial->p + 1, + (__le32*)partial->bh->b_data+addr_per_block, + (chain+n-1) - partial); + BUFFER_TRACE(partial->bh, "call brelse"); + brelse (partial->bh); + partial--; + } +do_indirects: + /* Kill the remaining (whole) subtrees */ + switch (offsets[0]) { + default: + nr = i_data[EXT3_IND_BLOCK]; + if (nr) { + ext3_free_branches(handle, inode, NULL, &nr, &nr+1, 1); + i_data[EXT3_IND_BLOCK] = 0; + } + case EXT3_IND_BLOCK: + nr = i_data[EXT3_DIND_BLOCK]; + if (nr) { + ext3_free_branches(handle, inode, NULL, &nr, &nr+1, 2); + i_data[EXT3_DIND_BLOCK] = 0; + } + case EXT3_DIND_BLOCK: + nr = i_data[EXT3_TIND_BLOCK]; + if (nr) { + ext3_free_branches(handle, inode, NULL, &nr, &nr+1, 3); + i_data[EXT3_TIND_BLOCK] = 0; + } + case EXT3_TIND_BLOCK: + ; + } + + ext3_discard_reservation(inode); + + mutex_unlock(&ei->truncate_mutex); + inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; + ext3_mark_inode_dirty(handle, inode); + + /* + * In a multi-transaction truncate, we only make the final transaction + * synchronous + */ + if (IS_SYNC(inode)) + handle->h_sync = 1; +out_stop: + /* + * If this was a simple ftruncate(), and the file will remain alive + * then we need to clear up the orphan record which we created above. + * However, if this was a real unlink then we were called by + * ext3_delete_inode(), and we allow that function to clean up the + * orphan info for us. + */ + if (inode->i_nlink) + ext3_orphan_del(handle, inode); + + ext3_journal_stop(handle); +} + +static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, + unsigned long ino, struct ext3_iloc *iloc) +{ + unsigned long desc, group_desc, block_group; + unsigned long offset; + ext3_fsblk_t block; + struct buffer_head *bh; + struct ext3_group_desc * gdp; + + if (!ext3_valid_inum(sb, ino)) { + /* + * This error is already checked for in namei.c unless we are + * looking at an NFS filehandle, in which case no error + * report is needed + */ + return 0; + } + + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); + if (block_group >= EXT3_SB(sb)->s_groups_count) { + ext3_error(sb,"ext3_get_inode_block","group >= groups count"); + return 0; + } + smp_rmb(); + group_desc = block_group >> EXT3_DESC_PER_BLOCK_BITS(sb); + desc = block_group & (EXT3_DESC_PER_BLOCK(sb) - 1); + bh = EXT3_SB(sb)->s_group_desc[group_desc]; + if (!bh) { + ext3_error (sb, "ext3_get_inode_block", + "Descriptor not loaded"); + return 0; + } + + gdp = (struct ext3_group_desc *)bh->b_data; + /* + * Figure out the offset within the block group inode table + */ + offset = ((ino - 1) % EXT3_INODES_PER_GROUP(sb)) * + EXT3_INODE_SIZE(sb); + block = le32_to_cpu(gdp[desc].bg_inode_table) + + (offset >> EXT3_BLOCK_SIZE_BITS(sb)); + + iloc->block_group = block_group; + iloc->offset = offset & (EXT3_BLOCK_SIZE(sb) - 1); + return block; +} + +/* + * ext3_get_inode_loc returns with an extra refcount against the inode's + * underlying buffer_head on success. If 'in_mem' is true, we have all + * data in memory that is needed to recreate the on-disk version of this + * inode. + */ +static int __ext3_get_inode_loc(struct inode *inode, + struct ext3_iloc *iloc, int in_mem) +{ + ext3_fsblk_t block; + struct buffer_head *bh; + + block = ext3_get_inode_block(inode->i_sb, inode->i_ino, iloc); + if (!block) + return -EIO; + + bh = sb_getblk(inode->i_sb, block); + if (!bh) { + ext3_error (inode->i_sb, "ext3_get_inode_loc", + "unable to read inode block - " + "inode=%lu, block="E3FSBLK, + inode->i_ino, block); + return -EIO; + } + if (!buffer_uptodate(bh)) { + lock_buffer(bh); + if (buffer_uptodate(bh)) { + /* someone brought it uptodate while we waited */ + unlock_buffer(bh); + goto has_buffer; + } + + /* + * If we have all information of the inode in memory and this + * is the only valid inode in the block, we need not read the + * block. + */ + if (in_mem) { + struct buffer_head *bitmap_bh; + struct ext3_group_desc *desc; + int inodes_per_buffer; + int inode_offset, i; + int block_group; + int start; + + block_group = (inode->i_ino - 1) / + EXT3_INODES_PER_GROUP(inode->i_sb); + inodes_per_buffer = bh->b_size / + EXT3_INODE_SIZE(inode->i_sb); + inode_offset = ((inode->i_ino - 1) % + EXT3_INODES_PER_GROUP(inode->i_sb)); + start = inode_offset & ~(inodes_per_buffer - 1); + + /* Is the inode bitmap in cache? */ + desc = ext3_get_group_desc(inode->i_sb, + block_group, NULL); + if (!desc) + goto make_io; + + bitmap_bh = sb_getblk(inode->i_sb, + le32_to_cpu(desc->bg_inode_bitmap)); + if (!bitmap_bh) + goto make_io; + + /* + * If the inode bitmap isn't in cache then the + * optimisation may end up performing two reads instead + * of one, so skip it. + */ + if (!buffer_uptodate(bitmap_bh)) { + brelse(bitmap_bh); + goto make_io; + } + for (i = start; i < start + inodes_per_buffer; i++) { + if (i == inode_offset) + continue; + if (ext3_test_bit(i, bitmap_bh->b_data)) + break; + } + brelse(bitmap_bh); + if (i == start + inodes_per_buffer) { + /* all other inodes are free, so skip I/O */ + memset(bh->b_data, 0, bh->b_size); + set_buffer_uptodate(bh); + unlock_buffer(bh); + goto has_buffer; + } + } + +make_io: + /* + * There are other valid inodes in the buffer, this inode + * has in-inode xattrs, or we don't have this inode in memory. + * Read the block from disk. + */ + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ_META, bh); + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) { + ext3_error(inode->i_sb, "ext3_get_inode_loc", + "unable to read inode block - " + "inode=%lu, block="E3FSBLK, + inode->i_ino, block); + brelse(bh); + return -EIO; + } + } +has_buffer: + iloc->bh = bh; + return 0; +} + +int ext3_get_inode_loc(struct inode *inode, struct ext3_iloc *iloc) +{ + /* We have all inode data except xattrs in memory here. */ + return __ext3_get_inode_loc(inode, iloc, + !(EXT3_I(inode)->i_state & EXT3_STATE_XATTR)); +} + +void ext3_set_inode_flags(struct inode *inode) +{ + unsigned int flags = EXT3_I(inode)->i_flags; + + inode->i_flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC); + if (flags & EXT3_SYNC_FL) + inode->i_flags |= S_SYNC; + if (flags & EXT3_APPEND_FL) + inode->i_flags |= S_APPEND; + if (flags & EXT3_IMMUTABLE_FL) + inode->i_flags |= S_IMMUTABLE; + if (flags & EXT3_NOATIME_FL) + inode->i_flags |= S_NOATIME; + if (flags & EXT3_DIRSYNC_FL) + inode->i_flags |= S_DIRSYNC; +} + +void ext3_read_inode(struct inode * inode) +{ + struct ext3_iloc iloc; + struct ext3_inode *raw_inode; + struct ext3_inode_info *ei = EXT3_I(inode); + struct buffer_head *bh; + int block; + +#ifdef CONFIG_EXT3_FS_POSIX_ACL + ei->i_acl = EXT3_ACL_NOT_CACHED; + ei->i_default_acl = EXT3_ACL_NOT_CACHED; +#endif + ei->i_block_alloc_info = NULL; + + if (__ext3_get_inode_loc(inode, &iloc, 0)) + goto bad_inode; + bh = iloc.bh; + raw_inode = ext3_raw_inode(&iloc); + inode->i_mode = le16_to_cpu(raw_inode->i_mode); + inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); + inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low); + if(!(test_opt (inode->i_sb, NO_UID32))) { + inode->i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16; + inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; + } + inode->i_nlink = le16_to_cpu(raw_inode->i_links_count); + inode->i_size = le32_to_cpu(raw_inode->i_size); + inode->i_atime.tv_sec = le32_to_cpu(raw_inode->i_atime); + inode->i_ctime.tv_sec = le32_to_cpu(raw_inode->i_ctime); + inode->i_mtime.tv_sec = le32_to_cpu(raw_inode->i_mtime); + inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_mtime.tv_nsec = 0; + + ei->i_state = 0; + ei->i_dir_start_lookup = 0; + ei->i_dtime = le32_to_cpu(raw_inode->i_dtime); + /* We now have enough fields to check if the inode was active or not. + * This is needed because nfsd might try to access dead inodes + * the test is that same one that e2fsck uses + * NeilBrown 1999oct15 + */ + if (inode->i_nlink == 0) { + if (inode->i_mode == 0 || + !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ORPHAN_FS)) { + /* this inode is deleted */ + brelse (bh); + goto bad_inode; + } + /* The only unlinked inodes we let through here have + * valid i_mode and are being read by the orphan + * recovery code: that's fine, we're about to complete + * the process of deleting those. */ + } + inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); + ei->i_flags = le32_to_cpu(raw_inode->i_flags); +#ifdef EXT3_FRAGMENTS + ei->i_faddr = le32_to_cpu(raw_inode->i_faddr); + ei->i_frag_no = raw_inode->i_frag; + ei->i_frag_size = raw_inode->i_fsize; +#endif + ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl); + if (!S_ISREG(inode->i_mode)) { + ei->i_dir_acl = le32_to_cpu(raw_inode->i_dir_acl); + } else { + inode->i_size |= + ((__u64)le32_to_cpu(raw_inode->i_size_high)) << 32; + } + ei->i_disksize = inode->i_size; + inode->i_generation = le32_to_cpu(raw_inode->i_generation); + ei->i_block_group = iloc.block_group; + /* + * NOTE! The in-memory inode i_data array is in little-endian order + * even on big-endian machines: we do NOT byteswap the block numbers! + */ + for (block = 0; block < EXT3_N_BLOCKS; block++) + ei->i_data[block] = raw_inode->i_block[block]; + INIT_LIST_HEAD(&ei->i_orphan); + + if (inode->i_ino >= EXT3_FIRST_INO(inode->i_sb) + 1 && + EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) { + /* + * When mke2fs creates big inodes it does not zero out + * the unused bytes above EXT3_GOOD_OLD_INODE_SIZE, + * so ignore those first few inodes. + */ + ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); + if (EXT3_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > + EXT3_INODE_SIZE(inode->i_sb)) + goto bad_inode; + if (ei->i_extra_isize == 0) { + /* The extra space is currently unused. Use it. */ + ei->i_extra_isize = sizeof(struct ext3_inode) - + EXT3_GOOD_OLD_INODE_SIZE; + } else { + __le32 *magic = (void *)raw_inode + + EXT3_GOOD_OLD_INODE_SIZE + + ei->i_extra_isize; + if (*magic == cpu_to_le32(EXT3_XATTR_MAGIC)) + ei->i_state |= EXT3_STATE_XATTR; + } + } else + ei->i_extra_isize = 0; + + if (S_ISREG(inode->i_mode)) { + inode->i_op = &ext3_file_inode_operations; + inode->i_fop = &ext3_file_operations; + ext3_set_aops(inode); + } else if (S_ISDIR(inode->i_mode)) { + inode->i_op = &ext3_dir_inode_operations; + inode->i_fop = &ext3_dir_operations; + } else if (S_ISLNK(inode->i_mode)) { + if (ext3_inode_is_fast_symlink(inode)) + inode->i_op = &ext3_fast_symlink_inode_operations; + else { + inode->i_op = &ext3_symlink_inode_operations; + ext3_set_aops(inode); + } + } else { + inode->i_op = &ext3_special_inode_operations; + if (raw_inode->i_block[0]) + init_special_inode(inode, inode->i_mode, + old_decode_dev(le32_to_cpu(raw_inode->i_block[0]))); + else + init_special_inode(inode, inode->i_mode, + new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); + } + brelse (iloc.bh); + ext3_set_inode_flags(inode); + return; + +bad_inode: + make_bad_inode(inode); + return; +} + +/* + * Post the struct inode info into an on-disk inode location in the + * buffer-cache. This gobbles the caller's reference to the + * buffer_head in the inode location struct. + * + * The caller must have write access to iloc->bh. + */ +static int ext3_do_update_inode(handle_t *handle, + struct inode *inode, + struct ext3_iloc *iloc) +{ + struct ext3_inode *raw_inode = ext3_raw_inode(iloc); + struct ext3_inode_info *ei = EXT3_I(inode); + struct buffer_head *bh = iloc->bh; + int err = 0, rc, block; + + /* For fields not not tracking in the in-memory inode, + * initialise them to zero for new inodes. */ + if (ei->i_state & EXT3_STATE_NEW) + memset(raw_inode, 0, EXT3_SB(inode->i_sb)->s_inode_size); + + raw_inode->i_mode = cpu_to_le16(inode->i_mode); + if(!(test_opt(inode->i_sb, NO_UID32))) { + raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid)); + raw_inode->i_gid_low = cpu_to_le16(low_16_bits(inode->i_gid)); +/* + * Fix up interoperability with old kernels. Otherwise, old inodes get + * re-used with the upper 16 bits of the uid/gid intact + */ + if(!ei->i_dtime) { + raw_inode->i_uid_high = + cpu_to_le16(high_16_bits(inode->i_uid)); + raw_inode->i_gid_high = + cpu_to_le16(high_16_bits(inode->i_gid)); + } else { + raw_inode->i_uid_high = 0; + raw_inode->i_gid_high = 0; + } + } else { + raw_inode->i_uid_low = + cpu_to_le16(fs_high2lowuid(inode->i_uid)); + raw_inode->i_gid_low = + cpu_to_le16(fs_high2lowgid(inode->i_gid)); + raw_inode->i_uid_high = 0; + raw_inode->i_gid_high = 0; + } + raw_inode->i_links_count = cpu_to_le16(inode->i_nlink); + raw_inode->i_size = cpu_to_le32(ei->i_disksize); + raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec); + raw_inode->i_ctime = cpu_to_le32(inode->i_ctime.tv_sec); + raw_inode->i_mtime = cpu_to_le32(inode->i_mtime.tv_sec); + raw_inode->i_blocks = cpu_to_le32(inode->i_blocks); + raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); + raw_inode->i_flags = cpu_to_le32(ei->i_flags); +#ifdef EXT3_FRAGMENTS + raw_inode->i_faddr = cpu_to_le32(ei->i_faddr); + raw_inode->i_frag = ei->i_frag_no; + raw_inode->i_fsize = ei->i_frag_size; +#endif + raw_inode->i_file_acl = cpu_to_le32(ei->i_file_acl); + if (!S_ISREG(inode->i_mode)) { + raw_inode->i_dir_acl = cpu_to_le32(ei->i_dir_acl); + } else { + raw_inode->i_size_high = + cpu_to_le32(ei->i_disksize >> 32); + if (ei->i_disksize > 0x7fffffffULL) { + struct super_block *sb = inode->i_sb; + if (!EXT3_HAS_RO_COMPAT_FEATURE(sb, + EXT3_FEATURE_RO_COMPAT_LARGE_FILE) || + EXT3_SB(sb)->s_es->s_rev_level == + cpu_to_le32(EXT3_GOOD_OLD_REV)) { + /* If this is the first large file + * created, add a flag to the superblock. + */ + err = ext3_journal_get_write_access(handle, + EXT3_SB(sb)->s_sbh); + if (err) + goto out_brelse; + ext3_update_dynamic_rev(sb); + EXT3_SET_RO_COMPAT_FEATURE(sb, + EXT3_FEATURE_RO_COMPAT_LARGE_FILE); + sb->s_dirt = 1; + handle->h_sync = 1; + err = ext3_journal_dirty_metadata(handle, + EXT3_SB(sb)->s_sbh); + } + } + } + raw_inode->i_generation = cpu_to_le32(inode->i_generation); + if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) { + if (old_valid_dev(inode->i_rdev)) { + raw_inode->i_block[0] = + cpu_to_le32(old_encode_dev(inode->i_rdev)); + raw_inode->i_block[1] = 0; + } else { + raw_inode->i_block[0] = 0; + raw_inode->i_block[1] = + cpu_to_le32(new_encode_dev(inode->i_rdev)); + raw_inode->i_block[2] = 0; + } + } else for (block = 0; block < EXT3_N_BLOCKS; block++) + raw_inode->i_block[block] = ei->i_data[block]; + + if (ei->i_extra_isize) + raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize); + + BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); + rc = ext3_journal_dirty_metadata(handle, bh); + if (!err) + err = rc; + ei->i_state &= ~EXT3_STATE_NEW; + +out_brelse: + brelse (bh); + ext3_std_error(inode->i_sb, err); + return err; +} + +/* + * ext3_write_inode() + * + * We are called from a few places: + * + * - Within generic_file_write() for O_SYNC files. + * Here, there will be no transaction running. We wait for any running + * trasnaction to commit. + * + * - Within sys_sync(), kupdate and such. + * We wait on commit, if tol to. + * + * - Within prune_icache() (PF_MEMALLOC == true) + * Here we simply return. We can't afford to block kswapd on the + * journal commit. + * + * In all cases it is actually safe for us to return without doing anything, + * because the inode has been copied into a raw inode buffer in + * ext3_mark_inode_dirty(). This is a correctness thing for O_SYNC and for + * knfsd. + * + * Note that we are absolutely dependent upon all inode dirtiers doing the + * right thing: they *must* call mark_inode_dirty() after dirtying info in + * which we are interested. + * + * It would be a bug for them to not do this. The code: + * + * mark_inode_dirty(inode) + * stuff(); + * inode->i_size = expr; + * + * is in error because a kswapd-driven write_inode() could occur while + * `stuff()' is running, and the new i_size will be lost. Plus the inode + * will no longer be on the superblock's dirty inode list. + */ +int ext3_write_inode(struct inode *inode, int wait) +{ + if (current->flags & PF_MEMALLOC) + return 0; + + if (ext3_journal_current_handle()) { + jbd_debug(0, "called recursively, non-PF_MEMALLOC!\n"); + dump_stack(); + return -EIO; + } + + if (!wait) + return 0; + + return ext3_force_commit(inode->i_sb); +} + +/* + * ext3_setattr() + * + * Called from notify_change. + * + * We want to trap VFS attempts to truncate the file as soon as + * possible. In particular, we want to make sure that when the VFS + * shrinks i_size, we put the inode on the orphan list and modify + * i_disksize immediately, so that during the subsequent flushing of + * dirty pages and freeing of disk blocks, we can guarantee that any + * commit will leave the blocks being flushed in an unused state on + * disk. (On recovery, the inode will get truncated and the blocks will + * be freed, so we have a strong guarantee that no future commit will + * leave these blocks visible to the user.) + * + * Called with inode->sem down. + */ +int ext3_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct inode *inode = dentry->d_inode; + int error, rc = 0; + const unsigned int ia_valid = attr->ia_valid; + + error = inode_change_ok(inode, attr); + if (error) + return error; + + if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) || + (ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) { + handle_t *handle; + + /* (user+group)*(old+new) structure, inode write (sb, + * inode block, ? - but truncate inode update has it) */ + handle = ext3_journal_start(inode, 2*(EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)+ + EXT3_QUOTA_DEL_BLOCKS(inode->i_sb))+3); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + goto err_out; + } + error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0; + if (error) { + ext3_journal_stop(handle); + return error; + } + /* Update corresponding info in inode so that everything is in + * one transaction */ + if (attr->ia_valid & ATTR_UID) + inode->i_uid = attr->ia_uid; + if (attr->ia_valid & ATTR_GID) + inode->i_gid = attr->ia_gid; + error = ext3_mark_inode_dirty(handle, inode); + ext3_journal_stop(handle); + } + + if (S_ISREG(inode->i_mode) && + attr->ia_valid & ATTR_SIZE && attr->ia_size < inode->i_size) { + handle_t *handle; + + handle = ext3_journal_start(inode, 3); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + goto err_out; + } + + error = ext3_orphan_add(handle, inode); + EXT3_I(inode)->i_disksize = attr->ia_size; + rc = ext3_mark_inode_dirty(handle, inode); + if (!error) + error = rc; + ext3_journal_stop(handle); + } + + rc = inode_setattr(inode, attr); + + /* If inode_setattr's call to ext3_truncate failed to get a + * transaction handle at all, we need to clean up the in-core + * orphan list manually. */ + if (inode->i_nlink) + ext3_orphan_del(NULL, inode); + + if (!rc && (ia_valid & ATTR_MODE)) + rc = ext3_acl_chmod(inode); + +err_out: + ext3_std_error(inode->i_sb, error); + if (!error) + error = rc; + return error; +} + + +/* + * How many blocks doth make a writepage()? + * + * With N blocks per page, it may be: + * N data blocks + * 2 indirect block + * 2 dindirect + * 1 tindirect + * N+5 bitmap blocks (from the above) + * N+5 group descriptor summary blocks + * 1 inode block + * 1 superblock. + * 2 * EXT3_SINGLEDATA_TRANS_BLOCKS for the quote files + * + * 3 * (N + 5) + 2 + 2 * EXT3_SINGLEDATA_TRANS_BLOCKS + * + * With ordered or writeback data it's the same, less the N data blocks. + * + * If the inode's direct blocks can hold an integral number of pages then a + * page cannot straddle two indirect blocks, and we can only touch one indirect + * and dindirect block, and the "5" above becomes "3". + * + * This still overestimates under most circumstances. If we were to pass the + * start and end offsets in here as well we could do block_to_path() on each + * block and work out the exact number of indirects which are touched. Pah. + */ + +static int ext3_writepage_trans_blocks(struct inode *inode) +{ + int bpp = ext3_journal_blocks_per_page(inode); + int indirects = (EXT3_NDIR_BLOCKS % bpp) ? 5 : 3; + int ret; + + if (ext3_should_journal_data(inode)) + ret = 3 * (bpp + indirects) + 2; + else + ret = 2 * (bpp + indirects) + 2; + +#ifdef CONFIG_QUOTA + /* We know that structure was already allocated during DQUOT_INIT so + * we will be updating only the data blocks + inodes */ + ret += 2*EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb); +#endif + + return ret; +} + +/* + * The caller must have previously called ext3_reserve_inode_write(). + * Give this, we know that the caller already has write access to iloc->bh. + */ +int ext3_mark_iloc_dirty(handle_t *handle, + struct inode *inode, struct ext3_iloc *iloc) +{ + int err = 0; + + /* the do_update_inode consumes one bh->b_count */ + get_bh(iloc->bh); + + /* ext3_do_update_inode() does journal_dirty_metadata */ + err = ext3_do_update_inode(handle, inode, iloc); + put_bh(iloc->bh); + return err; +} + +/* + * On success, We end up with an outstanding reference count against + * iloc->bh. This _must_ be cleaned up later. + */ + +int +ext3_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext3_iloc *iloc) +{ + int err = 0; + if (handle) { + err = ext3_get_inode_loc(inode, iloc); + if (!err) { + BUFFER_TRACE(iloc->bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, iloc->bh); + if (err) { + brelse(iloc->bh); + iloc->bh = NULL; + } + } + } + ext3_std_error(inode->i_sb, err); + return err; +} + +/* + * What we do here is to mark the in-core inode as clean with respect to inode + * dirtiness (it may still be data-dirty). + * This means that the in-core inode may be reaped by prune_icache + * without having to perform any I/O. This is a very good thing, + * because *any* task may call prune_icache - even ones which + * have a transaction open against a different journal. + * + * Is this cheating? Not really. Sure, we haven't written the + * inode out, but prune_icache isn't a user-visible syncing function. + * Whenever the user wants stuff synced (sys_sync, sys_msync, sys_fsync) + * we start and wait on commits. + * + * Is this efficient/effective? Well, we're being nice to the system + * by cleaning up our inodes proactively so they can be reaped + * without I/O. But we are potentially leaving up to five seconds' + * worth of inodes floating about which prune_icache wants us to + * write out. One way to fix that would be to get prune_icache() + * to do a write_super() to free up some memory. It has the desired + * effect. + */ +int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode) +{ + struct ext3_iloc iloc; + int err; + + might_sleep(); + err = ext3_reserve_inode_write(handle, inode, &iloc); + if (!err) + err = ext3_mark_iloc_dirty(handle, inode, &iloc); + return err; +} + +/* + * ext3_dirty_inode() is called from __mark_inode_dirty() + * + * We're really interested in the case where a file is being extended. + * i_size has been changed by generic_commit_write() and we thus need + * to include the updated inode in the current transaction. + * + * Also, DQUOT_ALLOC_SPACE() will always dirty the inode when blocks + * are allocated to the file. + * + * If the inode is marked synchronous, we don't honour that here - doing + * so would cause a commit on atime updates, which we don't bother doing. + * We handle synchronous inodes at the highest possible level. + */ +void ext3_dirty_inode(struct inode *inode) +{ + handle_t *current_handle = ext3_journal_current_handle(); + handle_t *handle; + + handle = ext3_journal_start(inode, 2); + if (IS_ERR(handle)) + goto out; + if (current_handle && + current_handle->h_transaction != handle->h_transaction) { + /* This task has a transaction open against a different fs */ + printk(KERN_EMERG "%s: transactions do not match!\n", + __FUNCTION__); + } else { + jbd_debug(5, "marking dirty. outer handle=%p\n", + current_handle); + ext3_mark_inode_dirty(handle, inode); + } + ext3_journal_stop(handle); +out: + return; +} + +#if 0 +/* + * Bind an inode's backing buffer_head into this transaction, to prevent + * it from being flushed to disk early. Unlike + * ext3_reserve_inode_write, this leaves behind no bh reference and + * returns no iloc structure, so the caller needs to repeat the iloc + * lookup to mark the inode dirty later. + */ +static int ext3_pin_inode(handle_t *handle, struct inode *inode) +{ + struct ext3_iloc iloc; + + int err = 0; + if (handle) { + err = ext3_get_inode_loc(inode, &iloc); + if (!err) { + BUFFER_TRACE(iloc.bh, "get_write_access"); + err = journal_get_write_access(handle, iloc.bh); + if (!err) + err = ext3_journal_dirty_metadata(handle, + iloc.bh); + brelse(iloc.bh); + } + } + ext3_std_error(inode->i_sb, err); + return err; +} +#endif + +int ext3_change_inode_journal_flag(struct inode *inode, int val) +{ + journal_t *journal; + handle_t *handle; + int err; + + /* + * We have to be very careful here: changing a data block's + * journaling status dynamically is dangerous. If we write a + * data block to the journal, change the status and then delete + * that block, we risk forgetting to revoke the old log record + * from the journal and so a subsequent replay can corrupt data. + * So, first we make sure that the journal is empty and that + * nobody is changing anything. + */ + + journal = EXT3_JOURNAL(inode); + if (is_journal_aborted(journal) || IS_RDONLY(inode)) + return -EROFS; + + journal_lock_updates(journal); + journal_flush(journal); + + /* + * OK, there are no updates running now, and all cached data is + * synced to disk. We are now in a completely consistent state + * which doesn't have anything in the journal, and we know that + * no filesystem updates are running, so it is safe to modify + * the inode's in-core data-journaling state flag now. + */ + + if (val) + EXT3_I(inode)->i_flags |= EXT3_JOURNAL_DATA_FL; + else + EXT3_I(inode)->i_flags &= ~EXT3_JOURNAL_DATA_FL; + ext3_set_aops(inode); + + journal_unlock_updates(journal); + + /* Finally we can mark the inode as dirty. */ + + handle = ext3_journal_start(inode, 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + err = ext3_mark_inode_dirty(handle, inode); + handle->h_sync = 1; + ext3_journal_stop(handle); + ext3_std_error(inode->i_sb, err); + + return err; +} diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c new file mode 100644 index 0000000..12daa68 --- /dev/null +++ b/fs/ext4/ioctl.c @@ -0,0 +1,307 @@ +/* + * linux/fs/ext3/ioctl.c + * + * Copyright (C) 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, + unsigned long arg) +{ + struct ext3_inode_info *ei = EXT3_I(inode); + unsigned int flags; + unsigned short rsv_window_size; + + ext3_debug ("cmd = %u, arg = %lu\n", cmd, arg); + + switch (cmd) { + case EXT3_IOC_GETFLAGS: + flags = ei->i_flags & EXT3_FL_USER_VISIBLE; + return put_user(flags, (int __user *) arg); + case EXT3_IOC_SETFLAGS: { + handle_t *handle = NULL; + int err; + struct ext3_iloc iloc; + unsigned int oldflags; + unsigned int jflag; + + if (IS_RDONLY(inode)) + return -EROFS; + + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EACCES; + + if (get_user(flags, (int __user *) arg)) + return -EFAULT; + + if (!S_ISDIR(inode->i_mode)) + flags &= ~EXT3_DIRSYNC_FL; + + mutex_lock(&inode->i_mutex); + oldflags = ei->i_flags; + + /* The JOURNAL_DATA flag is modifiable only by root */ + jflag = flags & EXT3_JOURNAL_DATA_FL; + + /* + * The IMMUTABLE and APPEND_ONLY flags can only be changed by + * the relevant capability. + * + * This test looks nicer. Thanks to Pauline Middelink + */ + if ((flags ^ oldflags) & (EXT3_APPEND_FL | EXT3_IMMUTABLE_FL)) { + if (!capable(CAP_LINUX_IMMUTABLE)) { + mutex_unlock(&inode->i_mutex); + return -EPERM; + } + } + + /* + * The JOURNAL_DATA flag can only be changed by + * the relevant capability. + */ + if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) { + if (!capable(CAP_SYS_RESOURCE)) { + mutex_unlock(&inode->i_mutex); + return -EPERM; + } + } + + + handle = ext3_journal_start(inode, 1); + if (IS_ERR(handle)) { + mutex_unlock(&inode->i_mutex); + return PTR_ERR(handle); + } + if (IS_SYNC(inode)) + handle->h_sync = 1; + err = ext3_reserve_inode_write(handle, inode, &iloc); + if (err) + goto flags_err; + + flags = flags & EXT3_FL_USER_MODIFIABLE; + flags |= oldflags & ~EXT3_FL_USER_MODIFIABLE; + ei->i_flags = flags; + + ext3_set_inode_flags(inode); + inode->i_ctime = CURRENT_TIME_SEC; + + err = ext3_mark_iloc_dirty(handle, inode, &iloc); +flags_err: + ext3_journal_stop(handle); + if (err) { + mutex_unlock(&inode->i_mutex); + return err; + } + + if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) + err = ext3_change_inode_journal_flag(inode, jflag); + mutex_unlock(&inode->i_mutex); + return err; + } + case EXT3_IOC_GETVERSION: + case EXT3_IOC_GETVERSION_OLD: + return put_user(inode->i_generation, (int __user *) arg); + case EXT3_IOC_SETVERSION: + case EXT3_IOC_SETVERSION_OLD: { + handle_t *handle; + struct ext3_iloc iloc; + __u32 generation; + int err; + + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EPERM; + if (IS_RDONLY(inode)) + return -EROFS; + if (get_user(generation, (int __user *) arg)) + return -EFAULT; + + handle = ext3_journal_start(inode, 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + err = ext3_reserve_inode_write(handle, inode, &iloc); + if (err == 0) { + inode->i_ctime = CURRENT_TIME_SEC; + inode->i_generation = generation; + err = ext3_mark_iloc_dirty(handle, inode, &iloc); + } + ext3_journal_stop(handle); + return err; + } +#ifdef CONFIG_JBD_DEBUG + case EXT3_IOC_WAIT_FOR_READONLY: + /* + * This is racy - by the time we're woken up and running, + * the superblock could be released. And the module could + * have been unloaded. So sue me. + * + * Returns 1 if it slept, else zero. + */ + { + struct super_block *sb = inode->i_sb; + DECLARE_WAITQUEUE(wait, current); + int ret = 0; + + set_current_state(TASK_INTERRUPTIBLE); + add_wait_queue(&EXT3_SB(sb)->ro_wait_queue, &wait); + if (timer_pending(&EXT3_SB(sb)->turn_ro_timer)) { + schedule(); + ret = 1; + } + remove_wait_queue(&EXT3_SB(sb)->ro_wait_queue, &wait); + return ret; + } +#endif + case EXT3_IOC_GETRSVSZ: + if (test_opt(inode->i_sb, RESERVATION) + && S_ISREG(inode->i_mode) + && ei->i_block_alloc_info) { + rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size; + return put_user(rsv_window_size, (int __user *)arg); + } + return -ENOTTY; + case EXT3_IOC_SETRSVSZ: { + + if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) + return -ENOTTY; + + if (IS_RDONLY(inode)) + return -EROFS; + + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EACCES; + + if (get_user(rsv_window_size, (int __user *)arg)) + return -EFAULT; + + if (rsv_window_size > EXT3_MAX_RESERVE_BLOCKS) + rsv_window_size = EXT3_MAX_RESERVE_BLOCKS; + + /* + * need to allocate reservation structure for this inode + * before set the window size + */ + mutex_lock(&ei->truncate_mutex); + if (!ei->i_block_alloc_info) + ext3_init_block_alloc_info(inode); + + if (ei->i_block_alloc_info){ + struct ext3_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; + rsv->rsv_goal_size = rsv_window_size; + } + mutex_unlock(&ei->truncate_mutex); + return 0; + } + case EXT3_IOC_GROUP_EXTEND: { + ext3_fsblk_t n_blocks_count; + struct super_block *sb = inode->i_sb; + int err; + + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + + if (IS_RDONLY(inode)) + return -EROFS; + + if (get_user(n_blocks_count, (__u32 __user *)arg)) + return -EFAULT; + + err = ext3_group_extend(sb, EXT3_SB(sb)->s_es, n_blocks_count); + journal_lock_updates(EXT3_SB(sb)->s_journal); + journal_flush(EXT3_SB(sb)->s_journal); + journal_unlock_updates(EXT3_SB(sb)->s_journal); + + return err; + } + case EXT3_IOC_GROUP_ADD: { + struct ext3_new_group_data input; + struct super_block *sb = inode->i_sb; + int err; + + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + + if (IS_RDONLY(inode)) + return -EROFS; + + if (copy_from_user(&input, (struct ext3_new_group_input __user *)arg, + sizeof(input))) + return -EFAULT; + + err = ext3_group_add(sb, &input); + journal_lock_updates(EXT3_SB(sb)->s_journal); + journal_flush(EXT3_SB(sb)->s_journal); + journal_unlock_updates(EXT3_SB(sb)->s_journal); + + return err; + } + + + default: + return -ENOTTY; + } +} + +#ifdef CONFIG_COMPAT +long ext3_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + struct inode *inode = file->f_dentry->d_inode; + int ret; + + /* These are just misnamed, they actually get/put from/to user an int */ + switch (cmd) { + case EXT3_IOC32_GETFLAGS: + cmd = EXT3_IOC_GETFLAGS; + break; + case EXT3_IOC32_SETFLAGS: + cmd = EXT3_IOC_SETFLAGS; + break; + case EXT3_IOC32_GETVERSION: + cmd = EXT3_IOC_GETVERSION; + break; + case EXT3_IOC32_SETVERSION: + cmd = EXT3_IOC_SETVERSION; + break; + case EXT3_IOC32_GROUP_EXTEND: + cmd = EXT3_IOC_GROUP_EXTEND; + break; + case EXT3_IOC32_GETVERSION_OLD: + cmd = EXT3_IOC_GETVERSION_OLD; + break; + case EXT3_IOC32_SETVERSION_OLD: + cmd = EXT3_IOC_SETVERSION_OLD; + break; +#ifdef CONFIG_JBD_DEBUG + case EXT3_IOC32_WAIT_FOR_READONLY: + cmd = EXT3_IOC_WAIT_FOR_READONLY; + break; +#endif + case EXT3_IOC32_GETRSVSZ: + cmd = EXT3_IOC_GETRSVSZ; + break; + case EXT3_IOC32_SETRSVSZ: + cmd = EXT3_IOC_SETRSVSZ; + break; + case EXT3_IOC_GROUP_ADD: + break; + default: + return -ENOIOCTLCMD; + } + lock_kernel(); + ret = ext3_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg)); + unlock_kernel(); + return ret; +} +#endif diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c new file mode 100644 index 0000000..906731a --- /dev/null +++ b/fs/ext4/namei.c @@ -0,0 +1,2397 @@ +/* + * linux/fs/ext3/namei.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/namei.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * Directory entry file type support and forward compatibility hooks + * for B-tree directories by Theodore Ts'o (tytso@mit.edu), 1998 + * Hash Tree Directory indexing (c) + * Daniel Phillips, 2001 + * Hash Tree Directory indexing porting + * Christopher Li, 2002 + * Hash Tree Directory indexing cleanup + * Theodore Ts'o, 2002 + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "namei.h" +#include "xattr.h" +#include "acl.h" + +/* + * define how far ahead to read directories while searching them. + */ +#define NAMEI_RA_CHUNKS 2 +#define NAMEI_RA_BLOCKS 4 +#define NAMEI_RA_SIZE (NAMEI_RA_CHUNKS * NAMEI_RA_BLOCKS) +#define NAMEI_RA_INDEX(c,b) (((c) * NAMEI_RA_BLOCKS) + (b)) + +static struct buffer_head *ext3_append(handle_t *handle, + struct inode *inode, + u32 *block, int *err) +{ + struct buffer_head *bh; + + *block = inode->i_size >> inode->i_sb->s_blocksize_bits; + + if ((bh = ext3_bread(handle, inode, *block, 1, err))) { + inode->i_size += inode->i_sb->s_blocksize; + EXT3_I(inode)->i_disksize = inode->i_size; + ext3_journal_get_write_access(handle,bh); + } + return bh; +} + +#ifndef assert +#define assert(test) J_ASSERT(test) +#endif + +#ifndef swap +#define swap(x, y) do { typeof(x) z = x; x = y; y = z; } while (0) +#endif + +#ifdef DX_DEBUG +#define dxtrace(command) command +#else +#define dxtrace(command) +#endif + +struct fake_dirent +{ + __le32 inode; + __le16 rec_len; + u8 name_len; + u8 file_type; +}; + +struct dx_countlimit +{ + __le16 limit; + __le16 count; +}; + +struct dx_entry +{ + __le32 hash; + __le32 block; +}; + +/* + * dx_root_info is laid out so that if it should somehow get overlaid by a + * dirent the two low bits of the hash version will be zero. Therefore, the + * hash version mod 4 should never be 0. Sincerely, the paranoia department. + */ + +struct dx_root +{ + struct fake_dirent dot; + char dot_name[4]; + struct fake_dirent dotdot; + char dotdot_name[4]; + struct dx_root_info + { + __le32 reserved_zero; + u8 hash_version; + u8 info_length; /* 8 */ + u8 indirect_levels; + u8 unused_flags; + } + info; + struct dx_entry entries[0]; +}; + +struct dx_node +{ + struct fake_dirent fake; + struct dx_entry entries[0]; +}; + + +struct dx_frame +{ + struct buffer_head *bh; + struct dx_entry *entries; + struct dx_entry *at; +}; + +struct dx_map_entry +{ + u32 hash; + u32 offs; +}; + +#ifdef CONFIG_EXT3_INDEX +static inline unsigned dx_get_block (struct dx_entry *entry); +static void dx_set_block (struct dx_entry *entry, unsigned value); +static inline unsigned dx_get_hash (struct dx_entry *entry); +static void dx_set_hash (struct dx_entry *entry, unsigned value); +static unsigned dx_get_count (struct dx_entry *entries); +static unsigned dx_get_limit (struct dx_entry *entries); +static void dx_set_count (struct dx_entry *entries, unsigned value); +static void dx_set_limit (struct dx_entry *entries, unsigned value); +static unsigned dx_root_limit (struct inode *dir, unsigned infosize); +static unsigned dx_node_limit (struct inode *dir); +static struct dx_frame *dx_probe(struct dentry *dentry, + struct inode *dir, + struct dx_hash_info *hinfo, + struct dx_frame *frame, + int *err); +static void dx_release (struct dx_frame *frames); +static int dx_make_map (struct ext3_dir_entry_2 *de, int size, + struct dx_hash_info *hinfo, struct dx_map_entry map[]); +static void dx_sort_map(struct dx_map_entry *map, unsigned count); +static struct ext3_dir_entry_2 *dx_move_dirents (char *from, char *to, + struct dx_map_entry *offsets, int count); +static struct ext3_dir_entry_2* dx_pack_dirents (char *base, int size); +static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block); +static int ext3_htree_next_block(struct inode *dir, __u32 hash, + struct dx_frame *frame, + struct dx_frame *frames, + __u32 *start_hash); +static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, + struct ext3_dir_entry_2 **res_dir, int *err); +static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, + struct inode *inode); + +/* + * Future: use high four bits of block for coalesce-on-delete flags + * Mask them off for now. + */ + +static inline unsigned dx_get_block (struct dx_entry *entry) +{ + return le32_to_cpu(entry->block) & 0x00ffffff; +} + +static inline void dx_set_block (struct dx_entry *entry, unsigned value) +{ + entry->block = cpu_to_le32(value); +} + +static inline unsigned dx_get_hash (struct dx_entry *entry) +{ + return le32_to_cpu(entry->hash); +} + +static inline void dx_set_hash (struct dx_entry *entry, unsigned value) +{ + entry->hash = cpu_to_le32(value); +} + +static inline unsigned dx_get_count (struct dx_entry *entries) +{ + return le16_to_cpu(((struct dx_countlimit *) entries)->count); +} + +static inline unsigned dx_get_limit (struct dx_entry *entries) +{ + return le16_to_cpu(((struct dx_countlimit *) entries)->limit); +} + +static inline void dx_set_count (struct dx_entry *entries, unsigned value) +{ + ((struct dx_countlimit *) entries)->count = cpu_to_le16(value); +} + +static inline void dx_set_limit (struct dx_entry *entries, unsigned value) +{ + ((struct dx_countlimit *) entries)->limit = cpu_to_le16(value); +} + +static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize) +{ + unsigned entry_space = dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(1) - + EXT3_DIR_REC_LEN(2) - infosize; + return 0? 20: entry_space / sizeof(struct dx_entry); +} + +static inline unsigned dx_node_limit (struct inode *dir) +{ + unsigned entry_space = dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(0); + return 0? 22: entry_space / sizeof(struct dx_entry); +} + +/* + * Debug + */ +#ifdef DX_DEBUG +static void dx_show_index (char * label, struct dx_entry *entries) +{ + int i, n = dx_get_count (entries); + printk("%s index ", label); + for (i = 0; i < n; i++) + { + printk("%x->%u ", i? dx_get_hash(entries + i): 0, dx_get_block(entries + i)); + } + printk("\n"); +} + +struct stats +{ + unsigned names; + unsigned space; + unsigned bcount; +}; + +static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext3_dir_entry_2 *de, + int size, int show_names) +{ + unsigned names = 0, space = 0; + char *base = (char *) de; + struct dx_hash_info h = *hinfo; + + printk("names: "); + while ((char *) de < base + size) + { + if (de->inode) + { + if (show_names) + { + int len = de->name_len; + char *name = de->name; + while (len--) printk("%c", *name++); + ext3fs_dirhash(de->name, de->name_len, &h); + printk(":%x.%u ", h.hash, + ((char *) de - base)); + } + space += EXT3_DIR_REC_LEN(de->name_len); + names++; + } + de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + } + printk("(%i)\n", names); + return (struct stats) { names, space, 1 }; +} + +struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, + struct dx_entry *entries, int levels) +{ + unsigned blocksize = dir->i_sb->s_blocksize; + unsigned count = dx_get_count (entries), names = 0, space = 0, i; + unsigned bcount = 0; + struct buffer_head *bh; + int err; + printk("%i indexed blocks...\n", count); + for (i = 0; i < count; i++, entries++) + { + u32 block = dx_get_block(entries), hash = i? dx_get_hash(entries): 0; + u32 range = i < count - 1? (dx_get_hash(entries + 1) - hash): ~hash; + struct stats stats; + printk("%s%3u:%03u hash %8x/%8x ",levels?"":" ", i, block, hash, range); + if (!(bh = ext3_bread (NULL,dir, block, 0,&err))) continue; + stats = levels? + dx_show_entries(hinfo, dir, ((struct dx_node *) bh->b_data)->entries, levels - 1): + dx_show_leaf(hinfo, (struct ext3_dir_entry_2 *) bh->b_data, blocksize, 0); + names += stats.names; + space += stats.space; + bcount += stats.bcount; + brelse (bh); + } + if (bcount) + printk("%snames %u, fullness %u (%u%%)\n", levels?"":" ", + names, space/bcount,(space/bcount)*100/blocksize); + return (struct stats) { names, space, bcount}; +} +#endif /* DX_DEBUG */ + +/* + * Probe for a directory leaf block to search. + * + * dx_probe can return ERR_BAD_DX_DIR, which means there was a format + * error in the directory index, and the caller should fall back to + * searching the directory normally. The callers of dx_probe **MUST** + * check for this error code, and make sure it never gets reflected + * back to userspace. + */ +static struct dx_frame * +dx_probe(struct dentry *dentry, struct inode *dir, + struct dx_hash_info *hinfo, struct dx_frame *frame_in, int *err) +{ + unsigned count, indirect; + struct dx_entry *at, *entries, *p, *q, *m; + struct dx_root *root; + struct buffer_head *bh; + struct dx_frame *frame = frame_in; + u32 hash; + + frame->bh = NULL; + if (dentry) + dir = dentry->d_parent->d_inode; + if (!(bh = ext3_bread (NULL,dir, 0, 0, err))) + goto fail; + root = (struct dx_root *) bh->b_data; + if (root->info.hash_version != DX_HASH_TEA && + root->info.hash_version != DX_HASH_HALF_MD4 && + root->info.hash_version != DX_HASH_LEGACY) { + ext3_warning(dir->i_sb, __FUNCTION__, + "Unrecognised inode hash code %d", + root->info.hash_version); + brelse(bh); + *err = ERR_BAD_DX_DIR; + goto fail; + } + hinfo->hash_version = root->info.hash_version; + hinfo->seed = EXT3_SB(dir->i_sb)->s_hash_seed; + if (dentry) + ext3fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo); + hash = hinfo->hash; + + if (root->info.unused_flags & 1) { + ext3_warning(dir->i_sb, __FUNCTION__, + "Unimplemented inode hash flags: %#06x", + root->info.unused_flags); + brelse(bh); + *err = ERR_BAD_DX_DIR; + goto fail; + } + + if ((indirect = root->info.indirect_levels) > 1) { + ext3_warning(dir->i_sb, __FUNCTION__, + "Unimplemented inode hash depth: %#06x", + root->info.indirect_levels); + brelse(bh); + *err = ERR_BAD_DX_DIR; + goto fail; + } + + entries = (struct dx_entry *) (((char *)&root->info) + + root->info.info_length); + assert(dx_get_limit(entries) == dx_root_limit(dir, + root->info.info_length)); + dxtrace (printk("Look up %x", hash)); + while (1) + { + count = dx_get_count(entries); + assert (count && count <= dx_get_limit(entries)); + p = entries + 1; + q = entries + count - 1; + while (p <= q) + { + m = p + (q - p)/2; + dxtrace(printk(".")); + if (dx_get_hash(m) > hash) + q = m - 1; + else + p = m + 1; + } + + if (0) // linear search cross check + { + unsigned n = count - 1; + at = entries; + while (n--) + { + dxtrace(printk(",")); + if (dx_get_hash(++at) > hash) + { + at--; + break; + } + } + assert (at == p - 1); + } + + at = p - 1; + dxtrace(printk(" %x->%u\n", at == entries? 0: dx_get_hash(at), dx_get_block(at))); + frame->bh = bh; + frame->entries = entries; + frame->at = at; + if (!indirect--) return frame; + if (!(bh = ext3_bread (NULL,dir, dx_get_block(at), 0, err))) + goto fail2; + at = entries = ((struct dx_node *) bh->b_data)->entries; + assert (dx_get_limit(entries) == dx_node_limit (dir)); + frame++; + } +fail2: + while (frame >= frame_in) { + brelse(frame->bh); + frame--; + } +fail: + return NULL; +} + +static void dx_release (struct dx_frame *frames) +{ + if (frames[0].bh == NULL) + return; + + if (((struct dx_root *) frames[0].bh->b_data)->info.indirect_levels) + brelse(frames[1].bh); + brelse(frames[0].bh); +} + +/* + * This function increments the frame pointer to search the next leaf + * block, and reads in the necessary intervening nodes if the search + * should be necessary. Whether or not the search is necessary is + * controlled by the hash parameter. If the hash value is even, then + * the search is only continued if the next block starts with that + * hash value. This is used if we are searching for a specific file. + * + * If the hash value is HASH_NB_ALWAYS, then always go to the next block. + * + * This function returns 1 if the caller should continue to search, + * or 0 if it should not. If there is an error reading one of the + * index blocks, it will a negative error code. + * + * If start_hash is non-null, it will be filled in with the starting + * hash of the next page. + */ +static int ext3_htree_next_block(struct inode *dir, __u32 hash, + struct dx_frame *frame, + struct dx_frame *frames, + __u32 *start_hash) +{ + struct dx_frame *p; + struct buffer_head *bh; + int err, num_frames = 0; + __u32 bhash; + + p = frame; + /* + * Find the next leaf page by incrementing the frame pointer. + * If we run out of entries in the interior node, loop around and + * increment pointer in the parent node. When we break out of + * this loop, num_frames indicates the number of interior + * nodes need to be read. + */ + while (1) { + if (++(p->at) < p->entries + dx_get_count(p->entries)) + break; + if (p == frames) + return 0; + num_frames++; + p--; + } + + /* + * If the hash is 1, then continue only if the next page has a + * continuation hash of any value. This is used for readdir + * handling. Otherwise, check to see if the hash matches the + * desired contiuation hash. If it doesn't, return since + * there's no point to read in the successive index pages. + */ + bhash = dx_get_hash(p->at); + if (start_hash) + *start_hash = bhash; + if ((hash & 1) == 0) { + if ((bhash & ~1) != hash) + return 0; + } + /* + * If the hash is HASH_NB_ALWAYS, we always go to the next + * block so no check is necessary + */ + while (num_frames--) { + if (!(bh = ext3_bread(NULL, dir, dx_get_block(p->at), + 0, &err))) + return err; /* Failure */ + p++; + brelse (p->bh); + p->bh = bh; + p->at = p->entries = ((struct dx_node *) bh->b_data)->entries; + } + return 1; +} + + +/* + * p is at least 6 bytes before the end of page + */ +static inline struct ext3_dir_entry_2 *ext3_next_entry(struct ext3_dir_entry_2 *p) +{ + return (struct ext3_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len)); +} + +/* + * This function fills a red-black tree with information from a + * directory block. It returns the number directory entries loaded + * into the tree. If there is an error it is returned in err. + */ +static int htree_dirblock_to_tree(struct file *dir_file, + struct inode *dir, int block, + struct dx_hash_info *hinfo, + __u32 start_hash, __u32 start_minor_hash) +{ + struct buffer_head *bh; + struct ext3_dir_entry_2 *de, *top; + int err, count = 0; + + dxtrace(printk("In htree dirblock_to_tree: block %d\n", block)); + if (!(bh = ext3_bread (NULL, dir, block, 0, &err))) + return err; + + de = (struct ext3_dir_entry_2 *) bh->b_data; + top = (struct ext3_dir_entry_2 *) ((char *) de + + dir->i_sb->s_blocksize - + EXT3_DIR_REC_LEN(0)); + for (; de < top; de = ext3_next_entry(de)) { + ext3fs_dirhash(de->name, de->name_len, hinfo); + if ((hinfo->hash < start_hash) || + ((hinfo->hash == start_hash) && + (hinfo->minor_hash < start_minor_hash))) + continue; + if (de->inode == 0) + continue; + if ((err = ext3_htree_store_dirent(dir_file, + hinfo->hash, hinfo->minor_hash, de)) != 0) { + brelse(bh); + return err; + } + count++; + } + brelse(bh); + return count; +} + + +/* + * This function fills a red-black tree with information from a + * directory. We start scanning the directory in hash order, starting + * at start_hash and start_minor_hash. + * + * This function returns the number of entries inserted into the tree, + * or a negative error code. + */ +int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, + __u32 start_minor_hash, __u32 *next_hash) +{ + struct dx_hash_info hinfo; + struct ext3_dir_entry_2 *de; + struct dx_frame frames[2], *frame; + struct inode *dir; + int block, err; + int count = 0; + int ret; + __u32 hashval; + + dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, + start_minor_hash)); + dir = dir_file->f_dentry->d_inode; + if (!(EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) { + hinfo.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; + hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; + count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo, + start_hash, start_minor_hash); + *next_hash = ~0; + return count; + } + hinfo.hash = start_hash; + hinfo.minor_hash = 0; + frame = dx_probe(NULL, dir_file->f_dentry->d_inode, &hinfo, frames, &err); + if (!frame) + return err; + + /* Add '.' and '..' from the htree header */ + if (!start_hash && !start_minor_hash) { + de = (struct ext3_dir_entry_2 *) frames[0].bh->b_data; + if ((err = ext3_htree_store_dirent(dir_file, 0, 0, de)) != 0) + goto errout; + count++; + } + if (start_hash < 2 || (start_hash ==2 && start_minor_hash==0)) { + de = (struct ext3_dir_entry_2 *) frames[0].bh->b_data; + de = ext3_next_entry(de); + if ((err = ext3_htree_store_dirent(dir_file, 2, 0, de)) != 0) + goto errout; + count++; + } + + while (1) { + block = dx_get_block(frame->at); + ret = htree_dirblock_to_tree(dir_file, dir, block, &hinfo, + start_hash, start_minor_hash); + if (ret < 0) { + err = ret; + goto errout; + } + count += ret; + hashval = ~0; + ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS, + frame, frames, &hashval); + *next_hash = hashval; + if (ret < 0) { + err = ret; + goto errout; + } + /* + * Stop if: (a) there are no more entries, or + * (b) we have inserted at least one entry and the + * next hash value is not a continuation + */ + if ((ret == 0) || + (count && ((hashval & 1) == 0))) + break; + } + dx_release(frames); + dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", + count, *next_hash)); + return count; +errout: + dx_release(frames); + return (err); +} + + +/* + * Directory block splitting, compacting + */ + +static int dx_make_map (struct ext3_dir_entry_2 *de, int size, + struct dx_hash_info *hinfo, struct dx_map_entry *map_tail) +{ + int count = 0; + char *base = (char *) de; + struct dx_hash_info h = *hinfo; + + while ((char *) de < base + size) + { + if (de->name_len && de->inode) { + ext3fs_dirhash(de->name, de->name_len, &h); + map_tail--; + map_tail->hash = h.hash; + map_tail->offs = (u32) ((char *) de - base); + count++; + cond_resched(); + } + /* XXX: do we need to check rec_len == 0 case? -Chris */ + de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + } + return count; +} + +static void dx_sort_map (struct dx_map_entry *map, unsigned count) +{ + struct dx_map_entry *p, *q, *top = map + count - 1; + int more; + /* Combsort until bubble sort doesn't suck */ + while (count > 2) + { + count = count*10/13; + if (count - 9 < 2) /* 9, 10 -> 11 */ + count = 11; + for (p = top, q = p - count; q >= map; p--, q--) + if (p->hash < q->hash) + swap(*p, *q); + } + /* Garden variety bubble sort */ + do { + more = 0; + q = top; + while (q-- > map) + { + if (q[1].hash >= q[0].hash) + continue; + swap(*(q+1), *q); + more = 1; + } + } while(more); +} + +static void dx_insert_block(struct dx_frame *frame, u32 hash, u32 block) +{ + struct dx_entry *entries = frame->entries; + struct dx_entry *old = frame->at, *new = old + 1; + int count = dx_get_count(entries); + + assert(count < dx_get_limit(entries)); + assert(old < entries + count); + memmove(new + 1, new, (char *)(entries + count) - (char *)(new)); + dx_set_hash(new, hash); + dx_set_block(new, block); + dx_set_count(entries, count + 1); +} +#endif + + +static void ext3_update_dx_flag(struct inode *inode) +{ + if (!EXT3_HAS_COMPAT_FEATURE(inode->i_sb, + EXT3_FEATURE_COMPAT_DIR_INDEX)) + EXT3_I(inode)->i_flags &= ~EXT3_INDEX_FL; +} + +/* + * NOTE! unlike strncmp, ext3_match returns 1 for success, 0 for failure. + * + * `len <= EXT3_NAME_LEN' is guaranteed by caller. + * `de != NULL' is guaranteed by caller. + */ +static inline int ext3_match (int len, const char * const name, + struct ext3_dir_entry_2 * de) +{ + if (len != de->name_len) + return 0; + if (!de->inode) + return 0; + return !memcmp(name, de->name, len); +} + +/* + * Returns 0 if not found, -1 on failure, and 1 on success + */ +static inline int search_dirblock(struct buffer_head * bh, + struct inode *dir, + struct dentry *dentry, + unsigned long offset, + struct ext3_dir_entry_2 ** res_dir) +{ + struct ext3_dir_entry_2 * de; + char * dlimit; + int de_len; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + + de = (struct ext3_dir_entry_2 *) bh->b_data; + dlimit = bh->b_data + dir->i_sb->s_blocksize; + while ((char *) de < dlimit) { + /* this code is executed quadratically often */ + /* do minimal checking `by hand' */ + + if ((char *) de + namelen <= dlimit && + ext3_match (namelen, name, de)) { + /* found a match - just to be sure, do a full check */ + if (!ext3_check_dir_entry("ext3_find_entry", + dir, de, bh, offset)) + return -1; + *res_dir = de; + return 1; + } + /* prevent looping on a bad block */ + de_len = le16_to_cpu(de->rec_len); + if (de_len <= 0) + return -1; + offset += de_len; + de = (struct ext3_dir_entry_2 *) ((char *) de + de_len); + } + return 0; +} + + +/* + * ext3_find_entry() + * + * finds an entry in the specified directory with the wanted name. It + * returns the cache buffer in which the entry was found, and the entry + * itself (as a parameter - res_dir). It does NOT read the inode of the + * entry - you'll have to do that yourself if you want to. + * + * The returned buffer_head has ->b_count elevated. The caller is expected + * to brelse() it when appropriate. + */ +static struct buffer_head * ext3_find_entry (struct dentry *dentry, + struct ext3_dir_entry_2 ** res_dir) +{ + struct super_block * sb; + struct buffer_head * bh_use[NAMEI_RA_SIZE]; + struct buffer_head * bh, *ret = NULL; + unsigned long start, block, b; + int ra_max = 0; /* Number of bh's in the readahead + buffer, bh_use[] */ + int ra_ptr = 0; /* Current index into readahead + buffer */ + int num = 0; + int nblocks, i, err; + struct inode *dir = dentry->d_parent->d_inode; + int namelen; + const u8 *name; + unsigned blocksize; + + *res_dir = NULL; + sb = dir->i_sb; + blocksize = sb->s_blocksize; + namelen = dentry->d_name.len; + name = dentry->d_name.name; + if (namelen > EXT3_NAME_LEN) + return NULL; +#ifdef CONFIG_EXT3_INDEX + if (is_dx(dir)) { + bh = ext3_dx_find_entry(dentry, res_dir, &err); + /* + * On success, or if the error was file not found, + * return. Otherwise, fall back to doing a search the + * old fashioned way. + */ + if (bh || (err != ERR_BAD_DX_DIR)) + return bh; + dxtrace(printk("ext3_find_entry: dx failed, falling back\n")); + } +#endif + nblocks = dir->i_size >> EXT3_BLOCK_SIZE_BITS(sb); + start = EXT3_I(dir)->i_dir_start_lookup; + if (start >= nblocks) + start = 0; + block = start; +restart: + do { + /* + * We deal with the read-ahead logic here. + */ + if (ra_ptr >= ra_max) { + /* Refill the readahead buffer */ + ra_ptr = 0; + b = block; + for (ra_max = 0; ra_max < NAMEI_RA_SIZE; ra_max++) { + /* + * Terminate if we reach the end of the + * directory and must wrap, or if our + * search has finished at this block. + */ + if (b >= nblocks || (num && block == start)) { + bh_use[ra_max] = NULL; + break; + } + num++; + bh = ext3_getblk(NULL, dir, b++, 0, &err); + bh_use[ra_max] = bh; + if (bh) + ll_rw_block(READ_META, 1, &bh); + } + } + if ((bh = bh_use[ra_ptr++]) == NULL) + goto next; + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) { + /* read error, skip block & hope for the best */ + ext3_error(sb, __FUNCTION__, "reading directory #%lu " + "offset %lu", dir->i_ino, block); + brelse(bh); + goto next; + } + i = search_dirblock(bh, dir, dentry, + block << EXT3_BLOCK_SIZE_BITS(sb), res_dir); + if (i == 1) { + EXT3_I(dir)->i_dir_start_lookup = block; + ret = bh; + goto cleanup_and_exit; + } else { + brelse(bh); + if (i < 0) + goto cleanup_and_exit; + } + next: + if (++block >= nblocks) + block = 0; + } while (block != start); + + /* + * If the directory has grown while we were searching, then + * search the last part of the directory before giving up. + */ + block = nblocks; + nblocks = dir->i_size >> EXT3_BLOCK_SIZE_BITS(sb); + if (block < nblocks) { + start = 0; + goto restart; + } + +cleanup_and_exit: + /* Clean up the read-ahead blocks */ + for (; ra_ptr < ra_max; ra_ptr++) + brelse (bh_use[ra_ptr]); + return ret; +} + +#ifdef CONFIG_EXT3_INDEX +static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, + struct ext3_dir_entry_2 **res_dir, int *err) +{ + struct super_block * sb; + struct dx_hash_info hinfo; + u32 hash; + struct dx_frame frames[2], *frame; + struct ext3_dir_entry_2 *de, *top; + struct buffer_head *bh; + unsigned long block; + int retval; + int namelen = dentry->d_name.len; + const u8 *name = dentry->d_name.name; + struct inode *dir = dentry->d_parent->d_inode; + + sb = dir->i_sb; + /* NFS may look up ".." - look at dx_root directory block */ + if (namelen > 2 || name[0] != '.'||(name[1] != '.' && name[1] != '\0')){ + if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err))) + return NULL; + } else { + frame = frames; + frame->bh = NULL; /* for dx_release() */ + frame->at = (struct dx_entry *)frames; /* hack for zero entry*/ + dx_set_block(frame->at, 0); /* dx_root block is 0 */ + } + hash = hinfo.hash; + do { + block = dx_get_block(frame->at); + if (!(bh = ext3_bread (NULL,dir, block, 0, err))) + goto errout; + de = (struct ext3_dir_entry_2 *) bh->b_data; + top = (struct ext3_dir_entry_2 *) ((char *) de + sb->s_blocksize - + EXT3_DIR_REC_LEN(0)); + for (; de < top; de = ext3_next_entry(de)) + if (ext3_match (namelen, name, de)) { + if (!ext3_check_dir_entry("ext3_find_entry", + dir, de, bh, + (block<b_data))) { + brelse (bh); + goto errout; + } + *res_dir = de; + dx_release (frames); + return bh; + } + brelse (bh); + /* Check to see if we should continue to search */ + retval = ext3_htree_next_block(dir, hash, frame, + frames, NULL); + if (retval < 0) { + ext3_warning(sb, __FUNCTION__, + "error reading index page in directory #%lu", + dir->i_ino); + *err = retval; + goto errout; + } + } while (retval == 1); + + *err = -ENOENT; +errout: + dxtrace(printk("%s not found\n", name)); + dx_release (frames); + return NULL; +} +#endif + +static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +{ + struct inode * inode; + struct ext3_dir_entry_2 * de; + struct buffer_head * bh; + + if (dentry->d_name.len > EXT3_NAME_LEN) + return ERR_PTR(-ENAMETOOLONG); + + bh = ext3_find_entry(dentry, &de); + inode = NULL; + if (bh) { + unsigned long ino = le32_to_cpu(de->inode); + brelse (bh); + if (!ext3_valid_inum(dir->i_sb, ino)) { + ext3_error(dir->i_sb, "ext3_lookup", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(dir->i_sb, ino); + + if (!inode) + return ERR_PTR(-EACCES); + } + return d_splice_alias(inode, dentry); +} + + +struct dentry *ext3_get_parent(struct dentry *child) +{ + unsigned long ino; + struct dentry *parent; + struct inode *inode; + struct dentry dotdot; + struct ext3_dir_entry_2 * de; + struct buffer_head *bh; + + dotdot.d_name.name = ".."; + dotdot.d_name.len = 2; + dotdot.d_parent = child; /* confusing, isn't it! */ + + bh = ext3_find_entry(&dotdot, &de); + inode = NULL; + if (!bh) + return ERR_PTR(-ENOENT); + ino = le32_to_cpu(de->inode); + brelse(bh); + + if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { + ext3_error(child->d_inode->i_sb, "ext3_get_parent", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(child->d_inode->i_sb, ino); + + if (!inode) + return ERR_PTR(-EACCES); + + parent = d_alloc_anon(inode); + if (!parent) { + iput(inode); + parent = ERR_PTR(-ENOMEM); + } + return parent; +} + +#define S_SHIFT 12 +static unsigned char ext3_type_by_mode[S_IFMT >> S_SHIFT] = { + [S_IFREG >> S_SHIFT] = EXT3_FT_REG_FILE, + [S_IFDIR >> S_SHIFT] = EXT3_FT_DIR, + [S_IFCHR >> S_SHIFT] = EXT3_FT_CHRDEV, + [S_IFBLK >> S_SHIFT] = EXT3_FT_BLKDEV, + [S_IFIFO >> S_SHIFT] = EXT3_FT_FIFO, + [S_IFSOCK >> S_SHIFT] = EXT3_FT_SOCK, + [S_IFLNK >> S_SHIFT] = EXT3_FT_SYMLINK, +}; + +static inline void ext3_set_de_type(struct super_block *sb, + struct ext3_dir_entry_2 *de, + umode_t mode) { + if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_FILETYPE)) + de->file_type = ext3_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; +} + +#ifdef CONFIG_EXT3_INDEX +static struct ext3_dir_entry_2 * +dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count) +{ + unsigned rec_len = 0; + + while (count--) { + struct ext3_dir_entry_2 *de = (struct ext3_dir_entry_2 *) (from + map->offs); + rec_len = EXT3_DIR_REC_LEN(de->name_len); + memcpy (to, de, rec_len); + ((struct ext3_dir_entry_2 *) to)->rec_len = + cpu_to_le16(rec_len); + de->inode = 0; + map++; + to += rec_len; + } + return (struct ext3_dir_entry_2 *) (to - rec_len); +} + +static struct ext3_dir_entry_2* dx_pack_dirents(char *base, int size) +{ + struct ext3_dir_entry_2 *next, *to, *prev, *de = (struct ext3_dir_entry_2 *) base; + unsigned rec_len = 0; + + prev = to = de; + while ((char*)de < base + size) { + next = (struct ext3_dir_entry_2 *) ((char *) de + + le16_to_cpu(de->rec_len)); + if (de->inode && de->name_len) { + rec_len = EXT3_DIR_REC_LEN(de->name_len); + if (de > to) + memmove(to, de, rec_len); + to->rec_len = cpu_to_le16(rec_len); + prev = to; + to = (struct ext3_dir_entry_2 *) (((char *) to) + rec_len); + } + de = next; + } + return prev; +} + +static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, + struct buffer_head **bh,struct dx_frame *frame, + struct dx_hash_info *hinfo, int *error) +{ + unsigned blocksize = dir->i_sb->s_blocksize; + unsigned count, continued; + struct buffer_head *bh2; + u32 newblock; + u32 hash2; + struct dx_map_entry *map; + char *data1 = (*bh)->b_data, *data2; + unsigned split; + struct ext3_dir_entry_2 *de = NULL, *de2; + int err; + + bh2 = ext3_append (handle, dir, &newblock, error); + if (!(bh2)) { + brelse(*bh); + *bh = NULL; + goto errout; + } + + BUFFER_TRACE(*bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, *bh); + if (err) { + journal_error: + brelse(*bh); + brelse(bh2); + *bh = NULL; + ext3_std_error(dir->i_sb, err); + goto errout; + } + BUFFER_TRACE(frame->bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, frame->bh); + if (err) + goto journal_error; + + data2 = bh2->b_data; + + /* create map in the end of data2 block */ + map = (struct dx_map_entry *) (data2 + blocksize); + count = dx_make_map ((struct ext3_dir_entry_2 *) data1, + blocksize, hinfo, map); + map -= count; + split = count/2; // need to adjust to actual middle + dx_sort_map (map, count); + hash2 = map[split].hash; + continued = hash2 == map[split - 1].hash; + dxtrace(printk("Split block %i at %x, %i/%i\n", + dx_get_block(frame->at), hash2, split, count-split)); + + /* Fancy dance to stay within two buffers */ + de2 = dx_move_dirents(data1, data2, map + split, count - split); + de = dx_pack_dirents(data1,blocksize); + de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); + de2->rec_len = cpu_to_le16(data2 + blocksize - (char *) de2); + dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data1, blocksize, 1)); + dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data2, blocksize, 1)); + + /* Which block gets the new entry? */ + if (hinfo->hash >= hash2) + { + swap(*bh, bh2); + de = de2; + } + dx_insert_block (frame, hash2 + continued, newblock); + err = ext3_journal_dirty_metadata (handle, bh2); + if (err) + goto journal_error; + err = ext3_journal_dirty_metadata (handle, frame->bh); + if (err) + goto journal_error; + brelse (bh2); + dxtrace(dx_show_index ("frame", frame->entries)); +errout: + return de; +} +#endif + + +/* + * Add a new entry into a directory (leaf) block. If de is non-NULL, + * it points to a directory entry which is guaranteed to be large + * enough for new directory entry. If de is NULL, then + * add_dirent_to_buf will attempt search the directory block for + * space. It will return -ENOSPC if no space is available, and -EIO + * and -EEXIST if directory entry already exists. + * + * NOTE! bh is NOT released in the case where ENOSPC is returned. In + * all other cases bh is released. + */ +static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, + struct inode *inode, struct ext3_dir_entry_2 *de, + struct buffer_head * bh) +{ + struct inode *dir = dentry->d_parent->d_inode; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + unsigned long offset = 0; + unsigned short reclen; + int nlen, rlen, err; + char *top; + + reclen = EXT3_DIR_REC_LEN(namelen); + if (!de) { + de = (struct ext3_dir_entry_2 *)bh->b_data; + top = bh->b_data + dir->i_sb->s_blocksize - reclen; + while ((char *) de <= top) { + if (!ext3_check_dir_entry("ext3_add_entry", dir, de, + bh, offset)) { + brelse (bh); + return -EIO; + } + if (ext3_match (namelen, name, de)) { + brelse (bh); + return -EEXIST; + } + nlen = EXT3_DIR_REC_LEN(de->name_len); + rlen = le16_to_cpu(de->rec_len); + if ((de->inode? rlen - nlen: rlen) >= reclen) + break; + de = (struct ext3_dir_entry_2 *)((char *)de + rlen); + offset += rlen; + } + if ((char *) de > top) + return -ENOSPC; + } + BUFFER_TRACE(bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, bh); + if (err) { + ext3_std_error(dir->i_sb, err); + brelse(bh); + return err; + } + + /* By now the buffer is marked for journaling */ + nlen = EXT3_DIR_REC_LEN(de->name_len); + rlen = le16_to_cpu(de->rec_len); + if (de->inode) { + struct ext3_dir_entry_2 *de1 = (struct ext3_dir_entry_2 *)((char *)de + nlen); + de1->rec_len = cpu_to_le16(rlen - nlen); + de->rec_len = cpu_to_le16(nlen); + de = de1; + } + de->file_type = EXT3_FT_UNKNOWN; + if (inode) { + de->inode = cpu_to_le32(inode->i_ino); + ext3_set_de_type(dir->i_sb, de, inode->i_mode); + } else + de->inode = 0; + de->name_len = namelen; + memcpy (de->name, name, namelen); + /* + * XXX shouldn't update any times until successful + * completion of syscall, but too many callers depend + * on this. + * + * XXX similarly, too many callers depend on + * ext3_new_inode() setting the times, but error + * recovery deletes the inode, so the worst that can + * happen is that the times are slightly out of date + * and/or different from the directory change time. + */ + dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC; + ext3_update_dx_flag(dir); + dir->i_version++; + ext3_mark_inode_dirty(handle, dir); + BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); + err = ext3_journal_dirty_metadata(handle, bh); + if (err) + ext3_std_error(dir->i_sb, err); + brelse(bh); + return 0; +} + +#ifdef CONFIG_EXT3_INDEX +/* + * This converts a one block unindexed directory to a 3 block indexed + * directory, and adds the dentry to the indexed directory. + */ +static int make_indexed_dir(handle_t *handle, struct dentry *dentry, + struct inode *inode, struct buffer_head *bh) +{ + struct inode *dir = dentry->d_parent->d_inode; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + struct buffer_head *bh2; + struct dx_root *root; + struct dx_frame frames[2], *frame; + struct dx_entry *entries; + struct ext3_dir_entry_2 *de, *de2; + char *data1, *top; + unsigned len; + int retval; + unsigned blocksize; + struct dx_hash_info hinfo; + u32 block; + struct fake_dirent *fde; + + blocksize = dir->i_sb->s_blocksize; + dxtrace(printk("Creating index\n")); + retval = ext3_journal_get_write_access(handle, bh); + if (retval) { + ext3_std_error(dir->i_sb, retval); + brelse(bh); + return retval; + } + root = (struct dx_root *) bh->b_data; + + bh2 = ext3_append (handle, dir, &block, &retval); + if (!(bh2)) { + brelse(bh); + return retval; + } + EXT3_I(dir)->i_flags |= EXT3_INDEX_FL; + data1 = bh2->b_data; + + /* The 0th block becomes the root, move the dirents out */ + fde = &root->dotdot; + de = (struct ext3_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len)); + len = ((char *) root) + blocksize - (char *) de; + memcpy (data1, de, len); + de = (struct ext3_dir_entry_2 *) data1; + top = data1 + len; + while ((char *)(de2=(void*)de+le16_to_cpu(de->rec_len)) < top) + de = de2; + de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); + /* Initialize the root; the dot dirents already exist */ + de = (struct ext3_dir_entry_2 *) (&root->dotdot); + de->rec_len = cpu_to_le16(blocksize - EXT3_DIR_REC_LEN(2)); + memset (&root->info, 0, sizeof(root->info)); + root->info.info_length = sizeof(root->info); + root->info.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; + entries = root->entries; + dx_set_block (entries, 1); + dx_set_count (entries, 1); + dx_set_limit (entries, dx_root_limit(dir, sizeof(root->info))); + + /* Initialize as for dx_probe */ + hinfo.hash_version = root->info.hash_version; + hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; + ext3fs_dirhash(name, namelen, &hinfo); + frame = frames; + frame->entries = entries; + frame->at = entries; + frame->bh = bh; + bh = bh2; + de = do_split(handle,dir, &bh, frame, &hinfo, &retval); + dx_release (frames); + if (!(de)) + return retval; + + return add_dirent_to_buf(handle, dentry, inode, de, bh); +} +#endif + +/* + * ext3_add_entry() + * + * adds a file entry to the specified directory, using the same + * semantics as ext3_find_entry(). It returns NULL if it failed. + * + * NOTE!! The inode part of 'de' is left at 0 - which means you + * may not sleep between calling this and putting something into + * the entry, as someone else might have used it while you slept. + */ +static int ext3_add_entry (handle_t *handle, struct dentry *dentry, + struct inode *inode) +{ + struct inode *dir = dentry->d_parent->d_inode; + unsigned long offset; + struct buffer_head * bh; + struct ext3_dir_entry_2 *de; + struct super_block * sb; + int retval; +#ifdef CONFIG_EXT3_INDEX + int dx_fallback=0; +#endif + unsigned blocksize; + u32 block, blocks; + + sb = dir->i_sb; + blocksize = sb->s_blocksize; + if (!dentry->d_name.len) + return -EINVAL; +#ifdef CONFIG_EXT3_INDEX + if (is_dx(dir)) { + retval = ext3_dx_add_entry(handle, dentry, inode); + if (!retval || (retval != ERR_BAD_DX_DIR)) + return retval; + EXT3_I(dir)->i_flags &= ~EXT3_INDEX_FL; + dx_fallback++; + ext3_mark_inode_dirty(handle, dir); + } +#endif + blocks = dir->i_size >> sb->s_blocksize_bits; + for (block = 0, offset = 0; block < blocks; block++) { + bh = ext3_bread(handle, dir, block, 0, &retval); + if(!bh) + return retval; + retval = add_dirent_to_buf(handle, dentry, inode, NULL, bh); + if (retval != -ENOSPC) + return retval; + +#ifdef CONFIG_EXT3_INDEX + if (blocks == 1 && !dx_fallback && + EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_DIR_INDEX)) + return make_indexed_dir(handle, dentry, inode, bh); +#endif + brelse(bh); + } + bh = ext3_append(handle, dir, &block, &retval); + if (!bh) + return retval; + de = (struct ext3_dir_entry_2 *) bh->b_data; + de->inode = 0; + de->rec_len = cpu_to_le16(blocksize); + return add_dirent_to_buf(handle, dentry, inode, de, bh); +} + +#ifdef CONFIG_EXT3_INDEX +/* + * Returns 0 for success, or a negative error value + */ +static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, + struct inode *inode) +{ + struct dx_frame frames[2], *frame; + struct dx_entry *entries, *at; + struct dx_hash_info hinfo; + struct buffer_head * bh; + struct inode *dir = dentry->d_parent->d_inode; + struct super_block * sb = dir->i_sb; + struct ext3_dir_entry_2 *de; + int err; + + frame = dx_probe(dentry, NULL, &hinfo, frames, &err); + if (!frame) + return err; + entries = frame->entries; + at = frame->at; + + if (!(bh = ext3_bread(handle,dir, dx_get_block(frame->at), 0, &err))) + goto cleanup; + + BUFFER_TRACE(bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, bh); + if (err) + goto journal_error; + + err = add_dirent_to_buf(handle, dentry, inode, NULL, bh); + if (err != -ENOSPC) { + bh = NULL; + goto cleanup; + } + + /* Block full, should compress but for now just split */ + dxtrace(printk("using %u of %u node entries\n", + dx_get_count(entries), dx_get_limit(entries))); + /* Need to split index? */ + if (dx_get_count(entries) == dx_get_limit(entries)) { + u32 newblock; + unsigned icount = dx_get_count(entries); + int levels = frame - frames; + struct dx_entry *entries2; + struct dx_node *node2; + struct buffer_head *bh2; + + if (levels && (dx_get_count(frames->entries) == + dx_get_limit(frames->entries))) { + ext3_warning(sb, __FUNCTION__, + "Directory index full!"); + err = -ENOSPC; + goto cleanup; + } + bh2 = ext3_append (handle, dir, &newblock, &err); + if (!(bh2)) + goto cleanup; + node2 = (struct dx_node *)(bh2->b_data); + entries2 = node2->entries; + node2->fake.rec_len = cpu_to_le16(sb->s_blocksize); + node2->fake.inode = 0; + BUFFER_TRACE(frame->bh, "get_write_access"); + err = ext3_journal_get_write_access(handle, frame->bh); + if (err) + goto journal_error; + if (levels) { + unsigned icount1 = icount/2, icount2 = icount - icount1; + unsigned hash2 = dx_get_hash(entries + icount1); + dxtrace(printk("Split index %i/%i\n", icount1, icount2)); + + BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */ + err = ext3_journal_get_write_access(handle, + frames[0].bh); + if (err) + goto journal_error; + + memcpy ((char *) entries2, (char *) (entries + icount1), + icount2 * sizeof(struct dx_entry)); + dx_set_count (entries, icount1); + dx_set_count (entries2, icount2); + dx_set_limit (entries2, dx_node_limit(dir)); + + /* Which index block gets the new entry? */ + if (at - entries >= icount1) { + frame->at = at = at - entries - icount1 + entries2; + frame->entries = entries = entries2; + swap(frame->bh, bh2); + } + dx_insert_block (frames + 0, hash2, newblock); + dxtrace(dx_show_index ("node", frames[1].entries)); + dxtrace(dx_show_index ("node", + ((struct dx_node *) bh2->b_data)->entries)); + err = ext3_journal_dirty_metadata(handle, bh2); + if (err) + goto journal_error; + brelse (bh2); + } else { + dxtrace(printk("Creating second level index...\n")); + memcpy((char *) entries2, (char *) entries, + icount * sizeof(struct dx_entry)); + dx_set_limit(entries2, dx_node_limit(dir)); + + /* Set up root */ + dx_set_count(entries, 1); + dx_set_block(entries + 0, newblock); + ((struct dx_root *) frames[0].bh->b_data)->info.indirect_levels = 1; + + /* Add new access path frame */ + frame = frames + 1; + frame->at = at = at - entries + entries2; + frame->entries = entries = entries2; + frame->bh = bh2; + err = ext3_journal_get_write_access(handle, + frame->bh); + if (err) + goto journal_error; + } + ext3_journal_dirty_metadata(handle, frames[0].bh); + } + de = do_split(handle, dir, &bh, frame, &hinfo, &err); + if (!de) + goto cleanup; + err = add_dirent_to_buf(handle, dentry, inode, de, bh); + bh = NULL; + goto cleanup; + +journal_error: + ext3_std_error(dir->i_sb, err); +cleanup: + if (bh) + brelse(bh); + dx_release(frames); + return err; +} +#endif + +/* + * ext3_delete_entry deletes a directory entry by merging it with the + * previous entry + */ +static int ext3_delete_entry (handle_t *handle, + struct inode * dir, + struct ext3_dir_entry_2 * de_del, + struct buffer_head * bh) +{ + struct ext3_dir_entry_2 * de, * pde; + int i; + + i = 0; + pde = NULL; + de = (struct ext3_dir_entry_2 *) bh->b_data; + while (i < bh->b_size) { + if (!ext3_check_dir_entry("ext3_delete_entry", dir, de, bh, i)) + return -EIO; + if (de == de_del) { + BUFFER_TRACE(bh, "get_write_access"); + ext3_journal_get_write_access(handle, bh); + if (pde) + pde->rec_len = + cpu_to_le16(le16_to_cpu(pde->rec_len) + + le16_to_cpu(de->rec_len)); + else + de->inode = 0; + dir->i_version++; + BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, bh); + return 0; + } + i += le16_to_cpu(de->rec_len); + pde = de; + de = (struct ext3_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + } + return -ENOENT; +} + +/* + * ext3_mark_inode_dirty is somewhat expensive, so unlike ext2 we + * do not perform it in these functions. We perform it at the call site, + * if it is needed. + */ +static inline void ext3_inc_count(handle_t *handle, struct inode *inode) +{ + inc_nlink(inode); +} + +static inline void ext3_dec_count(handle_t *handle, struct inode *inode) +{ + drop_nlink(inode); +} + +static int ext3_add_nondir(handle_t *handle, + struct dentry *dentry, struct inode *inode) +{ + int err = ext3_add_entry(handle, dentry, inode); + if (!err) { + ext3_mark_inode_dirty(handle, inode); + d_instantiate(dentry, inode); + return 0; + } + ext3_dec_count(handle, inode); + iput(inode); + return err; +} + +/* + * By the time this is called, we already have created + * the directory cache entry for the new file, but it + * is so far negative - it has no inode. + * + * If the create succeeds, we fill in the inode information + * with d_instantiate(). + */ +static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, + struct nameidata *nd) +{ + handle_t *handle; + struct inode * inode; + int err, retries = 0; + +retry: + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext3_new_inode (handle, dir, mode); + err = PTR_ERR(inode); + if (!IS_ERR(inode)) { + inode->i_op = &ext3_file_inode_operations; + inode->i_fop = &ext3_file_operations; + ext3_set_aops(inode); + err = ext3_add_nondir(handle, dentry, inode); + } + ext3_journal_stop(handle); + if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +static int ext3_mknod (struct inode * dir, struct dentry *dentry, + int mode, dev_t rdev) +{ + handle_t *handle; + struct inode *inode; + int err, retries = 0; + + if (!new_valid_dev(rdev)) + return -EINVAL; + +retry: + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext3_new_inode (handle, dir, mode); + err = PTR_ERR(inode); + if (!IS_ERR(inode)) { + init_special_inode(inode, inode->i_mode, rdev); +#ifdef CONFIG_EXT3_FS_XATTR + inode->i_op = &ext3_special_inode_operations; +#endif + err = ext3_add_nondir(handle, dentry, inode); + } + ext3_journal_stop(handle); + if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode) +{ + handle_t *handle; + struct inode * inode; + struct buffer_head * dir_block; + struct ext3_dir_entry_2 * de; + int err, retries = 0; + + if (dir->i_nlink >= EXT3_LINK_MAX) + return -EMLINK; + +retry: + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext3_new_inode (handle, dir, S_IFDIR | mode); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out_stop; + + inode->i_op = &ext3_dir_inode_operations; + inode->i_fop = &ext3_dir_operations; + inode->i_size = EXT3_I(inode)->i_disksize = inode->i_sb->s_blocksize; + dir_block = ext3_bread (handle, inode, 0, 1, &err); + if (!dir_block) { + drop_nlink(inode); /* is this nlink == 0? */ + ext3_mark_inode_dirty(handle, inode); + iput (inode); + goto out_stop; + } + BUFFER_TRACE(dir_block, "get_write_access"); + ext3_journal_get_write_access(handle, dir_block); + de = (struct ext3_dir_entry_2 *) dir_block->b_data; + de->inode = cpu_to_le32(inode->i_ino); + de->name_len = 1; + de->rec_len = cpu_to_le16(EXT3_DIR_REC_LEN(de->name_len)); + strcpy (de->name, "."); + ext3_set_de_type(dir->i_sb, de, S_IFDIR); + de = (struct ext3_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + de->inode = cpu_to_le32(dir->i_ino); + de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT3_DIR_REC_LEN(1)); + de->name_len = 2; + strcpy (de->name, ".."); + ext3_set_de_type(dir->i_sb, de, S_IFDIR); + inode->i_nlink = 2; + BUFFER_TRACE(dir_block, "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, dir_block); + brelse (dir_block); + ext3_mark_inode_dirty(handle, inode); + err = ext3_add_entry (handle, dentry, inode); + if (err) { + inode->i_nlink = 0; + ext3_mark_inode_dirty(handle, inode); + iput (inode); + goto out_stop; + } + inc_nlink(dir); + ext3_update_dx_flag(dir); + ext3_mark_inode_dirty(handle, dir); + d_instantiate(dentry, inode); +out_stop: + ext3_journal_stop(handle); + if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +/* + * routine to check that the specified directory is empty (for rmdir) + */ +static int empty_dir (struct inode * inode) +{ + unsigned long offset; + struct buffer_head * bh; + struct ext3_dir_entry_2 * de, * de1; + struct super_block * sb; + int err = 0; + + sb = inode->i_sb; + if (inode->i_size < EXT3_DIR_REC_LEN(1) + EXT3_DIR_REC_LEN(2) || + !(bh = ext3_bread (NULL, inode, 0, 0, &err))) { + if (err) + ext3_error(inode->i_sb, __FUNCTION__, + "error %d reading directory #%lu offset 0", + err, inode->i_ino); + else + ext3_warning(inode->i_sb, __FUNCTION__, + "bad directory (dir #%lu) - no data block", + inode->i_ino); + return 1; + } + de = (struct ext3_dir_entry_2 *) bh->b_data; + de1 = (struct ext3_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + if (le32_to_cpu(de->inode) != inode->i_ino || + !le32_to_cpu(de1->inode) || + strcmp (".", de->name) || + strcmp ("..", de1->name)) { + ext3_warning (inode->i_sb, "empty_dir", + "bad directory (dir #%lu) - no `.' or `..'", + inode->i_ino); + brelse (bh); + return 1; + } + offset = le16_to_cpu(de->rec_len) + le16_to_cpu(de1->rec_len); + de = (struct ext3_dir_entry_2 *) + ((char *) de1 + le16_to_cpu(de1->rec_len)); + while (offset < inode->i_size ) { + if (!bh || + (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { + err = 0; + brelse (bh); + bh = ext3_bread (NULL, inode, + offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err); + if (!bh) { + if (err) + ext3_error(sb, __FUNCTION__, + "error %d reading directory" + " #%lu offset %lu", + err, inode->i_ino, offset); + offset += sb->s_blocksize; + continue; + } + de = (struct ext3_dir_entry_2 *) bh->b_data; + } + if (!ext3_check_dir_entry("empty_dir", inode, de, bh, offset)) { + de = (struct ext3_dir_entry_2 *)(bh->b_data + + sb->s_blocksize); + offset = (offset | (sb->s_blocksize - 1)) + 1; + continue; + } + if (le32_to_cpu(de->inode)) { + brelse (bh); + return 0; + } + offset += le16_to_cpu(de->rec_len); + de = (struct ext3_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + } + brelse (bh); + return 1; +} + +/* ext3_orphan_add() links an unlinked or truncated inode into a list of + * such inodes, starting at the superblock, in case we crash before the + * file is closed/deleted, or in case the inode truncate spans multiple + * transactions and the last transaction is not recovered after a crash. + * + * At filesystem recovery time, we walk this list deleting unlinked + * inodes and truncating linked inodes in ext3_orphan_cleanup(). + */ +int ext3_orphan_add(handle_t *handle, struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct ext3_iloc iloc; + int err = 0, rc; + + lock_super(sb); + if (!list_empty(&EXT3_I(inode)->i_orphan)) + goto out_unlock; + + /* Orphan handling is only valid for files with data blocks + * being truncated, or files being unlinked. */ + + /* @@@ FIXME: Observation from aviro: + * I think I can trigger J_ASSERT in ext3_orphan_add(). We block + * here (on lock_super()), so race with ext3_link() which might bump + * ->i_nlink. For, say it, character device. Not a regular file, + * not a directory, not a symlink and ->i_nlink > 0. + */ + J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); + + BUFFER_TRACE(EXT3_SB(sb)->s_sbh, "get_write_access"); + err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh); + if (err) + goto out_unlock; + + err = ext3_reserve_inode_write(handle, inode, &iloc); + if (err) + goto out_unlock; + + /* Insert this inode at the head of the on-disk orphan list... */ + NEXT_ORPHAN(inode) = le32_to_cpu(EXT3_SB(sb)->s_es->s_last_orphan); + EXT3_SB(sb)->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); + err = ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + rc = ext3_mark_iloc_dirty(handle, inode, &iloc); + if (!err) + err = rc; + + /* Only add to the head of the in-memory list if all the + * previous operations succeeded. If the orphan_add is going to + * fail (possibly taking the journal offline), we can't risk + * leaving the inode on the orphan list: stray orphan-list + * entries can cause panics at unmount time. + * + * This is safe: on error we're going to ignore the orphan list + * anyway on the next recovery. */ + if (!err) + list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan); + + jbd_debug(4, "superblock will point to %lu\n", inode->i_ino); + jbd_debug(4, "orphan inode %lu will point to %d\n", + inode->i_ino, NEXT_ORPHAN(inode)); +out_unlock: + unlock_super(sb); + ext3_std_error(inode->i_sb, err); + return err; +} + +/* + * ext3_orphan_del() removes an unlinked or truncated inode from the list + * of such inodes stored on disk, because it is finally being cleaned up. + */ +int ext3_orphan_del(handle_t *handle, struct inode *inode) +{ + struct list_head *prev; + struct ext3_inode_info *ei = EXT3_I(inode); + struct ext3_sb_info *sbi; + unsigned long ino_next; + struct ext3_iloc iloc; + int err = 0; + + lock_super(inode->i_sb); + if (list_empty(&ei->i_orphan)) { + unlock_super(inode->i_sb); + return 0; + } + + ino_next = NEXT_ORPHAN(inode); + prev = ei->i_orphan.prev; + sbi = EXT3_SB(inode->i_sb); + + jbd_debug(4, "remove inode %lu from orphan list\n", inode->i_ino); + + list_del_init(&ei->i_orphan); + + /* If we're on an error path, we may not have a valid + * transaction handle with which to update the orphan list on + * disk, but we still need to remove the inode from the linked + * list in memory. */ + if (!handle) + goto out; + + err = ext3_reserve_inode_write(handle, inode, &iloc); + if (err) + goto out_err; + + if (prev == &sbi->s_orphan) { + jbd_debug(4, "superblock will point to %lu\n", ino_next); + BUFFER_TRACE(sbi->s_sbh, "get_write_access"); + err = ext3_journal_get_write_access(handle, sbi->s_sbh); + if (err) + goto out_brelse; + sbi->s_es->s_last_orphan = cpu_to_le32(ino_next); + err = ext3_journal_dirty_metadata(handle, sbi->s_sbh); + } else { + struct ext3_iloc iloc2; + struct inode *i_prev = + &list_entry(prev, struct ext3_inode_info, i_orphan)->vfs_inode; + + jbd_debug(4, "orphan inode %lu will point to %lu\n", + i_prev->i_ino, ino_next); + err = ext3_reserve_inode_write(handle, i_prev, &iloc2); + if (err) + goto out_brelse; + NEXT_ORPHAN(i_prev) = ino_next; + err = ext3_mark_iloc_dirty(handle, i_prev, &iloc2); + } + if (err) + goto out_brelse; + NEXT_ORPHAN(inode) = 0; + err = ext3_mark_iloc_dirty(handle, inode, &iloc); + +out_err: + ext3_std_error(inode->i_sb, err); +out: + unlock_super(inode->i_sb); + return err; + +out_brelse: + brelse(iloc.bh); + goto out_err; +} + +static int ext3_rmdir (struct inode * dir, struct dentry *dentry) +{ + int retval; + struct inode * inode; + struct buffer_head * bh; + struct ext3_dir_entry_2 * de; + handle_t *handle; + + /* Initialize quotas before so that eventual writes go in + * separate transaction */ + DQUOT_INIT(dentry->d_inode); + handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + retval = -ENOENT; + bh = ext3_find_entry (dentry, &de); + if (!bh) + goto end_rmdir; + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = dentry->d_inode; + + retval = -EIO; + if (le32_to_cpu(de->inode) != inode->i_ino) + goto end_rmdir; + + retval = -ENOTEMPTY; + if (!empty_dir (inode)) + goto end_rmdir; + + retval = ext3_delete_entry(handle, dir, de, bh); + if (retval) + goto end_rmdir; + if (inode->i_nlink != 2) + ext3_warning (inode->i_sb, "ext3_rmdir", + "empty directory has nlink!=2 (%d)", + inode->i_nlink); + inode->i_version++; + clear_nlink(inode); + /* There's no need to set i_disksize: the fact that i_nlink is + * zero will ensure that the right thing happens during any + * recovery. */ + inode->i_size = 0; + ext3_orphan_add(handle, inode); + inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; + ext3_mark_inode_dirty(handle, inode); + drop_nlink(dir); + ext3_update_dx_flag(dir); + ext3_mark_inode_dirty(handle, dir); + +end_rmdir: + ext3_journal_stop(handle); + brelse (bh); + return retval; +} + +static int ext3_unlink(struct inode * dir, struct dentry *dentry) +{ + int retval; + struct inode * inode; + struct buffer_head * bh; + struct ext3_dir_entry_2 * de; + handle_t *handle; + + /* Initialize quotas before so that eventual writes go + * in separate transaction */ + DQUOT_INIT(dentry->d_inode); + handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + retval = -ENOENT; + bh = ext3_find_entry (dentry, &de); + if (!bh) + goto end_unlink; + + inode = dentry->d_inode; + + retval = -EIO; + if (le32_to_cpu(de->inode) != inode->i_ino) + goto end_unlink; + + if (!inode->i_nlink) { + ext3_warning (inode->i_sb, "ext3_unlink", + "Deleting nonexistent file (%lu), %d", + inode->i_ino, inode->i_nlink); + inode->i_nlink = 1; + } + retval = ext3_delete_entry(handle, dir, de, bh); + if (retval) + goto end_unlink; + dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; + ext3_update_dx_flag(dir); + ext3_mark_inode_dirty(handle, dir); + drop_nlink(inode); + if (!inode->i_nlink) + ext3_orphan_add(handle, inode); + inode->i_ctime = dir->i_ctime; + ext3_mark_inode_dirty(handle, inode); + retval = 0; + +end_unlink: + ext3_journal_stop(handle); + brelse (bh); + return retval; +} + +static int ext3_symlink (struct inode * dir, + struct dentry *dentry, const char * symname) +{ + handle_t *handle; + struct inode * inode; + int l, err, retries = 0; + + l = strlen(symname)+1; + if (l > dir->i_sb->s_blocksize) + return -ENAMETOOLONG; + +retry: + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 + + 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext3_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out_stop; + + if (l > sizeof (EXT3_I(inode)->i_data)) { + inode->i_op = &ext3_symlink_inode_operations; + ext3_set_aops(inode); + /* + * page_symlink() calls into ext3_prepare/commit_write. + * We have a transaction open. All is sweetness. It also sets + * i_size in generic_commit_write(). + */ + err = __page_symlink(inode, symname, l, + mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS); + if (err) { + ext3_dec_count(handle, inode); + ext3_mark_inode_dirty(handle, inode); + iput (inode); + goto out_stop; + } + } else { + inode->i_op = &ext3_fast_symlink_inode_operations; + memcpy((char*)&EXT3_I(inode)->i_data,symname,l); + inode->i_size = l-1; + } + EXT3_I(inode)->i_disksize = inode->i_size; + err = ext3_add_nondir(handle, dentry, inode); +out_stop: + ext3_journal_stop(handle); + if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +static int ext3_link (struct dentry * old_dentry, + struct inode * dir, struct dentry *dentry) +{ + handle_t *handle; + struct inode *inode = old_dentry->d_inode; + int err, retries = 0; + + if (inode->i_nlink >= EXT3_LINK_MAX) + return -EMLINK; + +retry: + handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode->i_ctime = CURRENT_TIME_SEC; + ext3_inc_count(handle, inode); + atomic_inc(&inode->i_count); + + err = ext3_add_nondir(handle, dentry, inode); + ext3_journal_stop(handle); + if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +#define PARENT_INO(buffer) \ + ((struct ext3_dir_entry_2 *) ((char *) buffer + \ + le16_to_cpu(((struct ext3_dir_entry_2 *) buffer)->rec_len)))->inode + +/* + * Anybody can rename anything with this: the permission checks are left to the + * higher-level routines. + */ +static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, + struct inode * new_dir,struct dentry *new_dentry) +{ + handle_t *handle; + struct inode * old_inode, * new_inode; + struct buffer_head * old_bh, * new_bh, * dir_bh; + struct ext3_dir_entry_2 * old_de, * new_de; + int retval; + + old_bh = new_bh = dir_bh = NULL; + + /* Initialize quotas before so that eventual writes go + * in separate transaction */ + if (new_dentry->d_inode) + DQUOT_INIT(new_dentry->d_inode); + handle = ext3_journal_start(old_dir, 2 * + EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) + handle->h_sync = 1; + + old_bh = ext3_find_entry (old_dentry, &old_de); + /* + * Check for inode number is _not_ due to possible IO errors. + * We might rmdir the source, keep it as pwd of some process + * and merrily kill the link to whatever was created under the + * same name. Goodbye sticky bit ;-< + */ + old_inode = old_dentry->d_inode; + retval = -ENOENT; + if (!old_bh || le32_to_cpu(old_de->inode) != old_inode->i_ino) + goto end_rename; + + new_inode = new_dentry->d_inode; + new_bh = ext3_find_entry (new_dentry, &new_de); + if (new_bh) { + if (!new_inode) { + brelse (new_bh); + new_bh = NULL; + } + } + if (S_ISDIR(old_inode->i_mode)) { + if (new_inode) { + retval = -ENOTEMPTY; + if (!empty_dir (new_inode)) + goto end_rename; + } + retval = -EIO; + dir_bh = ext3_bread (handle, old_inode, 0, 0, &retval); + if (!dir_bh) + goto end_rename; + if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) + goto end_rename; + retval = -EMLINK; + if (!new_inode && new_dir!=old_dir && + new_dir->i_nlink >= EXT3_LINK_MAX) + goto end_rename; + } + if (!new_bh) { + retval = ext3_add_entry (handle, new_dentry, old_inode); + if (retval) + goto end_rename; + } else { + BUFFER_TRACE(new_bh, "get write access"); + ext3_journal_get_write_access(handle, new_bh); + new_de->inode = cpu_to_le32(old_inode->i_ino); + if (EXT3_HAS_INCOMPAT_FEATURE(new_dir->i_sb, + EXT3_FEATURE_INCOMPAT_FILETYPE)) + new_de->file_type = old_de->file_type; + new_dir->i_version++; + BUFFER_TRACE(new_bh, "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, new_bh); + brelse(new_bh); + new_bh = NULL; + } + + /* + * Like most other Unix systems, set the ctime for inodes on a + * rename. + */ + old_inode->i_ctime = CURRENT_TIME_SEC; + ext3_mark_inode_dirty(handle, old_inode); + + /* + * ok, that's it + */ + if (le32_to_cpu(old_de->inode) != old_inode->i_ino || + old_de->name_len != old_dentry->d_name.len || + strncmp(old_de->name, old_dentry->d_name.name, old_de->name_len) || + (retval = ext3_delete_entry(handle, old_dir, + old_de, old_bh)) == -ENOENT) { + /* old_de could have moved from under us during htree split, so + * make sure that we are deleting the right entry. We might + * also be pointing to a stale entry in the unused part of + * old_bh so just checking inum and the name isn't enough. */ + struct buffer_head *old_bh2; + struct ext3_dir_entry_2 *old_de2; + + old_bh2 = ext3_find_entry(old_dentry, &old_de2); + if (old_bh2) { + retval = ext3_delete_entry(handle, old_dir, + old_de2, old_bh2); + brelse(old_bh2); + } + } + if (retval) { + ext3_warning(old_dir->i_sb, "ext3_rename", + "Deleting old file (%lu), %d, error=%d", + old_dir->i_ino, old_dir->i_nlink, retval); + } + + if (new_inode) { + drop_nlink(new_inode); + new_inode->i_ctime = CURRENT_TIME_SEC; + } + old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC; + ext3_update_dx_flag(old_dir); + if (dir_bh) { + BUFFER_TRACE(dir_bh, "get_write_access"); + ext3_journal_get_write_access(handle, dir_bh); + PARENT_INO(dir_bh->b_data) = cpu_to_le32(new_dir->i_ino); + BUFFER_TRACE(dir_bh, "call ext3_journal_dirty_metadata"); + ext3_journal_dirty_metadata(handle, dir_bh); + drop_nlink(old_dir); + if (new_inode) { + drop_nlink(new_inode); + } else { + inc_nlink(new_dir); + ext3_update_dx_flag(new_dir); + ext3_mark_inode_dirty(handle, new_dir); + } + } + ext3_mark_inode_dirty(handle, old_dir); + if (new_inode) { + ext3_mark_inode_dirty(handle, new_inode); + if (!new_inode->i_nlink) + ext3_orphan_add(handle, new_inode); + } + retval = 0; + +end_rename: + brelse (dir_bh); + brelse (old_bh); + brelse (new_bh); + ext3_journal_stop(handle); + return retval; +} + +/* + * directories can handle most operations... + */ +struct inode_operations ext3_dir_inode_operations = { + .create = ext3_create, + .lookup = ext3_lookup, + .link = ext3_link, + .unlink = ext3_unlink, + .symlink = ext3_symlink, + .mkdir = ext3_mkdir, + .rmdir = ext3_rmdir, + .mknod = ext3_mknod, + .rename = ext3_rename, + .setattr = ext3_setattr, +#ifdef CONFIG_EXT3_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext3_listxattr, + .removexattr = generic_removexattr, +#endif + .permission = ext3_permission, +}; + +struct inode_operations ext3_special_inode_operations = { + .setattr = ext3_setattr, +#ifdef CONFIG_EXT3_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext3_listxattr, + .removexattr = generic_removexattr, +#endif + .permission = ext3_permission, +}; diff --git a/fs/ext4/namei.h b/fs/ext4/namei.h new file mode 100644 index 0000000..f2ce2b0 --- /dev/null +++ b/fs/ext4/namei.h @@ -0,0 +1,8 @@ +/* linux/fs/ext3/namei.h + * + * Copyright (C) 2005 Simtec Electronics + * Ben Dooks + * +*/ + +extern struct dentry *ext3_get_parent(struct dentry *child); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c new file mode 100644 index 0000000..b73cba1 --- /dev/null +++ b/fs/ext4/resize.c @@ -0,0 +1,1042 @@ +/* + * linux/fs/ext3/resize.c + * + * Support for resizing an ext3 filesystem while it is mounted. + * + * Copyright (C) 2001, 2002 Andreas Dilger + * + * This could probably be made into a module, because it is not often in use. + */ + + +#define EXT3FS_DEBUG + +#include +#include +#include + +#include +#include + + +#define outside(b, first, last) ((b) < (first) || (b) >= (last)) +#define inside(b, first, last) ((b) >= (first) && (b) < (last)) + +static int verify_group_input(struct super_block *sb, + struct ext3_new_group_data *input) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext3_super_block *es = sbi->s_es; + ext3_fsblk_t start = le32_to_cpu(es->s_blocks_count); + ext3_fsblk_t end = start + input->blocks_count; + unsigned group = input->group; + ext3_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; + unsigned overhead = ext3_bg_has_super(sb, group) ? + (1 + ext3_bg_num_gdb(sb, group) + + le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + ext3_fsblk_t metaend = start + overhead; + struct buffer_head *bh = NULL; + ext3_grpblk_t free_blocks_count; + int err = -EINVAL; + + input->free_blocks_count = free_blocks_count = + input->blocks_count - 2 - overhead - sbi->s_itb_per_group; + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT3-fs: adding %s group %u: %u blocks " + "(%d free, %u reserved)\n", + ext3_bg_has_super(sb, input->group) ? "normal" : + "no-super", input->group, input->blocks_count, + free_blocks_count, input->reserved_blocks); + + if (group != sbi->s_groups_count) + ext3_warning(sb, __FUNCTION__, + "Cannot add at group %u (only %lu groups)", + input->group, sbi->s_groups_count); + else if ((start - le32_to_cpu(es->s_first_data_block)) % + EXT3_BLOCKS_PER_GROUP(sb)) + ext3_warning(sb, __FUNCTION__, "Last group not full"); + else if (input->reserved_blocks > input->blocks_count / 5) + ext3_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", + input->reserved_blocks); + else if (free_blocks_count < 0) + ext3_warning(sb, __FUNCTION__, "Bad blocks count %u", + input->blocks_count); + else if (!(bh = sb_bread(sb, end - 1))) + ext3_warning(sb, __FUNCTION__, + "Cannot read last block ("E3FSBLK")", + end - 1); + else if (outside(input->block_bitmap, start, end)) + ext3_warning(sb, __FUNCTION__, + "Block bitmap not in group (block %u)", + input->block_bitmap); + else if (outside(input->inode_bitmap, start, end)) + ext3_warning(sb, __FUNCTION__, + "Inode bitmap not in group (block %u)", + input->inode_bitmap); + else if (outside(input->inode_table, start, end) || + outside(itend - 1, start, end)) + ext3_warning(sb, __FUNCTION__, + "Inode table not in group (blocks %u-"E3FSBLK")", + input->inode_table, itend - 1); + else if (input->inode_bitmap == input->block_bitmap) + ext3_warning(sb, __FUNCTION__, + "Block bitmap same as inode bitmap (%u)", + input->block_bitmap); + else if (inside(input->block_bitmap, input->inode_table, itend)) + ext3_warning(sb, __FUNCTION__, + "Block bitmap (%u) in inode table (%u-"E3FSBLK")", + input->block_bitmap, input->inode_table, itend-1); + else if (inside(input->inode_bitmap, input->inode_table, itend)) + ext3_warning(sb, __FUNCTION__, + "Inode bitmap (%u) in inode table (%u-"E3FSBLK")", + input->inode_bitmap, input->inode_table, itend-1); + else if (inside(input->block_bitmap, start, metaend)) + ext3_warning(sb, __FUNCTION__, + "Block bitmap (%u) in GDT table" + " ("E3FSBLK"-"E3FSBLK")", + input->block_bitmap, start, metaend - 1); + else if (inside(input->inode_bitmap, start, metaend)) + ext3_warning(sb, __FUNCTION__, + "Inode bitmap (%u) in GDT table" + " ("E3FSBLK"-"E3FSBLK")", + input->inode_bitmap, start, metaend - 1); + else if (inside(input->inode_table, start, metaend) || + inside(itend - 1, start, metaend)) + ext3_warning(sb, __FUNCTION__, + "Inode table (%u-"E3FSBLK") overlaps" + "GDT table ("E3FSBLK"-"E3FSBLK")", + input->inode_table, itend - 1, start, metaend - 1); + else + err = 0; + brelse(bh); + + return err; +} + +static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, + ext3_fsblk_t blk) +{ + struct buffer_head *bh; + int err; + + bh = sb_getblk(sb, blk); + if (!bh) + return ERR_PTR(-EIO); + if ((err = ext3_journal_get_write_access(handle, bh))) { + brelse(bh); + bh = ERR_PTR(err); + } else { + lock_buffer(bh); + memset(bh->b_data, 0, sb->s_blocksize); + set_buffer_uptodate(bh); + unlock_buffer(bh); + } + + return bh; +} + +/* + * To avoid calling the atomic setbit hundreds or thousands of times, we only + * need to use it within a single byte (to ensure we get endianness right). + * We can use memset for the rest of the bitmap as there are no other users. + */ +static void mark_bitmap_end(int start_bit, int end_bit, char *bitmap) +{ + int i; + + if (start_bit >= end_bit) + return; + + ext3_debug("mark end bits +%d through +%d used\n", start_bit, end_bit); + for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++) + ext3_set_bit(i, bitmap); + if (i < end_bit) + memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3); +} + +/* + * Set up the block and inode bitmaps, and the inode table for the new group. + * This doesn't need to be part of the main transaction, since we are only + * changing blocks outside the actual filesystem. We still do journaling to + * ensure the recovery is correct in case of a failure just after resize. + * If any part of this fails, we simply abort the resize. + */ +static int setup_new_group_blocks(struct super_block *sb, + struct ext3_new_group_data *input) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + ext3_fsblk_t start = ext3_group_first_block_no(sb, input->group); + int reserved_gdb = ext3_bg_has_super(sb, input->group) ? + le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) : 0; + unsigned long gdblocks = ext3_bg_num_gdb(sb, input->group); + struct buffer_head *bh; + handle_t *handle; + ext3_fsblk_t block; + ext3_grpblk_t bit; + int i; + int err = 0, err2; + + handle = ext3_journal_start_sb(sb, reserved_gdb + gdblocks + + 2 + sbi->s_itb_per_group); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + lock_super(sb); + if (input->group != sbi->s_groups_count) { + err = -EBUSY; + goto exit_journal; + } + + if (IS_ERR(bh = bclean(handle, sb, input->block_bitmap))) { + err = PTR_ERR(bh); + goto exit_journal; + } + + if (ext3_bg_has_super(sb, input->group)) { + ext3_debug("mark backup superblock %#04lx (+0)\n", start); + ext3_set_bit(0, bh->b_data); + } + + /* Copy all of the GDT blocks into the backup in this group */ + for (i = 0, bit = 1, block = start + 1; + i < gdblocks; i++, block++, bit++) { + struct buffer_head *gdb; + + ext3_debug("update backup group %#04lx (+%d)\n", block, bit); + + gdb = sb_getblk(sb, block); + if (!gdb) { + err = -EIO; + goto exit_bh; + } + if ((err = ext3_journal_get_write_access(handle, gdb))) { + brelse(gdb); + goto exit_bh; + } + lock_buffer(bh); + memcpy(gdb->b_data, sbi->s_group_desc[i]->b_data, bh->b_size); + set_buffer_uptodate(gdb); + unlock_buffer(bh); + ext3_journal_dirty_metadata(handle, gdb); + ext3_set_bit(bit, bh->b_data); + brelse(gdb); + } + + /* Zero out all of the reserved backup group descriptor table blocks */ + for (i = 0, bit = gdblocks + 1, block = start + bit; + i < reserved_gdb; i++, block++, bit++) { + struct buffer_head *gdb; + + ext3_debug("clear reserved block %#04lx (+%d)\n", block, bit); + + if (IS_ERR(gdb = bclean(handle, sb, block))) { + err = PTR_ERR(bh); + goto exit_bh; + } + ext3_journal_dirty_metadata(handle, gdb); + ext3_set_bit(bit, bh->b_data); + brelse(gdb); + } + ext3_debug("mark block bitmap %#04x (+%ld)\n", input->block_bitmap, + input->block_bitmap - start); + ext3_set_bit(input->block_bitmap - start, bh->b_data); + ext3_debug("mark inode bitmap %#04x (+%ld)\n", input->inode_bitmap, + input->inode_bitmap - start); + ext3_set_bit(input->inode_bitmap - start, bh->b_data); + + /* Zero out all of the inode table blocks */ + for (i = 0, block = input->inode_table, bit = block - start; + i < sbi->s_itb_per_group; i++, bit++, block++) { + struct buffer_head *it; + + ext3_debug("clear inode block %#04lx (+%d)\n", block, bit); + if (IS_ERR(it = bclean(handle, sb, block))) { + err = PTR_ERR(it); + goto exit_bh; + } + ext3_journal_dirty_metadata(handle, it); + brelse(it); + ext3_set_bit(bit, bh->b_data); + } + mark_bitmap_end(input->blocks_count, EXT3_BLOCKS_PER_GROUP(sb), + bh->b_data); + ext3_journal_dirty_metadata(handle, bh); + brelse(bh); + + /* Mark unused entries in inode bitmap used */ + ext3_debug("clear inode bitmap %#04x (+%ld)\n", + input->inode_bitmap, input->inode_bitmap - start); + if (IS_ERR(bh = bclean(handle, sb, input->inode_bitmap))) { + err = PTR_ERR(bh); + goto exit_journal; + } + + mark_bitmap_end(EXT3_INODES_PER_GROUP(sb), EXT3_BLOCKS_PER_GROUP(sb), + bh->b_data); + ext3_journal_dirty_metadata(handle, bh); +exit_bh: + brelse(bh); + +exit_journal: + unlock_super(sb); + if ((err2 = ext3_journal_stop(handle)) && !err) + err = err2; + + return err; +} + +/* + * Iterate through the groups which hold BACKUP superblock/GDT copies in an + * ext3 filesystem. The counters should be initialized to 1, 5, and 7 before + * calling this for the first time. In a sparse filesystem it will be the + * sequence of powers of 3, 5, and 7: 1, 3, 5, 7, 9, 25, 27, 49, 81, ... + * For a non-sparse filesystem it will be every group: 1, 2, 3, 4, ... + */ +static unsigned ext3_list_backups(struct super_block *sb, unsigned *three, + unsigned *five, unsigned *seven) +{ + unsigned *min = three; + int mult = 3; + unsigned ret; + + if (!EXT3_HAS_RO_COMPAT_FEATURE(sb, + EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER)) { + ret = *min; + *min += 1; + return ret; + } + + if (*five < *min) { + min = five; + mult = 5; + } + if (*seven < *min) { + min = seven; + mult = 7; + } + + ret = *min; + *min *= mult; + + return ret; +} + +/* + * Check that all of the backup GDT blocks are held in the primary GDT block. + * It is assumed that they are stored in group order. Returns the number of + * groups in current filesystem that have BACKUPS, or -ve error code. + */ +static int verify_reserved_gdb(struct super_block *sb, + struct buffer_head *primary) +{ + const ext3_fsblk_t blk = primary->b_blocknr; + const unsigned long end = EXT3_SB(sb)->s_groups_count; + unsigned three = 1; + unsigned five = 5; + unsigned seven = 7; + unsigned grp; + __le32 *p = (__le32 *)primary->b_data; + int gdbackups = 0; + + while ((grp = ext3_list_backups(sb, &three, &five, &seven)) < end) { + if (le32_to_cpu(*p++) != grp * EXT3_BLOCKS_PER_GROUP(sb) + blk){ + ext3_warning(sb, __FUNCTION__, + "reserved GDT "E3FSBLK + " missing grp %d ("E3FSBLK")", + blk, grp, + grp * EXT3_BLOCKS_PER_GROUP(sb) + blk); + return -EINVAL; + } + if (++gdbackups > EXT3_ADDR_PER_BLOCK(sb)) + return -EFBIG; + } + + return gdbackups; +} + +/* + * Called when we need to bring a reserved group descriptor table block into + * use from the resize inode. The primary copy of the new GDT block currently + * is an indirect block (under the double indirect block in the resize inode). + * The new backup GDT blocks will be stored as leaf blocks in this indirect + * block, in group order. Even though we know all the block numbers we need, + * we check to ensure that the resize inode has actually reserved these blocks. + * + * Don't need to update the block bitmaps because the blocks are still in use. + * + * We get all of the error cases out of the way, so that we are sure to not + * fail once we start modifying the data on disk, because JBD has no rollback. + */ +static int add_new_gdb(handle_t *handle, struct inode *inode, + struct ext3_new_group_data *input, + struct buffer_head **primary) +{ + struct super_block *sb = inode->i_sb; + struct ext3_super_block *es = EXT3_SB(sb)->s_es; + unsigned long gdb_num = input->group / EXT3_DESC_PER_BLOCK(sb); + ext3_fsblk_t gdblock = EXT3_SB(sb)->s_sbh->b_blocknr + 1 + gdb_num; + struct buffer_head **o_group_desc, **n_group_desc; + struct buffer_head *dind; + int gdbackups; + struct ext3_iloc iloc; + __le32 *data; + int err; + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG + "EXT3-fs: ext3_add_new_gdb: adding group block %lu\n", + gdb_num); + + /* + * If we are not using the primary superblock/GDT copy don't resize, + * because the user tools have no way of handling this. Probably a + * bad time to do it anyways. + */ + if (EXT3_SB(sb)->s_sbh->b_blocknr != + le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block)) { + ext3_warning(sb, __FUNCTION__, + "won't resize using backup superblock at %llu", + (unsigned long long)EXT3_SB(sb)->s_sbh->b_blocknr); + return -EPERM; + } + + *primary = sb_bread(sb, gdblock); + if (!*primary) + return -EIO; + + if ((gdbackups = verify_reserved_gdb(sb, *primary)) < 0) { + err = gdbackups; + goto exit_bh; + } + + data = EXT3_I(inode)->i_data + EXT3_DIND_BLOCK; + dind = sb_bread(sb, le32_to_cpu(*data)); + if (!dind) { + err = -EIO; + goto exit_bh; + } + + data = (__le32 *)dind->b_data; + if (le32_to_cpu(data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)]) != gdblock) { + ext3_warning(sb, __FUNCTION__, + "new group %u GDT block "E3FSBLK" not reserved", + input->group, gdblock); + err = -EINVAL; + goto exit_dind; + } + + if ((err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh))) + goto exit_dind; + + if ((err = ext3_journal_get_write_access(handle, *primary))) + goto exit_sbh; + + if ((err = ext3_journal_get_write_access(handle, dind))) + goto exit_primary; + + /* ext3_reserve_inode_write() gets a reference on the iloc */ + if ((err = ext3_reserve_inode_write(handle, inode, &iloc))) + goto exit_dindj; + + n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *), + GFP_KERNEL); + if (!n_group_desc) { + err = -ENOMEM; + ext3_warning (sb, __FUNCTION__, + "not enough memory for %lu groups", gdb_num + 1); + goto exit_inode; + } + + /* + * Finally, we have all of the possible failures behind us... + * + * Remove new GDT block from inode double-indirect block and clear out + * the new GDT block for use (which also "frees" the backup GDT blocks + * from the reserved inode). We don't need to change the bitmaps for + * these blocks, because they are marked as in-use from being in the + * reserved inode, and will become GDT blocks (primary and backup). + */ + data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)] = 0; + ext3_journal_dirty_metadata(handle, dind); + brelse(dind); + inode->i_blocks -= (gdbackups + 1) * sb->s_blocksize >> 9; + ext3_mark_iloc_dirty(handle, inode, &iloc); + memset((*primary)->b_data, 0, sb->s_blocksize); + ext3_journal_dirty_metadata(handle, *primary); + + o_group_desc = EXT3_SB(sb)->s_group_desc; + memcpy(n_group_desc, o_group_desc, + EXT3_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); + n_group_desc[gdb_num] = *primary; + EXT3_SB(sb)->s_group_desc = n_group_desc; + EXT3_SB(sb)->s_gdb_count++; + kfree(o_group_desc); + + es->s_reserved_gdt_blocks = + cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1); + ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + + return 0; + +exit_inode: + //ext3_journal_release_buffer(handle, iloc.bh); + brelse(iloc.bh); +exit_dindj: + //ext3_journal_release_buffer(handle, dind); +exit_primary: + //ext3_journal_release_buffer(handle, *primary); +exit_sbh: + //ext3_journal_release_buffer(handle, *primary); +exit_dind: + brelse(dind); +exit_bh: + brelse(*primary); + + ext3_debug("leaving with error %d\n", err); + return err; +} + +/* + * Called when we are adding a new group which has a backup copy of each of + * the GDT blocks (i.e. sparse group) and there are reserved GDT blocks. + * We need to add these reserved backup GDT blocks to the resize inode, so + * that they are kept for future resizing and not allocated to files. + * + * Each reserved backup GDT block will go into a different indirect block. + * The indirect blocks are actually the primary reserved GDT blocks, + * so we know in advance what their block numbers are. We only get the + * double-indirect block to verify it is pointing to the primary reserved + * GDT blocks so we don't overwrite a data block by accident. The reserved + * backup GDT blocks are stored in their reserved primary GDT block. + */ +static int reserve_backup_gdb(handle_t *handle, struct inode *inode, + struct ext3_new_group_data *input) +{ + struct super_block *sb = inode->i_sb; + int reserved_gdb =le16_to_cpu(EXT3_SB(sb)->s_es->s_reserved_gdt_blocks); + struct buffer_head **primary; + struct buffer_head *dind; + struct ext3_iloc iloc; + ext3_fsblk_t blk; + __le32 *data, *end; + int gdbackups = 0; + int res, i; + int err; + + primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_KERNEL); + if (!primary) + return -ENOMEM; + + data = EXT3_I(inode)->i_data + EXT3_DIND_BLOCK; + dind = sb_bread(sb, le32_to_cpu(*data)); + if (!dind) { + err = -EIO; + goto exit_free; + } + + blk = EXT3_SB(sb)->s_sbh->b_blocknr + 1 + EXT3_SB(sb)->s_gdb_count; + data = (__le32 *)dind->b_data + EXT3_SB(sb)->s_gdb_count; + end = (__le32 *)dind->b_data + EXT3_ADDR_PER_BLOCK(sb); + + /* Get each reserved primary GDT block and verify it holds backups */ + for (res = 0; res < reserved_gdb; res++, blk++) { + if (le32_to_cpu(*data) != blk) { + ext3_warning(sb, __FUNCTION__, + "reserved block "E3FSBLK + " not at offset %ld", + blk, + (long)(data - (__le32 *)dind->b_data)); + err = -EINVAL; + goto exit_bh; + } + primary[res] = sb_bread(sb, blk); + if (!primary[res]) { + err = -EIO; + goto exit_bh; + } + if ((gdbackups = verify_reserved_gdb(sb, primary[res])) < 0) { + brelse(primary[res]); + err = gdbackups; + goto exit_bh; + } + if (++data >= end) + data = (__le32 *)dind->b_data; + } + + for (i = 0; i < reserved_gdb; i++) { + if ((err = ext3_journal_get_write_access(handle, primary[i]))) { + /* + int j; + for (j = 0; j < i; j++) + ext3_journal_release_buffer(handle, primary[j]); + */ + goto exit_bh; + } + } + + if ((err = ext3_reserve_inode_write(handle, inode, &iloc))) + goto exit_bh; + + /* + * Finally we can add each of the reserved backup GDT blocks from + * the new group to its reserved primary GDT block. + */ + blk = input->group * EXT3_BLOCKS_PER_GROUP(sb); + for (i = 0; i < reserved_gdb; i++) { + int err2; + data = (__le32 *)primary[i]->b_data; + /* printk("reserving backup %lu[%u] = %lu\n", + primary[i]->b_blocknr, gdbackups, + blk + primary[i]->b_blocknr); */ + data[gdbackups] = cpu_to_le32(blk + primary[i]->b_blocknr); + err2 = ext3_journal_dirty_metadata(handle, primary[i]); + if (!err) + err = err2; + } + inode->i_blocks += reserved_gdb * sb->s_blocksize >> 9; + ext3_mark_iloc_dirty(handle, inode, &iloc); + +exit_bh: + while (--res >= 0) + brelse(primary[res]); + brelse(dind); + +exit_free: + kfree(primary); + + return err; +} + +/* + * Update the backup copies of the ext3 metadata. These don't need to be part + * of the main resize transaction, because e2fsck will re-write them if there + * is a problem (basically only OOM will cause a problem). However, we + * _should_ update the backups if possible, in case the primary gets trashed + * for some reason and we need to run e2fsck from a backup superblock. The + * important part is that the new block and inode counts are in the backup + * superblocks, and the location of the new group metadata in the GDT backups. + * + * We do not need lock_super() for this, because these blocks are not + * otherwise touched by the filesystem code when it is mounted. We don't + * need to worry about last changing from sbi->s_groups_count, because the + * worst that can happen is that we do not copy the full number of backups + * at this time. The resize which changed s_groups_count will backup again. + */ +static void update_backups(struct super_block *sb, + int blk_off, char *data, int size) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + const unsigned long last = sbi->s_groups_count; + const int bpg = EXT3_BLOCKS_PER_GROUP(sb); + unsigned three = 1; + unsigned five = 5; + unsigned seven = 7; + unsigned group; + int rest = sb->s_blocksize - size; + handle_t *handle; + int err = 0, err2; + + handle = ext3_journal_start_sb(sb, EXT3_MAX_TRANS_DATA); + if (IS_ERR(handle)) { + group = 1; + err = PTR_ERR(handle); + goto exit_err; + } + + while ((group = ext3_list_backups(sb, &three, &five, &seven)) < last) { + struct buffer_head *bh; + + /* Out of journal space, and can't get more - abort - so sad */ + if (handle->h_buffer_credits == 0 && + ext3_journal_extend(handle, EXT3_MAX_TRANS_DATA) && + (err = ext3_journal_restart(handle, EXT3_MAX_TRANS_DATA))) + break; + + bh = sb_getblk(sb, group * bpg + blk_off); + if (!bh) { + err = -EIO; + break; + } + ext3_debug("update metadata backup %#04lx\n", + (unsigned long)bh->b_blocknr); + if ((err = ext3_journal_get_write_access(handle, bh))) + break; + lock_buffer(bh); + memcpy(bh->b_data, data, size); + if (rest) + memset(bh->b_data + size, 0, rest); + set_buffer_uptodate(bh); + unlock_buffer(bh); + ext3_journal_dirty_metadata(handle, bh); + brelse(bh); + } + if ((err2 = ext3_journal_stop(handle)) && !err) + err = err2; + + /* + * Ugh! Need to have e2fsck write the backup copies. It is too + * late to revert the resize, we shouldn't fail just because of + * the backup copies (they are only needed in case of corruption). + * + * However, if we got here we have a journal problem too, so we + * can't really start a transaction to mark the superblock. + * Chicken out and just set the flag on the hope it will be written + * to disk, and if not - we will simply wait until next fsck. + */ +exit_err: + if (err) { + ext3_warning(sb, __FUNCTION__, + "can't update backup for group %d (err %d), " + "forcing fsck on next reboot", group, err); + sbi->s_mount_state &= ~EXT3_VALID_FS; + sbi->s_es->s_state &= cpu_to_le16(~EXT3_VALID_FS); + mark_buffer_dirty(sbi->s_sbh); + } +} + +/* Add group descriptor data to an existing or new group descriptor block. + * Ensure we handle all possible error conditions _before_ we start modifying + * the filesystem, because we cannot abort the transaction and not have it + * write the data to disk. + * + * If we are on a GDT block boundary, we need to get the reserved GDT block. + * Otherwise, we may need to add backup GDT blocks for a sparse group. + * + * We only need to hold the superblock lock while we are actually adding + * in the new group's counts to the superblock. Prior to that we have + * not really "added" the group at all. We re-check that we are still + * adding in the last group in case things have changed since verifying. + */ +int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext3_super_block *es = sbi->s_es; + int reserved_gdb = ext3_bg_has_super(sb, input->group) ? + le16_to_cpu(es->s_reserved_gdt_blocks) : 0; + struct buffer_head *primary = NULL; + struct ext3_group_desc *gdp; + struct inode *inode = NULL; + handle_t *handle; + int gdb_off, gdb_num; + int err, err2; + + gdb_num = input->group / EXT3_DESC_PER_BLOCK(sb); + gdb_off = input->group % EXT3_DESC_PER_BLOCK(sb); + + if (gdb_off == 0 && !EXT3_HAS_RO_COMPAT_FEATURE(sb, + EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER)) { + ext3_warning(sb, __FUNCTION__, + "Can't resize non-sparse filesystem further"); + return -EPERM; + } + + if (le32_to_cpu(es->s_blocks_count) + input->blocks_count < + le32_to_cpu(es->s_blocks_count)) { + ext3_warning(sb, __FUNCTION__, "blocks_count overflow\n"); + return -EINVAL; + } + + if (le32_to_cpu(es->s_inodes_count) + EXT3_INODES_PER_GROUP(sb) < + le32_to_cpu(es->s_inodes_count)) { + ext3_warning(sb, __FUNCTION__, "inodes_count overflow\n"); + return -EINVAL; + } + + if (reserved_gdb || gdb_off == 0) { + if (!EXT3_HAS_COMPAT_FEATURE(sb, + EXT3_FEATURE_COMPAT_RESIZE_INODE)){ + ext3_warning(sb, __FUNCTION__, + "No reserved GDT blocks, can't resize"); + return -EPERM; + } + inode = iget(sb, EXT3_RESIZE_INO); + if (!inode || is_bad_inode(inode)) { + ext3_warning(sb, __FUNCTION__, + "Error opening resize inode"); + iput(inode); + return -ENOENT; + } + } + + if ((err = verify_group_input(sb, input))) + goto exit_put; + + if ((err = setup_new_group_blocks(sb, input))) + goto exit_put; + + /* + * We will always be modifying at least the superblock and a GDT + * block. If we are adding a group past the last current GDT block, + * we will also modify the inode and the dindirect block. If we + * are adding a group with superblock/GDT backups we will also + * modify each of the reserved GDT dindirect blocks. + */ + handle = ext3_journal_start_sb(sb, + ext3_bg_has_super(sb, input->group) ? + 3 + reserved_gdb : 4); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto exit_put; + } + + lock_super(sb); + if (input->group != sbi->s_groups_count) { + ext3_warning(sb, __FUNCTION__, + "multiple resizers run on filesystem!"); + err = -EBUSY; + goto exit_journal; + } + + if ((err = ext3_journal_get_write_access(handle, sbi->s_sbh))) + goto exit_journal; + + /* + * We will only either add reserved group blocks to a backup group + * or remove reserved blocks for the first group in a new group block. + * Doing both would be mean more complex code, and sane people don't + * use non-sparse filesystems anymore. This is already checked above. + */ + if (gdb_off) { + primary = sbi->s_group_desc[gdb_num]; + if ((err = ext3_journal_get_write_access(handle, primary))) + goto exit_journal; + + if (reserved_gdb && ext3_bg_num_gdb(sb, input->group) && + (err = reserve_backup_gdb(handle, inode, input))) + goto exit_journal; + } else if ((err = add_new_gdb(handle, inode, input, &primary))) + goto exit_journal; + + /* + * OK, now we've set up the new group. Time to make it active. + * + * Current kernels don't lock all allocations via lock_super(), + * so we have to be safe wrt. concurrent accesses the group + * data. So we need to be careful to set all of the relevant + * group descriptor data etc. *before* we enable the group. + * + * The key field here is sbi->s_groups_count: as long as + * that retains its old value, nobody is going to access the new + * group. + * + * So first we update all the descriptor metadata for the new + * group; then we update the total disk blocks count; then we + * update the groups count to enable the group; then finally we + * update the free space counts so that the system can start + * using the new disk blocks. + */ + + /* Update group descriptor block for new group */ + gdp = (struct ext3_group_desc *)primary->b_data + gdb_off; + + gdp->bg_block_bitmap = cpu_to_le32(input->block_bitmap); + gdp->bg_inode_bitmap = cpu_to_le32(input->inode_bitmap); + gdp->bg_inode_table = cpu_to_le32(input->inode_table); + gdp->bg_free_blocks_count = cpu_to_le16(input->free_blocks_count); + gdp->bg_free_inodes_count = cpu_to_le16(EXT3_INODES_PER_GROUP(sb)); + + /* + * Make the new blocks and inodes valid next. We do this before + * increasing the group count so that once the group is enabled, + * all of its blocks and inodes are already valid. + * + * We always allocate group-by-group, then block-by-block or + * inode-by-inode within a group, so enabling these + * blocks/inodes before the group is live won't actually let us + * allocate the new space yet. + */ + es->s_blocks_count = cpu_to_le32(le32_to_cpu(es->s_blocks_count) + + input->blocks_count); + es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) + + EXT3_INODES_PER_GROUP(sb)); + + /* + * We need to protect s_groups_count against other CPUs seeing + * inconsistent state in the superblock. + * + * The precise rules we use are: + * + * * Writers of s_groups_count *must* hold lock_super + * AND + * * Writers must perform a smp_wmb() after updating all dependent + * data and before modifying the groups count + * + * * Readers must hold lock_super() over the access + * OR + * * Readers must perform an smp_rmb() after reading the groups count + * and before reading any dependent data. + * + * NB. These rules can be relaxed when checking the group count + * while freeing data, as we can only allocate from a block + * group after serialising against the group count, and we can + * only then free after serialising in turn against that + * allocation. + */ + smp_wmb(); + + /* Update the global fs size fields */ + sbi->s_groups_count++; + + ext3_journal_dirty_metadata(handle, primary); + + /* Update the reserved block counts only once the new group is + * active. */ + es->s_r_blocks_count = cpu_to_le32(le32_to_cpu(es->s_r_blocks_count) + + input->reserved_blocks); + + /* Update the free space counts */ + percpu_counter_mod(&sbi->s_freeblocks_counter, + input->free_blocks_count); + percpu_counter_mod(&sbi->s_freeinodes_counter, + EXT3_INODES_PER_GROUP(sb)); + + ext3_journal_dirty_metadata(handle, sbi->s_sbh); + sb->s_dirt = 1; + +exit_journal: + unlock_super(sb); + if ((err2 = ext3_journal_stop(handle)) && !err) + err = err2; + if (!err) { + update_backups(sb, sbi->s_sbh->b_blocknr, (char *)es, + sizeof(struct ext3_super_block)); + update_backups(sb, primary->b_blocknr, primary->b_data, + primary->b_size); + } +exit_put: + iput(inode); + return err; +} /* ext3_group_add */ + +/* Extend the filesystem to the new number of blocks specified. This entry + * point is only used to extend the current filesystem to the end of the last + * existing group. It can be accessed via ioctl, or by "remount,resize=" + * for emergencies (because it has no dependencies on reserved blocks). + * + * If we _really_ wanted, we could use default values to call ext3_group_add() + * allow the "remount" trick to work for arbitrary resizing, assuming enough + * GDT blocks are reserved to grow to the desired size. + */ +int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, + ext3_fsblk_t n_blocks_count) +{ + ext3_fsblk_t o_blocks_count; + unsigned long o_groups_count; + ext3_grpblk_t last; + ext3_grpblk_t add; + struct buffer_head * bh; + handle_t *handle; + int err; + unsigned long freed_blocks; + + /* We don't need to worry about locking wrt other resizers just + * yet: we're going to revalidate es->s_blocks_count after + * taking lock_super() below. */ + o_blocks_count = le32_to_cpu(es->s_blocks_count); + o_groups_count = EXT3_SB(sb)->s_groups_count; + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT3-fs: extending last group from "E3FSBLK" uto "E3FSBLK" blocks\n", + o_blocks_count, n_blocks_count); + + if (n_blocks_count == 0 || n_blocks_count == o_blocks_count) + return 0; + + if (n_blocks_count > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { + printk(KERN_ERR "EXT3-fs: filesystem on %s:" + " too large to resize to %lu blocks safely\n", + sb->s_id, n_blocks_count); + if (sizeof(sector_t) < 8) + ext3_warning(sb, __FUNCTION__, + "CONFIG_LBD not enabled\n"); + return -EINVAL; + } + + if (n_blocks_count < o_blocks_count) { + ext3_warning(sb, __FUNCTION__, + "can't shrink FS - resize aborted"); + return -EBUSY; + } + + /* Handle the remaining blocks in the last group only. */ + last = (o_blocks_count - le32_to_cpu(es->s_first_data_block)) % + EXT3_BLOCKS_PER_GROUP(sb); + + if (last == 0) { + ext3_warning(sb, __FUNCTION__, + "need to use ext2online to resize further"); + return -EPERM; + } + + add = EXT3_BLOCKS_PER_GROUP(sb) - last; + + if (o_blocks_count + add < o_blocks_count) { + ext3_warning(sb, __FUNCTION__, "blocks_count overflow"); + return -EINVAL; + } + + if (o_blocks_count + add > n_blocks_count) + add = n_blocks_count - o_blocks_count; + + if (o_blocks_count + add < n_blocks_count) + ext3_warning(sb, __FUNCTION__, + "will only finish group ("E3FSBLK + " blocks, %u new)", + o_blocks_count + add, add); + + /* See if the device is actually as big as what was requested */ + bh = sb_bread(sb, o_blocks_count + add -1); + if (!bh) { + ext3_warning(sb, __FUNCTION__, + "can't read last block, resize aborted"); + return -ENOSPC; + } + brelse(bh); + + /* We will update the superblock, one block bitmap, and + * one group descriptor via ext3_free_blocks(). + */ + handle = ext3_journal_start_sb(sb, 3); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + ext3_warning(sb, __FUNCTION__, "error %d on journal start",err); + goto exit_put; + } + + lock_super(sb); + if (o_blocks_count != le32_to_cpu(es->s_blocks_count)) { + ext3_warning(sb, __FUNCTION__, + "multiple resizers run on filesystem!"); + unlock_super(sb); + err = -EBUSY; + goto exit_put; + } + + if ((err = ext3_journal_get_write_access(handle, + EXT3_SB(sb)->s_sbh))) { + ext3_warning(sb, __FUNCTION__, + "error %d on journal write access", err); + unlock_super(sb); + ext3_journal_stop(handle); + goto exit_put; + } + es->s_blocks_count = cpu_to_le32(o_blocks_count + add); + ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + sb->s_dirt = 1; + unlock_super(sb); + ext3_debug("freeing blocks %lu through "E3FSBLK"\n", o_blocks_count, + o_blocks_count + add); + ext3_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); + ext3_debug("freed blocks "E3FSBLK" through "E3FSBLK"\n", o_blocks_count, + o_blocks_count + add); + if ((err = ext3_journal_stop(handle))) + goto exit_put; + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT3-fs: extended group to %u blocks\n", + le32_to_cpu(es->s_blocks_count)); + update_backups(sb, EXT3_SB(sb)->s_sbh->b_blocknr, (char *)es, + sizeof(struct ext3_super_block)); +exit_put: + return err; +} /* ext3_group_extend */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c new file mode 100644 index 0000000..8bfd56e --- /dev/null +++ b/fs/ext4/super.c @@ -0,0 +1,2754 @@ +/* + * linux/fs/ext3/super.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/inode.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include "xattr.h" +#include "acl.h" +#include "namei.h" + +static int ext3_load_journal(struct super_block *, struct ext3_super_block *, + unsigned long journal_devnum); +static int ext3_create_journal(struct super_block *, struct ext3_super_block *, + unsigned int); +static void ext3_commit_super (struct super_block * sb, + struct ext3_super_block * es, + int sync); +static void ext3_mark_recovery_complete(struct super_block * sb, + struct ext3_super_block * es); +static void ext3_clear_journal_err(struct super_block * sb, + struct ext3_super_block * es); +static int ext3_sync_fs(struct super_block *sb, int wait); +static const char *ext3_decode_error(struct super_block * sb, int errno, + char nbuf[16]); +static int ext3_remount (struct super_block * sb, int * flags, char * data); +static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf); +static void ext3_unlockfs(struct super_block *sb); +static void ext3_write_super (struct super_block * sb); +static void ext3_write_super_lockfs(struct super_block *sb); + +/* + * Wrappers for journal_start/end. + * + * The only special thing we need to do here is to make sure that all + * journal_end calls result in the superblock being marked dirty, so + * that sync() will call the filesystem's write_super callback if + * appropriate. + */ +handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks) +{ + journal_t *journal; + + if (sb->s_flags & MS_RDONLY) + return ERR_PTR(-EROFS); + + /* Special case here: if the journal has aborted behind our + * backs (eg. EIO in the commit thread), then we still need to + * take the FS itself readonly cleanly. */ + journal = EXT3_SB(sb)->s_journal; + if (is_journal_aborted(journal)) { + ext3_abort(sb, __FUNCTION__, + "Detected aborted journal"); + return ERR_PTR(-EROFS); + } + + return journal_start(journal, nblocks); +} + +/* + * The only special thing we need to do here is to make sure that all + * journal_stop calls result in the superblock being marked dirty, so + * that sync() will call the filesystem's write_super callback if + * appropriate. + */ +int __ext3_journal_stop(const char *where, handle_t *handle) +{ + struct super_block *sb; + int err; + int rc; + + sb = handle->h_transaction->t_journal->j_private; + err = handle->h_err; + rc = journal_stop(handle); + + if (!err) + err = rc; + if (err) + __ext3_std_error(sb, where, err); + return err; +} + +void ext3_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err) +{ + char nbuf[16]; + const char *errstr = ext3_decode_error(NULL, err, nbuf); + + if (bh) + BUFFER_TRACE(bh, "abort"); + + if (!handle->h_err) + handle->h_err = err; + + if (is_handle_aborted(handle)) + return; + + printk(KERN_ERR "%s: aborting transaction: %s in %s\n", + caller, errstr, err_fn); + + journal_abort_handle(handle); +} + +/* Deal with the reporting of failure conditions on a filesystem such as + * inconsistencies detected or read IO failures. + * + * On ext2, we can store the error state of the filesystem in the + * superblock. That is not possible on ext3, because we may have other + * write ordering constraints on the superblock which prevent us from + * writing it out straight away; and given that the journal is about to + * be aborted, we can't rely on the current, or future, transactions to + * write out the superblock safely. + * + * We'll just use the journal_abort() error code to record an error in + * the journal instead. On recovery, the journal will compain about + * that error until we've noted it down and cleared it. + */ + +static void ext3_handle_error(struct super_block *sb) +{ + struct ext3_super_block *es = EXT3_SB(sb)->s_es; + + EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; + es->s_state |= cpu_to_le16(EXT3_ERROR_FS); + + if (sb->s_flags & MS_RDONLY) + return; + + if (!test_opt (sb, ERRORS_CONT)) { + journal_t *journal = EXT3_SB(sb)->s_journal; + + EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT; + if (journal) + journal_abort(journal, -EIO); + } + if (test_opt (sb, ERRORS_RO)) { + printk (KERN_CRIT "Remounting filesystem read-only\n"); + sb->s_flags |= MS_RDONLY; + } + ext3_commit_super(sb, es, 1); + if (test_opt(sb, ERRORS_PANIC)) + panic("EXT3-fs (device %s): panic forced after error\n", + sb->s_id); +} + +void ext3_error (struct super_block * sb, const char * function, + const char * fmt, ...) +{ + va_list args; + + va_start(args, fmt); + printk(KERN_CRIT "EXT3-fs error (device %s): %s: ",sb->s_id, function); + vprintk(fmt, args); + printk("\n"); + va_end(args); + + ext3_handle_error(sb); +} + +static const char *ext3_decode_error(struct super_block * sb, int errno, + char nbuf[16]) +{ + char *errstr = NULL; + + switch (errno) { + case -EIO: + errstr = "IO failure"; + break; + case -ENOMEM: + errstr = "Out of memory"; + break; + case -EROFS: + if (!sb || EXT3_SB(sb)->s_journal->j_flags & JFS_ABORT) + errstr = "Journal has aborted"; + else + errstr = "Readonly filesystem"; + break; + default: + /* If the caller passed in an extra buffer for unknown + * errors, textualise them now. Else we just return + * NULL. */ + if (nbuf) { + /* Check for truncated error codes... */ + if (snprintf(nbuf, 16, "error %d", -errno) >= 0) + errstr = nbuf; + } + break; + } + + return errstr; +} + +/* __ext3_std_error decodes expected errors from journaling functions + * automatically and invokes the appropriate error response. */ + +void __ext3_std_error (struct super_block * sb, const char * function, + int errno) +{ + char nbuf[16]; + const char *errstr; + + /* Special case: if the error is EROFS, and we're not already + * inside a transaction, then there's really no point in logging + * an error. */ + if (errno == -EROFS && journal_current_handle() == NULL && + (sb->s_flags & MS_RDONLY)) + return; + + errstr = ext3_decode_error(sb, errno, nbuf); + printk (KERN_CRIT "EXT3-fs error (device %s) in %s: %s\n", + sb->s_id, function, errstr); + + ext3_handle_error(sb); +} + +/* + * ext3_abort is a much stronger failure handler than ext3_error. The + * abort function may be used to deal with unrecoverable failures such + * as journal IO errors or ENOMEM at a critical moment in log management. + * + * We unconditionally force the filesystem into an ABORT|READONLY state, + * unless the error response on the fs has been set to panic in which + * case we take the easy way out and panic immediately. + */ + +void ext3_abort (struct super_block * sb, const char * function, + const char * fmt, ...) +{ + va_list args; + + printk (KERN_CRIT "ext3_abort called.\n"); + + va_start(args, fmt); + printk(KERN_CRIT "EXT3-fs error (device %s): %s: ",sb->s_id, function); + vprintk(fmt, args); + printk("\n"); + va_end(args); + + if (test_opt(sb, ERRORS_PANIC)) + panic("EXT3-fs panic from previous error\n"); + + if (sb->s_flags & MS_RDONLY) + return; + + printk(KERN_CRIT "Remounting filesystem read-only\n"); + EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; + sb->s_flags |= MS_RDONLY; + EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT; + journal_abort(EXT3_SB(sb)->s_journal, -EIO); +} + +void ext3_warning (struct super_block * sb, const char * function, + const char * fmt, ...) +{ + va_list args; + + va_start(args, fmt); + printk(KERN_WARNING "EXT3-fs warning (device %s): %s: ", + sb->s_id, function); + vprintk(fmt, args); + printk("\n"); + va_end(args); +} + +void ext3_update_dynamic_rev(struct super_block *sb) +{ + struct ext3_super_block *es = EXT3_SB(sb)->s_es; + + if (le32_to_cpu(es->s_rev_level) > EXT3_GOOD_OLD_REV) + return; + + ext3_warning(sb, __FUNCTION__, + "updating to rev %d because of new feature flag, " + "running e2fsck is recommended", + EXT3_DYNAMIC_REV); + + es->s_first_ino = cpu_to_le32(EXT3_GOOD_OLD_FIRST_INO); + es->s_inode_size = cpu_to_le16(EXT3_GOOD_OLD_INODE_SIZE); + es->s_rev_level = cpu_to_le32(EXT3_DYNAMIC_REV); + /* leave es->s_feature_*compat flags alone */ + /* es->s_uuid will be set by e2fsck if empty */ + + /* + * The rest of the superblock fields should be zero, and if not it + * means they are likely already in use, so leave them alone. We + * can leave it up to e2fsck to clean up any inconsistencies there. + */ +} + +/* + * Open the external journal device + */ +static struct block_device *ext3_blkdev_get(dev_t dev) +{ + struct block_device *bdev; + char b[BDEVNAME_SIZE]; + + bdev = open_by_devnum(dev, FMODE_READ|FMODE_WRITE); + if (IS_ERR(bdev)) + goto fail; + return bdev; + +fail: + printk(KERN_ERR "EXT3: failed to open journal device %s: %ld\n", + __bdevname(dev, b), PTR_ERR(bdev)); + return NULL; +} + +/* + * Release the journal device + */ +static int ext3_blkdev_put(struct block_device *bdev) +{ + bd_release(bdev); + return blkdev_put(bdev); +} + +static int ext3_blkdev_remove(struct ext3_sb_info *sbi) +{ + struct block_device *bdev; + int ret = -ENODEV; + + bdev = sbi->journal_bdev; + if (bdev) { + ret = ext3_blkdev_put(bdev); + sbi->journal_bdev = NULL; + } + return ret; +} + +static inline struct inode *orphan_list_entry(struct list_head *l) +{ + return &list_entry(l, struct ext3_inode_info, i_orphan)->vfs_inode; +} + +static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi) +{ + struct list_head *l; + + printk(KERN_ERR "sb orphan head is %d\n", + le32_to_cpu(sbi->s_es->s_last_orphan)); + + printk(KERN_ERR "sb_info orphan list:\n"); + list_for_each(l, &sbi->s_orphan) { + struct inode *inode = orphan_list_entry(l); + printk(KERN_ERR " " + "inode %s:%lu at %p: mode %o, nlink %d, next %d\n", + inode->i_sb->s_id, inode->i_ino, inode, + inode->i_mode, inode->i_nlink, + NEXT_ORPHAN(inode)); + } +} + +static void ext3_put_super (struct super_block * sb) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext3_super_block *es = sbi->s_es; + int i; + + ext3_xattr_put_super(sb); + journal_destroy(sbi->s_journal); + if (!(sb->s_flags & MS_RDONLY)) { + EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + es->s_state = cpu_to_le16(sbi->s_mount_state); + BUFFER_TRACE(sbi->s_sbh, "marking dirty"); + mark_buffer_dirty(sbi->s_sbh); + ext3_commit_super(sb, es, 1); + } + + for (i = 0; i < sbi->s_gdb_count; i++) + brelse(sbi->s_group_desc[i]); + kfree(sbi->s_group_desc); + percpu_counter_destroy(&sbi->s_freeblocks_counter); + percpu_counter_destroy(&sbi->s_freeinodes_counter); + percpu_counter_destroy(&sbi->s_dirs_counter); + brelse(sbi->s_sbh); +#ifdef CONFIG_QUOTA + for (i = 0; i < MAXQUOTAS; i++) + kfree(sbi->s_qf_names[i]); +#endif + + /* Debugging code just in case the in-memory inode orphan list + * isn't empty. The on-disk one can be non-empty if we've + * detected an error and taken the fs readonly, but the + * in-memory list had better be clean by this point. */ + if (!list_empty(&sbi->s_orphan)) + dump_orphan_list(sb, sbi); + J_ASSERT(list_empty(&sbi->s_orphan)); + + invalidate_bdev(sb->s_bdev, 0); + if (sbi->journal_bdev && sbi->journal_bdev != sb->s_bdev) { + /* + * Invalidate the journal device's buffers. We don't want them + * floating about in memory - the physical journal device may + * hotswapped, and it breaks the `ro-after' testing code. + */ + sync_blockdev(sbi->journal_bdev); + invalidate_bdev(sbi->journal_bdev, 0); + ext3_blkdev_remove(sbi); + } + sb->s_fs_info = NULL; + kfree(sbi); + return; +} + +static kmem_cache_t *ext3_inode_cachep; + +/* + * Called inside transaction, so use GFP_NOFS + */ +static struct inode *ext3_alloc_inode(struct super_block *sb) +{ + struct ext3_inode_info *ei; + + ei = kmem_cache_alloc(ext3_inode_cachep, SLAB_NOFS); + if (!ei) + return NULL; +#ifdef CONFIG_EXT3_FS_POSIX_ACL + ei->i_acl = EXT3_ACL_NOT_CACHED; + ei->i_default_acl = EXT3_ACL_NOT_CACHED; +#endif + ei->i_block_alloc_info = NULL; + ei->vfs_inode.i_version = 1; + return &ei->vfs_inode; +} + +static void ext3_destroy_inode(struct inode *inode) +{ + kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); +} + +static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +{ + struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; + + if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == + SLAB_CTOR_CONSTRUCTOR) { + INIT_LIST_HEAD(&ei->i_orphan); +#ifdef CONFIG_EXT3_FS_XATTR + init_rwsem(&ei->xattr_sem); +#endif + mutex_init(&ei->truncate_mutex); + inode_init_once(&ei->vfs_inode); + } +} + +static int init_inodecache(void) +{ + ext3_inode_cachep = kmem_cache_create("ext3_inode_cache", + sizeof(struct ext3_inode_info), + 0, (SLAB_RECLAIM_ACCOUNT| + SLAB_MEM_SPREAD), + init_once, NULL); + if (ext3_inode_cachep == NULL) + return -ENOMEM; + return 0; +} + +static void destroy_inodecache(void) +{ + kmem_cache_destroy(ext3_inode_cachep); +} + +static void ext3_clear_inode(struct inode *inode) +{ + struct ext3_block_alloc_info *rsv = EXT3_I(inode)->i_block_alloc_info; +#ifdef CONFIG_EXT3_FS_POSIX_ACL + if (EXT3_I(inode)->i_acl && + EXT3_I(inode)->i_acl != EXT3_ACL_NOT_CACHED) { + posix_acl_release(EXT3_I(inode)->i_acl); + EXT3_I(inode)->i_acl = EXT3_ACL_NOT_CACHED; + } + if (EXT3_I(inode)->i_default_acl && + EXT3_I(inode)->i_default_acl != EXT3_ACL_NOT_CACHED) { + posix_acl_release(EXT3_I(inode)->i_default_acl); + EXT3_I(inode)->i_default_acl = EXT3_ACL_NOT_CACHED; + } +#endif + ext3_discard_reservation(inode); + EXT3_I(inode)->i_block_alloc_info = NULL; + if (unlikely(rsv)) + kfree(rsv); +} + +static inline void ext3_show_quota_options(struct seq_file *seq, struct super_block *sb) +{ +#if defined(CONFIG_QUOTA) + struct ext3_sb_info *sbi = EXT3_SB(sb); + + if (sbi->s_jquota_fmt) + seq_printf(seq, ",jqfmt=%s", + (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0"); + + if (sbi->s_qf_names[USRQUOTA]) + seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]); + + if (sbi->s_qf_names[GRPQUOTA]) + seq_printf(seq, ",grpjquota=%s", sbi->s_qf_names[GRPQUOTA]); + + if (sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA) + seq_puts(seq, ",usrquota"); + + if (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA) + seq_puts(seq, ",grpquota"); +#endif +} + +static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) +{ + struct super_block *sb = vfs->mnt_sb; + + if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA) + seq_puts(seq, ",data=journal"); + else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA) + seq_puts(seq, ",data=ordered"); + else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA) + seq_puts(seq, ",data=writeback"); + + ext3_show_quota_options(seq, sb); + + return 0; +} + + +static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino < EXT3_FIRST_INO(sb) && ino != EXT3_ROOT_INO) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * + * ext3_read_inode will return a bad_inode if the inode had been + * deleted, so we should be safe. + * + * Currently we don't know the generation for parent directory, so + * a generation of 0 means "accept any" + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) || + (generation && inode->i_generation != generation)) { + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + +#ifdef CONFIG_QUOTA +#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group") +#define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) + +static int ext3_dquot_initialize(struct inode *inode, int type); +static int ext3_dquot_drop(struct inode *inode); +static int ext3_write_dquot(struct dquot *dquot); +static int ext3_acquire_dquot(struct dquot *dquot); +static int ext3_release_dquot(struct dquot *dquot); +static int ext3_mark_dquot_dirty(struct dquot *dquot); +static int ext3_write_info(struct super_block *sb, int type); +static int ext3_quota_on(struct super_block *sb, int type, int format_id, char *path); +static int ext3_quota_on_mount(struct super_block *sb, int type); +static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, + size_t len, loff_t off); +static ssize_t ext3_quota_write(struct super_block *sb, int type, + const char *data, size_t len, loff_t off); + +static struct dquot_operations ext3_quota_operations = { + .initialize = ext3_dquot_initialize, + .drop = ext3_dquot_drop, + .alloc_space = dquot_alloc_space, + .alloc_inode = dquot_alloc_inode, + .free_space = dquot_free_space, + .free_inode = dquot_free_inode, + .transfer = dquot_transfer, + .write_dquot = ext3_write_dquot, + .acquire_dquot = ext3_acquire_dquot, + .release_dquot = ext3_release_dquot, + .mark_dirty = ext3_mark_dquot_dirty, + .write_info = ext3_write_info +}; + +static struct quotactl_ops ext3_qctl_operations = { + .quota_on = ext3_quota_on, + .quota_off = vfs_quota_off, + .quota_sync = vfs_quota_sync, + .get_info = vfs_get_dqinfo, + .set_info = vfs_set_dqinfo, + .get_dqblk = vfs_get_dqblk, + .set_dqblk = vfs_set_dqblk +}; +#endif + +static struct super_operations ext3_sops = { + .alloc_inode = ext3_alloc_inode, + .destroy_inode = ext3_destroy_inode, + .read_inode = ext3_read_inode, + .write_inode = ext3_write_inode, + .dirty_inode = ext3_dirty_inode, + .delete_inode = ext3_delete_inode, + .put_super = ext3_put_super, + .write_super = ext3_write_super, + .sync_fs = ext3_sync_fs, + .write_super_lockfs = ext3_write_super_lockfs, + .unlockfs = ext3_unlockfs, + .statfs = ext3_statfs, + .remount_fs = ext3_remount, + .clear_inode = ext3_clear_inode, + .show_options = ext3_show_options, +#ifdef CONFIG_QUOTA + .quota_read = ext3_quota_read, + .quota_write = ext3_quota_write, +#endif +}; + +static struct export_operations ext3_export_ops = { + .get_parent = ext3_get_parent, + .get_dentry = ext3_get_dentry, +}; + +enum { + Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid, + Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, Opt_err_ro, + Opt_nouid32, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov, + Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, + Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh, + Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, + Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, + Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, + Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, + Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, + Opt_grpquota +}; + +static match_table_t tokens = { + {Opt_bsd_df, "bsddf"}, + {Opt_minix_df, "minixdf"}, + {Opt_grpid, "grpid"}, + {Opt_grpid, "bsdgroups"}, + {Opt_nogrpid, "nogrpid"}, + {Opt_nogrpid, "sysvgroups"}, + {Opt_resgid, "resgid=%u"}, + {Opt_resuid, "resuid=%u"}, + {Opt_sb, "sb=%u"}, + {Opt_err_cont, "errors=continue"}, + {Opt_err_panic, "errors=panic"}, + {Opt_err_ro, "errors=remount-ro"}, + {Opt_nouid32, "nouid32"}, + {Opt_nocheck, "nocheck"}, + {Opt_nocheck, "check=none"}, + {Opt_debug, "debug"}, + {Opt_oldalloc, "oldalloc"}, + {Opt_orlov, "orlov"}, + {Opt_user_xattr, "user_xattr"}, + {Opt_nouser_xattr, "nouser_xattr"}, + {Opt_acl, "acl"}, + {Opt_noacl, "noacl"}, + {Opt_reservation, "reservation"}, + {Opt_noreservation, "noreservation"}, + {Opt_noload, "noload"}, + {Opt_nobh, "nobh"}, + {Opt_bh, "bh"}, + {Opt_commit, "commit=%u"}, + {Opt_journal_update, "journal=update"}, + {Opt_journal_inum, "journal=%u"}, + {Opt_journal_dev, "journal_dev=%u"}, + {Opt_abort, "abort"}, + {Opt_data_journal, "data=journal"}, + {Opt_data_ordered, "data=ordered"}, + {Opt_data_writeback, "data=writeback"}, + {Opt_offusrjquota, "usrjquota="}, + {Opt_usrjquota, "usrjquota=%s"}, + {Opt_offgrpjquota, "grpjquota="}, + {Opt_grpjquota, "grpjquota=%s"}, + {Opt_jqfmt_vfsold, "jqfmt=vfsold"}, + {Opt_jqfmt_vfsv0, "jqfmt=vfsv0"}, + {Opt_grpquota, "grpquota"}, + {Opt_noquota, "noquota"}, + {Opt_quota, "quota"}, + {Opt_usrquota, "usrquota"}, + {Opt_barrier, "barrier=%u"}, + {Opt_err, NULL}, + {Opt_resize, "resize"}, +}; + +static ext3_fsblk_t get_sb_block(void **data) +{ + ext3_fsblk_t sb_block; + char *options = (char *) *data; + + if (!options || strncmp(options, "sb=", 3) != 0) + return 1; /* Default location */ + options += 3; + /*todo: use simple_strtoll with >32bit ext3 */ + sb_block = simple_strtoul(options, &options, 0); + if (*options && *options != ',') { + printk("EXT3-fs: Invalid sb specification: %s\n", + (char *) *data); + return 1; + } + if (*options == ',') + options++; + *data = (void *) options; + return sb_block; +} + +static int parse_options (char *options, struct super_block *sb, + unsigned int *inum, unsigned long *journal_devnum, + ext3_fsblk_t *n_blocks_count, int is_remount) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + char * p; + substring_t args[MAX_OPT_ARGS]; + int data_opt = 0; + int option; +#ifdef CONFIG_QUOTA + int qtype; + char *qname; +#endif + + if (!options) + return 1; + + while ((p = strsep (&options, ",")) != NULL) { + int token; + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_bsd_df: + clear_opt (sbi->s_mount_opt, MINIX_DF); + break; + case Opt_minix_df: + set_opt (sbi->s_mount_opt, MINIX_DF); + break; + case Opt_grpid: + set_opt (sbi->s_mount_opt, GRPID); + break; + case Opt_nogrpid: + clear_opt (sbi->s_mount_opt, GRPID); + break; + case Opt_resuid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_resuid = option; + break; + case Opt_resgid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_resgid = option; + break; + case Opt_sb: + /* handled by get_sb_block() instead of here */ + /* *sb_block = match_int(&args[0]); */ + break; + case Opt_err_panic: + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_RO); + set_opt (sbi->s_mount_opt, ERRORS_PANIC); + break; + case Opt_err_ro: + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_RO); + break; + case Opt_err_cont: + clear_opt (sbi->s_mount_opt, ERRORS_RO); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_CONT); + break; + case Opt_nouid32: + set_opt (sbi->s_mount_opt, NO_UID32); + break; + case Opt_nocheck: + clear_opt (sbi->s_mount_opt, CHECK); + break; + case Opt_debug: + set_opt (sbi->s_mount_opt, DEBUG); + break; + case Opt_oldalloc: + set_opt (sbi->s_mount_opt, OLDALLOC); + break; + case Opt_orlov: + clear_opt (sbi->s_mount_opt, OLDALLOC); + break; +#ifdef CONFIG_EXT3_FS_XATTR + case Opt_user_xattr: + set_opt (sbi->s_mount_opt, XATTR_USER); + break; + case Opt_nouser_xattr: + clear_opt (sbi->s_mount_opt, XATTR_USER); + break; +#else + case Opt_user_xattr: + case Opt_nouser_xattr: + printk("EXT3 (no)user_xattr options not supported\n"); + break; +#endif +#ifdef CONFIG_EXT3_FS_POSIX_ACL + case Opt_acl: + set_opt(sbi->s_mount_opt, POSIX_ACL); + break; + case Opt_noacl: + clear_opt(sbi->s_mount_opt, POSIX_ACL); + break; +#else + case Opt_acl: + case Opt_noacl: + printk("EXT3 (no)acl options not supported\n"); + break; +#endif + case Opt_reservation: + set_opt(sbi->s_mount_opt, RESERVATION); + break; + case Opt_noreservation: + clear_opt(sbi->s_mount_opt, RESERVATION); + break; + case Opt_journal_update: + /* @@@ FIXME */ + /* Eventually we will want to be able to create + a journal file here. For now, only allow the + user to specify an existing inode to be the + journal file. */ + if (is_remount) { + printk(KERN_ERR "EXT3-fs: cannot specify " + "journal on remount\n"); + return 0; + } + set_opt (sbi->s_mount_opt, UPDATE_JOURNAL); + break; + case Opt_journal_inum: + if (is_remount) { + printk(KERN_ERR "EXT3-fs: cannot specify " + "journal on remount\n"); + return 0; + } + if (match_int(&args[0], &option)) + return 0; + *inum = option; + break; + case Opt_journal_dev: + if (is_remount) { + printk(KERN_ERR "EXT3-fs: cannot specify " + "journal on remount\n"); + return 0; + } + if (match_int(&args[0], &option)) + return 0; + *journal_devnum = option; + break; + case Opt_noload: + set_opt (sbi->s_mount_opt, NOLOAD); + break; + case Opt_commit: + if (match_int(&args[0], &option)) + return 0; + if (option < 0) + return 0; + if (option == 0) + option = JBD_DEFAULT_MAX_COMMIT_AGE; + sbi->s_commit_interval = HZ * option; + break; + case Opt_data_journal: + data_opt = EXT3_MOUNT_JOURNAL_DATA; + goto datacheck; + case Opt_data_ordered: + data_opt = EXT3_MOUNT_ORDERED_DATA; + goto datacheck; + case Opt_data_writeback: + data_opt = EXT3_MOUNT_WRITEBACK_DATA; + datacheck: + if (is_remount) { + if ((sbi->s_mount_opt & EXT3_MOUNT_DATA_FLAGS) + != data_opt) { + printk(KERN_ERR + "EXT3-fs: cannot change data " + "mode on remount\n"); + return 0; + } + } else { + sbi->s_mount_opt &= ~EXT3_MOUNT_DATA_FLAGS; + sbi->s_mount_opt |= data_opt; + } + break; +#ifdef CONFIG_QUOTA + case Opt_usrjquota: + qtype = USRQUOTA; + goto set_qf_name; + case Opt_grpjquota: + qtype = GRPQUOTA; +set_qf_name: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR + "EXT3-fs: Cannot change journalled " + "quota options when quota turned on.\n"); + return 0; + } + qname = match_strdup(&args[0]); + if (!qname) { + printk(KERN_ERR + "EXT3-fs: not enough memory for " + "storing quotafile name.\n"); + return 0; + } + if (sbi->s_qf_names[qtype] && + strcmp(sbi->s_qf_names[qtype], qname)) { + printk(KERN_ERR + "EXT3-fs: %s quota file already " + "specified.\n", QTYPE2NAME(qtype)); + kfree(qname); + return 0; + } + sbi->s_qf_names[qtype] = qname; + if (strchr(sbi->s_qf_names[qtype], '/')) { + printk(KERN_ERR + "EXT3-fs: quotafile must be on " + "filesystem root.\n"); + kfree(sbi->s_qf_names[qtype]); + sbi->s_qf_names[qtype] = NULL; + return 0; + } + set_opt(sbi->s_mount_opt, QUOTA); + break; + case Opt_offusrjquota: + qtype = USRQUOTA; + goto clear_qf_name; + case Opt_offgrpjquota: + qtype = GRPQUOTA; +clear_qf_name: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR "EXT3-fs: Cannot change " + "journalled quota options when " + "quota turned on.\n"); + return 0; + } + /* + * The space will be released later when all options + * are confirmed to be correct + */ + sbi->s_qf_names[qtype] = NULL; + break; + case Opt_jqfmt_vfsold: + sbi->s_jquota_fmt = QFMT_VFS_OLD; + break; + case Opt_jqfmt_vfsv0: + sbi->s_jquota_fmt = QFMT_VFS_V0; + break; + case Opt_quota: + case Opt_usrquota: + set_opt(sbi->s_mount_opt, QUOTA); + set_opt(sbi->s_mount_opt, USRQUOTA); + break; + case Opt_grpquota: + set_opt(sbi->s_mount_opt, QUOTA); + set_opt(sbi->s_mount_opt, GRPQUOTA); + break; + case Opt_noquota: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR "EXT3-fs: Cannot change quota " + "options when quota turned on.\n"); + return 0; + } + clear_opt(sbi->s_mount_opt, QUOTA); + clear_opt(sbi->s_mount_opt, USRQUOTA); + clear_opt(sbi->s_mount_opt, GRPQUOTA); + break; +#else + case Opt_quota: + case Opt_usrquota: + case Opt_grpquota: + case Opt_usrjquota: + case Opt_grpjquota: + case Opt_offusrjquota: + case Opt_offgrpjquota: + case Opt_jqfmt_vfsold: + case Opt_jqfmt_vfsv0: + printk(KERN_ERR + "EXT3-fs: journalled quota options not " + "supported.\n"); + break; + case Opt_noquota: + break; +#endif + case Opt_abort: + set_opt(sbi->s_mount_opt, ABORT); + break; + case Opt_barrier: + if (match_int(&args[0], &option)) + return 0; + if (option) + set_opt(sbi->s_mount_opt, BARRIER); + else + clear_opt(sbi->s_mount_opt, BARRIER); + break; + case Opt_ignore: + break; + case Opt_resize: + if (!is_remount) { + printk("EXT3-fs: resize option only available " + "for remount\n"); + return 0; + } + if (match_int(&args[0], &option) != 0) + return 0; + *n_blocks_count = option; + break; + case Opt_nobh: + set_opt(sbi->s_mount_opt, NOBH); + break; + case Opt_bh: + clear_opt(sbi->s_mount_opt, NOBH); + break; + default: + printk (KERN_ERR + "EXT3-fs: Unrecognized mount option \"%s\" " + "or missing value\n", p); + return 0; + } + } +#ifdef CONFIG_QUOTA + if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) { + if ((sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA) && + sbi->s_qf_names[USRQUOTA]) + clear_opt(sbi->s_mount_opt, USRQUOTA); + + if ((sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA) && + sbi->s_qf_names[GRPQUOTA]) + clear_opt(sbi->s_mount_opt, GRPQUOTA); + + if ((sbi->s_qf_names[USRQUOTA] && + (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA)) || + (sbi->s_qf_names[GRPQUOTA] && + (sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA))) { + printk(KERN_ERR "EXT3-fs: old and new quota " + "format mixing.\n"); + return 0; + } + + if (!sbi->s_jquota_fmt) { + printk(KERN_ERR "EXT3-fs: journalled quota format " + "not specified.\n"); + return 0; + } + } else { + if (sbi->s_jquota_fmt) { + printk(KERN_ERR "EXT3-fs: journalled quota format " + "specified with no journalling " + "enabled.\n"); + return 0; + } + } +#endif + return 1; +} + +static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, + int read_only) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + int res = 0; + + if (le32_to_cpu(es->s_rev_level) > EXT3_MAX_SUPP_REV) { + printk (KERN_ERR "EXT3-fs warning: revision level too high, " + "forcing read-only mode\n"); + res = MS_RDONLY; + } + if (read_only) + return res; + if (!(sbi->s_mount_state & EXT3_VALID_FS)) + printk (KERN_WARNING "EXT3-fs warning: mounting unchecked fs, " + "running e2fsck is recommended\n"); + else if ((sbi->s_mount_state & EXT3_ERROR_FS)) + printk (KERN_WARNING + "EXT3-fs warning: mounting fs with errors, " + "running e2fsck is recommended\n"); + else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 && + le16_to_cpu(es->s_mnt_count) >= + (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count)) + printk (KERN_WARNING + "EXT3-fs warning: maximal mount count reached, " + "running e2fsck is recommended\n"); + else if (le32_to_cpu(es->s_checkinterval) && + (le32_to_cpu(es->s_lastcheck) + + le32_to_cpu(es->s_checkinterval) <= get_seconds())) + printk (KERN_WARNING + "EXT3-fs warning: checktime reached, " + "running e2fsck is recommended\n"); +#if 0 + /* @@@ We _will_ want to clear the valid bit if we find + inconsistencies, to force a fsck at reboot. But for + a plain journaled filesystem we can keep it set as + valid forever! :) */ + es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT3_VALID_FS); +#endif + if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) + es->s_max_mnt_count = cpu_to_le16(EXT3_DFL_MAX_MNT_COUNT); + es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1); + es->s_mtime = cpu_to_le32(get_seconds()); + ext3_update_dynamic_rev(sb); + EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + + ext3_commit_super(sb, es, 1); + if (test_opt(sb, DEBUG)) + printk(KERN_INFO "[EXT3 FS bs=%lu, gc=%lu, " + "bpg=%lu, ipg=%lu, mo=%04lx]\n", + sb->s_blocksize, + sbi->s_groups_count, + EXT3_BLOCKS_PER_GROUP(sb), + EXT3_INODES_PER_GROUP(sb), + sbi->s_mount_opt); + + printk(KERN_INFO "EXT3 FS on %s, ", sb->s_id); + if (EXT3_SB(sb)->s_journal->j_inode == NULL) { + char b[BDEVNAME_SIZE]; + + printk("external journal on %s\n", + bdevname(EXT3_SB(sb)->s_journal->j_dev, b)); + } else { + printk("internal journal\n"); + } + return res; +} + +/* Called at mount-time, super-block is locked */ +static int ext3_check_descriptors (struct super_block * sb) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + ext3_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); + ext3_fsblk_t last_block; + struct ext3_group_desc * gdp = NULL; + int desc_block = 0; + int i; + + ext3_debug ("Checking group descriptors"); + + for (i = 0; i < sbi->s_groups_count; i++) + { + if (i == sbi->s_groups_count - 1) + last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; + else + last_block = first_block + + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + + if ((i % EXT3_DESC_PER_BLOCK(sb)) == 0) + gdp = (struct ext3_group_desc *) + sbi->s_group_desc[desc_block++]->b_data; + if (le32_to_cpu(gdp->bg_block_bitmap) < first_block || + le32_to_cpu(gdp->bg_block_bitmap) > last_block) + { + ext3_error (sb, "ext3_check_descriptors", + "Block bitmap for group %d" + " not in group (block %lu)!", + i, (unsigned long) + le32_to_cpu(gdp->bg_block_bitmap)); + return 0; + } + if (le32_to_cpu(gdp->bg_inode_bitmap) < first_block || + le32_to_cpu(gdp->bg_inode_bitmap) > last_block) + { + ext3_error (sb, "ext3_check_descriptors", + "Inode bitmap for group %d" + " not in group (block %lu)!", + i, (unsigned long) + le32_to_cpu(gdp->bg_inode_bitmap)); + return 0; + } + if (le32_to_cpu(gdp->bg_inode_table) < first_block || + le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group > + last_block) + { + ext3_error (sb, "ext3_check_descriptors", + "Inode table for group %d" + " not in group (block %lu)!", + i, (unsigned long) + le32_to_cpu(gdp->bg_inode_table)); + return 0; + } + first_block += EXT3_BLOCKS_PER_GROUP(sb); + gdp++; + } + + sbi->s_es->s_free_blocks_count=cpu_to_le32(ext3_count_free_blocks(sb)); + sbi->s_es->s_free_inodes_count=cpu_to_le32(ext3_count_free_inodes(sb)); + return 1; +} + + +/* ext3_orphan_cleanup() walks a singly-linked list of inodes (starting at + * the superblock) which were deleted from all directories, but held open by + * a process at the time of a crash. We walk the list and try to delete these + * inodes at recovery time (only with a read-write filesystem). + * + * In order to keep the orphan inode chain consistent during traversal (in + * case of crash during recovery), we link each inode into the superblock + * orphan list_head and handle it the same way as an inode deletion during + * normal operation (which journals the operations for us). + * + * We only do an iget() and an iput() on each inode, which is very safe if we + * accidentally point at an in-use or already deleted inode. The worst that + * can happen in this case is that we get a "bit already cleared" message from + * ext3_free_inode(). The only reason we would point at a wrong inode is if + * e2fsck was run on this filesystem, and it must have already done the orphan + * inode cleanup for us, so we can safely abort without any further action. + */ +static void ext3_orphan_cleanup (struct super_block * sb, + struct ext3_super_block * es) +{ + unsigned int s_flags = sb->s_flags; + int nr_orphans = 0, nr_truncates = 0; +#ifdef CONFIG_QUOTA + int i; +#endif + if (!es->s_last_orphan) { + jbd_debug(4, "no orphan inodes to clean up\n"); + return; + } + + if (EXT3_SB(sb)->s_mount_state & EXT3_ERROR_FS) { + if (es->s_last_orphan) + jbd_debug(1, "Errors on filesystem, " + "clearing orphan list.\n"); + es->s_last_orphan = 0; + jbd_debug(1, "Skipping orphan recovery on fs with errors.\n"); + return; + } + + if (s_flags & MS_RDONLY) { + printk(KERN_INFO "EXT3-fs: %s: orphan cleanup on readonly fs\n", + sb->s_id); + sb->s_flags &= ~MS_RDONLY; + } +#ifdef CONFIG_QUOTA + /* Needed for iput() to work correctly and not trash data */ + sb->s_flags |= MS_ACTIVE; + /* Turn on quotas so that they are updated correctly */ + for (i = 0; i < MAXQUOTAS; i++) { + if (EXT3_SB(sb)->s_qf_names[i]) { + int ret = ext3_quota_on_mount(sb, i); + if (ret < 0) + printk(KERN_ERR + "EXT3-fs: Cannot turn on journalled " + "quota: error %d\n", ret); + } + } +#endif + + while (es->s_last_orphan) { + struct inode *inode; + + if (!(inode = + ext3_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) { + es->s_last_orphan = 0; + break; + } + + list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan); + DQUOT_INIT(inode); + if (inode->i_nlink) { + printk(KERN_DEBUG + "%s: truncating inode %lu to %Ld bytes\n", + __FUNCTION__, inode->i_ino, inode->i_size); + jbd_debug(2, "truncating inode %lu to %Ld bytes\n", + inode->i_ino, inode->i_size); + ext3_truncate(inode); + nr_truncates++; + } else { + printk(KERN_DEBUG + "%s: deleting unreferenced inode %lu\n", + __FUNCTION__, inode->i_ino); + jbd_debug(2, "deleting unreferenced inode %lu\n", + inode->i_ino); + nr_orphans++; + } + iput(inode); /* The delete magic happens here! */ + } + +#define PLURAL(x) (x), ((x)==1) ? "" : "s" + + if (nr_orphans) + printk(KERN_INFO "EXT3-fs: %s: %d orphan inode%s deleted\n", + sb->s_id, PLURAL(nr_orphans)); + if (nr_truncates) + printk(KERN_INFO "EXT3-fs: %s: %d truncate%s cleaned up\n", + sb->s_id, PLURAL(nr_truncates)); +#ifdef CONFIG_QUOTA + /* Turn quotas off */ + for (i = 0; i < MAXQUOTAS; i++) { + if (sb_dqopt(sb)->files[i]) + vfs_quota_off(sb, i); + } +#endif + sb->s_flags = s_flags; /* Restore MS_RDONLY status */ +} + +#define log2(n) ffz(~(n)) + +/* + * Maximal file size. There is a direct, and {,double-,triple-}indirect + * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. + * We need to be 1 filesystem block less than the 2^32 sector limit. + */ +static loff_t ext3_max_size(int bits) +{ + loff_t res = EXT3_NDIR_BLOCKS; + /* This constant is calculated to be the largest file size for a + * dense, 4k-blocksize file such that the total number of + * sectors in the file, including data and all indirect blocks, + * does not exceed 2^32. */ + const loff_t upper_limit = 0x1ff7fffd000LL; + + res += 1LL << (bits-2); + res += 1LL << (2*(bits-2)); + res += 1LL << (3*(bits-2)); + res <<= bits; + if (res > upper_limit) + res = upper_limit; + return res; +} + +static ext3_fsblk_t descriptor_loc(struct super_block *sb, + ext3_fsblk_t logic_sb_block, + int nr) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + unsigned long bg, first_meta_bg; + int has_super = 0; + + first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg); + + if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_META_BG) || + nr < first_meta_bg) + return (logic_sb_block + nr + 1); + bg = sbi->s_desc_per_block * nr; + if (ext3_bg_has_super(sb, bg)) + has_super = 1; + return (has_super + ext3_group_first_block_no(sb, bg)); +} + + +static int ext3_fill_super (struct super_block *sb, void *data, int silent) +{ + struct buffer_head * bh; + struct ext3_super_block *es = NULL; + struct ext3_sb_info *sbi; + ext3_fsblk_t block; + ext3_fsblk_t sb_block = get_sb_block(&data); + ext3_fsblk_t logic_sb_block; + unsigned long offset = 0; + unsigned int journal_inum = 0; + unsigned long journal_devnum = 0; + unsigned long def_mount_opts; + struct inode *root; + int blocksize; + int hblock; + int db_count; + int i; + int needs_recovery; + __le32 features; + + sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); + if (!sbi) + return -ENOMEM; + sb->s_fs_info = sbi; + sbi->s_mount_opt = 0; + sbi->s_resuid = EXT3_DEF_RESUID; + sbi->s_resgid = EXT3_DEF_RESGID; + + unlock_kernel(); + + blocksize = sb_min_blocksize(sb, EXT3_MIN_BLOCK_SIZE); + if (!blocksize) { + printk(KERN_ERR "EXT3-fs: unable to set blocksize\n"); + goto out_fail; + } + + /* + * The ext3 superblock will not be buffer aligned for other than 1kB + * block sizes. We need to calculate the offset from buffer start. + */ + if (blocksize != EXT3_MIN_BLOCK_SIZE) { + logic_sb_block = (sb_block * EXT3_MIN_BLOCK_SIZE) / blocksize; + offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize; + } else { + logic_sb_block = sb_block; + } + + if (!(bh = sb_bread(sb, logic_sb_block))) { + printk (KERN_ERR "EXT3-fs: unable to read superblock\n"); + goto out_fail; + } + /* + * Note: s_es must be initialized as soon as possible because + * some ext3 macro-instructions depend on its value + */ + es = (struct ext3_super_block *) (((char *)bh->b_data) + offset); + sbi->s_es = es; + sb->s_magic = le16_to_cpu(es->s_magic); + if (sb->s_magic != EXT3_SUPER_MAGIC) + goto cantfind_ext3; + + /* Set defaults before we parse the mount options */ + def_mount_opts = le32_to_cpu(es->s_default_mount_opts); + if (def_mount_opts & EXT3_DEFM_DEBUG) + set_opt(sbi->s_mount_opt, DEBUG); + if (def_mount_opts & EXT3_DEFM_BSDGROUPS) + set_opt(sbi->s_mount_opt, GRPID); + if (def_mount_opts & EXT3_DEFM_UID16) + set_opt(sbi->s_mount_opt, NO_UID32); + if (def_mount_opts & EXT3_DEFM_XATTR_USER) + set_opt(sbi->s_mount_opt, XATTR_USER); + if (def_mount_opts & EXT3_DEFM_ACL) + set_opt(sbi->s_mount_opt, POSIX_ACL); + if ((def_mount_opts & EXT3_DEFM_JMODE) == EXT3_DEFM_JMODE_DATA) + sbi->s_mount_opt |= EXT3_MOUNT_JOURNAL_DATA; + else if ((def_mount_opts & EXT3_DEFM_JMODE) == EXT3_DEFM_JMODE_ORDERED) + sbi->s_mount_opt |= EXT3_MOUNT_ORDERED_DATA; + else if ((def_mount_opts & EXT3_DEFM_JMODE) == EXT3_DEFM_JMODE_WBACK) + sbi->s_mount_opt |= EXT3_MOUNT_WRITEBACK_DATA; + + if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_PANIC) + set_opt(sbi->s_mount_opt, ERRORS_PANIC); + else if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_RO) + set_opt(sbi->s_mount_opt, ERRORS_RO); + + sbi->s_resuid = le16_to_cpu(es->s_def_resuid); + sbi->s_resgid = le16_to_cpu(es->s_def_resgid); + + set_opt(sbi->s_mount_opt, RESERVATION); + + if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum, + NULL, 0)) + goto failed_mount; + + sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + ((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + + if (le32_to_cpu(es->s_rev_level) == EXT3_GOOD_OLD_REV && + (EXT3_HAS_COMPAT_FEATURE(sb, ~0U) || + EXT3_HAS_RO_COMPAT_FEATURE(sb, ~0U) || + EXT3_HAS_INCOMPAT_FEATURE(sb, ~0U))) + printk(KERN_WARNING + "EXT3-fs warning: feature flags set on rev 0 fs, " + "running e2fsck is recommended\n"); + /* + * Check feature flags regardless of the revision level, since we + * previously didn't change the revision level when setting the flags, + * so there is a chance incompat flags are set on a rev 0 filesystem. + */ + features = EXT3_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP); + if (features) { + printk(KERN_ERR "EXT3-fs: %s: couldn't mount because of " + "unsupported optional features (%x).\n", + sb->s_id, le32_to_cpu(features)); + goto failed_mount; + } + features = EXT3_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP); + if (!(sb->s_flags & MS_RDONLY) && features) { + printk(KERN_ERR "EXT3-fs: %s: couldn't mount RDWR because of " + "unsupported optional features (%x).\n", + sb->s_id, le32_to_cpu(features)); + goto failed_mount; + } + blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size); + + if (blocksize < EXT3_MIN_BLOCK_SIZE || + blocksize > EXT3_MAX_BLOCK_SIZE) { + printk(KERN_ERR + "EXT3-fs: Unsupported filesystem blocksize %d on %s.\n", + blocksize, sb->s_id); + goto failed_mount; + } + + hblock = bdev_hardsect_size(sb->s_bdev); + if (sb->s_blocksize != blocksize) { + /* + * Make sure the blocksize for the filesystem is larger + * than the hardware sectorsize for the machine. + */ + if (blocksize < hblock) { + printk(KERN_ERR "EXT3-fs: blocksize %d too small for " + "device blocksize %d.\n", blocksize, hblock); + goto failed_mount; + } + + brelse (bh); + sb_set_blocksize(sb, blocksize); + logic_sb_block = (sb_block * EXT3_MIN_BLOCK_SIZE) / blocksize; + offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize; + bh = sb_bread(sb, logic_sb_block); + if (!bh) { + printk(KERN_ERR + "EXT3-fs: Can't read superblock on 2nd try.\n"); + goto failed_mount; + } + es = (struct ext3_super_block *)(((char *)bh->b_data) + offset); + sbi->s_es = es; + if (es->s_magic != cpu_to_le16(EXT3_SUPER_MAGIC)) { + printk (KERN_ERR + "EXT3-fs: Magic mismatch, very weird !\n"); + goto failed_mount; + } + } + + sb->s_maxbytes = ext3_max_size(sb->s_blocksize_bits); + + if (le32_to_cpu(es->s_rev_level) == EXT3_GOOD_OLD_REV) { + sbi->s_inode_size = EXT3_GOOD_OLD_INODE_SIZE; + sbi->s_first_ino = EXT3_GOOD_OLD_FIRST_INO; + } else { + sbi->s_inode_size = le16_to_cpu(es->s_inode_size); + sbi->s_first_ino = le32_to_cpu(es->s_first_ino); + if ((sbi->s_inode_size < EXT3_GOOD_OLD_INODE_SIZE) || + (sbi->s_inode_size & (sbi->s_inode_size - 1)) || + (sbi->s_inode_size > blocksize)) { + printk (KERN_ERR + "EXT3-fs: unsupported inode size: %d\n", + sbi->s_inode_size); + goto failed_mount; + } + } + sbi->s_frag_size = EXT3_MIN_FRAG_SIZE << + le32_to_cpu(es->s_log_frag_size); + if (blocksize != sbi->s_frag_size) { + printk(KERN_ERR + "EXT3-fs: fragsize %lu != blocksize %u (unsupported)\n", + sbi->s_frag_size, blocksize); + goto failed_mount; + } + sbi->s_frags_per_block = 1; + sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group); + sbi->s_frags_per_group = le32_to_cpu(es->s_frags_per_group); + sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group); + if (EXT3_INODE_SIZE(sb) == 0) + goto cantfind_ext3; + sbi->s_inodes_per_block = blocksize / EXT3_INODE_SIZE(sb); + if (sbi->s_inodes_per_block == 0) + goto cantfind_ext3; + sbi->s_itb_per_group = sbi->s_inodes_per_group / + sbi->s_inodes_per_block; + sbi->s_desc_per_block = blocksize / sizeof(struct ext3_group_desc); + sbi->s_sbh = bh; + sbi->s_mount_state = le16_to_cpu(es->s_state); + sbi->s_addr_per_block_bits = log2(EXT3_ADDR_PER_BLOCK(sb)); + sbi->s_desc_per_block_bits = log2(EXT3_DESC_PER_BLOCK(sb)); + for (i=0; i < 4; i++) + sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); + sbi->s_def_hash_version = es->s_def_hash_version; + + if (sbi->s_blocks_per_group > blocksize * 8) { + printk (KERN_ERR + "EXT3-fs: #blocks per group too big: %lu\n", + sbi->s_blocks_per_group); + goto failed_mount; + } + if (sbi->s_frags_per_group > blocksize * 8) { + printk (KERN_ERR + "EXT3-fs: #fragments per group too big: %lu\n", + sbi->s_frags_per_group); + goto failed_mount; + } + if (sbi->s_inodes_per_group > blocksize * 8) { + printk (KERN_ERR + "EXT3-fs: #inodes per group too big: %lu\n", + sbi->s_inodes_per_group); + goto failed_mount; + } + + if (le32_to_cpu(es->s_blocks_count) > + (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { + printk(KERN_ERR "EXT3-fs: filesystem on %s:" + " too large to mount safely\n", sb->s_id); + if (sizeof(sector_t) < 8) + printk(KERN_WARNING "EXT3-fs: CONFIG_LBD not " + "enabled\n"); + goto failed_mount; + } + + if (EXT3_BLOCKS_PER_GROUP(sb) == 0) + goto cantfind_ext3; + sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) - + le32_to_cpu(es->s_first_data_block) - 1) + / EXT3_BLOCKS_PER_GROUP(sb)) + 1; + db_count = (sbi->s_groups_count + EXT3_DESC_PER_BLOCK(sb) - 1) / + EXT3_DESC_PER_BLOCK(sb); + sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *), + GFP_KERNEL); + if (sbi->s_group_desc == NULL) { + printk (KERN_ERR "EXT3-fs: not enough memory\n"); + goto failed_mount; + } + + bgl_lock_init(&sbi->s_blockgroup_lock); + + for (i = 0; i < db_count; i++) { + block = descriptor_loc(sb, logic_sb_block, i); + sbi->s_group_desc[i] = sb_bread(sb, block); + if (!sbi->s_group_desc[i]) { + printk (KERN_ERR "EXT3-fs: " + "can't read group descriptor %d\n", i); + db_count = i; + goto failed_mount2; + } + } + if (!ext3_check_descriptors (sb)) { + printk(KERN_ERR "EXT3-fs: group descriptors corrupted!\n"); + goto failed_mount2; + } + sbi->s_gdb_count = db_count; + get_random_bytes(&sbi->s_next_generation, sizeof(u32)); + spin_lock_init(&sbi->s_next_gen_lock); + + percpu_counter_init(&sbi->s_freeblocks_counter, + ext3_count_free_blocks(sb)); + percpu_counter_init(&sbi->s_freeinodes_counter, + ext3_count_free_inodes(sb)); + percpu_counter_init(&sbi->s_dirs_counter, + ext3_count_dirs(sb)); + + /* per fileystem reservation list head & lock */ + spin_lock_init(&sbi->s_rsv_window_lock); + sbi->s_rsv_window_root = RB_ROOT; + /* Add a single, static dummy reservation to the start of the + * reservation window list --- it gives us a placeholder for + * append-at-start-of-list which makes the allocation logic + * _much_ simpler. */ + sbi->s_rsv_window_head.rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + sbi->s_rsv_window_head.rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + sbi->s_rsv_window_head.rsv_alloc_hit = 0; + sbi->s_rsv_window_head.rsv_goal_size = 0; + ext3_rsv_window_add(sb, &sbi->s_rsv_window_head); + + /* + * set up enough so that it can read an inode + */ + sb->s_op = &ext3_sops; + sb->s_export_op = &ext3_export_ops; + sb->s_xattr = ext3_xattr_handlers; +#ifdef CONFIG_QUOTA + sb->s_qcop = &ext3_qctl_operations; + sb->dq_op = &ext3_quota_operations; +#endif + INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */ + + sb->s_root = NULL; + + needs_recovery = (es->s_last_orphan != 0 || + EXT3_HAS_INCOMPAT_FEATURE(sb, + EXT3_FEATURE_INCOMPAT_RECOVER)); + + /* + * The first inode we look at is the journal inode. Don't try + * root first: it may be modified in the journal! + */ + if (!test_opt(sb, NOLOAD) && + EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL)) { + if (ext3_load_journal(sb, es, journal_devnum)) + goto failed_mount3; + } else if (journal_inum) { + if (ext3_create_journal(sb, es, journal_inum)) + goto failed_mount3; + } else { + if (!silent) + printk (KERN_ERR + "ext3: No journal on filesystem on %s\n", + sb->s_id); + goto failed_mount3; + } + + /* We have now updated the journal if required, so we can + * validate the data journaling mode. */ + switch (test_opt(sb, DATA_FLAGS)) { + case 0: + /* No mode set, assume a default based on the journal + capabilities: ORDERED_DATA if the journal can + cope, else JOURNAL_DATA */ + if (journal_check_available_features + (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) + set_opt(sbi->s_mount_opt, ORDERED_DATA); + else + set_opt(sbi->s_mount_opt, JOURNAL_DATA); + break; + + case EXT3_MOUNT_ORDERED_DATA: + case EXT3_MOUNT_WRITEBACK_DATA: + if (!journal_check_available_features + (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) { + printk(KERN_ERR "EXT3-fs: Journal does not support " + "requested data journaling mode\n"); + goto failed_mount4; + } + default: + break; + } + + if (test_opt(sb, NOBH)) { + if (!(test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)) { + printk(KERN_WARNING "EXT3-fs: Ignoring nobh option - " + "its supported only with writeback mode\n"); + clear_opt(sbi->s_mount_opt, NOBH); + } + } + /* + * The journal_load will have done any necessary log recovery, + * so we can safely mount the rest of the filesystem now. + */ + + root = iget(sb, EXT3_ROOT_INO); + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + printk(KERN_ERR "EXT3-fs: get root inode failed\n"); + iput(root); + goto failed_mount4; + } + if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { + dput(sb->s_root); + sb->s_root = NULL; + printk(KERN_ERR "EXT3-fs: corrupt root inode, run e2fsck\n"); + goto failed_mount4; + } + + ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY); + /* + * akpm: core read_super() calls in here with the superblock locked. + * That deadlocks, because orphan cleanup needs to lock the superblock + * in numerous places. Here we just pop the lock - it's relatively + * harmless, because we are now ready to accept write_super() requests, + * and aviro says that's the only reason for hanging onto the + * superblock lock. + */ + EXT3_SB(sb)->s_mount_state |= EXT3_ORPHAN_FS; + ext3_orphan_cleanup(sb, es); + EXT3_SB(sb)->s_mount_state &= ~EXT3_ORPHAN_FS; + if (needs_recovery) + printk (KERN_INFO "EXT3-fs: recovery complete.\n"); + ext3_mark_recovery_complete(sb, es); + printk (KERN_INFO "EXT3-fs: mounted filesystem with %s data mode.\n", + test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA ? "journal": + test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA ? "ordered": + "writeback"); + + lock_kernel(); + return 0; + +cantfind_ext3: + if (!silent) + printk(KERN_ERR "VFS: Can't find ext3 filesystem on dev %s.\n", + sb->s_id); + goto failed_mount; + +failed_mount4: + journal_destroy(sbi->s_journal); +failed_mount3: + percpu_counter_destroy(&sbi->s_freeblocks_counter); + percpu_counter_destroy(&sbi->s_freeinodes_counter); + percpu_counter_destroy(&sbi->s_dirs_counter); +failed_mount2: + for (i = 0; i < db_count; i++) + brelse(sbi->s_group_desc[i]); + kfree(sbi->s_group_desc); +failed_mount: +#ifdef CONFIG_QUOTA + for (i = 0; i < MAXQUOTAS; i++) + kfree(sbi->s_qf_names[i]); +#endif + ext3_blkdev_remove(sbi); + brelse(bh); +out_fail: + sb->s_fs_info = NULL; + kfree(sbi); + lock_kernel(); + return -EINVAL; +} + +/* + * Setup any per-fs journal parameters now. We'll do this both on + * initial mount, once the journal has been initialised but before we've + * done any recovery; and again on any subsequent remount. + */ +static void ext3_init_journal_params(struct super_block *sb, journal_t *journal) +{ + struct ext3_sb_info *sbi = EXT3_SB(sb); + + if (sbi->s_commit_interval) + journal->j_commit_interval = sbi->s_commit_interval; + /* We could also set up an ext3-specific default for the commit + * interval here, but for now we'll just fall back to the jbd + * default. */ + + spin_lock(&journal->j_state_lock); + if (test_opt(sb, BARRIER)) + journal->j_flags |= JFS_BARRIER; + else + journal->j_flags &= ~JFS_BARRIER; + spin_unlock(&journal->j_state_lock); +} + +static journal_t *ext3_get_journal(struct super_block *sb, + unsigned int journal_inum) +{ + struct inode *journal_inode; + journal_t *journal; + + /* First, test for the existence of a valid inode on disk. Bad + * things happen if we iget() an unused inode, as the subsequent + * iput() will try to delete it. */ + + journal_inode = iget(sb, journal_inum); + if (!journal_inode) { + printk(KERN_ERR "EXT3-fs: no journal found.\n"); + return NULL; + } + if (!journal_inode->i_nlink) { + make_bad_inode(journal_inode); + iput(journal_inode); + printk(KERN_ERR "EXT3-fs: journal inode is deleted.\n"); + return NULL; + } + + jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", + journal_inode, journal_inode->i_size); + if (is_bad_inode(journal_inode) || !S_ISREG(journal_inode->i_mode)) { + printk(KERN_ERR "EXT3-fs: invalid journal inode.\n"); + iput(journal_inode); + return NULL; + } + + journal = journal_init_inode(journal_inode); + if (!journal) { + printk(KERN_ERR "EXT3-fs: Could not load journal inode\n"); + iput(journal_inode); + return NULL; + } + journal->j_private = sb; + ext3_init_journal_params(sb, journal); + return journal; +} + +static journal_t *ext3_get_dev_journal(struct super_block *sb, + dev_t j_dev) +{ + struct buffer_head * bh; + journal_t *journal; + ext3_fsblk_t start; + ext3_fsblk_t len; + int hblock, blocksize; + ext3_fsblk_t sb_block; + unsigned long offset; + struct ext3_super_block * es; + struct block_device *bdev; + + bdev = ext3_blkdev_get(j_dev); + if (bdev == NULL) + return NULL; + + if (bd_claim(bdev, sb)) { + printk(KERN_ERR + "EXT3: failed to claim external journal device.\n"); + blkdev_put(bdev); + return NULL; + } + + blocksize = sb->s_blocksize; + hblock = bdev_hardsect_size(bdev); + if (blocksize < hblock) { + printk(KERN_ERR + "EXT3-fs: blocksize too small for journal device.\n"); + goto out_bdev; + } + + sb_block = EXT3_MIN_BLOCK_SIZE / blocksize; + offset = EXT3_MIN_BLOCK_SIZE % blocksize; + set_blocksize(bdev, blocksize); + if (!(bh = __bread(bdev, sb_block, blocksize))) { + printk(KERN_ERR "EXT3-fs: couldn't read superblock of " + "external journal\n"); + goto out_bdev; + } + + es = (struct ext3_super_block *) (((char *)bh->b_data) + offset); + if ((le16_to_cpu(es->s_magic) != EXT3_SUPER_MAGIC) || + !(le32_to_cpu(es->s_feature_incompat) & + EXT3_FEATURE_INCOMPAT_JOURNAL_DEV)) { + printk(KERN_ERR "EXT3-fs: external journal has " + "bad superblock\n"); + brelse(bh); + goto out_bdev; + } + + if (memcmp(EXT3_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) { + printk(KERN_ERR "EXT3-fs: journal UUID does not match\n"); + brelse(bh); + goto out_bdev; + } + + len = le32_to_cpu(es->s_blocks_count); + start = sb_block + 1; + brelse(bh); /* we're done with the superblock */ + + journal = journal_init_dev(bdev, sb->s_bdev, + start, len, blocksize); + if (!journal) { + printk(KERN_ERR "EXT3-fs: failed to create device journal\n"); + goto out_bdev; + } + journal->j_private = sb; + ll_rw_block(READ, 1, &journal->j_sb_buffer); + wait_on_buffer(journal->j_sb_buffer); + if (!buffer_uptodate(journal->j_sb_buffer)) { + printk(KERN_ERR "EXT3-fs: I/O error on journal device\n"); + goto out_journal; + } + if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) { + printk(KERN_ERR "EXT3-fs: External journal has more than one " + "user (unsupported) - %d\n", + be32_to_cpu(journal->j_superblock->s_nr_users)); + goto out_journal; + } + EXT3_SB(sb)->journal_bdev = bdev; + ext3_init_journal_params(sb, journal); + return journal; +out_journal: + journal_destroy(journal); +out_bdev: + ext3_blkdev_put(bdev); + return NULL; +} + +static int ext3_load_journal(struct super_block *sb, + struct ext3_super_block *es, + unsigned long journal_devnum) +{ + journal_t *journal; + unsigned int journal_inum = le32_to_cpu(es->s_journal_inum); + dev_t journal_dev; + int err = 0; + int really_read_only; + + if (journal_devnum && + journal_devnum != le32_to_cpu(es->s_journal_dev)) { + printk(KERN_INFO "EXT3-fs: external journal device major/minor " + "numbers have changed\n"); + journal_dev = new_decode_dev(journal_devnum); + } else + journal_dev = new_decode_dev(le32_to_cpu(es->s_journal_dev)); + + really_read_only = bdev_read_only(sb->s_bdev); + + /* + * Are we loading a blank journal or performing recovery after a + * crash? For recovery, we need to check in advance whether we + * can get read-write access to the device. + */ + + if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER)) { + if (sb->s_flags & MS_RDONLY) { + printk(KERN_INFO "EXT3-fs: INFO: recovery " + "required on readonly filesystem.\n"); + if (really_read_only) { + printk(KERN_ERR "EXT3-fs: write access " + "unavailable, cannot proceed.\n"); + return -EROFS; + } + printk (KERN_INFO "EXT3-fs: write access will " + "be enabled during recovery.\n"); + } + } + + if (journal_inum && journal_dev) { + printk(KERN_ERR "EXT3-fs: filesystem has both journal " + "and inode journals!\n"); + return -EINVAL; + } + + if (journal_inum) { + if (!(journal = ext3_get_journal(sb, journal_inum))) + return -EINVAL; + } else { + if (!(journal = ext3_get_dev_journal(sb, journal_dev))) + return -EINVAL; + } + + if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { + err = journal_update_format(journal); + if (err) { + printk(KERN_ERR "EXT3-fs: error updating journal.\n"); + journal_destroy(journal); + return err; + } + } + + if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER)) + err = journal_wipe(journal, !really_read_only); + if (!err) + err = journal_load(journal); + + if (err) { + printk(KERN_ERR "EXT3-fs: error loading journal.\n"); + journal_destroy(journal); + return err; + } + + EXT3_SB(sb)->s_journal = journal; + ext3_clear_journal_err(sb, es); + + if (journal_devnum && + journal_devnum != le32_to_cpu(es->s_journal_dev)) { + es->s_journal_dev = cpu_to_le32(journal_devnum); + sb->s_dirt = 1; + + /* Make sure we flush the recovery flag to disk. */ + ext3_commit_super(sb, es, 1); + } + + return 0; +} + +static int ext3_create_journal(struct super_block * sb, + struct ext3_super_block * es, + unsigned int journal_inum) +{ + journal_t *journal; + + if (sb->s_flags & MS_RDONLY) { + printk(KERN_ERR "EXT3-fs: readonly filesystem when trying to " + "create journal.\n"); + return -EROFS; + } + + if (!(journal = ext3_get_journal(sb, journal_inum))) + return -EINVAL; + + printk(KERN_INFO "EXT3-fs: creating new journal on inode %u\n", + journal_inum); + + if (journal_create(journal)) { + printk(KERN_ERR "EXT3-fs: error creating journal.\n"); + journal_destroy(journal); + return -EIO; + } + + EXT3_SB(sb)->s_journal = journal; + + ext3_update_dynamic_rev(sb); + EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + EXT3_SET_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL); + + es->s_journal_inum = cpu_to_le32(journal_inum); + sb->s_dirt = 1; + + /* Make sure we flush the recovery flag to disk. */ + ext3_commit_super(sb, es, 1); + + return 0; +} + +static void ext3_commit_super (struct super_block * sb, + struct ext3_super_block * es, + int sync) +{ + struct buffer_head *sbh = EXT3_SB(sb)->s_sbh; + + if (!sbh) + return; + es->s_wtime = cpu_to_le32(get_seconds()); + es->s_free_blocks_count = cpu_to_le32(ext3_count_free_blocks(sb)); + es->s_free_inodes_count = cpu_to_le32(ext3_count_free_inodes(sb)); + BUFFER_TRACE(sbh, "marking dirty"); + mark_buffer_dirty(sbh); + if (sync) + sync_dirty_buffer(sbh); +} + + +/* + * Have we just finished recovery? If so, and if we are mounting (or + * remounting) the filesystem readonly, then we will end up with a + * consistent fs on disk. Record that fact. + */ +static void ext3_mark_recovery_complete(struct super_block * sb, + struct ext3_super_block * es) +{ + journal_t *journal = EXT3_SB(sb)->s_journal; + + journal_lock_updates(journal); + journal_flush(journal); + if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER) && + sb->s_flags & MS_RDONLY) { + EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + sb->s_dirt = 0; + ext3_commit_super(sb, es, 1); + } + journal_unlock_updates(journal); +} + +/* + * If we are mounting (or read-write remounting) a filesystem whose journal + * has recorded an error from a previous lifetime, move that error to the + * main filesystem now. + */ +static void ext3_clear_journal_err(struct super_block * sb, + struct ext3_super_block * es) +{ + journal_t *journal; + int j_errno; + const char *errstr; + + journal = EXT3_SB(sb)->s_journal; + + /* + * Now check for any error status which may have been recorded in the + * journal by a prior ext3_error() or ext3_abort() + */ + + j_errno = journal_errno(journal); + if (j_errno) { + char nbuf[16]; + + errstr = ext3_decode_error(sb, j_errno, nbuf); + ext3_warning(sb, __FUNCTION__, "Filesystem error recorded " + "from previous mount: %s", errstr); + ext3_warning(sb, __FUNCTION__, "Marking fs in need of " + "filesystem check."); + + EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; + es->s_state |= cpu_to_le16(EXT3_ERROR_FS); + ext3_commit_super (sb, es, 1); + + journal_clear_err(journal); + } +} + +/* + * Force the running and committing transactions to commit, + * and wait on the commit. + */ +int ext3_force_commit(struct super_block *sb) +{ + journal_t *journal; + int ret; + + if (sb->s_flags & MS_RDONLY) + return 0; + + journal = EXT3_SB(sb)->s_journal; + sb->s_dirt = 0; + ret = ext3_journal_force_commit(journal); + return ret; +} + +/* + * Ext3 always journals updates to the superblock itself, so we don't + * have to propagate any other updates to the superblock on disk at this + * point. Just start an async writeback to get the buffers on their way + * to the disk. + * + * This implicitly triggers the writebehind on sync(). + */ + +static void ext3_write_super (struct super_block * sb) +{ + if (mutex_trylock(&sb->s_lock) != 0) + BUG(); + sb->s_dirt = 0; +} + +static int ext3_sync_fs(struct super_block *sb, int wait) +{ + tid_t target; + + sb->s_dirt = 0; + if (journal_start_commit(EXT3_SB(sb)->s_journal, &target)) { + if (wait) + log_wait_commit(EXT3_SB(sb)->s_journal, target); + } + return 0; +} + +/* + * LVM calls this function before a (read-only) snapshot is created. This + * gives us a chance to flush the journal completely and mark the fs clean. + */ +static void ext3_write_super_lockfs(struct super_block *sb) +{ + sb->s_dirt = 0; + + if (!(sb->s_flags & MS_RDONLY)) { + journal_t *journal = EXT3_SB(sb)->s_journal; + + /* Now we set up the journal barrier. */ + journal_lock_updates(journal); + journal_flush(journal); + + /* Journal blocked and flushed, clear needs_recovery flag. */ + EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1); + } +} + +/* + * Called by LVM after the snapshot is done. We need to reset the RECOVER + * flag here, even though the filesystem is not technically dirty yet. + */ +static void ext3_unlockfs(struct super_block *sb) +{ + if (!(sb->s_flags & MS_RDONLY)) { + lock_super(sb); + /* Reser the needs_recovery flag before the fs is unlocked. */ + EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1); + unlock_super(sb); + journal_unlock_updates(EXT3_SB(sb)->s_journal); + } +} + +static int ext3_remount (struct super_block * sb, int * flags, char * data) +{ + struct ext3_super_block * es; + struct ext3_sb_info *sbi = EXT3_SB(sb); + ext3_fsblk_t n_blocks_count = 0; + unsigned long old_sb_flags; + struct ext3_mount_options old_opts; + int err; +#ifdef CONFIG_QUOTA + int i; +#endif + + /* Store the original options */ + old_sb_flags = sb->s_flags; + old_opts.s_mount_opt = sbi->s_mount_opt; + old_opts.s_resuid = sbi->s_resuid; + old_opts.s_resgid = sbi->s_resgid; + old_opts.s_commit_interval = sbi->s_commit_interval; +#ifdef CONFIG_QUOTA + old_opts.s_jquota_fmt = sbi->s_jquota_fmt; + for (i = 0; i < MAXQUOTAS; i++) + old_opts.s_qf_names[i] = sbi->s_qf_names[i]; +#endif + + /* + * Allow the "check" option to be passed as a remount option. + */ + if (!parse_options(data, sb, NULL, NULL, &n_blocks_count, 1)) { + err = -EINVAL; + goto restore_opts; + } + + if (sbi->s_mount_opt & EXT3_MOUNT_ABORT) + ext3_abort(sb, __FUNCTION__, "Abort forced by user"); + + sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + ((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + + es = sbi->s_es; + + ext3_init_journal_params(sb, sbi->s_journal); + + if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY) || + n_blocks_count > le32_to_cpu(es->s_blocks_count)) { + if (sbi->s_mount_opt & EXT3_MOUNT_ABORT) { + err = -EROFS; + goto restore_opts; + } + + if (*flags & MS_RDONLY) { + /* + * First of all, the unconditional stuff we have to do + * to disable replay of the journal when we next remount + */ + sb->s_flags |= MS_RDONLY; + + /* + * OK, test if we are remounting a valid rw partition + * readonly, and if so set the rdonly flag and then + * mark the partition as valid again. + */ + if (!(es->s_state & cpu_to_le16(EXT3_VALID_FS)) && + (sbi->s_mount_state & EXT3_VALID_FS)) + es->s_state = cpu_to_le16(sbi->s_mount_state); + + ext3_mark_recovery_complete(sb, es); + } else { + __le32 ret; + if ((ret = EXT3_HAS_RO_COMPAT_FEATURE(sb, + ~EXT3_FEATURE_RO_COMPAT_SUPP))) { + printk(KERN_WARNING "EXT3-fs: %s: couldn't " + "remount RDWR because of unsupported " + "optional features (%x).\n", + sb->s_id, le32_to_cpu(ret)); + err = -EROFS; + goto restore_opts; + } + /* + * Mounting a RDONLY partition read-write, so reread + * and store the current valid flag. (It may have + * been changed by e2fsck since we originally mounted + * the partition.) + */ + ext3_clear_journal_err(sb, es); + sbi->s_mount_state = le16_to_cpu(es->s_state); + if ((err = ext3_group_extend(sb, es, n_blocks_count))) + goto restore_opts; + if (!ext3_setup_super (sb, es, 0)) + sb->s_flags &= ~MS_RDONLY; + } + } +#ifdef CONFIG_QUOTA + /* Release old quota file names */ + for (i = 0; i < MAXQUOTAS; i++) + if (old_opts.s_qf_names[i] && + old_opts.s_qf_names[i] != sbi->s_qf_names[i]) + kfree(old_opts.s_qf_names[i]); +#endif + return 0; +restore_opts: + sb->s_flags = old_sb_flags; + sbi->s_mount_opt = old_opts.s_mount_opt; + sbi->s_resuid = old_opts.s_resuid; + sbi->s_resgid = old_opts.s_resgid; + sbi->s_commit_interval = old_opts.s_commit_interval; +#ifdef CONFIG_QUOTA + sbi->s_jquota_fmt = old_opts.s_jquota_fmt; + for (i = 0; i < MAXQUOTAS; i++) { + if (sbi->s_qf_names[i] && + old_opts.s_qf_names[i] != sbi->s_qf_names[i]) + kfree(sbi->s_qf_names[i]); + sbi->s_qf_names[i] = old_opts.s_qf_names[i]; + } +#endif + return err; +} + +static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) +{ + struct super_block *sb = dentry->d_sb; + struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext3_super_block *es = sbi->s_es; + ext3_fsblk_t overhead; + int i; + + if (test_opt (sb, MINIX_DF)) + overhead = 0; + else { + unsigned long ngroups; + ngroups = EXT3_SB(sb)->s_groups_count; + smp_rmb(); + + /* + * Compute the overhead (FS structures) + */ + + /* + * All of the blocks before first_data_block are + * overhead + */ + overhead = le32_to_cpu(es->s_first_data_block); + + /* + * Add the overhead attributed to the superblock and + * block group descriptors. If the sparse superblocks + * feature is turned on, then not all groups have this. + */ + for (i = 0; i < ngroups; i++) { + overhead += ext3_bg_has_super(sb, i) + + ext3_bg_num_gdb(sb, i); + cond_resched(); + } + + /* + * Every block group has an inode bitmap, a block + * bitmap, and an inode table. + */ + overhead += (ngroups * (2 + EXT3_SB(sb)->s_itb_per_group)); + } + + buf->f_type = EXT3_SUPER_MAGIC; + buf->f_bsize = sb->s_blocksize; + buf->f_blocks = le32_to_cpu(es->s_blocks_count) - overhead; + buf->f_bfree = percpu_counter_sum(&sbi->s_freeblocks_counter); + buf->f_bavail = buf->f_bfree - le32_to_cpu(es->s_r_blocks_count); + if (buf->f_bfree < le32_to_cpu(es->s_r_blocks_count)) + buf->f_bavail = 0; + buf->f_files = le32_to_cpu(es->s_inodes_count); + buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter); + buf->f_namelen = EXT3_NAME_LEN; + return 0; +} + +/* Helper function for writing quotas on sync - we need to start transaction before quota file + * is locked for write. Otherwise the are possible deadlocks: + * Process 1 Process 2 + * ext3_create() quota_sync() + * journal_start() write_dquot() + * DQUOT_INIT() down(dqio_mutex) + * down(dqio_mutex) journal_start() + * + */ + +#ifdef CONFIG_QUOTA + +static inline struct inode *dquot_to_inode(struct dquot *dquot) +{ + return sb_dqopt(dquot->dq_sb)->files[dquot->dq_type]; +} + +static int ext3_dquot_initialize(struct inode *inode, int type) +{ + handle_t *handle; + int ret, err; + + /* We may create quota structure so we need to reserve enough blocks */ + handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_initialize(inode, type); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext3_dquot_drop(struct inode *inode) +{ + handle_t *handle; + int ret, err; + + /* We may delete quota structure so we need to reserve enough blocks */ + handle = ext3_journal_start(inode, 2*EXT3_QUOTA_DEL_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_drop(inode); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext3_write_dquot(struct dquot *dquot) +{ + int ret, err; + handle_t *handle; + struct inode *inode; + + inode = dquot_to_inode(dquot); + handle = ext3_journal_start(inode, + EXT3_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_commit(dquot); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext3_acquire_dquot(struct dquot *dquot) +{ + int ret, err; + handle_t *handle; + + handle = ext3_journal_start(dquot_to_inode(dquot), + EXT3_QUOTA_INIT_BLOCKS(dquot->dq_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_acquire(dquot); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext3_release_dquot(struct dquot *dquot) +{ + int ret, err; + handle_t *handle; + + handle = ext3_journal_start(dquot_to_inode(dquot), + EXT3_QUOTA_DEL_BLOCKS(dquot->dq_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_release(dquot); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext3_mark_dquot_dirty(struct dquot *dquot) +{ + /* Are we journalling quotas? */ + if (EXT3_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] || + EXT3_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) { + dquot_mark_dquot_dirty(dquot); + return ext3_write_dquot(dquot); + } else { + return dquot_mark_dquot_dirty(dquot); + } +} + +static int ext3_write_info(struct super_block *sb, int type) +{ + int ret, err; + handle_t *handle; + + /* Data block + inode block */ + handle = ext3_journal_start(sb->s_root->d_inode, 2); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_commit_info(sb, type); + err = ext3_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +/* + * Turn on quotas during mount time - we need to find + * the quota file and such... + */ +static int ext3_quota_on_mount(struct super_block *sb, int type) +{ + return vfs_quota_on_mount(sb, EXT3_SB(sb)->s_qf_names[type], + EXT3_SB(sb)->s_jquota_fmt, type); +} + +/* + * Standard function to be called on quota_on + */ +static int ext3_quota_on(struct super_block *sb, int type, int format_id, + char *path) +{ + int err; + struct nameidata nd; + + if (!test_opt(sb, QUOTA)) + return -EINVAL; + /* Not journalling quota? */ + if (!EXT3_SB(sb)->s_qf_names[USRQUOTA] && + !EXT3_SB(sb)->s_qf_names[GRPQUOTA]) + return vfs_quota_on(sb, type, format_id, path); + err = path_lookup(path, LOOKUP_FOLLOW, &nd); + if (err) + return err; + /* Quotafile not on the same filesystem? */ + if (nd.mnt->mnt_sb != sb) { + path_release(&nd); + return -EXDEV; + } + /* Quotafile not of fs root? */ + if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode) + printk(KERN_WARNING + "EXT3-fs: Quota file not on filesystem root. " + "Journalled quota will not work.\n"); + path_release(&nd); + return vfs_quota_on(sb, type, format_id, path); +} + +/* Read data from quotafile - avoid pagecache and such because we cannot afford + * acquiring the locks... As quota files are never truncated and quota code + * itself serializes the operations (and noone else should touch the files) + * we don't have to be afraid of races */ +static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, + size_t len, loff_t off) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + sector_t blk = off >> EXT3_BLOCK_SIZE_BITS(sb); + int err = 0; + int offset = off & (sb->s_blocksize - 1); + int tocopy; + size_t toread; + struct buffer_head *bh; + loff_t i_size = i_size_read(inode); + + if (off > i_size) + return 0; + if (off+len > i_size) + len = i_size-off; + toread = len; + while (toread > 0) { + tocopy = sb->s_blocksize - offset < toread ? + sb->s_blocksize - offset : toread; + bh = ext3_bread(NULL, inode, blk, 0, &err); + if (err) + return err; + if (!bh) /* A hole? */ + memset(data, 0, tocopy); + else + memcpy(data, bh->b_data+offset, tocopy); + brelse(bh); + offset = 0; + toread -= tocopy; + data += tocopy; + blk++; + } + return len; +} + +/* Write to quotafile (we know the transaction is already started and has + * enough credits) */ +static ssize_t ext3_quota_write(struct super_block *sb, int type, + const char *data, size_t len, loff_t off) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + sector_t blk = off >> EXT3_BLOCK_SIZE_BITS(sb); + int err = 0; + int offset = off & (sb->s_blocksize - 1); + int tocopy; + int journal_quota = EXT3_SB(sb)->s_qf_names[type] != NULL; + size_t towrite = len; + struct buffer_head *bh; + handle_t *handle = journal_current_handle(); + + mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA); + while (towrite > 0) { + tocopy = sb->s_blocksize - offset < towrite ? + sb->s_blocksize - offset : towrite; + bh = ext3_bread(handle, inode, blk, 1, &err); + if (!bh) + goto out; + if (journal_quota) { + err = ext3_journal_get_write_access(handle, bh); + if (err) { + brelse(bh); + goto out; + } + } + lock_buffer(bh); + memcpy(bh->b_data+offset, data, tocopy); + flush_dcache_page(bh->b_page); + unlock_buffer(bh); + if (journal_quota) + err = ext3_journal_dirty_metadata(handle, bh); + else { + /* Always do at least ordered writes for quotas */ + err = ext3_journal_dirty_data(handle, bh); + mark_buffer_dirty(bh); + } + brelse(bh); + if (err) + goto out; + offset = 0; + towrite -= tocopy; + data += tocopy; + blk++; + } +out: + if (len == towrite) + return err; + if (inode->i_size < off+len-towrite) { + i_size_write(inode, off+len-towrite); + EXT3_I(inode)->i_disksize = inode->i_size; + } + inode->i_version++; + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + ext3_mark_inode_dirty(handle, inode); + mutex_unlock(&inode->i_mutex); + return len - towrite; +} + +#endif + +static int ext3_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data, struct vfsmount *mnt) +{ + return get_sb_bdev(fs_type, flags, dev_name, data, ext3_fill_super, mnt); +} + +static struct file_system_type ext3_fs_type = { + .owner = THIS_MODULE, + .name = "ext3", + .get_sb = ext3_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +static int __init init_ext3_fs(void) +{ + int err = init_ext3_xattr(); + if (err) + return err; + err = init_inodecache(); + if (err) + goto out1; + err = register_filesystem(&ext3_fs_type); + if (err) + goto out; + return 0; +out: + destroy_inodecache(); +out1: + exit_ext3_xattr(); + return err; +} + +static void __exit exit_ext3_fs(void) +{ + unregister_filesystem(&ext3_fs_type); + destroy_inodecache(); + exit_ext3_xattr(); +} + +MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); +MODULE_DESCRIPTION("Second Extended Filesystem with journaling extensions"); +MODULE_LICENSE("GPL"); +module_init(init_ext3_fs) +module_exit(exit_ext3_fs) diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c new file mode 100644 index 0000000..4f79122 --- /dev/null +++ b/fs/ext4/symlink.c @@ -0,0 +1,54 @@ +/* + * linux/fs/ext3/symlink.c + * + * Only fast symlinks left here - the rest is done by generic code. AV, 1999 + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/symlink.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * ext3 symlink handling code + */ + +#include +#include +#include +#include +#include "xattr.h" + +static void * ext3_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + struct ext3_inode_info *ei = EXT3_I(dentry->d_inode); + nd_set_link(nd, (char*)ei->i_data); + return NULL; +} + +struct inode_operations ext3_symlink_inode_operations = { + .readlink = generic_readlink, + .follow_link = page_follow_link_light, + .put_link = page_put_link, +#ifdef CONFIG_EXT3_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext3_listxattr, + .removexattr = generic_removexattr, +#endif +}; + +struct inode_operations ext3_fast_symlink_inode_operations = { + .readlink = generic_readlink, + .follow_link = ext3_follow_link, +#ifdef CONFIG_EXT3_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext3_listxattr, + .removexattr = generic_removexattr, +#endif +}; diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c new file mode 100644 index 0000000..f86f248 --- /dev/null +++ b/fs/ext4/xattr.c @@ -0,0 +1,1317 @@ +/* + * linux/fs/ext3/xattr.c + * + * Copyright (C) 2001-2003 Andreas Gruenbacher, + * + * Fix by Harrison Xing . + * Ext3 code with a lot of help from Eric Jarman . + * Extended attributes for symlinks and special files added per + * suggestion of Luka Renko . + * xattr consolidation Copyright (c) 2004 James Morris , + * Red Hat Inc. + * ea-in-inode support by Alex Tomas aka bzzz + * and Andreas Gruenbacher . + */ + +/* + * Extended attributes are stored directly in inodes (on file systems with + * inodes bigger than 128 bytes) and on additional disk blocks. The i_file_acl + * field contains the block number if an inode uses an additional block. All + * attributes must fit in the inode and one additional block. Blocks that + * contain the identical set of attributes may be shared among several inodes. + * Identical blocks are detected by keeping a cache of blocks that have + * recently been accessed. + * + * The attributes in inodes and on blocks have a different header; the entries + * are stored in the same format: + * + * +------------------+ + * | header | + * | entry 1 | | + * | entry 2 | | growing downwards + * | entry 3 | v + * | four null bytes | + * | . . . | + * | value 1 | ^ + * | value 3 | | growing upwards + * | value 2 | | + * +------------------+ + * + * The header is followed by multiple entry descriptors. In disk blocks, the + * entry descriptors are kept sorted. In inodes, they are unsorted. The + * attribute values are aligned to the end of the block in no specific order. + * + * Locking strategy + * ---------------- + * EXT3_I(inode)->i_file_acl is protected by EXT3_I(inode)->xattr_sem. + * EA blocks are only changed if they are exclusive to an inode, so + * holding xattr_sem also means that nothing but the EA block's reference + * count can change. Multiple writers to the same block are synchronized + * by the buffer lock. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include "xattr.h" +#include "acl.h" + +#define BHDR(bh) ((struct ext3_xattr_header *)((bh)->b_data)) +#define ENTRY(ptr) ((struct ext3_xattr_entry *)(ptr)) +#define BFIRST(bh) ENTRY(BHDR(bh)+1) +#define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0) + +#define IHDR(inode, raw_inode) \ + ((struct ext3_xattr_ibody_header *) \ + ((void *)raw_inode + \ + EXT3_GOOD_OLD_INODE_SIZE + \ + EXT3_I(inode)->i_extra_isize)) +#define IFIRST(hdr) ((struct ext3_xattr_entry *)((hdr)+1)) + +#ifdef EXT3_XATTR_DEBUG +# define ea_idebug(inode, f...) do { \ + printk(KERN_DEBUG "inode %s:%lu: ", \ + inode->i_sb->s_id, inode->i_ino); \ + printk(f); \ + printk("\n"); \ + } while (0) +# define ea_bdebug(bh, f...) do { \ + char b[BDEVNAME_SIZE]; \ + printk(KERN_DEBUG "block %s:%lu: ", \ + bdevname(bh->b_bdev, b), \ + (unsigned long) bh->b_blocknr); \ + printk(f); \ + printk("\n"); \ + } while (0) +#else +# define ea_idebug(f...) +# define ea_bdebug(f...) +#endif + +static void ext3_xattr_cache_insert(struct buffer_head *); +static struct buffer_head *ext3_xattr_cache_find(struct inode *, + struct ext3_xattr_header *, + struct mb_cache_entry **); +static void ext3_xattr_rehash(struct ext3_xattr_header *, + struct ext3_xattr_entry *); + +static struct mb_cache *ext3_xattr_cache; + +static struct xattr_handler *ext3_xattr_handler_map[] = { + [EXT3_XATTR_INDEX_USER] = &ext3_xattr_user_handler, +#ifdef CONFIG_EXT3_FS_POSIX_ACL + [EXT3_XATTR_INDEX_POSIX_ACL_ACCESS] = &ext3_xattr_acl_access_handler, + [EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext3_xattr_acl_default_handler, +#endif + [EXT3_XATTR_INDEX_TRUSTED] = &ext3_xattr_trusted_handler, +#ifdef CONFIG_EXT3_FS_SECURITY + [EXT3_XATTR_INDEX_SECURITY] = &ext3_xattr_security_handler, +#endif +}; + +struct xattr_handler *ext3_xattr_handlers[] = { + &ext3_xattr_user_handler, + &ext3_xattr_trusted_handler, +#ifdef CONFIG_EXT3_FS_POSIX_ACL + &ext3_xattr_acl_access_handler, + &ext3_xattr_acl_default_handler, +#endif +#ifdef CONFIG_EXT3_FS_SECURITY + &ext3_xattr_security_handler, +#endif + NULL +}; + +static inline struct xattr_handler * +ext3_xattr_handler(int name_index) +{ + struct xattr_handler *handler = NULL; + + if (name_index > 0 && name_index < ARRAY_SIZE(ext3_xattr_handler_map)) + handler = ext3_xattr_handler_map[name_index]; + return handler; +} + +/* + * Inode operation listxattr() + * + * dentry->d_inode->i_mutex: don't care + */ +ssize_t +ext3_listxattr(struct dentry *dentry, char *buffer, size_t size) +{ + return ext3_xattr_list(dentry->d_inode, buffer, size); +} + +static int +ext3_xattr_check_names(struct ext3_xattr_entry *entry, void *end) +{ + while (!IS_LAST_ENTRY(entry)) { + struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(entry); + if ((void *)next >= end) + return -EIO; + entry = next; + } + return 0; +} + +static inline int +ext3_xattr_check_block(struct buffer_head *bh) +{ + int error; + + if (BHDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + BHDR(bh)->h_blocks != cpu_to_le32(1)) + return -EIO; + error = ext3_xattr_check_names(BFIRST(bh), bh->b_data + bh->b_size); + return error; +} + +static inline int +ext3_xattr_check_entry(struct ext3_xattr_entry *entry, size_t size) +{ + size_t value_size = le32_to_cpu(entry->e_value_size); + + if (entry->e_value_block != 0 || value_size > size || + le16_to_cpu(entry->e_value_offs) + value_size > size) + return -EIO; + return 0; +} + +static int +ext3_xattr_find_entry(struct ext3_xattr_entry **pentry, int name_index, + const char *name, size_t size, int sorted) +{ + struct ext3_xattr_entry *entry; + size_t name_len; + int cmp = 1; + + if (name == NULL) + return -EINVAL; + name_len = strlen(name); + entry = *pentry; + for (; !IS_LAST_ENTRY(entry); entry = EXT3_XATTR_NEXT(entry)) { + cmp = name_index - entry->e_name_index; + if (!cmp) + cmp = name_len - entry->e_name_len; + if (!cmp) + cmp = memcmp(name, entry->e_name, name_len); + if (cmp <= 0 && (sorted || cmp == 0)) + break; + } + *pentry = entry; + if (!cmp && ext3_xattr_check_entry(entry, size)) + return -EIO; + return cmp ? -ENODATA : 0; +} + +static int +ext3_xattr_block_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + struct ext3_xattr_entry *entry; + size_t size; + int error; + + ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld", + name_index, name, buffer, (long)buffer_size); + + error = -ENODATA; + if (!EXT3_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %u", EXT3_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); + if (ext3_xattr_check_block(bh)) { +bad_block: ext3_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block "E3FSBLK, inode->i_ino, + EXT3_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + ext3_xattr_cache_insert(bh); + entry = BFIRST(bh); + error = ext3_xattr_find_entry(&entry, name_index, name, bh->b_size, 1); + if (error == -EIO) + goto bad_block; + if (error) + goto cleanup; + size = le32_to_cpu(entry->e_value_size); + if (buffer) { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + memcpy(buffer, bh->b_data + le16_to_cpu(entry->e_value_offs), + size); + } + error = size; + +cleanup: + brelse(bh); + return error; +} + +static int +ext3_xattr_ibody_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + struct ext3_xattr_ibody_header *header; + struct ext3_xattr_entry *entry; + struct ext3_inode *raw_inode; + struct ext3_iloc iloc; + size_t size; + void *end; + int error; + + if (!(EXT3_I(inode)->i_state & EXT3_STATE_XATTR)) + return -ENODATA; + error = ext3_get_inode_loc(inode, &iloc); + if (error) + return error; + raw_inode = ext3_raw_inode(&iloc); + header = IHDR(inode, raw_inode); + entry = IFIRST(header); + end = (void *)raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; + error = ext3_xattr_check_names(entry, end); + if (error) + goto cleanup; + error = ext3_xattr_find_entry(&entry, name_index, name, + end - (void *)entry, 0); + if (error) + goto cleanup; + size = le32_to_cpu(entry->e_value_size); + if (buffer) { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + memcpy(buffer, (void *)IFIRST(header) + + le16_to_cpu(entry->e_value_offs), size); + } + error = size; + +cleanup: + brelse(iloc.bh); + return error; +} + +/* + * ext3_xattr_get() + * + * Copy an extended attribute into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext3_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + int error; + + down_read(&EXT3_I(inode)->xattr_sem); + error = ext3_xattr_ibody_get(inode, name_index, name, buffer, + buffer_size); + if (error == -ENODATA) + error = ext3_xattr_block_get(inode, name_index, name, buffer, + buffer_size); + up_read(&EXT3_I(inode)->xattr_sem); + return error; +} + +static int +ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry, + char *buffer, size_t buffer_size) +{ + size_t rest = buffer_size; + + for (; !IS_LAST_ENTRY(entry); entry = EXT3_XATTR_NEXT(entry)) { + struct xattr_handler *handler = + ext3_xattr_handler(entry->e_name_index); + + if (handler) { + size_t size = handler->list(inode, buffer, rest, + entry->e_name, + entry->e_name_len); + if (buffer) { + if (size > rest) + return -ERANGE; + buffer += size; + } + rest -= size; + } + } + return buffer_size - rest; +} + +static int +ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + int error; + + ea_idebug(inode, "buffer=%p, buffer_size=%ld", + buffer, (long)buffer_size); + + error = 0; + if (!EXT3_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %u", EXT3_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); + if (ext3_xattr_check_block(bh)) { + ext3_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block "E3FSBLK, inode->i_ino, + EXT3_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + ext3_xattr_cache_insert(bh); + error = ext3_xattr_list_entries(inode, BFIRST(bh), buffer, buffer_size); + +cleanup: + brelse(bh); + + return error; +} + +static int +ext3_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + struct ext3_xattr_ibody_header *header; + struct ext3_inode *raw_inode; + struct ext3_iloc iloc; + void *end; + int error; + + if (!(EXT3_I(inode)->i_state & EXT3_STATE_XATTR)) + return 0; + error = ext3_get_inode_loc(inode, &iloc); + if (error) + return error; + raw_inode = ext3_raw_inode(&iloc); + header = IHDR(inode, raw_inode); + end = (void *)raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; + error = ext3_xattr_check_names(IFIRST(header), end); + if (error) + goto cleanup; + error = ext3_xattr_list_entries(inode, IFIRST(header), + buffer, buffer_size); + +cleanup: + brelse(iloc.bh); + return error; +} + +/* + * ext3_xattr_list() + * + * Copy a list of attribute names into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + int i_error, b_error; + + down_read(&EXT3_I(inode)->xattr_sem); + i_error = ext3_xattr_ibody_list(inode, buffer, buffer_size); + if (i_error < 0) { + b_error = 0; + } else { + if (buffer) { + buffer += i_error; + buffer_size -= i_error; + } + b_error = ext3_xattr_block_list(inode, buffer, buffer_size); + if (b_error < 0) + i_error = 0; + } + up_read(&EXT3_I(inode)->xattr_sem); + return i_error + b_error; +} + +/* + * If the EXT3_FEATURE_COMPAT_EXT_ATTR feature of this file system is + * not set, set it. + */ +static void ext3_xattr_update_super_block(handle_t *handle, + struct super_block *sb) +{ + if (EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_EXT_ATTR)) + return; + + lock_super(sb); + if (ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh) == 0) { + EXT3_SB(sb)->s_es->s_feature_compat |= + cpu_to_le32(EXT3_FEATURE_COMPAT_EXT_ATTR); + sb->s_dirt = 1; + ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + } + unlock_super(sb); +} + +/* + * Release the xattr block BH: If the reference count is > 1, decrement + * it; otherwise free the block. + */ +static void +ext3_xattr_release_block(handle_t *handle, struct inode *inode, + struct buffer_head *bh) +{ + struct mb_cache_entry *ce = NULL; + + ce = mb_cache_entry_get(ext3_xattr_cache, bh->b_bdev, bh->b_blocknr); + if (BHDR(bh)->h_refcount == cpu_to_le32(1)) { + ea_bdebug(bh, "refcount now=0; freeing"); + if (ce) + mb_cache_entry_free(ce); + ext3_free_blocks(handle, inode, bh->b_blocknr, 1); + get_bh(bh); + ext3_forget(handle, 1, inode, bh, bh->b_blocknr); + } else { + if (ext3_journal_get_write_access(handle, bh) == 0) { + lock_buffer(bh); + BHDR(bh)->h_refcount = cpu_to_le32( + le32_to_cpu(BHDR(bh)->h_refcount) - 1); + ext3_journal_dirty_metadata(handle, bh); + if (IS_SYNC(inode)) + handle->h_sync = 1; + DQUOT_FREE_BLOCK(inode, 1); + unlock_buffer(bh); + ea_bdebug(bh, "refcount now=%d; releasing", + le32_to_cpu(BHDR(bh)->h_refcount)); + } + if (ce) + mb_cache_entry_release(ce); + } +} + +struct ext3_xattr_info { + int name_index; + const char *name; + const void *value; + size_t value_len; +}; + +struct ext3_xattr_search { + struct ext3_xattr_entry *first; + void *base; + void *end; + struct ext3_xattr_entry *here; + int not_found; +}; + +static int +ext3_xattr_set_entry(struct ext3_xattr_info *i, struct ext3_xattr_search *s) +{ + struct ext3_xattr_entry *last; + size_t free, min_offs = s->end - s->base, name_len = strlen(i->name); + + /* Compute min_offs and last. */ + last = s->first; + for (; !IS_LAST_ENTRY(last); last = EXT3_XATTR_NEXT(last)) { + if (!last->e_value_block && last->e_value_size) { + size_t offs = le16_to_cpu(last->e_value_offs); + if (offs < min_offs) + min_offs = offs; + } + } + free = min_offs - ((void *)last - s->base) - sizeof(__u32); + if (!s->not_found) { + if (!s->here->e_value_block && s->here->e_value_size) { + size_t size = le32_to_cpu(s->here->e_value_size); + free += EXT3_XATTR_SIZE(size); + } + free += EXT3_XATTR_LEN(name_len); + } + if (i->value) { + if (free < EXT3_XATTR_SIZE(i->value_len) || + free < EXT3_XATTR_LEN(name_len) + + EXT3_XATTR_SIZE(i->value_len)) + return -ENOSPC; + } + + if (i->value && s->not_found) { + /* Insert the new name. */ + size_t size = EXT3_XATTR_LEN(name_len); + size_t rest = (void *)last - (void *)s->here + sizeof(__u32); + memmove((void *)s->here + size, s->here, rest); + memset(s->here, 0, size); + s->here->e_name_index = i->name_index; + s->here->e_name_len = name_len; + memcpy(s->here->e_name, i->name, name_len); + } else { + if (!s->here->e_value_block && s->here->e_value_size) { + void *first_val = s->base + min_offs; + size_t offs = le16_to_cpu(s->here->e_value_offs); + void *val = s->base + offs; + size_t size = EXT3_XATTR_SIZE( + le32_to_cpu(s->here->e_value_size)); + + if (i->value && size == EXT3_XATTR_SIZE(i->value_len)) { + /* The old and the new value have the same + size. Just replace. */ + s->here->e_value_size = + cpu_to_le32(i->value_len); + memset(val + size - EXT3_XATTR_PAD, 0, + EXT3_XATTR_PAD); /* Clear pad bytes. */ + memcpy(val, i->value, i->value_len); + return 0; + } + + /* Remove the old value. */ + memmove(first_val + size, first_val, val - first_val); + memset(first_val, 0, size); + s->here->e_value_size = 0; + s->here->e_value_offs = 0; + min_offs += size; + + /* Adjust all value offsets. */ + last = s->first; + while (!IS_LAST_ENTRY(last)) { + size_t o = le16_to_cpu(last->e_value_offs); + if (!last->e_value_block && + last->e_value_size && o < offs) + last->e_value_offs = + cpu_to_le16(o + size); + last = EXT3_XATTR_NEXT(last); + } + } + if (!i->value) { + /* Remove the old name. */ + size_t size = EXT3_XATTR_LEN(name_len); + last = ENTRY((void *)last - size); + memmove(s->here, (void *)s->here + size, + (void *)last - (void *)s->here + sizeof(__u32)); + memset(last, 0, size); + } + } + + if (i->value) { + /* Insert the new value. */ + s->here->e_value_size = cpu_to_le32(i->value_len); + if (i->value_len) { + size_t size = EXT3_XATTR_SIZE(i->value_len); + void *val = s->base + min_offs - size; + s->here->e_value_offs = cpu_to_le16(min_offs - size); + memset(val + size - EXT3_XATTR_PAD, 0, + EXT3_XATTR_PAD); /* Clear the pad bytes. */ + memcpy(val, i->value, i->value_len); + } + } + return 0; +} + +struct ext3_xattr_block_find { + struct ext3_xattr_search s; + struct buffer_head *bh; +}; + +static int +ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i, + struct ext3_xattr_block_find *bs) +{ + struct super_block *sb = inode->i_sb; + int error; + + ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld", + i->name_index, i->name, i->value, (long)i->value_len); + + if (EXT3_I(inode)->i_file_acl) { + /* The inode already has an extended attribute block. */ + bs->bh = sb_bread(sb, EXT3_I(inode)->i_file_acl); + error = -EIO; + if (!bs->bh) + goto cleanup; + ea_bdebug(bs->bh, "b_count=%d, refcount=%d", + atomic_read(&(bs->bh->b_count)), + le32_to_cpu(BHDR(bs->bh)->h_refcount)); + if (ext3_xattr_check_block(bs->bh)) { + ext3_error(sb, __FUNCTION__, + "inode %lu: bad block "E3FSBLK, inode->i_ino, + EXT3_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* Find the named attribute. */ + bs->s.base = BHDR(bs->bh); + bs->s.first = BFIRST(bs->bh); + bs->s.end = bs->bh->b_data + bs->bh->b_size; + bs->s.here = bs->s.first; + error = ext3_xattr_find_entry(&bs->s.here, i->name_index, + i->name, bs->bh->b_size, 1); + if (error && error != -ENODATA) + goto cleanup; + bs->s.not_found = error; + } + error = 0; + +cleanup: + return error; +} + +static int +ext3_xattr_block_set(handle_t *handle, struct inode *inode, + struct ext3_xattr_info *i, + struct ext3_xattr_block_find *bs) +{ + struct super_block *sb = inode->i_sb; + struct buffer_head *new_bh = NULL; + struct ext3_xattr_search *s = &bs->s; + struct mb_cache_entry *ce = NULL; + int error; + +#define header(x) ((struct ext3_xattr_header *)(x)) + + if (i->value && i->value_len > sb->s_blocksize) + return -ENOSPC; + if (s->base) { + ce = mb_cache_entry_get(ext3_xattr_cache, bs->bh->b_bdev, + bs->bh->b_blocknr); + if (header(s->base)->h_refcount == cpu_to_le32(1)) { + if (ce) { + mb_cache_entry_free(ce); + ce = NULL; + } + ea_bdebug(bs->bh, "modifying in-place"); + error = ext3_journal_get_write_access(handle, bs->bh); + if (error) + goto cleanup; + lock_buffer(bs->bh); + error = ext3_xattr_set_entry(i, s); + if (!error) { + if (!IS_LAST_ENTRY(s->first)) + ext3_xattr_rehash(header(s->base), + s->here); + ext3_xattr_cache_insert(bs->bh); + } + unlock_buffer(bs->bh); + if (error == -EIO) + goto bad_block; + if (!error) + error = ext3_journal_dirty_metadata(handle, + bs->bh); + if (error) + goto cleanup; + goto inserted; + } else { + int offset = (char *)s->here - bs->bh->b_data; + + if (ce) { + mb_cache_entry_release(ce); + ce = NULL; + } + ea_bdebug(bs->bh, "cloning"); + s->base = kmalloc(bs->bh->b_size, GFP_KERNEL); + error = -ENOMEM; + if (s->base == NULL) + goto cleanup; + memcpy(s->base, BHDR(bs->bh), bs->bh->b_size); + s->first = ENTRY(header(s->base)+1); + header(s->base)->h_refcount = cpu_to_le32(1); + s->here = ENTRY(s->base + offset); + s->end = s->base + bs->bh->b_size; + } + } else { + /* Allocate a buffer where we construct the new block. */ + s->base = kmalloc(sb->s_blocksize, GFP_KERNEL); + /* assert(header == s->base) */ + error = -ENOMEM; + if (s->base == NULL) + goto cleanup; + memset(s->base, 0, sb->s_blocksize); + header(s->base)->h_magic = cpu_to_le32(EXT3_XATTR_MAGIC); + header(s->base)->h_blocks = cpu_to_le32(1); + header(s->base)->h_refcount = cpu_to_le32(1); + s->first = ENTRY(header(s->base)+1); + s->here = ENTRY(header(s->base)+1); + s->end = s->base + sb->s_blocksize; + } + + error = ext3_xattr_set_entry(i, s); + if (error == -EIO) + goto bad_block; + if (error) + goto cleanup; + if (!IS_LAST_ENTRY(s->first)) + ext3_xattr_rehash(header(s->base), s->here); + +inserted: + if (!IS_LAST_ENTRY(s->first)) { + new_bh = ext3_xattr_cache_find(inode, header(s->base), &ce); + if (new_bh) { + /* We found an identical block in the cache. */ + if (new_bh == bs->bh) + ea_bdebug(new_bh, "keeping"); + else { + /* The old block is released after updating + the inode. */ + error = -EDQUOT; + if (DQUOT_ALLOC_BLOCK(inode, 1)) + goto cleanup; + error = ext3_journal_get_write_access(handle, + new_bh); + if (error) + goto cleanup_dquot; + lock_buffer(new_bh); + BHDR(new_bh)->h_refcount = cpu_to_le32(1 + + le32_to_cpu(BHDR(new_bh)->h_refcount)); + ea_bdebug(new_bh, "reusing; refcount now=%d", + le32_to_cpu(BHDR(new_bh)->h_refcount)); + unlock_buffer(new_bh); + error = ext3_journal_dirty_metadata(handle, + new_bh); + if (error) + goto cleanup_dquot; + } + mb_cache_entry_release(ce); + ce = NULL; + } else if (bs->bh && s->base == bs->bh->b_data) { + /* We were modifying this block in-place. */ + ea_bdebug(bs->bh, "keeping this block"); + new_bh = bs->bh; + get_bh(new_bh); + } else { + /* We need to allocate a new block */ + ext3_fsblk_t goal = le32_to_cpu( + EXT3_SB(sb)->s_es->s_first_data_block) + + (ext3_fsblk_t)EXT3_I(inode)->i_block_group * + EXT3_BLOCKS_PER_GROUP(sb); + ext3_fsblk_t block = ext3_new_block(handle, inode, + goal, &error); + if (error) + goto cleanup; + ea_idebug(inode, "creating block %d", block); + + new_bh = sb_getblk(sb, block); + if (!new_bh) { +getblk_failed: + ext3_free_blocks(handle, inode, block, 1); + error = -EIO; + goto cleanup; + } + lock_buffer(new_bh); + error = ext3_journal_get_create_access(handle, new_bh); + if (error) { + unlock_buffer(new_bh); + goto getblk_failed; + } + memcpy(new_bh->b_data, s->base, new_bh->b_size); + set_buffer_uptodate(new_bh); + unlock_buffer(new_bh); + ext3_xattr_cache_insert(new_bh); + error = ext3_journal_dirty_metadata(handle, new_bh); + if (error) + goto cleanup; + } + } + + /* Update the inode. */ + EXT3_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0; + + /* Drop the previous xattr block. */ + if (bs->bh && bs->bh != new_bh) + ext3_xattr_release_block(handle, inode, bs->bh); + error = 0; + +cleanup: + if (ce) + mb_cache_entry_release(ce); + brelse(new_bh); + if (!(bs->bh && s->base == bs->bh->b_data)) + kfree(s->base); + + return error; + +cleanup_dquot: + DQUOT_FREE_BLOCK(inode, 1); + goto cleanup; + +bad_block: + ext3_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block "E3FSBLK, inode->i_ino, + EXT3_I(inode)->i_file_acl); + goto cleanup; + +#undef header +} + +struct ext3_xattr_ibody_find { + struct ext3_xattr_search s; + struct ext3_iloc iloc; +}; + +static int +ext3_xattr_ibody_find(struct inode *inode, struct ext3_xattr_info *i, + struct ext3_xattr_ibody_find *is) +{ + struct ext3_xattr_ibody_header *header; + struct ext3_inode *raw_inode; + int error; + + if (EXT3_I(inode)->i_extra_isize == 0) + return 0; + raw_inode = ext3_raw_inode(&is->iloc); + header = IHDR(inode, raw_inode); + is->s.base = is->s.first = IFIRST(header); + is->s.here = is->s.first; + is->s.end = (void *)raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; + if (EXT3_I(inode)->i_state & EXT3_STATE_XATTR) { + error = ext3_xattr_check_names(IFIRST(header), is->s.end); + if (error) + return error; + /* Find the named attribute. */ + error = ext3_xattr_find_entry(&is->s.here, i->name_index, + i->name, is->s.end - + (void *)is->s.base, 0); + if (error && error != -ENODATA) + return error; + is->s.not_found = error; + } + return 0; +} + +static int +ext3_xattr_ibody_set(handle_t *handle, struct inode *inode, + struct ext3_xattr_info *i, + struct ext3_xattr_ibody_find *is) +{ + struct ext3_xattr_ibody_header *header; + struct ext3_xattr_search *s = &is->s; + int error; + + if (EXT3_I(inode)->i_extra_isize == 0) + return -ENOSPC; + error = ext3_xattr_set_entry(i, s); + if (error) + return error; + header = IHDR(inode, ext3_raw_inode(&is->iloc)); + if (!IS_LAST_ENTRY(s->first)) { + header->h_magic = cpu_to_le32(EXT3_XATTR_MAGIC); + EXT3_I(inode)->i_state |= EXT3_STATE_XATTR; + } else { + header->h_magic = cpu_to_le32(0); + EXT3_I(inode)->i_state &= ~EXT3_STATE_XATTR; + } + return 0; +} + +/* + * ext3_xattr_set_handle() + * + * Create, replace or remove an extended attribute for this inode. Buffer + * is NULL to remove an existing extended attribute, and non-NULL to + * either replace an existing extended attribute, or create a new extended + * attribute. The flags XATTR_REPLACE and XATTR_CREATE + * specify that an extended attribute must exist and must not exist + * previous to the call, respectively. + * + * Returns 0, or a negative error number on failure. + */ +int +ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, + const char *name, const void *value, size_t value_len, + int flags) +{ + struct ext3_xattr_info i = { + .name_index = name_index, + .name = name, + .value = value, + .value_len = value_len, + + }; + struct ext3_xattr_ibody_find is = { + .s = { .not_found = -ENODATA, }, + }; + struct ext3_xattr_block_find bs = { + .s = { .not_found = -ENODATA, }, + }; + int error; + + if (!name) + return -EINVAL; + if (strlen(name) > 255) + return -ERANGE; + down_write(&EXT3_I(inode)->xattr_sem); + error = ext3_get_inode_loc(inode, &is.iloc); + if (error) + goto cleanup; + + if (EXT3_I(inode)->i_state & EXT3_STATE_NEW) { + struct ext3_inode *raw_inode = ext3_raw_inode(&is.iloc); + memset(raw_inode, 0, EXT3_SB(inode->i_sb)->s_inode_size); + EXT3_I(inode)->i_state &= ~EXT3_STATE_NEW; + } + + error = ext3_xattr_ibody_find(inode, &i, &is); + if (error) + goto cleanup; + if (is.s.not_found) + error = ext3_xattr_block_find(inode, &i, &bs); + if (error) + goto cleanup; + if (is.s.not_found && bs.s.not_found) { + error = -ENODATA; + if (flags & XATTR_REPLACE) + goto cleanup; + error = 0; + if (!value) + goto cleanup; + } else { + error = -EEXIST; + if (flags & XATTR_CREATE) + goto cleanup; + } + error = ext3_journal_get_write_access(handle, is.iloc.bh); + if (error) + goto cleanup; + if (!value) { + if (!is.s.not_found) + error = ext3_xattr_ibody_set(handle, inode, &i, &is); + else if (!bs.s.not_found) + error = ext3_xattr_block_set(handle, inode, &i, &bs); + } else { + error = ext3_xattr_ibody_set(handle, inode, &i, &is); + if (!error && !bs.s.not_found) { + i.value = NULL; + error = ext3_xattr_block_set(handle, inode, &i, &bs); + } else if (error == -ENOSPC) { + error = ext3_xattr_block_set(handle, inode, &i, &bs); + if (error) + goto cleanup; + if (!is.s.not_found) { + i.value = NULL; + error = ext3_xattr_ibody_set(handle, inode, &i, + &is); + } + } + } + if (!error) { + ext3_xattr_update_super_block(handle, inode->i_sb); + inode->i_ctime = CURRENT_TIME_SEC; + error = ext3_mark_iloc_dirty(handle, inode, &is.iloc); + /* + * The bh is consumed by ext3_mark_iloc_dirty, even with + * error != 0. + */ + is.iloc.bh = NULL; + if (IS_SYNC(inode)) + handle->h_sync = 1; + } + +cleanup: + brelse(is.iloc.bh); + brelse(bs.bh); + up_write(&EXT3_I(inode)->xattr_sem); + return error; +} + +/* + * ext3_xattr_set() + * + * Like ext3_xattr_set_handle, but start from an inode. This extended + * attribute modification is a filesystem transaction by itself. + * + * Returns 0, or a negative error number on failure. + */ +int +ext3_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t value_len, int flags) +{ + handle_t *handle; + int error, retries = 0; + +retry: + handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + } else { + int error2; + + error = ext3_xattr_set_handle(handle, inode, name_index, name, + value, value_len, flags); + error2 = ext3_journal_stop(handle); + if (error == -ENOSPC && + ext3_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + if (error == 0) + error = error2; + } + + return error; +} + +/* + * ext3_xattr_delete_inode() + * + * Free extended attribute resources associated with this inode. This + * is called immediately before an inode is freed. We have exclusive + * access to the inode. + */ +void +ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) +{ + struct buffer_head *bh = NULL; + + if (!EXT3_I(inode)->i_file_acl) + goto cleanup; + bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + if (!bh) { + ext3_error(inode->i_sb, __FUNCTION__, + "inode %lu: block "E3FSBLK" read error", inode->i_ino, + EXT3_I(inode)->i_file_acl); + goto cleanup; + } + if (BHDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + BHDR(bh)->h_blocks != cpu_to_le32(1)) { + ext3_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block "E3FSBLK, inode->i_ino, + EXT3_I(inode)->i_file_acl); + goto cleanup; + } + ext3_xattr_release_block(handle, inode, bh); + EXT3_I(inode)->i_file_acl = 0; + +cleanup: + brelse(bh); +} + +/* + * ext3_xattr_put_super() + * + * This is called when a file system is unmounted. + */ +void +ext3_xattr_put_super(struct super_block *sb) +{ + mb_cache_shrink(sb->s_bdev); +} + +/* + * ext3_xattr_cache_insert() + * + * Create a new entry in the extended attribute cache, and insert + * it unless such an entry is already in the cache. + * + * Returns 0, or a negative error number on failure. + */ +static void +ext3_xattr_cache_insert(struct buffer_head *bh) +{ + __u32 hash = le32_to_cpu(BHDR(bh)->h_hash); + struct mb_cache_entry *ce; + int error; + + ce = mb_cache_entry_alloc(ext3_xattr_cache); + if (!ce) { + ea_bdebug(bh, "out of memory"); + return; + } + error = mb_cache_entry_insert(ce, bh->b_bdev, bh->b_blocknr, &hash); + if (error) { + mb_cache_entry_free(ce); + if (error == -EBUSY) { + ea_bdebug(bh, "already in cache"); + error = 0; + } + } else { + ea_bdebug(bh, "inserting [%x]", (int)hash); + mb_cache_entry_release(ce); + } +} + +/* + * ext3_xattr_cmp() + * + * Compare two extended attribute blocks for equality. + * + * Returns 0 if the blocks are equal, 1 if they differ, and + * a negative error number on errors. + */ +static int +ext3_xattr_cmp(struct ext3_xattr_header *header1, + struct ext3_xattr_header *header2) +{ + struct ext3_xattr_entry *entry1, *entry2; + + entry1 = ENTRY(header1+1); + entry2 = ENTRY(header2+1); + while (!IS_LAST_ENTRY(entry1)) { + if (IS_LAST_ENTRY(entry2)) + return 1; + if (entry1->e_hash != entry2->e_hash || + entry1->e_name_index != entry2->e_name_index || + entry1->e_name_len != entry2->e_name_len || + entry1->e_value_size != entry2->e_value_size || + memcmp(entry1->e_name, entry2->e_name, entry1->e_name_len)) + return 1; + if (entry1->e_value_block != 0 || entry2->e_value_block != 0) + return -EIO; + if (memcmp((char *)header1 + le16_to_cpu(entry1->e_value_offs), + (char *)header2 + le16_to_cpu(entry2->e_value_offs), + le32_to_cpu(entry1->e_value_size))) + return 1; + + entry1 = EXT3_XATTR_NEXT(entry1); + entry2 = EXT3_XATTR_NEXT(entry2); + } + if (!IS_LAST_ENTRY(entry2)) + return 1; + return 0; +} + +/* + * ext3_xattr_cache_find() + * + * Find an identical extended attribute block. + * + * Returns a pointer to the block found, or NULL if such a block was + * not found or an error occurred. + */ +static struct buffer_head * +ext3_xattr_cache_find(struct inode *inode, struct ext3_xattr_header *header, + struct mb_cache_entry **pce) +{ + __u32 hash = le32_to_cpu(header->h_hash); + struct mb_cache_entry *ce; + + if (!header->h_hash) + return NULL; /* never share */ + ea_idebug(inode, "looking for cached blocks [%x]", (int)hash); +again: + ce = mb_cache_entry_find_first(ext3_xattr_cache, 0, + inode->i_sb->s_bdev, hash); + while (ce) { + struct buffer_head *bh; + + if (IS_ERR(ce)) { + if (PTR_ERR(ce) == -EAGAIN) + goto again; + break; + } + bh = sb_bread(inode->i_sb, ce->e_block); + if (!bh) { + ext3_error(inode->i_sb, __FUNCTION__, + "inode %lu: block %lu read error", + inode->i_ino, (unsigned long) ce->e_block); + } else if (le32_to_cpu(BHDR(bh)->h_refcount) >= + EXT3_XATTR_REFCOUNT_MAX) { + ea_idebug(inode, "block %lu refcount %d>=%d", + (unsigned long) ce->e_block, + le32_to_cpu(BHDR(bh)->h_refcount), + EXT3_XATTR_REFCOUNT_MAX); + } else if (ext3_xattr_cmp(header, BHDR(bh)) == 0) { + *pce = ce; + return bh; + } + brelse(bh); + ce = mb_cache_entry_find_next(ce, 0, inode->i_sb->s_bdev, hash); + } + return NULL; +} + +#define NAME_HASH_SHIFT 5 +#define VALUE_HASH_SHIFT 16 + +/* + * ext3_xattr_hash_entry() + * + * Compute the hash of an extended attribute. + */ +static inline void ext3_xattr_hash_entry(struct ext3_xattr_header *header, + struct ext3_xattr_entry *entry) +{ + __u32 hash = 0; + char *name = entry->e_name; + int n; + + for (n=0; n < entry->e_name_len; n++) { + hash = (hash << NAME_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^ + *name++; + } + + if (entry->e_value_block == 0 && entry->e_value_size != 0) { + __le32 *value = (__le32 *)((char *)header + + le16_to_cpu(entry->e_value_offs)); + for (n = (le32_to_cpu(entry->e_value_size) + + EXT3_XATTR_ROUND) >> EXT3_XATTR_PAD_BITS; n; n--) { + hash = (hash << VALUE_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - VALUE_HASH_SHIFT)) ^ + le32_to_cpu(*value++); + } + } + entry->e_hash = cpu_to_le32(hash); +} + +#undef NAME_HASH_SHIFT +#undef VALUE_HASH_SHIFT + +#define BLOCK_HASH_SHIFT 16 + +/* + * ext3_xattr_rehash() + * + * Re-compute the extended attribute hash value after an entry has changed. + */ +static void ext3_xattr_rehash(struct ext3_xattr_header *header, + struct ext3_xattr_entry *entry) +{ + struct ext3_xattr_entry *here; + __u32 hash = 0; + + ext3_xattr_hash_entry(header, entry); + here = ENTRY(header+1); + while (!IS_LAST_ENTRY(here)) { + if (!here->e_hash) { + /* Block is not shared if an entry's hash value == 0 */ + hash = 0; + break; + } + hash = (hash << BLOCK_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - BLOCK_HASH_SHIFT)) ^ + le32_to_cpu(here->e_hash); + here = EXT3_XATTR_NEXT(here); + } + header->h_hash = cpu_to_le32(hash); +} + +#undef BLOCK_HASH_SHIFT + +int __init +init_ext3_xattr(void) +{ + ext3_xattr_cache = mb_cache_create("ext3_xattr", NULL, + sizeof(struct mb_cache_entry) + + sizeof(((struct mb_cache_entry *) 0)->e_indexes[0]), 1, 6); + if (!ext3_xattr_cache) + return -ENOMEM; + return 0; +} + +void +exit_ext3_xattr(void) +{ + if (ext3_xattr_cache) + mb_cache_destroy(ext3_xattr_cache); + ext3_xattr_cache = NULL; +} diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h new file mode 100644 index 0000000..6b1ae1c --- /dev/null +++ b/fs/ext4/xattr.h @@ -0,0 +1,145 @@ +/* + File: fs/ext3/xattr.h + + On-disk format of extended attributes for the ext3 filesystem. + + (C) 2001 Andreas Gruenbacher, +*/ + +#include + +/* Magic value in attribute blocks */ +#define EXT3_XATTR_MAGIC 0xEA020000 + +/* Maximum number of references to one attribute block */ +#define EXT3_XATTR_REFCOUNT_MAX 1024 + +/* Name indexes */ +#define EXT3_XATTR_INDEX_USER 1 +#define EXT3_XATTR_INDEX_POSIX_ACL_ACCESS 2 +#define EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT 3 +#define EXT3_XATTR_INDEX_TRUSTED 4 +#define EXT3_XATTR_INDEX_LUSTRE 5 +#define EXT3_XATTR_INDEX_SECURITY 6 + +struct ext3_xattr_header { + __le32 h_magic; /* magic number for identification */ + __le32 h_refcount; /* reference count */ + __le32 h_blocks; /* number of disk blocks used */ + __le32 h_hash; /* hash value of all attributes */ + __u32 h_reserved[4]; /* zero right now */ +}; + +struct ext3_xattr_ibody_header { + __le32 h_magic; /* magic number for identification */ +}; + +struct ext3_xattr_entry { + __u8 e_name_len; /* length of name */ + __u8 e_name_index; /* attribute name index */ + __le16 e_value_offs; /* offset in disk block of value */ + __le32 e_value_block; /* disk block attribute is stored on (n/i) */ + __le32 e_value_size; /* size of attribute value */ + __le32 e_hash; /* hash value of name and value */ + char e_name[0]; /* attribute name */ +}; + +#define EXT3_XATTR_PAD_BITS 2 +#define EXT3_XATTR_PAD (1<e_name_len)) ) +#define EXT3_XATTR_SIZE(size) \ + (((size) + EXT3_XATTR_ROUND) & ~EXT3_XATTR_ROUND) + +# ifdef CONFIG_EXT3_FS_XATTR + +extern struct xattr_handler ext3_xattr_user_handler; +extern struct xattr_handler ext3_xattr_trusted_handler; +extern struct xattr_handler ext3_xattr_acl_access_handler; +extern struct xattr_handler ext3_xattr_acl_default_handler; +extern struct xattr_handler ext3_xattr_security_handler; + +extern ssize_t ext3_listxattr(struct dentry *, char *, size_t); + +extern int ext3_xattr_get(struct inode *, int, const char *, void *, size_t); +extern int ext3_xattr_list(struct inode *, char *, size_t); +extern int ext3_xattr_set(struct inode *, int, const char *, const void *, size_t, int); +extern int ext3_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); + +extern void ext3_xattr_delete_inode(handle_t *, struct inode *); +extern void ext3_xattr_put_super(struct super_block *); + +extern int init_ext3_xattr(void); +extern void exit_ext3_xattr(void); + +extern struct xattr_handler *ext3_xattr_handlers[]; + +# else /* CONFIG_EXT3_FS_XATTR */ + +static inline int +ext3_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline int +ext3_xattr_list(struct inode *inode, void *buffer, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline int +ext3_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline int +ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, + const char *name, const void *value, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline void +ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) +{ +} + +static inline void +ext3_xattr_put_super(struct super_block *sb) +{ +} + +static inline int +init_ext3_xattr(void) +{ + return 0; +} + +static inline void +exit_ext3_xattr(void) +{ +} + +#define ext3_xattr_handlers NULL + +# endif /* CONFIG_EXT3_FS_XATTR */ + +#ifdef CONFIG_EXT3_FS_SECURITY +extern int ext3_init_security(handle_t *handle, struct inode *inode, + struct inode *dir); +#else +static inline int ext3_init_security(handle_t *handle, struct inode *inode, + struct inode *dir) +{ + return 0; +} +#endif diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c new file mode 100644 index 0000000..b9c40c1 --- /dev/null +++ b/fs/ext4/xattr_security.c @@ -0,0 +1,77 @@ +/* + * linux/fs/ext3/xattr_security.c + * Handler for storing security labels as extended attributes. + */ + +#include +#include +#include +#include +#include +#include +#include +#include "xattr.h" + +static size_t +ext3_xattr_security_list(struct inode *inode, char *list, size_t list_size, + const char *name, size_t name_len) +{ + const size_t prefix_len = sizeof(XATTR_SECURITY_PREFIX)-1; + const size_t total_len = prefix_len + name_len + 1; + + + if (list && total_len <= list_size) { + memcpy(list, XATTR_SECURITY_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return total_len; +} + +static int +ext3_xattr_security_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext3_xattr_get(inode, EXT3_XATTR_INDEX_SECURITY, name, + buffer, size); +} + +static int +ext3_xattr_security_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext3_xattr_set(inode, EXT3_XATTR_INDEX_SECURITY, name, + value, size, flags); +} + +int +ext3_init_security(handle_t *handle, struct inode *inode, struct inode *dir) +{ + int err; + size_t len; + void *value; + char *name; + + err = security_inode_init_security(inode, dir, &name, &value, &len); + if (err) { + if (err == -EOPNOTSUPP) + return 0; + return err; + } + err = ext3_xattr_set_handle(handle, inode, EXT3_XATTR_INDEX_SECURITY, + name, value, len, 0); + kfree(name); + kfree(value); + return err; +} + +struct xattr_handler ext3_xattr_security_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .list = ext3_xattr_security_list, + .get = ext3_xattr_security_get, + .set = ext3_xattr_security_set, +}; diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c new file mode 100644 index 0000000..86d91f1 --- /dev/null +++ b/fs/ext4/xattr_trusted.c @@ -0,0 +1,62 @@ +/* + * linux/fs/ext3/xattr_trusted.c + * Handler for trusted extended attributes. + * + * Copyright (C) 2003 by Andreas Gruenbacher, + */ + +#include +#include +#include +#include +#include +#include +#include +#include "xattr.h" + +#define XATTR_TRUSTED_PREFIX "trusted." + +static size_t +ext3_xattr_trusted_list(struct inode *inode, char *list, size_t list_size, + const char *name, size_t name_len) +{ + const size_t prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1; + const size_t total_len = prefix_len + name_len + 1; + + if (!capable(CAP_SYS_ADMIN)) + return 0; + + if (list && total_len <= list_size) { + memcpy(list, XATTR_TRUSTED_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return total_len; +} + +static int +ext3_xattr_trusted_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext3_xattr_get(inode, EXT3_XATTR_INDEX_TRUSTED, name, + buffer, size); +} + +static int +ext3_xattr_trusted_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext3_xattr_set(inode, EXT3_XATTR_INDEX_TRUSTED, name, + value, size, flags); +} + +struct xattr_handler ext3_xattr_trusted_handler = { + .prefix = XATTR_TRUSTED_PREFIX, + .list = ext3_xattr_trusted_list, + .get = ext3_xattr_trusted_get, + .set = ext3_xattr_trusted_set, +}; diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c new file mode 100644 index 0000000..a85a0a1 --- /dev/null +++ b/fs/ext4/xattr_user.c @@ -0,0 +1,64 @@ +/* + * linux/fs/ext3/xattr_user.c + * Handler for extended user attributes. + * + * Copyright (C) 2001 by Andreas Gruenbacher, + */ + +#include +#include +#include +#include +#include +#include +#include "xattr.h" + +#define XATTR_USER_PREFIX "user." + +static size_t +ext3_xattr_user_list(struct inode *inode, char *list, size_t list_size, + const char *name, size_t name_len) +{ + const size_t prefix_len = sizeof(XATTR_USER_PREFIX)-1; + const size_t total_len = prefix_len + name_len + 1; + + if (!test_opt(inode->i_sb, XATTR_USER)) + return 0; + + if (list && total_len <= list_size) { + memcpy(list, XATTR_USER_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return total_len; +} + +static int +ext3_xattr_user_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + return ext3_xattr_get(inode, EXT3_XATTR_INDEX_USER, name, buffer, size); +} + +static int +ext3_xattr_user_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + return ext3_xattr_set(inode, EXT3_XATTR_INDEX_USER, name, + value, size, flags); +} + +struct xattr_handler ext3_xattr_user_handler = { + .prefix = XATTR_USER_PREFIX, + .list = ext3_xattr_user_list, + .get = ext3_xattr_user_get, + .set = ext3_xattr_user_set, +}; diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h new file mode 100644 index 0000000..11cca1b --- /dev/null +++ b/include/linux/ext4_fs.h @@ -0,0 +1,885 @@ +/* + * linux/include/linux/ext3_fs.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _LINUX_EXT3_FS_H +#define _LINUX_EXT3_FS_H + +#include +#include + +/* + * The second extended filesystem constants/structures + */ + +/* + * Define EXT3FS_DEBUG to produce debug messages + */ +#undef EXT3FS_DEBUG + +/* + * Define EXT3_RESERVATION to reserve data blocks for expanding files + */ +#define EXT3_DEFAULT_RESERVE_BLOCKS 8 +/*max window size: 1024(direct blocks) + 3([t,d]indirect blocks) */ +#define EXT3_MAX_RESERVE_BLOCKS 1027 +#define EXT3_RESERVE_WINDOW_NOT_ALLOCATED 0 +/* + * Always enable hashed directories + */ +#define CONFIG_EXT3_INDEX + +/* + * Debug code + */ +#ifdef EXT3FS_DEBUG +#define ext3_debug(f, a...) \ + do { \ + printk (KERN_DEBUG "EXT3-fs DEBUG (%s, %d): %s:", \ + __FILE__, __LINE__, __FUNCTION__); \ + printk (KERN_DEBUG f, ## a); \ + } while (0) +#else +#define ext3_debug(f, a...) do {} while (0) +#endif + +/* + * Special inodes numbers + */ +#define EXT3_BAD_INO 1 /* Bad blocks inode */ +#define EXT3_ROOT_INO 2 /* Root inode */ +#define EXT3_BOOT_LOADER_INO 5 /* Boot loader inode */ +#define EXT3_UNDEL_DIR_INO 6 /* Undelete directory inode */ +#define EXT3_RESIZE_INO 7 /* Reserved group descriptors inode */ +#define EXT3_JOURNAL_INO 8 /* Journal inode */ + +/* First non-reserved inode for old ext3 filesystems */ +#define EXT3_GOOD_OLD_FIRST_INO 11 + +/* + * Maximal count of links to a file + */ +#define EXT3_LINK_MAX 32000 + +/* + * Macro-instructions used to manage several block sizes + */ +#define EXT3_MIN_BLOCK_SIZE 1024 +#define EXT3_MAX_BLOCK_SIZE 4096 +#define EXT3_MIN_BLOCK_LOG_SIZE 10 +#ifdef __KERNEL__ +# define EXT3_BLOCK_SIZE(s) ((s)->s_blocksize) +#else +# define EXT3_BLOCK_SIZE(s) (EXT3_MIN_BLOCK_SIZE << (s)->s_log_block_size) +#endif +#define EXT3_ADDR_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / sizeof (__u32)) +#ifdef __KERNEL__ +# define EXT3_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) +#else +# define EXT3_BLOCK_SIZE_BITS(s) ((s)->s_log_block_size + 10) +#endif +#ifdef __KERNEL__ +#define EXT3_ADDR_PER_BLOCK_BITS(s) (EXT3_SB(s)->s_addr_per_block_bits) +#define EXT3_INODE_SIZE(s) (EXT3_SB(s)->s_inode_size) +#define EXT3_FIRST_INO(s) (EXT3_SB(s)->s_first_ino) +#else +#define EXT3_INODE_SIZE(s) (((s)->s_rev_level == EXT3_GOOD_OLD_REV) ? \ + EXT3_GOOD_OLD_INODE_SIZE : \ + (s)->s_inode_size) +#define EXT3_FIRST_INO(s) (((s)->s_rev_level == EXT3_GOOD_OLD_REV) ? \ + EXT3_GOOD_OLD_FIRST_INO : \ + (s)->s_first_ino) +#endif + +/* + * Macro-instructions used to manage fragments + */ +#define EXT3_MIN_FRAG_SIZE 1024 +#define EXT3_MAX_FRAG_SIZE 4096 +#define EXT3_MIN_FRAG_LOG_SIZE 10 +#ifdef __KERNEL__ +# define EXT3_FRAG_SIZE(s) (EXT3_SB(s)->s_frag_size) +# define EXT3_FRAGS_PER_BLOCK(s) (EXT3_SB(s)->s_frags_per_block) +#else +# define EXT3_FRAG_SIZE(s) (EXT3_MIN_FRAG_SIZE << (s)->s_log_frag_size) +# define EXT3_FRAGS_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / EXT3_FRAG_SIZE(s)) +#endif + +/* + * Structure of a blocks group descriptor + */ +struct ext3_group_desc +{ + __le32 bg_block_bitmap; /* Blocks bitmap block */ + __le32 bg_inode_bitmap; /* Inodes bitmap block */ + __le32 bg_inode_table; /* Inodes table block */ + __le16 bg_free_blocks_count; /* Free blocks count */ + __le16 bg_free_inodes_count; /* Free inodes count */ + __le16 bg_used_dirs_count; /* Directories count */ + __u16 bg_pad; + __le32 bg_reserved[3]; +}; + +/* + * Macro-instructions used to manage group descriptors + */ +#ifdef __KERNEL__ +# define EXT3_BLOCKS_PER_GROUP(s) (EXT3_SB(s)->s_blocks_per_group) +# define EXT3_DESC_PER_BLOCK(s) (EXT3_SB(s)->s_desc_per_block) +# define EXT3_INODES_PER_GROUP(s) (EXT3_SB(s)->s_inodes_per_group) +# define EXT3_DESC_PER_BLOCK_BITS(s) (EXT3_SB(s)->s_desc_per_block_bits) +#else +# define EXT3_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group) +# define EXT3_DESC_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / sizeof (struct ext3_group_desc)) +# define EXT3_INODES_PER_GROUP(s) ((s)->s_inodes_per_group) +#endif + +/* + * Constants relative to the data blocks + */ +#define EXT3_NDIR_BLOCKS 12 +#define EXT3_IND_BLOCK EXT3_NDIR_BLOCKS +#define EXT3_DIND_BLOCK (EXT3_IND_BLOCK + 1) +#define EXT3_TIND_BLOCK (EXT3_DIND_BLOCK + 1) +#define EXT3_N_BLOCKS (EXT3_TIND_BLOCK + 1) + +/* + * Inode flags + */ +#define EXT3_SECRM_FL 0x00000001 /* Secure deletion */ +#define EXT3_UNRM_FL 0x00000002 /* Undelete */ +#define EXT3_COMPR_FL 0x00000004 /* Compress file */ +#define EXT3_SYNC_FL 0x00000008 /* Synchronous updates */ +#define EXT3_IMMUTABLE_FL 0x00000010 /* Immutable file */ +#define EXT3_APPEND_FL 0x00000020 /* writes to file may only append */ +#define EXT3_NODUMP_FL 0x00000040 /* do not dump file */ +#define EXT3_NOATIME_FL 0x00000080 /* do not update atime */ +/* Reserved for compression usage... */ +#define EXT3_DIRTY_FL 0x00000100 +#define EXT3_COMPRBLK_FL 0x00000200 /* One or more compressed clusters */ +#define EXT3_NOCOMPR_FL 0x00000400 /* Don't compress */ +#define EXT3_ECOMPR_FL 0x00000800 /* Compression error */ +/* End compression flags --- maybe not all used */ +#define EXT3_INDEX_FL 0x00001000 /* hash-indexed directory */ +#define EXT3_IMAGIC_FL 0x00002000 /* AFS directory */ +#define EXT3_JOURNAL_DATA_FL 0x00004000 /* file data should be journaled */ +#define EXT3_NOTAIL_FL 0x00008000 /* file tail should not be merged */ +#define EXT3_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ +#define EXT3_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ +#define EXT3_RESERVED_FL 0x80000000 /* reserved for ext3 lib */ + +#define EXT3_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */ +#define EXT3_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ + +/* + * Inode dynamic state flags + */ +#define EXT3_STATE_JDATA 0x00000001 /* journaled data exists */ +#define EXT3_STATE_NEW 0x00000002 /* inode is newly created */ +#define EXT3_STATE_XATTR 0x00000004 /* has in-inode xattrs */ + +/* Used to pass group descriptor data when online resize is done */ +struct ext3_new_group_input { + __u32 group; /* Group number for this data */ + __u32 block_bitmap; /* Absolute block number of block bitmap */ + __u32 inode_bitmap; /* Absolute block number of inode bitmap */ + __u32 inode_table; /* Absolute block number of inode table start */ + __u32 blocks_count; /* Total number of blocks in this group */ + __u16 reserved_blocks; /* Number of reserved blocks in this group */ + __u16 unused; +}; + +/* The struct ext3_new_group_input in kernel space, with free_blocks_count */ +struct ext3_new_group_data { + __u32 group; + __u32 block_bitmap; + __u32 inode_bitmap; + __u32 inode_table; + __u32 blocks_count; + __u16 reserved_blocks; + __u16 unused; + __u32 free_blocks_count; +}; + + +/* + * ioctl commands + */ +#define EXT3_IOC_GETFLAGS FS_IOC_GETFLAGS +#define EXT3_IOC_SETFLAGS FS_IOC_SETFLAGS +#define EXT3_IOC_GETVERSION _IOR('f', 3, long) +#define EXT3_IOC_SETVERSION _IOW('f', 4, long) +#define EXT3_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) +#define EXT3_IOC_GROUP_ADD _IOW('f', 8,struct ext3_new_group_input) +#define EXT3_IOC_GETVERSION_OLD FS_IOC_GETVERSION +#define EXT3_IOC_SETVERSION_OLD FS_IOC_SETVERSION +#ifdef CONFIG_JBD_DEBUG +#define EXT3_IOC_WAIT_FOR_READONLY _IOR('f', 99, long) +#endif +#define EXT3_IOC_GETRSVSZ _IOR('f', 5, long) +#define EXT3_IOC_SETRSVSZ _IOW('f', 6, long) + +/* + * ioctl commands in 32 bit emulation + */ +#define EXT3_IOC32_GETFLAGS FS_IOC32_GETFLAGS +#define EXT3_IOC32_SETFLAGS FS_IOC32_SETFLAGS +#define EXT3_IOC32_GETVERSION _IOR('f', 3, int) +#define EXT3_IOC32_SETVERSION _IOW('f', 4, int) +#define EXT3_IOC32_GETRSVSZ _IOR('f', 5, int) +#define EXT3_IOC32_SETRSVSZ _IOW('f', 6, int) +#define EXT3_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int) +#ifdef CONFIG_JBD_DEBUG +#define EXT3_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int) +#endif +#define EXT3_IOC32_GETVERSION_OLD FS_IOC32_GETVERSION +#define EXT3_IOC32_SETVERSION_OLD FS_IOC32_SETVERSION + + +/* + * Mount options + */ +struct ext3_mount_options { + unsigned long s_mount_opt; + uid_t s_resuid; + gid_t s_resgid; + unsigned long s_commit_interval; +#ifdef CONFIG_QUOTA + int s_jquota_fmt; + char *s_qf_names[MAXQUOTAS]; +#endif +}; + +/* + * Structure of an inode on the disk + */ +struct ext3_inode { + __le16 i_mode; /* File mode */ + __le16 i_uid; /* Low 16 bits of Owner Uid */ + __le32 i_size; /* Size in bytes */ + __le32 i_atime; /* Access time */ + __le32 i_ctime; /* Creation time */ + __le32 i_mtime; /* Modification time */ + __le32 i_dtime; /* Deletion Time */ + __le16 i_gid; /* Low 16 bits of Group Id */ + __le16 i_links_count; /* Links count */ + __le32 i_blocks; /* Blocks count */ + __le32 i_flags; /* File flags */ + union { + struct { + __u32 l_i_reserved1; + } linux1; + struct { + __u32 h_i_translator; + } hurd1; + struct { + __u32 m_i_reserved1; + } masix1; + } osd1; /* OS dependent 1 */ + __le32 i_block[EXT3_N_BLOCKS];/* Pointers to blocks */ + __le32 i_generation; /* File version (for NFS) */ + __le32 i_file_acl; /* File ACL */ + __le32 i_dir_acl; /* Directory ACL */ + __le32 i_faddr; /* Fragment address */ + union { + struct { + __u8 l_i_frag; /* Fragment number */ + __u8 l_i_fsize; /* Fragment size */ + __u16 i_pad1; + __le16 l_i_uid_high; /* these 2 fields */ + __le16 l_i_gid_high; /* were reserved2[0] */ + __u32 l_i_reserved2; + } linux2; + struct { + __u8 h_i_frag; /* Fragment number */ + __u8 h_i_fsize; /* Fragment size */ + __u16 h_i_mode_high; + __u16 h_i_uid_high; + __u16 h_i_gid_high; + __u32 h_i_author; + } hurd2; + struct { + __u8 m_i_frag; /* Fragment number */ + __u8 m_i_fsize; /* Fragment size */ + __u16 m_pad1; + __u32 m_i_reserved2[2]; + } masix2; + } osd2; /* OS dependent 2 */ + __le16 i_extra_isize; + __le16 i_pad1; +}; + +#define i_size_high i_dir_acl + +#if defined(__KERNEL__) || defined(__linux__) +#define i_reserved1 osd1.linux1.l_i_reserved1 +#define i_frag osd2.linux2.l_i_frag +#define i_fsize osd2.linux2.l_i_fsize +#define i_uid_low i_uid +#define i_gid_low i_gid +#define i_uid_high osd2.linux2.l_i_uid_high +#define i_gid_high osd2.linux2.l_i_gid_high +#define i_reserved2 osd2.linux2.l_i_reserved2 + +#elif defined(__GNU__) + +#define i_translator osd1.hurd1.h_i_translator +#define i_frag osd2.hurd2.h_i_frag; +#define i_fsize osd2.hurd2.h_i_fsize; +#define i_uid_high osd2.hurd2.h_i_uid_high +#define i_gid_high osd2.hurd2.h_i_gid_high +#define i_author osd2.hurd2.h_i_author + +#elif defined(__masix__) + +#define i_reserved1 osd1.masix1.m_i_reserved1 +#define i_frag osd2.masix2.m_i_frag +#define i_fsize osd2.masix2.m_i_fsize +#define i_reserved2 osd2.masix2.m_i_reserved2 + +#endif /* defined(__KERNEL__) || defined(__linux__) */ + +/* + * File system states + */ +#define EXT3_VALID_FS 0x0001 /* Unmounted cleanly */ +#define EXT3_ERROR_FS 0x0002 /* Errors detected */ +#define EXT3_ORPHAN_FS 0x0004 /* Orphans being recovered */ + +/* + * Mount flags + */ +#define EXT3_MOUNT_CHECK 0x00001 /* Do mount-time checks */ +#define EXT3_MOUNT_OLDALLOC 0x00002 /* Don't use the new Orlov allocator */ +#define EXT3_MOUNT_GRPID 0x00004 /* Create files with directory's group */ +#define EXT3_MOUNT_DEBUG 0x00008 /* Some debugging messages */ +#define EXT3_MOUNT_ERRORS_CONT 0x00010 /* Continue on errors */ +#define EXT3_MOUNT_ERRORS_RO 0x00020 /* Remount fs ro on errors */ +#define EXT3_MOUNT_ERRORS_PANIC 0x00040 /* Panic on errors */ +#define EXT3_MOUNT_MINIX_DF 0x00080 /* Mimics the Minix statfs */ +#define EXT3_MOUNT_NOLOAD 0x00100 /* Don't use existing journal*/ +#define EXT3_MOUNT_ABORT 0x00200 /* Fatal error detected */ +#define EXT3_MOUNT_DATA_FLAGS 0x00C00 /* Mode for data writes: */ +#define EXT3_MOUNT_JOURNAL_DATA 0x00400 /* Write data to journal */ +#define EXT3_MOUNT_ORDERED_DATA 0x00800 /* Flush data before commit */ +#define EXT3_MOUNT_WRITEBACK_DATA 0x00C00 /* No data ordering */ +#define EXT3_MOUNT_UPDATE_JOURNAL 0x01000 /* Update the journal format */ +#define EXT3_MOUNT_NO_UID32 0x02000 /* Disable 32-bit UIDs */ +#define EXT3_MOUNT_XATTR_USER 0x04000 /* Extended user attributes */ +#define EXT3_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */ +#define EXT3_MOUNT_RESERVATION 0x10000 /* Preallocation */ +#define EXT3_MOUNT_BARRIER 0x20000 /* Use block barriers */ +#define EXT3_MOUNT_NOBH 0x40000 /* No bufferheads */ +#define EXT3_MOUNT_QUOTA 0x80000 /* Some quota option set */ +#define EXT3_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ +#define EXT3_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ + +/* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ +#ifndef _LINUX_EXT2_FS_H +#define clear_opt(o, opt) o &= ~EXT3_MOUNT_##opt +#define set_opt(o, opt) o |= EXT3_MOUNT_##opt +#define test_opt(sb, opt) (EXT3_SB(sb)->s_mount_opt & \ + EXT3_MOUNT_##opt) +#else +#define EXT2_MOUNT_NOLOAD EXT3_MOUNT_NOLOAD +#define EXT2_MOUNT_ABORT EXT3_MOUNT_ABORT +#define EXT2_MOUNT_DATA_FLAGS EXT3_MOUNT_DATA_FLAGS +#endif + +#define ext3_set_bit ext2_set_bit +#define ext3_set_bit_atomic ext2_set_bit_atomic +#define ext3_clear_bit ext2_clear_bit +#define ext3_clear_bit_atomic ext2_clear_bit_atomic +#define ext3_test_bit ext2_test_bit +#define ext3_find_first_zero_bit ext2_find_first_zero_bit +#define ext3_find_next_zero_bit ext2_find_next_zero_bit + +/* + * Maximal mount counts between two filesystem checks + */ +#define EXT3_DFL_MAX_MNT_COUNT 20 /* Allow 20 mounts */ +#define EXT3_DFL_CHECKINTERVAL 0 /* Don't use interval check */ + +/* + * Behaviour when detecting errors + */ +#define EXT3_ERRORS_CONTINUE 1 /* Continue execution */ +#define EXT3_ERRORS_RO 2 /* Remount fs read-only */ +#define EXT3_ERRORS_PANIC 3 /* Panic */ +#define EXT3_ERRORS_DEFAULT EXT3_ERRORS_CONTINUE + +/* + * Structure of the super block + */ +struct ext3_super_block { +/*00*/ __le32 s_inodes_count; /* Inodes count */ + __le32 s_blocks_count; /* Blocks count */ + __le32 s_r_blocks_count; /* Reserved blocks count */ + __le32 s_free_blocks_count; /* Free blocks count */ +/*10*/ __le32 s_free_inodes_count; /* Free inodes count */ + __le32 s_first_data_block; /* First Data Block */ + __le32 s_log_block_size; /* Block size */ + __le32 s_log_frag_size; /* Fragment size */ +/*20*/ __le32 s_blocks_per_group; /* # Blocks per group */ + __le32 s_frags_per_group; /* # Fragments per group */ + __le32 s_inodes_per_group; /* # Inodes per group */ + __le32 s_mtime; /* Mount time */ +/*30*/ __le32 s_wtime; /* Write time */ + __le16 s_mnt_count; /* Mount count */ + __le16 s_max_mnt_count; /* Maximal mount count */ + __le16 s_magic; /* Magic signature */ + __le16 s_state; /* File system state */ + __le16 s_errors; /* Behaviour when detecting errors */ + __le16 s_minor_rev_level; /* minor revision level */ +/*40*/ __le32 s_lastcheck; /* time of last check */ + __le32 s_checkinterval; /* max. time between checks */ + __le32 s_creator_os; /* OS */ + __le32 s_rev_level; /* Revision level */ +/*50*/ __le16 s_def_resuid; /* Default uid for reserved blocks */ + __le16 s_def_resgid; /* Default gid for reserved blocks */ + /* + * These fields are for EXT3_DYNAMIC_REV superblocks only. + * + * Note: the difference between the compatible feature set and + * the incompatible feature set is that if there is a bit set + * in the incompatible feature set that the kernel doesn't + * know about, it should refuse to mount the filesystem. + * + * e2fsck's requirements are more strict; if it doesn't know + * about a feature in either the compatible or incompatible + * feature set, it must abort and not try to meddle with + * things it doesn't understand... + */ + __le32 s_first_ino; /* First non-reserved inode */ + __le16 s_inode_size; /* size of inode structure */ + __le16 s_block_group_nr; /* block group # of this superblock */ + __le32 s_feature_compat; /* compatible feature set */ +/*60*/ __le32 s_feature_incompat; /* incompatible feature set */ + __le32 s_feature_ro_compat; /* readonly-compatible feature set */ +/*68*/ __u8 s_uuid[16]; /* 128-bit uuid for volume */ +/*78*/ char s_volume_name[16]; /* volume name */ +/*88*/ char s_last_mounted[64]; /* directory where last mounted */ +/*C8*/ __le32 s_algorithm_usage_bitmap; /* For compression */ + /* + * Performance hints. Directory preallocation should only + * happen if the EXT3_FEATURE_COMPAT_DIR_PREALLOC flag is on. + */ + __u8 s_prealloc_blocks; /* Nr of blocks to try to preallocate*/ + __u8 s_prealloc_dir_blocks; /* Nr to preallocate for dirs */ + __le16 s_reserved_gdt_blocks; /* Per group desc for online growth */ + /* + * Journaling support valid if EXT3_FEATURE_COMPAT_HAS_JOURNAL set. + */ +/*D0*/ __u8 s_journal_uuid[16]; /* uuid of journal superblock */ +/*E0*/ __le32 s_journal_inum; /* inode number of journal file */ + __le32 s_journal_dev; /* device number of journal file */ + __le32 s_last_orphan; /* start of list of inodes to delete */ + __le32 s_hash_seed[4]; /* HTREE hash seed */ + __u8 s_def_hash_version; /* Default hash version to use */ + __u8 s_reserved_char_pad; + __u16 s_reserved_word_pad; + __le32 s_default_mount_opts; + __le32 s_first_meta_bg; /* First metablock block group */ + __u32 s_reserved[190]; /* Padding to the end of the block */ +}; + +#ifdef __KERNEL__ +#include +#include +static inline struct ext3_sb_info * EXT3_SB(struct super_block *sb) +{ + return sb->s_fs_info; +} +static inline struct ext3_inode_info *EXT3_I(struct inode *inode) +{ + return container_of(inode, struct ext3_inode_info, vfs_inode); +} + +static inline int ext3_valid_inum(struct super_block *sb, unsigned long ino) +{ + return ino == EXT3_ROOT_INO || + ino == EXT3_JOURNAL_INO || + ino == EXT3_RESIZE_INO || + (ino >= EXT3_FIRST_INO(sb) && + ino <= le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)); +} +#else +/* Assume that user mode programs are passing in an ext3fs superblock, not + * a kernel struct super_block. This will allow us to call the feature-test + * macros from user land. */ +#define EXT3_SB(sb) (sb) +#endif + +#define NEXT_ORPHAN(inode) EXT3_I(inode)->i_dtime + +/* + * Codes for operating systems + */ +#define EXT3_OS_LINUX 0 +#define EXT3_OS_HURD 1 +#define EXT3_OS_MASIX 2 +#define EXT3_OS_FREEBSD 3 +#define EXT3_OS_LITES 4 + +/* + * Revision levels + */ +#define EXT3_GOOD_OLD_REV 0 /* The good old (original) format */ +#define EXT3_DYNAMIC_REV 1 /* V2 format w/ dynamic inode sizes */ + +#define EXT3_CURRENT_REV EXT3_GOOD_OLD_REV +#define EXT3_MAX_SUPP_REV EXT3_DYNAMIC_REV + +#define EXT3_GOOD_OLD_INODE_SIZE 128 + +/* + * Feature set definitions + */ + +#define EXT3_HAS_COMPAT_FEATURE(sb,mask) \ + ( EXT3_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) +#define EXT3_HAS_RO_COMPAT_FEATURE(sb,mask) \ + ( EXT3_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) +#define EXT3_HAS_INCOMPAT_FEATURE(sb,mask) \ + ( EXT3_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) +#define EXT3_SET_COMPAT_FEATURE(sb,mask) \ + EXT3_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) +#define EXT3_SET_RO_COMPAT_FEATURE(sb,mask) \ + EXT3_SB(sb)->s_es->s_feature_ro_compat |= cpu_to_le32(mask) +#define EXT3_SET_INCOMPAT_FEATURE(sb,mask) \ + EXT3_SB(sb)->s_es->s_feature_incompat |= cpu_to_le32(mask) +#define EXT3_CLEAR_COMPAT_FEATURE(sb,mask) \ + EXT3_SB(sb)->s_es->s_feature_compat &= ~cpu_to_le32(mask) +#define EXT3_CLEAR_RO_COMPAT_FEATURE(sb,mask) \ + EXT3_SB(sb)->s_es->s_feature_ro_compat &= ~cpu_to_le32(mask) +#define EXT3_CLEAR_INCOMPAT_FEATURE(sb,mask) \ + EXT3_SB(sb)->s_es->s_feature_incompat &= ~cpu_to_le32(mask) + +#define EXT3_FEATURE_COMPAT_DIR_PREALLOC 0x0001 +#define EXT3_FEATURE_COMPAT_IMAGIC_INODES 0x0002 +#define EXT3_FEATURE_COMPAT_HAS_JOURNAL 0x0004 +#define EXT3_FEATURE_COMPAT_EXT_ATTR 0x0008 +#define EXT3_FEATURE_COMPAT_RESIZE_INODE 0x0010 +#define EXT3_FEATURE_COMPAT_DIR_INDEX 0x0020 + +#define EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001 +#define EXT3_FEATURE_RO_COMPAT_LARGE_FILE 0x0002 +#define EXT3_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 + +#define EXT3_FEATURE_INCOMPAT_COMPRESSION 0x0001 +#define EXT3_FEATURE_INCOMPAT_FILETYPE 0x0002 +#define EXT3_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ +#define EXT3_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ +#define EXT3_FEATURE_INCOMPAT_META_BG 0x0010 + +#define EXT3_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR +#define EXT3_FEATURE_INCOMPAT_SUPP (EXT3_FEATURE_INCOMPAT_FILETYPE| \ + EXT3_FEATURE_INCOMPAT_RECOVER| \ + EXT3_FEATURE_INCOMPAT_META_BG) +#define EXT3_FEATURE_RO_COMPAT_SUPP (EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER| \ + EXT3_FEATURE_RO_COMPAT_LARGE_FILE| \ + EXT3_FEATURE_RO_COMPAT_BTREE_DIR) + +/* + * Default values for user and/or group using reserved blocks + */ +#define EXT3_DEF_RESUID 0 +#define EXT3_DEF_RESGID 0 + +/* + * Default mount options + */ +#define EXT3_DEFM_DEBUG 0x0001 +#define EXT3_DEFM_BSDGROUPS 0x0002 +#define EXT3_DEFM_XATTR_USER 0x0004 +#define EXT3_DEFM_ACL 0x0008 +#define EXT3_DEFM_UID16 0x0010 +#define EXT3_DEFM_JMODE 0x0060 +#define EXT3_DEFM_JMODE_DATA 0x0020 +#define EXT3_DEFM_JMODE_ORDERED 0x0040 +#define EXT3_DEFM_JMODE_WBACK 0x0060 + +/* + * Structure of a directory entry + */ +#define EXT3_NAME_LEN 255 + +struct ext3_dir_entry { + __le32 inode; /* Inode number */ + __le16 rec_len; /* Directory entry length */ + __le16 name_len; /* Name length */ + char name[EXT3_NAME_LEN]; /* File name */ +}; + +/* + * The new version of the directory entry. Since EXT3 structures are + * stored in intel byte order, and the name_len field could never be + * bigger than 255 chars, it's safe to reclaim the extra byte for the + * file_type field. + */ +struct ext3_dir_entry_2 { + __le32 inode; /* Inode number */ + __le16 rec_len; /* Directory entry length */ + __u8 name_len; /* Name length */ + __u8 file_type; + char name[EXT3_NAME_LEN]; /* File name */ +}; + +/* + * Ext3 directory file types. Only the low 3 bits are used. The + * other bits are reserved for now. + */ +#define EXT3_FT_UNKNOWN 0 +#define EXT3_FT_REG_FILE 1 +#define EXT3_FT_DIR 2 +#define EXT3_FT_CHRDEV 3 +#define EXT3_FT_BLKDEV 4 +#define EXT3_FT_FIFO 5 +#define EXT3_FT_SOCK 6 +#define EXT3_FT_SYMLINK 7 + +#define EXT3_FT_MAX 8 + +/* + * EXT3_DIR_PAD defines the directory entries boundaries + * + * NOTE: It must be a multiple of 4 + */ +#define EXT3_DIR_PAD 4 +#define EXT3_DIR_ROUND (EXT3_DIR_PAD - 1) +#define EXT3_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT3_DIR_ROUND) & \ + ~EXT3_DIR_ROUND) +/* + * Hash Tree Directory indexing + * (c) Daniel Phillips, 2001 + */ + +#ifdef CONFIG_EXT3_INDEX + #define is_dx(dir) (EXT3_HAS_COMPAT_FEATURE(dir->i_sb, \ + EXT3_FEATURE_COMPAT_DIR_INDEX) && \ + (EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) +#define EXT3_DIR_LINK_MAX(dir) (!is_dx(dir) && (dir)->i_nlink >= EXT3_LINK_MAX) +#define EXT3_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2 || (dir)->i_nlink == 1) +#else + #define is_dx(dir) 0 +#define EXT3_DIR_LINK_MAX(dir) ((dir)->i_nlink >= EXT3_LINK_MAX) +#define EXT3_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2) +#endif + +/* Legal values for the dx_root hash_version field: */ + +#define DX_HASH_LEGACY 0 +#define DX_HASH_HALF_MD4 1 +#define DX_HASH_TEA 2 + +#ifdef __KERNEL__ + +/* hash info structure used by the directory hash */ +struct dx_hash_info +{ + u32 hash; + u32 minor_hash; + int hash_version; + u32 *seed; +}; + +#define EXT3_HTREE_EOF 0x7fffffff + +/* + * Control parameters used by ext3_htree_next_block + */ +#define HASH_NB_ALWAYS 1 + + +/* + * Describe an inode's exact location on disk and in memory + */ +struct ext3_iloc +{ + struct buffer_head *bh; + unsigned long offset; + unsigned long block_group; +}; + +static inline struct ext3_inode *ext3_raw_inode(struct ext3_iloc *iloc) +{ + return (struct ext3_inode *) (iloc->bh->b_data + iloc->offset); +} + +/* + * This structure is stuffed into the struct file's private_data field + * for directories. It is where we put information so that we can do + * readdir operations in hash tree order. + */ +struct dir_private_info { + struct rb_root root; + struct rb_node *curr_node; + struct fname *extra_fname; + loff_t last_pos; + __u32 curr_hash; + __u32 curr_minor_hash; + __u32 next_hash; +}; + +/* calculate the first block number of the group */ +static inline ext3_fsblk_t +ext3_group_first_block_no(struct super_block *sb, unsigned long group_no) +{ + return group_no * (ext3_fsblk_t)EXT3_BLOCKS_PER_GROUP(sb) + + le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block); +} + +/* + * Special error return code only used by dx_probe() and its callers. + */ +#define ERR_BAD_DX_DIR -75000 + +/* + * Function prototypes + */ + +/* + * Ok, these declarations are also in but none of the + * ext3 source programs needs to include it so they are duplicated here. + */ +# define NORET_TYPE /**/ +# define ATTRIB_NORET __attribute__((noreturn)) +# define NORET_AND noreturn, + +/* balloc.c */ +extern int ext3_bg_has_super(struct super_block *sb, int group); +extern unsigned long ext3_bg_num_gdb(struct super_block *sb, int group); +extern ext3_fsblk_t ext3_new_block (handle_t *handle, struct inode *inode, + ext3_fsblk_t goal, int *errp); +extern ext3_fsblk_t ext3_new_blocks (handle_t *handle, struct inode *inode, + ext3_fsblk_t goal, unsigned long *count, int *errp); +extern void ext3_free_blocks (handle_t *handle, struct inode *inode, + ext3_fsblk_t block, unsigned long count); +extern void ext3_free_blocks_sb (handle_t *handle, struct super_block *sb, + ext3_fsblk_t block, unsigned long count, + unsigned long *pdquot_freed_blocks); +extern ext3_fsblk_t ext3_count_free_blocks (struct super_block *); +extern void ext3_check_blocks_bitmap (struct super_block *); +extern struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, + unsigned int block_group, + struct buffer_head ** bh); +extern int ext3_should_retry_alloc(struct super_block *sb, int *retries); +extern void ext3_init_block_alloc_info(struct inode *); +extern void ext3_rsv_window_add(struct super_block *sb, struct ext3_reserve_window_node *rsv); + +/* dir.c */ +extern int ext3_check_dir_entry(const char *, struct inode *, + struct ext3_dir_entry_2 *, + struct buffer_head *, unsigned long); +extern int ext3_htree_store_dirent(struct file *dir_file, __u32 hash, + __u32 minor_hash, + struct ext3_dir_entry_2 *dirent); +extern void ext3_htree_free_dir_info(struct dir_private_info *p); + +/* fsync.c */ +extern int ext3_sync_file (struct file *, struct dentry *, int); + +/* hash.c */ +extern int ext3fs_dirhash(const char *name, int len, struct + dx_hash_info *hinfo); + +/* ialloc.c */ +extern struct inode * ext3_new_inode (handle_t *, struct inode *, int); +extern void ext3_free_inode (handle_t *, struct inode *); +extern struct inode * ext3_orphan_get (struct super_block *, unsigned long); +extern unsigned long ext3_count_free_inodes (struct super_block *); +extern unsigned long ext3_count_dirs (struct super_block *); +extern void ext3_check_inodes_bitmap (struct super_block *); +extern unsigned long ext3_count_free (struct buffer_head *, unsigned); + + +/* inode.c */ +int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, + struct buffer_head *bh, ext3_fsblk_t blocknr); +struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *); +struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *); +int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, + sector_t iblock, unsigned long maxblocks, struct buffer_head *bh_result, + int create, int extend_disksize); + +extern void ext3_read_inode (struct inode *); +extern int ext3_write_inode (struct inode *, int); +extern int ext3_setattr (struct dentry *, struct iattr *); +extern void ext3_delete_inode (struct inode *); +extern int ext3_sync_inode (handle_t *, struct inode *); +extern void ext3_discard_reservation (struct inode *); +extern void ext3_dirty_inode(struct inode *); +extern int ext3_change_inode_journal_flag(struct inode *, int); +extern int ext3_get_inode_loc(struct inode *, struct ext3_iloc *); +extern void ext3_truncate (struct inode *); +extern void ext3_set_inode_flags(struct inode *); +extern void ext3_set_aops(struct inode *inode); + +/* ioctl.c */ +extern int ext3_ioctl (struct inode *, struct file *, unsigned int, + unsigned long); +extern long ext3_compat_ioctl (struct file *, unsigned int, unsigned long); + +/* namei.c */ +extern int ext3_orphan_add(handle_t *, struct inode *); +extern int ext3_orphan_del(handle_t *, struct inode *); +extern int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, + __u32 start_minor_hash, __u32 *next_hash); + +/* resize.c */ +extern int ext3_group_add(struct super_block *sb, + struct ext3_new_group_data *input); +extern int ext3_group_extend(struct super_block *sb, + struct ext3_super_block *es, + ext3_fsblk_t n_blocks_count); + +/* super.c */ +extern void ext3_error (struct super_block *, const char *, const char *, ...) + __attribute__ ((format (printf, 3, 4))); +extern void __ext3_std_error (struct super_block *, const char *, int); +extern void ext3_abort (struct super_block *, const char *, const char *, ...) + __attribute__ ((format (printf, 3, 4))); +extern void ext3_warning (struct super_block *, const char *, const char *, ...) + __attribute__ ((format (printf, 3, 4))); +extern void ext3_update_dynamic_rev (struct super_block *sb); + +#define ext3_std_error(sb, errno) \ +do { \ + if ((errno)) \ + __ext3_std_error((sb), __FUNCTION__, (errno)); \ +} while (0) + +/* + * Inodes and files operations + */ + +/* dir.c */ +extern const struct file_operations ext3_dir_operations; + +/* file.c */ +extern struct inode_operations ext3_file_inode_operations; +extern const struct file_operations ext3_file_operations; + +/* namei.c */ +extern struct inode_operations ext3_dir_inode_operations; +extern struct inode_operations ext3_special_inode_operations; + +/* symlink.c */ +extern struct inode_operations ext3_symlink_inode_operations; +extern struct inode_operations ext3_fast_symlink_inode_operations; + + +#endif /* __KERNEL__ */ + +#endif /* _LINUX_EXT3_FS_H */ diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h new file mode 100644 index 0000000..4395e52 --- /dev/null +++ b/include/linux/ext4_fs_i.h @@ -0,0 +1,147 @@ +/* + * linux/include/linux/ext3_fs_i.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs_i.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _LINUX_EXT3_FS_I +#define _LINUX_EXT3_FS_I + +#include +#include +#include +#include + +/* data type for block offset of block group */ +typedef int ext3_grpblk_t; + +/* data type for filesystem-wide blocks number */ +typedef unsigned long ext3_fsblk_t; + +#define E3FSBLK "%lu" + +struct ext3_reserve_window { + ext3_fsblk_t _rsv_start; /* First byte reserved */ + ext3_fsblk_t _rsv_end; /* Last byte reserved or 0 */ +}; + +struct ext3_reserve_window_node { + struct rb_node rsv_node; + __u32 rsv_goal_size; + __u32 rsv_alloc_hit; + struct ext3_reserve_window rsv_window; +}; + +struct ext3_block_alloc_info { + /* information about reservation window */ + struct ext3_reserve_window_node rsv_window_node; + /* + * was i_next_alloc_block in ext3_inode_info + * is the logical (file-relative) number of the + * most-recently-allocated block in this file. + * We use this for detecting linearly ascending allocation requests. + */ + __u32 last_alloc_logical_block; + /* + * Was i_next_alloc_goal in ext3_inode_info + * is the *physical* companion to i_next_alloc_block. + * it the the physical block number of the block which was most-recentl + * allocated to this file. This give us the goal (target) for the next + * allocation when we detect linearly ascending requests. + */ + ext3_fsblk_t last_alloc_physical_block; +}; + +#define rsv_start rsv_window._rsv_start +#define rsv_end rsv_window._rsv_end + +/* + * third extended file system inode data in memory + */ +struct ext3_inode_info { + __le32 i_data[15]; /* unconverted */ + __u32 i_flags; +#ifdef EXT3_FRAGMENTS + __u32 i_faddr; + __u8 i_frag_no; + __u8 i_frag_size; +#endif + ext3_fsblk_t i_file_acl; + __u32 i_dir_acl; + __u32 i_dtime; + + /* + * i_block_group is the number of the block group which contains + * this file's inode. Constant across the lifetime of the inode, + * it is ued for making block allocation decisions - we try to + * place a file's data blocks near its inode block, and new inodes + * near to their parent directory's inode. + */ + __u32 i_block_group; + __u32 i_state; /* Dynamic state flags for ext3 */ + + /* block reservation info */ + struct ext3_block_alloc_info *i_block_alloc_info; + + __u32 i_dir_start_lookup; +#ifdef CONFIG_EXT3_FS_XATTR + /* + * Extended attributes can be read independently of the main file + * data. Taking i_mutex even when reading would cause contention + * between readers of EAs and writers of regular file data, so + * instead we synchronize on xattr_sem when reading or changing + * EAs. + */ + struct rw_semaphore xattr_sem; +#endif +#ifdef CONFIG_EXT3_FS_POSIX_ACL + struct posix_acl *i_acl; + struct posix_acl *i_default_acl; +#endif + + struct list_head i_orphan; /* unlinked but open inodes */ + + /* + * i_disksize keeps track of what the inode size is ON DISK, not + * in memory. During truncate, i_size is set to the new size by + * the VFS prior to calling ext3_truncate(), but the filesystem won't + * set i_disksize to 0 until the truncate is actually under way. + * + * The intent is that i_disksize always represents the blocks which + * are used by this file. This allows recovery to restart truncate + * on orphans if we crash during truncate. We actually write i_disksize + * into the on-disk inode when writing inodes out, instead of i_size. + * + * The only time when i_disksize and i_size may be different is when + * a truncate is in progress. The only things which change i_disksize + * are ext3_get_block (growth) and ext3_truncate (shrinkth). + */ + loff_t i_disksize; + + /* on-disk additional length */ + __u16 i_extra_isize; + + /* + * truncate_mutex is for serialising ext3_truncate() against + * ext3_getblock(). In the 2.4 ext2 design, great chunks of inode's + * data tree are chopped off during truncate. We can't do that in + * ext3 because whenever we perform intermediate commits during + * truncate, the inode and all the metadata blocks *must* be in a + * consistent state which allows truncation of the orphans to restart + * during recovery. Hence we must fix the get_block-vs-truncate race + * by other means, so we have truncate_mutex. + */ + struct mutex truncate_mutex; + struct inode vfs_inode; +}; + +#endif /* _LINUX_EXT3_FS_I */ diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h new file mode 100644 index 0000000..f61309c --- /dev/null +++ b/include/linux/ext4_fs_sb.h @@ -0,0 +1,83 @@ +/* + * linux/include/linux/ext3_fs_sb.h + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/include/linux/minix_fs_sb.h + * + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +#ifndef _LINUX_EXT3_FS_SB +#define _LINUX_EXT3_FS_SB + +#ifdef __KERNEL__ +#include +#include +#include +#include +#endif +#include + +/* + * third extended-fs super-block data in memory + */ +struct ext3_sb_info { + unsigned long s_frag_size; /* Size of a fragment in bytes */ + unsigned long s_frags_per_block;/* Number of fragments per block */ + unsigned long s_inodes_per_block;/* Number of inodes per block */ + unsigned long s_frags_per_group;/* Number of fragments in a group */ + unsigned long s_blocks_per_group;/* Number of blocks in a group */ + unsigned long s_inodes_per_group;/* Number of inodes in a group */ + unsigned long s_itb_per_group; /* Number of inode table blocks per group */ + unsigned long s_gdb_count; /* Number of group descriptor blocks */ + unsigned long s_desc_per_block; /* Number of group descriptors per block */ + unsigned long s_groups_count; /* Number of groups in the fs */ + struct buffer_head * s_sbh; /* Buffer containing the super block */ + struct ext3_super_block * s_es; /* Pointer to the super block in the buffer */ + struct buffer_head ** s_group_desc; + unsigned long s_mount_opt; + uid_t s_resuid; + gid_t s_resgid; + unsigned short s_mount_state; + unsigned short s_pad; + int s_addr_per_block_bits; + int s_desc_per_block_bits; + int s_inode_size; + int s_first_ino; + spinlock_t s_next_gen_lock; + u32 s_next_generation; + u32 s_hash_seed[4]; + int s_def_hash_version; + struct percpu_counter s_freeblocks_counter; + struct percpu_counter s_freeinodes_counter; + struct percpu_counter s_dirs_counter; + struct blockgroup_lock s_blockgroup_lock; + + /* root of the per fs reservation window tree */ + spinlock_t s_rsv_window_lock; + struct rb_root s_rsv_window_root; + struct ext3_reserve_window_node s_rsv_window_head; + + /* Journaling */ + struct inode * s_journal_inode; + struct journal_s * s_journal; + struct list_head s_orphan; + unsigned long s_commit_interval; + struct block_device *journal_bdev; +#ifdef CONFIG_JBD_DEBUG + struct timer_list turn_ro_timer; /* For turning read-only (crash simulation) */ + wait_queue_head_t ro_wait_queue; /* For people waiting for the fs to go read-only */ +#endif +#ifdef CONFIG_QUOTA + char *s_qf_names[MAXQUOTAS]; /* Names of quota files with journalled quota */ + int s_jquota_fmt; /* Format of quota to use */ +#endif +}; + +#endif /* _LINUX_EXT3_FS_SB */ diff --git a/include/linux/ext4_jbd.h b/include/linux/ext4_jbd.h new file mode 100644 index 0000000..ce0e610 --- /dev/null +++ b/include/linux/ext4_jbd.h @@ -0,0 +1,268 @@ +/* + * linux/include/linux/ext3_jbd.h + * + * Written by Stephen C. Tweedie , 1999 + * + * Copyright 1998--1999 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Ext3-specific journaling extensions. + */ + +#ifndef _LINUX_EXT3_JBD_H +#define _LINUX_EXT3_JBD_H + +#include +#include +#include + +#define EXT3_JOURNAL(inode) (EXT3_SB((inode)->i_sb)->s_journal) + +/* Define the number of blocks we need to account to a transaction to + * modify one block of data. + * + * We may have to touch one inode, one bitmap buffer, up to three + * indirection blocks, the group and superblock summaries, and the data + * block to complete the transaction. */ + +#define EXT3_SINGLEDATA_TRANS_BLOCKS 8U + +/* Extended attribute operations touch at most two data buffers, + * two bitmap buffers, and two group summaries, in addition to the inode + * and the superblock, which are already accounted for. */ + +#define EXT3_XATTR_TRANS_BLOCKS 6U + +/* Define the minimum size for a transaction which modifies data. This + * needs to take into account the fact that we may end up modifying two + * quota files too (one for the group, one for the user quota). The + * superblock only gets updated once, of course, so don't bother + * counting that again for the quota updates. */ + +#define EXT3_DATA_TRANS_BLOCKS(sb) (EXT3_SINGLEDATA_TRANS_BLOCKS + \ + EXT3_XATTR_TRANS_BLOCKS - 2 + \ + 2*EXT3_QUOTA_TRANS_BLOCKS(sb)) + +/* Delete operations potentially hit one directory's namespace plus an + * entire inode, plus arbitrary amounts of bitmap/indirection data. Be + * generous. We can grow the delete transaction later if necessary. */ + +#define EXT3_DELETE_TRANS_BLOCKS(sb) (2 * EXT3_DATA_TRANS_BLOCKS(sb) + 64) + +/* Define an arbitrary limit for the amount of data we will anticipate + * writing to any given transaction. For unbounded transactions such as + * write(2) and truncate(2) we can write more than this, but we always + * start off at the maximum transaction size and grow the transaction + * optimistically as we go. */ + +#define EXT3_MAX_TRANS_DATA 64U + +/* We break up a large truncate or write transaction once the handle's + * buffer credits gets this low, we need either to extend the + * transaction or to start a new one. Reserve enough space here for + * inode, bitmap, superblock, group and indirection updates for at least + * one block, plus two quota updates. Quota allocations are not + * needed. */ + +#define EXT3_RESERVE_TRANS_BLOCKS 12U + +#define EXT3_INDEX_EXTRA_TRANS_BLOCKS 8 + +#ifdef CONFIG_QUOTA +/* Amount of blocks needed for quota update - we know that the structure was + * allocated so we need to update only inode+data */ +#define EXT3_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) +/* Amount of blocks needed for quota insert/delete - we do some block writes + * but inode, sb and group updates are done only once */ +#define EXT3_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ + (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) +#define EXT3_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ + (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) +#else +#define EXT3_QUOTA_TRANS_BLOCKS(sb) 0 +#define EXT3_QUOTA_INIT_BLOCKS(sb) 0 +#define EXT3_QUOTA_DEL_BLOCKS(sb) 0 +#endif + +int +ext3_mark_iloc_dirty(handle_t *handle, + struct inode *inode, + struct ext3_iloc *iloc); + +/* + * On success, We end up with an outstanding reference count against + * iloc->bh. This _must_ be cleaned up later. + */ + +int ext3_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext3_iloc *iloc); + +int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode); + +/* + * Wrapper functions with which ext3 calls into JBD. The intent here is + * to allow these to be turned into appropriate stubs so ext3 can control + * ext2 filesystems, so ext2+ext3 systems only nee one fs. This work hasn't + * been done yet. + */ + +void ext3_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err); + +static inline int +__ext3_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_undo_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext3_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_write_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline void +ext3_journal_release_buffer(handle_t *handle, struct buffer_head *bh) +{ + journal_release_buffer(handle, bh); +} + +static inline int +__ext3_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) +{ + int err = journal_forget(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext3_journal_revoke(const char *where, handle_t *handle, + unsigned long blocknr, struct buffer_head *bh) +{ + int err = journal_revoke(handle, blocknr, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext3_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_get_create_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext3_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_dirty_metadata(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + + +#define ext3_journal_get_undo_access(handle, bh) \ + __ext3_journal_get_undo_access(__FUNCTION__, (handle), (bh)) +#define ext3_journal_get_write_access(handle, bh) \ + __ext3_journal_get_write_access(__FUNCTION__, (handle), (bh)) +#define ext3_journal_revoke(handle, blocknr, bh) \ + __ext3_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) +#define ext3_journal_get_create_access(handle, bh) \ + __ext3_journal_get_create_access(__FUNCTION__, (handle), (bh)) +#define ext3_journal_dirty_metadata(handle, bh) \ + __ext3_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) +#define ext3_journal_forget(handle, bh) \ + __ext3_journal_forget(__FUNCTION__, (handle), (bh)) + +int ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh); + +handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks); +int __ext3_journal_stop(const char *where, handle_t *handle); + +static inline handle_t *ext3_journal_start(struct inode *inode, int nblocks) +{ + return ext3_journal_start_sb(inode->i_sb, nblocks); +} + +#define ext3_journal_stop(handle) \ + __ext3_journal_stop(__FUNCTION__, (handle)) + +static inline handle_t *ext3_journal_current_handle(void) +{ + return journal_current_handle(); +} + +static inline int ext3_journal_extend(handle_t *handle, int nblocks) +{ + return journal_extend(handle, nblocks); +} + +static inline int ext3_journal_restart(handle_t *handle, int nblocks) +{ + return journal_restart(handle, nblocks); +} + +static inline int ext3_journal_blocks_per_page(struct inode *inode) +{ + return journal_blocks_per_page(inode); +} + +static inline int ext3_journal_force_commit(journal_t *journal) +{ + return journal_force_commit(journal); +} + +/* super.c */ +int ext3_force_commit(struct super_block *sb); + +static inline int ext3_should_journal_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 1; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA) + return 1; + if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL) + return 1; + return 0; +} + +static inline int ext3_should_order_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL) + return 0; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA) + return 1; + return 0; +} + +static inline int ext3_should_writeback_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL) + return 0; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA) + return 1; + return 0; +} + +#endif /* _LINUX_EXT3_JBD_H */ -- cgit v0.10.2 From 617ba13b31fbf505cc21799826639ef24ed94af0 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:20:53 -0700 Subject: [PATCH] ext4: rename ext4 symbols to avoid duplication of ext3 symbols Mingming Cao originally did this work, and Shaggy reproduced it using some scripts from her. Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index 704cd44..09c4878 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -1,12 +1,12 @@ # -# Makefile for the linux ext3-filesystem routines. +# Makefile for the linux ext4-filesystem routines. # -obj-$(CONFIG_EXT3_FS) += ext3.o +obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o -ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ +ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ ioctl.o namei.o super.o symlink.o hash.o resize.o -ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o -ext3-$(CONFIG_EXT3_FS_POSIX_ACL) += acl.o -ext3-$(CONFIG_EXT3_FS_SECURITY) += xattr_security.o +ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o +ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o +ext4dev-$(CONFIG_EXT4DEV_FS_SECURITY) += xattr_security.o diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index 1e5038d..d143489 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/acl.c + * linux/fs/ext4/acl.c * * Copyright (C) 2001-2003 Andreas Gruenbacher, */ @@ -9,8 +9,8 @@ #include #include #include -#include -#include +#include +#include #include "xattr.h" #include "acl.h" @@ -18,7 +18,7 @@ * Convert from filesystem to in-memory representation. */ static struct posix_acl * -ext3_acl_from_disk(const void *value, size_t size) +ext4_acl_from_disk(const void *value, size_t size) { const char *end = (char *)value + size; int n, count; @@ -26,13 +26,13 @@ ext3_acl_from_disk(const void *value, size_t size) if (!value) return NULL; - if (size < sizeof(ext3_acl_header)) + if (size < sizeof(ext4_acl_header)) return ERR_PTR(-EINVAL); - if (((ext3_acl_header *)value)->a_version != - cpu_to_le32(EXT3_ACL_VERSION)) + if (((ext4_acl_header *)value)->a_version != + cpu_to_le32(EXT4_ACL_VERSION)) return ERR_PTR(-EINVAL); - value = (char *)value + sizeof(ext3_acl_header); - count = ext3_acl_count(size); + value = (char *)value + sizeof(ext4_acl_header); + count = ext4_acl_count(size); if (count < 0) return ERR_PTR(-EINVAL); if (count == 0) @@ -41,9 +41,9 @@ ext3_acl_from_disk(const void *value, size_t size) if (!acl) return ERR_PTR(-ENOMEM); for (n=0; n < count; n++) { - ext3_acl_entry *entry = - (ext3_acl_entry *)value; - if ((char *)value + sizeof(ext3_acl_entry_short) > end) + ext4_acl_entry *entry = + (ext4_acl_entry *)value; + if ((char *)value + sizeof(ext4_acl_entry_short) > end) goto fail; acl->a_entries[n].e_tag = le16_to_cpu(entry->e_tag); acl->a_entries[n].e_perm = le16_to_cpu(entry->e_perm); @@ -53,13 +53,13 @@ ext3_acl_from_disk(const void *value, size_t size) case ACL_MASK: case ACL_OTHER: value = (char *)value + - sizeof(ext3_acl_entry_short); + sizeof(ext4_acl_entry_short); acl->a_entries[n].e_id = ACL_UNDEFINED_ID; break; case ACL_USER: case ACL_GROUP: - value = (char *)value + sizeof(ext3_acl_entry); + value = (char *)value + sizeof(ext4_acl_entry); if ((char *)value > end) goto fail; acl->a_entries[n].e_id = @@ -83,21 +83,21 @@ fail: * Convert from in-memory to filesystem representation. */ static void * -ext3_acl_to_disk(const struct posix_acl *acl, size_t *size) +ext4_acl_to_disk(const struct posix_acl *acl, size_t *size) { - ext3_acl_header *ext_acl; + ext4_acl_header *ext_acl; char *e; size_t n; - *size = ext3_acl_size(acl->a_count); - ext_acl = kmalloc(sizeof(ext3_acl_header) + acl->a_count * - sizeof(ext3_acl_entry), GFP_KERNEL); + *size = ext4_acl_size(acl->a_count); + ext_acl = kmalloc(sizeof(ext4_acl_header) + acl->a_count * + sizeof(ext4_acl_entry), GFP_KERNEL); if (!ext_acl) return ERR_PTR(-ENOMEM); - ext_acl->a_version = cpu_to_le32(EXT3_ACL_VERSION); - e = (char *)ext_acl + sizeof(ext3_acl_header); + ext_acl->a_version = cpu_to_le32(EXT4_ACL_VERSION); + e = (char *)ext_acl + sizeof(ext4_acl_header); for (n=0; n < acl->a_count; n++) { - ext3_acl_entry *entry = (ext3_acl_entry *)e; + ext4_acl_entry *entry = (ext4_acl_entry *)e; entry->e_tag = cpu_to_le16(acl->a_entries[n].e_tag); entry->e_perm = cpu_to_le16(acl->a_entries[n].e_perm); switch(acl->a_entries[n].e_tag) { @@ -105,14 +105,14 @@ ext3_acl_to_disk(const struct posix_acl *acl, size_t *size) case ACL_GROUP: entry->e_id = cpu_to_le32(acl->a_entries[n].e_id); - e += sizeof(ext3_acl_entry); + e += sizeof(ext4_acl_entry); break; case ACL_USER_OBJ: case ACL_GROUP_OBJ: case ACL_MASK: case ACL_OTHER: - e += sizeof(ext3_acl_entry_short); + e += sizeof(ext4_acl_entry_short); break; default: @@ -127,12 +127,12 @@ fail: } static inline struct posix_acl * -ext3_iget_acl(struct inode *inode, struct posix_acl **i_acl) +ext4_iget_acl(struct inode *inode, struct posix_acl **i_acl) { - struct posix_acl *acl = EXT3_ACL_NOT_CACHED; + struct posix_acl *acl = EXT4_ACL_NOT_CACHED; spin_lock(&inode->i_lock); - if (*i_acl != EXT3_ACL_NOT_CACHED) + if (*i_acl != EXT4_ACL_NOT_CACHED) acl = posix_acl_dup(*i_acl); spin_unlock(&inode->i_lock); @@ -140,11 +140,11 @@ ext3_iget_acl(struct inode *inode, struct posix_acl **i_acl) } static inline void -ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl, +ext4_iset_acl(struct inode *inode, struct posix_acl **i_acl, struct posix_acl *acl) { spin_lock(&inode->i_lock); - if (*i_acl != EXT3_ACL_NOT_CACHED) + if (*i_acl != EXT4_ACL_NOT_CACHED) posix_acl_release(*i_acl); *i_acl = posix_acl_dup(acl); spin_unlock(&inode->i_lock); @@ -156,9 +156,9 @@ ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl, * inode->i_mutex: don't care */ static struct posix_acl * -ext3_get_acl(struct inode *inode, int type) +ext4_get_acl(struct inode *inode, int type) { - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); int name_index; char *value = NULL; struct posix_acl *acl; @@ -169,31 +169,31 @@ ext3_get_acl(struct inode *inode, int type) switch(type) { case ACL_TYPE_ACCESS: - acl = ext3_iget_acl(inode, &ei->i_acl); - if (acl != EXT3_ACL_NOT_CACHED) + acl = ext4_iget_acl(inode, &ei->i_acl); + if (acl != EXT4_ACL_NOT_CACHED) return acl; - name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; break; case ACL_TYPE_DEFAULT: - acl = ext3_iget_acl(inode, &ei->i_default_acl); - if (acl != EXT3_ACL_NOT_CACHED) + acl = ext4_iget_acl(inode, &ei->i_default_acl); + if (acl != EXT4_ACL_NOT_CACHED) return acl; - name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; break; default: return ERR_PTR(-EINVAL); } - retval = ext3_xattr_get(inode, name_index, "", NULL, 0); + retval = ext4_xattr_get(inode, name_index, "", NULL, 0); if (retval > 0) { value = kmalloc(retval, GFP_KERNEL); if (!value) return ERR_PTR(-ENOMEM); - retval = ext3_xattr_get(inode, name_index, "", value, retval); + retval = ext4_xattr_get(inode, name_index, "", value, retval); } if (retval > 0) - acl = ext3_acl_from_disk(value, retval); + acl = ext4_acl_from_disk(value, retval); else if (retval == -ENODATA || retval == -ENOSYS) acl = NULL; else @@ -203,11 +203,11 @@ ext3_get_acl(struct inode *inode, int type) if (!IS_ERR(acl)) { switch(type) { case ACL_TYPE_ACCESS: - ext3_iset_acl(inode, &ei->i_acl, acl); + ext4_iset_acl(inode, &ei->i_acl, acl); break; case ACL_TYPE_DEFAULT: - ext3_iset_acl(inode, &ei->i_default_acl, acl); + ext4_iset_acl(inode, &ei->i_default_acl, acl); break; } } @@ -217,13 +217,13 @@ ext3_get_acl(struct inode *inode, int type) /* * Set the access or default ACL of an inode. * - * inode->i_mutex: down unless called from ext3_new_inode + * inode->i_mutex: down unless called from ext4_new_inode */ static int -ext3_set_acl(handle_t *handle, struct inode *inode, int type, +ext4_set_acl(handle_t *handle, struct inode *inode, int type, struct posix_acl *acl) { - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); int name_index; void *value = NULL; size_t size = 0; @@ -234,7 +234,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type, switch(type) { case ACL_TYPE_ACCESS: - name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; if (acl) { mode_t mode = inode->i_mode; error = posix_acl_equiv_mode(acl, &mode); @@ -242,7 +242,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type, return error; else { inode->i_mode = mode; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); if (error == 0) acl = NULL; } @@ -250,7 +250,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type, break; case ACL_TYPE_DEFAULT: - name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; if (!S_ISDIR(inode->i_mode)) return acl ? -EACCES : 0; break; @@ -259,23 +259,23 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type, return -EINVAL; } if (acl) { - value = ext3_acl_to_disk(acl, &size); + value = ext4_acl_to_disk(acl, &size); if (IS_ERR(value)) return (int)PTR_ERR(value); } - error = ext3_xattr_set_handle(handle, inode, name_index, "", + error = ext4_xattr_set_handle(handle, inode, name_index, "", value, size, 0); kfree(value); if (!error) { switch(type) { case ACL_TYPE_ACCESS: - ext3_iset_acl(inode, &ei->i_acl, acl); + ext4_iset_acl(inode, &ei->i_acl, acl); break; case ACL_TYPE_DEFAULT: - ext3_iset_acl(inode, &ei->i_default_acl, acl); + ext4_iset_acl(inode, &ei->i_default_acl, acl); break; } } @@ -283,9 +283,9 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type, } static int -ext3_check_acl(struct inode *inode, int mask) +ext4_check_acl(struct inode *inode, int mask) { - struct posix_acl *acl = ext3_get_acl(inode, ACL_TYPE_ACCESS); + struct posix_acl *acl = ext4_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); @@ -299,26 +299,26 @@ ext3_check_acl(struct inode *inode, int mask) } int -ext3_permission(struct inode *inode, int mask, struct nameidata *nd) +ext4_permission(struct inode *inode, int mask, struct nameidata *nd) { - return generic_permission(inode, mask, ext3_check_acl); + return generic_permission(inode, mask, ext4_check_acl); } /* - * Initialize the ACLs of a new inode. Called from ext3_new_inode. + * Initialize the ACLs of a new inode. Called from ext4_new_inode. * * dir->i_mutex: down * inode->i_mutex: up (access to inode is still exclusive) */ int -ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) +ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) { struct posix_acl *acl = NULL; int error = 0; if (!S_ISLNK(inode->i_mode)) { if (test_opt(dir->i_sb, POSIX_ACL)) { - acl = ext3_get_acl(dir, ACL_TYPE_DEFAULT); + acl = ext4_get_acl(dir, ACL_TYPE_DEFAULT); if (IS_ERR(acl)) return PTR_ERR(acl); } @@ -330,7 +330,7 @@ ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) mode_t mode; if (S_ISDIR(inode->i_mode)) { - error = ext3_set_acl(handle, inode, + error = ext4_set_acl(handle, inode, ACL_TYPE_DEFAULT, acl); if (error) goto cleanup; @@ -346,7 +346,7 @@ ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) inode->i_mode = mode; if (error > 0) { /* This is an extended ACL */ - error = ext3_set_acl(handle, inode, + error = ext4_set_acl(handle, inode, ACL_TYPE_ACCESS, clone); } } @@ -372,7 +372,7 @@ cleanup: * inode->i_mutex: down */ int -ext3_acl_chmod(struct inode *inode) +ext4_acl_chmod(struct inode *inode) { struct posix_acl *acl, *clone; int error; @@ -381,7 +381,7 @@ ext3_acl_chmod(struct inode *inode) return -EOPNOTSUPP; if (!test_opt(inode->i_sb, POSIX_ACL)) return 0; - acl = ext3_get_acl(inode, ACL_TYPE_ACCESS); + acl = ext4_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl) || !acl) return PTR_ERR(acl); clone = posix_acl_clone(acl, GFP_KERNEL); @@ -394,17 +394,17 @@ ext3_acl_chmod(struct inode *inode) int retries = 0; retry: - handle = ext3_journal_start(inode, - EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); + handle = ext4_journal_start(inode, + EXT4_DATA_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) { error = PTR_ERR(handle); - ext3_std_error(inode->i_sb, error); + ext4_std_error(inode->i_sb, error); goto out; } - error = ext3_set_acl(handle, inode, ACL_TYPE_ACCESS, clone); - ext3_journal_stop(handle); + error = ext4_set_acl(handle, inode, ACL_TYPE_ACCESS, clone); + ext4_journal_stop(handle); if (error == -ENOSPC && - ext3_should_retry_alloc(inode->i_sb, &retries)) + ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; } out: @@ -416,7 +416,7 @@ out: * Extended attribute handlers */ static size_t -ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len, +ext4_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len, const char *name, size_t name_len) { const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS); @@ -429,7 +429,7 @@ ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len, } static size_t -ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len, +ext4_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len, const char *name, size_t name_len) { const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT); @@ -442,7 +442,7 @@ ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len, } static int -ext3_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size) +ext4_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size) { struct posix_acl *acl; int error; @@ -450,7 +450,7 @@ ext3_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size) if (!test_opt(inode->i_sb, POSIX_ACL)) return -EOPNOTSUPP; - acl = ext3_get_acl(inode, type); + acl = ext4_get_acl(inode, type); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl == NULL) @@ -462,25 +462,25 @@ ext3_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size) } static int -ext3_xattr_get_acl_access(struct inode *inode, const char *name, +ext4_xattr_get_acl_access(struct inode *inode, const char *name, void *buffer, size_t size) { if (strcmp(name, "") != 0) return -EINVAL; - return ext3_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size); + return ext4_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size); } static int -ext3_xattr_get_acl_default(struct inode *inode, const char *name, +ext4_xattr_get_acl_default(struct inode *inode, const char *name, void *buffer, size_t size) { if (strcmp(name, "") != 0) return -EINVAL; - return ext3_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size); + return ext4_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size); } static int -ext3_xattr_set_acl(struct inode *inode, int type, const void *value, +ext4_xattr_set_acl(struct inode *inode, int type, const void *value, size_t size) { handle_t *handle; @@ -505,12 +505,12 @@ ext3_xattr_set_acl(struct inode *inode, int type, const void *value, acl = NULL; retry: - handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); + handle = ext4_journal_start(inode, EXT4_DATA_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); - error = ext3_set_acl(handle, inode, type, acl); - ext3_journal_stop(handle); - if (error == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries)) + error = ext4_set_acl(handle, inode, type, acl); + ext4_journal_stop(handle); + if (error == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; release_and_out: @@ -519,33 +519,33 @@ release_and_out: } static int -ext3_xattr_set_acl_access(struct inode *inode, const char *name, +ext4_xattr_set_acl_access(struct inode *inode, const char *name, const void *value, size_t size, int flags) { if (strcmp(name, "") != 0) return -EINVAL; - return ext3_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size); + return ext4_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size); } static int -ext3_xattr_set_acl_default(struct inode *inode, const char *name, +ext4_xattr_set_acl_default(struct inode *inode, const char *name, const void *value, size_t size, int flags) { if (strcmp(name, "") != 0) return -EINVAL; - return ext3_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size); + return ext4_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size); } -struct xattr_handler ext3_xattr_acl_access_handler = { +struct xattr_handler ext4_xattr_acl_access_handler = { .prefix = POSIX_ACL_XATTR_ACCESS, - .list = ext3_xattr_list_acl_access, - .get = ext3_xattr_get_acl_access, - .set = ext3_xattr_set_acl_access, + .list = ext4_xattr_list_acl_access, + .get = ext4_xattr_get_acl_access, + .set = ext4_xattr_set_acl_access, }; -struct xattr_handler ext3_xattr_acl_default_handler = { +struct xattr_handler ext4_xattr_acl_default_handler = { .prefix = POSIX_ACL_XATTR_DEFAULT, - .list = ext3_xattr_list_acl_default, - .get = ext3_xattr_get_acl_default, - .set = ext3_xattr_set_acl_default, + .list = ext4_xattr_list_acl_default, + .get = ext4_xattr_get_acl_default, + .set = ext4_xattr_set_acl_default, }; diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index 0d1e627..26a5c1a 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -1,81 +1,81 @@ /* - File: fs/ext3/acl.h + File: fs/ext4/acl.h (C) 2001 Andreas Gruenbacher, */ #include -#define EXT3_ACL_VERSION 0x0001 +#define EXT4_ACL_VERSION 0x0001 typedef struct { __le16 e_tag; __le16 e_perm; __le32 e_id; -} ext3_acl_entry; +} ext4_acl_entry; typedef struct { __le16 e_tag; __le16 e_perm; -} ext3_acl_entry_short; +} ext4_acl_entry_short; typedef struct { __le32 a_version; -} ext3_acl_header; +} ext4_acl_header; -static inline size_t ext3_acl_size(int count) +static inline size_t ext4_acl_size(int count) { if (count <= 4) { - return sizeof(ext3_acl_header) + - count * sizeof(ext3_acl_entry_short); + return sizeof(ext4_acl_header) + + count * sizeof(ext4_acl_entry_short); } else { - return sizeof(ext3_acl_header) + - 4 * sizeof(ext3_acl_entry_short) + - (count - 4) * sizeof(ext3_acl_entry); + return sizeof(ext4_acl_header) + + 4 * sizeof(ext4_acl_entry_short) + + (count - 4) * sizeof(ext4_acl_entry); } } -static inline int ext3_acl_count(size_t size) +static inline int ext4_acl_count(size_t size) { ssize_t s; - size -= sizeof(ext3_acl_header); - s = size - 4 * sizeof(ext3_acl_entry_short); + size -= sizeof(ext4_acl_header); + s = size - 4 * sizeof(ext4_acl_entry_short); if (s < 0) { - if (size % sizeof(ext3_acl_entry_short)) + if (size % sizeof(ext4_acl_entry_short)) return -1; - return size / sizeof(ext3_acl_entry_short); + return size / sizeof(ext4_acl_entry_short); } else { - if (s % sizeof(ext3_acl_entry)) + if (s % sizeof(ext4_acl_entry)) return -1; - return s / sizeof(ext3_acl_entry) + 4; + return s / sizeof(ext4_acl_entry) + 4; } } -#ifdef CONFIG_EXT3_FS_POSIX_ACL +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL -/* Value for inode->u.ext3_i.i_acl and inode->u.ext3_i.i_default_acl +/* Value for inode->u.ext4_i.i_acl and inode->u.ext4_i.i_default_acl if the ACL has not been cached */ -#define EXT3_ACL_NOT_CACHED ((void *)-1) +#define EXT4_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext3_permission (struct inode *, int, struct nameidata *); -extern int ext3_acl_chmod (struct inode *); -extern int ext3_init_acl (handle_t *, struct inode *, struct inode *); +extern int ext4_permission (struct inode *, int, struct nameidata *); +extern int ext4_acl_chmod (struct inode *); +extern int ext4_init_acl (handle_t *, struct inode *, struct inode *); -#else /* CONFIG_EXT3_FS_POSIX_ACL */ +#else /* CONFIG_EXT4DEV_FS_POSIX_ACL */ #include -#define ext3_permission NULL +#define ext4_permission NULL static inline int -ext3_acl_chmod(struct inode *inode) +ext4_acl_chmod(struct inode *inode) { return 0; } static inline int -ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) +ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) { return 0; } -#endif /* CONFIG_EXT3_FS_POSIX_ACL */ +#endif /* CONFIG_EXT4DEV_FS_POSIX_ACL */ diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index b41a7d7..357e4e5 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/balloc.c + * linux/fs/ext4/balloc.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -15,8 +15,8 @@ #include #include #include -#include -#include +#include +#include #include #include @@ -32,30 +32,30 @@ * The file system contains group descriptors which are located after the * super block. Each descriptor contains the number of the bitmap block and * the free blocks count in the block. The descriptors are loaded in memory - * when a file system is mounted (see ext3_read_super). + * when a file system is mounted (see ext4_read_super). */ #define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) /** - * ext3_get_group_desc() -- load group descriptor from disk + * ext4_get_group_desc() -- load group descriptor from disk * @sb: super block * @block_group: given block group * @bh: pointer to the buffer head to store the block * group descriptor */ -struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, +struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, unsigned int block_group, struct buffer_head ** bh) { unsigned long group_desc; unsigned long offset; - struct ext3_group_desc * desc; - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_group_desc * desc; + struct ext4_sb_info *sbi = EXT4_SB(sb); if (block_group >= sbi->s_groups_count) { - ext3_error (sb, "ext3_get_group_desc", + ext4_error (sb, "ext4_get_group_desc", "block_group >= groups_count - " "block_group = %d, groups_count = %lu", block_group, sbi->s_groups_count); @@ -64,17 +64,17 @@ struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, } smp_rmb(); - group_desc = block_group >> EXT3_DESC_PER_BLOCK_BITS(sb); - offset = block_group & (EXT3_DESC_PER_BLOCK(sb) - 1); + group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); + offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); if (!sbi->s_group_desc[group_desc]) { - ext3_error (sb, "ext3_get_group_desc", + ext4_error (sb, "ext4_get_group_desc", "Group descriptor not loaded - " "block_group = %d, group_desc = %lu, desc = %lu", block_group, group_desc, offset); return NULL; } - desc = (struct ext3_group_desc *) sbi->s_group_desc[group_desc]->b_data; + desc = (struct ext4_group_desc *) sbi->s_group_desc[group_desc]->b_data; if (bh) *bh = sbi->s_group_desc[group_desc]; return desc + offset; @@ -93,15 +93,15 @@ struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, static struct buffer_head * read_block_bitmap(struct super_block *sb, unsigned int block_group) { - struct ext3_group_desc * desc; + struct ext4_group_desc * desc; struct buffer_head * bh = NULL; - desc = ext3_get_group_desc (sb, block_group, NULL); + desc = ext4_get_group_desc (sb, block_group, NULL); if (!desc) goto error_out; bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); if (!bh) - ext3_error (sb, "read_block_bitmap", + ext4_error (sb, "read_block_bitmap", "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); @@ -134,7 +134,7 @@ static void __rsv_window_dump(struct rb_root *root, int verbose, const char *fn) { struct rb_node *n; - struct ext3_reserve_window_node *rsv, *prev; + struct ext4_reserve_window_node *rsv, *prev; int bad; restart: @@ -144,7 +144,7 @@ restart: printk("Block Allocation Reservation Windows Map (%s):\n", fn); while (n) { - rsv = list_entry(n, struct ext3_reserve_window_node, rsv_node); + rsv = list_entry(n, struct ext4_reserve_window_node, rsv_node); if (verbose) printk("reservation window 0x%p " "start: %lu, end: %lu\n", @@ -196,13 +196,13 @@ restart: * otherwise, return 0; */ static int -goal_in_my_reservation(struct ext3_reserve_window *rsv, ext3_grpblk_t grp_goal, +goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal, unsigned int group, struct super_block * sb) { - ext3_fsblk_t group_first_block, group_last_block; + ext4_fsblk_t group_first_block, group_last_block; - group_first_block = ext3_group_first_block_no(sb, group); - group_last_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + group_first_block = ext4_group_first_block_no(sb, group); + group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); if ((rsv->_rsv_start > group_last_block) || (rsv->_rsv_end < group_first_block)) @@ -222,17 +222,17 @@ goal_in_my_reservation(struct ext3_reserve_window *rsv, ext3_grpblk_t grp_goal, * if the goal is not in any window. * Returns NULL if there are no windows or if all windows start after the goal. */ -static struct ext3_reserve_window_node * -search_reserve_window(struct rb_root *root, ext3_fsblk_t goal) +static struct ext4_reserve_window_node * +search_reserve_window(struct rb_root *root, ext4_fsblk_t goal) { struct rb_node *n = root->rb_node; - struct ext3_reserve_window_node *rsv; + struct ext4_reserve_window_node *rsv; if (!n) return NULL; do { - rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); + rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); if (goal < rsv->rsv_start) n = n->rb_left; @@ -249,33 +249,33 @@ search_reserve_window(struct rb_root *root, ext3_fsblk_t goal) */ if (rsv->rsv_start > goal) { n = rb_prev(&rsv->rsv_node); - rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); + rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); } return rsv; } /** - * ext3_rsv_window_add() -- Insert a window to the block reservation rb tree. + * ext4_rsv_window_add() -- Insert a window to the block reservation rb tree. * @sb: super block * @rsv: reservation window to add * * Must be called with rsv_lock hold. */ -void ext3_rsv_window_add(struct super_block *sb, - struct ext3_reserve_window_node *rsv) +void ext4_rsv_window_add(struct super_block *sb, + struct ext4_reserve_window_node *rsv) { - struct rb_root *root = &EXT3_SB(sb)->s_rsv_window_root; + struct rb_root *root = &EXT4_SB(sb)->s_rsv_window_root; struct rb_node *node = &rsv->rsv_node; - ext3_fsblk_t start = rsv->rsv_start; + ext4_fsblk_t start = rsv->rsv_start; struct rb_node ** p = &root->rb_node; struct rb_node * parent = NULL; - struct ext3_reserve_window_node *this; + struct ext4_reserve_window_node *this; while (*p) { parent = *p; - this = rb_entry(parent, struct ext3_reserve_window_node, rsv_node); + this = rb_entry(parent, struct ext4_reserve_window_node, rsv_node); if (start < this->rsv_start) p = &(*p)->rb_left; @@ -292,7 +292,7 @@ void ext3_rsv_window_add(struct super_block *sb, } /** - * ext3_rsv_window_remove() -- unlink a window from the reservation rb tree + * ext4_rsv_window_remove() -- unlink a window from the reservation rb tree * @sb: super block * @rsv: reservation window to remove * @@ -301,59 +301,59 @@ void ext3_rsv_window_add(struct super_block *sb, * rsv_lock hold. */ static void rsv_window_remove(struct super_block *sb, - struct ext3_reserve_window_node *rsv) + struct ext4_reserve_window_node *rsv) { - rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; rsv->rsv_alloc_hit = 0; - rb_erase(&rsv->rsv_node, &EXT3_SB(sb)->s_rsv_window_root); + rb_erase(&rsv->rsv_node, &EXT4_SB(sb)->s_rsv_window_root); } /* * rsv_is_empty() -- Check if the reservation window is allocated. * @rsv: given reservation window to check * - * returns 1 if the end block is EXT3_RESERVE_WINDOW_NOT_ALLOCATED. + * returns 1 if the end block is EXT4_RESERVE_WINDOW_NOT_ALLOCATED. */ -static inline int rsv_is_empty(struct ext3_reserve_window *rsv) +static inline int rsv_is_empty(struct ext4_reserve_window *rsv) { /* a valid reservation end block could not be 0 */ - return rsv->_rsv_end == EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + return rsv->_rsv_end == EXT4_RESERVE_WINDOW_NOT_ALLOCATED; } /** - * ext3_init_block_alloc_info() + * ext4_init_block_alloc_info() * @inode: file inode structure * * Allocate and initialize the reservation window structure, and - * link the window to the ext3 inode structure at last + * link the window to the ext4 inode structure at last * * The reservation window structure is only dynamically allocated - * and linked to ext3 inode the first time the open file - * needs a new block. So, before every ext3_new_block(s) call, for + * and linked to ext4 inode the first time the open file + * needs a new block. So, before every ext4_new_block(s) call, for * regular files, we should check whether the reservation window * structure exists or not. In the latter case, this function is called. * Fail to do so will result in block reservation being turned off for that * open file. * - * This function is called from ext3_get_blocks_handle(), also called + * This function is called from ext4_get_blocks_handle(), also called * when setting the reservation window size through ioctl before the file * is open for write (needs block allocation). * * Needs truncate_mutex protection prior to call this function. */ -void ext3_init_block_alloc_info(struct inode *inode) +void ext4_init_block_alloc_info(struct inode *inode) { - struct ext3_inode_info *ei = EXT3_I(inode); - struct ext3_block_alloc_info *block_i = ei->i_block_alloc_info; + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; struct super_block *sb = inode->i_sb; block_i = kmalloc(sizeof(*block_i), GFP_NOFS); if (block_i) { - struct ext3_reserve_window_node *rsv = &block_i->rsv_window_node; + struct ext4_reserve_window_node *rsv = &block_i->rsv_window_node; - rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; /* * if filesystem is mounted with NORESERVATION, the goal @@ -363,7 +363,7 @@ void ext3_init_block_alloc_info(struct inode *inode) if (!test_opt(sb, RESERVATION)) rsv->rsv_goal_size = 0; else - rsv->rsv_goal_size = EXT3_DEFAULT_RESERVE_BLOCKS; + rsv->rsv_goal_size = EXT4_DEFAULT_RESERVE_BLOCKS; rsv->rsv_alloc_hit = 0; block_i->last_alloc_logical_block = 0; block_i->last_alloc_physical_block = 0; @@ -372,24 +372,24 @@ void ext3_init_block_alloc_info(struct inode *inode) } /** - * ext3_discard_reservation() + * ext4_discard_reservation() * @inode: inode * * Discard(free) block reservation window on last file close, or truncate * or at last iput(). * * It is being called in three cases: - * ext3_release_file(): last writer close the file - * ext3_clear_inode(): last iput(), when nobody link to this file. - * ext3_truncate(): when the block indirect map is about to change. + * ext4_release_file(): last writer close the file + * ext4_clear_inode(): last iput(), when nobody link to this file. + * ext4_truncate(): when the block indirect map is about to change. * */ -void ext3_discard_reservation(struct inode *inode) +void ext4_discard_reservation(struct inode *inode) { - struct ext3_inode_info *ei = EXT3_I(inode); - struct ext3_block_alloc_info *block_i = ei->i_block_alloc_info; - struct ext3_reserve_window_node *rsv; - spinlock_t *rsv_lock = &EXT3_SB(inode->i_sb)->s_rsv_window_lock; + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; + struct ext4_reserve_window_node *rsv; + spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock; if (!block_i) return; @@ -404,62 +404,62 @@ void ext3_discard_reservation(struct inode *inode) } /** - * ext3_free_blocks_sb() -- Free given blocks and update quota + * ext4_free_blocks_sb() -- Free given blocks and update quota * @handle: handle to this transaction * @sb: super block * @block: start physcial block to free * @count: number of blocks to free * @pdquot_freed_blocks: pointer to quota */ -void ext3_free_blocks_sb(handle_t *handle, struct super_block *sb, - ext3_fsblk_t block, unsigned long count, +void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count, unsigned long *pdquot_freed_blocks) { struct buffer_head *bitmap_bh = NULL; struct buffer_head *gd_bh; unsigned long block_group; - ext3_grpblk_t bit; + ext4_grpblk_t bit; unsigned long i; unsigned long overflow; - struct ext3_group_desc * desc; - struct ext3_super_block * es; - struct ext3_sb_info *sbi; + struct ext4_group_desc * desc; + struct ext4_super_block * es; + struct ext4_sb_info *sbi; int err = 0, ret; - ext3_grpblk_t group_freed; + ext4_grpblk_t group_freed; *pdquot_freed_blocks = 0; - sbi = EXT3_SB(sb); + sbi = EXT4_SB(sb); es = sbi->s_es; if (block < le32_to_cpu(es->s_first_data_block) || block + count < block || block + count > le32_to_cpu(es->s_blocks_count)) { - ext3_error (sb, "ext3_free_blocks", + ext4_error (sb, "ext4_free_blocks", "Freeing blocks not in datazone - " "block = "E3FSBLK", count = %lu", block, count); goto error_return; } - ext3_debug ("freeing block(s) %lu-%lu\n", block, block + count - 1); + ext4_debug ("freeing block(s) %lu-%lu\n", block, block + count - 1); do_more: overflow = 0; block_group = (block - le32_to_cpu(es->s_first_data_block)) / - EXT3_BLOCKS_PER_GROUP(sb); + EXT4_BLOCKS_PER_GROUP(sb); bit = (block - le32_to_cpu(es->s_first_data_block)) % - EXT3_BLOCKS_PER_GROUP(sb); + EXT4_BLOCKS_PER_GROUP(sb); /* * Check to see if we are freeing blocks across a group * boundary. */ - if (bit + count > EXT3_BLOCKS_PER_GROUP(sb)) { - overflow = bit + count - EXT3_BLOCKS_PER_GROUP(sb); + if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) { + overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb); count -= overflow; } brelse(bitmap_bh); bitmap_bh = read_block_bitmap(sb, block_group); if (!bitmap_bh) goto error_return; - desc = ext3_get_group_desc (sb, block_group, &gd_bh); + desc = ext4_get_group_desc (sb, block_group, &gd_bh); if (!desc) goto error_return; @@ -469,7 +469,7 @@ do_more: sbi->s_itb_per_group) || in_range (block + count - 1, le32_to_cpu(desc->bg_inode_table), sbi->s_itb_per_group)) - ext3_error (sb, "ext3_free_blocks", + ext4_error (sb, "ext4_free_blocks", "Freeing blocks in system zones - " "Block = "E3FSBLK", count = %lu", block, count); @@ -480,7 +480,7 @@ do_more: */ /* @@@ check errors */ BUFFER_TRACE(bitmap_bh, "getting undo access"); - err = ext3_journal_get_undo_access(handle, bitmap_bh); + err = ext4_journal_get_undo_access(handle, bitmap_bh); if (err) goto error_return; @@ -490,7 +490,7 @@ do_more: * using it */ BUFFER_TRACE(gd_bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, gd_bh); + err = ext4_journal_get_write_access(handle, gd_bh); if (err) goto error_return; @@ -542,7 +542,7 @@ do_more: BUFFER_TRACE(bitmap_bh, "set in b_committed_data"); J_ASSERT_BH(bitmap_bh, bh2jh(bitmap_bh)->b_committed_data != NULL); - ext3_set_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, + ext4_set_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, bh2jh(bitmap_bh)->b_committed_data); /* @@ -551,10 +551,10 @@ do_more: * the allocator uses. */ BUFFER_TRACE(bitmap_bh, "clear bit"); - if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group), + if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, bitmap_bh->b_data)) { jbd_unlock_bh_state(bitmap_bh); - ext3_error(sb, __FUNCTION__, + ext4_error(sb, __FUNCTION__, "bit already cleared for block "E3FSBLK, block + i); jbd_lock_bh_state(bitmap_bh); @@ -574,11 +574,11 @@ do_more: /* We dirtied the bitmap block */ BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); - err = ext3_journal_dirty_metadata(handle, bitmap_bh); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); /* And the group descriptor block */ BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); - ret = ext3_journal_dirty_metadata(handle, gd_bh); + ret = ext4_journal_dirty_metadata(handle, gd_bh); if (!err) err = ret; *pdquot_freed_blocks += group_freed; @@ -590,40 +590,40 @@ do_more: sb->s_dirt = 1; error_return: brelse(bitmap_bh); - ext3_std_error(sb, err); + ext4_std_error(sb, err); return; } /** - * ext3_free_blocks() -- Free given blocks and update quota + * ext4_free_blocks() -- Free given blocks and update quota * @handle: handle for this transaction * @inode: inode * @block: start physical block to free * @count: number of blocks to count */ -void ext3_free_blocks(handle_t *handle, struct inode *inode, - ext3_fsblk_t block, unsigned long count) +void ext4_free_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t block, unsigned long count) { struct super_block * sb; unsigned long dquot_freed_blocks; sb = inode->i_sb; if (!sb) { - printk ("ext3_free_blocks: nonexistent device"); + printk ("ext4_free_blocks: nonexistent device"); return; } - ext3_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks); + ext4_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks); if (dquot_freed_blocks) DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); return; } /** - * ext3_test_allocatable() + * ext4_test_allocatable() * @nr: given allocation block group * @bh: bufferhead contains the bitmap of the given block group * - * For ext3 allocations, we must not reuse any blocks which are + * For ext4 allocations, we must not reuse any blocks which are * allocated in the bitmap buffer's "last committed data" copy. This * prevents deletes from freeing up the page for reuse until we have * committed the delete transaction. @@ -638,19 +638,19 @@ void ext3_free_blocks(handle_t *handle, struct inode *inode, * data-writes at some point, and disable it for metadata allocations or * sync-data inodes. */ -static int ext3_test_allocatable(ext3_grpblk_t nr, struct buffer_head *bh) +static int ext4_test_allocatable(ext4_grpblk_t nr, struct buffer_head *bh) { int ret; struct journal_head *jh = bh2jh(bh); - if (ext3_test_bit(nr, bh->b_data)) + if (ext4_test_bit(nr, bh->b_data)) return 0; jbd_lock_bh_state(bh); if (!jh->b_committed_data) ret = 1; else - ret = !ext3_test_bit(nr, jh->b_committed_data); + ret = !ext4_test_bit(nr, jh->b_committed_data); jbd_unlock_bh_state(bh); return ret; } @@ -665,22 +665,22 @@ static int ext3_test_allocatable(ext3_grpblk_t nr, struct buffer_head *bh) * bitmap on disk and the last-committed copy in journal, until we find a * bit free in both bitmaps. */ -static ext3_grpblk_t -bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, - ext3_grpblk_t maxblocks) +static ext4_grpblk_t +bitmap_search_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, + ext4_grpblk_t maxblocks) { - ext3_grpblk_t next; + ext4_grpblk_t next; struct journal_head *jh = bh2jh(bh); while (start < maxblocks) { - next = ext3_find_next_zero_bit(bh->b_data, maxblocks, start); + next = ext4_find_next_zero_bit(bh->b_data, maxblocks, start); if (next >= maxblocks) return -1; - if (ext3_test_allocatable(next, bh)) + if (ext4_test_allocatable(next, bh)) return next; jbd_lock_bh_state(bh); if (jh->b_committed_data) - start = ext3_find_next_zero_bit(jh->b_committed_data, + start = ext4_find_next_zero_bit(jh->b_committed_data, maxblocks, next); jbd_unlock_bh_state(bh); } @@ -700,11 +700,11 @@ bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, * the initial goal; then for a free byte somewhere in the bitmap; then * for any free bit in the bitmap. */ -static ext3_grpblk_t -find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, - ext3_grpblk_t maxblocks) +static ext4_grpblk_t +find_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, + ext4_grpblk_t maxblocks) { - ext3_grpblk_t here, next; + ext4_grpblk_t here, next; char *p, *r; if (start > 0) { @@ -713,16 +713,16 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, * block within the next XX blocks. * * end_goal is more or less random, but it has to be - * less than EXT3_BLOCKS_PER_GROUP. Aligning up to the + * less than EXT4_BLOCKS_PER_GROUP. Aligning up to the * next 64-bit boundary is simple.. */ - ext3_grpblk_t end_goal = (start + 63) & ~63; + ext4_grpblk_t end_goal = (start + 63) & ~63; if (end_goal > maxblocks) end_goal = maxblocks; - here = ext3_find_next_zero_bit(bh->b_data, end_goal, start); - if (here < end_goal && ext3_test_allocatable(here, bh)) + here = ext4_find_next_zero_bit(bh->b_data, end_goal, start); + if (here < end_goal && ext4_test_allocatable(here, bh)) return here; - ext3_debug("Bit not found near goal\n"); + ext4_debug("Bit not found near goal\n"); } here = start; @@ -733,7 +733,7 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, r = memscan(p, 0, (maxblocks - here + 7) >> 3); next = (r - ((char *)bh->b_data)) << 3; - if (next < maxblocks && next >= start && ext3_test_allocatable(next, bh)) + if (next < maxblocks && next >= start && ext4_test_allocatable(next, bh)) return next; /* @@ -757,16 +757,16 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, * zero (failure). */ static inline int -claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh) +claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh) { struct journal_head *jh = bh2jh(bh); int ret; - if (ext3_set_bit_atomic(lock, block, bh->b_data)) + if (ext4_set_bit_atomic(lock, block, bh->b_data)) return 0; jbd_lock_bh_state(bh); - if (jh->b_committed_data && ext3_test_bit(block,jh->b_committed_data)) { - ext3_clear_bit_atomic(lock, block, bh->b_data); + if (jh->b_committed_data && ext4_test_bit(block,jh->b_committed_data)) { + ext4_clear_bit_atomic(lock, block, bh->b_data); ret = 0; } else { ret = 1; @@ -776,7 +776,7 @@ claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh) } /** - * ext3_try_to_allocate() + * ext4_try_to_allocate() * @sb: superblock * @handle: handle to this transaction * @group: given allocation block group @@ -797,29 +797,29 @@ claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh) * * If we failed to allocate the desired block then we may end up crossing to a * new bitmap. In that case we must release write access to the old one via - * ext3_journal_release_buffer(), else we'll run out of credits. + * ext4_journal_release_buffer(), else we'll run out of credits. */ -static ext3_grpblk_t -ext3_try_to_allocate(struct super_block *sb, handle_t *handle, int group, - struct buffer_head *bitmap_bh, ext3_grpblk_t grp_goal, - unsigned long *count, struct ext3_reserve_window *my_rsv) +static ext4_grpblk_t +ext4_try_to_allocate(struct super_block *sb, handle_t *handle, int group, + struct buffer_head *bitmap_bh, ext4_grpblk_t grp_goal, + unsigned long *count, struct ext4_reserve_window *my_rsv) { - ext3_fsblk_t group_first_block; - ext3_grpblk_t start, end; + ext4_fsblk_t group_first_block; + ext4_grpblk_t start, end; unsigned long num = 0; /* we do allocation within the reservation window if we have a window */ if (my_rsv) { - group_first_block = ext3_group_first_block_no(sb, group); + group_first_block = ext4_group_first_block_no(sb, group); if (my_rsv->_rsv_start >= group_first_block) start = my_rsv->_rsv_start - group_first_block; else /* reservation window cross group boundary */ start = 0; end = my_rsv->_rsv_end - group_first_block + 1; - if (end > EXT3_BLOCKS_PER_GROUP(sb)) + if (end > EXT4_BLOCKS_PER_GROUP(sb)) /* reservation window crosses group boundary */ - end = EXT3_BLOCKS_PER_GROUP(sb); + end = EXT4_BLOCKS_PER_GROUP(sb); if ((start <= grp_goal) && (grp_goal < end)) start = grp_goal; else @@ -829,13 +829,13 @@ ext3_try_to_allocate(struct super_block *sb, handle_t *handle, int group, start = grp_goal; else start = 0; - end = EXT3_BLOCKS_PER_GROUP(sb); + end = EXT4_BLOCKS_PER_GROUP(sb); } - BUG_ON(start > EXT3_BLOCKS_PER_GROUP(sb)); + BUG_ON(start > EXT4_BLOCKS_PER_GROUP(sb)); repeat: - if (grp_goal < 0 || !ext3_test_allocatable(grp_goal, bitmap_bh)) { + if (grp_goal < 0 || !ext4_test_allocatable(grp_goal, bitmap_bh)) { grp_goal = find_next_usable_block(start, bitmap_bh, end); if (grp_goal < 0) goto fail_access; @@ -843,7 +843,7 @@ repeat: int i; for (i = 0; i < 7 && grp_goal > start && - ext3_test_allocatable(grp_goal - 1, + ext4_test_allocatable(grp_goal - 1, bitmap_bh); i++, grp_goal--) ; @@ -851,7 +851,7 @@ repeat: } start = grp_goal; - if (!claim_block(sb_bgl_lock(EXT3_SB(sb), group), + if (!claim_block(sb_bgl_lock(EXT4_SB(sb), group), grp_goal, bitmap_bh)) { /* * The block was allocated by another thread, or it was @@ -866,8 +866,8 @@ repeat: num++; grp_goal++; while (num < *count && grp_goal < end - && ext3_test_allocatable(grp_goal, bitmap_bh) - && claim_block(sb_bgl_lock(EXT3_SB(sb), group), + && ext4_test_allocatable(grp_goal, bitmap_bh) + && claim_block(sb_bgl_lock(EXT4_SB(sb), group), grp_goal, bitmap_bh)) { num++; grp_goal++; @@ -913,15 +913,15 @@ fail_access: * */ static int find_next_reservable_window( - struct ext3_reserve_window_node *search_head, - struct ext3_reserve_window_node *my_rsv, + struct ext4_reserve_window_node *search_head, + struct ext4_reserve_window_node *my_rsv, struct super_block * sb, - ext3_fsblk_t start_block, - ext3_fsblk_t last_block) + ext4_fsblk_t start_block, + ext4_fsblk_t last_block) { struct rb_node *next; - struct ext3_reserve_window_node *rsv, *prev; - ext3_fsblk_t cur; + struct ext4_reserve_window_node *rsv, *prev; + ext4_fsblk_t cur; int size = my_rsv->rsv_goal_size; /* TODO: make the start of the reservation window byte-aligned */ @@ -949,7 +949,7 @@ static int find_next_reservable_window( prev = rsv; next = rb_next(&rsv->rsv_node); - rsv = list_entry(next,struct ext3_reserve_window_node,rsv_node); + rsv = list_entry(next,struct ext4_reserve_window_node,rsv_node); /* * Reached the last reservation, we can just append to the @@ -992,7 +992,7 @@ static int find_next_reservable_window( my_rsv->rsv_alloc_hit = 0; if (prev != my_rsv) - ext3_rsv_window_add(sb, my_rsv); + ext4_rsv_window_add(sb, my_rsv); return 0; } @@ -1034,20 +1034,20 @@ static int find_next_reservable_window( * @bitmap_bh: the block group block bitmap * */ -static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv, - ext3_grpblk_t grp_goal, struct super_block *sb, +static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv, + ext4_grpblk_t grp_goal, struct super_block *sb, unsigned int group, struct buffer_head *bitmap_bh) { - struct ext3_reserve_window_node *search_head; - ext3_fsblk_t group_first_block, group_end_block, start_block; - ext3_grpblk_t first_free_block; - struct rb_root *fs_rsv_root = &EXT3_SB(sb)->s_rsv_window_root; + struct ext4_reserve_window_node *search_head; + ext4_fsblk_t group_first_block, group_end_block, start_block; + ext4_grpblk_t first_free_block; + struct rb_root *fs_rsv_root = &EXT4_SB(sb)->s_rsv_window_root; unsigned long size; int ret; - spinlock_t *rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock; + spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; - group_first_block = ext3_group_first_block_no(sb, group); - group_end_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + group_first_block = ext4_group_first_block_no(sb, group); + group_end_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); if (grp_goal < 0) start_block = group_first_block; @@ -1085,8 +1085,8 @@ static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv, * otherwise we keep the same size window */ size = size * 2; - if (size > EXT3_MAX_RESERVE_BLOCKS) - size = EXT3_MAX_RESERVE_BLOCKS; + if (size > EXT4_MAX_RESERVE_BLOCKS) + size = EXT4_MAX_RESERVE_BLOCKS; my_rsv->rsv_goal_size= size; } } @@ -1170,20 +1170,20 @@ retry: * Attempt to expand the reservation window large enough to have * required number of free blocks * - * Since ext3_try_to_allocate() will always allocate blocks within + * Since ext4_try_to_allocate() will always allocate blocks within * the reservation window range, if the window size is too small, * multiple blocks allocation has to stop at the end of the reservation * window. To make this more efficient, given the total number of * blocks needed and the current size of the window, we try to * expand the reservation window size if necessary on a best-effort - * basis before ext3_new_blocks() tries to allocate blocks, + * basis before ext4_new_blocks() tries to allocate blocks, */ -static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv, +static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv, struct super_block *sb, int size) { - struct ext3_reserve_window_node *next_rsv; + struct ext4_reserve_window_node *next_rsv; struct rb_node *next; - spinlock_t *rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock; + spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; if (!spin_trylock(rsv_lock)) return; @@ -1193,7 +1193,7 @@ static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv, if (!next) my_rsv->rsv_end += size; else { - next_rsv = list_entry(next, struct ext3_reserve_window_node, rsv_node); + next_rsv = list_entry(next, struct ext4_reserve_window_node, rsv_node); if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) my_rsv->rsv_end += size; @@ -1204,7 +1204,7 @@ static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv, } /** - * ext3_try_to_allocate_with_rsv() + * ext4_try_to_allocate_with_rsv() * @sb: superblock * @handle: handle to this transaction * @group: given allocation block group @@ -1232,15 +1232,15 @@ static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv, * We use a red-black tree for the per-filesystem reservation list. * */ -static ext3_grpblk_t -ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, +static ext4_grpblk_t +ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, unsigned int group, struct buffer_head *bitmap_bh, - ext3_grpblk_t grp_goal, - struct ext3_reserve_window_node * my_rsv, + ext4_grpblk_t grp_goal, + struct ext4_reserve_window_node * my_rsv, unsigned long *count, int *errp) { - ext3_fsblk_t group_first_block, group_last_block; - ext3_grpblk_t ret = 0; + ext4_fsblk_t group_first_block, group_last_block; + ext4_grpblk_t ret = 0; int fatal; unsigned long num = *count; @@ -1252,7 +1252,7 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, * if the buffer is in BJ_Forget state in the committing transaction. */ BUFFER_TRACE(bitmap_bh, "get undo access for new block"); - fatal = ext3_journal_get_undo_access(handle, bitmap_bh); + fatal = ext4_journal_get_undo_access(handle, bitmap_bh); if (fatal) { *errp = fatal; return -1; @@ -1265,18 +1265,18 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, * or last attempt to allocate a block with reservation turned on failed */ if (my_rsv == NULL ) { - ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, + ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, grp_goal, count, NULL); goto out; } /* * grp_goal is a group relative block number (if there is a goal) - * 0 < grp_goal < EXT3_BLOCKS_PER_GROUP(sb) + * 0 < grp_goal < EXT4_BLOCKS_PER_GROUP(sb) * first block is a filesystem wide block number * first block is the block number of the first block in this group */ - group_first_block = ext3_group_first_block_no(sb, group); - group_last_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1); + group_first_block = ext4_group_first_block_no(sb, group); + group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); /* * Basically we will allocate a new block from inode's reservation @@ -1314,10 +1314,10 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, if ((my_rsv->rsv_start > group_last_block) || (my_rsv->rsv_end < group_first_block)) { - rsv_window_dump(&EXT3_SB(sb)->s_rsv_window_root, 1); + rsv_window_dump(&EXT4_SB(sb)->s_rsv_window_root, 1); BUG(); } - ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, + ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, grp_goal, &num, &my_rsv->rsv_window); if (ret >= 0) { my_rsv->rsv_alloc_hit += num; @@ -1330,7 +1330,7 @@ out: if (ret >= 0) { BUFFER_TRACE(bitmap_bh, "journal_dirty_metadata for " "bitmap block"); - fatal = ext3_journal_dirty_metadata(handle, bitmap_bh); + fatal = ext4_journal_dirty_metadata(handle, bitmap_bh); if (fatal) { *errp = fatal; return -1; @@ -1339,19 +1339,19 @@ out: } BUFFER_TRACE(bitmap_bh, "journal_release_buffer"); - ext3_journal_release_buffer(handle, bitmap_bh); + ext4_journal_release_buffer(handle, bitmap_bh); return ret; } /** - * ext3_has_free_blocks() + * ext4_has_free_blocks() * @sbi: in-core super block structure. * * Check if filesystem has at least 1 free block available for allocation. */ -static int ext3_has_free_blocks(struct ext3_sb_info *sbi) +static int ext4_has_free_blocks(struct ext4_sb_info *sbi) { - ext3_fsblk_t free_blocks, root_blocks; + ext4_fsblk_t free_blocks, root_blocks; free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); root_blocks = le32_to_cpu(sbi->s_es->s_r_blocks_count); @@ -1364,63 +1364,63 @@ static int ext3_has_free_blocks(struct ext3_sb_info *sbi) } /** - * ext3_should_retry_alloc() + * ext4_should_retry_alloc() * @sb: super block * @retries number of attemps has been made * - * ext3_should_retry_alloc() is called when ENOSPC is returned, and if + * ext4_should_retry_alloc() is called when ENOSPC is returned, and if * it is profitable to retry the operation, this function will wait * for the current or commiting transaction to complete, and then * return TRUE. * * if the total number of retries exceed three times, return FALSE. */ -int ext3_should_retry_alloc(struct super_block *sb, int *retries) +int ext4_should_retry_alloc(struct super_block *sb, int *retries) { - if (!ext3_has_free_blocks(EXT3_SB(sb)) || (*retries)++ > 3) + if (!ext4_has_free_blocks(EXT4_SB(sb)) || (*retries)++ > 3) return 0; jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); - return journal_force_commit_nested(EXT3_SB(sb)->s_journal); + return journal_force_commit_nested(EXT4_SB(sb)->s_journal); } /** - * ext3_new_blocks() -- core block(s) allocation function + * ext4_new_blocks() -- core block(s) allocation function * @handle: handle to this transaction * @inode: file inode * @goal: given target block(filesystem wide) * @count: target number of blocks to allocate * @errp: error code * - * ext3_new_blocks uses a goal block to assist allocation. It tries to + * ext4_new_blocks uses a goal block to assist allocation. It tries to * allocate block(s) from the block group contains the goal block first. If that * fails, it will try to allocate block(s) from other block groups without * any specific goal block. * */ -ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode, - ext3_fsblk_t goal, unsigned long *count, int *errp) +ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp) { struct buffer_head *bitmap_bh = NULL; struct buffer_head *gdp_bh; int group_no; int goal_group; - ext3_grpblk_t grp_target_blk; /* blockgroup relative goal block */ - ext3_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ - ext3_fsblk_t ret_block; /* filesyetem-wide allocated block */ + ext4_grpblk_t grp_target_blk; /* blockgroup relative goal block */ + ext4_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ + ext4_fsblk_t ret_block; /* filesyetem-wide allocated block */ int bgi; /* blockgroup iteration index */ int fatal = 0, err; int performed_allocation = 0; - ext3_grpblk_t free_blocks; /* number of free blocks in a group */ + ext4_grpblk_t free_blocks; /* number of free blocks in a group */ struct super_block *sb; - struct ext3_group_desc *gdp; - struct ext3_super_block *es; - struct ext3_sb_info *sbi; - struct ext3_reserve_window_node *my_rsv = NULL; - struct ext3_block_alloc_info *block_i; + struct ext4_group_desc *gdp; + struct ext4_super_block *es; + struct ext4_sb_info *sbi; + struct ext4_reserve_window_node *my_rsv = NULL; + struct ext4_block_alloc_info *block_i; unsigned short windowsz = 0; -#ifdef EXT3FS_DEBUG +#ifdef EXT4FS_DEBUG static int goal_hits, goal_attempts; #endif unsigned long ngroups; @@ -1429,7 +1429,7 @@ ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode, *errp = -ENOSPC; sb = inode->i_sb; if (!sb) { - printk("ext3_new_block: nonexistent device"); + printk("ext4_new_block: nonexistent device"); return 0; } @@ -1441,22 +1441,22 @@ ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode, return 0; } - sbi = EXT3_SB(sb); - es = EXT3_SB(sb)->s_es; - ext3_debug("goal=%lu.\n", goal); + sbi = EXT4_SB(sb); + es = EXT4_SB(sb)->s_es; + ext4_debug("goal=%lu.\n", goal); /* * Allocate a block from reservation only when * filesystem is mounted with reservation(default,-o reservation), and * it's a regular file, and * the desired window size is greater than 0 (One could use ioctl - * command EXT3_IOC_SETRSVSZ to set the window size to 0 to turn off + * command EXT4_IOC_SETRSVSZ to set the window size to 0 to turn off * reservation on that particular file) */ - block_i = EXT3_I(inode)->i_block_alloc_info; + block_i = EXT4_I(inode)->i_block_alloc_info; if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0)) my_rsv = &block_i->rsv_window_node; - if (!ext3_has_free_blocks(sbi)) { + if (!ext4_has_free_blocks(sbi)) { *errp = -ENOSPC; goto out; } @@ -1468,10 +1468,10 @@ ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode, goal >= le32_to_cpu(es->s_blocks_count)) goal = le32_to_cpu(es->s_first_data_block); group_no = (goal - le32_to_cpu(es->s_first_data_block)) / - EXT3_BLOCKS_PER_GROUP(sb); + EXT4_BLOCKS_PER_GROUP(sb); goal_group = group_no; retry_alloc: - gdp = ext3_get_group_desc(sb, group_no, &gdp_bh); + gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); if (!gdp) goto io_error; @@ -1486,11 +1486,11 @@ retry_alloc: if (free_blocks > 0) { grp_target_blk = ((goal - le32_to_cpu(es->s_first_data_block)) % - EXT3_BLOCKS_PER_GROUP(sb)); + EXT4_BLOCKS_PER_GROUP(sb)); bitmap_bh = read_block_bitmap(sb, group_no); if (!bitmap_bh) goto io_error; - grp_alloc_blk = ext3_try_to_allocate_with_rsv(sb, handle, + grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, group_no, bitmap_bh, grp_target_blk, my_rsv, &num, &fatal); if (fatal) @@ -1499,7 +1499,7 @@ retry_alloc: goto allocated; } - ngroups = EXT3_SB(sb)->s_groups_count; + ngroups = EXT4_SB(sb)->s_groups_count; smp_rmb(); /* @@ -1510,7 +1510,7 @@ retry_alloc: group_no++; if (group_no >= ngroups) group_no = 0; - gdp = ext3_get_group_desc(sb, group_no, &gdp_bh); + gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); if (!gdp) { *errp = -EIO; goto out; @@ -1531,7 +1531,7 @@ retry_alloc: /* * try to allocate block(s) from this group, without a goal(-1). */ - grp_alloc_blk = ext3_try_to_allocate_with_rsv(sb, handle, + grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, group_no, bitmap_bh, -1, my_rsv, &num, &fatal); if (fatal) @@ -1557,23 +1557,23 @@ retry_alloc: allocated: - ext3_debug("using block group %d(%d)\n", + ext4_debug("using block group %d(%d)\n", group_no, gdp->bg_free_blocks_count); BUFFER_TRACE(gdp_bh, "get_write_access"); - fatal = ext3_journal_get_write_access(handle, gdp_bh); + fatal = ext4_journal_get_write_access(handle, gdp_bh); if (fatal) goto out; - ret_block = grp_alloc_blk + ext3_group_first_block_no(sb, group_no); + ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no); if (in_range(le32_to_cpu(gdp->bg_block_bitmap), ret_block, num) || in_range(le32_to_cpu(gdp->bg_inode_bitmap), ret_block, num) || in_range(ret_block, le32_to_cpu(gdp->bg_inode_table), - EXT3_SB(sb)->s_itb_per_group) || + EXT4_SB(sb)->s_itb_per_group) || in_range(ret_block + num - 1, le32_to_cpu(gdp->bg_inode_table), - EXT3_SB(sb)->s_itb_per_group)) - ext3_error(sb, "ext3_new_block", + EXT4_SB(sb)->s_itb_per_group)) + ext4_error(sb, "ext4_new_block", "Allocating block in system zone - " "blocks from "E3FSBLK", length %lu", ret_block, num); @@ -1598,20 +1598,20 @@ allocated: int i; for (i = 0; i < num; i++) { - if (ext3_test_bit(grp_alloc_blk+i, + if (ext4_test_bit(grp_alloc_blk+i, bh2jh(bitmap_bh)->b_committed_data)) { printk("%s: block was unexpectedly set in " "b_committed_data\n", __FUNCTION__); } } } - ext3_debug("found bit %d\n", grp_alloc_blk); + ext4_debug("found bit %d\n", grp_alloc_blk); spin_unlock(sb_bgl_lock(sbi, group_no)); jbd_unlock_bh_state(bitmap_bh); #endif if (ret_block + num - 1 >= le32_to_cpu(es->s_blocks_count)) { - ext3_error(sb, "ext3_new_block", + ext4_error(sb, "ext4_new_block", "block("E3FSBLK") >= blocks count(%d) - " "block_group = %d, es == %p ", ret_block, le32_to_cpu(es->s_blocks_count), group_no, es); @@ -1623,7 +1623,7 @@ allocated: * list of some description. We don't know in advance whether * the caller wants to use it as metadata or data. */ - ext3_debug("allocating block %lu. Goal hits %d of %d.\n", + ext4_debug("allocating block %lu. Goal hits %d of %d.\n", ret_block, goal_hits, goal_attempts); spin_lock(sb_bgl_lock(sbi, group_no)); @@ -1633,7 +1633,7 @@ allocated: percpu_counter_mod(&sbi->s_freeblocks_counter, -num); BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor"); - err = ext3_journal_dirty_metadata(handle, gdp_bh); + err = ext4_journal_dirty_metadata(handle, gdp_bh); if (!fatal) fatal = err; @@ -1652,7 +1652,7 @@ io_error: out: if (fatal) { *errp = fatal; - ext3_std_error(sb, fatal); + ext4_std_error(sb, fatal); } /* * Undo the block allocation @@ -1663,40 +1663,40 @@ out: return 0; } -ext3_fsblk_t ext3_new_block(handle_t *handle, struct inode *inode, - ext3_fsblk_t goal, int *errp) +ext4_fsblk_t ext4_new_block(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int *errp) { unsigned long count = 1; - return ext3_new_blocks(handle, inode, goal, &count, errp); + return ext4_new_blocks(handle, inode, goal, &count, errp); } /** - * ext3_count_free_blocks() -- count filesystem free blocks + * ext4_count_free_blocks() -- count filesystem free blocks * @sb: superblock * * Adds up the number of free blocks from each block group. */ -ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb) +ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) { - ext3_fsblk_t desc_count; - struct ext3_group_desc *gdp; + ext4_fsblk_t desc_count; + struct ext4_group_desc *gdp; int i; - unsigned long ngroups = EXT3_SB(sb)->s_groups_count; -#ifdef EXT3FS_DEBUG - struct ext3_super_block *es; - ext3_fsblk_t bitmap_count; + unsigned long ngroups = EXT4_SB(sb)->s_groups_count; +#ifdef EXT4FS_DEBUG + struct ext4_super_block *es; + ext4_fsblk_t bitmap_count; unsigned long x; struct buffer_head *bitmap_bh = NULL; - es = EXT3_SB(sb)->s_es; + es = EXT4_SB(sb)->s_es; desc_count = 0; bitmap_count = 0; gdp = NULL; smp_rmb(); for (i = 0; i < ngroups; i++) { - gdp = ext3_get_group_desc(sb, i, NULL); + gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_blocks_count); @@ -1705,13 +1705,13 @@ ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb) if (bitmap_bh == NULL) continue; - x = ext3_count_free(bitmap_bh, sb->s_blocksize); + x = ext4_count_free(bitmap_bh, sb->s_blocksize); printk("group %d: stored = %d, counted = %lu\n", i, le16_to_cpu(gdp->bg_free_blocks_count), x); bitmap_count += x; } brelse(bitmap_bh); - printk("ext3_count_free_blocks: stored = "E3FSBLK + printk("ext4_count_free_blocks: stored = "E3FSBLK ", computed = "E3FSBLK", "E3FSBLK"\n", le32_to_cpu(es->s_free_blocks_count), desc_count, bitmap_count); @@ -1720,7 +1720,7 @@ ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb) desc_count = 0; smp_rmb(); for (i = 0; i < ngroups; i++) { - gdp = ext3_get_group_desc(sb, i, NULL); + gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_blocks_count); @@ -1731,11 +1731,11 @@ ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb) } static inline int -block_in_use(ext3_fsblk_t block, struct super_block *sb, unsigned char *map) +block_in_use(ext4_fsblk_t block, struct super_block *sb, unsigned char *map) { - return ext3_test_bit ((block - - le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block)) % - EXT3_BLOCKS_PER_GROUP(sb), map); + return ext4_test_bit ((block - + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) % + EXT4_BLOCKS_PER_GROUP(sb), map); } static inline int test_root(int a, int b) @@ -1747,7 +1747,7 @@ static inline int test_root(int a, int b) return num == a; } -static int ext3_group_sparse(int group) +static int ext4_group_sparse(int group) { if (group <= 1) return 1; @@ -1758,44 +1758,44 @@ static int ext3_group_sparse(int group) } /** - * ext3_bg_has_super - number of blocks used by the superblock in group + * ext4_bg_has_super - number of blocks used by the superblock in group * @sb: superblock for filesystem * @group: group number to check * * Return the number of blocks used by the superblock (primary or backup) * in this group. Currently this will be only 0 or 1. */ -int ext3_bg_has_super(struct super_block *sb, int group) +int ext4_bg_has_super(struct super_block *sb, int group) { - if (EXT3_HAS_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER) && - !ext3_group_sparse(group)) + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) && + !ext4_group_sparse(group)) return 0; return 1; } -static unsigned long ext3_bg_num_gdb_meta(struct super_block *sb, int group) +static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb, int group) { - unsigned long metagroup = group / EXT3_DESC_PER_BLOCK(sb); - unsigned long first = metagroup * EXT3_DESC_PER_BLOCK(sb); - unsigned long last = first + EXT3_DESC_PER_BLOCK(sb) - 1; + unsigned long metagroup = group / EXT4_DESC_PER_BLOCK(sb); + unsigned long first = metagroup * EXT4_DESC_PER_BLOCK(sb); + unsigned long last = first + EXT4_DESC_PER_BLOCK(sb) - 1; if (group == first || group == first + 1 || group == last) return 1; return 0; } -static unsigned long ext3_bg_num_gdb_nometa(struct super_block *sb, int group) +static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, int group) { - if (EXT3_HAS_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER) && - !ext3_group_sparse(group)) + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) && + !ext4_group_sparse(group)) return 0; - return EXT3_SB(sb)->s_gdb_count; + return EXT4_SB(sb)->s_gdb_count; } /** - * ext3_bg_num_gdb - number of blocks used by the group table in group + * ext4_bg_num_gdb - number of blocks used by the group table in group * @sb: superblock for filesystem * @group: group number to check * @@ -1803,16 +1803,16 @@ static unsigned long ext3_bg_num_gdb_nometa(struct super_block *sb, int group) * (primary or backup) in this group. In the future there may be a * different number of descriptor blocks in each group. */ -unsigned long ext3_bg_num_gdb(struct super_block *sb, int group) +unsigned long ext4_bg_num_gdb(struct super_block *sb, int group) { unsigned long first_meta_bg = - le32_to_cpu(EXT3_SB(sb)->s_es->s_first_meta_bg); - unsigned long metagroup = group / EXT3_DESC_PER_BLOCK(sb); + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_meta_bg); + unsigned long metagroup = group / EXT4_DESC_PER_BLOCK(sb); - if (!EXT3_HAS_INCOMPAT_FEATURE(sb,EXT3_FEATURE_INCOMPAT_META_BG) || + if (!EXT4_HAS_INCOMPAT_FEATURE(sb,EXT4_FEATURE_INCOMPAT_META_BG) || metagroup < first_meta_bg) - return ext3_bg_num_gdb_nometa(sb,group); + return ext4_bg_num_gdb_nometa(sb,group); - return ext3_bg_num_gdb_meta(sb,group); + return ext4_bg_num_gdb_meta(sb,group); } diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index b9176ee..f4b3570 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/bitmap.c + * linux/fs/ext4/bitmap.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -9,13 +9,13 @@ #include #include -#include +#include -#ifdef EXT3FS_DEBUG +#ifdef EXT4FS_DEBUG static int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; -unsigned long ext3_count_free (struct buffer_head * map, unsigned int numchars) +unsigned long ext4_count_free (struct buffer_head * map, unsigned int numchars) { unsigned int i; unsigned long sum = 0; @@ -28,5 +28,5 @@ unsigned long ext3_count_free (struct buffer_head * map, unsigned int numchars) return (sum); } -#endif /* EXT3FS_DEBUG */ +#endif /* EXT4FS_DEBUG */ diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index d0b54f3..ec114d7 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/dir.c + * linux/fs/ext4/dir.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -12,7 +12,7 @@ * * Copyright (C) 1991, 1992 Linus Torvalds * - * ext3 directory handling functions + * ext4 directory handling functions * * Big-endian to little-endian byte-swapping/bitmaps by * David S. Miller (davem@caip.rutgers.edu), 1995 @@ -23,69 +23,69 @@ #include #include -#include +#include #include #include #include #include -static unsigned char ext3_filetype_table[] = { +static unsigned char ext4_filetype_table[] = { DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK }; -static int ext3_readdir(struct file *, void *, filldir_t); -static int ext3_dx_readdir(struct file * filp, +static int ext4_readdir(struct file *, void *, filldir_t); +static int ext4_dx_readdir(struct file * filp, void * dirent, filldir_t filldir); -static int ext3_release_dir (struct inode * inode, +static int ext4_release_dir (struct inode * inode, struct file * filp); -const struct file_operations ext3_dir_operations = { +const struct file_operations ext4_dir_operations = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ext3_readdir, /* we take BKL. needed?*/ - .ioctl = ext3_ioctl, /* BKL held */ + .readdir = ext4_readdir, /* we take BKL. needed?*/ + .ioctl = ext4_ioctl, /* BKL held */ #ifdef CONFIG_COMPAT - .compat_ioctl = ext3_compat_ioctl, + .compat_ioctl = ext4_compat_ioctl, #endif - .fsync = ext3_sync_file, /* BKL held */ -#ifdef CONFIG_EXT3_INDEX - .release = ext3_release_dir, + .fsync = ext4_sync_file, /* BKL held */ +#ifdef CONFIG_EXT4_INDEX + .release = ext4_release_dir, #endif }; static unsigned char get_dtype(struct super_block *sb, int filetype) { - if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_FILETYPE) || - (filetype >= EXT3_FT_MAX)) + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FILETYPE) || + (filetype >= EXT4_FT_MAX)) return DT_UNKNOWN; - return (ext3_filetype_table[filetype]); + return (ext4_filetype_table[filetype]); } -int ext3_check_dir_entry (const char * function, struct inode * dir, - struct ext3_dir_entry_2 * de, +int ext4_check_dir_entry (const char * function, struct inode * dir, + struct ext4_dir_entry_2 * de, struct buffer_head * bh, unsigned long offset) { const char * error_msg = NULL; const int rlen = le16_to_cpu(de->rec_len); - if (rlen < EXT3_DIR_REC_LEN(1)) + if (rlen < EXT4_DIR_REC_LEN(1)) error_msg = "rec_len is smaller than minimal"; else if (rlen % 4 != 0) error_msg = "rec_len % 4 != 0"; - else if (rlen < EXT3_DIR_REC_LEN(de->name_len)) + else if (rlen < EXT4_DIR_REC_LEN(de->name_len)) error_msg = "rec_len is too small for name_len"; else if (((char *) de - bh->b_data) + rlen > dir->i_sb->s_blocksize) error_msg = "directory entry across blocks"; else if (le32_to_cpu(de->inode) > - le32_to_cpu(EXT3_SB(dir->i_sb)->s_es->s_inodes_count)) + le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count)) error_msg = "inode out of bounds"; if (error_msg != NULL) - ext3_error (dir->i_sb, function, + ext4_error (dir->i_sb, function, "bad entry in directory #%lu: %s - " "offset=%lu, inode=%lu, rec_len=%d, name_len=%d", dir->i_ino, error_msg, offset, @@ -94,13 +94,13 @@ int ext3_check_dir_entry (const char * function, struct inode * dir, return error_msg == NULL ? 1 : 0; } -static int ext3_readdir(struct file * filp, +static int ext4_readdir(struct file * filp, void * dirent, filldir_t filldir) { int error = 0; unsigned long offset; int i, stored; - struct ext3_dir_entry_2 *de; + struct ext4_dir_entry_2 *de; struct super_block *sb; int err; struct inode *inode = filp->f_dentry->d_inode; @@ -108,12 +108,12 @@ static int ext3_readdir(struct file * filp, sb = inode->i_sb; -#ifdef CONFIG_EXT3_INDEX - if (EXT3_HAS_COMPAT_FEATURE(inode->i_sb, - EXT3_FEATURE_COMPAT_DIR_INDEX) && - ((EXT3_I(inode)->i_flags & EXT3_INDEX_FL) || +#ifdef CONFIG_EXT4_INDEX + if (EXT4_HAS_COMPAT_FEATURE(inode->i_sb, + EXT4_FEATURE_COMPAT_DIR_INDEX) && + ((EXT4_I(inode)->i_flags & EXT4_INDEX_FL) || ((inode->i_size >> sb->s_blocksize_bits) == 1))) { - err = ext3_dx_readdir(filp, dirent, filldir); + err = ext4_dx_readdir(filp, dirent, filldir); if (err != ERR_BAD_DX_DIR) { ret = err; goto out; @@ -122,19 +122,19 @@ static int ext3_readdir(struct file * filp, * We don't set the inode dirty flag since it's not * critical that it get flushed back to the disk. */ - EXT3_I(filp->f_dentry->d_inode)->i_flags &= ~EXT3_INDEX_FL; + EXT4_I(filp->f_dentry->d_inode)->i_flags &= ~EXT4_INDEX_FL; } #endif stored = 0; offset = filp->f_pos & (sb->s_blocksize - 1); while (!error && !stored && filp->f_pos < inode->i_size) { - unsigned long blk = filp->f_pos >> EXT3_BLOCK_SIZE_BITS(sb); + unsigned long blk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb); struct buffer_head map_bh; struct buffer_head *bh = NULL; map_bh.b_state = 0; - err = ext3_get_blocks_handle(NULL, inode, blk, 1, + err = ext4_get_blocks_handle(NULL, inode, blk, 1, &map_bh, 0, 0); if (err > 0) { page_cache_readahead(sb->s_bdev->bd_inode->i_mapping, @@ -143,7 +143,7 @@ static int ext3_readdir(struct file * filp, map_bh.b_blocknr >> (PAGE_CACHE_SHIFT - inode->i_blkbits), 1); - bh = ext3_bread(NULL, inode, blk, 0, &err); + bh = ext4_bread(NULL, inode, blk, 0, &err); } /* @@ -151,7 +151,7 @@ static int ext3_readdir(struct file * filp, * of recovering data when there's a bad sector */ if (!bh) { - ext3_error (sb, "ext3_readdir", + ext4_error (sb, "ext4_readdir", "directory #%lu contains a hole at offset %lu", inode->i_ino, (unsigned long)filp->f_pos); filp->f_pos += sb->s_blocksize - offset; @@ -165,7 +165,7 @@ revalidate: * to make sure. */ if (filp->f_version != inode->i_version) { for (i = 0; i < sb->s_blocksize && i < offset; ) { - de = (struct ext3_dir_entry_2 *) + de = (struct ext4_dir_entry_2 *) (bh->b_data + i); /* It's too expensive to do a full * dirent test each time round this @@ -174,7 +174,7 @@ revalidate: * failure will be detected in the * dirent test below. */ if (le16_to_cpu(de->rec_len) < - EXT3_DIR_REC_LEN(1)) + EXT4_DIR_REC_LEN(1)) break; i += le16_to_cpu(de->rec_len); } @@ -186,8 +186,8 @@ revalidate: while (!error && filp->f_pos < inode->i_size && offset < sb->s_blocksize) { - de = (struct ext3_dir_entry_2 *) (bh->b_data + offset); - if (!ext3_check_dir_entry ("ext3_readdir", inode, de, + de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); + if (!ext4_check_dir_entry ("ext4_readdir", inode, de, bh, offset)) { /* On error, skip the f_pos to the next block. */ @@ -228,7 +228,7 @@ out: return ret; } -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX /* * These functions convert from the major/minor hash to an f_pos * value. @@ -323,7 +323,7 @@ static struct dir_private_info *create_dir_info(loff_t pos) return p; } -void ext3_htree_free_dir_info(struct dir_private_info *p) +void ext4_htree_free_dir_info(struct dir_private_info *p) { free_rb_tree_fname(&p->root); kfree(p); @@ -332,9 +332,9 @@ void ext3_htree_free_dir_info(struct dir_private_info *p) /* * Given a directory entry, enter it into the fname rb tree. */ -int ext3_htree_store_dirent(struct file *dir_file, __u32 hash, +int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, __u32 minor_hash, - struct ext3_dir_entry_2 *dirent) + struct ext4_dir_entry_2 *dirent) { struct rb_node **p, *parent = NULL; struct fname * fname, *new_fn; @@ -390,7 +390,7 @@ int ext3_htree_store_dirent(struct file *dir_file, __u32 hash, /* - * This is a helper function for ext3_dx_readdir. It calls filldir + * This is a helper function for ext4_dx_readdir. It calls filldir * for all entres on the fname linked list. (Normally there is only * one entry on the linked list, unless there are 62 bit hash collisions.) */ @@ -425,7 +425,7 @@ static int call_filldir(struct file * filp, void * dirent, return 0; } -static int ext3_dx_readdir(struct file * filp, +static int ext4_dx_readdir(struct file * filp, void * dirent, filldir_t filldir) { struct dir_private_info *info = filp->private_data; @@ -440,7 +440,7 @@ static int ext3_dx_readdir(struct file * filp, filp->private_data = info; } - if (filp->f_pos == EXT3_HTREE_EOF) + if (filp->f_pos == EXT4_HTREE_EOF) return 0; /* EOF */ /* Some one has messed with f_pos; reset the world */ @@ -474,13 +474,13 @@ static int ext3_dx_readdir(struct file * filp, info->curr_node = NULL; free_rb_tree_fname(&info->root); filp->f_version = inode->i_version; - ret = ext3_htree_fill_tree(filp, info->curr_hash, + ret = ext4_htree_fill_tree(filp, info->curr_hash, info->curr_minor_hash, &info->next_hash); if (ret < 0) return ret; if (ret == 0) { - filp->f_pos = EXT3_HTREE_EOF; + filp->f_pos = EXT4_HTREE_EOF; break; } info->curr_node = rb_first(&info->root); @@ -495,7 +495,7 @@ static int ext3_dx_readdir(struct file * filp, info->curr_node = rb_next(info->curr_node); if (!info->curr_node) { if (info->next_hash == ~0) { - filp->f_pos = EXT3_HTREE_EOF; + filp->f_pos = EXT4_HTREE_EOF; break; } info->curr_hash = info->next_hash; @@ -507,10 +507,10 @@ finished: return 0; } -static int ext3_release_dir (struct inode * inode, struct file * filp) +static int ext4_release_dir (struct inode * inode, struct file * filp) { if (filp->private_data) - ext3_htree_free_dir_info(filp->private_data); + ext4_htree_free_dir_info(filp->private_data); return 0; } diff --git a/fs/ext4/file.c b/fs/ext4/file.c index e96c388..d938fbe 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/file.c + * linux/fs/ext4/file.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -12,7 +12,7 @@ * * Copyright (C) 1991, 1992 Linus Torvalds * - * ext3 fs regular file handling primitives + * ext4 fs regular file handling primitives * * 64-bit file support on 64-bit platforms by Jakub Jelinek * (jj@sunsite.ms.mff.cuni.cz) @@ -21,34 +21,34 @@ #include #include #include -#include -#include +#include +#include #include "xattr.h" #include "acl.h" /* * Called when an inode is released. Note that this is different - * from ext3_file_open: open gets called at every open, but release + * from ext4_file_open: open gets called at every open, but release * gets called only when /all/ the files are closed. */ -static int ext3_release_file (struct inode * inode, struct file * filp) +static int ext4_release_file (struct inode * inode, struct file * filp) { /* if we are the last writer on the inode, drop the block reservation */ if ((filp->f_mode & FMODE_WRITE) && (atomic_read(&inode->i_writecount) == 1)) { - mutex_lock(&EXT3_I(inode)->truncate_mutex); - ext3_discard_reservation(inode); - mutex_unlock(&EXT3_I(inode)->truncate_mutex); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + ext4_discard_reservation(inode); + mutex_unlock(&EXT4_I(inode)->truncate_mutex); } if (is_dx(inode) && filp->private_data) - ext3_htree_free_dir_info(filp->private_data); + ext4_htree_free_dir_info(filp->private_data); return 0; } static ssize_t -ext3_file_write(struct kiocb *iocb, const struct iovec *iov, +ext4_file_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { struct file *file = iocb->ki_filp; @@ -79,7 +79,7 @@ ext3_file_write(struct kiocb *iocb, const struct iovec *iov, * Open question --- do we care about flushing timestamps too * if the inode is IS_SYNC? */ - if (!ext3_should_journal_data(inode)) + if (!ext4_should_journal_data(inode)) return ret; goto force_commit; @@ -100,40 +100,40 @@ ext3_file_write(struct kiocb *iocb, const struct iovec *iov, */ force_commit: - err = ext3_force_commit(inode->i_sb); + err = ext4_force_commit(inode->i_sb); if (err) return err; return ret; } -const struct file_operations ext3_file_operations = { +const struct file_operations ext4_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, .write = do_sync_write, .aio_read = generic_file_aio_read, - .aio_write = ext3_file_write, - .ioctl = ext3_ioctl, + .aio_write = ext4_file_write, + .ioctl = ext4_ioctl, #ifdef CONFIG_COMPAT - .compat_ioctl = ext3_compat_ioctl, + .compat_ioctl = ext4_compat_ioctl, #endif .mmap = generic_file_mmap, .open = generic_file_open, - .release = ext3_release_file, - .fsync = ext3_sync_file, + .release = ext4_release_file, + .fsync = ext4_sync_file, .sendfile = generic_file_sendfile, .splice_read = generic_file_splice_read, .splice_write = generic_file_splice_write, }; -struct inode_operations ext3_file_inode_operations = { - .truncate = ext3_truncate, - .setattr = ext3_setattr, -#ifdef CONFIG_EXT3_FS_XATTR +struct inode_operations ext4_file_inode_operations = { + .truncate = ext4_truncate, + .setattr = ext4_setattr, +#ifdef CONFIG_EXT4DEV_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, - .listxattr = ext3_listxattr, + .listxattr = ext4_listxattr, .removexattr = generic_removexattr, #endif - .permission = ext3_permission, + .permission = ext4_permission, }; diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index dd1fd3c..272faa2 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/fsync.c + * linux/fs/ext4/fsync.c * * Copyright (C) 1993 Stephen Tweedie (sct@redhat.com) * from @@ -9,7 +9,7 @@ * from * linux/fs/minix/truncate.c Copyright (C) 1991, 1992 Linus Torvalds * - * ext3fs fsync primitive + * ext4fs fsync primitive * * Big-endian to little-endian byte-swapping/bitmaps by * David S. Miller (davem@caip.rutgers.edu), 1995 @@ -27,11 +27,11 @@ #include #include #include -#include -#include +#include +#include /* - * akpm: A new design for ext3_sync_file(). + * akpm: A new design for ext4_sync_file(). * * This is only called from sys_fsync(), sys_fdatasync() and sys_msync(). * There cannot be a transaction open by this task. @@ -42,12 +42,12 @@ * inode to disk. */ -int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync) +int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) { struct inode *inode = dentry->d_inode; int ret = 0; - J_ASSERT(ext3_journal_current_handle() == 0); + J_ASSERT(ext4_journal_current_handle() == 0); /* * data=writeback: @@ -61,14 +61,14 @@ int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync) * * data=journal: * filemap_fdatawrite won't do anything (the buffers are clean). - * ext3_force_commit will write the file data into the journal and + * ext4_force_commit will write the file data into the journal and * will wait on that. * filemap_fdatawait() will encounter a ton of newly-dirtied pages * (they were dirtied by commit). But that's OK - the blocks are * safe in-journal, which is all fsync() needs to ensure. */ - if (ext3_should_journal_data(inode)) { - ret = ext3_force_commit(inode->i_sb); + if (ext4_should_journal_data(inode)) { + ret = ext4_force_commit(inode->i_sb); goto out; } diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c index deeb27b..d15bb42 100644 --- a/fs/ext4/hash.c +++ b/fs/ext4/hash.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/hash.c + * linux/fs/ext4/hash.c * * Copyright (C) 2002 by Theodore Ts'o * @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #define DELTA 0x9E3779B9 @@ -89,7 +89,7 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) * represented, and whether or not the returned hash is 32 bits or 64 * bits. 32 bit hashes will return 0 for the minor hash. */ -int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) +int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) { __u32 hash; __u32 minor_hash = 0; @@ -144,8 +144,8 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) return -1; } hash = hash & ~1; - if (hash == (EXT3_HTREE_EOF << 1)) - hash = (EXT3_HTREE_EOF-1) << 1; + if (hash == (EXT4_HTREE_EOF << 1)) + hash = (EXT4_HTREE_EOF-1) << 1; hinfo->hash = hash; hinfo->minor_hash = minor_hash; return 0; diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index e45dbd6..4b92066 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/ialloc.c + * linux/fs/ext4/ialloc.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -15,8 +15,8 @@ #include #include #include -#include -#include +#include +#include #include #include #include @@ -53,16 +53,16 @@ static struct buffer_head * read_inode_bitmap(struct super_block * sb, unsigned long block_group) { - struct ext3_group_desc *desc; + struct ext4_group_desc *desc; struct buffer_head *bh = NULL; - desc = ext3_get_group_desc(sb, block_group, NULL); + desc = ext4_get_group_desc(sb, block_group, NULL); if (!desc) goto error_out; bh = sb_bread(sb, le32_to_cpu(desc->bg_inode_bitmap)); if (!bh) - ext3_error(sb, "read_inode_bitmap", + ext4_error(sb, "read_inode_bitmap", "Cannot read inode bitmap - " "block_group = %lu, inode_bitmap = %u", block_group, le32_to_cpu(desc->bg_inode_bitmap)); @@ -86,7 +86,7 @@ error_out: * though), and then we'd have two inodes sharing the * same inode number and space on the harddisk. */ -void ext3_free_inode (handle_t *handle, struct inode * inode) +void ext4_free_inode (handle_t *handle, struct inode * inode) { struct super_block * sb = inode->i_sb; int is_directory; @@ -95,36 +95,36 @@ void ext3_free_inode (handle_t *handle, struct inode * inode) struct buffer_head *bh2; unsigned long block_group; unsigned long bit; - struct ext3_group_desc * gdp; - struct ext3_super_block * es; - struct ext3_sb_info *sbi; + struct ext4_group_desc * gdp; + struct ext4_super_block * es; + struct ext4_sb_info *sbi; int fatal = 0, err; if (atomic_read(&inode->i_count) > 1) { - printk ("ext3_free_inode: inode has count=%d\n", + printk ("ext4_free_inode: inode has count=%d\n", atomic_read(&inode->i_count)); return; } if (inode->i_nlink) { - printk ("ext3_free_inode: inode has nlink=%d\n", + printk ("ext4_free_inode: inode has nlink=%d\n", inode->i_nlink); return; } if (!sb) { - printk("ext3_free_inode: inode on nonexistent device\n"); + printk("ext4_free_inode: inode on nonexistent device\n"); return; } - sbi = EXT3_SB(sb); + sbi = EXT4_SB(sb); ino = inode->i_ino; - ext3_debug ("freeing inode %lu\n", ino); + ext4_debug ("freeing inode %lu\n", ino); /* * Note: we must free any quota before locking the superblock, * as writing the quota to disk may need the lock as well. */ DQUOT_INIT(inode); - ext3_xattr_delete_inode(handle, inode); + ext4_xattr_delete_inode(handle, inode); DQUOT_FREE_INODE(inode); DQUOT_DROP(inode); @@ -133,33 +133,33 @@ void ext3_free_inode (handle_t *handle, struct inode * inode) /* Do this BEFORE marking the inode not in use or returning an error */ clear_inode (inode); - es = EXT3_SB(sb)->s_es; - if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { - ext3_error (sb, "ext3_free_inode", + es = EXT4_SB(sb)->s_es; + if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { + ext4_error (sb, "ext4_free_inode", "reserved or nonexistent inode %lu", ino); goto error_return; } - block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); - bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb); + block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); + bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); bitmap_bh = read_inode_bitmap(sb, block_group); if (!bitmap_bh) goto error_return; BUFFER_TRACE(bitmap_bh, "get_write_access"); - fatal = ext3_journal_get_write_access(handle, bitmap_bh); + fatal = ext4_journal_get_write_access(handle, bitmap_bh); if (fatal) goto error_return; /* Ok, now we can actually update the inode bitmaps.. */ - if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group), + if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit, bitmap_bh->b_data)) - ext3_error (sb, "ext3_free_inode", + ext4_error (sb, "ext4_free_inode", "bit already cleared for inode %lu", ino); else { - gdp = ext3_get_group_desc (sb, block_group, &bh2); + gdp = ext4_get_group_desc (sb, block_group, &bh2); BUFFER_TRACE(bh2, "get_write_access"); - fatal = ext3_journal_get_write_access(handle, bh2); + fatal = ext4_journal_get_write_access(handle, bh2); if (fatal) goto error_return; if (gdp) { @@ -175,18 +175,18 @@ void ext3_free_inode (handle_t *handle, struct inode * inode) percpu_counter_dec(&sbi->s_dirs_counter); } - BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bh2); + BUFFER_TRACE(bh2, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh2); if (!fatal) fatal = err; } - BUFFER_TRACE(bitmap_bh, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bitmap_bh); + BUFFER_TRACE(bitmap_bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); if (!fatal) fatal = err; sb->s_dirt = 1; error_return: brelse(bitmap_bh); - ext3_std_error(sb, fatal); + ext4_std_error(sb, fatal); } /* @@ -201,17 +201,17 @@ error_return: */ static int find_group_dir(struct super_block *sb, struct inode *parent) { - int ngroups = EXT3_SB(sb)->s_groups_count; + int ngroups = EXT4_SB(sb)->s_groups_count; unsigned int freei, avefreei; - struct ext3_group_desc *desc, *best_desc = NULL; + struct ext4_group_desc *desc, *best_desc = NULL; struct buffer_head *bh; int group, best_group = -1; - freei = percpu_counter_read_positive(&EXT3_SB(sb)->s_freeinodes_counter); + freei = percpu_counter_read_positive(&EXT4_SB(sb)->s_freeinodes_counter); avefreei = freei / ngroups; for (group = 0; group < ngroups; group++) { - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) @@ -256,19 +256,19 @@ static int find_group_dir(struct super_block *sb, struct inode *parent) static int find_group_orlov(struct super_block *sb, struct inode *parent) { - int parent_group = EXT3_I(parent)->i_block_group; - struct ext3_sb_info *sbi = EXT3_SB(sb); - struct ext3_super_block *es = sbi->s_es; + int parent_group = EXT4_I(parent)->i_block_group; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; int ngroups = sbi->s_groups_count; - int inodes_per_group = EXT3_INODES_PER_GROUP(sb); + int inodes_per_group = EXT4_INODES_PER_GROUP(sb); unsigned int freei, avefreei; - ext3_fsblk_t freeb, avefreeb; - ext3_fsblk_t blocks_per_dir; + ext4_fsblk_t freeb, avefreeb; + ext4_fsblk_t blocks_per_dir; unsigned int ndirs; int max_debt, max_dirs, min_inodes; - ext3_grpblk_t min_blocks; + ext4_grpblk_t min_blocks; int group = -1, i; - struct ext3_group_desc *desc; + struct ext4_group_desc *desc; struct buffer_head *bh; freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter); @@ -278,7 +278,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter); if ((parent == sb->s_root->d_inode) || - (EXT3_I(parent)->i_flags & EXT3_TOPDIR_FL)) { + (EXT4_I(parent)->i_flags & EXT4_TOPDIR_FL)) { int best_ndir = inodes_per_group; int best_group = -1; @@ -286,7 +286,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) parent_group = (unsigned)group % ngroups; for (i = 0; i < ngroups; i++) { group = (parent_group + i) % ngroups; - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir) @@ -307,9 +307,9 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) max_dirs = ndirs / ngroups + inodes_per_group / 16; min_inodes = avefreei - inodes_per_group / 4; - min_blocks = avefreeb - EXT3_BLOCKS_PER_GROUP(sb) / 4; + min_blocks = avefreeb - EXT4_BLOCKS_PER_GROUP(sb) / 4; - max_debt = EXT3_BLOCKS_PER_GROUP(sb) / max(blocks_per_dir, (ext3_fsblk_t)BLOCK_COST); + max_debt = EXT4_BLOCKS_PER_GROUP(sb) / max(blocks_per_dir, (ext4_fsblk_t)BLOCK_COST); if (max_debt * INODE_COST > inodes_per_group) max_debt = inodes_per_group / INODE_COST; if (max_debt > 255) @@ -319,7 +319,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) for (i = 0; i < ngroups; i++) { group = (parent_group + i) % ngroups; - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs) @@ -334,7 +334,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) fallback: for (i = 0; i < ngroups; i++) { group = (parent_group + i) % ngroups; - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei) @@ -355,9 +355,9 @@ fallback: static int find_group_other(struct super_block *sb, struct inode *parent) { - int parent_group = EXT3_I(parent)->i_block_group; - int ngroups = EXT3_SB(sb)->s_groups_count; - struct ext3_group_desc *desc; + int parent_group = EXT4_I(parent)->i_block_group; + int ngroups = EXT4_SB(sb)->s_groups_count; + struct ext4_group_desc *desc; struct buffer_head *bh; int group, i; @@ -365,7 +365,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent) * Try to place the inode in its parent directory */ group = parent_group; - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (desc && le16_to_cpu(desc->bg_free_inodes_count) && le16_to_cpu(desc->bg_free_blocks_count)) return group; @@ -389,7 +389,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent) group += i; if (group >= ngroups) group -= ngroups; - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (desc && le16_to_cpu(desc->bg_free_inodes_count) && le16_to_cpu(desc->bg_free_blocks_count)) return group; @@ -403,7 +403,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent) for (i = 0; i < ngroups; i++) { if (++group >= ngroups) group = 0; - desc = ext3_get_group_desc (sb, group, &bh); + desc = ext4_get_group_desc (sb, group, &bh); if (desc && le16_to_cpu(desc->bg_free_inodes_count)) return group; } @@ -421,7 +421,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent) * For other inodes, search forward from the parent directory's block * group to find a free inode. */ -struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) +struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) { struct super_block *sb; struct buffer_head *bitmap_bh = NULL; @@ -429,10 +429,10 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) int group; unsigned long ino = 0; struct inode * inode; - struct ext3_group_desc * gdp = NULL; - struct ext3_super_block * es; - struct ext3_inode_info *ei; - struct ext3_sb_info *sbi; + struct ext4_group_desc * gdp = NULL; + struct ext4_super_block * es; + struct ext4_inode_info *ei; + struct ext4_sb_info *sbi; int err = 0; struct inode *ret; int i; @@ -445,9 +445,9 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) inode = new_inode(sb); if (!inode) return ERR_PTR(-ENOMEM); - ei = EXT3_I(inode); + ei = EXT4_I(inode); - sbi = EXT3_SB(sb); + sbi = EXT4_SB(sb); es = sbi->s_es; if (S_ISDIR(mode)) { if (test_opt (sb, OLDALLOC)) @@ -464,7 +464,7 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) for (i = 0; i < sbi->s_groups_count; i++) { err = -EIO; - gdp = ext3_get_group_desc(sb, group, &bh2); + gdp = ext4_get_group_desc(sb, group, &bh2); if (!gdp) goto fail; @@ -476,21 +476,21 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode) ino = 0; repeat_in_this_group: - ino = ext3_find_next_zero_bit((unsigned long *) - bitmap_bh->b_data, EXT3_INODES_PER_GROUP(sb), ino); - if (ino < EXT3_INODES_PER_GROUP(sb)) { + ino = ext4_find_next_zero_bit((unsigned long *) + bitmap_bh->b_data, EXT4_INODES_PER_GROUP(sb), ino); + if (ino < EXT4_INODES_PER_GROUP(sb)) { BUFFER_TRACE(bitmap_bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, bitmap_bh); + err = ext4_journal_get_write_access(handle, bitmap_bh); if (err) goto fail; - if (!ext3_set_bit_atomic(sb_bgl_lock(sbi, group), + if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), ino, bitmap_bh->b_data)) { /* we won it */ BUFFER_TRACE(bitmap_bh, - "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, + "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); if (err) goto fail; @@ -499,7 +499,7 @@ repeat_in_this_group: /* we lost it */ journal_release_buffer(handle, bitmap_bh); - if (++ino < EXT3_INODES_PER_GROUP(sb)) + if (++ino < EXT4_INODES_PER_GROUP(sb)) goto repeat_in_this_group; } @@ -517,9 +517,9 @@ repeat_in_this_group: goto out; got: - ino += group * EXT3_INODES_PER_GROUP(sb) + 1; - if (ino < EXT3_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { - ext3_error (sb, "ext3_new_inode", + ino += group * EXT4_INODES_PER_GROUP(sb) + 1; + if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { + ext4_error (sb, "ext4_new_inode", "reserved inode or inode > inodes count - " "block_group = %d, inode=%lu", group, ino); err = -EIO; @@ -527,7 +527,7 @@ got: } BUFFER_TRACE(bh2, "get_write_access"); - err = ext3_journal_get_write_access(handle, bh2); + err = ext4_journal_get_write_access(handle, bh2); if (err) goto fail; spin_lock(sb_bgl_lock(sbi, group)); gdp->bg_free_inodes_count = @@ -537,8 +537,8 @@ got: cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); } spin_unlock(sb_bgl_lock(sbi, group)); - BUFFER_TRACE(bh2, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bh2); + BUFFER_TRACE(bh2, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh2); if (err) goto fail; percpu_counter_dec(&sbi->s_freeinodes_counter); @@ -566,13 +566,13 @@ got: ei->i_dir_start_lookup = 0; ei->i_disksize = 0; - ei->i_flags = EXT3_I(dir)->i_flags & ~EXT3_INDEX_FL; + ei->i_flags = EXT4_I(dir)->i_flags & ~EXT4_INDEX_FL; if (S_ISLNK(mode)) - ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL); + ei->i_flags &= ~(EXT4_IMMUTABLE_FL|EXT4_APPEND_FL); /* dirsync only applies to directories */ if (!S_ISDIR(mode)) - ei->i_flags &= ~EXT3_DIRSYNC_FL; -#ifdef EXT3_FRAGMENTS + ei->i_flags &= ~EXT4_DIRSYNC_FL; +#ifdef EXT4_FRAGMENTS ei->i_faddr = 0; ei->i_frag_no = 0; ei->i_frag_size = 0; @@ -583,7 +583,7 @@ got: ei->i_block_alloc_info = NULL; ei->i_block_group = group; - ext3_set_inode_flags(inode); + ext4_set_inode_flags(inode); if (IS_DIRSYNC(inode)) handle->h_sync = 1; insert_inode_hash(inode); @@ -591,10 +591,10 @@ got: inode->i_generation = sbi->s_next_generation++; spin_unlock(&sbi->s_next_gen_lock); - ei->i_state = EXT3_STATE_NEW; + ei->i_state = EXT4_STATE_NEW; ei->i_extra_isize = - (EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) ? - sizeof(struct ext3_inode) - EXT3_GOOD_OLD_INODE_SIZE : 0; + (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) ? + sizeof(struct ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE : 0; ret = inode; if(DQUOT_ALLOC_INODE(inode)) { @@ -602,24 +602,24 @@ got: goto fail_drop; } - err = ext3_init_acl(handle, inode, dir); + err = ext4_init_acl(handle, inode, dir); if (err) goto fail_free_drop; - err = ext3_init_security(handle,inode, dir); + err = ext4_init_security(handle,inode, dir); if (err) goto fail_free_drop; - err = ext3_mark_inode_dirty(handle, inode); + err = ext4_mark_inode_dirty(handle, inode); if (err) { - ext3_std_error(sb, err); + ext4_std_error(sb, err); goto fail_free_drop; } - ext3_debug("allocating inode %lu\n", inode->i_ino); + ext4_debug("allocating inode %lu\n", inode->i_ino); goto really_out; fail: - ext3_std_error(sb, err); + ext4_std_error(sb, err); out: iput(inode); ret = ERR_PTR(err); @@ -640,9 +640,9 @@ fail_drop: } /* Verify that we are loading a valid orphan from disk */ -struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) +struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) { - unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count); + unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count); unsigned long block_group; int bit; struct buffer_head *bitmap_bh = NULL; @@ -650,16 +650,16 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) /* Error cases - e2fsck has already cleaned up for us */ if (ino > max_ino) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "bad orphan ino %lu! e2fsck was run?", ino); goto out; } - block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); - bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb); + block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); + bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); bitmap_bh = read_inode_bitmap(sb, block_group); if (!bitmap_bh) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "inode bitmap error for orphan %lu", ino); goto out; } @@ -668,14 +668,14 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) * is a valid orphan (no e2fsck run on fs). Orphans also include * inodes that were being truncated, so we can't check i_nlink==0. */ - if (!ext3_test_bit(bit, bitmap_bh->b_data) || + if (!ext4_test_bit(bit, bitmap_bh->b_data) || !(inode = iget(sb, ino)) || is_bad_inode(inode) || NEXT_ORPHAN(inode) > max_ino) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "bad orphan inode %lu! e2fsck was run?", ino); - printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n", + printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n", bit, (unsigned long long)bitmap_bh->b_blocknr, - ext3_test_bit(bit, bitmap_bh->b_data)); + ext4_test_bit(bit, bitmap_bh->b_data)); printk(KERN_NOTICE "inode=%p\n", inode); if (inode) { printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", @@ -695,22 +695,22 @@ out: return inode; } -unsigned long ext3_count_free_inodes (struct super_block * sb) +unsigned long ext4_count_free_inodes (struct super_block * sb) { unsigned long desc_count; - struct ext3_group_desc *gdp; + struct ext4_group_desc *gdp; int i; -#ifdef EXT3FS_DEBUG - struct ext3_super_block *es; +#ifdef EXT4FS_DEBUG + struct ext4_super_block *es; unsigned long bitmap_count, x; struct buffer_head *bitmap_bh = NULL; - es = EXT3_SB(sb)->s_es; + es = EXT4_SB(sb)->s_es; desc_count = 0; bitmap_count = 0; gdp = NULL; - for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { - gdp = ext3_get_group_desc (sb, i, NULL); + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + gdp = ext4_get_group_desc (sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_inodes_count); @@ -719,19 +719,19 @@ unsigned long ext3_count_free_inodes (struct super_block * sb) if (!bitmap_bh) continue; - x = ext3_count_free(bitmap_bh, EXT3_INODES_PER_GROUP(sb) / 8); + x = ext4_count_free(bitmap_bh, EXT4_INODES_PER_GROUP(sb) / 8); printk("group %d: stored = %d, counted = %lu\n", i, le16_to_cpu(gdp->bg_free_inodes_count), x); bitmap_count += x; } brelse(bitmap_bh); - printk("ext3_count_free_inodes: stored = %u, computed = %lu, %lu\n", + printk("ext4_count_free_inodes: stored = %u, computed = %lu, %lu\n", le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); return desc_count; #else desc_count = 0; - for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { - gdp = ext3_get_group_desc (sb, i, NULL); + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + gdp = ext4_get_group_desc (sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_inodes_count); @@ -742,13 +742,13 @@ unsigned long ext3_count_free_inodes (struct super_block * sb) } /* Called at mount-time, super-block is locked */ -unsigned long ext3_count_dirs (struct super_block * sb) +unsigned long ext4_count_dirs (struct super_block * sb) { unsigned long count = 0; int i; - for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { - struct ext3_group_desc *gdp = ext3_get_group_desc (sb, i, NULL); + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL); if (!gdp) continue; count += le16_to_cpu(gdp->bg_used_dirs_count); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 03ba5bc..7275d60 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/inode.c + * linux/fs/ext4/inode.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -19,13 +19,13 @@ * 64-bit file support on 64-bit platforms by Jakub Jelinek * (jj@sunsite.ms.mff.cuni.cz) * - * Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000 + * Assorted race fixes, rewrite of ext4_get_block() by Al Viro, 2000 */ #include #include #include -#include +#include #include #include #include @@ -40,21 +40,21 @@ #include "xattr.h" #include "acl.h" -static int ext3_writepage_trans_blocks(struct inode *inode); +static int ext4_writepage_trans_blocks(struct inode *inode); /* * Test whether an inode is a fast symlink. */ -static int ext3_inode_is_fast_symlink(struct inode *inode) +static int ext4_inode_is_fast_symlink(struct inode *inode) { - int ea_blocks = EXT3_I(inode)->i_file_acl ? + int ea_blocks = EXT4_I(inode)->i_file_acl ? (inode->i_sb->s_blocksize >> 9) : 0; return (S_ISLNK(inode->i_mode) && inode->i_blocks - ea_blocks == 0); } /* - * The ext3 forget function must perform a revoke if we are freeing data + * The ext4 forget function must perform a revoke if we are freeing data * which has been journaled. Metadata (eg. indirect blocks) must be * revoked in all cases. * @@ -62,8 +62,8 @@ static int ext3_inode_is_fast_symlink(struct inode *inode) * but there may still be a record of it in the journal, and that record * still needs to be revoked. */ -int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, - struct buffer_head *bh, ext3_fsblk_t blocknr) +int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, + struct buffer_head *bh, ext4_fsblk_t blocknr) { int err; @@ -81,11 +81,11 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, * support it. Otherwise, only skip the revoke on un-journaled * data blocks. */ - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA || - (!is_metadata && !ext3_should_journal_data(inode))) { + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA || + (!is_metadata && !ext4_should_journal_data(inode))) { if (bh) { BUFFER_TRACE(bh, "call journal_forget"); - return ext3_journal_forget(handle, bh); + return ext4_journal_forget(handle, bh); } return 0; } @@ -93,10 +93,10 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, /* * data!=journal && (is_metadata || should_journal_data(inode)) */ - BUFFER_TRACE(bh, "call ext3_journal_revoke"); - err = ext3_journal_revoke(handle, blocknr, bh); + BUFFER_TRACE(bh, "call ext4_journal_revoke"); + err = ext4_journal_revoke(handle, blocknr, bh); if (err) - ext3_abort(inode->i_sb, __FUNCTION__, + ext4_abort(inode->i_sb, __FUNCTION__, "error %d when attempting revoke", err); BUFFER_TRACE(bh, "exit"); return err; @@ -115,7 +115,7 @@ static unsigned long blocks_for_truncate(struct inode *inode) /* Give ourselves just enough room to cope with inodes in which * i_blocks is corrupt: we've seen disk corruptions in the past * which resulted in random data in an inode which looked enough - * like a regular file for ext3 to try to delete it. Things + * like a regular file for ext4 to try to delete it. Things * will go a bit crazy if that happens, but at least we should * try not to panic the whole kernel. */ if (needed < 2) @@ -123,10 +123,10 @@ static unsigned long blocks_for_truncate(struct inode *inode) /* But we need to bound the transaction so we don't overflow the * journal. */ - if (needed > EXT3_MAX_TRANS_DATA) - needed = EXT3_MAX_TRANS_DATA; + if (needed > EXT4_MAX_TRANS_DATA) + needed = EXT4_MAX_TRANS_DATA; - return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed; + return EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + needed; } /* @@ -143,11 +143,11 @@ static handle_t *start_transaction(struct inode *inode) { handle_t *result; - result = ext3_journal_start(inode, blocks_for_truncate(inode)); + result = ext4_journal_start(inode, blocks_for_truncate(inode)); if (!IS_ERR(result)) return result; - ext3_std_error(inode->i_sb, PTR_ERR(result)); + ext4_std_error(inode->i_sb, PTR_ERR(result)); return result; } @@ -159,9 +159,9 @@ static handle_t *start_transaction(struct inode *inode) */ static int try_to_extend_transaction(handle_t *handle, struct inode *inode) { - if (handle->h_buffer_credits > EXT3_RESERVE_TRANS_BLOCKS) + if (handle->h_buffer_credits > EXT4_RESERVE_TRANS_BLOCKS) return 0; - if (!ext3_journal_extend(handle, blocks_for_truncate(inode))) + if (!ext4_journal_extend(handle, blocks_for_truncate(inode))) return 0; return 1; } @@ -171,16 +171,16 @@ static int try_to_extend_transaction(handle_t *handle, struct inode *inode) * so before we call here everything must be consistently dirtied against * this transaction. */ -static int ext3_journal_test_restart(handle_t *handle, struct inode *inode) +static int ext4_journal_test_restart(handle_t *handle, struct inode *inode) { jbd_debug(2, "restarting handle %p\n", handle); - return ext3_journal_restart(handle, blocks_for_truncate(inode)); + return ext4_journal_restart(handle, blocks_for_truncate(inode)); } /* * Called at the last iput() if i_nlink is zero. */ -void ext3_delete_inode (struct inode * inode) +void ext4_delete_inode (struct inode * inode) { handle_t *handle; @@ -196,7 +196,7 @@ void ext3_delete_inode (struct inode * inode) * make sure that the in-core orphan linked list is properly * cleaned up. */ - ext3_orphan_del(NULL, inode); + ext4_orphan_del(NULL, inode); goto no_delete; } @@ -204,17 +204,17 @@ void ext3_delete_inode (struct inode * inode) handle->h_sync = 1; inode->i_size = 0; if (inode->i_blocks) - ext3_truncate(inode); + ext4_truncate(inode); /* - * Kill off the orphan record which ext3_truncate created. + * Kill off the orphan record which ext4_truncate created. * AKPM: I think this can be inside the above `if'. - * Note that ext3_orphan_del() has to be able to cope with the + * Note that ext4_orphan_del() has to be able to cope with the * deletion of a non-existent orphan - this is because we don't - * know if ext3_truncate() actually created an orphan record. + * know if ext4_truncate() actually created an orphan record. * (Well, we could do this if we need to, but heck - it works) */ - ext3_orphan_del(handle, inode); - EXT3_I(inode)->i_dtime = get_seconds(); + ext4_orphan_del(handle, inode); + EXT4_I(inode)->i_dtime = get_seconds(); /* * One subtle ordering requirement: if anything has gone wrong @@ -223,12 +223,12 @@ void ext3_delete_inode (struct inode * inode) * having errors), but we can't free the inode if the mark_dirty * fails. */ - if (ext3_mark_inode_dirty(handle, inode)) + if (ext4_mark_inode_dirty(handle, inode)) /* If that failed, just do the required in-core inode clear. */ clear_inode(inode); else - ext3_free_inode(handle, inode); - ext3_journal_stop(handle); + ext4_free_inode(handle, inode); + ext4_journal_stop(handle); return; no_delete: clear_inode(inode); /* We must guarantee clearing of inode... */ @@ -254,14 +254,14 @@ static int verify_chain(Indirect *from, Indirect *to) } /** - * ext3_block_to_path - parse the block number into array of offsets + * ext4_block_to_path - parse the block number into array of offsets * @inode: inode in question (we are only interested in its superblock) * @i_block: block number to be parsed * @offsets: array to store the offsets in * @boundary: set this non-zero if the referred-to block is likely to be * followed (on disk) by an indirect block. * - * To store the locations of file's data ext3 uses a data structure common + * To store the locations of file's data ext4 uses a data structure common * for UNIX filesystems - tree of pointers anchored in the inode, with * data blocks at leaves and indirect blocks in intermediate nodes. * This function translates the block number into path in that tree - @@ -284,39 +284,39 @@ static int verify_chain(Indirect *from, Indirect *to) * get there at all. */ -static int ext3_block_to_path(struct inode *inode, +static int ext4_block_to_path(struct inode *inode, long i_block, int offsets[4], int *boundary) { - int ptrs = EXT3_ADDR_PER_BLOCK(inode->i_sb); - int ptrs_bits = EXT3_ADDR_PER_BLOCK_BITS(inode->i_sb); - const long direct_blocks = EXT3_NDIR_BLOCKS, + int ptrs = EXT4_ADDR_PER_BLOCK(inode->i_sb); + int ptrs_bits = EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb); + const long direct_blocks = EXT4_NDIR_BLOCKS, indirect_blocks = ptrs, double_blocks = (1 << (ptrs_bits * 2)); int n = 0; int final = 0; if (i_block < 0) { - ext3_warning (inode->i_sb, "ext3_block_to_path", "block < 0"); + ext4_warning (inode->i_sb, "ext4_block_to_path", "block < 0"); } else if (i_block < direct_blocks) { offsets[n++] = i_block; final = direct_blocks; } else if ( (i_block -= direct_blocks) < indirect_blocks) { - offsets[n++] = EXT3_IND_BLOCK; + offsets[n++] = EXT4_IND_BLOCK; offsets[n++] = i_block; final = ptrs; } else if ((i_block -= indirect_blocks) < double_blocks) { - offsets[n++] = EXT3_DIND_BLOCK; + offsets[n++] = EXT4_DIND_BLOCK; offsets[n++] = i_block >> ptrs_bits; offsets[n++] = i_block & (ptrs - 1); final = ptrs; } else if (((i_block -= double_blocks) >> (ptrs_bits * 2)) < ptrs) { - offsets[n++] = EXT3_TIND_BLOCK; + offsets[n++] = EXT4_TIND_BLOCK; offsets[n++] = i_block >> (ptrs_bits * 2); offsets[n++] = (i_block >> ptrs_bits) & (ptrs - 1); offsets[n++] = i_block & (ptrs - 1); final = ptrs; } else { - ext3_warning(inode->i_sb, "ext3_block_to_path", "block > big"); + ext4_warning(inode->i_sb, "ext4_block_to_path", "block > big"); } if (boundary) *boundary = final - 1 - (i_block & (ptrs - 1)); @@ -324,7 +324,7 @@ static int ext3_block_to_path(struct inode *inode, } /** - * ext3_get_branch - read the chain of indirect blocks leading to data + * ext4_get_branch - read the chain of indirect blocks leading to data * @inode: inode in question * @depth: depth of the chain (1 - direct pointer, etc.) * @offsets: offsets of pointers in inode/indirect blocks @@ -352,7 +352,7 @@ static int ext3_block_to_path(struct inode *inode, * or when it reads all @depth-1 indirect blocks successfully and finds * the whole chain, all way to the data (returns %NULL, *err == 0). */ -static Indirect *ext3_get_branch(struct inode *inode, int depth, int *offsets, +static Indirect *ext4_get_branch(struct inode *inode, int depth, int *offsets, Indirect chain[4], int *err) { struct super_block *sb = inode->i_sb; @@ -361,7 +361,7 @@ static Indirect *ext3_get_branch(struct inode *inode, int depth, int *offsets, *err = 0; /* i_data is not going away, no lock needed */ - add_chain (chain, NULL, EXT3_I(inode)->i_data + *offsets); + add_chain (chain, NULL, EXT4_I(inode)->i_data + *offsets); if (!p->key) goto no_block; while (--depth) { @@ -389,7 +389,7 @@ no_block: } /** - * ext3_find_near - find a place for allocation with sufficient locality + * ext4_find_near - find a place for allocation with sufficient locality * @inode: owner * @ind: descriptor of indirect block. * @@ -408,13 +408,13 @@ no_block: * * Caller must make sure that @ind is valid and will stay that way. */ -static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind) +static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) { - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; __le32 *p; - ext3_fsblk_t bg_start; - ext3_grpblk_t colour; + ext4_fsblk_t bg_start; + ext4_grpblk_t colour; /* Try to find previous block */ for (p = ind->p - 1; p >= start; p--) { @@ -430,14 +430,14 @@ static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind) * It is going to be referred to from the inode itself? OK, just put it * into the same cylinder group then. */ - bg_start = ext3_group_first_block_no(inode->i_sb, ei->i_block_group); + bg_start = ext4_group_first_block_no(inode->i_sb, ei->i_block_group); colour = (current->pid % 16) * - (EXT3_BLOCKS_PER_GROUP(inode->i_sb) / 16); + (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); return bg_start + colour; } /** - * ext3_find_goal - find a prefered place for allocation. + * ext4_find_goal - find a prefered place for allocation. * @inode: owner * @block: block we want * @chain: chain of indirect blocks @@ -448,12 +448,12 @@ static ext3_fsblk_t ext3_find_near(struct inode *inode, Indirect *ind) * stores it in *@goal and returns zero. */ -static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, +static ext4_fsblk_t ext4_find_goal(struct inode *inode, long block, Indirect chain[4], Indirect *partial) { - struct ext3_block_alloc_info *block_i; + struct ext4_block_alloc_info *block_i; - block_i = EXT3_I(inode)->i_block_alloc_info; + block_i = EXT4_I(inode)->i_block_alloc_info; /* * try the heuristic for sequential allocation, @@ -464,11 +464,11 @@ static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, return block_i->last_alloc_physical_block + 1; } - return ext3_find_near(inode, partial); + return ext4_find_near(inode, partial); } /** - * ext3_blks_to_allocate: Look up the block map and count the number + * ext4_blks_to_allocate: Look up the block map and count the number * of direct blocks need to be allocated for the given branch. * * @branch: chain of indirect blocks @@ -479,7 +479,7 @@ static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block, * return the total number of blocks to be allocate, including the * direct and indirect blocks. */ -static int ext3_blks_to_allocate(Indirect *branch, int k, unsigned long blks, +static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned long blks, int blocks_to_boundary) { unsigned long count = 0; @@ -506,7 +506,7 @@ static int ext3_blks_to_allocate(Indirect *branch, int k, unsigned long blks, } /** - * ext3_alloc_blocks: multiple allocate blocks needed for a branch + * ext4_alloc_blocks: multiple allocate blocks needed for a branch * @indirect_blks: the number of blocks need to allocate for indirect * blocks * @@ -515,14 +515,14 @@ static int ext3_blks_to_allocate(Indirect *branch, int k, unsigned long blks, * @blks: on return it will store the total number of allocated * direct blocks */ -static int ext3_alloc_blocks(handle_t *handle, struct inode *inode, - ext3_fsblk_t goal, int indirect_blks, int blks, - ext3_fsblk_t new_blocks[4], int *err) +static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int indirect_blks, int blks, + ext4_fsblk_t new_blocks[4], int *err) { int target, i; unsigned long count = 0; int index = 0; - ext3_fsblk_t current_block = 0; + ext4_fsblk_t current_block = 0; int ret = 0; /* @@ -538,7 +538,7 @@ static int ext3_alloc_blocks(handle_t *handle, struct inode *inode, while (1) { count = target; /* allocating blocks for indirect blocks and direct blocks */ - current_block = ext3_new_blocks(handle,inode,goal,&count,err); + current_block = ext4_new_blocks(handle,inode,goal,&count,err); if (*err) goto failed_out; @@ -562,12 +562,12 @@ static int ext3_alloc_blocks(handle_t *handle, struct inode *inode, return ret; failed_out: for (i = 0; i key). Upon the exit we have the same - * picture as after the successful ext3_get_block(), except that in one + * picture as after the successful ext4_get_block(), except that in one * place chain is disconnected - *branch->p is still zero (we did not * set the last link), but branch->key contains the number that should * be placed into *branch->p to fill that gap. * * If allocation fails we free all blocks we've allocated (and forget * their buffer_heads) and return the error value the from failed - * ext3_alloc_block() (normally -ENOSPC). Otherwise we set the chain + * ext4_alloc_block() (normally -ENOSPC). Otherwise we set the chain * as described above and return 0. */ -static int ext3_alloc_branch(handle_t *handle, struct inode *inode, - int indirect_blks, int *blks, ext3_fsblk_t goal, +static int ext4_alloc_branch(handle_t *handle, struct inode *inode, + int indirect_blks, int *blks, ext4_fsblk_t goal, int *offsets, Indirect *branch) { int blocksize = inode->i_sb->s_blocksize; @@ -600,10 +600,10 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, int err = 0; struct buffer_head *bh; int num; - ext3_fsblk_t new_blocks[4]; - ext3_fsblk_t current_block; + ext4_fsblk_t new_blocks[4]; + ext4_fsblk_t current_block; - num = ext3_alloc_blocks(handle, inode, goal, indirect_blks, + num = ext4_alloc_blocks(handle, inode, goal, indirect_blks, *blks, new_blocks, &err); if (err) return err; @@ -622,7 +622,7 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, branch[n].bh = bh; lock_buffer(bh); BUFFER_TRACE(bh, "call get_create_access"); - err = ext3_journal_get_create_access(handle, bh); + err = ext4_journal_get_create_access(handle, bh); if (err) { unlock_buffer(bh); brelse(bh); @@ -647,8 +647,8 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, set_buffer_uptodate(bh); unlock_buffer(bh); - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh); if (err) goto failed; } @@ -658,22 +658,22 @@ failed: /* Allocation failed, free what we already allocated */ for (i = 1; i <= n ; i++) { BUFFER_TRACE(branch[i].bh, "call journal_forget"); - ext3_journal_forget(handle, branch[i].bh); + ext4_journal_forget(handle, branch[i].bh); } for (i = 0; i i_blocks, etc.). In case of success we end up with the full * chain to new block and return 0. */ -static int ext3_splice_branch(handle_t *handle, struct inode *inode, +static int ext4_splice_branch(handle_t *handle, struct inode *inode, long block, Indirect *where, int num, int blks) { int i; int err = 0; - struct ext3_block_alloc_info *block_i; - ext3_fsblk_t current_block; + struct ext4_block_alloc_info *block_i; + ext4_fsblk_t current_block; - block_i = EXT3_I(inode)->i_block_alloc_info; + block_i = EXT4_I(inode)->i_block_alloc_info; /* * If we're splicing into a [td]indirect block (as opposed to the * inode) then we need to get write access to the [td]indirect block @@ -698,7 +698,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, */ if (where->bh) { BUFFER_TRACE(where->bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, where->bh); + err = ext4_journal_get_write_access(handle, where->bh); if (err) goto err_out; } @@ -730,7 +730,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, /* We are done with atomic stuff, now do the rest of housekeeping */ inode->i_ctime = CURRENT_TIME_SEC; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); /* had we spliced it onto indirect block? */ if (where->bh) { @@ -740,11 +740,11 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, * onto an indirect block at the very end of the file (the * file is growing) then we *will* alter the inode to reflect * the new i_size. But that is not done here - it is done in - * generic_commit_write->__mark_inode_dirty->ext3_dirty_inode. + * generic_commit_write->__mark_inode_dirty->ext4_dirty_inode. */ jbd_debug(5, "splicing indirect only\n"); - BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, where->bh); + BUFFER_TRACE(where->bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, where->bh); if (err) goto err_out; } else { @@ -759,10 +759,10 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, err_out: for (i = 1; i <= num; i++) { BUFFER_TRACE(where[i].bh, "call journal_forget"); - ext3_journal_forget(handle, where[i].bh); - ext3_free_blocks(handle,inode,le32_to_cpu(where[i-1].key),1); + ext4_journal_forget(handle, where[i].bh); + ext4_free_blocks(handle,inode,le32_to_cpu(where[i-1].key),1); } - ext3_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks); + ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks); return err; } @@ -786,7 +786,7 @@ err_out: * return = 0, if plain lookup failed. * return < 0, error case. */ -int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, +int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, sector_t iblock, unsigned long maxblocks, struct buffer_head *bh_result, int create, int extend_disksize) @@ -795,22 +795,22 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, int offsets[4]; Indirect chain[4]; Indirect *partial; - ext3_fsblk_t goal; + ext4_fsblk_t goal; int indirect_blks; int blocks_to_boundary = 0; int depth; - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); int count = 0; - ext3_fsblk_t first_block = 0; + ext4_fsblk_t first_block = 0; J_ASSERT(handle != NULL || create == 0); - depth = ext3_block_to_path(inode,iblock,offsets,&blocks_to_boundary); + depth = ext4_block_to_path(inode,iblock,offsets,&blocks_to_boundary); if (depth == 0) goto out; - partial = ext3_get_branch(inode, depth, offsets, chain, &err); + partial = ext4_get_branch(inode, depth, offsets, chain, &err); /* Simplest case - block found, no allocation needed */ if (!partial) { @@ -819,7 +819,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, count++; /*map more blocks*/ while (count < maxblocks && count <= blocks_to_boundary) { - ext3_fsblk_t blk; + ext4_fsblk_t blk; if (!verify_chain(chain, partial)) { /* @@ -852,7 +852,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, /* * If the indirect block is missing while we are reading - * the chain(ext3_get_branch() returns -EAGAIN err), or + * the chain(ext4_get_branch() returns -EAGAIN err), or * if the chain has been changed after we grab the semaphore, * (either because another process truncated this branch, or * another get_block allocated this branch) re-grab the chain to see if @@ -867,7 +867,7 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, brelse(partial->bh); partial--; } - partial = ext3_get_branch(inode, depth, offsets, chain, &err); + partial = ext4_get_branch(inode, depth, offsets, chain, &err); if (!partial) { count++; mutex_unlock(&ei->truncate_mutex); @@ -883,9 +883,9 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, * allocation info here if necessary */ if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) - ext3_init_block_alloc_info(inode); + ext4_init_block_alloc_info(inode); - goal = ext3_find_goal(inode, iblock, chain, partial); + goal = ext4_find_goal(inode, iblock, chain, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ indirect_blks = (chain + depth) - partial - 1; @@ -894,28 +894,28 @@ int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, * Next look up the indirect map to count the totoal number of * direct blocks to allocate for this branch. */ - count = ext3_blks_to_allocate(partial, indirect_blks, + count = ext4_blks_to_allocate(partial, indirect_blks, maxblocks, blocks_to_boundary); /* - * Block out ext3_truncate while we alter the tree + * Block out ext4_truncate while we alter the tree */ - err = ext3_alloc_branch(handle, inode, indirect_blks, &count, goal, + err = ext4_alloc_branch(handle, inode, indirect_blks, &count, goal, offsets + (partial - chain), partial); /* - * The ext3_splice_branch call will free and forget any buffers + * The ext4_splice_branch call will free and forget any buffers * on the new chain if there is a failure, but that risks using * up transaction credits, especially for bitmaps where the * credits cannot be returned. Can we handle this somehow? We * may need to return -EAGAIN upwards in the worst case. --sct */ if (!err) - err = ext3_splice_branch(handle, inode, iblock, + err = ext4_splice_branch(handle, inode, iblock, partial, indirect_blks, count); /* * i_disksize growing is protected by truncate_mutex. Don't forget to * protect it if you're about to implement concurrent - * ext3_get_block() -bzzz + * ext4_get_block() -bzzz */ if (!err && extend_disksize && inode->i_size > ei->i_disksize) ei->i_disksize = inode->i_size; @@ -942,9 +942,9 @@ out: return err; } -#define DIO_CREDITS (EXT3_RESERVE_TRANS_BLOCKS + 32) +#define DIO_CREDITS (EXT4_RESERVE_TRANS_BLOCKS + 32) -static int ext3_get_block(struct inode *inode, sector_t iblock, +static int ext4_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { handle_t *handle = journal_current_handle(); @@ -962,29 +962,29 @@ static int ext3_get_block(struct inode *inode, sector_t iblock, * Huge direct-io writes can hold off commits for long * periods of time. Let this commit run. */ - ext3_journal_stop(handle); - handle = ext3_journal_start(inode, DIO_CREDITS); + ext4_journal_stop(handle); + handle = ext4_journal_start(inode, DIO_CREDITS); if (IS_ERR(handle)) ret = PTR_ERR(handle); goto get_block; } - if (handle->h_buffer_credits <= EXT3_RESERVE_TRANS_BLOCKS) { + if (handle->h_buffer_credits <= EXT4_RESERVE_TRANS_BLOCKS) { /* * Getting low on buffer credits... */ - ret = ext3_journal_extend(handle, DIO_CREDITS); + ret = ext4_journal_extend(handle, DIO_CREDITS); if (ret > 0) { /* * Couldn't extend the transaction. Start a new one. */ - ret = ext3_journal_restart(handle, DIO_CREDITS); + ret = ext4_journal_restart(handle, DIO_CREDITS); } } get_block: if (ret == 0) { - ret = ext3_get_blocks_handle(handle, inode, iblock, + ret = ext4_get_blocks_handle(handle, inode, iblock, max_blocks, bh_result, create, 0); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); @@ -997,7 +997,7 @@ get_block: /* * `handle' can be NULL if create is zero */ -struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, +struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, long block, int create, int *errp) { struct buffer_head dummy; @@ -1008,10 +1008,10 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, dummy.b_state = 0; dummy.b_blocknr = -1000; buffer_trace_init(&dummy.b_history); - err = ext3_get_blocks_handle(handle, inode, block, 1, + err = ext4_get_blocks_handle(handle, inode, block, 1, &dummy, create, 1); /* - * ext3_get_blocks_handle() returns number of blocks + * ext4_get_blocks_handle() returns number of blocks * mapped. 0 in case of a HOLE. */ if (err > 0) { @@ -1035,19 +1035,19 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode *inode, * Now that we do not always journal data, we should * keep in mind whether this should always journal the * new buffer as metadata. For now, regular file - * writes use ext3_get_block instead, so it's not a + * writes use ext4_get_block instead, so it's not a * problem. */ lock_buffer(bh); BUFFER_TRACE(bh, "call get_create_access"); - fatal = ext3_journal_get_create_access(handle, bh); + fatal = ext4_journal_get_create_access(handle, bh); if (!fatal && !buffer_uptodate(bh)) { memset(bh->b_data,0,inode->i_sb->s_blocksize); set_buffer_uptodate(bh); } unlock_buffer(bh); - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh); if (!fatal) fatal = err; } else { @@ -1064,12 +1064,12 @@ err: return NULL; } -struct buffer_head *ext3_bread(handle_t *handle, struct inode *inode, +struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, int block, int create, int *err) { struct buffer_head * bh; - bh = ext3_getblk(handle, inode, block, create, err); + bh = ext4_getblk(handle, inode, block, create, err); if (!bh) return bh; if (buffer_uptodate(bh)) @@ -1118,17 +1118,17 @@ static int walk_page_buffers( handle_t *handle, /* * To preserve ordering, it is essential that the hole instantiation and * the data write be encapsulated in a single transaction. We cannot - * close off a transaction and start a new one between the ext3_get_block() + * close off a transaction and start a new one between the ext4_get_block() * and the commit_write(). So doing the journal_start at the start of * prepare_write() is the right place. * - * Also, this function can nest inside ext3_writepage() -> - * block_write_full_page(). In that case, we *know* that ext3_writepage() + * Also, this function can nest inside ext4_writepage() -> + * block_write_full_page(). In that case, we *know* that ext4_writepage() * has generated enough buffer credits to do the whole page. So we won't * block on the journal in that case, which is good, because the caller may * be PF_MEMALLOC. * - * By accident, ext3 can be reentered when a transaction is open via + * By accident, ext4 can be reentered when a transaction is open via * quota file writes. If we were to commit the transaction while thus * reentered, there can be a deadlock - we would be holding a quota * lock, and the commit would never complete if another thread had a @@ -1145,48 +1145,48 @@ static int do_journal_get_write_access(handle_t *handle, { if (!buffer_mapped(bh) || buffer_freed(bh)) return 0; - return ext3_journal_get_write_access(handle, bh); + return ext4_journal_get_write_access(handle, bh); } -static int ext3_prepare_write(struct file *file, struct page *page, +static int ext4_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) { struct inode *inode = page->mapping->host; - int ret, needed_blocks = ext3_writepage_trans_blocks(inode); + int ret, needed_blocks = ext4_writepage_trans_blocks(inode); handle_t *handle; int retries = 0; retry: - handle = ext3_journal_start(inode, needed_blocks); + handle = ext4_journal_start(inode, needed_blocks); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; } - if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) - ret = nobh_prepare_write(page, from, to, ext3_get_block); + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_prepare_write(page, from, to, ext4_get_block); else - ret = block_prepare_write(page, from, to, ext3_get_block); + ret = block_prepare_write(page, from, to, ext4_get_block); if (ret) goto prepare_write_failed; - if (ext3_should_journal_data(inode)) { + if (ext4_should_journal_data(inode)) { ret = walk_page_buffers(handle, page_buffers(page), from, to, NULL, do_journal_get_write_access); } prepare_write_failed: if (ret) - ext3_journal_stop(handle); - if (ret == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries)) + ext4_journal_stop(handle); + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; out: return ret; } -int ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh) +int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh) { int err = journal_dirty_data(handle, bh); if (err) - ext3_journal_abort_handle(__FUNCTION__, __FUNCTION__, + ext4_journal_abort_handle(__FUNCTION__, __FUNCTION__, bh, handle,err); return err; } @@ -1197,25 +1197,25 @@ static int commit_write_fn(handle_t *handle, struct buffer_head *bh) if (!buffer_mapped(bh) || buffer_freed(bh)) return 0; set_buffer_uptodate(bh); - return ext3_journal_dirty_metadata(handle, bh); + return ext4_journal_dirty_metadata(handle, bh); } /* * We need to pick up the new inode size which generic_commit_write gave us * `file' can be NULL - eg, when called from page_symlink(). * - * ext3 never places buffers on inode->i_mapping->private_list. metadata + * ext4 never places buffers on inode->i_mapping->private_list. metadata * buffers are managed internally. */ -static int ext3_ordered_commit_write(struct file *file, struct page *page, +static int ext4_ordered_commit_write(struct file *file, struct page *page, unsigned from, unsigned to) { - handle_t *handle = ext3_journal_current_handle(); + handle_t *handle = ext4_journal_current_handle(); struct inode *inode = page->mapping->host; int ret = 0, ret2; ret = walk_page_buffers(handle, page_buffers(page), - from, to, NULL, ext3_journal_dirty_data); + from, to, NULL, ext4_journal_dirty_data); if (ret == 0) { /* @@ -1226,43 +1226,43 @@ static int ext3_ordered_commit_write(struct file *file, struct page *page, loff_t new_i_size; new_i_size = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; - if (new_i_size > EXT3_I(inode)->i_disksize) - EXT3_I(inode)->i_disksize = new_i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = new_i_size; ret = generic_commit_write(file, page, from, to); } - ret2 = ext3_journal_stop(handle); + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; return ret; } -static int ext3_writeback_commit_write(struct file *file, struct page *page, +static int ext4_writeback_commit_write(struct file *file, struct page *page, unsigned from, unsigned to) { - handle_t *handle = ext3_journal_current_handle(); + handle_t *handle = ext4_journal_current_handle(); struct inode *inode = page->mapping->host; int ret = 0, ret2; loff_t new_i_size; new_i_size = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; - if (new_i_size > EXT3_I(inode)->i_disksize) - EXT3_I(inode)->i_disksize = new_i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = new_i_size; - if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) ret = nobh_commit_write(file, page, from, to); else ret = generic_commit_write(file, page, from, to); - ret2 = ext3_journal_stop(handle); + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; return ret; } -static int ext3_journalled_commit_write(struct file *file, +static int ext4_journalled_commit_write(struct file *file, struct page *page, unsigned from, unsigned to) { - handle_t *handle = ext3_journal_current_handle(); + handle_t *handle = ext4_journal_current_handle(); struct inode *inode = page->mapping->host; int ret = 0, ret2; int partial = 0; @@ -1279,14 +1279,14 @@ static int ext3_journalled_commit_write(struct file *file, SetPageUptodate(page); if (pos > inode->i_size) i_size_write(inode, pos); - EXT3_I(inode)->i_state |= EXT3_STATE_JDATA; - if (inode->i_size > EXT3_I(inode)->i_disksize) { - EXT3_I(inode)->i_disksize = inode->i_size; - ret2 = ext3_mark_inode_dirty(handle, inode); + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + if (inode->i_size > EXT4_I(inode)->i_disksize) { + EXT4_I(inode)->i_disksize = inode->i_size; + ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; } - ret2 = ext3_journal_stop(handle); + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; return ret; @@ -1297,7 +1297,7 @@ static int ext3_journalled_commit_write(struct file *file, * the swapper to find the on-disk block of a specific piece of data. * * Naturally, this is dangerous if the block concerned is still in the - * journal. If somebody makes a swapfile on an ext3 data-journaling + * journal. If somebody makes a swapfile on an ext4 data-journaling * filesystem and enables swap, then they may get a nasty shock when the * data getting swapped to that swapfile suddenly gets overwritten by * the original zero's written out previously to the journal and @@ -1306,13 +1306,13 @@ static int ext3_journalled_commit_write(struct file *file, * So, if we see any bmap calls here on a modified, data-journaled file, * take extra steps to flush any blocks which might be in the cache. */ -static sector_t ext3_bmap(struct address_space *mapping, sector_t block) +static sector_t ext4_bmap(struct address_space *mapping, sector_t block) { struct inode *inode = mapping->host; journal_t *journal; int err; - if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) { + if (EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { /* * This is a REALLY heavyweight approach, but the use of * bmap on dirty files is expected to be extremely rare: @@ -1324,15 +1324,15 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block) * in trouble if mortal users could trigger this path at * will.) * - * NB. EXT3_STATE_JDATA is not set on files other than + * NB. EXT4_STATE_JDATA is not set on files other than * regular files. If somebody wants to bmap a directory * or symlink and gets confused because the buffer * hasn't yet been flushed to disk, they deserve * everything they get. */ - EXT3_I(inode)->i_state &= ~EXT3_STATE_JDATA; - journal = EXT3_JOURNAL(inode); + EXT4_I(inode)->i_state &= ~EXT4_STATE_JDATA; + journal = EXT4_JOURNAL(inode); journal_lock_updates(journal); err = journal_flush(journal); journal_unlock_updates(journal); @@ -1341,7 +1341,7 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block) return 0; } - return generic_block_bmap(mapping,block,ext3_get_block); + return generic_block_bmap(mapping,block,ext4_get_block); } static int bget_one(handle_t *handle, struct buffer_head *bh) @@ -1359,14 +1359,14 @@ static int bput_one(handle_t *handle, struct buffer_head *bh) static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) { if (buffer_mapped(bh)) - return ext3_journal_dirty_data(handle, bh); + return ext4_journal_dirty_data(handle, bh); return 0; } /* * Note that we always start a transaction even if we're not journalling * data. This is to preserve ordering: any hole instantiation within - * __block_write_full_page -> ext3_get_block() should be journalled + * __block_write_full_page -> ext4_get_block() should be journalled * along with the data so we don't crash and then get metadata which * refers to old data. * @@ -1374,14 +1374,14 @@ static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) * * Problem: * - * ext3_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() -> - * ext3_writepage() + * ext4_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() -> + * ext4_writepage() * * Similar for: * - * ext3_file_write() -> generic_file_write() -> __alloc_pages() -> ... + * ext4_file_write() -> generic_file_write() -> __alloc_pages() -> ... * - * Same applies to ext3_get_block(). We will deadlock on various things like + * Same applies to ext4_get_block(). We will deadlock on various things like * lock_journal and i_truncate_mutex. * * Setting PF_MEMALLOC here doesn't work - too many internal memory @@ -1415,7 +1415,7 @@ static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) * AKPM2: if all the page's buffers are mapped to disk and !data=journal, * we don't need to open a transaction here. */ -static int ext3_ordered_writepage(struct page *page, +static int ext4_ordered_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; @@ -1430,10 +1430,10 @@ static int ext3_ordered_writepage(struct page *page, * We give up here if we're reentered, because it might be for a * different filesystem. */ - if (ext3_journal_current_handle()) + if (ext4_journal_current_handle()) goto out_fail; - handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); @@ -1448,7 +1448,7 @@ static int ext3_ordered_writepage(struct page *page, walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, bget_one); - ret = block_write_full_page(page, ext3_get_block, wbc); + ret = block_write_full_page(page, ext4_get_block, wbc); /* * The page can become unlocked at any point now, and @@ -1470,7 +1470,7 @@ static int ext3_ordered_writepage(struct page *page, } walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, bput_one); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; @@ -1481,7 +1481,7 @@ out_fail: return ret; } -static int ext3_writeback_writepage(struct page *page, +static int ext4_writeback_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; @@ -1489,21 +1489,21 @@ static int ext3_writeback_writepage(struct page *page, int ret = 0; int err; - if (ext3_journal_current_handle()) + if (ext4_journal_current_handle()) goto out_fail; - handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out_fail; } - if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) - ret = nobh_writepage(page, ext3_get_block, wbc); + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_writepage(page, ext4_get_block, wbc); else - ret = block_write_full_page(page, ext3_get_block, wbc); + ret = block_write_full_page(page, ext4_get_block, wbc); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; @@ -1514,7 +1514,7 @@ out_fail: return ret; } -static int ext3_journalled_writepage(struct page *page, +static int ext4_journalled_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; @@ -1522,10 +1522,10 @@ static int ext3_journalled_writepage(struct page *page, int ret = 0; int err; - if (ext3_journal_current_handle()) + if (ext4_journal_current_handle()) goto no_write; - handle = ext3_journal_start(inode, ext3_writepage_trans_blocks(inode)); + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto no_write; @@ -1538,9 +1538,9 @@ static int ext3_journalled_writepage(struct page *page, */ ClearPageChecked(page); ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, - ext3_get_block); + ext4_get_block); if (ret != 0) { - ext3_journal_stop(handle); + ext4_journal_stop(handle); goto out_unlock; } ret = walk_page_buffers(handle, page_buffers(page), 0, @@ -1550,7 +1550,7 @@ static int ext3_journalled_writepage(struct page *page, PAGE_CACHE_SIZE, NULL, commit_write_fn); if (ret == 0) ret = err; - EXT3_I(inode)->i_state |= EXT3_STATE_JDATA; + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; unlock_page(page); } else { /* @@ -1558,9 +1558,9 @@ static int ext3_journalled_writepage(struct page *page, * really know unless we go poke around in the buffer_heads. * But block_write_full_page will do the right thing. */ - ret = block_write_full_page(page, ext3_get_block, wbc); + ret = block_write_full_page(page, ext4_get_block, wbc); } - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; out: @@ -1573,21 +1573,21 @@ out_unlock: goto out; } -static int ext3_readpage(struct file *file, struct page *page) +static int ext4_readpage(struct file *file, struct page *page) { - return mpage_readpage(page, ext3_get_block); + return mpage_readpage(page, ext4_get_block); } static int -ext3_readpages(struct file *file, struct address_space *mapping, +ext4_readpages(struct file *file, struct address_space *mapping, struct list_head *pages, unsigned nr_pages) { - return mpage_readpages(mapping, pages, nr_pages, ext3_get_block); + return mpage_readpages(mapping, pages, nr_pages, ext4_get_block); } -static void ext3_invalidatepage(struct page *page, unsigned long offset) +static void ext4_invalidatepage(struct page *page, unsigned long offset) { - journal_t *journal = EXT3_JOURNAL(page->mapping->host); + journal_t *journal = EXT4_JOURNAL(page->mapping->host); /* * If it's a full truncate we just forget about the pending dirtying @@ -1598,9 +1598,9 @@ static void ext3_invalidatepage(struct page *page, unsigned long offset) journal_invalidatepage(journal, page, offset); } -static int ext3_releasepage(struct page *page, gfp_t wait) +static int ext4_releasepage(struct page *page, gfp_t wait) { - journal_t *journal = EXT3_JOURNAL(page->mapping->host); + journal_t *journal = EXT4_JOURNAL(page->mapping->host); WARN_ON(PageChecked(page)); if (!page_has_buffers(page)) @@ -1616,13 +1616,13 @@ static int ext3_releasepage(struct page *page, gfp_t wait) * If the O_DIRECT write is intantiating holes inside i_size and the machine * crashes then stale disk data _may_ be exposed inside the file. */ -static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, +static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, unsigned long nr_segs) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); handle_t *handle = NULL; ssize_t ret; int orphan = 0; @@ -1631,13 +1631,13 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, if (rw == WRITE) { loff_t final_size = offset + count; - handle = ext3_journal_start(inode, DIO_CREDITS); + handle = ext4_journal_start(inode, DIO_CREDITS); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; } if (final_size > inode->i_size) { - ret = ext3_orphan_add(handle, inode); + ret = ext4_orphan_add(handle, inode); if (ret) goto out_stop; orphan = 1; @@ -1647,10 +1647,10 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, offset, nr_segs, - ext3_get_block, NULL); + ext4_get_block, NULL); /* - * Reacquire the handle: ext3_get_block() can restart the transaction + * Reacquire the handle: ext4_get_block() can restart the transaction */ handle = journal_current_handle(); @@ -1659,7 +1659,7 @@ out_stop: int err; if (orphan && inode->i_nlink) - ext3_orphan_del(handle, inode); + ext4_orphan_del(handle, inode); if (orphan && ret > 0) { loff_t end = offset + ret; if (end > inode->i_size) { @@ -1669,13 +1669,13 @@ out_stop: * We're going to return a positive `ret' * here due to non-zero-length I/O, so there's * no way of reporting error returns from - * ext3_mark_inode_dirty() to userspace. So + * ext4_mark_inode_dirty() to userspace. So * ignore it. */ - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); } } - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (ret == 0) ret = err; } @@ -1684,7 +1684,7 @@ out: } /* - * Pages can be marked dirty completely asynchronously from ext3's journalling + * Pages can be marked dirty completely asynchronously from ext4's journalling * activity. By filemap_sync_pte(), try_to_unmap_one(), etc. We cannot do * much here because ->set_page_dirty is called under VFS locks. The page is * not necessarily locked. @@ -1696,73 +1696,73 @@ out: * So what we do is to mark the page "pending dirty" and next time writepage * is called, propagate that into the buffers appropriately. */ -static int ext3_journalled_set_page_dirty(struct page *page) +static int ext4_journalled_set_page_dirty(struct page *page) { SetPageChecked(page); return __set_page_dirty_nobuffers(page); } -static const struct address_space_operations ext3_ordered_aops = { - .readpage = ext3_readpage, - .readpages = ext3_readpages, - .writepage = ext3_ordered_writepage, +static const struct address_space_operations ext4_ordered_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_ordered_writepage, .sync_page = block_sync_page, - .prepare_write = ext3_prepare_write, - .commit_write = ext3_ordered_commit_write, - .bmap = ext3_bmap, - .invalidatepage = ext3_invalidatepage, - .releasepage = ext3_releasepage, - .direct_IO = ext3_direct_IO, + .prepare_write = ext4_prepare_write, + .commit_write = ext4_ordered_commit_write, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, .migratepage = buffer_migrate_page, }; -static const struct address_space_operations ext3_writeback_aops = { - .readpage = ext3_readpage, - .readpages = ext3_readpages, - .writepage = ext3_writeback_writepage, +static const struct address_space_operations ext4_writeback_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_writeback_writepage, .sync_page = block_sync_page, - .prepare_write = ext3_prepare_write, - .commit_write = ext3_writeback_commit_write, - .bmap = ext3_bmap, - .invalidatepage = ext3_invalidatepage, - .releasepage = ext3_releasepage, - .direct_IO = ext3_direct_IO, + .prepare_write = ext4_prepare_write, + .commit_write = ext4_writeback_commit_write, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, .migratepage = buffer_migrate_page, }; -static const struct address_space_operations ext3_journalled_aops = { - .readpage = ext3_readpage, - .readpages = ext3_readpages, - .writepage = ext3_journalled_writepage, +static const struct address_space_operations ext4_journalled_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_journalled_writepage, .sync_page = block_sync_page, - .prepare_write = ext3_prepare_write, - .commit_write = ext3_journalled_commit_write, - .set_page_dirty = ext3_journalled_set_page_dirty, - .bmap = ext3_bmap, - .invalidatepage = ext3_invalidatepage, - .releasepage = ext3_releasepage, + .prepare_write = ext4_prepare_write, + .commit_write = ext4_journalled_commit_write, + .set_page_dirty = ext4_journalled_set_page_dirty, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, }; -void ext3_set_aops(struct inode *inode) +void ext4_set_aops(struct inode *inode) { - if (ext3_should_order_data(inode)) - inode->i_mapping->a_ops = &ext3_ordered_aops; - else if (ext3_should_writeback_data(inode)) - inode->i_mapping->a_ops = &ext3_writeback_aops; + if (ext4_should_order_data(inode)) + inode->i_mapping->a_ops = &ext4_ordered_aops; + else if (ext4_should_writeback_data(inode)) + inode->i_mapping->a_ops = &ext4_writeback_aops; else - inode->i_mapping->a_ops = &ext3_journalled_aops; + inode->i_mapping->a_ops = &ext4_journalled_aops; } /* - * ext3_block_truncate_page() zeroes out a mapping from file offset `from' + * ext4_block_truncate_page() zeroes out a mapping from file offset `from' * up to the end of the block which corresponds to `from'. * This required during truncate. We need to physically zero the tail end * of that block so it doesn't yield old data if the file is later grown. */ -static int ext3_block_truncate_page(handle_t *handle, struct page *page, +static int ext4_block_truncate_page(handle_t *handle, struct page *page, struct address_space *mapping, loff_t from) { - ext3_fsblk_t index = from >> PAGE_CACHE_SHIFT; + ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; unsigned offset = from & (PAGE_CACHE_SIZE-1); unsigned blocksize, iblock, length, pos; struct inode *inode = mapping->host; @@ -1779,7 +1779,7 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, * read-in the page - otherwise we create buffers to do the IO. */ if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && - ext3_should_writeback_data(inode) && PageUptodate(page)) { + ext4_should_writeback_data(inode) && PageUptodate(page)) { kaddr = kmap_atomic(page, KM_USER0); memset(kaddr + offset, 0, length); flush_dcache_page(page); @@ -1808,7 +1808,7 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, if (!buffer_mapped(bh)) { BUFFER_TRACE(bh, "unmapped"); - ext3_get_block(inode, iblock, bh, 0); + ext4_get_block(inode, iblock, bh, 0); /* unmapped? It's a hole - nothing to do */ if (!buffer_mapped(bh)) { BUFFER_TRACE(bh, "still unmapped"); @@ -1829,9 +1829,9 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, goto unlock; } - if (ext3_should_journal_data(inode)) { + if (ext4_should_journal_data(inode)) { BUFFER_TRACE(bh, "get write access"); - err = ext3_journal_get_write_access(handle, bh); + err = ext4_journal_get_write_access(handle, bh); if (err) goto unlock; } @@ -1844,11 +1844,11 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, BUFFER_TRACE(bh, "zeroed end of block"); err = 0; - if (ext3_should_journal_data(inode)) { - err = ext3_journal_dirty_metadata(handle, bh); + if (ext4_should_journal_data(inode)) { + err = ext4_journal_dirty_metadata(handle, bh); } else { - if (ext3_should_order_data(inode)) - err = ext3_journal_dirty_data(handle, bh); + if (ext4_should_order_data(inode)) + err = ext4_journal_dirty_data(handle, bh); mark_buffer_dirty(bh); } @@ -1872,14 +1872,14 @@ static inline int all_zeroes(__le32 *p, __le32 *q) } /** - * ext3_find_shared - find the indirect blocks for partial truncation. + * ext4_find_shared - find the indirect blocks for partial truncation. * @inode: inode in question * @depth: depth of the affected branch - * @offsets: offsets of pointers in that branch (see ext3_block_to_path) + * @offsets: offsets of pointers in that branch (see ext4_block_to_path) * @chain: place to store the pointers to partial indirect blocks * @top: place to the (detached) top of branch * - * This is a helper function used by ext3_truncate(). + * This is a helper function used by ext4_truncate(). * * When we do truncate() we may have to clean the ends of several * indirect blocks but leave the blocks themselves alive. Block is @@ -1887,7 +1887,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * from it (and it is on the path to the first completely truncated * data block, indeed). We have to free the top of that path along * with everything to the right of the path. Since no allocation - * past the truncation point is possible until ext3_truncate() + * past the truncation point is possible until ext4_truncate() * finishes, we may safely do the latter, but top of branch may * require special attention - pageout below the truncation point * might try to populate it. @@ -1906,7 +1906,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * c) free the subtrees growing from the inode past the @chain[0]. * (no partially truncated stuff there). */ -static Indirect *ext3_find_shared(struct inode *inode, int depth, +static Indirect *ext4_find_shared(struct inode *inode, int depth, int offsets[4], Indirect chain[4], __le32 *top) { Indirect *partial, *p; @@ -1916,7 +1916,7 @@ static Indirect *ext3_find_shared(struct inode *inode, int depth, /* Make k index the deepest non-null offest + 1 */ for (k = depth; k > 1 && !offsets[k-1]; k--) ; - partial = ext3_get_branch(inode, k, offsets, chain, &err); + partial = ext4_get_branch(inode, k, offsets, chain, &err); /* Writer: pointers */ if (!partial) partial = chain + k-1; @@ -1939,7 +1939,7 @@ static Indirect *ext3_find_shared(struct inode *inode, int depth, p->p--; } else { *top = *p->p; - /* Nope, don't do this in ext3. Must leave the tree intact */ + /* Nope, don't do this in ext4. Must leave the tree intact */ #if 0 *p->p = 0; #endif @@ -1962,21 +1962,21 @@ no_top: * We release `count' blocks on disk, but (last - first) may be greater * than `count' because there can be holes in there. */ -static void ext3_clear_blocks(handle_t *handle, struct inode *inode, - struct buffer_head *bh, ext3_fsblk_t block_to_free, +static void ext4_clear_blocks(handle_t *handle, struct inode *inode, + struct buffer_head *bh, ext4_fsblk_t block_to_free, unsigned long count, __le32 *first, __le32 *last) { __le32 *p; if (try_to_extend_transaction(handle, inode)) { if (bh) { - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, bh); } - ext3_mark_inode_dirty(handle, inode); - ext3_journal_test_restart(handle, inode); + ext4_mark_inode_dirty(handle, inode); + ext4_journal_test_restart(handle, inode); if (bh) { BUFFER_TRACE(bh, "retaking write access"); - ext3_journal_get_write_access(handle, bh); + ext4_journal_get_write_access(handle, bh); } } @@ -1995,15 +1995,15 @@ static void ext3_clear_blocks(handle_t *handle, struct inode *inode, *p = 0; bh = sb_find_get_block(inode->i_sb, nr); - ext3_forget(handle, 0, inode, bh, nr); + ext4_forget(handle, 0, inode, bh, nr); } } - ext3_free_blocks(handle, inode, block_to_free, count); + ext4_free_blocks(handle, inode, block_to_free, count); } /** - * ext3_free_data - free a list of data blocks + * ext4_free_data - free a list of data blocks * @handle: handle for this transaction * @inode: inode we are dealing with * @this_bh: indirect buffer_head which contains *@first and *@last @@ -2021,23 +2021,23 @@ static void ext3_clear_blocks(handle_t *handle, struct inode *inode, * @this_bh will be %NULL if @first and @last point into the inode's direct * block pointers. */ -static void ext3_free_data(handle_t *handle, struct inode *inode, +static void ext4_free_data(handle_t *handle, struct inode *inode, struct buffer_head *this_bh, __le32 *first, __le32 *last) { - ext3_fsblk_t block_to_free = 0; /* Starting block # of a run */ + ext4_fsblk_t block_to_free = 0; /* Starting block # of a run */ unsigned long count = 0; /* Number of blocks in the run */ __le32 *block_to_free_p = NULL; /* Pointer into inode/ind corresponding to block_to_free */ - ext3_fsblk_t nr; /* Current block # */ + ext4_fsblk_t nr; /* Current block # */ __le32 *p; /* Pointer into inode/ind for current block */ int err; if (this_bh) { /* For indirect block */ BUFFER_TRACE(this_bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, this_bh); + err = ext4_journal_get_write_access(handle, this_bh); /* Important: if we can't update the indirect pointers * to the blocks, we can't free them. */ if (err) @@ -2055,7 +2055,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, } else if (nr == block_to_free + count) { count++; } else { - ext3_clear_blocks(handle, inode, this_bh, + ext4_clear_blocks(handle, inode, this_bh, block_to_free, count, block_to_free_p, p); block_to_free = nr; @@ -2066,17 +2066,17 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, } if (count > 0) - ext3_clear_blocks(handle, inode, this_bh, block_to_free, + ext4_clear_blocks(handle, inode, this_bh, block_to_free, count, block_to_free_p, p); if (this_bh) { - BUFFER_TRACE(this_bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, this_bh); + BUFFER_TRACE(this_bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, this_bh); } } /** - * ext3_free_branches - free an array of branches + * ext4_free_branches - free an array of branches * @handle: JBD handle for this transaction * @inode: inode we are dealing with * @parent_bh: the buffer_head which contains *@first and *@last @@ -2088,11 +2088,11 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, * stored as little-endian 32-bit) and updating @inode->i_blocks * appropriately. */ -static void ext3_free_branches(handle_t *handle, struct inode *inode, +static void ext4_free_branches(handle_t *handle, struct inode *inode, struct buffer_head *parent_bh, __le32 *first, __le32 *last, int depth) { - ext3_fsblk_t nr; + ext4_fsblk_t nr; __le32 *p; if (is_handle_aborted(handle)) @@ -2100,7 +2100,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, if (depth--) { struct buffer_head *bh; - int addr_per_block = EXT3_ADDR_PER_BLOCK(inode->i_sb); + int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb); p = last; while (--p >= first) { nr = le32_to_cpu(*p); @@ -2115,7 +2115,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, * (should be rare). */ if (!bh) { - ext3_error(inode->i_sb, "ext3_free_branches", + ext4_error(inode->i_sb, "ext4_free_branches", "Read failure, inode=%lu, block="E3FSBLK, inode->i_ino, nr); continue; @@ -2123,7 +2123,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, /* This zaps the entire block. Bottom up. */ BUFFER_TRACE(bh, "free child branches"); - ext3_free_branches(handle, inode, bh, + ext4_free_branches(handle, inode, bh, (__le32*)bh->b_data, (__le32*)bh->b_data + addr_per_block, depth); @@ -2138,7 +2138,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, * transaction. But if it's part of the committing * transaction then journal_forget() will simply * brelse() it. That means that if the underlying - * block is reallocated in ext3_get_block(), + * block is reallocated in ext4_get_block(), * unmap_underlying_metadata() will find this block * and will try to get rid of it. damn, damn. * @@ -2147,7 +2147,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, * revoke records must be emitted *before* clearing * this block's bit in the bitmaps. */ - ext3_forget(handle, 1, inode, bh, bh->b_blocknr); + ext4_forget(handle, 1, inode, bh, bh->b_blocknr); /* * Everything below this this pointer has been @@ -2168,11 +2168,11 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, if (is_handle_aborted(handle)) return; if (try_to_extend_transaction(handle, inode)) { - ext3_mark_inode_dirty(handle, inode); - ext3_journal_test_restart(handle, inode); + ext4_mark_inode_dirty(handle, inode); + ext4_journal_test_restart(handle, inode); } - ext3_free_blocks(handle, inode, nr, 1); + ext4_free_blocks(handle, inode, nr, 1); if (parent_bh) { /* @@ -2180,12 +2180,12 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, * pointed to by an indirect block: journal it */ BUFFER_TRACE(parent_bh, "get_write_access"); - if (!ext3_journal_get_write_access(handle, + if (!ext4_journal_get_write_access(handle, parent_bh)){ *p = 0; BUFFER_TRACE(parent_bh, - "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, + "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, parent_bh); } } @@ -2193,15 +2193,15 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, } else { /* We have reached the bottom of the tree. */ BUFFER_TRACE(parent_bh, "free data blocks"); - ext3_free_data(handle, inode, parent_bh, first, last); + ext4_free_data(handle, inode, parent_bh, first, last); } } /* - * ext3_truncate() + * ext4_truncate() * - * We block out ext3_get_block() block instantiations across the entire - * transaction, and VFS/VM ensures that ext3_truncate() cannot run + * We block out ext4_get_block() block instantiations across the entire + * transaction, and VFS/VM ensures that ext4_truncate() cannot run * simultaneously on behalf of the same inode. * * As we work through the truncate and commmit bits of it to the journal there @@ -2218,19 +2218,19 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode, * truncate against the orphan inode list. * * The committed inode has the new, desired i_size (which is the same as - * i_disksize in this case). After a crash, ext3_orphan_cleanup() will see + * i_disksize in this case). After a crash, ext4_orphan_cleanup() will see * that this inode's truncate did not complete and it will again call - * ext3_truncate() to have another go. So there will be instantiated blocks - * to the right of the truncation point in a crashed ext3 filesystem. But + * ext4_truncate() to have another go. So there will be instantiated blocks + * to the right of the truncation point in a crashed ext4 filesystem. But * that's fine - as long as they are linked from the inode, the post-crash - * ext3_truncate() run will find them and release them. + * ext4_truncate() run will find them and release them. */ -void ext3_truncate(struct inode *inode) +void ext4_truncate(struct inode *inode) { handle_t *handle; - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); __le32 *i_data = ei->i_data; - int addr_per_block = EXT3_ADDR_PER_BLOCK(inode->i_sb); + int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb); struct address_space *mapping = inode->i_mapping; int offsets[4]; Indirect chain[4]; @@ -2244,7 +2244,7 @@ void ext3_truncate(struct inode *inode) if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) return; - if (ext3_inode_is_fast_symlink(inode)) + if (ext4_inode_is_fast_symlink(inode)) return; if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) return; @@ -2275,12 +2275,12 @@ void ext3_truncate(struct inode *inode) } last_block = (inode->i_size + blocksize-1) - >> EXT3_BLOCK_SIZE_BITS(inode->i_sb); + >> EXT4_BLOCK_SIZE_BITS(inode->i_sb); if (page) - ext3_block_truncate_page(handle, page, mapping, inode->i_size); + ext4_block_truncate_page(handle, page, mapping, inode->i_size); - n = ext3_block_to_path(inode, last_block, offsets, NULL); + n = ext4_block_to_path(inode, last_block, offsets, NULL); if (n == 0) goto out_stop; /* error */ @@ -2293,7 +2293,7 @@ void ext3_truncate(struct inode *inode) * Implication: the file must always be in a sane, consistent * truncatable state while each transaction commits. */ - if (ext3_orphan_add(handle, inode)) + if (ext4_orphan_add(handle, inode)) goto out_stop; /* @@ -2301,28 +2301,28 @@ void ext3_truncate(struct inode *inode) * occurs before the truncate completes, so it is now safe to propagate * the new, shorter inode size (held for now in i_size) into the * on-disk inode. We do this via i_disksize, which is the value which - * ext3 *really* writes onto the disk inode. + * ext4 *really* writes onto the disk inode. */ ei->i_disksize = inode->i_size; /* - * From here we block out all ext3_get_block() callers who want to + * From here we block out all ext4_get_block() callers who want to * modify the block allocation tree. */ mutex_lock(&ei->truncate_mutex); if (n == 1) { /* direct blocks */ - ext3_free_data(handle, inode, NULL, i_data+offsets[0], - i_data + EXT3_NDIR_BLOCKS); + ext4_free_data(handle, inode, NULL, i_data+offsets[0], + i_data + EXT4_NDIR_BLOCKS); goto do_indirects; } - partial = ext3_find_shared(inode, n, offsets, chain, &nr); + partial = ext4_find_shared(inode, n, offsets, chain, &nr); /* Kill the top of shared branch (not detached) */ if (nr) { if (partial == chain) { /* Shared branch grows from the inode */ - ext3_free_branches(handle, inode, NULL, + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, (chain+n-1) - partial); *partial->p = 0; /* @@ -2332,14 +2332,14 @@ void ext3_truncate(struct inode *inode) } else { /* Shared branch grows from an indirect block */ BUFFER_TRACE(partial->bh, "get_write_access"); - ext3_free_branches(handle, inode, partial->bh, + ext4_free_branches(handle, inode, partial->bh, partial->p, partial->p+1, (chain+n-1) - partial); } } /* Clear the ends of indirect blocks on the shared branch */ while (partial > chain) { - ext3_free_branches(handle, inode, partial->bh, partial->p + 1, + ext4_free_branches(handle, inode, partial->bh, partial->p + 1, (__le32*)partial->bh->b_data+addr_per_block, (chain+n-1) - partial); BUFFER_TRACE(partial->bh, "call brelse"); @@ -2350,32 +2350,32 @@ do_indirects: /* Kill the remaining (whole) subtrees */ switch (offsets[0]) { default: - nr = i_data[EXT3_IND_BLOCK]; + nr = i_data[EXT4_IND_BLOCK]; if (nr) { - ext3_free_branches(handle, inode, NULL, &nr, &nr+1, 1); - i_data[EXT3_IND_BLOCK] = 0; + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 1); + i_data[EXT4_IND_BLOCK] = 0; } - case EXT3_IND_BLOCK: - nr = i_data[EXT3_DIND_BLOCK]; + case EXT4_IND_BLOCK: + nr = i_data[EXT4_DIND_BLOCK]; if (nr) { - ext3_free_branches(handle, inode, NULL, &nr, &nr+1, 2); - i_data[EXT3_DIND_BLOCK] = 0; + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 2); + i_data[EXT4_DIND_BLOCK] = 0; } - case EXT3_DIND_BLOCK: - nr = i_data[EXT3_TIND_BLOCK]; + case EXT4_DIND_BLOCK: + nr = i_data[EXT4_TIND_BLOCK]; if (nr) { - ext3_free_branches(handle, inode, NULL, &nr, &nr+1, 3); - i_data[EXT3_TIND_BLOCK] = 0; + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 3); + i_data[EXT4_TIND_BLOCK] = 0; } - case EXT3_TIND_BLOCK: + case EXT4_TIND_BLOCK: ; } - ext3_discard_reservation(inode); + ext4_discard_reservation(inode); mutex_unlock(&ei->truncate_mutex); inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); /* * In a multi-transaction truncate, we only make the final transaction @@ -2388,25 +2388,25 @@ out_stop: * If this was a simple ftruncate(), and the file will remain alive * then we need to clear up the orphan record which we created above. * However, if this was a real unlink then we were called by - * ext3_delete_inode(), and we allow that function to clean up the + * ext4_delete_inode(), and we allow that function to clean up the * orphan info for us. */ if (inode->i_nlink) - ext3_orphan_del(handle, inode); + ext4_orphan_del(handle, inode); - ext3_journal_stop(handle); + ext4_journal_stop(handle); } -static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, - unsigned long ino, struct ext3_iloc *iloc) +static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, + unsigned long ino, struct ext4_iloc *iloc) { unsigned long desc, group_desc, block_group; unsigned long offset; - ext3_fsblk_t block; + ext4_fsblk_t block; struct buffer_head *bh; - struct ext3_group_desc * gdp; + struct ext4_group_desc * gdp; - if (!ext3_valid_inum(sb, ino)) { + if (!ext4_valid_inum(sb, ino)) { /* * This error is already checked for in namei.c unless we are * looking at an NFS filehandle, in which case no error @@ -2415,54 +2415,54 @@ static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, return 0; } - block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); - if (block_group >= EXT3_SB(sb)->s_groups_count) { - ext3_error(sb,"ext3_get_inode_block","group >= groups count"); + block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); + if (block_group >= EXT4_SB(sb)->s_groups_count) { + ext4_error(sb,"ext4_get_inode_block","group >= groups count"); return 0; } smp_rmb(); - group_desc = block_group >> EXT3_DESC_PER_BLOCK_BITS(sb); - desc = block_group & (EXT3_DESC_PER_BLOCK(sb) - 1); - bh = EXT3_SB(sb)->s_group_desc[group_desc]; + group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); + desc = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); + bh = EXT4_SB(sb)->s_group_desc[group_desc]; if (!bh) { - ext3_error (sb, "ext3_get_inode_block", + ext4_error (sb, "ext4_get_inode_block", "Descriptor not loaded"); return 0; } - gdp = (struct ext3_group_desc *)bh->b_data; + gdp = (struct ext4_group_desc *)bh->b_data; /* * Figure out the offset within the block group inode table */ - offset = ((ino - 1) % EXT3_INODES_PER_GROUP(sb)) * - EXT3_INODE_SIZE(sb); + offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * + EXT4_INODE_SIZE(sb); block = le32_to_cpu(gdp[desc].bg_inode_table) + - (offset >> EXT3_BLOCK_SIZE_BITS(sb)); + (offset >> EXT4_BLOCK_SIZE_BITS(sb)); iloc->block_group = block_group; - iloc->offset = offset & (EXT3_BLOCK_SIZE(sb) - 1); + iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); return block; } /* - * ext3_get_inode_loc returns with an extra refcount against the inode's + * ext4_get_inode_loc returns with an extra refcount against the inode's * underlying buffer_head on success. If 'in_mem' is true, we have all * data in memory that is needed to recreate the on-disk version of this * inode. */ -static int __ext3_get_inode_loc(struct inode *inode, - struct ext3_iloc *iloc, int in_mem) +static int __ext4_get_inode_loc(struct inode *inode, + struct ext4_iloc *iloc, int in_mem) { - ext3_fsblk_t block; + ext4_fsblk_t block; struct buffer_head *bh; - block = ext3_get_inode_block(inode->i_sb, inode->i_ino, iloc); + block = ext4_get_inode_block(inode->i_sb, inode->i_ino, iloc); if (!block) return -EIO; bh = sb_getblk(inode->i_sb, block); if (!bh) { - ext3_error (inode->i_sb, "ext3_get_inode_loc", + ext4_error (inode->i_sb, "ext4_get_inode_loc", "unable to read inode block - " "inode=%lu, block="E3FSBLK, inode->i_ino, block); @@ -2483,22 +2483,22 @@ static int __ext3_get_inode_loc(struct inode *inode, */ if (in_mem) { struct buffer_head *bitmap_bh; - struct ext3_group_desc *desc; + struct ext4_group_desc *desc; int inodes_per_buffer; int inode_offset, i; int block_group; int start; block_group = (inode->i_ino - 1) / - EXT3_INODES_PER_GROUP(inode->i_sb); + EXT4_INODES_PER_GROUP(inode->i_sb); inodes_per_buffer = bh->b_size / - EXT3_INODE_SIZE(inode->i_sb); + EXT4_INODE_SIZE(inode->i_sb); inode_offset = ((inode->i_ino - 1) % - EXT3_INODES_PER_GROUP(inode->i_sb)); + EXT4_INODES_PER_GROUP(inode->i_sb)); start = inode_offset & ~(inodes_per_buffer - 1); /* Is the inode bitmap in cache? */ - desc = ext3_get_group_desc(inode->i_sb, + desc = ext4_get_group_desc(inode->i_sb, block_group, NULL); if (!desc) goto make_io; @@ -2520,7 +2520,7 @@ static int __ext3_get_inode_loc(struct inode *inode, for (i = start; i < start + inodes_per_buffer; i++) { if (i == inode_offset) continue; - if (ext3_test_bit(i, bitmap_bh->b_data)) + if (ext4_test_bit(i, bitmap_bh->b_data)) break; } brelse(bitmap_bh); @@ -2544,7 +2544,7 @@ make_io: submit_bh(READ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { - ext3_error(inode->i_sb, "ext3_get_inode_loc", + ext4_error(inode->i_sb, "ext4_get_inode_loc", "unable to read inode block - " "inode=%lu, block="E3FSBLK, inode->i_ino, block); @@ -2557,48 +2557,48 @@ has_buffer: return 0; } -int ext3_get_inode_loc(struct inode *inode, struct ext3_iloc *iloc) +int ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc) { /* We have all inode data except xattrs in memory here. */ - return __ext3_get_inode_loc(inode, iloc, - !(EXT3_I(inode)->i_state & EXT3_STATE_XATTR)); + return __ext4_get_inode_loc(inode, iloc, + !(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)); } -void ext3_set_inode_flags(struct inode *inode) +void ext4_set_inode_flags(struct inode *inode) { - unsigned int flags = EXT3_I(inode)->i_flags; + unsigned int flags = EXT4_I(inode)->i_flags; inode->i_flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC); - if (flags & EXT3_SYNC_FL) + if (flags & EXT4_SYNC_FL) inode->i_flags |= S_SYNC; - if (flags & EXT3_APPEND_FL) + if (flags & EXT4_APPEND_FL) inode->i_flags |= S_APPEND; - if (flags & EXT3_IMMUTABLE_FL) + if (flags & EXT4_IMMUTABLE_FL) inode->i_flags |= S_IMMUTABLE; - if (flags & EXT3_NOATIME_FL) + if (flags & EXT4_NOATIME_FL) inode->i_flags |= S_NOATIME; - if (flags & EXT3_DIRSYNC_FL) + if (flags & EXT4_DIRSYNC_FL) inode->i_flags |= S_DIRSYNC; } -void ext3_read_inode(struct inode * inode) +void ext4_read_inode(struct inode * inode) { - struct ext3_iloc iloc; - struct ext3_inode *raw_inode; - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_iloc iloc; + struct ext4_inode *raw_inode; + struct ext4_inode_info *ei = EXT4_I(inode); struct buffer_head *bh; int block; -#ifdef CONFIG_EXT3_FS_POSIX_ACL - ei->i_acl = EXT3_ACL_NOT_CACHED; - ei->i_default_acl = EXT3_ACL_NOT_CACHED; +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + ei->i_acl = EXT4_ACL_NOT_CACHED; + ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif ei->i_block_alloc_info = NULL; - if (__ext3_get_inode_loc(inode, &iloc, 0)) + if (__ext4_get_inode_loc(inode, &iloc, 0)) goto bad_inode; bh = iloc.bh; - raw_inode = ext3_raw_inode(&iloc); + raw_inode = ext4_raw_inode(&iloc); inode->i_mode = le16_to_cpu(raw_inode->i_mode); inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low); @@ -2623,7 +2623,7 @@ void ext3_read_inode(struct inode * inode) */ if (inode->i_nlink == 0) { if (inode->i_mode == 0 || - !(EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ORPHAN_FS)) { + !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) { /* this inode is deleted */ brelse (bh); goto bad_inode; @@ -2635,7 +2635,7 @@ void ext3_read_inode(struct inode * inode) } inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); ei->i_flags = le32_to_cpu(raw_inode->i_flags); -#ifdef EXT3_FRAGMENTS +#ifdef EXT4_FRAGMENTS ei->i_faddr = le32_to_cpu(raw_inode->i_faddr); ei->i_frag_no = raw_inode->i_frag; ei->i_frag_size = raw_inode->i_fsize; @@ -2654,51 +2654,51 @@ void ext3_read_inode(struct inode * inode) * NOTE! The in-memory inode i_data array is in little-endian order * even on big-endian machines: we do NOT byteswap the block numbers! */ - for (block = 0; block < EXT3_N_BLOCKS; block++) + for (block = 0; block < EXT4_N_BLOCKS; block++) ei->i_data[block] = raw_inode->i_block[block]; INIT_LIST_HEAD(&ei->i_orphan); - if (inode->i_ino >= EXT3_FIRST_INO(inode->i_sb) + 1 && - EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) { + if (inode->i_ino >= EXT4_FIRST_INO(inode->i_sb) + 1 && + EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) { /* * When mke2fs creates big inodes it does not zero out - * the unused bytes above EXT3_GOOD_OLD_INODE_SIZE, + * the unused bytes above EXT4_GOOD_OLD_INODE_SIZE, * so ignore those first few inodes. */ ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); - if (EXT3_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > - EXT3_INODE_SIZE(inode->i_sb)) + if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > + EXT4_INODE_SIZE(inode->i_sb)) goto bad_inode; if (ei->i_extra_isize == 0) { /* The extra space is currently unused. Use it. */ - ei->i_extra_isize = sizeof(struct ext3_inode) - - EXT3_GOOD_OLD_INODE_SIZE; + ei->i_extra_isize = sizeof(struct ext4_inode) - + EXT4_GOOD_OLD_INODE_SIZE; } else { __le32 *magic = (void *)raw_inode + - EXT3_GOOD_OLD_INODE_SIZE + + EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize; - if (*magic == cpu_to_le32(EXT3_XATTR_MAGIC)) - ei->i_state |= EXT3_STATE_XATTR; + if (*magic == cpu_to_le32(EXT4_XATTR_MAGIC)) + ei->i_state |= EXT4_STATE_XATTR; } } else ei->i_extra_isize = 0; if (S_ISREG(inode->i_mode)) { - inode->i_op = &ext3_file_inode_operations; - inode->i_fop = &ext3_file_operations; - ext3_set_aops(inode); + inode->i_op = &ext4_file_inode_operations; + inode->i_fop = &ext4_file_operations; + ext4_set_aops(inode); } else if (S_ISDIR(inode->i_mode)) { - inode->i_op = &ext3_dir_inode_operations; - inode->i_fop = &ext3_dir_operations; + inode->i_op = &ext4_dir_inode_operations; + inode->i_fop = &ext4_dir_operations; } else if (S_ISLNK(inode->i_mode)) { - if (ext3_inode_is_fast_symlink(inode)) - inode->i_op = &ext3_fast_symlink_inode_operations; + if (ext4_inode_is_fast_symlink(inode)) + inode->i_op = &ext4_fast_symlink_inode_operations; else { - inode->i_op = &ext3_symlink_inode_operations; - ext3_set_aops(inode); + inode->i_op = &ext4_symlink_inode_operations; + ext4_set_aops(inode); } } else { - inode->i_op = &ext3_special_inode_operations; + inode->i_op = &ext4_special_inode_operations; if (raw_inode->i_block[0]) init_special_inode(inode, inode->i_mode, old_decode_dev(le32_to_cpu(raw_inode->i_block[0]))); @@ -2707,7 +2707,7 @@ void ext3_read_inode(struct inode * inode) new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); } brelse (iloc.bh); - ext3_set_inode_flags(inode); + ext4_set_inode_flags(inode); return; bad_inode: @@ -2722,19 +2722,19 @@ bad_inode: * * The caller must have write access to iloc->bh. */ -static int ext3_do_update_inode(handle_t *handle, +static int ext4_do_update_inode(handle_t *handle, struct inode *inode, - struct ext3_iloc *iloc) + struct ext4_iloc *iloc) { - struct ext3_inode *raw_inode = ext3_raw_inode(iloc); - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode *raw_inode = ext4_raw_inode(iloc); + struct ext4_inode_info *ei = EXT4_I(inode); struct buffer_head *bh = iloc->bh; int err = 0, rc, block; /* For fields not not tracking in the in-memory inode, * initialise them to zero for new inodes. */ - if (ei->i_state & EXT3_STATE_NEW) - memset(raw_inode, 0, EXT3_SB(inode->i_sb)->s_inode_size); + if (ei->i_state & EXT4_STATE_NEW) + memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size); raw_inode->i_mode = cpu_to_le16(inode->i_mode); if(!(test_opt(inode->i_sb, NO_UID32))) { @@ -2769,7 +2769,7 @@ static int ext3_do_update_inode(handle_t *handle, raw_inode->i_blocks = cpu_to_le32(inode->i_blocks); raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); raw_inode->i_flags = cpu_to_le32(ei->i_flags); -#ifdef EXT3_FRAGMENTS +#ifdef EXT4_FRAGMENTS raw_inode->i_faddr = cpu_to_le32(ei->i_faddr); raw_inode->i_frag = ei->i_frag_no; raw_inode->i_fsize = ei->i_frag_size; @@ -2782,24 +2782,24 @@ static int ext3_do_update_inode(handle_t *handle, cpu_to_le32(ei->i_disksize >> 32); if (ei->i_disksize > 0x7fffffffULL) { struct super_block *sb = inode->i_sb; - if (!EXT3_HAS_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_LARGE_FILE) || - EXT3_SB(sb)->s_es->s_rev_level == - cpu_to_le32(EXT3_GOOD_OLD_REV)) { + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_LARGE_FILE) || + EXT4_SB(sb)->s_es->s_rev_level == + cpu_to_le32(EXT4_GOOD_OLD_REV)) { /* If this is the first large file * created, add a flag to the superblock. */ - err = ext3_journal_get_write_access(handle, - EXT3_SB(sb)->s_sbh); + err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh); if (err) goto out_brelse; - ext3_update_dynamic_rev(sb); - EXT3_SET_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_LARGE_FILE); + ext4_update_dynamic_rev(sb); + EXT4_SET_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_LARGE_FILE); sb->s_dirt = 1; handle->h_sync = 1; - err = ext3_journal_dirty_metadata(handle, - EXT3_SB(sb)->s_sbh); + err = ext4_journal_dirty_metadata(handle, + EXT4_SB(sb)->s_sbh); } } } @@ -2815,26 +2815,26 @@ static int ext3_do_update_inode(handle_t *handle, cpu_to_le32(new_encode_dev(inode->i_rdev)); raw_inode->i_block[2] = 0; } - } else for (block = 0; block < EXT3_N_BLOCKS; block++) + } else for (block = 0; block < EXT4_N_BLOCKS; block++) raw_inode->i_block[block] = ei->i_data[block]; if (ei->i_extra_isize) raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize); - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - rc = ext3_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + rc = ext4_journal_dirty_metadata(handle, bh); if (!err) err = rc; - ei->i_state &= ~EXT3_STATE_NEW; + ei->i_state &= ~EXT4_STATE_NEW; out_brelse: brelse (bh); - ext3_std_error(inode->i_sb, err); + ext4_std_error(inode->i_sb, err); return err; } /* - * ext3_write_inode() + * ext4_write_inode() * * We are called from a few places: * @@ -2851,7 +2851,7 @@ out_brelse: * * In all cases it is actually safe for us to return without doing anything, * because the inode has been copied into a raw inode buffer in - * ext3_mark_inode_dirty(). This is a correctness thing for O_SYNC and for + * ext4_mark_inode_dirty(). This is a correctness thing for O_SYNC and for * knfsd. * * Note that we are absolutely dependent upon all inode dirtiers doing the @@ -2868,12 +2868,12 @@ out_brelse: * `stuff()' is running, and the new i_size will be lost. Plus the inode * will no longer be on the superblock's dirty inode list. */ -int ext3_write_inode(struct inode *inode, int wait) +int ext4_write_inode(struct inode *inode, int wait) { if (current->flags & PF_MEMALLOC) return 0; - if (ext3_journal_current_handle()) { + if (ext4_journal_current_handle()) { jbd_debug(0, "called recursively, non-PF_MEMALLOC!\n"); dump_stack(); return -EIO; @@ -2882,11 +2882,11 @@ int ext3_write_inode(struct inode *inode, int wait) if (!wait) return 0; - return ext3_force_commit(inode->i_sb); + return ext4_force_commit(inode->i_sb); } /* - * ext3_setattr() + * ext4_setattr() * * Called from notify_change. * @@ -2902,7 +2902,7 @@ int ext3_write_inode(struct inode *inode, int wait) * * Called with inode->sem down. */ -int ext3_setattr(struct dentry *dentry, struct iattr *attr) +int ext4_setattr(struct dentry *dentry, struct iattr *attr) { struct inode *inode = dentry->d_inode; int error, rc = 0; @@ -2918,15 +2918,15 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr) /* (user+group)*(old+new) structure, inode write (sb, * inode block, ? - but truncate inode update has it) */ - handle = ext3_journal_start(inode, 2*(EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)+ - EXT3_QUOTA_DEL_BLOCKS(inode->i_sb))+3); + handle = ext4_journal_start(inode, 2*(EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)+ + EXT4_QUOTA_DEL_BLOCKS(inode->i_sb))+3); if (IS_ERR(handle)) { error = PTR_ERR(handle); goto err_out; } error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0; if (error) { - ext3_journal_stop(handle); + ext4_journal_stop(handle); return error; } /* Update corresponding info in inode so that everything is in @@ -2935,41 +2935,41 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr) inode->i_uid = attr->ia_uid; if (attr->ia_valid & ATTR_GID) inode->i_gid = attr->ia_gid; - error = ext3_mark_inode_dirty(handle, inode); - ext3_journal_stop(handle); + error = ext4_mark_inode_dirty(handle, inode); + ext4_journal_stop(handle); } if (S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE && attr->ia_size < inode->i_size) { handle_t *handle; - handle = ext3_journal_start(inode, 3); + handle = ext4_journal_start(inode, 3); if (IS_ERR(handle)) { error = PTR_ERR(handle); goto err_out; } - error = ext3_orphan_add(handle, inode); - EXT3_I(inode)->i_disksize = attr->ia_size; - rc = ext3_mark_inode_dirty(handle, inode); + error = ext4_orphan_add(handle, inode); + EXT4_I(inode)->i_disksize = attr->ia_size; + rc = ext4_mark_inode_dirty(handle, inode); if (!error) error = rc; - ext3_journal_stop(handle); + ext4_journal_stop(handle); } rc = inode_setattr(inode, attr); - /* If inode_setattr's call to ext3_truncate failed to get a + /* If inode_setattr's call to ext4_truncate failed to get a * transaction handle at all, we need to clean up the in-core * orphan list manually. */ if (inode->i_nlink) - ext3_orphan_del(NULL, inode); + ext4_orphan_del(NULL, inode); if (!rc && (ia_valid & ATTR_MODE)) - rc = ext3_acl_chmod(inode); + rc = ext4_acl_chmod(inode); err_out: - ext3_std_error(inode->i_sb, error); + ext4_std_error(inode->i_sb, error); if (!error) error = rc; return error; @@ -2988,9 +2988,9 @@ err_out: * N+5 group descriptor summary blocks * 1 inode block * 1 superblock. - * 2 * EXT3_SINGLEDATA_TRANS_BLOCKS for the quote files + * 2 * EXT4_SINGLEDATA_TRANS_BLOCKS for the quote files * - * 3 * (N + 5) + 2 + 2 * EXT3_SINGLEDATA_TRANS_BLOCKS + * 3 * (N + 5) + 2 + 2 * EXT4_SINGLEDATA_TRANS_BLOCKS * * With ordered or writeback data it's the same, less the N data blocks. * @@ -3003,13 +3003,13 @@ err_out: * block and work out the exact number of indirects which are touched. Pah. */ -static int ext3_writepage_trans_blocks(struct inode *inode) +static int ext4_writepage_trans_blocks(struct inode *inode) { - int bpp = ext3_journal_blocks_per_page(inode); - int indirects = (EXT3_NDIR_BLOCKS % bpp) ? 5 : 3; + int bpp = ext4_journal_blocks_per_page(inode); + int indirects = (EXT4_NDIR_BLOCKS % bpp) ? 5 : 3; int ret; - if (ext3_should_journal_data(inode)) + if (ext4_should_journal_data(inode)) ret = 3 * (bpp + indirects) + 2; else ret = 2 * (bpp + indirects) + 2; @@ -3017,26 +3017,26 @@ static int ext3_writepage_trans_blocks(struct inode *inode) #ifdef CONFIG_QUOTA /* We know that structure was already allocated during DQUOT_INIT so * we will be updating only the data blocks + inodes */ - ret += 2*EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb); + ret += 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); #endif return ret; } /* - * The caller must have previously called ext3_reserve_inode_write(). + * The caller must have previously called ext4_reserve_inode_write(). * Give this, we know that the caller already has write access to iloc->bh. */ -int ext3_mark_iloc_dirty(handle_t *handle, - struct inode *inode, struct ext3_iloc *iloc) +int ext4_mark_iloc_dirty(handle_t *handle, + struct inode *inode, struct ext4_iloc *iloc) { int err = 0; /* the do_update_inode consumes one bh->b_count */ get_bh(iloc->bh); - /* ext3_do_update_inode() does journal_dirty_metadata */ - err = ext3_do_update_inode(handle, inode, iloc); + /* ext4_do_update_inode() does journal_dirty_metadata */ + err = ext4_do_update_inode(handle, inode, iloc); put_bh(iloc->bh); return err; } @@ -3047,22 +3047,22 @@ int ext3_mark_iloc_dirty(handle_t *handle, */ int -ext3_reserve_inode_write(handle_t *handle, struct inode *inode, - struct ext3_iloc *iloc) +ext4_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext4_iloc *iloc) { int err = 0; if (handle) { - err = ext3_get_inode_loc(inode, iloc); + err = ext4_get_inode_loc(inode, iloc); if (!err) { BUFFER_TRACE(iloc->bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, iloc->bh); + err = ext4_journal_get_write_access(handle, iloc->bh); if (err) { brelse(iloc->bh); iloc->bh = NULL; } } } - ext3_std_error(inode->i_sb, err); + ext4_std_error(inode->i_sb, err); return err; } @@ -3087,20 +3087,20 @@ ext3_reserve_inode_write(handle_t *handle, struct inode *inode, * to do a write_super() to free up some memory. It has the desired * effect. */ -int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode) +int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) { - struct ext3_iloc iloc; + struct ext4_iloc iloc; int err; might_sleep(); - err = ext3_reserve_inode_write(handle, inode, &iloc); + err = ext4_reserve_inode_write(handle, inode, &iloc); if (!err) - err = ext3_mark_iloc_dirty(handle, inode, &iloc); + err = ext4_mark_iloc_dirty(handle, inode, &iloc); return err; } /* - * ext3_dirty_inode() is called from __mark_inode_dirty() + * ext4_dirty_inode() is called from __mark_inode_dirty() * * We're really interested in the case where a file is being extended. * i_size has been changed by generic_commit_write() and we thus need @@ -3113,12 +3113,12 @@ int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode) * so would cause a commit on atime updates, which we don't bother doing. * We handle synchronous inodes at the highest possible level. */ -void ext3_dirty_inode(struct inode *inode) +void ext4_dirty_inode(struct inode *inode) { - handle_t *current_handle = ext3_journal_current_handle(); + handle_t *current_handle = ext4_journal_current_handle(); handle_t *handle; - handle = ext3_journal_start(inode, 2); + handle = ext4_journal_start(inode, 2); if (IS_ERR(handle)) goto out; if (current_handle && @@ -3129,9 +3129,9 @@ void ext3_dirty_inode(struct inode *inode) } else { jbd_debug(5, "marking dirty. outer handle=%p\n", current_handle); - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); } - ext3_journal_stop(handle); + ext4_journal_stop(handle); out: return; } @@ -3140,32 +3140,32 @@ out: /* * Bind an inode's backing buffer_head into this transaction, to prevent * it from being flushed to disk early. Unlike - * ext3_reserve_inode_write, this leaves behind no bh reference and + * ext4_reserve_inode_write, this leaves behind no bh reference and * returns no iloc structure, so the caller needs to repeat the iloc * lookup to mark the inode dirty later. */ -static int ext3_pin_inode(handle_t *handle, struct inode *inode) +static int ext4_pin_inode(handle_t *handle, struct inode *inode) { - struct ext3_iloc iloc; + struct ext4_iloc iloc; int err = 0; if (handle) { - err = ext3_get_inode_loc(inode, &iloc); + err = ext4_get_inode_loc(inode, &iloc); if (!err) { BUFFER_TRACE(iloc.bh, "get_write_access"); err = journal_get_write_access(handle, iloc.bh); if (!err) - err = ext3_journal_dirty_metadata(handle, + err = ext4_journal_dirty_metadata(handle, iloc.bh); brelse(iloc.bh); } } - ext3_std_error(inode->i_sb, err); + ext4_std_error(inode->i_sb, err); return err; } #endif -int ext3_change_inode_journal_flag(struct inode *inode, int val) +int ext4_change_inode_journal_flag(struct inode *inode, int val) { journal_t *journal; handle_t *handle; @@ -3181,7 +3181,7 @@ int ext3_change_inode_journal_flag(struct inode *inode, int val) * nobody is changing anything. */ - journal = EXT3_JOURNAL(inode); + journal = EXT4_JOURNAL(inode); if (is_journal_aborted(journal) || IS_RDONLY(inode)) return -EROFS; @@ -3197,23 +3197,23 @@ int ext3_change_inode_journal_flag(struct inode *inode, int val) */ if (val) - EXT3_I(inode)->i_flags |= EXT3_JOURNAL_DATA_FL; + EXT4_I(inode)->i_flags |= EXT4_JOURNAL_DATA_FL; else - EXT3_I(inode)->i_flags &= ~EXT3_JOURNAL_DATA_FL; - ext3_set_aops(inode); + EXT4_I(inode)->i_flags &= ~EXT4_JOURNAL_DATA_FL; + ext4_set_aops(inode); journal_unlock_updates(journal); /* Finally we can mark the inode as dirty. */ - handle = ext3_journal_start(inode, 1); + handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) return PTR_ERR(handle); - err = ext3_mark_inode_dirty(handle, inode); + err = ext4_mark_inode_dirty(handle, inode); handle->h_sync = 1; - ext3_journal_stop(handle); - ext3_std_error(inode->i_sb, err); + ext4_journal_stop(handle); + ext4_std_error(inode->i_sb, err); return err; } diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 12daa68..a567af1 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/ioctl.c + * linux/fs/ext4/ioctl.c * * Copyright (C) 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -10,30 +10,30 @@ #include #include #include -#include -#include +#include +#include #include #include #include #include -int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, +int ext4_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, unsigned long arg) { - struct ext3_inode_info *ei = EXT3_I(inode); + struct ext4_inode_info *ei = EXT4_I(inode); unsigned int flags; unsigned short rsv_window_size; - ext3_debug ("cmd = %u, arg = %lu\n", cmd, arg); + ext4_debug ("cmd = %u, arg = %lu\n", cmd, arg); switch (cmd) { - case EXT3_IOC_GETFLAGS: - flags = ei->i_flags & EXT3_FL_USER_VISIBLE; + case EXT4_IOC_GETFLAGS: + flags = ei->i_flags & EXT4_FL_USER_VISIBLE; return put_user(flags, (int __user *) arg); - case EXT3_IOC_SETFLAGS: { + case EXT4_IOC_SETFLAGS: { handle_t *handle = NULL; int err; - struct ext3_iloc iloc; + struct ext4_iloc iloc; unsigned int oldflags; unsigned int jflag; @@ -47,13 +47,13 @@ int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, return -EFAULT; if (!S_ISDIR(inode->i_mode)) - flags &= ~EXT3_DIRSYNC_FL; + flags &= ~EXT4_DIRSYNC_FL; mutex_lock(&inode->i_mutex); oldflags = ei->i_flags; /* The JOURNAL_DATA flag is modifiable only by root */ - jflag = flags & EXT3_JOURNAL_DATA_FL; + jflag = flags & EXT4_JOURNAL_DATA_FL; /* * The IMMUTABLE and APPEND_ONLY flags can only be changed by @@ -61,7 +61,7 @@ int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, * * This test looks nicer. Thanks to Pauline Middelink */ - if ((flags ^ oldflags) & (EXT3_APPEND_FL | EXT3_IMMUTABLE_FL)) { + if ((flags ^ oldflags) & (EXT4_APPEND_FL | EXT4_IMMUTABLE_FL)) { if (!capable(CAP_LINUX_IMMUTABLE)) { mutex_unlock(&inode->i_mutex); return -EPERM; @@ -72,7 +72,7 @@ int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, * The JOURNAL_DATA flag can only be changed by * the relevant capability. */ - if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) { + if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) { if (!capable(CAP_SYS_RESOURCE)) { mutex_unlock(&inode->i_mutex); return -EPERM; @@ -80,44 +80,44 @@ int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, } - handle = ext3_journal_start(inode, 1); + handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) { mutex_unlock(&inode->i_mutex); return PTR_ERR(handle); } if (IS_SYNC(inode)) handle->h_sync = 1; - err = ext3_reserve_inode_write(handle, inode, &iloc); + err = ext4_reserve_inode_write(handle, inode, &iloc); if (err) goto flags_err; - flags = flags & EXT3_FL_USER_MODIFIABLE; - flags |= oldflags & ~EXT3_FL_USER_MODIFIABLE; + flags = flags & EXT4_FL_USER_MODIFIABLE; + flags |= oldflags & ~EXT4_FL_USER_MODIFIABLE; ei->i_flags = flags; - ext3_set_inode_flags(inode); + ext4_set_inode_flags(inode); inode->i_ctime = CURRENT_TIME_SEC; - err = ext3_mark_iloc_dirty(handle, inode, &iloc); + err = ext4_mark_iloc_dirty(handle, inode, &iloc); flags_err: - ext3_journal_stop(handle); + ext4_journal_stop(handle); if (err) { mutex_unlock(&inode->i_mutex); return err; } - if ((jflag ^ oldflags) & (EXT3_JOURNAL_DATA_FL)) - err = ext3_change_inode_journal_flag(inode, jflag); + if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) + err = ext4_change_inode_journal_flag(inode, jflag); mutex_unlock(&inode->i_mutex); return err; } - case EXT3_IOC_GETVERSION: - case EXT3_IOC_GETVERSION_OLD: + case EXT4_IOC_GETVERSION: + case EXT4_IOC_GETVERSION_OLD: return put_user(inode->i_generation, (int __user *) arg); - case EXT3_IOC_SETVERSION: - case EXT3_IOC_SETVERSION_OLD: { + case EXT4_IOC_SETVERSION: + case EXT4_IOC_SETVERSION_OLD: { handle_t *handle; - struct ext3_iloc iloc; + struct ext4_iloc iloc; __u32 generation; int err; @@ -128,20 +128,20 @@ flags_err: if (get_user(generation, (int __user *) arg)) return -EFAULT; - handle = ext3_journal_start(inode, 1); + handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) return PTR_ERR(handle); - err = ext3_reserve_inode_write(handle, inode, &iloc); + err = ext4_reserve_inode_write(handle, inode, &iloc); if (err == 0) { inode->i_ctime = CURRENT_TIME_SEC; inode->i_generation = generation; - err = ext3_mark_iloc_dirty(handle, inode, &iloc); + err = ext4_mark_iloc_dirty(handle, inode, &iloc); } - ext3_journal_stop(handle); + ext4_journal_stop(handle); return err; } #ifdef CONFIG_JBD_DEBUG - case EXT3_IOC_WAIT_FOR_READONLY: + case EXT4_IOC_WAIT_FOR_READONLY: /* * This is racy - by the time we're woken up and running, * the superblock could be released. And the module could @@ -155,16 +155,16 @@ flags_err: int ret = 0; set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&EXT3_SB(sb)->ro_wait_queue, &wait); - if (timer_pending(&EXT3_SB(sb)->turn_ro_timer)) { + add_wait_queue(&EXT4_SB(sb)->ro_wait_queue, &wait); + if (timer_pending(&EXT4_SB(sb)->turn_ro_timer)) { schedule(); ret = 1; } - remove_wait_queue(&EXT3_SB(sb)->ro_wait_queue, &wait); + remove_wait_queue(&EXT4_SB(sb)->ro_wait_queue, &wait); return ret; } #endif - case EXT3_IOC_GETRSVSZ: + case EXT4_IOC_GETRSVSZ: if (test_opt(inode->i_sb, RESERVATION) && S_ISREG(inode->i_mode) && ei->i_block_alloc_info) { @@ -172,7 +172,7 @@ flags_err: return put_user(rsv_window_size, (int __user *)arg); } return -ENOTTY; - case EXT3_IOC_SETRSVSZ: { + case EXT4_IOC_SETRSVSZ: { if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) return -ENOTTY; @@ -186,8 +186,8 @@ flags_err: if (get_user(rsv_window_size, (int __user *)arg)) return -EFAULT; - if (rsv_window_size > EXT3_MAX_RESERVE_BLOCKS) - rsv_window_size = EXT3_MAX_RESERVE_BLOCKS; + if (rsv_window_size > EXT4_MAX_RESERVE_BLOCKS) + rsv_window_size = EXT4_MAX_RESERVE_BLOCKS; /* * need to allocate reservation structure for this inode @@ -195,17 +195,17 @@ flags_err: */ mutex_lock(&ei->truncate_mutex); if (!ei->i_block_alloc_info) - ext3_init_block_alloc_info(inode); + ext4_init_block_alloc_info(inode); if (ei->i_block_alloc_info){ - struct ext3_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; + struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; rsv->rsv_goal_size = rsv_window_size; } mutex_unlock(&ei->truncate_mutex); return 0; } - case EXT3_IOC_GROUP_EXTEND: { - ext3_fsblk_t n_blocks_count; + case EXT4_IOC_GROUP_EXTEND: { + ext4_fsblk_t n_blocks_count; struct super_block *sb = inode->i_sb; int err; @@ -218,15 +218,15 @@ flags_err: if (get_user(n_blocks_count, (__u32 __user *)arg)) return -EFAULT; - err = ext3_group_extend(sb, EXT3_SB(sb)->s_es, n_blocks_count); - journal_lock_updates(EXT3_SB(sb)->s_journal); - journal_flush(EXT3_SB(sb)->s_journal); - journal_unlock_updates(EXT3_SB(sb)->s_journal); + err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count); + journal_lock_updates(EXT4_SB(sb)->s_journal); + journal_flush(EXT4_SB(sb)->s_journal); + journal_unlock_updates(EXT4_SB(sb)->s_journal); return err; } - case EXT3_IOC_GROUP_ADD: { - struct ext3_new_group_data input; + case EXT4_IOC_GROUP_ADD: { + struct ext4_new_group_data input; struct super_block *sb = inode->i_sb; int err; @@ -236,14 +236,14 @@ flags_err: if (IS_RDONLY(inode)) return -EROFS; - if (copy_from_user(&input, (struct ext3_new_group_input __user *)arg, + if (copy_from_user(&input, (struct ext4_new_group_input __user *)arg, sizeof(input))) return -EFAULT; - err = ext3_group_add(sb, &input); - journal_lock_updates(EXT3_SB(sb)->s_journal); - journal_flush(EXT3_SB(sb)->s_journal); - journal_unlock_updates(EXT3_SB(sb)->s_journal); + err = ext4_group_add(sb, &input); + journal_lock_updates(EXT4_SB(sb)->s_journal); + journal_flush(EXT4_SB(sb)->s_journal); + journal_unlock_updates(EXT4_SB(sb)->s_journal); return err; } @@ -255,52 +255,52 @@ flags_err: } #ifdef CONFIG_COMPAT -long ext3_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct inode *inode = file->f_dentry->d_inode; int ret; /* These are just misnamed, they actually get/put from/to user an int */ switch (cmd) { - case EXT3_IOC32_GETFLAGS: - cmd = EXT3_IOC_GETFLAGS; + case EXT4_IOC32_GETFLAGS: + cmd = EXT4_IOC_GETFLAGS; break; - case EXT3_IOC32_SETFLAGS: - cmd = EXT3_IOC_SETFLAGS; + case EXT4_IOC32_SETFLAGS: + cmd = EXT4_IOC_SETFLAGS; break; - case EXT3_IOC32_GETVERSION: - cmd = EXT3_IOC_GETVERSION; + case EXT4_IOC32_GETVERSION: + cmd = EXT4_IOC_GETVERSION; break; - case EXT3_IOC32_SETVERSION: - cmd = EXT3_IOC_SETVERSION; + case EXT4_IOC32_SETVERSION: + cmd = EXT4_IOC_SETVERSION; break; - case EXT3_IOC32_GROUP_EXTEND: - cmd = EXT3_IOC_GROUP_EXTEND; + case EXT4_IOC32_GROUP_EXTEND: + cmd = EXT4_IOC_GROUP_EXTEND; break; - case EXT3_IOC32_GETVERSION_OLD: - cmd = EXT3_IOC_GETVERSION_OLD; + case EXT4_IOC32_GETVERSION_OLD: + cmd = EXT4_IOC_GETVERSION_OLD; break; - case EXT3_IOC32_SETVERSION_OLD: - cmd = EXT3_IOC_SETVERSION_OLD; + case EXT4_IOC32_SETVERSION_OLD: + cmd = EXT4_IOC_SETVERSION_OLD; break; #ifdef CONFIG_JBD_DEBUG - case EXT3_IOC32_WAIT_FOR_READONLY: - cmd = EXT3_IOC_WAIT_FOR_READONLY; + case EXT4_IOC32_WAIT_FOR_READONLY: + cmd = EXT4_IOC_WAIT_FOR_READONLY; break; #endif - case EXT3_IOC32_GETRSVSZ: - cmd = EXT3_IOC_GETRSVSZ; + case EXT4_IOC32_GETRSVSZ: + cmd = EXT4_IOC_GETRSVSZ; break; - case EXT3_IOC32_SETRSVSZ: - cmd = EXT3_IOC_SETRSVSZ; + case EXT4_IOC32_SETRSVSZ: + cmd = EXT4_IOC_SETRSVSZ; break; - case EXT3_IOC_GROUP_ADD: + case EXT4_IOC_GROUP_ADD: break; default: return -ENOIOCTLCMD; } lock_kernel(); - ret = ext3_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg)); + ret = ext4_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg)); unlock_kernel(); return ret; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 906731a..956b381 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/namei.c + * linux/fs/ext4/namei.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -28,8 +28,8 @@ #include #include #include -#include -#include +#include +#include #include #include #include @@ -50,7 +50,7 @@ #define NAMEI_RA_SIZE (NAMEI_RA_CHUNKS * NAMEI_RA_BLOCKS) #define NAMEI_RA_INDEX(c,b) (((c) * NAMEI_RA_BLOCKS) + (b)) -static struct buffer_head *ext3_append(handle_t *handle, +static struct buffer_head *ext4_append(handle_t *handle, struct inode *inode, u32 *block, int *err) { @@ -58,10 +58,10 @@ static struct buffer_head *ext3_append(handle_t *handle, *block = inode->i_size >> inode->i_sb->s_blocksize_bits; - if ((bh = ext3_bread(handle, inode, *block, 1, err))) { + if ((bh = ext4_bread(handle, inode, *block, 1, err))) { inode->i_size += inode->i_sb->s_blocksize; - EXT3_I(inode)->i_disksize = inode->i_size; - ext3_journal_get_write_access(handle,bh); + EXT4_I(inode)->i_disksize = inode->i_size; + ext4_journal_get_write_access(handle,bh); } return bh; } @@ -144,7 +144,7 @@ struct dx_map_entry u32 offs; }; -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX static inline unsigned dx_get_block (struct dx_entry *entry); static void dx_set_block (struct dx_entry *entry, unsigned value); static inline unsigned dx_get_hash (struct dx_entry *entry); @@ -161,20 +161,20 @@ static struct dx_frame *dx_probe(struct dentry *dentry, struct dx_frame *frame, int *err); static void dx_release (struct dx_frame *frames); -static int dx_make_map (struct ext3_dir_entry_2 *de, int size, +static int dx_make_map (struct ext4_dir_entry_2 *de, int size, struct dx_hash_info *hinfo, struct dx_map_entry map[]); static void dx_sort_map(struct dx_map_entry *map, unsigned count); -static struct ext3_dir_entry_2 *dx_move_dirents (char *from, char *to, +static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to, struct dx_map_entry *offsets, int count); -static struct ext3_dir_entry_2* dx_pack_dirents (char *base, int size); +static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size); static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block); -static int ext3_htree_next_block(struct inode *dir, __u32 hash, +static int ext4_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames, __u32 *start_hash); -static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, - struct ext3_dir_entry_2 **res_dir, int *err); -static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, +static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, + struct ext4_dir_entry_2 **res_dir, int *err); +static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode); /* @@ -224,14 +224,14 @@ static inline void dx_set_limit (struct dx_entry *entries, unsigned value) static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize) { - unsigned entry_space = dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(1) - - EXT3_DIR_REC_LEN(2) - infosize; + unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1) - + EXT4_DIR_REC_LEN(2) - infosize; return 0? 20: entry_space / sizeof(struct dx_entry); } static inline unsigned dx_node_limit (struct inode *dir) { - unsigned entry_space = dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(0); + unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0); return 0? 22: entry_space / sizeof(struct dx_entry); } @@ -257,7 +257,7 @@ struct stats unsigned bcount; }; -static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext3_dir_entry_2 *de, +static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext4_dir_entry_2 *de, int size, int show_names) { unsigned names = 0, space = 0; @@ -274,14 +274,14 @@ static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext3_dir_ent int len = de->name_len; char *name = de->name; while (len--) printk("%c", *name++); - ext3fs_dirhash(de->name, de->name_len, &h); + ext4fs_dirhash(de->name, de->name_len, &h); printk(":%x.%u ", h.hash, ((char *) de - base)); } - space += EXT3_DIR_REC_LEN(de->name_len); + space += EXT4_DIR_REC_LEN(de->name_len); names++; } - de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); } printk("(%i)\n", names); return (struct stats) { names, space, 1 }; @@ -302,10 +302,10 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, u32 range = i < count - 1? (dx_get_hash(entries + 1) - hash): ~hash; struct stats stats; printk("%s%3u:%03u hash %8x/%8x ",levels?"":" ", i, block, hash, range); - if (!(bh = ext3_bread (NULL,dir, block, 0,&err))) continue; + if (!(bh = ext4_bread (NULL,dir, block, 0,&err))) continue; stats = levels? dx_show_entries(hinfo, dir, ((struct dx_node *) bh->b_data)->entries, levels - 1): - dx_show_leaf(hinfo, (struct ext3_dir_entry_2 *) bh->b_data, blocksize, 0); + dx_show_leaf(hinfo, (struct ext4_dir_entry_2 *) bh->b_data, blocksize, 0); names += stats.names; space += stats.space; bcount += stats.bcount; @@ -341,13 +341,13 @@ dx_probe(struct dentry *dentry, struct inode *dir, frame->bh = NULL; if (dentry) dir = dentry->d_parent->d_inode; - if (!(bh = ext3_bread (NULL,dir, 0, 0, err))) + if (!(bh = ext4_bread (NULL,dir, 0, 0, err))) goto fail; root = (struct dx_root *) bh->b_data; if (root->info.hash_version != DX_HASH_TEA && root->info.hash_version != DX_HASH_HALF_MD4 && root->info.hash_version != DX_HASH_LEGACY) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __FUNCTION__, "Unrecognised inode hash code %d", root->info.hash_version); brelse(bh); @@ -355,13 +355,13 @@ dx_probe(struct dentry *dentry, struct inode *dir, goto fail; } hinfo->hash_version = root->info.hash_version; - hinfo->seed = EXT3_SB(dir->i_sb)->s_hash_seed; + hinfo->seed = EXT4_SB(dir->i_sb)->s_hash_seed; if (dentry) - ext3fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo); + ext4fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo); hash = hinfo->hash; if (root->info.unused_flags & 1) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __FUNCTION__, "Unimplemented inode hash flags: %#06x", root->info.unused_flags); brelse(bh); @@ -370,7 +370,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, } if ((indirect = root->info.indirect_levels) > 1) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext4_warning(dir->i_sb, __FUNCTION__, "Unimplemented inode hash depth: %#06x", root->info.indirect_levels); brelse(bh); @@ -421,7 +421,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, frame->entries = entries; frame->at = at; if (!indirect--) return frame; - if (!(bh = ext3_bread (NULL,dir, dx_get_block(at), 0, err))) + if (!(bh = ext4_bread (NULL,dir, dx_get_block(at), 0, err))) goto fail2; at = entries = ((struct dx_node *) bh->b_data)->entries; assert (dx_get_limit(entries) == dx_node_limit (dir)); @@ -463,7 +463,7 @@ static void dx_release (struct dx_frame *frames) * If start_hash is non-null, it will be filled in with the starting * hash of the next page. */ -static int ext3_htree_next_block(struct inode *dir, __u32 hash, +static int ext4_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames, __u32 *start_hash) @@ -509,7 +509,7 @@ static int ext3_htree_next_block(struct inode *dir, __u32 hash, * block so no check is necessary */ while (num_frames--) { - if (!(bh = ext3_bread(NULL, dir, dx_get_block(p->at), + if (!(bh = ext4_bread(NULL, dir, dx_get_block(p->at), 0, &err))) return err; /* Failure */ p++; @@ -524,9 +524,9 @@ static int ext3_htree_next_block(struct inode *dir, __u32 hash, /* * p is at least 6 bytes before the end of page */ -static inline struct ext3_dir_entry_2 *ext3_next_entry(struct ext3_dir_entry_2 *p) +static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 *p) { - return (struct ext3_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len)); + return (struct ext4_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len)); } /* @@ -540,26 +540,26 @@ static int htree_dirblock_to_tree(struct file *dir_file, __u32 start_hash, __u32 start_minor_hash) { struct buffer_head *bh; - struct ext3_dir_entry_2 *de, *top; + struct ext4_dir_entry_2 *de, *top; int err, count = 0; dxtrace(printk("In htree dirblock_to_tree: block %d\n", block)); - if (!(bh = ext3_bread (NULL, dir, block, 0, &err))) + if (!(bh = ext4_bread (NULL, dir, block, 0, &err))) return err; - de = (struct ext3_dir_entry_2 *) bh->b_data; - top = (struct ext3_dir_entry_2 *) ((char *) de + + de = (struct ext4_dir_entry_2 *) bh->b_data; + top = (struct ext4_dir_entry_2 *) ((char *) de + dir->i_sb->s_blocksize - - EXT3_DIR_REC_LEN(0)); - for (; de < top; de = ext3_next_entry(de)) { - ext3fs_dirhash(de->name, de->name_len, hinfo); + EXT4_DIR_REC_LEN(0)); + for (; de < top; de = ext4_next_entry(de)) { + ext4fs_dirhash(de->name, de->name_len, hinfo); if ((hinfo->hash < start_hash) || ((hinfo->hash == start_hash) && (hinfo->minor_hash < start_minor_hash))) continue; if (de->inode == 0) continue; - if ((err = ext3_htree_store_dirent(dir_file, + if ((err = ext4_htree_store_dirent(dir_file, hinfo->hash, hinfo->minor_hash, de)) != 0) { brelse(bh); return err; @@ -579,11 +579,11 @@ static int htree_dirblock_to_tree(struct file *dir_file, * This function returns the number of entries inserted into the tree, * or a negative error code. */ -int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, +int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, __u32 start_minor_hash, __u32 *next_hash) { struct dx_hash_info hinfo; - struct ext3_dir_entry_2 *de; + struct ext4_dir_entry_2 *de; struct dx_frame frames[2], *frame; struct inode *dir; int block, err; @@ -594,9 +594,9 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, start_minor_hash)); dir = dir_file->f_dentry->d_inode; - if (!(EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) { - hinfo.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; - hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; + if (!(EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) { + hinfo.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; + hinfo.seed = EXT4_SB(dir->i_sb)->s_hash_seed; count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo, start_hash, start_minor_hash); *next_hash = ~0; @@ -610,15 +610,15 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, /* Add '.' and '..' from the htree header */ if (!start_hash && !start_minor_hash) { - de = (struct ext3_dir_entry_2 *) frames[0].bh->b_data; - if ((err = ext3_htree_store_dirent(dir_file, 0, 0, de)) != 0) + de = (struct ext4_dir_entry_2 *) frames[0].bh->b_data; + if ((err = ext4_htree_store_dirent(dir_file, 0, 0, de)) != 0) goto errout; count++; } if (start_hash < 2 || (start_hash ==2 && start_minor_hash==0)) { - de = (struct ext3_dir_entry_2 *) frames[0].bh->b_data; - de = ext3_next_entry(de); - if ((err = ext3_htree_store_dirent(dir_file, 2, 0, de)) != 0) + de = (struct ext4_dir_entry_2 *) frames[0].bh->b_data; + de = ext4_next_entry(de); + if ((err = ext4_htree_store_dirent(dir_file, 2, 0, de)) != 0) goto errout; count++; } @@ -633,7 +633,7 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, } count += ret; hashval = ~0; - ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS, + ret = ext4_htree_next_block(dir, HASH_NB_ALWAYS, frame, frames, &hashval); *next_hash = hashval; if (ret < 0) { @@ -663,7 +663,7 @@ errout: * Directory block splitting, compacting */ -static int dx_make_map (struct ext3_dir_entry_2 *de, int size, +static int dx_make_map (struct ext4_dir_entry_2 *de, int size, struct dx_hash_info *hinfo, struct dx_map_entry *map_tail) { int count = 0; @@ -673,7 +673,7 @@ static int dx_make_map (struct ext3_dir_entry_2 *de, int size, while ((char *) de < base + size) { if (de->name_len && de->inode) { - ext3fs_dirhash(de->name, de->name_len, &h); + ext4fs_dirhash(de->name, de->name_len, &h); map_tail--; map_tail->hash = h.hash; map_tail->offs = (u32) ((char *) de - base); @@ -681,7 +681,7 @@ static int dx_make_map (struct ext3_dir_entry_2 *de, int size, cond_resched(); } /* XXX: do we need to check rec_len == 0 case? -Chris */ - de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); } return count; } @@ -730,21 +730,21 @@ static void dx_insert_block(struct dx_frame *frame, u32 hash, u32 block) #endif -static void ext3_update_dx_flag(struct inode *inode) +static void ext4_update_dx_flag(struct inode *inode) { - if (!EXT3_HAS_COMPAT_FEATURE(inode->i_sb, - EXT3_FEATURE_COMPAT_DIR_INDEX)) - EXT3_I(inode)->i_flags &= ~EXT3_INDEX_FL; + if (!EXT4_HAS_COMPAT_FEATURE(inode->i_sb, + EXT4_FEATURE_COMPAT_DIR_INDEX)) + EXT4_I(inode)->i_flags &= ~EXT4_INDEX_FL; } /* - * NOTE! unlike strncmp, ext3_match returns 1 for success, 0 for failure. + * NOTE! unlike strncmp, ext4_match returns 1 for success, 0 for failure. * - * `len <= EXT3_NAME_LEN' is guaranteed by caller. + * `len <= EXT4_NAME_LEN' is guaranteed by caller. * `de != NULL' is guaranteed by caller. */ -static inline int ext3_match (int len, const char * const name, - struct ext3_dir_entry_2 * de) +static inline int ext4_match (int len, const char * const name, + struct ext4_dir_entry_2 * de) { if (len != de->name_len) return 0; @@ -760,24 +760,24 @@ static inline int search_dirblock(struct buffer_head * bh, struct inode *dir, struct dentry *dentry, unsigned long offset, - struct ext3_dir_entry_2 ** res_dir) + struct ext4_dir_entry_2 ** res_dir) { - struct ext3_dir_entry_2 * de; + struct ext4_dir_entry_2 * de; char * dlimit; int de_len; const char *name = dentry->d_name.name; int namelen = dentry->d_name.len; - de = (struct ext3_dir_entry_2 *) bh->b_data; + de = (struct ext4_dir_entry_2 *) bh->b_data; dlimit = bh->b_data + dir->i_sb->s_blocksize; while ((char *) de < dlimit) { /* this code is executed quadratically often */ /* do minimal checking `by hand' */ if ((char *) de + namelen <= dlimit && - ext3_match (namelen, name, de)) { + ext4_match (namelen, name, de)) { /* found a match - just to be sure, do a full check */ - if (!ext3_check_dir_entry("ext3_find_entry", + if (!ext4_check_dir_entry("ext4_find_entry", dir, de, bh, offset)) return -1; *res_dir = de; @@ -788,14 +788,14 @@ static inline int search_dirblock(struct buffer_head * bh, if (de_len <= 0) return -1; offset += de_len; - de = (struct ext3_dir_entry_2 *) ((char *) de + de_len); + de = (struct ext4_dir_entry_2 *) ((char *) de + de_len); } return 0; } /* - * ext3_find_entry() + * ext4_find_entry() * * finds an entry in the specified directory with the wanted name. It * returns the cache buffer in which the entry was found, and the entry @@ -805,8 +805,8 @@ static inline int search_dirblock(struct buffer_head * bh, * The returned buffer_head has ->b_count elevated. The caller is expected * to brelse() it when appropriate. */ -static struct buffer_head * ext3_find_entry (struct dentry *dentry, - struct ext3_dir_entry_2 ** res_dir) +static struct buffer_head * ext4_find_entry (struct dentry *dentry, + struct ext4_dir_entry_2 ** res_dir) { struct super_block * sb; struct buffer_head * bh_use[NAMEI_RA_SIZE]; @@ -828,11 +828,11 @@ static struct buffer_head * ext3_find_entry (struct dentry *dentry, blocksize = sb->s_blocksize; namelen = dentry->d_name.len; name = dentry->d_name.name; - if (namelen > EXT3_NAME_LEN) + if (namelen > EXT4_NAME_LEN) return NULL; -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX if (is_dx(dir)) { - bh = ext3_dx_find_entry(dentry, res_dir, &err); + bh = ext4_dx_find_entry(dentry, res_dir, &err); /* * On success, or if the error was file not found, * return. Otherwise, fall back to doing a search the @@ -840,11 +840,11 @@ static struct buffer_head * ext3_find_entry (struct dentry *dentry, */ if (bh || (err != ERR_BAD_DX_DIR)) return bh; - dxtrace(printk("ext3_find_entry: dx failed, falling back\n")); + dxtrace(printk("ext4_find_entry: dx failed, falling back\n")); } #endif - nblocks = dir->i_size >> EXT3_BLOCK_SIZE_BITS(sb); - start = EXT3_I(dir)->i_dir_start_lookup; + nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); + start = EXT4_I(dir)->i_dir_start_lookup; if (start >= nblocks) start = 0; block = start; @@ -868,7 +868,7 @@ restart: break; } num++; - bh = ext3_getblk(NULL, dir, b++, 0, &err); + bh = ext4_getblk(NULL, dir, b++, 0, &err); bh_use[ra_max] = bh; if (bh) ll_rw_block(READ_META, 1, &bh); @@ -879,15 +879,15 @@ restart: wait_on_buffer(bh); if (!buffer_uptodate(bh)) { /* read error, skip block & hope for the best */ - ext3_error(sb, __FUNCTION__, "reading directory #%lu " + ext4_error(sb, __FUNCTION__, "reading directory #%lu " "offset %lu", dir->i_ino, block); brelse(bh); goto next; } i = search_dirblock(bh, dir, dentry, - block << EXT3_BLOCK_SIZE_BITS(sb), res_dir); + block << EXT4_BLOCK_SIZE_BITS(sb), res_dir); if (i == 1) { - EXT3_I(dir)->i_dir_start_lookup = block; + EXT4_I(dir)->i_dir_start_lookup = block; ret = bh; goto cleanup_and_exit; } else { @@ -905,7 +905,7 @@ restart: * search the last part of the directory before giving up. */ block = nblocks; - nblocks = dir->i_size >> EXT3_BLOCK_SIZE_BITS(sb); + nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); if (block < nblocks) { start = 0; goto restart; @@ -918,15 +918,15 @@ cleanup_and_exit: return ret; } -#ifdef CONFIG_EXT3_INDEX -static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, - struct ext3_dir_entry_2 **res_dir, int *err) +#ifdef CONFIG_EXT4_INDEX +static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, + struct ext4_dir_entry_2 **res_dir, int *err) { struct super_block * sb; struct dx_hash_info hinfo; u32 hash; struct dx_frame frames[2], *frame; - struct ext3_dir_entry_2 *de, *top; + struct ext4_dir_entry_2 *de, *top; struct buffer_head *bh; unsigned long block; int retval; @@ -948,16 +948,16 @@ static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, hash = hinfo.hash; do { block = dx_get_block(frame->at); - if (!(bh = ext3_bread (NULL,dir, block, 0, err))) + if (!(bh = ext4_bread (NULL,dir, block, 0, err))) goto errout; - de = (struct ext3_dir_entry_2 *) bh->b_data; - top = (struct ext3_dir_entry_2 *) ((char *) de + sb->s_blocksize - - EXT3_DIR_REC_LEN(0)); - for (; de < top; de = ext3_next_entry(de)) - if (ext3_match (namelen, name, de)) { - if (!ext3_check_dir_entry("ext3_find_entry", + de = (struct ext4_dir_entry_2 *) bh->b_data; + top = (struct ext4_dir_entry_2 *) ((char *) de + sb->s_blocksize - + EXT4_DIR_REC_LEN(0)); + for (; de < top; de = ext4_next_entry(de)) + if (ext4_match (namelen, name, de)) { + if (!ext4_check_dir_entry("ext4_find_entry", dir, de, bh, - (block<b_data))) { brelse (bh); goto errout; @@ -968,10 +968,10 @@ static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, } brelse (bh); /* Check to see if we should continue to search */ - retval = ext3_htree_next_block(dir, hash, frame, + retval = ext4_htree_next_block(dir, hash, frame, frames, NULL); if (retval < 0) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "error reading index page in directory #%lu", dir->i_ino); *err = retval; @@ -987,22 +987,22 @@ errout: } #endif -static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ext4_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) { struct inode * inode; - struct ext3_dir_entry_2 * de; + struct ext4_dir_entry_2 * de; struct buffer_head * bh; - if (dentry->d_name.len > EXT3_NAME_LEN) + if (dentry->d_name.len > EXT4_NAME_LEN) return ERR_PTR(-ENAMETOOLONG); - bh = ext3_find_entry(dentry, &de); + bh = ext4_find_entry(dentry, &de); inode = NULL; if (bh) { unsigned long ino = le32_to_cpu(de->inode); brelse (bh); - if (!ext3_valid_inum(dir->i_sb, ino)) { - ext3_error(dir->i_sb, "ext3_lookup", + if (!ext4_valid_inum(dir->i_sb, ino)) { + ext4_error(dir->i_sb, "ext4_lookup", "bad inode number: %lu", ino); inode = NULL; } else @@ -1015,28 +1015,28 @@ static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, str } -struct dentry *ext3_get_parent(struct dentry *child) +struct dentry *ext4_get_parent(struct dentry *child) { unsigned long ino; struct dentry *parent; struct inode *inode; struct dentry dotdot; - struct ext3_dir_entry_2 * de; + struct ext4_dir_entry_2 * de; struct buffer_head *bh; dotdot.d_name.name = ".."; dotdot.d_name.len = 2; dotdot.d_parent = child; /* confusing, isn't it! */ - bh = ext3_find_entry(&dotdot, &de); + bh = ext4_find_entry(&dotdot, &de); inode = NULL; if (!bh) return ERR_PTR(-ENOENT); ino = le32_to_cpu(de->inode); brelse(bh); - if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { - ext3_error(child->d_inode->i_sb, "ext3_get_parent", + if (!ext4_valid_inum(child->d_inode->i_sb, ino)) { + ext4_error(child->d_inode->i_sb, "ext4_get_parent", "bad inode number: %lu", ino); inode = NULL; } else @@ -1054,65 +1054,65 @@ struct dentry *ext3_get_parent(struct dentry *child) } #define S_SHIFT 12 -static unsigned char ext3_type_by_mode[S_IFMT >> S_SHIFT] = { - [S_IFREG >> S_SHIFT] = EXT3_FT_REG_FILE, - [S_IFDIR >> S_SHIFT] = EXT3_FT_DIR, - [S_IFCHR >> S_SHIFT] = EXT3_FT_CHRDEV, - [S_IFBLK >> S_SHIFT] = EXT3_FT_BLKDEV, - [S_IFIFO >> S_SHIFT] = EXT3_FT_FIFO, - [S_IFSOCK >> S_SHIFT] = EXT3_FT_SOCK, - [S_IFLNK >> S_SHIFT] = EXT3_FT_SYMLINK, +static unsigned char ext4_type_by_mode[S_IFMT >> S_SHIFT] = { + [S_IFREG >> S_SHIFT] = EXT4_FT_REG_FILE, + [S_IFDIR >> S_SHIFT] = EXT4_FT_DIR, + [S_IFCHR >> S_SHIFT] = EXT4_FT_CHRDEV, + [S_IFBLK >> S_SHIFT] = EXT4_FT_BLKDEV, + [S_IFIFO >> S_SHIFT] = EXT4_FT_FIFO, + [S_IFSOCK >> S_SHIFT] = EXT4_FT_SOCK, + [S_IFLNK >> S_SHIFT] = EXT4_FT_SYMLINK, }; -static inline void ext3_set_de_type(struct super_block *sb, - struct ext3_dir_entry_2 *de, +static inline void ext4_set_de_type(struct super_block *sb, + struct ext4_dir_entry_2 *de, umode_t mode) { - if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_FILETYPE)) - de->file_type = ext3_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FILETYPE)) + de->file_type = ext4_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; } -#ifdef CONFIG_EXT3_INDEX -static struct ext3_dir_entry_2 * +#ifdef CONFIG_EXT4_INDEX +static struct ext4_dir_entry_2 * dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count) { unsigned rec_len = 0; while (count--) { - struct ext3_dir_entry_2 *de = (struct ext3_dir_entry_2 *) (from + map->offs); - rec_len = EXT3_DIR_REC_LEN(de->name_len); + struct ext4_dir_entry_2 *de = (struct ext4_dir_entry_2 *) (from + map->offs); + rec_len = EXT4_DIR_REC_LEN(de->name_len); memcpy (to, de, rec_len); - ((struct ext3_dir_entry_2 *) to)->rec_len = + ((struct ext4_dir_entry_2 *) to)->rec_len = cpu_to_le16(rec_len); de->inode = 0; map++; to += rec_len; } - return (struct ext3_dir_entry_2 *) (to - rec_len); + return (struct ext4_dir_entry_2 *) (to - rec_len); } -static struct ext3_dir_entry_2* dx_pack_dirents(char *base, int size) +static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size) { - struct ext3_dir_entry_2 *next, *to, *prev, *de = (struct ext3_dir_entry_2 *) base; + struct ext4_dir_entry_2 *next, *to, *prev, *de = (struct ext4_dir_entry_2 *) base; unsigned rec_len = 0; prev = to = de; while ((char*)de < base + size) { - next = (struct ext3_dir_entry_2 *) ((char *) de + + next = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); if (de->inode && de->name_len) { - rec_len = EXT3_DIR_REC_LEN(de->name_len); + rec_len = EXT4_DIR_REC_LEN(de->name_len); if (de > to) memmove(to, de, rec_len); to->rec_len = cpu_to_le16(rec_len); prev = to; - to = (struct ext3_dir_entry_2 *) (((char *) to) + rec_len); + to = (struct ext4_dir_entry_2 *) (((char *) to) + rec_len); } de = next; } return prev; } -static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, +static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, struct buffer_head **bh,struct dx_frame *frame, struct dx_hash_info *hinfo, int *error) { @@ -1124,10 +1124,10 @@ static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, struct dx_map_entry *map; char *data1 = (*bh)->b_data, *data2; unsigned split; - struct ext3_dir_entry_2 *de = NULL, *de2; + struct ext4_dir_entry_2 *de = NULL, *de2; int err; - bh2 = ext3_append (handle, dir, &newblock, error); + bh2 = ext4_append (handle, dir, &newblock, error); if (!(bh2)) { brelse(*bh); *bh = NULL; @@ -1135,17 +1135,17 @@ static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, } BUFFER_TRACE(*bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, *bh); + err = ext4_journal_get_write_access(handle, *bh); if (err) { journal_error: brelse(*bh); brelse(bh2); *bh = NULL; - ext3_std_error(dir->i_sb, err); + ext4_std_error(dir->i_sb, err); goto errout; } BUFFER_TRACE(frame->bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, frame->bh); + err = ext4_journal_get_write_access(handle, frame->bh); if (err) goto journal_error; @@ -1153,7 +1153,7 @@ static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, /* create map in the end of data2 block */ map = (struct dx_map_entry *) (data2 + blocksize); - count = dx_make_map ((struct ext3_dir_entry_2 *) data1, + count = dx_make_map ((struct ext4_dir_entry_2 *) data1, blocksize, hinfo, map); map -= count; split = count/2; // need to adjust to actual middle @@ -1168,8 +1168,8 @@ static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, de = dx_pack_dirents(data1,blocksize); de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); de2->rec_len = cpu_to_le16(data2 + blocksize - (char *) de2); - dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data1, blocksize, 1)); - dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data2, blocksize, 1)); + dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1)); + dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data2, blocksize, 1)); /* Which block gets the new entry? */ if (hinfo->hash >= hash2) @@ -1178,10 +1178,10 @@ static struct ext3_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, de = de2; } dx_insert_block (frame, hash2 + continued, newblock); - err = ext3_journal_dirty_metadata (handle, bh2); + err = ext4_journal_dirty_metadata (handle, bh2); if (err) goto journal_error; - err = ext3_journal_dirty_metadata (handle, frame->bh); + err = ext4_journal_dirty_metadata (handle, frame->bh); if (err) goto journal_error; brelse (bh2); @@ -1204,7 +1204,7 @@ errout: * all other cases bh is released. */ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, - struct inode *inode, struct ext3_dir_entry_2 *de, + struct inode *inode, struct ext4_dir_entry_2 *de, struct buffer_head * bh) { struct inode *dir = dentry->d_parent->d_inode; @@ -1215,51 +1215,51 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, int nlen, rlen, err; char *top; - reclen = EXT3_DIR_REC_LEN(namelen); + reclen = EXT4_DIR_REC_LEN(namelen); if (!de) { - de = (struct ext3_dir_entry_2 *)bh->b_data; + de = (struct ext4_dir_entry_2 *)bh->b_data; top = bh->b_data + dir->i_sb->s_blocksize - reclen; while ((char *) de <= top) { - if (!ext3_check_dir_entry("ext3_add_entry", dir, de, + if (!ext4_check_dir_entry("ext4_add_entry", dir, de, bh, offset)) { brelse (bh); return -EIO; } - if (ext3_match (namelen, name, de)) { + if (ext4_match (namelen, name, de)) { brelse (bh); return -EEXIST; } - nlen = EXT3_DIR_REC_LEN(de->name_len); + nlen = EXT4_DIR_REC_LEN(de->name_len); rlen = le16_to_cpu(de->rec_len); if ((de->inode? rlen - nlen: rlen) >= reclen) break; - de = (struct ext3_dir_entry_2 *)((char *)de + rlen); + de = (struct ext4_dir_entry_2 *)((char *)de + rlen); offset += rlen; } if ((char *) de > top) return -ENOSPC; } BUFFER_TRACE(bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, bh); + err = ext4_journal_get_write_access(handle, bh); if (err) { - ext3_std_error(dir->i_sb, err); + ext4_std_error(dir->i_sb, err); brelse(bh); return err; } /* By now the buffer is marked for journaling */ - nlen = EXT3_DIR_REC_LEN(de->name_len); + nlen = EXT4_DIR_REC_LEN(de->name_len); rlen = le16_to_cpu(de->rec_len); if (de->inode) { - struct ext3_dir_entry_2 *de1 = (struct ext3_dir_entry_2 *)((char *)de + nlen); + struct ext4_dir_entry_2 *de1 = (struct ext4_dir_entry_2 *)((char *)de + nlen); de1->rec_len = cpu_to_le16(rlen - nlen); de->rec_len = cpu_to_le16(nlen); de = de1; } - de->file_type = EXT3_FT_UNKNOWN; + de->file_type = EXT4_FT_UNKNOWN; if (inode) { de->inode = cpu_to_le32(inode->i_ino); - ext3_set_de_type(dir->i_sb, de, inode->i_mode); + ext4_set_de_type(dir->i_sb, de, inode->i_mode); } else de->inode = 0; de->name_len = namelen; @@ -1270,24 +1270,24 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, * on this. * * XXX similarly, too many callers depend on - * ext3_new_inode() setting the times, but error + * ext4_new_inode() setting the times, but error * recovery deletes the inode, so the worst that can * happen is that the times are slightly out of date * and/or different from the directory change time. */ dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC; - ext3_update_dx_flag(dir); + ext4_update_dx_flag(dir); dir->i_version++; - ext3_mark_inode_dirty(handle, dir); - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - err = ext3_journal_dirty_metadata(handle, bh); + ext4_mark_inode_dirty(handle, dir); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh); if (err) - ext3_std_error(dir->i_sb, err); + ext4_std_error(dir->i_sb, err); brelse(bh); return 0; } -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX /* * This converts a one block unindexed directory to a 3 block indexed * directory, and adds the dentry to the indexed directory. @@ -1302,7 +1302,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, struct dx_root *root; struct dx_frame frames[2], *frame; struct dx_entry *entries; - struct ext3_dir_entry_2 *de, *de2; + struct ext4_dir_entry_2 *de, *de2; char *data1, *top; unsigned len; int retval; @@ -1313,38 +1313,38 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, blocksize = dir->i_sb->s_blocksize; dxtrace(printk("Creating index\n")); - retval = ext3_journal_get_write_access(handle, bh); + retval = ext4_journal_get_write_access(handle, bh); if (retval) { - ext3_std_error(dir->i_sb, retval); + ext4_std_error(dir->i_sb, retval); brelse(bh); return retval; } root = (struct dx_root *) bh->b_data; - bh2 = ext3_append (handle, dir, &block, &retval); + bh2 = ext4_append (handle, dir, &block, &retval); if (!(bh2)) { brelse(bh); return retval; } - EXT3_I(dir)->i_flags |= EXT3_INDEX_FL; + EXT4_I(dir)->i_flags |= EXT4_INDEX_FL; data1 = bh2->b_data; /* The 0th block becomes the root, move the dirents out */ fde = &root->dotdot; - de = (struct ext3_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len)); + de = (struct ext4_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len)); len = ((char *) root) + blocksize - (char *) de; memcpy (data1, de, len); - de = (struct ext3_dir_entry_2 *) data1; + de = (struct ext4_dir_entry_2 *) data1; top = data1 + len; while ((char *)(de2=(void*)de+le16_to_cpu(de->rec_len)) < top) de = de2; de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); /* Initialize the root; the dot dirents already exist */ - de = (struct ext3_dir_entry_2 *) (&root->dotdot); - de->rec_len = cpu_to_le16(blocksize - EXT3_DIR_REC_LEN(2)); + de = (struct ext4_dir_entry_2 *) (&root->dotdot); + de->rec_len = cpu_to_le16(blocksize - EXT4_DIR_REC_LEN(2)); memset (&root->info, 0, sizeof(root->info)); root->info.info_length = sizeof(root->info); - root->info.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; + root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; entries = root->entries; dx_set_block (entries, 1); dx_set_count (entries, 1); @@ -1352,8 +1352,8 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, /* Initialize as for dx_probe */ hinfo.hash_version = root->info.hash_version; - hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; - ext3fs_dirhash(name, namelen, &hinfo); + hinfo.seed = EXT4_SB(dir->i_sb)->s_hash_seed; + ext4fs_dirhash(name, namelen, &hinfo); frame = frames; frame->entries = entries; frame->at = entries; @@ -1369,25 +1369,25 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, #endif /* - * ext3_add_entry() + * ext4_add_entry() * * adds a file entry to the specified directory, using the same - * semantics as ext3_find_entry(). It returns NULL if it failed. + * semantics as ext4_find_entry(). It returns NULL if it failed. * * NOTE!! The inode part of 'de' is left at 0 - which means you * may not sleep between calling this and putting something into * the entry, as someone else might have used it while you slept. */ -static int ext3_add_entry (handle_t *handle, struct dentry *dentry, +static int ext4_add_entry (handle_t *handle, struct dentry *dentry, struct inode *inode) { struct inode *dir = dentry->d_parent->d_inode; unsigned long offset; struct buffer_head * bh; - struct ext3_dir_entry_2 *de; + struct ext4_dir_entry_2 *de; struct super_block * sb; int retval; -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX int dx_fallback=0; #endif unsigned blocksize; @@ -1397,46 +1397,46 @@ static int ext3_add_entry (handle_t *handle, struct dentry *dentry, blocksize = sb->s_blocksize; if (!dentry->d_name.len) return -EINVAL; -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX if (is_dx(dir)) { - retval = ext3_dx_add_entry(handle, dentry, inode); + retval = ext4_dx_add_entry(handle, dentry, inode); if (!retval || (retval != ERR_BAD_DX_DIR)) return retval; - EXT3_I(dir)->i_flags &= ~EXT3_INDEX_FL; + EXT4_I(dir)->i_flags &= ~EXT4_INDEX_FL; dx_fallback++; - ext3_mark_inode_dirty(handle, dir); + ext4_mark_inode_dirty(handle, dir); } #endif blocks = dir->i_size >> sb->s_blocksize_bits; for (block = 0, offset = 0; block < blocks; block++) { - bh = ext3_bread(handle, dir, block, 0, &retval); + bh = ext4_bread(handle, dir, block, 0, &retval); if(!bh) return retval; retval = add_dirent_to_buf(handle, dentry, inode, NULL, bh); if (retval != -ENOSPC) return retval; -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX if (blocks == 1 && !dx_fallback && - EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_DIR_INDEX)) + EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX)) return make_indexed_dir(handle, dentry, inode, bh); #endif brelse(bh); } - bh = ext3_append(handle, dir, &block, &retval); + bh = ext4_append(handle, dir, &block, &retval); if (!bh) return retval; - de = (struct ext3_dir_entry_2 *) bh->b_data; + de = (struct ext4_dir_entry_2 *) bh->b_data; de->inode = 0; de->rec_len = cpu_to_le16(blocksize); return add_dirent_to_buf(handle, dentry, inode, de, bh); } -#ifdef CONFIG_EXT3_INDEX +#ifdef CONFIG_EXT4_INDEX /* * Returns 0 for success, or a negative error value */ -static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, +static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode) { struct dx_frame frames[2], *frame; @@ -1445,7 +1445,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, struct buffer_head * bh; struct inode *dir = dentry->d_parent->d_inode; struct super_block * sb = dir->i_sb; - struct ext3_dir_entry_2 *de; + struct ext4_dir_entry_2 *de; int err; frame = dx_probe(dentry, NULL, &hinfo, frames, &err); @@ -1454,11 +1454,11 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, entries = frame->entries; at = frame->at; - if (!(bh = ext3_bread(handle,dir, dx_get_block(frame->at), 0, &err))) + if (!(bh = ext4_bread(handle,dir, dx_get_block(frame->at), 0, &err))) goto cleanup; BUFFER_TRACE(bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, bh); + err = ext4_journal_get_write_access(handle, bh); if (err) goto journal_error; @@ -1482,12 +1482,12 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, if (levels && (dx_get_count(frames->entries) == dx_get_limit(frames->entries))) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Directory index full!"); err = -ENOSPC; goto cleanup; } - bh2 = ext3_append (handle, dir, &newblock, &err); + bh2 = ext4_append (handle, dir, &newblock, &err); if (!(bh2)) goto cleanup; node2 = (struct dx_node *)(bh2->b_data); @@ -1495,7 +1495,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, node2->fake.rec_len = cpu_to_le16(sb->s_blocksize); node2->fake.inode = 0; BUFFER_TRACE(frame->bh, "get_write_access"); - err = ext3_journal_get_write_access(handle, frame->bh); + err = ext4_journal_get_write_access(handle, frame->bh); if (err) goto journal_error; if (levels) { @@ -1504,7 +1504,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, dxtrace(printk("Split index %i/%i\n", icount1, icount2)); BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */ - err = ext3_journal_get_write_access(handle, + err = ext4_journal_get_write_access(handle, frames[0].bh); if (err) goto journal_error; @@ -1525,7 +1525,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, dxtrace(dx_show_index ("node", frames[1].entries)); dxtrace(dx_show_index ("node", ((struct dx_node *) bh2->b_data)->entries)); - err = ext3_journal_dirty_metadata(handle, bh2); + err = ext4_journal_dirty_metadata(handle, bh2); if (err) goto journal_error; brelse (bh2); @@ -1545,12 +1545,12 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, frame->at = at = at - entries + entries2; frame->entries = entries = entries2; frame->bh = bh2; - err = ext3_journal_get_write_access(handle, + err = ext4_journal_get_write_access(handle, frame->bh); if (err) goto journal_error; } - ext3_journal_dirty_metadata(handle, frames[0].bh); + ext4_journal_dirty_metadata(handle, frames[0].bh); } de = do_split(handle, dir, &bh, frame, &hinfo, &err); if (!de) @@ -1560,7 +1560,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, goto cleanup; journal_error: - ext3_std_error(dir->i_sb, err); + ext4_std_error(dir->i_sb, err); cleanup: if (bh) brelse(bh); @@ -1570,26 +1570,26 @@ cleanup: #endif /* - * ext3_delete_entry deletes a directory entry by merging it with the + * ext4_delete_entry deletes a directory entry by merging it with the * previous entry */ -static int ext3_delete_entry (handle_t *handle, +static int ext4_delete_entry (handle_t *handle, struct inode * dir, - struct ext3_dir_entry_2 * de_del, + struct ext4_dir_entry_2 * de_del, struct buffer_head * bh) { - struct ext3_dir_entry_2 * de, * pde; + struct ext4_dir_entry_2 * de, * pde; int i; i = 0; pde = NULL; - de = (struct ext3_dir_entry_2 *) bh->b_data; + de = (struct ext4_dir_entry_2 *) bh->b_data; while (i < bh->b_size) { - if (!ext3_check_dir_entry("ext3_delete_entry", dir, de, bh, i)) + if (!ext4_check_dir_entry("ext4_delete_entry", dir, de, bh, i)) return -EIO; if (de == de_del) { BUFFER_TRACE(bh, "get_write_access"); - ext3_journal_get_write_access(handle, bh); + ext4_journal_get_write_access(handle, bh); if (pde) pde->rec_len = cpu_to_le16(le16_to_cpu(pde->rec_len) + @@ -1597,43 +1597,43 @@ static int ext3_delete_entry (handle_t *handle, else de->inode = 0; dir->i_version++; - BUFFER_TRACE(bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, bh); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, bh); return 0; } i += le16_to_cpu(de->rec_len); pde = de; - de = (struct ext3_dir_entry_2 *) + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); } return -ENOENT; } /* - * ext3_mark_inode_dirty is somewhat expensive, so unlike ext2 we + * ext4_mark_inode_dirty is somewhat expensive, so unlike ext2 we * do not perform it in these functions. We perform it at the call site, * if it is needed. */ -static inline void ext3_inc_count(handle_t *handle, struct inode *inode) +static inline void ext4_inc_count(handle_t *handle, struct inode *inode) { inc_nlink(inode); } -static inline void ext3_dec_count(handle_t *handle, struct inode *inode) +static inline void ext4_dec_count(handle_t *handle, struct inode *inode) { drop_nlink(inode); } -static int ext3_add_nondir(handle_t *handle, +static int ext4_add_nondir(handle_t *handle, struct dentry *dentry, struct inode *inode) { - int err = ext3_add_entry(handle, dentry, inode); + int err = ext4_add_entry(handle, dentry, inode); if (!err) { - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); d_instantiate(dentry, inode); return 0; } - ext3_dec_count(handle, inode); + ext4_dec_count(handle, inode); iput(inode); return err; } @@ -1646,7 +1646,7 @@ static int ext3_add_nondir(handle_t *handle, * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, +static int ext4_create (struct inode * dir, struct dentry * dentry, int mode, struct nameidata *nd) { handle_t *handle; @@ -1654,30 +1654,30 @@ static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, int err, retries = 0; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext3_new_inode (handle, dir, mode); + inode = ext4_new_inode (handle, dir, mode); err = PTR_ERR(inode); if (!IS_ERR(inode)) { - inode->i_op = &ext3_file_inode_operations; - inode->i_fop = &ext3_file_operations; - ext3_set_aops(inode); - err = ext3_add_nondir(handle, dentry, inode); + inode->i_op = &ext4_file_inode_operations; + inode->i_fop = &ext4_file_operations; + ext4_set_aops(inode); + err = ext4_add_nondir(handle, dentry, inode); } - ext3_journal_stop(handle); - if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } -static int ext3_mknod (struct inode * dir, struct dentry *dentry, +static int ext4_mknod (struct inode * dir, struct dentry *dentry, int mode, dev_t rdev) { handle_t *handle; @@ -1688,100 +1688,100 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry, return -EINVAL; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext3_new_inode (handle, dir, mode); + inode = ext4_new_inode (handle, dir, mode); err = PTR_ERR(inode); if (!IS_ERR(inode)) { init_special_inode(inode, inode->i_mode, rdev); -#ifdef CONFIG_EXT3_FS_XATTR - inode->i_op = &ext3_special_inode_operations; +#ifdef CONFIG_EXT4DEV_FS_XATTR + inode->i_op = &ext4_special_inode_operations; #endif - err = ext3_add_nondir(handle, dentry, inode); + err = ext4_add_nondir(handle, dentry, inode); } - ext3_journal_stop(handle); - if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } -static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ext4_mkdir(struct inode * dir, struct dentry * dentry, int mode) { handle_t *handle; struct inode * inode; struct buffer_head * dir_block; - struct ext3_dir_entry_2 * de; + struct ext4_dir_entry_2 * de; int err, retries = 0; - if (dir->i_nlink >= EXT3_LINK_MAX) + if (dir->i_nlink >= EXT4_LINK_MAX) return -EMLINK; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext3_new_inode (handle, dir, S_IFDIR | mode); + inode = ext4_new_inode (handle, dir, S_IFDIR | mode); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; - inode->i_op = &ext3_dir_inode_operations; - inode->i_fop = &ext3_dir_operations; - inode->i_size = EXT3_I(inode)->i_disksize = inode->i_sb->s_blocksize; - dir_block = ext3_bread (handle, inode, 0, 1, &err); + inode->i_op = &ext4_dir_inode_operations; + inode->i_fop = &ext4_dir_operations; + inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize; + dir_block = ext4_bread (handle, inode, 0, 1, &err); if (!dir_block) { drop_nlink(inode); /* is this nlink == 0? */ - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; } BUFFER_TRACE(dir_block, "get_write_access"); - ext3_journal_get_write_access(handle, dir_block); - de = (struct ext3_dir_entry_2 *) dir_block->b_data; + ext4_journal_get_write_access(handle, dir_block); + de = (struct ext4_dir_entry_2 *) dir_block->b_data; de->inode = cpu_to_le32(inode->i_ino); de->name_len = 1; - de->rec_len = cpu_to_le16(EXT3_DIR_REC_LEN(de->name_len)); + de->rec_len = cpu_to_le16(EXT4_DIR_REC_LEN(de->name_len)); strcpy (de->name, "."); - ext3_set_de_type(dir->i_sb, de, S_IFDIR); - de = (struct ext3_dir_entry_2 *) + ext4_set_de_type(dir->i_sb, de, S_IFDIR); + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); de->inode = cpu_to_le32(dir->i_ino); - de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT3_DIR_REC_LEN(1)); + de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT4_DIR_REC_LEN(1)); de->name_len = 2; strcpy (de->name, ".."); - ext3_set_de_type(dir->i_sb, de, S_IFDIR); + ext4_set_de_type(dir->i_sb, de, S_IFDIR); inode->i_nlink = 2; - BUFFER_TRACE(dir_block, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, dir_block); + BUFFER_TRACE(dir_block, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, dir_block); brelse (dir_block); - ext3_mark_inode_dirty(handle, inode); - err = ext3_add_entry (handle, dentry, inode); + ext4_mark_inode_dirty(handle, inode); + err = ext4_add_entry (handle, dentry, inode); if (err) { inode->i_nlink = 0; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; } inc_nlink(dir); - ext3_update_dx_flag(dir); - ext3_mark_inode_dirty(handle, dir); + ext4_update_dx_flag(dir); + ext4_mark_inode_dirty(handle, dir); d_instantiate(dentry, inode); out_stop: - ext3_journal_stop(handle); - if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } @@ -1793,59 +1793,59 @@ static int empty_dir (struct inode * inode) { unsigned long offset; struct buffer_head * bh; - struct ext3_dir_entry_2 * de, * de1; + struct ext4_dir_entry_2 * de, * de1; struct super_block * sb; int err = 0; sb = inode->i_sb; - if (inode->i_size < EXT3_DIR_REC_LEN(1) + EXT3_DIR_REC_LEN(2) || - !(bh = ext3_bread (NULL, inode, 0, 0, &err))) { + if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) || + !(bh = ext4_bread (NULL, inode, 0, 0, &err))) { if (err) - ext3_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __FUNCTION__, "error %d reading directory #%lu offset 0", err, inode->i_ino); else - ext3_warning(inode->i_sb, __FUNCTION__, + ext4_warning(inode->i_sb, __FUNCTION__, "bad directory (dir #%lu) - no data block", inode->i_ino); return 1; } - de = (struct ext3_dir_entry_2 *) bh->b_data; - de1 = (struct ext3_dir_entry_2 *) + de = (struct ext4_dir_entry_2 *) bh->b_data; + de1 = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); if (le32_to_cpu(de->inode) != inode->i_ino || !le32_to_cpu(de1->inode) || strcmp (".", de->name) || strcmp ("..", de1->name)) { - ext3_warning (inode->i_sb, "empty_dir", + ext4_warning (inode->i_sb, "empty_dir", "bad directory (dir #%lu) - no `.' or `..'", inode->i_ino); brelse (bh); return 1; } offset = le16_to_cpu(de->rec_len) + le16_to_cpu(de1->rec_len); - de = (struct ext3_dir_entry_2 *) + de = (struct ext4_dir_entry_2 *) ((char *) de1 + le16_to_cpu(de1->rec_len)); while (offset < inode->i_size ) { if (!bh || (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { err = 0; brelse (bh); - bh = ext3_bread (NULL, inode, - offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err); + bh = ext4_bread (NULL, inode, + offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err); if (!bh) { if (err) - ext3_error(sb, __FUNCTION__, + ext4_error(sb, __FUNCTION__, "error %d reading directory" " #%lu offset %lu", err, inode->i_ino, offset); offset += sb->s_blocksize; continue; } - de = (struct ext3_dir_entry_2 *) bh->b_data; + de = (struct ext4_dir_entry_2 *) bh->b_data; } - if (!ext3_check_dir_entry("empty_dir", inode, de, bh, offset)) { - de = (struct ext3_dir_entry_2 *)(bh->b_data + + if (!ext4_check_dir_entry("empty_dir", inode, de, bh, offset)) { + de = (struct ext4_dir_entry_2 *)(bh->b_data + sb->s_blocksize); offset = (offset | (sb->s_blocksize - 1)) + 1; continue; @@ -1855,57 +1855,57 @@ static int empty_dir (struct inode * inode) return 0; } offset += le16_to_cpu(de->rec_len); - de = (struct ext3_dir_entry_2 *) + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); } brelse (bh); return 1; } -/* ext3_orphan_add() links an unlinked or truncated inode into a list of +/* ext4_orphan_add() links an unlinked or truncated inode into a list of * such inodes, starting at the superblock, in case we crash before the * file is closed/deleted, or in case the inode truncate spans multiple * transactions and the last transaction is not recovered after a crash. * * At filesystem recovery time, we walk this list deleting unlinked - * inodes and truncating linked inodes in ext3_orphan_cleanup(). + * inodes and truncating linked inodes in ext4_orphan_cleanup(). */ -int ext3_orphan_add(handle_t *handle, struct inode *inode) +int ext4_orphan_add(handle_t *handle, struct inode *inode) { struct super_block *sb = inode->i_sb; - struct ext3_iloc iloc; + struct ext4_iloc iloc; int err = 0, rc; lock_super(sb); - if (!list_empty(&EXT3_I(inode)->i_orphan)) + if (!list_empty(&EXT4_I(inode)->i_orphan)) goto out_unlock; /* Orphan handling is only valid for files with data blocks * being truncated, or files being unlinked. */ /* @@@ FIXME: Observation from aviro: - * I think I can trigger J_ASSERT in ext3_orphan_add(). We block - * here (on lock_super()), so race with ext3_link() which might bump + * I think I can trigger J_ASSERT in ext4_orphan_add(). We block + * here (on lock_super()), so race with ext4_link() which might bump * ->i_nlink. For, say it, character device. Not a regular file, * not a directory, not a symlink and ->i_nlink > 0. */ J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); - BUFFER_TRACE(EXT3_SB(sb)->s_sbh, "get_write_access"); - err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh); + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get_write_access"); + err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); if (err) goto out_unlock; - err = ext3_reserve_inode_write(handle, inode, &iloc); + err = ext4_reserve_inode_write(handle, inode, &iloc); if (err) goto out_unlock; /* Insert this inode at the head of the on-disk orphan list... */ - NEXT_ORPHAN(inode) = le32_to_cpu(EXT3_SB(sb)->s_es->s_last_orphan); - EXT3_SB(sb)->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); - err = ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); - rc = ext3_mark_iloc_dirty(handle, inode, &iloc); + NEXT_ORPHAN(inode) = le32_to_cpu(EXT4_SB(sb)->s_es->s_last_orphan); + EXT4_SB(sb)->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); + err = ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + rc = ext4_mark_iloc_dirty(handle, inode, &iloc); if (!err) err = rc; @@ -1918,28 +1918,28 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode) * This is safe: on error we're going to ignore the orphan list * anyway on the next recovery. */ if (!err) - list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan); + list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan); jbd_debug(4, "superblock will point to %lu\n", inode->i_ino); jbd_debug(4, "orphan inode %lu will point to %d\n", inode->i_ino, NEXT_ORPHAN(inode)); out_unlock: unlock_super(sb); - ext3_std_error(inode->i_sb, err); + ext4_std_error(inode->i_sb, err); return err; } /* - * ext3_orphan_del() removes an unlinked or truncated inode from the list + * ext4_orphan_del() removes an unlinked or truncated inode from the list * of such inodes stored on disk, because it is finally being cleaned up. */ -int ext3_orphan_del(handle_t *handle, struct inode *inode) +int ext4_orphan_del(handle_t *handle, struct inode *inode) { struct list_head *prev; - struct ext3_inode_info *ei = EXT3_I(inode); - struct ext3_sb_info *sbi; + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_sb_info *sbi; unsigned long ino_next; - struct ext3_iloc iloc; + struct ext4_iloc iloc; int err = 0; lock_super(inode->i_sb); @@ -1950,7 +1950,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode) ino_next = NEXT_ORPHAN(inode); prev = ei->i_orphan.prev; - sbi = EXT3_SB(inode->i_sb); + sbi = EXT4_SB(inode->i_sb); jbd_debug(4, "remove inode %lu from orphan list\n", inode->i_ino); @@ -1963,38 +1963,38 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode) if (!handle) goto out; - err = ext3_reserve_inode_write(handle, inode, &iloc); + err = ext4_reserve_inode_write(handle, inode, &iloc); if (err) goto out_err; if (prev == &sbi->s_orphan) { jbd_debug(4, "superblock will point to %lu\n", ino_next); BUFFER_TRACE(sbi->s_sbh, "get_write_access"); - err = ext3_journal_get_write_access(handle, sbi->s_sbh); + err = ext4_journal_get_write_access(handle, sbi->s_sbh); if (err) goto out_brelse; sbi->s_es->s_last_orphan = cpu_to_le32(ino_next); - err = ext3_journal_dirty_metadata(handle, sbi->s_sbh); + err = ext4_journal_dirty_metadata(handle, sbi->s_sbh); } else { - struct ext3_iloc iloc2; + struct ext4_iloc iloc2; struct inode *i_prev = - &list_entry(prev, struct ext3_inode_info, i_orphan)->vfs_inode; + &list_entry(prev, struct ext4_inode_info, i_orphan)->vfs_inode; jbd_debug(4, "orphan inode %lu will point to %lu\n", i_prev->i_ino, ino_next); - err = ext3_reserve_inode_write(handle, i_prev, &iloc2); + err = ext4_reserve_inode_write(handle, i_prev, &iloc2); if (err) goto out_brelse; NEXT_ORPHAN(i_prev) = ino_next; - err = ext3_mark_iloc_dirty(handle, i_prev, &iloc2); + err = ext4_mark_iloc_dirty(handle, i_prev, &iloc2); } if (err) goto out_brelse; NEXT_ORPHAN(inode) = 0; - err = ext3_mark_iloc_dirty(handle, inode, &iloc); + err = ext4_mark_iloc_dirty(handle, inode, &iloc); out_err: - ext3_std_error(inode->i_sb, err); + ext4_std_error(inode->i_sb, err); out: unlock_super(inode->i_sb); return err; @@ -2004,23 +2004,23 @@ out_brelse: goto out_err; } -static int ext3_rmdir (struct inode * dir, struct dentry *dentry) +static int ext4_rmdir (struct inode * dir, struct dentry *dentry) { int retval; struct inode * inode; struct buffer_head * bh; - struct ext3_dir_entry_2 * de; + struct ext4_dir_entry_2 * de; handle_t *handle; /* Initialize quotas before so that eventual writes go in * separate transaction */ DQUOT_INIT(dentry->d_inode); - handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, EXT4_DELETE_TRANS_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); retval = -ENOENT; - bh = ext3_find_entry (dentry, &de); + bh = ext4_find_entry (dentry, &de); if (!bh) goto end_rmdir; @@ -2037,11 +2037,11 @@ static int ext3_rmdir (struct inode * dir, struct dentry *dentry) if (!empty_dir (inode)) goto end_rmdir; - retval = ext3_delete_entry(handle, dir, de, bh); + retval = ext4_delete_entry(handle, dir, de, bh); if (retval) goto end_rmdir; if (inode->i_nlink != 2) - ext3_warning (inode->i_sb, "ext3_rmdir", + ext4_warning (inode->i_sb, "ext4_rmdir", "empty directory has nlink!=2 (%d)", inode->i_nlink); inode->i_version++; @@ -2050,31 +2050,31 @@ static int ext3_rmdir (struct inode * dir, struct dentry *dentry) * zero will ensure that the right thing happens during any * recovery. */ inode->i_size = 0; - ext3_orphan_add(handle, inode); + ext4_orphan_add(handle, inode); inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); drop_nlink(dir); - ext3_update_dx_flag(dir); - ext3_mark_inode_dirty(handle, dir); + ext4_update_dx_flag(dir); + ext4_mark_inode_dirty(handle, dir); end_rmdir: - ext3_journal_stop(handle); + ext4_journal_stop(handle); brelse (bh); return retval; } -static int ext3_unlink(struct inode * dir, struct dentry *dentry) +static int ext4_unlink(struct inode * dir, struct dentry *dentry) { int retval; struct inode * inode; struct buffer_head * bh; - struct ext3_dir_entry_2 * de; + struct ext4_dir_entry_2 * de; handle_t *handle; /* Initialize quotas before so that eventual writes go * in separate transaction */ DQUOT_INIT(dentry->d_inode); - handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, EXT4_DELETE_TRANS_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2082,7 +2082,7 @@ static int ext3_unlink(struct inode * dir, struct dentry *dentry) handle->h_sync = 1; retval = -ENOENT; - bh = ext3_find_entry (dentry, &de); + bh = ext4_find_entry (dentry, &de); if (!bh) goto end_unlink; @@ -2093,31 +2093,31 @@ static int ext3_unlink(struct inode * dir, struct dentry *dentry) goto end_unlink; if (!inode->i_nlink) { - ext3_warning (inode->i_sb, "ext3_unlink", + ext4_warning (inode->i_sb, "ext4_unlink", "Deleting nonexistent file (%lu), %d", inode->i_ino, inode->i_nlink); inode->i_nlink = 1; } - retval = ext3_delete_entry(handle, dir, de, bh); + retval = ext4_delete_entry(handle, dir, de, bh); if (retval) goto end_unlink; dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; - ext3_update_dx_flag(dir); - ext3_mark_inode_dirty(handle, dir); + ext4_update_dx_flag(dir); + ext4_mark_inode_dirty(handle, dir); drop_nlink(inode); if (!inode->i_nlink) - ext3_orphan_add(handle, inode); + ext4_orphan_add(handle, inode); inode->i_ctime = dir->i_ctime; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); retval = 0; end_unlink: - ext3_journal_stop(handle); + ext4_journal_stop(handle); brelse (bh); return retval; } -static int ext3_symlink (struct inode * dir, +static int ext4_symlink (struct inode * dir, struct dentry *dentry, const char * symname) { handle_t *handle; @@ -2129,63 +2129,63 @@ static int ext3_symlink (struct inode * dir, return -ENAMETOOLONG; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 5 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext3_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); + inode = ext4_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; - if (l > sizeof (EXT3_I(inode)->i_data)) { - inode->i_op = &ext3_symlink_inode_operations; - ext3_set_aops(inode); + if (l > sizeof (EXT4_I(inode)->i_data)) { + inode->i_op = &ext4_symlink_inode_operations; + ext4_set_aops(inode); /* - * page_symlink() calls into ext3_prepare/commit_write. + * page_symlink() calls into ext4_prepare/commit_write. * We have a transaction open. All is sweetness. It also sets * i_size in generic_commit_write(). */ err = __page_symlink(inode, symname, l, mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS); if (err) { - ext3_dec_count(handle, inode); - ext3_mark_inode_dirty(handle, inode); + ext4_dec_count(handle, inode); + ext4_mark_inode_dirty(handle, inode); iput (inode); goto out_stop; } } else { - inode->i_op = &ext3_fast_symlink_inode_operations; - memcpy((char*)&EXT3_I(inode)->i_data,symname,l); + inode->i_op = &ext4_fast_symlink_inode_operations; + memcpy((char*)&EXT4_I(inode)->i_data,symname,l); inode->i_size = l-1; } - EXT3_I(inode)->i_disksize = inode->i_size; - err = ext3_add_nondir(handle, dentry, inode); + EXT4_I(inode)->i_disksize = inode->i_size; + err = ext4_add_nondir(handle, dentry, inode); out_stop: - ext3_journal_stop(handle); - if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } -static int ext3_link (struct dentry * old_dentry, +static int ext4_link (struct dentry * old_dentry, struct inode * dir, struct dentry *dentry) { handle_t *handle; struct inode *inode = old_dentry->d_inode; int err, retries = 0; - if (inode->i_nlink >= EXT3_LINK_MAX) + if (inode->i_nlink >= EXT4_LINK_MAX) return -EMLINK; retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS); + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2193,31 +2193,31 @@ retry: handle->h_sync = 1; inode->i_ctime = CURRENT_TIME_SEC; - ext3_inc_count(handle, inode); + ext4_inc_count(handle, inode); atomic_inc(&inode->i_count); - err = ext3_add_nondir(handle, dentry, inode); - ext3_journal_stop(handle); - if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + err = ext4_add_nondir(handle, dentry, inode); + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; } #define PARENT_INO(buffer) \ - ((struct ext3_dir_entry_2 *) ((char *) buffer + \ - le16_to_cpu(((struct ext3_dir_entry_2 *) buffer)->rec_len)))->inode + ((struct ext4_dir_entry_2 *) ((char *) buffer + \ + le16_to_cpu(((struct ext4_dir_entry_2 *) buffer)->rec_len)))->inode /* * Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ -static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, +static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, struct inode * new_dir,struct dentry *new_dentry) { handle_t *handle; struct inode * old_inode, * new_inode; struct buffer_head * old_bh, * new_bh, * dir_bh; - struct ext3_dir_entry_2 * old_de, * new_de; + struct ext4_dir_entry_2 * old_de, * new_de; int retval; old_bh = new_bh = dir_bh = NULL; @@ -2226,16 +2226,16 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, * in separate transaction */ if (new_dentry->d_inode) DQUOT_INIT(new_dentry->d_inode); - handle = ext3_journal_start(old_dir, 2 * - EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2); + handle = ext4_journal_start(old_dir, 2 * + EXT4_DATA_TRANS_BLOCKS(old_dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 2); if (IS_ERR(handle)) return PTR_ERR(handle); if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) handle->h_sync = 1; - old_bh = ext3_find_entry (old_dentry, &old_de); + old_bh = ext4_find_entry (old_dentry, &old_de); /* * Check for inode number is _not_ due to possible IO errors. * We might rmdir the source, keep it as pwd of some process @@ -2248,7 +2248,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, goto end_rename; new_inode = new_dentry->d_inode; - new_bh = ext3_find_entry (new_dentry, &new_de); + new_bh = ext4_find_entry (new_dentry, &new_de); if (new_bh) { if (!new_inode) { brelse (new_bh); @@ -2262,30 +2262,30 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, goto end_rename; } retval = -EIO; - dir_bh = ext3_bread (handle, old_inode, 0, 0, &retval); + dir_bh = ext4_bread (handle, old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) goto end_rename; retval = -EMLINK; if (!new_inode && new_dir!=old_dir && - new_dir->i_nlink >= EXT3_LINK_MAX) + new_dir->i_nlink >= EXT4_LINK_MAX) goto end_rename; } if (!new_bh) { - retval = ext3_add_entry (handle, new_dentry, old_inode); + retval = ext4_add_entry (handle, new_dentry, old_inode); if (retval) goto end_rename; } else { BUFFER_TRACE(new_bh, "get write access"); - ext3_journal_get_write_access(handle, new_bh); + ext4_journal_get_write_access(handle, new_bh); new_de->inode = cpu_to_le32(old_inode->i_ino); - if (EXT3_HAS_INCOMPAT_FEATURE(new_dir->i_sb, - EXT3_FEATURE_INCOMPAT_FILETYPE)) + if (EXT4_HAS_INCOMPAT_FEATURE(new_dir->i_sb, + EXT4_FEATURE_INCOMPAT_FILETYPE)) new_de->file_type = old_de->file_type; new_dir->i_version++; - BUFFER_TRACE(new_bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, new_bh); + BUFFER_TRACE(new_bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, new_bh); brelse(new_bh); new_bh = NULL; } @@ -2295,7 +2295,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, * rename. */ old_inode->i_ctime = CURRENT_TIME_SEC; - ext3_mark_inode_dirty(handle, old_inode); + ext4_mark_inode_dirty(handle, old_inode); /* * ok, that's it @@ -2303,24 +2303,24 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, if (le32_to_cpu(old_de->inode) != old_inode->i_ino || old_de->name_len != old_dentry->d_name.len || strncmp(old_de->name, old_dentry->d_name.name, old_de->name_len) || - (retval = ext3_delete_entry(handle, old_dir, + (retval = ext4_delete_entry(handle, old_dir, old_de, old_bh)) == -ENOENT) { /* old_de could have moved from under us during htree split, so * make sure that we are deleting the right entry. We might * also be pointing to a stale entry in the unused part of * old_bh so just checking inum and the name isn't enough. */ struct buffer_head *old_bh2; - struct ext3_dir_entry_2 *old_de2; + struct ext4_dir_entry_2 *old_de2; - old_bh2 = ext3_find_entry(old_dentry, &old_de2); + old_bh2 = ext4_find_entry(old_dentry, &old_de2); if (old_bh2) { - retval = ext3_delete_entry(handle, old_dir, + retval = ext4_delete_entry(handle, old_dir, old_de2, old_bh2); brelse(old_bh2); } } if (retval) { - ext3_warning(old_dir->i_sb, "ext3_rename", + ext4_warning(old_dir->i_sb, "ext4_rename", "Deleting old file (%lu), %d, error=%d", old_dir->i_ino, old_dir->i_nlink, retval); } @@ -2330,27 +2330,27 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, new_inode->i_ctime = CURRENT_TIME_SEC; } old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC; - ext3_update_dx_flag(old_dir); + ext4_update_dx_flag(old_dir); if (dir_bh) { BUFFER_TRACE(dir_bh, "get_write_access"); - ext3_journal_get_write_access(handle, dir_bh); + ext4_journal_get_write_access(handle, dir_bh); PARENT_INO(dir_bh->b_data) = cpu_to_le32(new_dir->i_ino); - BUFFER_TRACE(dir_bh, "call ext3_journal_dirty_metadata"); - ext3_journal_dirty_metadata(handle, dir_bh); + BUFFER_TRACE(dir_bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, dir_bh); drop_nlink(old_dir); if (new_inode) { drop_nlink(new_inode); } else { inc_nlink(new_dir); - ext3_update_dx_flag(new_dir); - ext3_mark_inode_dirty(handle, new_dir); + ext4_update_dx_flag(new_dir); + ext4_mark_inode_dirty(handle, new_dir); } } - ext3_mark_inode_dirty(handle, old_dir); + ext4_mark_inode_dirty(handle, old_dir); if (new_inode) { - ext3_mark_inode_dirty(handle, new_inode); + ext4_mark_inode_dirty(handle, new_inode); if (!new_inode->i_nlink) - ext3_orphan_add(handle, new_inode); + ext4_orphan_add(handle, new_inode); } retval = 0; @@ -2358,40 +2358,40 @@ end_rename: brelse (dir_bh); brelse (old_bh); brelse (new_bh); - ext3_journal_stop(handle); + ext4_journal_stop(handle); return retval; } /* * directories can handle most operations... */ -struct inode_operations ext3_dir_inode_operations = { - .create = ext3_create, - .lookup = ext3_lookup, - .link = ext3_link, - .unlink = ext3_unlink, - .symlink = ext3_symlink, - .mkdir = ext3_mkdir, - .rmdir = ext3_rmdir, - .mknod = ext3_mknod, - .rename = ext3_rename, - .setattr = ext3_setattr, -#ifdef CONFIG_EXT3_FS_XATTR +struct inode_operations ext4_dir_inode_operations = { + .create = ext4_create, + .lookup = ext4_lookup, + .link = ext4_link, + .unlink = ext4_unlink, + .symlink = ext4_symlink, + .mkdir = ext4_mkdir, + .rmdir = ext4_rmdir, + .mknod = ext4_mknod, + .rename = ext4_rename, + .setattr = ext4_setattr, +#ifdef CONFIG_EXT4DEV_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, - .listxattr = ext3_listxattr, + .listxattr = ext4_listxattr, .removexattr = generic_removexattr, #endif - .permission = ext3_permission, + .permission = ext4_permission, }; -struct inode_operations ext3_special_inode_operations = { - .setattr = ext3_setattr, -#ifdef CONFIG_EXT3_FS_XATTR +struct inode_operations ext4_special_inode_operations = { + .setattr = ext4_setattr, +#ifdef CONFIG_EXT4DEV_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, - .listxattr = ext3_listxattr, + .listxattr = ext4_listxattr, .removexattr = generic_removexattr, #endif - .permission = ext3_permission, + .permission = ext4_permission, }; diff --git a/fs/ext4/namei.h b/fs/ext4/namei.h index f2ce2b0..5e4dfff 100644 --- a/fs/ext4/namei.h +++ b/fs/ext4/namei.h @@ -1,8 +1,8 @@ -/* linux/fs/ext3/namei.h +/* linux/fs/ext4/namei.h * * Copyright (C) 2005 Simtec Electronics * Ben Dooks * */ -extern struct dentry *ext3_get_parent(struct dentry *child); +extern struct dentry *ext4_get_parent(struct dentry *child); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index b73cba1..4a47895 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -1,7 +1,7 @@ /* - * linux/fs/ext3/resize.c + * linux/fs/ext4/resize.c * - * Support for resizing an ext3 filesystem while it is mounted. + * Support for resizing an ext4 filesystem while it is mounted. * * Copyright (C) 2001, 2002 Andreas Dilger * @@ -9,11 +9,11 @@ */ -#define EXT3FS_DEBUG +#define EXT4FS_DEBUG #include #include -#include +#include #include #include @@ -23,87 +23,87 @@ #define inside(b, first, last) ((b) >= (first) && (b) < (last)) static int verify_group_input(struct super_block *sb, - struct ext3_new_group_data *input) + struct ext4_new_group_data *input) { - struct ext3_sb_info *sbi = EXT3_SB(sb); - struct ext3_super_block *es = sbi->s_es; - ext3_fsblk_t start = le32_to_cpu(es->s_blocks_count); - ext3_fsblk_t end = start + input->blocks_count; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_fsblk_t start = le32_to_cpu(es->s_blocks_count); + ext4_fsblk_t end = start + input->blocks_count; unsigned group = input->group; - ext3_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; - unsigned overhead = ext3_bg_has_super(sb, group) ? - (1 + ext3_bg_num_gdb(sb, group) + + ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; + unsigned overhead = ext4_bg_has_super(sb, group) ? + (1 + ext4_bg_num_gdb(sb, group) + le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; - ext3_fsblk_t metaend = start + overhead; + ext4_fsblk_t metaend = start + overhead; struct buffer_head *bh = NULL; - ext3_grpblk_t free_blocks_count; + ext4_grpblk_t free_blocks_count; int err = -EINVAL; input->free_blocks_count = free_blocks_count = input->blocks_count - 2 - overhead - sbi->s_itb_per_group; if (test_opt(sb, DEBUG)) - printk(KERN_DEBUG "EXT3-fs: adding %s group %u: %u blocks " + printk(KERN_DEBUG "EXT4-fs: adding %s group %u: %u blocks " "(%d free, %u reserved)\n", - ext3_bg_has_super(sb, input->group) ? "normal" : + ext4_bg_has_super(sb, input->group) ? "normal" : "no-super", input->group, input->blocks_count, free_blocks_count, input->reserved_blocks); if (group != sbi->s_groups_count) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Cannot add at group %u (only %lu groups)", input->group, sbi->s_groups_count); else if ((start - le32_to_cpu(es->s_first_data_block)) % - EXT3_BLOCKS_PER_GROUP(sb)) - ext3_warning(sb, __FUNCTION__, "Last group not full"); + EXT4_BLOCKS_PER_GROUP(sb)) + ext4_warning(sb, __FUNCTION__, "Last group not full"); else if (input->reserved_blocks > input->blocks_count / 5) - ext3_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", + ext4_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", input->reserved_blocks); else if (free_blocks_count < 0) - ext3_warning(sb, __FUNCTION__, "Bad blocks count %u", + ext4_warning(sb, __FUNCTION__, "Bad blocks count %u", input->blocks_count); else if (!(bh = sb_bread(sb, end - 1))) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Cannot read last block ("E3FSBLK")", end - 1); else if (outside(input->block_bitmap, start, end)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Block bitmap not in group (block %u)", input->block_bitmap); else if (outside(input->inode_bitmap, start, end)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Inode bitmap not in group (block %u)", input->inode_bitmap); else if (outside(input->inode_table, start, end) || outside(itend - 1, start, end)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Inode table not in group (blocks %u-"E3FSBLK")", input->inode_table, itend - 1); else if (input->inode_bitmap == input->block_bitmap) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Block bitmap same as inode bitmap (%u)", input->block_bitmap); else if (inside(input->block_bitmap, input->inode_table, itend)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Block bitmap (%u) in inode table (%u-"E3FSBLK")", input->block_bitmap, input->inode_table, itend-1); else if (inside(input->inode_bitmap, input->inode_table, itend)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Inode bitmap (%u) in inode table (%u-"E3FSBLK")", input->inode_bitmap, input->inode_table, itend-1); else if (inside(input->block_bitmap, start, metaend)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Block bitmap (%u) in GDT table" " ("E3FSBLK"-"E3FSBLK")", input->block_bitmap, start, metaend - 1); else if (inside(input->inode_bitmap, start, metaend)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Inode bitmap (%u) in GDT table" " ("E3FSBLK"-"E3FSBLK")", input->inode_bitmap, start, metaend - 1); else if (inside(input->inode_table, start, metaend) || inside(itend - 1, start, metaend)) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Inode table (%u-"E3FSBLK") overlaps" "GDT table ("E3FSBLK"-"E3FSBLK")", input->inode_table, itend - 1, start, metaend - 1); @@ -115,7 +115,7 @@ static int verify_group_input(struct super_block *sb, } static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, - ext3_fsblk_t blk) + ext4_fsblk_t blk) { struct buffer_head *bh; int err; @@ -123,7 +123,7 @@ static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, bh = sb_getblk(sb, blk); if (!bh) return ERR_PTR(-EIO); - if ((err = ext3_journal_get_write_access(handle, bh))) { + if ((err = ext4_journal_get_write_access(handle, bh))) { brelse(bh); bh = ERR_PTR(err); } else { @@ -148,9 +148,9 @@ static void mark_bitmap_end(int start_bit, int end_bit, char *bitmap) if (start_bit >= end_bit) return; - ext3_debug("mark end bits +%d through +%d used\n", start_bit, end_bit); + ext4_debug("mark end bits +%d through +%d used\n", start_bit, end_bit); for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++) - ext3_set_bit(i, bitmap); + ext4_set_bit(i, bitmap); if (i < end_bit) memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3); } @@ -163,21 +163,21 @@ static void mark_bitmap_end(int start_bit, int end_bit, char *bitmap) * If any part of this fails, we simply abort the resize. */ static int setup_new_group_blocks(struct super_block *sb, - struct ext3_new_group_data *input) + struct ext4_new_group_data *input) { - struct ext3_sb_info *sbi = EXT3_SB(sb); - ext3_fsblk_t start = ext3_group_first_block_no(sb, input->group); - int reserved_gdb = ext3_bg_has_super(sb, input->group) ? + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_fsblk_t start = ext4_group_first_block_no(sb, input->group); + int reserved_gdb = ext4_bg_has_super(sb, input->group) ? le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) : 0; - unsigned long gdblocks = ext3_bg_num_gdb(sb, input->group); + unsigned long gdblocks = ext4_bg_num_gdb(sb, input->group); struct buffer_head *bh; handle_t *handle; - ext3_fsblk_t block; - ext3_grpblk_t bit; + ext4_fsblk_t block; + ext4_grpblk_t bit; int i; int err = 0, err2; - handle = ext3_journal_start_sb(sb, reserved_gdb + gdblocks + + handle = ext4_journal_start_sb(sb, reserved_gdb + gdblocks + 2 + sbi->s_itb_per_group); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -193,9 +193,9 @@ static int setup_new_group_blocks(struct super_block *sb, goto exit_journal; } - if (ext3_bg_has_super(sb, input->group)) { - ext3_debug("mark backup superblock %#04lx (+0)\n", start); - ext3_set_bit(0, bh->b_data); + if (ext4_bg_has_super(sb, input->group)) { + ext4_debug("mark backup superblock %#04lx (+0)\n", start); + ext4_set_bit(0, bh->b_data); } /* Copy all of the GDT blocks into the backup in this group */ @@ -203,14 +203,14 @@ static int setup_new_group_blocks(struct super_block *sb, i < gdblocks; i++, block++, bit++) { struct buffer_head *gdb; - ext3_debug("update backup group %#04lx (+%d)\n", block, bit); + ext4_debug("update backup group %#04lx (+%d)\n", block, bit); gdb = sb_getblk(sb, block); if (!gdb) { err = -EIO; goto exit_bh; } - if ((err = ext3_journal_get_write_access(handle, gdb))) { + if ((err = ext4_journal_get_write_access(handle, gdb))) { brelse(gdb); goto exit_bh; } @@ -218,8 +218,8 @@ static int setup_new_group_blocks(struct super_block *sb, memcpy(gdb->b_data, sbi->s_group_desc[i]->b_data, bh->b_size); set_buffer_uptodate(gdb); unlock_buffer(bh); - ext3_journal_dirty_metadata(handle, gdb); - ext3_set_bit(bit, bh->b_data); + ext4_journal_dirty_metadata(handle, gdb); + ext4_set_bit(bit, bh->b_data); brelse(gdb); } @@ -228,59 +228,59 @@ static int setup_new_group_blocks(struct super_block *sb, i < reserved_gdb; i++, block++, bit++) { struct buffer_head *gdb; - ext3_debug("clear reserved block %#04lx (+%d)\n", block, bit); + ext4_debug("clear reserved block %#04lx (+%d)\n", block, bit); if (IS_ERR(gdb = bclean(handle, sb, block))) { err = PTR_ERR(bh); goto exit_bh; } - ext3_journal_dirty_metadata(handle, gdb); - ext3_set_bit(bit, bh->b_data); + ext4_journal_dirty_metadata(handle, gdb); + ext4_set_bit(bit, bh->b_data); brelse(gdb); } - ext3_debug("mark block bitmap %#04x (+%ld)\n", input->block_bitmap, + ext4_debug("mark block bitmap %#04x (+%ld)\n", input->block_bitmap, input->block_bitmap - start); - ext3_set_bit(input->block_bitmap - start, bh->b_data); - ext3_debug("mark inode bitmap %#04x (+%ld)\n", input->inode_bitmap, + ext4_set_bit(input->block_bitmap - start, bh->b_data); + ext4_debug("mark inode bitmap %#04x (+%ld)\n", input->inode_bitmap, input->inode_bitmap - start); - ext3_set_bit(input->inode_bitmap - start, bh->b_data); + ext4_set_bit(input->inode_bitmap - start, bh->b_data); /* Zero out all of the inode table blocks */ for (i = 0, block = input->inode_table, bit = block - start; i < sbi->s_itb_per_group; i++, bit++, block++) { struct buffer_head *it; - ext3_debug("clear inode block %#04lx (+%d)\n", block, bit); + ext4_debug("clear inode block %#04lx (+%d)\n", block, bit); if (IS_ERR(it = bclean(handle, sb, block))) { err = PTR_ERR(it); goto exit_bh; } - ext3_journal_dirty_metadata(handle, it); + ext4_journal_dirty_metadata(handle, it); brelse(it); - ext3_set_bit(bit, bh->b_data); + ext4_set_bit(bit, bh->b_data); } - mark_bitmap_end(input->blocks_count, EXT3_BLOCKS_PER_GROUP(sb), + mark_bitmap_end(input->blocks_count, EXT4_BLOCKS_PER_GROUP(sb), bh->b_data); - ext3_journal_dirty_metadata(handle, bh); + ext4_journal_dirty_metadata(handle, bh); brelse(bh); /* Mark unused entries in inode bitmap used */ - ext3_debug("clear inode bitmap %#04x (+%ld)\n", + ext4_debug("clear inode bitmap %#04x (+%ld)\n", input->inode_bitmap, input->inode_bitmap - start); if (IS_ERR(bh = bclean(handle, sb, input->inode_bitmap))) { err = PTR_ERR(bh); goto exit_journal; } - mark_bitmap_end(EXT3_INODES_PER_GROUP(sb), EXT3_BLOCKS_PER_GROUP(sb), + mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), EXT4_BLOCKS_PER_GROUP(sb), bh->b_data); - ext3_journal_dirty_metadata(handle, bh); + ext4_journal_dirty_metadata(handle, bh); exit_bh: brelse(bh); exit_journal: unlock_super(sb); - if ((err2 = ext3_journal_stop(handle)) && !err) + if ((err2 = ext4_journal_stop(handle)) && !err) err = err2; return err; @@ -288,20 +288,20 @@ exit_journal: /* * Iterate through the groups which hold BACKUP superblock/GDT copies in an - * ext3 filesystem. The counters should be initialized to 1, 5, and 7 before + * ext4 filesystem. The counters should be initialized to 1, 5, and 7 before * calling this for the first time. In a sparse filesystem it will be the * sequence of powers of 3, 5, and 7: 1, 3, 5, 7, 9, 25, 27, 49, 81, ... * For a non-sparse filesystem it will be every group: 1, 2, 3, 4, ... */ -static unsigned ext3_list_backups(struct super_block *sb, unsigned *three, +static unsigned ext4_list_backups(struct super_block *sb, unsigned *three, unsigned *five, unsigned *seven) { unsigned *min = three; int mult = 3; unsigned ret; - if (!EXT3_HAS_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER)) { + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER)) { ret = *min; *min += 1; return ret; @@ -330,8 +330,8 @@ static unsigned ext3_list_backups(struct super_block *sb, unsigned *three, static int verify_reserved_gdb(struct super_block *sb, struct buffer_head *primary) { - const ext3_fsblk_t blk = primary->b_blocknr; - const unsigned long end = EXT3_SB(sb)->s_groups_count; + const ext4_fsblk_t blk = primary->b_blocknr; + const unsigned long end = EXT4_SB(sb)->s_groups_count; unsigned three = 1; unsigned five = 5; unsigned seven = 7; @@ -339,16 +339,16 @@ static int verify_reserved_gdb(struct super_block *sb, __le32 *p = (__le32 *)primary->b_data; int gdbackups = 0; - while ((grp = ext3_list_backups(sb, &three, &five, &seven)) < end) { - if (le32_to_cpu(*p++) != grp * EXT3_BLOCKS_PER_GROUP(sb) + blk){ - ext3_warning(sb, __FUNCTION__, + while ((grp = ext4_list_backups(sb, &three, &five, &seven)) < end) { + if (le32_to_cpu(*p++) != grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){ + ext4_warning(sb, __FUNCTION__, "reserved GDT "E3FSBLK " missing grp %d ("E3FSBLK")", blk, grp, - grp * EXT3_BLOCKS_PER_GROUP(sb) + blk); + grp * EXT4_BLOCKS_PER_GROUP(sb) + blk); return -EINVAL; } - if (++gdbackups > EXT3_ADDR_PER_BLOCK(sb)) + if (++gdbackups > EXT4_ADDR_PER_BLOCK(sb)) return -EFBIG; } @@ -369,23 +369,23 @@ static int verify_reserved_gdb(struct super_block *sb, * fail once we start modifying the data on disk, because JBD has no rollback. */ static int add_new_gdb(handle_t *handle, struct inode *inode, - struct ext3_new_group_data *input, + struct ext4_new_group_data *input, struct buffer_head **primary) { struct super_block *sb = inode->i_sb; - struct ext3_super_block *es = EXT3_SB(sb)->s_es; - unsigned long gdb_num = input->group / EXT3_DESC_PER_BLOCK(sb); - ext3_fsblk_t gdblock = EXT3_SB(sb)->s_sbh->b_blocknr + 1 + gdb_num; + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + unsigned long gdb_num = input->group / EXT4_DESC_PER_BLOCK(sb); + ext4_fsblk_t gdblock = EXT4_SB(sb)->s_sbh->b_blocknr + 1 + gdb_num; struct buffer_head **o_group_desc, **n_group_desc; struct buffer_head *dind; int gdbackups; - struct ext3_iloc iloc; + struct ext4_iloc iloc; __le32 *data; int err; if (test_opt(sb, DEBUG)) printk(KERN_DEBUG - "EXT3-fs: ext3_add_new_gdb: adding group block %lu\n", + "EXT4-fs: ext4_add_new_gdb: adding group block %lu\n", gdb_num); /* @@ -393,11 +393,11 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, * because the user tools have no way of handling this. Probably a * bad time to do it anyways. */ - if (EXT3_SB(sb)->s_sbh->b_blocknr != - le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block)) { - ext3_warning(sb, __FUNCTION__, + if (EXT4_SB(sb)->s_sbh->b_blocknr != + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) { + ext4_warning(sb, __FUNCTION__, "won't resize using backup superblock at %llu", - (unsigned long long)EXT3_SB(sb)->s_sbh->b_blocknr); + (unsigned long long)EXT4_SB(sb)->s_sbh->b_blocknr); return -EPERM; } @@ -410,7 +410,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, goto exit_bh; } - data = EXT3_I(inode)->i_data + EXT3_DIND_BLOCK; + data = EXT4_I(inode)->i_data + EXT4_DIND_BLOCK; dind = sb_bread(sb, le32_to_cpu(*data)); if (!dind) { err = -EIO; @@ -418,32 +418,32 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, } data = (__le32 *)dind->b_data; - if (le32_to_cpu(data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)]) != gdblock) { - ext3_warning(sb, __FUNCTION__, + if (le32_to_cpu(data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)]) != gdblock) { + ext4_warning(sb, __FUNCTION__, "new group %u GDT block "E3FSBLK" not reserved", input->group, gdblock); err = -EINVAL; goto exit_dind; } - if ((err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh))) + if ((err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh))) goto exit_dind; - if ((err = ext3_journal_get_write_access(handle, *primary))) + if ((err = ext4_journal_get_write_access(handle, *primary))) goto exit_sbh; - if ((err = ext3_journal_get_write_access(handle, dind))) + if ((err = ext4_journal_get_write_access(handle, dind))) goto exit_primary; - /* ext3_reserve_inode_write() gets a reference on the iloc */ - if ((err = ext3_reserve_inode_write(handle, inode, &iloc))) + /* ext4_reserve_inode_write() gets a reference on the iloc */ + if ((err = ext4_reserve_inode_write(handle, inode, &iloc))) goto exit_dindj; n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *), GFP_KERNEL); if (!n_group_desc) { err = -ENOMEM; - ext3_warning (sb, __FUNCTION__, + ext4_warning (sb, __FUNCTION__, "not enough memory for %lu groups", gdb_num + 1); goto exit_inode; } @@ -457,43 +457,43 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, * these blocks, because they are marked as in-use from being in the * reserved inode, and will become GDT blocks (primary and backup). */ - data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)] = 0; - ext3_journal_dirty_metadata(handle, dind); + data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)] = 0; + ext4_journal_dirty_metadata(handle, dind); brelse(dind); inode->i_blocks -= (gdbackups + 1) * sb->s_blocksize >> 9; - ext3_mark_iloc_dirty(handle, inode, &iloc); + ext4_mark_iloc_dirty(handle, inode, &iloc); memset((*primary)->b_data, 0, sb->s_blocksize); - ext3_journal_dirty_metadata(handle, *primary); + ext4_journal_dirty_metadata(handle, *primary); - o_group_desc = EXT3_SB(sb)->s_group_desc; + o_group_desc = EXT4_SB(sb)->s_group_desc; memcpy(n_group_desc, o_group_desc, - EXT3_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); + EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); n_group_desc[gdb_num] = *primary; - EXT3_SB(sb)->s_group_desc = n_group_desc; - EXT3_SB(sb)->s_gdb_count++; + EXT4_SB(sb)->s_group_desc = n_group_desc; + EXT4_SB(sb)->s_gdb_count++; kfree(o_group_desc); es->s_reserved_gdt_blocks = cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1); - ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); return 0; exit_inode: - //ext3_journal_release_buffer(handle, iloc.bh); + //ext4_journal_release_buffer(handle, iloc.bh); brelse(iloc.bh); exit_dindj: - //ext3_journal_release_buffer(handle, dind); + //ext4_journal_release_buffer(handle, dind); exit_primary: - //ext3_journal_release_buffer(handle, *primary); + //ext4_journal_release_buffer(handle, *primary); exit_sbh: - //ext3_journal_release_buffer(handle, *primary); + //ext4_journal_release_buffer(handle, *primary); exit_dind: brelse(dind); exit_bh: brelse(*primary); - ext3_debug("leaving with error %d\n", err); + ext4_debug("leaving with error %d\n", err); return err; } @@ -511,14 +511,14 @@ exit_bh: * backup GDT blocks are stored in their reserved primary GDT block. */ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, - struct ext3_new_group_data *input) + struct ext4_new_group_data *input) { struct super_block *sb = inode->i_sb; - int reserved_gdb =le16_to_cpu(EXT3_SB(sb)->s_es->s_reserved_gdt_blocks); + int reserved_gdb =le16_to_cpu(EXT4_SB(sb)->s_es->s_reserved_gdt_blocks); struct buffer_head **primary; struct buffer_head *dind; - struct ext3_iloc iloc; - ext3_fsblk_t blk; + struct ext4_iloc iloc; + ext4_fsblk_t blk; __le32 *data, *end; int gdbackups = 0; int res, i; @@ -528,21 +528,21 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, if (!primary) return -ENOMEM; - data = EXT3_I(inode)->i_data + EXT3_DIND_BLOCK; + data = EXT4_I(inode)->i_data + EXT4_DIND_BLOCK; dind = sb_bread(sb, le32_to_cpu(*data)); if (!dind) { err = -EIO; goto exit_free; } - blk = EXT3_SB(sb)->s_sbh->b_blocknr + 1 + EXT3_SB(sb)->s_gdb_count; - data = (__le32 *)dind->b_data + EXT3_SB(sb)->s_gdb_count; - end = (__le32 *)dind->b_data + EXT3_ADDR_PER_BLOCK(sb); + blk = EXT4_SB(sb)->s_sbh->b_blocknr + 1 + EXT4_SB(sb)->s_gdb_count; + data = (__le32 *)dind->b_data + EXT4_SB(sb)->s_gdb_count; + end = (__le32 *)dind->b_data + EXT4_ADDR_PER_BLOCK(sb); /* Get each reserved primary GDT block and verify it holds backups */ for (res = 0; res < reserved_gdb; res++, blk++) { if (le32_to_cpu(*data) != blk) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "reserved block "E3FSBLK " not at offset %ld", blk, @@ -565,24 +565,24 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, } for (i = 0; i < reserved_gdb; i++) { - if ((err = ext3_journal_get_write_access(handle, primary[i]))) { + if ((err = ext4_journal_get_write_access(handle, primary[i]))) { /* int j; for (j = 0; j < i; j++) - ext3_journal_release_buffer(handle, primary[j]); + ext4_journal_release_buffer(handle, primary[j]); */ goto exit_bh; } } - if ((err = ext3_reserve_inode_write(handle, inode, &iloc))) + if ((err = ext4_reserve_inode_write(handle, inode, &iloc))) goto exit_bh; /* * Finally we can add each of the reserved backup GDT blocks from * the new group to its reserved primary GDT block. */ - blk = input->group * EXT3_BLOCKS_PER_GROUP(sb); + blk = input->group * EXT4_BLOCKS_PER_GROUP(sb); for (i = 0; i < reserved_gdb; i++) { int err2; data = (__le32 *)primary[i]->b_data; @@ -590,12 +590,12 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, primary[i]->b_blocknr, gdbackups, blk + primary[i]->b_blocknr); */ data[gdbackups] = cpu_to_le32(blk + primary[i]->b_blocknr); - err2 = ext3_journal_dirty_metadata(handle, primary[i]); + err2 = ext4_journal_dirty_metadata(handle, primary[i]); if (!err) err = err2; } inode->i_blocks += reserved_gdb * sb->s_blocksize >> 9; - ext3_mark_iloc_dirty(handle, inode, &iloc); + ext4_mark_iloc_dirty(handle, inode, &iloc); exit_bh: while (--res >= 0) @@ -609,7 +609,7 @@ exit_free: } /* - * Update the backup copies of the ext3 metadata. These don't need to be part + * Update the backup copies of the ext4 metadata. These don't need to be part * of the main resize transaction, because e2fsck will re-write them if there * is a problem (basically only OOM will cause a problem). However, we * _should_ update the backups if possible, in case the primary gets trashed @@ -626,9 +626,9 @@ exit_free: static void update_backups(struct super_block *sb, int blk_off, char *data, int size) { - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_sb_info *sbi = EXT4_SB(sb); const unsigned long last = sbi->s_groups_count; - const int bpg = EXT3_BLOCKS_PER_GROUP(sb); + const int bpg = EXT4_BLOCKS_PER_GROUP(sb); unsigned three = 1; unsigned five = 5; unsigned seven = 7; @@ -637,20 +637,20 @@ static void update_backups(struct super_block *sb, handle_t *handle; int err = 0, err2; - handle = ext3_journal_start_sb(sb, EXT3_MAX_TRANS_DATA); + handle = ext4_journal_start_sb(sb, EXT4_MAX_TRANS_DATA); if (IS_ERR(handle)) { group = 1; err = PTR_ERR(handle); goto exit_err; } - while ((group = ext3_list_backups(sb, &three, &five, &seven)) < last) { + while ((group = ext4_list_backups(sb, &three, &five, &seven)) < last) { struct buffer_head *bh; /* Out of journal space, and can't get more - abort - so sad */ if (handle->h_buffer_credits == 0 && - ext3_journal_extend(handle, EXT3_MAX_TRANS_DATA) && - (err = ext3_journal_restart(handle, EXT3_MAX_TRANS_DATA))) + ext4_journal_extend(handle, EXT4_MAX_TRANS_DATA) && + (err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA))) break; bh = sb_getblk(sb, group * bpg + blk_off); @@ -658,9 +658,9 @@ static void update_backups(struct super_block *sb, err = -EIO; break; } - ext3_debug("update metadata backup %#04lx\n", + ext4_debug("update metadata backup %#04lx\n", (unsigned long)bh->b_blocknr); - if ((err = ext3_journal_get_write_access(handle, bh))) + if ((err = ext4_journal_get_write_access(handle, bh))) break; lock_buffer(bh); memcpy(bh->b_data, data, size); @@ -668,10 +668,10 @@ static void update_backups(struct super_block *sb, memset(bh->b_data + size, 0, rest); set_buffer_uptodate(bh); unlock_buffer(bh); - ext3_journal_dirty_metadata(handle, bh); + ext4_journal_dirty_metadata(handle, bh); brelse(bh); } - if ((err2 = ext3_journal_stop(handle)) && !err) + if ((err2 = ext4_journal_stop(handle)) && !err) err = err2; /* @@ -686,11 +686,11 @@ static void update_backups(struct super_block *sb, */ exit_err: if (err) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "can't update backup for group %d (err %d), " "forcing fsck on next reboot", group, err); - sbi->s_mount_state &= ~EXT3_VALID_FS; - sbi->s_es->s_state &= cpu_to_le16(~EXT3_VALID_FS); + sbi->s_mount_state &= ~EXT4_VALID_FS; + sbi->s_es->s_state &= cpu_to_le16(~EXT4_VALID_FS); mark_buffer_dirty(sbi->s_sbh); } } @@ -708,51 +708,51 @@ exit_err: * not really "added" the group at all. We re-check that we are still * adding in the last group in case things have changed since verifying. */ -int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) +int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) { - struct ext3_sb_info *sbi = EXT3_SB(sb); - struct ext3_super_block *es = sbi->s_es; - int reserved_gdb = ext3_bg_has_super(sb, input->group) ? + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + int reserved_gdb = ext4_bg_has_super(sb, input->group) ? le16_to_cpu(es->s_reserved_gdt_blocks) : 0; struct buffer_head *primary = NULL; - struct ext3_group_desc *gdp; + struct ext4_group_desc *gdp; struct inode *inode = NULL; handle_t *handle; int gdb_off, gdb_num; int err, err2; - gdb_num = input->group / EXT3_DESC_PER_BLOCK(sb); - gdb_off = input->group % EXT3_DESC_PER_BLOCK(sb); + gdb_num = input->group / EXT4_DESC_PER_BLOCK(sb); + gdb_off = input->group % EXT4_DESC_PER_BLOCK(sb); - if (gdb_off == 0 && !EXT3_HAS_RO_COMPAT_FEATURE(sb, - EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER)) { - ext3_warning(sb, __FUNCTION__, + if (gdb_off == 0 && !EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER)) { + ext4_warning(sb, __FUNCTION__, "Can't resize non-sparse filesystem further"); return -EPERM; } if (le32_to_cpu(es->s_blocks_count) + input->blocks_count < le32_to_cpu(es->s_blocks_count)) { - ext3_warning(sb, __FUNCTION__, "blocks_count overflow\n"); + ext4_warning(sb, __FUNCTION__, "blocks_count overflow\n"); return -EINVAL; } - if (le32_to_cpu(es->s_inodes_count) + EXT3_INODES_PER_GROUP(sb) < + if (le32_to_cpu(es->s_inodes_count) + EXT4_INODES_PER_GROUP(sb) < le32_to_cpu(es->s_inodes_count)) { - ext3_warning(sb, __FUNCTION__, "inodes_count overflow\n"); + ext4_warning(sb, __FUNCTION__, "inodes_count overflow\n"); return -EINVAL; } if (reserved_gdb || gdb_off == 0) { - if (!EXT3_HAS_COMPAT_FEATURE(sb, - EXT3_FEATURE_COMPAT_RESIZE_INODE)){ - ext3_warning(sb, __FUNCTION__, + if (!EXT4_HAS_COMPAT_FEATURE(sb, + EXT4_FEATURE_COMPAT_RESIZE_INODE)){ + ext4_warning(sb, __FUNCTION__, "No reserved GDT blocks, can't resize"); return -EPERM; } - inode = iget(sb, EXT3_RESIZE_INO); + inode = iget(sb, EXT4_RESIZE_INO); if (!inode || is_bad_inode(inode)) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "Error opening resize inode"); iput(inode); return -ENOENT; @@ -772,8 +772,8 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) * are adding a group with superblock/GDT backups we will also * modify each of the reserved GDT dindirect blocks. */ - handle = ext3_journal_start_sb(sb, - ext3_bg_has_super(sb, input->group) ? + handle = ext4_journal_start_sb(sb, + ext4_bg_has_super(sb, input->group) ? 3 + reserved_gdb : 4); if (IS_ERR(handle)) { err = PTR_ERR(handle); @@ -782,13 +782,13 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) lock_super(sb); if (input->group != sbi->s_groups_count) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "multiple resizers run on filesystem!"); err = -EBUSY; goto exit_journal; } - if ((err = ext3_journal_get_write_access(handle, sbi->s_sbh))) + if ((err = ext4_journal_get_write_access(handle, sbi->s_sbh))) goto exit_journal; /* @@ -799,10 +799,10 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) */ if (gdb_off) { primary = sbi->s_group_desc[gdb_num]; - if ((err = ext3_journal_get_write_access(handle, primary))) + if ((err = ext4_journal_get_write_access(handle, primary))) goto exit_journal; - if (reserved_gdb && ext3_bg_num_gdb(sb, input->group) && + if (reserved_gdb && ext4_bg_num_gdb(sb, input->group) && (err = reserve_backup_gdb(handle, inode, input))) goto exit_journal; } else if ((err = add_new_gdb(handle, inode, input, &primary))) @@ -828,13 +828,13 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) */ /* Update group descriptor block for new group */ - gdp = (struct ext3_group_desc *)primary->b_data + gdb_off; + gdp = (struct ext4_group_desc *)primary->b_data + gdb_off; gdp->bg_block_bitmap = cpu_to_le32(input->block_bitmap); gdp->bg_inode_bitmap = cpu_to_le32(input->inode_bitmap); gdp->bg_inode_table = cpu_to_le32(input->inode_table); gdp->bg_free_blocks_count = cpu_to_le16(input->free_blocks_count); - gdp->bg_free_inodes_count = cpu_to_le16(EXT3_INODES_PER_GROUP(sb)); + gdp->bg_free_inodes_count = cpu_to_le16(EXT4_INODES_PER_GROUP(sb)); /* * Make the new blocks and inodes valid next. We do this before @@ -849,7 +849,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) es->s_blocks_count = cpu_to_le32(le32_to_cpu(es->s_blocks_count) + input->blocks_count); es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) + - EXT3_INODES_PER_GROUP(sb)); + EXT4_INODES_PER_GROUP(sb)); /* * We need to protect s_groups_count against other CPUs seeing @@ -878,7 +878,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) /* Update the global fs size fields */ sbi->s_groups_count++; - ext3_journal_dirty_metadata(handle, primary); + ext4_journal_dirty_metadata(handle, primary); /* Update the reserved block counts only once the new group is * active. */ @@ -889,42 +889,42 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) percpu_counter_mod(&sbi->s_freeblocks_counter, input->free_blocks_count); percpu_counter_mod(&sbi->s_freeinodes_counter, - EXT3_INODES_PER_GROUP(sb)); + EXT4_INODES_PER_GROUP(sb)); - ext3_journal_dirty_metadata(handle, sbi->s_sbh); + ext4_journal_dirty_metadata(handle, sbi->s_sbh); sb->s_dirt = 1; exit_journal: unlock_super(sb); - if ((err2 = ext3_journal_stop(handle)) && !err) + if ((err2 = ext4_journal_stop(handle)) && !err) err = err2; if (!err) { update_backups(sb, sbi->s_sbh->b_blocknr, (char *)es, - sizeof(struct ext3_super_block)); + sizeof(struct ext4_super_block)); update_backups(sb, primary->b_blocknr, primary->b_data, primary->b_size); } exit_put: iput(inode); return err; -} /* ext3_group_add */ +} /* ext4_group_add */ /* Extend the filesystem to the new number of blocks specified. This entry * point is only used to extend the current filesystem to the end of the last * existing group. It can be accessed via ioctl, or by "remount,resize=" * for emergencies (because it has no dependencies on reserved blocks). * - * If we _really_ wanted, we could use default values to call ext3_group_add() + * If we _really_ wanted, we could use default values to call ext4_group_add() * allow the "remount" trick to work for arbitrary resizing, assuming enough * GDT blocks are reserved to grow to the desired size. */ -int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, - ext3_fsblk_t n_blocks_count) +int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, + ext4_fsblk_t n_blocks_count) { - ext3_fsblk_t o_blocks_count; + ext4_fsblk_t o_blocks_count; unsigned long o_groups_count; - ext3_grpblk_t last; - ext3_grpblk_t add; + ext4_grpblk_t last; + ext4_grpblk_t add; struct buffer_head * bh; handle_t *handle; int err; @@ -934,45 +934,45 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, * yet: we're going to revalidate es->s_blocks_count after * taking lock_super() below. */ o_blocks_count = le32_to_cpu(es->s_blocks_count); - o_groups_count = EXT3_SB(sb)->s_groups_count; + o_groups_count = EXT4_SB(sb)->s_groups_count; if (test_opt(sb, DEBUG)) - printk(KERN_DEBUG "EXT3-fs: extending last group from "E3FSBLK" uto "E3FSBLK" blocks\n", + printk(KERN_DEBUG "EXT4-fs: extending last group from "E3FSBLK" uto "E3FSBLK" blocks\n", o_blocks_count, n_blocks_count); if (n_blocks_count == 0 || n_blocks_count == o_blocks_count) return 0; if (n_blocks_count > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { - printk(KERN_ERR "EXT3-fs: filesystem on %s:" + printk(KERN_ERR "EXT4-fs: filesystem on %s:" " too large to resize to %lu blocks safely\n", sb->s_id, n_blocks_count); if (sizeof(sector_t) < 8) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "CONFIG_LBD not enabled\n"); return -EINVAL; } if (n_blocks_count < o_blocks_count) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "can't shrink FS - resize aborted"); return -EBUSY; } /* Handle the remaining blocks in the last group only. */ last = (o_blocks_count - le32_to_cpu(es->s_first_data_block)) % - EXT3_BLOCKS_PER_GROUP(sb); + EXT4_BLOCKS_PER_GROUP(sb); if (last == 0) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "need to use ext2online to resize further"); return -EPERM; } - add = EXT3_BLOCKS_PER_GROUP(sb) - last; + add = EXT4_BLOCKS_PER_GROUP(sb) - last; if (o_blocks_count + add < o_blocks_count) { - ext3_warning(sb, __FUNCTION__, "blocks_count overflow"); + ext4_warning(sb, __FUNCTION__, "blocks_count overflow"); return -EINVAL; } @@ -980,7 +980,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, add = n_blocks_count - o_blocks_count; if (o_blocks_count + add < n_blocks_count) - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "will only finish group ("E3FSBLK " blocks, %u new)", o_blocks_count + add, add); @@ -988,55 +988,55 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, /* See if the device is actually as big as what was requested */ bh = sb_bread(sb, o_blocks_count + add -1); if (!bh) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "can't read last block, resize aborted"); return -ENOSPC; } brelse(bh); /* We will update the superblock, one block bitmap, and - * one group descriptor via ext3_free_blocks(). + * one group descriptor via ext4_free_blocks(). */ - handle = ext3_journal_start_sb(sb, 3); + handle = ext4_journal_start_sb(sb, 3); if (IS_ERR(handle)) { err = PTR_ERR(handle); - ext3_warning(sb, __FUNCTION__, "error %d on journal start",err); + ext4_warning(sb, __FUNCTION__, "error %d on journal start",err); goto exit_put; } lock_super(sb); if (o_blocks_count != le32_to_cpu(es->s_blocks_count)) { - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "multiple resizers run on filesystem!"); unlock_super(sb); err = -EBUSY; goto exit_put; } - if ((err = ext3_journal_get_write_access(handle, - EXT3_SB(sb)->s_sbh))) { - ext3_warning(sb, __FUNCTION__, + if ((err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh))) { + ext4_warning(sb, __FUNCTION__, "error %d on journal write access", err); unlock_super(sb); - ext3_journal_stop(handle); + ext4_journal_stop(handle); goto exit_put; } es->s_blocks_count = cpu_to_le32(o_blocks_count + add); - ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); sb->s_dirt = 1; unlock_super(sb); - ext3_debug("freeing blocks %lu through "E3FSBLK"\n", o_blocks_count, + ext4_debug("freeing blocks %lu through "E3FSBLK"\n", o_blocks_count, o_blocks_count + add); - ext3_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); - ext3_debug("freed blocks "E3FSBLK" through "E3FSBLK"\n", o_blocks_count, + ext4_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); + ext4_debug("freed blocks "E3FSBLK" through "E3FSBLK"\n", o_blocks_count, o_blocks_count + add); - if ((err = ext3_journal_stop(handle))) + if ((err = ext4_journal_stop(handle))) goto exit_put; if (test_opt(sb, DEBUG)) - printk(KERN_DEBUG "EXT3-fs: extended group to %u blocks\n", + printk(KERN_DEBUG "EXT4-fs: extended group to %u blocks\n", le32_to_cpu(es->s_blocks_count)); - update_backups(sb, EXT3_SB(sb)->s_sbh->b_blocknr, (char *)es, - sizeof(struct ext3_super_block)); + update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr, (char *)es, + sizeof(struct ext4_super_block)); exit_put: return err; -} /* ext3_group_extend */ +} /* ext4_group_extend */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 8bfd56e..9e32a2a 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/super.c + * linux/fs/ext4/super.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -21,8 +21,8 @@ #include #include #include -#include -#include +#include +#include #include #include #include @@ -42,25 +42,25 @@ #include "acl.h" #include "namei.h" -static int ext3_load_journal(struct super_block *, struct ext3_super_block *, +static int ext4_load_journal(struct super_block *, struct ext4_super_block *, unsigned long journal_devnum); -static int ext3_create_journal(struct super_block *, struct ext3_super_block *, +static int ext4_create_journal(struct super_block *, struct ext4_super_block *, unsigned int); -static void ext3_commit_super (struct super_block * sb, - struct ext3_super_block * es, +static void ext4_commit_super (struct super_block * sb, + struct ext4_super_block * es, int sync); -static void ext3_mark_recovery_complete(struct super_block * sb, - struct ext3_super_block * es); -static void ext3_clear_journal_err(struct super_block * sb, - struct ext3_super_block * es); -static int ext3_sync_fs(struct super_block *sb, int wait); -static const char *ext3_decode_error(struct super_block * sb, int errno, +static void ext4_mark_recovery_complete(struct super_block * sb, + struct ext4_super_block * es); +static void ext4_clear_journal_err(struct super_block * sb, + struct ext4_super_block * es); +static int ext4_sync_fs(struct super_block *sb, int wait); +static const char *ext4_decode_error(struct super_block * sb, int errno, char nbuf[16]); -static int ext3_remount (struct super_block * sb, int * flags, char * data); -static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf); -static void ext3_unlockfs(struct super_block *sb); -static void ext3_write_super (struct super_block * sb); -static void ext3_write_super_lockfs(struct super_block *sb); +static int ext4_remount (struct super_block * sb, int * flags, char * data); +static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf); +static void ext4_unlockfs(struct super_block *sb); +static void ext4_write_super (struct super_block * sb); +static void ext4_write_super_lockfs(struct super_block *sb); /* * Wrappers for journal_start/end. @@ -70,7 +70,7 @@ static void ext3_write_super_lockfs(struct super_block *sb); * that sync() will call the filesystem's write_super callback if * appropriate. */ -handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks) +handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks) { journal_t *journal; @@ -80,9 +80,9 @@ handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks) /* Special case here: if the journal has aborted behind our * backs (eg. EIO in the commit thread), then we still need to * take the FS itself readonly cleanly. */ - journal = EXT3_SB(sb)->s_journal; + journal = EXT4_SB(sb)->s_journal; if (is_journal_aborted(journal)) { - ext3_abort(sb, __FUNCTION__, + ext4_abort(sb, __FUNCTION__, "Detected aborted journal"); return ERR_PTR(-EROFS); } @@ -96,7 +96,7 @@ handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks) * that sync() will call the filesystem's write_super callback if * appropriate. */ -int __ext3_journal_stop(const char *where, handle_t *handle) +int __ext4_journal_stop(const char *where, handle_t *handle) { struct super_block *sb; int err; @@ -109,15 +109,15 @@ int __ext3_journal_stop(const char *where, handle_t *handle) if (!err) err = rc; if (err) - __ext3_std_error(sb, where, err); + __ext4_std_error(sb, where, err); return err; } -void ext3_journal_abort_handle(const char *caller, const char *err_fn, +void ext4_journal_abort_handle(const char *caller, const char *err_fn, struct buffer_head *bh, handle_t *handle, int err) { char nbuf[16]; - const char *errstr = ext3_decode_error(NULL, err, nbuf); + const char *errstr = ext4_decode_error(NULL, err, nbuf); if (bh) BUFFER_TRACE(bh, "abort"); @@ -138,7 +138,7 @@ void ext3_journal_abort_handle(const char *caller, const char *err_fn, * inconsistencies detected or read IO failures. * * On ext2, we can store the error state of the filesystem in the - * superblock. That is not possible on ext3, because we may have other + * superblock. That is not possible on ext4, because we may have other * write ordering constraints on the superblock which prevent us from * writing it out straight away; and given that the journal is about to * be aborted, we can't rely on the current, or future, transactions to @@ -149,20 +149,20 @@ void ext3_journal_abort_handle(const char *caller, const char *err_fn, * that error until we've noted it down and cleared it. */ -static void ext3_handle_error(struct super_block *sb) +static void ext4_handle_error(struct super_block *sb) { - struct ext3_super_block *es = EXT3_SB(sb)->s_es; + struct ext4_super_block *es = EXT4_SB(sb)->s_es; - EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; - es->s_state |= cpu_to_le16(EXT3_ERROR_FS); + EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; + es->s_state |= cpu_to_le16(EXT4_ERROR_FS); if (sb->s_flags & MS_RDONLY) return; if (!test_opt (sb, ERRORS_CONT)) { - journal_t *journal = EXT3_SB(sb)->s_journal; + journal_t *journal = EXT4_SB(sb)->s_journal; - EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT; + EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; if (journal) journal_abort(journal, -EIO); } @@ -170,27 +170,27 @@ static void ext3_handle_error(struct super_block *sb) printk (KERN_CRIT "Remounting filesystem read-only\n"); sb->s_flags |= MS_RDONLY; } - ext3_commit_super(sb, es, 1); + ext4_commit_super(sb, es, 1); if (test_opt(sb, ERRORS_PANIC)) - panic("EXT3-fs (device %s): panic forced after error\n", + panic("EXT4-fs (device %s): panic forced after error\n", sb->s_id); } -void ext3_error (struct super_block * sb, const char * function, +void ext4_error (struct super_block * sb, const char * function, const char * fmt, ...) { va_list args; va_start(args, fmt); - printk(KERN_CRIT "EXT3-fs error (device %s): %s: ",sb->s_id, function); + printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function); vprintk(fmt, args); printk("\n"); va_end(args); - ext3_handle_error(sb); + ext4_handle_error(sb); } -static const char *ext3_decode_error(struct super_block * sb, int errno, +static const char *ext4_decode_error(struct super_block * sb, int errno, char nbuf[16]) { char *errstr = NULL; @@ -203,7 +203,7 @@ static const char *ext3_decode_error(struct super_block * sb, int errno, errstr = "Out of memory"; break; case -EROFS: - if (!sb || EXT3_SB(sb)->s_journal->j_flags & JFS_ABORT) + if (!sb || EXT4_SB(sb)->s_journal->j_flags & JFS_ABORT) errstr = "Journal has aborted"; else errstr = "Readonly filesystem"; @@ -223,10 +223,10 @@ static const char *ext3_decode_error(struct super_block * sb, int errno, return errstr; } -/* __ext3_std_error decodes expected errors from journaling functions +/* __ext4_std_error decodes expected errors from journaling functions * automatically and invokes the appropriate error response. */ -void __ext3_std_error (struct super_block * sb, const char * function, +void __ext4_std_error (struct super_block * sb, const char * function, int errno) { char nbuf[16]; @@ -239,15 +239,15 @@ void __ext3_std_error (struct super_block * sb, const char * function, (sb->s_flags & MS_RDONLY)) return; - errstr = ext3_decode_error(sb, errno, nbuf); - printk (KERN_CRIT "EXT3-fs error (device %s) in %s: %s\n", + errstr = ext4_decode_error(sb, errno, nbuf); + printk (KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n", sb->s_id, function, errstr); - ext3_handle_error(sb); + ext4_handle_error(sb); } /* - * ext3_abort is a much stronger failure handler than ext3_error. The + * ext4_abort is a much stronger failure handler than ext4_error. The * abort function may be used to deal with unrecoverable failures such * as journal IO errors or ENOMEM at a critical moment in log management. * @@ -256,60 +256,60 @@ void __ext3_std_error (struct super_block * sb, const char * function, * case we take the easy way out and panic immediately. */ -void ext3_abort (struct super_block * sb, const char * function, +void ext4_abort (struct super_block * sb, const char * function, const char * fmt, ...) { va_list args; - printk (KERN_CRIT "ext3_abort called.\n"); + printk (KERN_CRIT "ext4_abort called.\n"); va_start(args, fmt); - printk(KERN_CRIT "EXT3-fs error (device %s): %s: ",sb->s_id, function); + printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function); vprintk(fmt, args); printk("\n"); va_end(args); if (test_opt(sb, ERRORS_PANIC)) - panic("EXT3-fs panic from previous error\n"); + panic("EXT4-fs panic from previous error\n"); if (sb->s_flags & MS_RDONLY) return; printk(KERN_CRIT "Remounting filesystem read-only\n"); - EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; + EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; sb->s_flags |= MS_RDONLY; - EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT; - journal_abort(EXT3_SB(sb)->s_journal, -EIO); + EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; + journal_abort(EXT4_SB(sb)->s_journal, -EIO); } -void ext3_warning (struct super_block * sb, const char * function, +void ext4_warning (struct super_block * sb, const char * function, const char * fmt, ...) { va_list args; va_start(args, fmt); - printk(KERN_WARNING "EXT3-fs warning (device %s): %s: ", + printk(KERN_WARNING "EXT4-fs warning (device %s): %s: ", sb->s_id, function); vprintk(fmt, args); printk("\n"); va_end(args); } -void ext3_update_dynamic_rev(struct super_block *sb) +void ext4_update_dynamic_rev(struct super_block *sb) { - struct ext3_super_block *es = EXT3_SB(sb)->s_es; + struct ext4_super_block *es = EXT4_SB(sb)->s_es; - if (le32_to_cpu(es->s_rev_level) > EXT3_GOOD_OLD_REV) + if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV) return; - ext3_warning(sb, __FUNCTION__, + ext4_warning(sb, __FUNCTION__, "updating to rev %d because of new feature flag, " "running e2fsck is recommended", - EXT3_DYNAMIC_REV); + EXT4_DYNAMIC_REV); - es->s_first_ino = cpu_to_le32(EXT3_GOOD_OLD_FIRST_INO); - es->s_inode_size = cpu_to_le16(EXT3_GOOD_OLD_INODE_SIZE); - es->s_rev_level = cpu_to_le32(EXT3_DYNAMIC_REV); + es->s_first_ino = cpu_to_le32(EXT4_GOOD_OLD_FIRST_INO); + es->s_inode_size = cpu_to_le16(EXT4_GOOD_OLD_INODE_SIZE); + es->s_rev_level = cpu_to_le32(EXT4_DYNAMIC_REV); /* leave es->s_feature_*compat flags alone */ /* es->s_uuid will be set by e2fsck if empty */ @@ -323,7 +323,7 @@ void ext3_update_dynamic_rev(struct super_block *sb) /* * Open the external journal device */ -static struct block_device *ext3_blkdev_get(dev_t dev) +static struct block_device *ext4_blkdev_get(dev_t dev) { struct block_device *bdev; char b[BDEVNAME_SIZE]; @@ -334,7 +334,7 @@ static struct block_device *ext3_blkdev_get(dev_t dev) return bdev; fail: - printk(KERN_ERR "EXT3: failed to open journal device %s: %ld\n", + printk(KERN_ERR "EXT4: failed to open journal device %s: %ld\n", __bdevname(dev, b), PTR_ERR(bdev)); return NULL; } @@ -342,20 +342,20 @@ fail: /* * Release the journal device */ -static int ext3_blkdev_put(struct block_device *bdev) +static int ext4_blkdev_put(struct block_device *bdev) { bd_release(bdev); return blkdev_put(bdev); } -static int ext3_blkdev_remove(struct ext3_sb_info *sbi) +static int ext4_blkdev_remove(struct ext4_sb_info *sbi) { struct block_device *bdev; int ret = -ENODEV; bdev = sbi->journal_bdev; if (bdev) { - ret = ext3_blkdev_put(bdev); + ret = ext4_blkdev_put(bdev); sbi->journal_bdev = NULL; } return ret; @@ -363,10 +363,10 @@ static int ext3_blkdev_remove(struct ext3_sb_info *sbi) static inline struct inode *orphan_list_entry(struct list_head *l) { - return &list_entry(l, struct ext3_inode_info, i_orphan)->vfs_inode; + return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode; } -static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi) +static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi) { struct list_head *l; @@ -384,20 +384,20 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi) } } -static void ext3_put_super (struct super_block * sb) +static void ext4_put_super (struct super_block * sb) { - struct ext3_sb_info *sbi = EXT3_SB(sb); - struct ext3_super_block *es = sbi->s_es; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; int i; - ext3_xattr_put_super(sb); + ext4_xattr_put_super(sb); journal_destroy(sbi->s_journal); if (!(sb->s_flags & MS_RDONLY)) { - EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); es->s_state = cpu_to_le16(sbi->s_mount_state); BUFFER_TRACE(sbi->s_sbh, "marking dirty"); mark_buffer_dirty(sbi->s_sbh); - ext3_commit_super(sb, es, 1); + ext4_commit_super(sb, es, 1); } for (i = 0; i < sbi->s_gdb_count; i++) @@ -429,47 +429,47 @@ static void ext3_put_super (struct super_block * sb) */ sync_blockdev(sbi->journal_bdev); invalidate_bdev(sbi->journal_bdev, 0); - ext3_blkdev_remove(sbi); + ext4_blkdev_remove(sbi); } sb->s_fs_info = NULL; kfree(sbi); return; } -static kmem_cache_t *ext3_inode_cachep; +static kmem_cache_t *ext4_inode_cachep; /* * Called inside transaction, so use GFP_NOFS */ -static struct inode *ext3_alloc_inode(struct super_block *sb) +static struct inode *ext4_alloc_inode(struct super_block *sb) { - struct ext3_inode_info *ei; + struct ext4_inode_info *ei; - ei = kmem_cache_alloc(ext3_inode_cachep, SLAB_NOFS); + ei = kmem_cache_alloc(ext4_inode_cachep, SLAB_NOFS); if (!ei) return NULL; -#ifdef CONFIG_EXT3_FS_POSIX_ACL - ei->i_acl = EXT3_ACL_NOT_CACHED; - ei->i_default_acl = EXT3_ACL_NOT_CACHED; +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + ei->i_acl = EXT4_ACL_NOT_CACHED; + ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif ei->i_block_alloc_info = NULL; ei->vfs_inode.i_version = 1; return &ei->vfs_inode; } -static void ext3_destroy_inode(struct inode *inode) +static void ext4_destroy_inode(struct inode *inode) { - kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); + kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) { - struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; + struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR) { INIT_LIST_HEAD(&ei->i_orphan); -#ifdef CONFIG_EXT3_FS_XATTR +#ifdef CONFIG_EXT4DEV_FS_XATTR init_rwsem(&ei->xattr_sem); #endif mutex_init(&ei->truncate_mutex); @@ -479,46 +479,46 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) static int init_inodecache(void) { - ext3_inode_cachep = kmem_cache_create("ext3_inode_cache", - sizeof(struct ext3_inode_info), + ext4_inode_cachep = kmem_cache_create("ext4_inode_cache", + sizeof(struct ext4_inode_info), 0, (SLAB_RECLAIM_ACCOUNT| SLAB_MEM_SPREAD), init_once, NULL); - if (ext3_inode_cachep == NULL) + if (ext4_inode_cachep == NULL) return -ENOMEM; return 0; } static void destroy_inodecache(void) { - kmem_cache_destroy(ext3_inode_cachep); + kmem_cache_destroy(ext4_inode_cachep); } -static void ext3_clear_inode(struct inode *inode) +static void ext4_clear_inode(struct inode *inode) { - struct ext3_block_alloc_info *rsv = EXT3_I(inode)->i_block_alloc_info; -#ifdef CONFIG_EXT3_FS_POSIX_ACL - if (EXT3_I(inode)->i_acl && - EXT3_I(inode)->i_acl != EXT3_ACL_NOT_CACHED) { - posix_acl_release(EXT3_I(inode)->i_acl); - EXT3_I(inode)->i_acl = EXT3_ACL_NOT_CACHED; - } - if (EXT3_I(inode)->i_default_acl && - EXT3_I(inode)->i_default_acl != EXT3_ACL_NOT_CACHED) { - posix_acl_release(EXT3_I(inode)->i_default_acl); - EXT3_I(inode)->i_default_acl = EXT3_ACL_NOT_CACHED; + struct ext4_block_alloc_info *rsv = EXT4_I(inode)->i_block_alloc_info; +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + if (EXT4_I(inode)->i_acl && + EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) { + posix_acl_release(EXT4_I(inode)->i_acl); + EXT4_I(inode)->i_acl = EXT4_ACL_NOT_CACHED; + } + if (EXT4_I(inode)->i_default_acl && + EXT4_I(inode)->i_default_acl != EXT4_ACL_NOT_CACHED) { + posix_acl_release(EXT4_I(inode)->i_default_acl); + EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED; } #endif - ext3_discard_reservation(inode); - EXT3_I(inode)->i_block_alloc_info = NULL; + ext4_discard_reservation(inode); + EXT4_I(inode)->i_block_alloc_info = NULL; if (unlikely(rsv)) kfree(rsv); } -static inline void ext3_show_quota_options(struct seq_file *seq, struct super_block *sb) +static inline void ext4_show_quota_options(struct seq_file *seq, struct super_block *sb) { #if defined(CONFIG_QUOTA) - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_sb_info *sbi = EXT4_SB(sb); if (sbi->s_jquota_fmt) seq_printf(seq, ",jqfmt=%s", @@ -530,32 +530,32 @@ static inline void ext3_show_quota_options(struct seq_file *seq, struct super_bl if (sbi->s_qf_names[GRPQUOTA]) seq_printf(seq, ",grpjquota=%s", sbi->s_qf_names[GRPQUOTA]); - if (sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA) + if (sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA) seq_puts(seq, ",usrquota"); - if (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA) + if (sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA) seq_puts(seq, ",grpquota"); #endif } -static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) +static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) { struct super_block *sb = vfs->mnt_sb; - if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA) + if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) seq_puts(seq, ",data=journal"); - else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA) + else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) seq_puts(seq, ",data=ordered"); - else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA) + else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) seq_puts(seq, ",data=writeback"); - ext3_show_quota_options(seq, sb); + ext4_show_quota_options(seq, sb); return 0; } -static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) +static struct dentry *ext4_get_dentry(struct super_block *sb, void *vobjp) { __u32 *objp = vobjp; unsigned long ino = objp[0]; @@ -563,14 +563,14 @@ static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) struct inode *inode; struct dentry *result; - if (ino < EXT3_FIRST_INO(sb) && ino != EXT3_ROOT_INO) + if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO) return ERR_PTR(-ESTALE); - if (ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) + if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count)) return ERR_PTR(-ESTALE); /* iget isn't really right if the inode is currently unallocated!! * - * ext3_read_inode will return a bad_inode if the inode had been + * ext4_read_inode will return a bad_inode if the inode had been * deleted, so we should be safe. * * Currently we don't know the generation for parent directory, so @@ -599,37 +599,37 @@ static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) #define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group") #define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) -static int ext3_dquot_initialize(struct inode *inode, int type); -static int ext3_dquot_drop(struct inode *inode); -static int ext3_write_dquot(struct dquot *dquot); -static int ext3_acquire_dquot(struct dquot *dquot); -static int ext3_release_dquot(struct dquot *dquot); -static int ext3_mark_dquot_dirty(struct dquot *dquot); -static int ext3_write_info(struct super_block *sb, int type); -static int ext3_quota_on(struct super_block *sb, int type, int format_id, char *path); -static int ext3_quota_on_mount(struct super_block *sb, int type); -static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, +static int ext4_dquot_initialize(struct inode *inode, int type); +static int ext4_dquot_drop(struct inode *inode); +static int ext4_write_dquot(struct dquot *dquot); +static int ext4_acquire_dquot(struct dquot *dquot); +static int ext4_release_dquot(struct dquot *dquot); +static int ext4_mark_dquot_dirty(struct dquot *dquot); +static int ext4_write_info(struct super_block *sb, int type); +static int ext4_quota_on(struct super_block *sb, int type, int format_id, char *path); +static int ext4_quota_on_mount(struct super_block *sb, int type); +static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off); -static ssize_t ext3_quota_write(struct super_block *sb, int type, +static ssize_t ext4_quota_write(struct super_block *sb, int type, const char *data, size_t len, loff_t off); -static struct dquot_operations ext3_quota_operations = { - .initialize = ext3_dquot_initialize, - .drop = ext3_dquot_drop, +static struct dquot_operations ext4_quota_operations = { + .initialize = ext4_dquot_initialize, + .drop = ext4_dquot_drop, .alloc_space = dquot_alloc_space, .alloc_inode = dquot_alloc_inode, .free_space = dquot_free_space, .free_inode = dquot_free_inode, .transfer = dquot_transfer, - .write_dquot = ext3_write_dquot, - .acquire_dquot = ext3_acquire_dquot, - .release_dquot = ext3_release_dquot, - .mark_dirty = ext3_mark_dquot_dirty, - .write_info = ext3_write_info + .write_dquot = ext4_write_dquot, + .acquire_dquot = ext4_acquire_dquot, + .release_dquot = ext4_release_dquot, + .mark_dirty = ext4_mark_dquot_dirty, + .write_info = ext4_write_info }; -static struct quotactl_ops ext3_qctl_operations = { - .quota_on = ext3_quota_on, +static struct quotactl_ops ext4_qctl_operations = { + .quota_on = ext4_quota_on, .quota_off = vfs_quota_off, .quota_sync = vfs_quota_sync, .get_info = vfs_get_dqinfo, @@ -639,31 +639,31 @@ static struct quotactl_ops ext3_qctl_operations = { }; #endif -static struct super_operations ext3_sops = { - .alloc_inode = ext3_alloc_inode, - .destroy_inode = ext3_destroy_inode, - .read_inode = ext3_read_inode, - .write_inode = ext3_write_inode, - .dirty_inode = ext3_dirty_inode, - .delete_inode = ext3_delete_inode, - .put_super = ext3_put_super, - .write_super = ext3_write_super, - .sync_fs = ext3_sync_fs, - .write_super_lockfs = ext3_write_super_lockfs, - .unlockfs = ext3_unlockfs, - .statfs = ext3_statfs, - .remount_fs = ext3_remount, - .clear_inode = ext3_clear_inode, - .show_options = ext3_show_options, +static struct super_operations ext4_sops = { + .alloc_inode = ext4_alloc_inode, + .destroy_inode = ext4_destroy_inode, + .read_inode = ext4_read_inode, + .write_inode = ext4_write_inode, + .dirty_inode = ext4_dirty_inode, + .delete_inode = ext4_delete_inode, + .put_super = ext4_put_super, + .write_super = ext4_write_super, + .sync_fs = ext4_sync_fs, + .write_super_lockfs = ext4_write_super_lockfs, + .unlockfs = ext4_unlockfs, + .statfs = ext4_statfs, + .remount_fs = ext4_remount, + .clear_inode = ext4_clear_inode, + .show_options = ext4_show_options, #ifdef CONFIG_QUOTA - .quota_read = ext3_quota_read, - .quota_write = ext3_quota_write, + .quota_read = ext4_quota_read, + .quota_write = ext4_quota_write, #endif }; -static struct export_operations ext3_export_ops = { - .get_parent = ext3_get_parent, - .get_dentry = ext3_get_dentry, +static struct export_operations ext4_export_ops = { + .get_parent = ext4_get_parent, + .get_dentry = ext4_get_dentry, }; enum { @@ -731,18 +731,18 @@ static match_table_t tokens = { {Opt_resize, "resize"}, }; -static ext3_fsblk_t get_sb_block(void **data) +static ext4_fsblk_t get_sb_block(void **data) { - ext3_fsblk_t sb_block; + ext4_fsblk_t sb_block; char *options = (char *) *data; if (!options || strncmp(options, "sb=", 3) != 0) return 1; /* Default location */ options += 3; - /*todo: use simple_strtoll with >32bit ext3 */ + /*todo: use simple_strtoll with >32bit ext4 */ sb_block = simple_strtoul(options, &options, 0); if (*options && *options != ',') { - printk("EXT3-fs: Invalid sb specification: %s\n", + printk("EXT4-fs: Invalid sb specification: %s\n", (char *) *data); return 1; } @@ -754,9 +754,9 @@ static ext3_fsblk_t get_sb_block(void **data) static int parse_options (char *options, struct super_block *sb, unsigned int *inum, unsigned long *journal_devnum, - ext3_fsblk_t *n_blocks_count, int is_remount) + ext4_fsblk_t *n_blocks_count, int is_remount) { - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_sb_info *sbi = EXT4_SB(sb); char * p; substring_t args[MAX_OPT_ARGS]; int data_opt = 0; @@ -832,7 +832,7 @@ static int parse_options (char *options, struct super_block *sb, case Opt_orlov: clear_opt (sbi->s_mount_opt, OLDALLOC); break; -#ifdef CONFIG_EXT3_FS_XATTR +#ifdef CONFIG_EXT4DEV_FS_XATTR case Opt_user_xattr: set_opt (sbi->s_mount_opt, XATTR_USER); break; @@ -842,10 +842,10 @@ static int parse_options (char *options, struct super_block *sb, #else case Opt_user_xattr: case Opt_nouser_xattr: - printk("EXT3 (no)user_xattr options not supported\n"); + printk("EXT4 (no)user_xattr options not supported\n"); break; #endif -#ifdef CONFIG_EXT3_FS_POSIX_ACL +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL case Opt_acl: set_opt(sbi->s_mount_opt, POSIX_ACL); break; @@ -855,7 +855,7 @@ static int parse_options (char *options, struct super_block *sb, #else case Opt_acl: case Opt_noacl: - printk("EXT3 (no)acl options not supported\n"); + printk("EXT4 (no)acl options not supported\n"); break; #endif case Opt_reservation: @@ -871,7 +871,7 @@ static int parse_options (char *options, struct super_block *sb, user to specify an existing inode to be the journal file. */ if (is_remount) { - printk(KERN_ERR "EXT3-fs: cannot specify " + printk(KERN_ERR "EXT4-fs: cannot specify " "journal on remount\n"); return 0; } @@ -879,7 +879,7 @@ static int parse_options (char *options, struct super_block *sb, break; case Opt_journal_inum: if (is_remount) { - printk(KERN_ERR "EXT3-fs: cannot specify " + printk(KERN_ERR "EXT4-fs: cannot specify " "journal on remount\n"); return 0; } @@ -889,7 +889,7 @@ static int parse_options (char *options, struct super_block *sb, break; case Opt_journal_dev: if (is_remount) { - printk(KERN_ERR "EXT3-fs: cannot specify " + printk(KERN_ERR "EXT4-fs: cannot specify " "journal on remount\n"); return 0; } @@ -910,24 +910,24 @@ static int parse_options (char *options, struct super_block *sb, sbi->s_commit_interval = HZ * option; break; case Opt_data_journal: - data_opt = EXT3_MOUNT_JOURNAL_DATA; + data_opt = EXT4_MOUNT_JOURNAL_DATA; goto datacheck; case Opt_data_ordered: - data_opt = EXT3_MOUNT_ORDERED_DATA; + data_opt = EXT4_MOUNT_ORDERED_DATA; goto datacheck; case Opt_data_writeback: - data_opt = EXT3_MOUNT_WRITEBACK_DATA; + data_opt = EXT4_MOUNT_WRITEBACK_DATA; datacheck: if (is_remount) { - if ((sbi->s_mount_opt & EXT3_MOUNT_DATA_FLAGS) + if ((sbi->s_mount_opt & EXT4_MOUNT_DATA_FLAGS) != data_opt) { printk(KERN_ERR - "EXT3-fs: cannot change data " + "EXT4-fs: cannot change data " "mode on remount\n"); return 0; } } else { - sbi->s_mount_opt &= ~EXT3_MOUNT_DATA_FLAGS; + sbi->s_mount_opt &= ~EXT4_MOUNT_DATA_FLAGS; sbi->s_mount_opt |= data_opt; } break; @@ -940,21 +940,21 @@ static int parse_options (char *options, struct super_block *sb, set_qf_name: if (sb_any_quota_enabled(sb)) { printk(KERN_ERR - "EXT3-fs: Cannot change journalled " + "EXT4-fs: Cannot change journalled " "quota options when quota turned on.\n"); return 0; } qname = match_strdup(&args[0]); if (!qname) { printk(KERN_ERR - "EXT3-fs: not enough memory for " + "EXT4-fs: not enough memory for " "storing quotafile name.\n"); return 0; } if (sbi->s_qf_names[qtype] && strcmp(sbi->s_qf_names[qtype], qname)) { printk(KERN_ERR - "EXT3-fs: %s quota file already " + "EXT4-fs: %s quota file already " "specified.\n", QTYPE2NAME(qtype)); kfree(qname); return 0; @@ -962,7 +962,7 @@ set_qf_name: sbi->s_qf_names[qtype] = qname; if (strchr(sbi->s_qf_names[qtype], '/')) { printk(KERN_ERR - "EXT3-fs: quotafile must be on " + "EXT4-fs: quotafile must be on " "filesystem root.\n"); kfree(sbi->s_qf_names[qtype]); sbi->s_qf_names[qtype] = NULL; @@ -977,7 +977,7 @@ set_qf_name: qtype = GRPQUOTA; clear_qf_name: if (sb_any_quota_enabled(sb)) { - printk(KERN_ERR "EXT3-fs: Cannot change " + printk(KERN_ERR "EXT4-fs: Cannot change " "journalled quota options when " "quota turned on.\n"); return 0; @@ -1005,7 +1005,7 @@ clear_qf_name: break; case Opt_noquota: if (sb_any_quota_enabled(sb)) { - printk(KERN_ERR "EXT3-fs: Cannot change quota " + printk(KERN_ERR "EXT4-fs: Cannot change quota " "options when quota turned on.\n"); return 0; } @@ -1024,7 +1024,7 @@ clear_qf_name: case Opt_jqfmt_vfsold: case Opt_jqfmt_vfsv0: printk(KERN_ERR - "EXT3-fs: journalled quota options not " + "EXT4-fs: journalled quota options not " "supported.\n"); break; case Opt_noquota: @@ -1045,7 +1045,7 @@ clear_qf_name: break; case Opt_resize: if (!is_remount) { - printk("EXT3-fs: resize option only available " + printk("EXT4-fs: resize option only available " "for remount\n"); return 0; } @@ -1061,38 +1061,38 @@ clear_qf_name: break; default: printk (KERN_ERR - "EXT3-fs: Unrecognized mount option \"%s\" " + "EXT4-fs: Unrecognized mount option \"%s\" " "or missing value\n", p); return 0; } } #ifdef CONFIG_QUOTA if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) { - if ((sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA) && + if ((sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA) && sbi->s_qf_names[USRQUOTA]) clear_opt(sbi->s_mount_opt, USRQUOTA); - if ((sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA) && + if ((sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA) && sbi->s_qf_names[GRPQUOTA]) clear_opt(sbi->s_mount_opt, GRPQUOTA); if ((sbi->s_qf_names[USRQUOTA] && - (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA)) || + (sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA)) || (sbi->s_qf_names[GRPQUOTA] && - (sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA))) { - printk(KERN_ERR "EXT3-fs: old and new quota " + (sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA))) { + printk(KERN_ERR "EXT4-fs: old and new quota " "format mixing.\n"); return 0; } if (!sbi->s_jquota_fmt) { - printk(KERN_ERR "EXT3-fs: journalled quota format " + printk(KERN_ERR "EXT4-fs: journalled quota format " "not specified.\n"); return 0; } } else { if (sbi->s_jquota_fmt) { - printk(KERN_ERR "EXT3-fs: journalled quota format " + printk(KERN_ERR "EXT4-fs: journalled quota format " "specified with no journalling " "enabled.\n"); return 0; @@ -1102,68 +1102,68 @@ clear_qf_name: return 1; } -static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, +static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, int read_only) { - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_sb_info *sbi = EXT4_SB(sb); int res = 0; - if (le32_to_cpu(es->s_rev_level) > EXT3_MAX_SUPP_REV) { - printk (KERN_ERR "EXT3-fs warning: revision level too high, " + if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) { + printk (KERN_ERR "EXT4-fs warning: revision level too high, " "forcing read-only mode\n"); res = MS_RDONLY; } if (read_only) return res; - if (!(sbi->s_mount_state & EXT3_VALID_FS)) - printk (KERN_WARNING "EXT3-fs warning: mounting unchecked fs, " + if (!(sbi->s_mount_state & EXT4_VALID_FS)) + printk (KERN_WARNING "EXT4-fs warning: mounting unchecked fs, " "running e2fsck is recommended\n"); - else if ((sbi->s_mount_state & EXT3_ERROR_FS)) + else if ((sbi->s_mount_state & EXT4_ERROR_FS)) printk (KERN_WARNING - "EXT3-fs warning: mounting fs with errors, " + "EXT4-fs warning: mounting fs with errors, " "running e2fsck is recommended\n"); else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 && le16_to_cpu(es->s_mnt_count) >= (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count)) printk (KERN_WARNING - "EXT3-fs warning: maximal mount count reached, " + "EXT4-fs warning: maximal mount count reached, " "running e2fsck is recommended\n"); else if (le32_to_cpu(es->s_checkinterval) && (le32_to_cpu(es->s_lastcheck) + le32_to_cpu(es->s_checkinterval) <= get_seconds())) printk (KERN_WARNING - "EXT3-fs warning: checktime reached, " + "EXT4-fs warning: checktime reached, " "running e2fsck is recommended\n"); #if 0 /* @@@ We _will_ want to clear the valid bit if we find inconsistencies, to force a fsck at reboot. But for a plain journaled filesystem we can keep it set as valid forever! :) */ - es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT3_VALID_FS); + es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT4_VALID_FS); #endif if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) - es->s_max_mnt_count = cpu_to_le16(EXT3_DFL_MAX_MNT_COUNT); + es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT); es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1); es->s_mtime = cpu_to_le32(get_seconds()); - ext3_update_dynamic_rev(sb); - EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + ext4_update_dynamic_rev(sb); + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); - ext3_commit_super(sb, es, 1); + ext4_commit_super(sb, es, 1); if (test_opt(sb, DEBUG)) - printk(KERN_INFO "[EXT3 FS bs=%lu, gc=%lu, " + printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%lu, " "bpg=%lu, ipg=%lu, mo=%04lx]\n", sb->s_blocksize, sbi->s_groups_count, - EXT3_BLOCKS_PER_GROUP(sb), - EXT3_INODES_PER_GROUP(sb), + EXT4_BLOCKS_PER_GROUP(sb), + EXT4_INODES_PER_GROUP(sb), sbi->s_mount_opt); - printk(KERN_INFO "EXT3 FS on %s, ", sb->s_id); - if (EXT3_SB(sb)->s_journal->j_inode == NULL) { + printk(KERN_INFO "EXT4 FS on %s, ", sb->s_id); + if (EXT4_SB(sb)->s_journal->j_inode == NULL) { char b[BDEVNAME_SIZE]; printk("external journal on %s\n", - bdevname(EXT3_SB(sb)->s_journal->j_dev, b)); + bdevname(EXT4_SB(sb)->s_journal->j_dev, b)); } else { printk("internal journal\n"); } @@ -1171,16 +1171,16 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, } /* Called at mount-time, super-block is locked */ -static int ext3_check_descriptors (struct super_block * sb) +static int ext4_check_descriptors (struct super_block * sb) { - struct ext3_sb_info *sbi = EXT3_SB(sb); - ext3_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); - ext3_fsblk_t last_block; - struct ext3_group_desc * gdp = NULL; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); + ext4_fsblk_t last_block; + struct ext4_group_desc * gdp = NULL; int desc_block = 0; int i; - ext3_debug ("Checking group descriptors"); + ext4_debug ("Checking group descriptors"); for (i = 0; i < sbi->s_groups_count; i++) { @@ -1188,15 +1188,15 @@ static int ext3_check_descriptors (struct super_block * sb) last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; else last_block = first_block + - (EXT3_BLOCKS_PER_GROUP(sb) - 1); + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - if ((i % EXT3_DESC_PER_BLOCK(sb)) == 0) - gdp = (struct ext3_group_desc *) + if ((i % EXT4_DESC_PER_BLOCK(sb)) == 0) + gdp = (struct ext4_group_desc *) sbi->s_group_desc[desc_block++]->b_data; if (le32_to_cpu(gdp->bg_block_bitmap) < first_block || le32_to_cpu(gdp->bg_block_bitmap) > last_block) { - ext3_error (sb, "ext3_check_descriptors", + ext4_error (sb, "ext4_check_descriptors", "Block bitmap for group %d" " not in group (block %lu)!", i, (unsigned long) @@ -1206,7 +1206,7 @@ static int ext3_check_descriptors (struct super_block * sb) if (le32_to_cpu(gdp->bg_inode_bitmap) < first_block || le32_to_cpu(gdp->bg_inode_bitmap) > last_block) { - ext3_error (sb, "ext3_check_descriptors", + ext4_error (sb, "ext4_check_descriptors", "Inode bitmap for group %d" " not in group (block %lu)!", i, (unsigned long) @@ -1217,24 +1217,24 @@ static int ext3_check_descriptors (struct super_block * sb) le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group > last_block) { - ext3_error (sb, "ext3_check_descriptors", + ext4_error (sb, "ext4_check_descriptors", "Inode table for group %d" " not in group (block %lu)!", i, (unsigned long) le32_to_cpu(gdp->bg_inode_table)); return 0; } - first_block += EXT3_BLOCKS_PER_GROUP(sb); + first_block += EXT4_BLOCKS_PER_GROUP(sb); gdp++; } - sbi->s_es->s_free_blocks_count=cpu_to_le32(ext3_count_free_blocks(sb)); - sbi->s_es->s_free_inodes_count=cpu_to_le32(ext3_count_free_inodes(sb)); + sbi->s_es->s_free_blocks_count=cpu_to_le32(ext4_count_free_blocks(sb)); + sbi->s_es->s_free_inodes_count=cpu_to_le32(ext4_count_free_inodes(sb)); return 1; } -/* ext3_orphan_cleanup() walks a singly-linked list of inodes (starting at +/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at * the superblock) which were deleted from all directories, but held open by * a process at the time of a crash. We walk the list and try to delete these * inodes at recovery time (only with a read-write filesystem). @@ -1247,12 +1247,12 @@ static int ext3_check_descriptors (struct super_block * sb) * We only do an iget() and an iput() on each inode, which is very safe if we * accidentally point at an in-use or already deleted inode. The worst that * can happen in this case is that we get a "bit already cleared" message from - * ext3_free_inode(). The only reason we would point at a wrong inode is if + * ext4_free_inode(). The only reason we would point at a wrong inode is if * e2fsck was run on this filesystem, and it must have already done the orphan * inode cleanup for us, so we can safely abort without any further action. */ -static void ext3_orphan_cleanup (struct super_block * sb, - struct ext3_super_block * es) +static void ext4_orphan_cleanup (struct super_block * sb, + struct ext4_super_block * es) { unsigned int s_flags = sb->s_flags; int nr_orphans = 0, nr_truncates = 0; @@ -1264,7 +1264,7 @@ static void ext3_orphan_cleanup (struct super_block * sb, return; } - if (EXT3_SB(sb)->s_mount_state & EXT3_ERROR_FS) { + if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) { if (es->s_last_orphan) jbd_debug(1, "Errors on filesystem, " "clearing orphan list.\n"); @@ -1274,7 +1274,7 @@ static void ext3_orphan_cleanup (struct super_block * sb, } if (s_flags & MS_RDONLY) { - printk(KERN_INFO "EXT3-fs: %s: orphan cleanup on readonly fs\n", + printk(KERN_INFO "EXT4-fs: %s: orphan cleanup on readonly fs\n", sb->s_id); sb->s_flags &= ~MS_RDONLY; } @@ -1283,11 +1283,11 @@ static void ext3_orphan_cleanup (struct super_block * sb, sb->s_flags |= MS_ACTIVE; /* Turn on quotas so that they are updated correctly */ for (i = 0; i < MAXQUOTAS; i++) { - if (EXT3_SB(sb)->s_qf_names[i]) { - int ret = ext3_quota_on_mount(sb, i); + if (EXT4_SB(sb)->s_qf_names[i]) { + int ret = ext4_quota_on_mount(sb, i); if (ret < 0) printk(KERN_ERR - "EXT3-fs: Cannot turn on journalled " + "EXT4-fs: Cannot turn on journalled " "quota: error %d\n", ret); } } @@ -1297,12 +1297,12 @@ static void ext3_orphan_cleanup (struct super_block * sb, struct inode *inode; if (!(inode = - ext3_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) { + ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) { es->s_last_orphan = 0; break; } - list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan); + list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan); DQUOT_INIT(inode); if (inode->i_nlink) { printk(KERN_DEBUG @@ -1310,7 +1310,7 @@ static void ext3_orphan_cleanup (struct super_block * sb, __FUNCTION__, inode->i_ino, inode->i_size); jbd_debug(2, "truncating inode %lu to %Ld bytes\n", inode->i_ino, inode->i_size); - ext3_truncate(inode); + ext4_truncate(inode); nr_truncates++; } else { printk(KERN_DEBUG @@ -1326,10 +1326,10 @@ static void ext3_orphan_cleanup (struct super_block * sb, #define PLURAL(x) (x), ((x)==1) ? "" : "s" if (nr_orphans) - printk(KERN_INFO "EXT3-fs: %s: %d orphan inode%s deleted\n", + printk(KERN_INFO "EXT4-fs: %s: %d orphan inode%s deleted\n", sb->s_id, PLURAL(nr_orphans)); if (nr_truncates) - printk(KERN_INFO "EXT3-fs: %s: %d truncate%s cleaned up\n", + printk(KERN_INFO "EXT4-fs: %s: %d truncate%s cleaned up\n", sb->s_id, PLURAL(nr_truncates)); #ifdef CONFIG_QUOTA /* Turn quotas off */ @@ -1348,9 +1348,9 @@ static void ext3_orphan_cleanup (struct super_block * sb, * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. * We need to be 1 filesystem block less than the 2^32 sector limit. */ -static loff_t ext3_max_size(int bits) +static loff_t ext4_max_size(int bits) { - loff_t res = EXT3_NDIR_BLOCKS; + loff_t res = EXT4_NDIR_BLOCKS; /* This constant is calculated to be the largest file size for a * dense, 4k-blocksize file such that the total number of * sectors in the file, including data and all indirect blocks, @@ -1366,34 +1366,34 @@ static loff_t ext3_max_size(int bits) return res; } -static ext3_fsblk_t descriptor_loc(struct super_block *sb, - ext3_fsblk_t logic_sb_block, +static ext4_fsblk_t descriptor_loc(struct super_block *sb, + ext4_fsblk_t logic_sb_block, int nr) { - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_sb_info *sbi = EXT4_SB(sb); unsigned long bg, first_meta_bg; int has_super = 0; first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg); - if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_META_BG) || + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG) || nr < first_meta_bg) return (logic_sb_block + nr + 1); bg = sbi->s_desc_per_block * nr; - if (ext3_bg_has_super(sb, bg)) + if (ext4_bg_has_super(sb, bg)) has_super = 1; - return (has_super + ext3_group_first_block_no(sb, bg)); + return (has_super + ext4_group_first_block_no(sb, bg)); } -static int ext3_fill_super (struct super_block *sb, void *data, int silent) +static int ext4_fill_super (struct super_block *sb, void *data, int silent) { struct buffer_head * bh; - struct ext3_super_block *es = NULL; - struct ext3_sb_info *sbi; - ext3_fsblk_t block; - ext3_fsblk_t sb_block = get_sb_block(&data); - ext3_fsblk_t logic_sb_block; + struct ext4_super_block *es = NULL; + struct ext4_sb_info *sbi; + ext4_fsblk_t block; + ext4_fsblk_t sb_block = get_sb_block(&data); + ext4_fsblk_t logic_sb_block; unsigned long offset = 0; unsigned int journal_inum = 0; unsigned long journal_devnum = 0; @@ -1411,64 +1411,64 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) return -ENOMEM; sb->s_fs_info = sbi; sbi->s_mount_opt = 0; - sbi->s_resuid = EXT3_DEF_RESUID; - sbi->s_resgid = EXT3_DEF_RESGID; + sbi->s_resuid = EXT4_DEF_RESUID; + sbi->s_resgid = EXT4_DEF_RESGID; unlock_kernel(); - blocksize = sb_min_blocksize(sb, EXT3_MIN_BLOCK_SIZE); + blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE); if (!blocksize) { - printk(KERN_ERR "EXT3-fs: unable to set blocksize\n"); + printk(KERN_ERR "EXT4-fs: unable to set blocksize\n"); goto out_fail; } /* - * The ext3 superblock will not be buffer aligned for other than 1kB + * The ext4 superblock will not be buffer aligned for other than 1kB * block sizes. We need to calculate the offset from buffer start. */ - if (blocksize != EXT3_MIN_BLOCK_SIZE) { - logic_sb_block = (sb_block * EXT3_MIN_BLOCK_SIZE) / blocksize; - offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize; + if (blocksize != EXT4_MIN_BLOCK_SIZE) { + logic_sb_block = (sb_block * EXT4_MIN_BLOCK_SIZE) / blocksize; + offset = (sb_block * EXT4_MIN_BLOCK_SIZE) % blocksize; } else { logic_sb_block = sb_block; } if (!(bh = sb_bread(sb, logic_sb_block))) { - printk (KERN_ERR "EXT3-fs: unable to read superblock\n"); + printk (KERN_ERR "EXT4-fs: unable to read superblock\n"); goto out_fail; } /* * Note: s_es must be initialized as soon as possible because - * some ext3 macro-instructions depend on its value + * some ext4 macro-instructions depend on its value */ - es = (struct ext3_super_block *) (((char *)bh->b_data) + offset); + es = (struct ext4_super_block *) (((char *)bh->b_data) + offset); sbi->s_es = es; sb->s_magic = le16_to_cpu(es->s_magic); - if (sb->s_magic != EXT3_SUPER_MAGIC) - goto cantfind_ext3; + if (sb->s_magic != EXT4_SUPER_MAGIC) + goto cantfind_ext4; /* Set defaults before we parse the mount options */ def_mount_opts = le32_to_cpu(es->s_default_mount_opts); - if (def_mount_opts & EXT3_DEFM_DEBUG) + if (def_mount_opts & EXT4_DEFM_DEBUG) set_opt(sbi->s_mount_opt, DEBUG); - if (def_mount_opts & EXT3_DEFM_BSDGROUPS) + if (def_mount_opts & EXT4_DEFM_BSDGROUPS) set_opt(sbi->s_mount_opt, GRPID); - if (def_mount_opts & EXT3_DEFM_UID16) + if (def_mount_opts & EXT4_DEFM_UID16) set_opt(sbi->s_mount_opt, NO_UID32); - if (def_mount_opts & EXT3_DEFM_XATTR_USER) + if (def_mount_opts & EXT4_DEFM_XATTR_USER) set_opt(sbi->s_mount_opt, XATTR_USER); - if (def_mount_opts & EXT3_DEFM_ACL) + if (def_mount_opts & EXT4_DEFM_ACL) set_opt(sbi->s_mount_opt, POSIX_ACL); - if ((def_mount_opts & EXT3_DEFM_JMODE) == EXT3_DEFM_JMODE_DATA) - sbi->s_mount_opt |= EXT3_MOUNT_JOURNAL_DATA; - else if ((def_mount_opts & EXT3_DEFM_JMODE) == EXT3_DEFM_JMODE_ORDERED) - sbi->s_mount_opt |= EXT3_MOUNT_ORDERED_DATA; - else if ((def_mount_opts & EXT3_DEFM_JMODE) == EXT3_DEFM_JMODE_WBACK) - sbi->s_mount_opt |= EXT3_MOUNT_WRITEBACK_DATA; - - if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_PANIC) + if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA) + sbi->s_mount_opt |= EXT4_MOUNT_JOURNAL_DATA; + else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED) + sbi->s_mount_opt |= EXT4_MOUNT_ORDERED_DATA; + else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK) + sbi->s_mount_opt |= EXT4_MOUNT_WRITEBACK_DATA; + + if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC) set_opt(sbi->s_mount_opt, ERRORS_PANIC); - else if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_RO) + else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_RO) set_opt(sbi->s_mount_opt, ERRORS_RO); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); @@ -1481,40 +1481,40 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount; sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | - ((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + ((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); - if (le32_to_cpu(es->s_rev_level) == EXT3_GOOD_OLD_REV && - (EXT3_HAS_COMPAT_FEATURE(sb, ~0U) || - EXT3_HAS_RO_COMPAT_FEATURE(sb, ~0U) || - EXT3_HAS_INCOMPAT_FEATURE(sb, ~0U))) + if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV && + (EXT4_HAS_COMPAT_FEATURE(sb, ~0U) || + EXT4_HAS_RO_COMPAT_FEATURE(sb, ~0U) || + EXT4_HAS_INCOMPAT_FEATURE(sb, ~0U))) printk(KERN_WARNING - "EXT3-fs warning: feature flags set on rev 0 fs, " + "EXT4-fs warning: feature flags set on rev 0 fs, " "running e2fsck is recommended\n"); /* * Check feature flags regardless of the revision level, since we * previously didn't change the revision level when setting the flags, * so there is a chance incompat flags are set on a rev 0 filesystem. */ - features = EXT3_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP); + features = EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT4_FEATURE_INCOMPAT_SUPP); if (features) { - printk(KERN_ERR "EXT3-fs: %s: couldn't mount because of " + printk(KERN_ERR "EXT4-fs: %s: couldn't mount because of " "unsupported optional features (%x).\n", sb->s_id, le32_to_cpu(features)); goto failed_mount; } - features = EXT3_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP); + features = EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT4_FEATURE_RO_COMPAT_SUPP); if (!(sb->s_flags & MS_RDONLY) && features) { - printk(KERN_ERR "EXT3-fs: %s: couldn't mount RDWR because of " + printk(KERN_ERR "EXT4-fs: %s: couldn't mount RDWR because of " "unsupported optional features (%x).\n", sb->s_id, le32_to_cpu(features)); goto failed_mount; } blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size); - if (blocksize < EXT3_MIN_BLOCK_SIZE || - blocksize > EXT3_MAX_BLOCK_SIZE) { + if (blocksize < EXT4_MIN_BLOCK_SIZE || + blocksize > EXT4_MAX_BLOCK_SIZE) { printk(KERN_ERR - "EXT3-fs: Unsupported filesystem blocksize %d on %s.\n", + "EXT4-fs: Unsupported filesystem blocksize %d on %s.\n", blocksize, sb->s_id); goto failed_mount; } @@ -1526,52 +1526,52 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) * than the hardware sectorsize for the machine. */ if (blocksize < hblock) { - printk(KERN_ERR "EXT3-fs: blocksize %d too small for " + printk(KERN_ERR "EXT4-fs: blocksize %d too small for " "device blocksize %d.\n", blocksize, hblock); goto failed_mount; } brelse (bh); sb_set_blocksize(sb, blocksize); - logic_sb_block = (sb_block * EXT3_MIN_BLOCK_SIZE) / blocksize; - offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize; + logic_sb_block = (sb_block * EXT4_MIN_BLOCK_SIZE) / blocksize; + offset = (sb_block * EXT4_MIN_BLOCK_SIZE) % blocksize; bh = sb_bread(sb, logic_sb_block); if (!bh) { printk(KERN_ERR - "EXT3-fs: Can't read superblock on 2nd try.\n"); + "EXT4-fs: Can't read superblock on 2nd try.\n"); goto failed_mount; } - es = (struct ext3_super_block *)(((char *)bh->b_data) + offset); + es = (struct ext4_super_block *)(((char *)bh->b_data) + offset); sbi->s_es = es; - if (es->s_magic != cpu_to_le16(EXT3_SUPER_MAGIC)) { + if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) { printk (KERN_ERR - "EXT3-fs: Magic mismatch, very weird !\n"); + "EXT4-fs: Magic mismatch, very weird !\n"); goto failed_mount; } } - sb->s_maxbytes = ext3_max_size(sb->s_blocksize_bits); + sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits); - if (le32_to_cpu(es->s_rev_level) == EXT3_GOOD_OLD_REV) { - sbi->s_inode_size = EXT3_GOOD_OLD_INODE_SIZE; - sbi->s_first_ino = EXT3_GOOD_OLD_FIRST_INO; + if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) { + sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE; + sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO; } else { sbi->s_inode_size = le16_to_cpu(es->s_inode_size); sbi->s_first_ino = le32_to_cpu(es->s_first_ino); - if ((sbi->s_inode_size < EXT3_GOOD_OLD_INODE_SIZE) || + if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) || (sbi->s_inode_size & (sbi->s_inode_size - 1)) || (sbi->s_inode_size > blocksize)) { printk (KERN_ERR - "EXT3-fs: unsupported inode size: %d\n", + "EXT4-fs: unsupported inode size: %d\n", sbi->s_inode_size); goto failed_mount; } } - sbi->s_frag_size = EXT3_MIN_FRAG_SIZE << + sbi->s_frag_size = EXT4_MIN_FRAG_SIZE << le32_to_cpu(es->s_log_frag_size); if (blocksize != sbi->s_frag_size) { printk(KERN_ERR - "EXT3-fs: fragsize %lu != blocksize %u (unsupported)\n", + "EXT4-fs: fragsize %lu != blocksize %u (unsupported)\n", sbi->s_frag_size, blocksize); goto failed_mount; } @@ -1579,62 +1579,62 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group); sbi->s_frags_per_group = le32_to_cpu(es->s_frags_per_group); sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group); - if (EXT3_INODE_SIZE(sb) == 0) - goto cantfind_ext3; - sbi->s_inodes_per_block = blocksize / EXT3_INODE_SIZE(sb); + if (EXT4_INODE_SIZE(sb) == 0) + goto cantfind_ext4; + sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb); if (sbi->s_inodes_per_block == 0) - goto cantfind_ext3; + goto cantfind_ext4; sbi->s_itb_per_group = sbi->s_inodes_per_group / sbi->s_inodes_per_block; - sbi->s_desc_per_block = blocksize / sizeof(struct ext3_group_desc); + sbi->s_desc_per_block = blocksize / sizeof(struct ext4_group_desc); sbi->s_sbh = bh; sbi->s_mount_state = le16_to_cpu(es->s_state); - sbi->s_addr_per_block_bits = log2(EXT3_ADDR_PER_BLOCK(sb)); - sbi->s_desc_per_block_bits = log2(EXT3_DESC_PER_BLOCK(sb)); + sbi->s_addr_per_block_bits = log2(EXT4_ADDR_PER_BLOCK(sb)); + sbi->s_desc_per_block_bits = log2(EXT4_DESC_PER_BLOCK(sb)); for (i=0; i < 4; i++) sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); sbi->s_def_hash_version = es->s_def_hash_version; if (sbi->s_blocks_per_group > blocksize * 8) { printk (KERN_ERR - "EXT3-fs: #blocks per group too big: %lu\n", + "EXT4-fs: #blocks per group too big: %lu\n", sbi->s_blocks_per_group); goto failed_mount; } if (sbi->s_frags_per_group > blocksize * 8) { printk (KERN_ERR - "EXT3-fs: #fragments per group too big: %lu\n", + "EXT4-fs: #fragments per group too big: %lu\n", sbi->s_frags_per_group); goto failed_mount; } if (sbi->s_inodes_per_group > blocksize * 8) { printk (KERN_ERR - "EXT3-fs: #inodes per group too big: %lu\n", + "EXT4-fs: #inodes per group too big: %lu\n", sbi->s_inodes_per_group); goto failed_mount; } if (le32_to_cpu(es->s_blocks_count) > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { - printk(KERN_ERR "EXT3-fs: filesystem on %s:" + printk(KERN_ERR "EXT4-fs: filesystem on %s:" " too large to mount safely\n", sb->s_id); if (sizeof(sector_t) < 8) - printk(KERN_WARNING "EXT3-fs: CONFIG_LBD not " + printk(KERN_WARNING "EXT4-fs: CONFIG_LBD not " "enabled\n"); goto failed_mount; } - if (EXT3_BLOCKS_PER_GROUP(sb) == 0) - goto cantfind_ext3; + if (EXT4_BLOCKS_PER_GROUP(sb) == 0) + goto cantfind_ext4; sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) - le32_to_cpu(es->s_first_data_block) - 1) - / EXT3_BLOCKS_PER_GROUP(sb)) + 1; - db_count = (sbi->s_groups_count + EXT3_DESC_PER_BLOCK(sb) - 1) / - EXT3_DESC_PER_BLOCK(sb); + / EXT4_BLOCKS_PER_GROUP(sb)) + 1; + db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / + EXT4_DESC_PER_BLOCK(sb); sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *), GFP_KERNEL); if (sbi->s_group_desc == NULL) { - printk (KERN_ERR "EXT3-fs: not enough memory\n"); + printk (KERN_ERR "EXT4-fs: not enough memory\n"); goto failed_mount; } @@ -1644,14 +1644,14 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) block = descriptor_loc(sb, logic_sb_block, i); sbi->s_group_desc[i] = sb_bread(sb, block); if (!sbi->s_group_desc[i]) { - printk (KERN_ERR "EXT3-fs: " + printk (KERN_ERR "EXT4-fs: " "can't read group descriptor %d\n", i); db_count = i; goto failed_mount2; } } - if (!ext3_check_descriptors (sb)) { - printk(KERN_ERR "EXT3-fs: group descriptors corrupted!\n"); + if (!ext4_check_descriptors (sb)) { + printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n"); goto failed_mount2; } sbi->s_gdb_count = db_count; @@ -1659,11 +1659,11 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) spin_lock_init(&sbi->s_next_gen_lock); percpu_counter_init(&sbi->s_freeblocks_counter, - ext3_count_free_blocks(sb)); + ext4_count_free_blocks(sb)); percpu_counter_init(&sbi->s_freeinodes_counter, - ext3_count_free_inodes(sb)); + ext4_count_free_inodes(sb)); percpu_counter_init(&sbi->s_dirs_counter, - ext3_count_dirs(sb)); + ext4_count_dirs(sb)); /* per fileystem reservation list head & lock */ spin_lock_init(&sbi->s_rsv_window_lock); @@ -1672,45 +1672,45 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) * reservation window list --- it gives us a placeholder for * append-at-start-of-list which makes the allocation logic * _much_ simpler. */ - sbi->s_rsv_window_head.rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; - sbi->s_rsv_window_head.rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED; + sbi->s_rsv_window_head.rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + sbi->s_rsv_window_head.rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; sbi->s_rsv_window_head.rsv_alloc_hit = 0; sbi->s_rsv_window_head.rsv_goal_size = 0; - ext3_rsv_window_add(sb, &sbi->s_rsv_window_head); + ext4_rsv_window_add(sb, &sbi->s_rsv_window_head); /* * set up enough so that it can read an inode */ - sb->s_op = &ext3_sops; - sb->s_export_op = &ext3_export_ops; - sb->s_xattr = ext3_xattr_handlers; + sb->s_op = &ext4_sops; + sb->s_export_op = &ext4_export_ops; + sb->s_xattr = ext4_xattr_handlers; #ifdef CONFIG_QUOTA - sb->s_qcop = &ext3_qctl_operations; - sb->dq_op = &ext3_quota_operations; + sb->s_qcop = &ext4_qctl_operations; + sb->dq_op = &ext4_quota_operations; #endif INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */ sb->s_root = NULL; needs_recovery = (es->s_last_orphan != 0 || - EXT3_HAS_INCOMPAT_FEATURE(sb, - EXT3_FEATURE_INCOMPAT_RECOVER)); + EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_RECOVER)); /* * The first inode we look at is the journal inode. Don't try * root first: it may be modified in the journal! */ if (!test_opt(sb, NOLOAD) && - EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL)) { - if (ext3_load_journal(sb, es, journal_devnum)) + EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) { + if (ext4_load_journal(sb, es, journal_devnum)) goto failed_mount3; } else if (journal_inum) { - if (ext3_create_journal(sb, es, journal_inum)) + if (ext4_create_journal(sb, es, journal_inum)) goto failed_mount3; } else { if (!silent) printk (KERN_ERR - "ext3: No journal on filesystem on %s\n", + "ext4: No journal on filesystem on %s\n", sb->s_id); goto failed_mount3; } @@ -1729,11 +1729,11 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, JOURNAL_DATA); break; - case EXT3_MOUNT_ORDERED_DATA: - case EXT3_MOUNT_WRITEBACK_DATA: + case EXT4_MOUNT_ORDERED_DATA: + case EXT4_MOUNT_WRITEBACK_DATA: if (!journal_check_available_features (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) { - printk(KERN_ERR "EXT3-fs: Journal does not support " + printk(KERN_ERR "EXT4-fs: Journal does not support " "requested data journaling mode\n"); goto failed_mount4; } @@ -1742,8 +1742,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) } if (test_opt(sb, NOBH)) { - if (!(test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)) { - printk(KERN_WARNING "EXT3-fs: Ignoring nobh option - " + if (!(test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)) { + printk(KERN_WARNING "EXT4-fs: Ignoring nobh option - " "its supported only with writeback mode\n"); clear_opt(sbi->s_mount_opt, NOBH); } @@ -1753,21 +1753,21 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) * so we can safely mount the rest of the filesystem now. */ - root = iget(sb, EXT3_ROOT_INO); + root = iget(sb, EXT4_ROOT_INO); sb->s_root = d_alloc_root(root); if (!sb->s_root) { - printk(KERN_ERR "EXT3-fs: get root inode failed\n"); + printk(KERN_ERR "EXT4-fs: get root inode failed\n"); iput(root); goto failed_mount4; } if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { dput(sb->s_root); sb->s_root = NULL; - printk(KERN_ERR "EXT3-fs: corrupt root inode, run e2fsck\n"); + printk(KERN_ERR "EXT4-fs: corrupt root inode, run e2fsck\n"); goto failed_mount4; } - ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY); + ext4_setup_super (sb, es, sb->s_flags & MS_RDONLY); /* * akpm: core read_super() calls in here with the superblock locked. * That deadlocks, because orphan cleanup needs to lock the superblock @@ -1776,23 +1776,23 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) * and aviro says that's the only reason for hanging onto the * superblock lock. */ - EXT3_SB(sb)->s_mount_state |= EXT3_ORPHAN_FS; - ext3_orphan_cleanup(sb, es); - EXT3_SB(sb)->s_mount_state &= ~EXT3_ORPHAN_FS; + EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS; + ext4_orphan_cleanup(sb, es); + EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS; if (needs_recovery) - printk (KERN_INFO "EXT3-fs: recovery complete.\n"); - ext3_mark_recovery_complete(sb, es); - printk (KERN_INFO "EXT3-fs: mounted filesystem with %s data mode.\n", - test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA ? "journal": - test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA ? "ordered": + printk (KERN_INFO "EXT4-fs: recovery complete.\n"); + ext4_mark_recovery_complete(sb, es); + printk (KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n", + test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal": + test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered": "writeback"); lock_kernel(); return 0; -cantfind_ext3: +cantfind_ext4: if (!silent) - printk(KERN_ERR "VFS: Can't find ext3 filesystem on dev %s.\n", + printk(KERN_ERR "VFS: Can't find ext4 filesystem on dev %s.\n", sb->s_id); goto failed_mount; @@ -1811,7 +1811,7 @@ failed_mount: for (i = 0; i < MAXQUOTAS; i++) kfree(sbi->s_qf_names[i]); #endif - ext3_blkdev_remove(sbi); + ext4_blkdev_remove(sbi); brelse(bh); out_fail: sb->s_fs_info = NULL; @@ -1825,13 +1825,13 @@ out_fail: * initial mount, once the journal has been initialised but before we've * done any recovery; and again on any subsequent remount. */ -static void ext3_init_journal_params(struct super_block *sb, journal_t *journal) +static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) { - struct ext3_sb_info *sbi = EXT3_SB(sb); + struct ext4_sb_info *sbi = EXT4_SB(sb); if (sbi->s_commit_interval) journal->j_commit_interval = sbi->s_commit_interval; - /* We could also set up an ext3-specific default for the commit + /* We could also set up an ext4-specific default for the commit * interval here, but for now we'll just fall back to the jbd * default. */ @@ -1843,7 +1843,7 @@ static void ext3_init_journal_params(struct super_block *sb, journal_t *journal) spin_unlock(&journal->j_state_lock); } -static journal_t *ext3_get_journal(struct super_block *sb, +static journal_t *ext4_get_journal(struct super_block *sb, unsigned int journal_inum) { struct inode *journal_inode; @@ -1855,55 +1855,55 @@ static journal_t *ext3_get_journal(struct super_block *sb, journal_inode = iget(sb, journal_inum); if (!journal_inode) { - printk(KERN_ERR "EXT3-fs: no journal found.\n"); + printk(KERN_ERR "EXT4-fs: no journal found.\n"); return NULL; } if (!journal_inode->i_nlink) { make_bad_inode(journal_inode); iput(journal_inode); - printk(KERN_ERR "EXT3-fs: journal inode is deleted.\n"); + printk(KERN_ERR "EXT4-fs: journal inode is deleted.\n"); return NULL; } jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", journal_inode, journal_inode->i_size); if (is_bad_inode(journal_inode) || !S_ISREG(journal_inode->i_mode)) { - printk(KERN_ERR "EXT3-fs: invalid journal inode.\n"); + printk(KERN_ERR "EXT4-fs: invalid journal inode.\n"); iput(journal_inode); return NULL; } journal = journal_init_inode(journal_inode); if (!journal) { - printk(KERN_ERR "EXT3-fs: Could not load journal inode\n"); + printk(KERN_ERR "EXT4-fs: Could not load journal inode\n"); iput(journal_inode); return NULL; } journal->j_private = sb; - ext3_init_journal_params(sb, journal); + ext4_init_journal_params(sb, journal); return journal; } -static journal_t *ext3_get_dev_journal(struct super_block *sb, +static journal_t *ext4_get_dev_journal(struct super_block *sb, dev_t j_dev) { struct buffer_head * bh; journal_t *journal; - ext3_fsblk_t start; - ext3_fsblk_t len; + ext4_fsblk_t start; + ext4_fsblk_t len; int hblock, blocksize; - ext3_fsblk_t sb_block; + ext4_fsblk_t sb_block; unsigned long offset; - struct ext3_super_block * es; + struct ext4_super_block * es; struct block_device *bdev; - bdev = ext3_blkdev_get(j_dev); + bdev = ext4_blkdev_get(j_dev); if (bdev == NULL) return NULL; if (bd_claim(bdev, sb)) { printk(KERN_ERR - "EXT3: failed to claim external journal device.\n"); + "EXT4: failed to claim external journal device.\n"); blkdev_put(bdev); return NULL; } @@ -1912,31 +1912,31 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb, hblock = bdev_hardsect_size(bdev); if (blocksize < hblock) { printk(KERN_ERR - "EXT3-fs: blocksize too small for journal device.\n"); + "EXT4-fs: blocksize too small for journal device.\n"); goto out_bdev; } - sb_block = EXT3_MIN_BLOCK_SIZE / blocksize; - offset = EXT3_MIN_BLOCK_SIZE % blocksize; + sb_block = EXT4_MIN_BLOCK_SIZE / blocksize; + offset = EXT4_MIN_BLOCK_SIZE % blocksize; set_blocksize(bdev, blocksize); if (!(bh = __bread(bdev, sb_block, blocksize))) { - printk(KERN_ERR "EXT3-fs: couldn't read superblock of " + printk(KERN_ERR "EXT4-fs: couldn't read superblock of " "external journal\n"); goto out_bdev; } - es = (struct ext3_super_block *) (((char *)bh->b_data) + offset); - if ((le16_to_cpu(es->s_magic) != EXT3_SUPER_MAGIC) || + es = (struct ext4_super_block *) (((char *)bh->b_data) + offset); + if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) || !(le32_to_cpu(es->s_feature_incompat) & - EXT3_FEATURE_INCOMPAT_JOURNAL_DEV)) { - printk(KERN_ERR "EXT3-fs: external journal has " + EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) { + printk(KERN_ERR "EXT4-fs: external journal has " "bad superblock\n"); brelse(bh); goto out_bdev; } - if (memcmp(EXT3_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) { - printk(KERN_ERR "EXT3-fs: journal UUID does not match\n"); + if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) { + printk(KERN_ERR "EXT4-fs: journal UUID does not match\n"); brelse(bh); goto out_bdev; } @@ -1948,34 +1948,34 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb, journal = journal_init_dev(bdev, sb->s_bdev, start, len, blocksize); if (!journal) { - printk(KERN_ERR "EXT3-fs: failed to create device journal\n"); + printk(KERN_ERR "EXT4-fs: failed to create device journal\n"); goto out_bdev; } journal->j_private = sb; ll_rw_block(READ, 1, &journal->j_sb_buffer); wait_on_buffer(journal->j_sb_buffer); if (!buffer_uptodate(journal->j_sb_buffer)) { - printk(KERN_ERR "EXT3-fs: I/O error on journal device\n"); + printk(KERN_ERR "EXT4-fs: I/O error on journal device\n"); goto out_journal; } if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) { - printk(KERN_ERR "EXT3-fs: External journal has more than one " + printk(KERN_ERR "EXT4-fs: External journal has more than one " "user (unsupported) - %d\n", be32_to_cpu(journal->j_superblock->s_nr_users)); goto out_journal; } - EXT3_SB(sb)->journal_bdev = bdev; - ext3_init_journal_params(sb, journal); + EXT4_SB(sb)->journal_bdev = bdev; + ext4_init_journal_params(sb, journal); return journal; out_journal: journal_destroy(journal); out_bdev: - ext3_blkdev_put(bdev); + ext4_blkdev_put(bdev); return NULL; } -static int ext3_load_journal(struct super_block *sb, - struct ext3_super_block *es, +static int ext4_load_journal(struct super_block *sb, + struct ext4_super_block *es, unsigned long journal_devnum) { journal_t *journal; @@ -1986,7 +1986,7 @@ static int ext3_load_journal(struct super_block *sb, if (journal_devnum && journal_devnum != le32_to_cpu(es->s_journal_dev)) { - printk(KERN_INFO "EXT3-fs: external journal device major/minor " + printk(KERN_INFO "EXT4-fs: external journal device major/minor " "numbers have changed\n"); journal_dev = new_decode_dev(journal_devnum); } else @@ -2000,56 +2000,56 @@ static int ext3_load_journal(struct super_block *sb, * can get read-write access to the device. */ - if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER)) { + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) { if (sb->s_flags & MS_RDONLY) { - printk(KERN_INFO "EXT3-fs: INFO: recovery " + printk(KERN_INFO "EXT4-fs: INFO: recovery " "required on readonly filesystem.\n"); if (really_read_only) { - printk(KERN_ERR "EXT3-fs: write access " + printk(KERN_ERR "EXT4-fs: write access " "unavailable, cannot proceed.\n"); return -EROFS; } - printk (KERN_INFO "EXT3-fs: write access will " + printk (KERN_INFO "EXT4-fs: write access will " "be enabled during recovery.\n"); } } if (journal_inum && journal_dev) { - printk(KERN_ERR "EXT3-fs: filesystem has both journal " + printk(KERN_ERR "EXT4-fs: filesystem has both journal " "and inode journals!\n"); return -EINVAL; } if (journal_inum) { - if (!(journal = ext3_get_journal(sb, journal_inum))) + if (!(journal = ext4_get_journal(sb, journal_inum))) return -EINVAL; } else { - if (!(journal = ext3_get_dev_journal(sb, journal_dev))) + if (!(journal = ext4_get_dev_journal(sb, journal_dev))) return -EINVAL; } if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { err = journal_update_format(journal); if (err) { - printk(KERN_ERR "EXT3-fs: error updating journal.\n"); + printk(KERN_ERR "EXT4-fs: error updating journal.\n"); journal_destroy(journal); return err; } } - if (!EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER)) + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) err = journal_wipe(journal, !really_read_only); if (!err) err = journal_load(journal); if (err) { - printk(KERN_ERR "EXT3-fs: error loading journal.\n"); + printk(KERN_ERR "EXT4-fs: error loading journal.\n"); journal_destroy(journal); return err; } - EXT3_SB(sb)->s_journal = journal; - ext3_clear_journal_err(sb, es); + EXT4_SB(sb)->s_journal = journal; + ext4_clear_journal_err(sb, es); if (journal_devnum && journal_devnum != le32_to_cpu(es->s_journal_dev)) { @@ -2057,62 +2057,62 @@ static int ext3_load_journal(struct super_block *sb, sb->s_dirt = 1; /* Make sure we flush the recovery flag to disk. */ - ext3_commit_super(sb, es, 1); + ext4_commit_super(sb, es, 1); } return 0; } -static int ext3_create_journal(struct super_block * sb, - struct ext3_super_block * es, +static int ext4_create_journal(struct super_block * sb, + struct ext4_super_block * es, unsigned int journal_inum) { journal_t *journal; if (sb->s_flags & MS_RDONLY) { - printk(KERN_ERR "EXT3-fs: readonly filesystem when trying to " + printk(KERN_ERR "EXT4-fs: readonly filesystem when trying to " "create journal.\n"); return -EROFS; } - if (!(journal = ext3_get_journal(sb, journal_inum))) + if (!(journal = ext4_get_journal(sb, journal_inum))) return -EINVAL; - printk(KERN_INFO "EXT3-fs: creating new journal on inode %u\n", + printk(KERN_INFO "EXT4-fs: creating new journal on inode %u\n", journal_inum); if (journal_create(journal)) { - printk(KERN_ERR "EXT3-fs: error creating journal.\n"); + printk(KERN_ERR "EXT4-fs: error creating journal.\n"); journal_destroy(journal); return -EIO; } - EXT3_SB(sb)->s_journal = journal; + EXT4_SB(sb)->s_journal = journal; - ext3_update_dynamic_rev(sb); - EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); - EXT3_SET_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL); + ext4_update_dynamic_rev(sb); + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL); es->s_journal_inum = cpu_to_le32(journal_inum); sb->s_dirt = 1; /* Make sure we flush the recovery flag to disk. */ - ext3_commit_super(sb, es, 1); + ext4_commit_super(sb, es, 1); return 0; } -static void ext3_commit_super (struct super_block * sb, - struct ext3_super_block * es, +static void ext4_commit_super (struct super_block * sb, + struct ext4_super_block * es, int sync) { - struct buffer_head *sbh = EXT3_SB(sb)->s_sbh; + struct buffer_head *sbh = EXT4_SB(sb)->s_sbh; if (!sbh) return; es->s_wtime = cpu_to_le32(get_seconds()); - es->s_free_blocks_count = cpu_to_le32(ext3_count_free_blocks(sb)); - es->s_free_inodes_count = cpu_to_le32(ext3_count_free_inodes(sb)); + es->s_free_blocks_count = cpu_to_le32(ext4_count_free_blocks(sb)); + es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb)); BUFFER_TRACE(sbh, "marking dirty"); mark_buffer_dirty(sbh); if (sync) @@ -2125,18 +2125,18 @@ static void ext3_commit_super (struct super_block * sb, * remounting) the filesystem readonly, then we will end up with a * consistent fs on disk. Record that fact. */ -static void ext3_mark_recovery_complete(struct super_block * sb, - struct ext3_super_block * es) +static void ext4_mark_recovery_complete(struct super_block * sb, + struct ext4_super_block * es) { - journal_t *journal = EXT3_SB(sb)->s_journal; + journal_t *journal = EXT4_SB(sb)->s_journal; journal_lock_updates(journal); journal_flush(journal); - if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER) && + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) && sb->s_flags & MS_RDONLY) { - EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); + EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); sb->s_dirt = 0; - ext3_commit_super(sb, es, 1); + ext4_commit_super(sb, es, 1); } journal_unlock_updates(journal); } @@ -2146,33 +2146,33 @@ static void ext3_mark_recovery_complete(struct super_block * sb, * has recorded an error from a previous lifetime, move that error to the * main filesystem now. */ -static void ext3_clear_journal_err(struct super_block * sb, - struct ext3_super_block * es) +static void ext4_clear_journal_err(struct super_block * sb, + struct ext4_super_block * es) { journal_t *journal; int j_errno; const char *errstr; - journal = EXT3_SB(sb)->s_journal; + journal = EXT4_SB(sb)->s_journal; /* * Now check for any error status which may have been recorded in the - * journal by a prior ext3_error() or ext3_abort() + * journal by a prior ext4_error() or ext4_abort() */ j_errno = journal_errno(journal); if (j_errno) { char nbuf[16]; - errstr = ext3_decode_error(sb, j_errno, nbuf); - ext3_warning(sb, __FUNCTION__, "Filesystem error recorded " + errstr = ext4_decode_error(sb, j_errno, nbuf); + ext4_warning(sb, __FUNCTION__, "Filesystem error recorded " "from previous mount: %s", errstr); - ext3_warning(sb, __FUNCTION__, "Marking fs in need of " + ext4_warning(sb, __FUNCTION__, "Marking fs in need of " "filesystem check."); - EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; - es->s_state |= cpu_to_le16(EXT3_ERROR_FS); - ext3_commit_super (sb, es, 1); + EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; + es->s_state |= cpu_to_le16(EXT4_ERROR_FS); + ext4_commit_super (sb, es, 1); journal_clear_err(journal); } @@ -2182,7 +2182,7 @@ static void ext3_clear_journal_err(struct super_block * sb, * Force the running and committing transactions to commit, * and wait on the commit. */ -int ext3_force_commit(struct super_block *sb) +int ext4_force_commit(struct super_block *sb) { journal_t *journal; int ret; @@ -2190,14 +2190,14 @@ int ext3_force_commit(struct super_block *sb) if (sb->s_flags & MS_RDONLY) return 0; - journal = EXT3_SB(sb)->s_journal; + journal = EXT4_SB(sb)->s_journal; sb->s_dirt = 0; - ret = ext3_journal_force_commit(journal); + ret = ext4_journal_force_commit(journal); return ret; } /* - * Ext3 always journals updates to the superblock itself, so we don't + * Ext4 always journals updates to the superblock itself, so we don't * have to propagate any other updates to the superblock on disk at this * point. Just start an async writeback to get the buffers on their way * to the disk. @@ -2205,21 +2205,21 @@ int ext3_force_commit(struct super_block *sb) * This implicitly triggers the writebehind on sync(). */ -static void ext3_write_super (struct super_block * sb) +static void ext4_write_super (struct super_block * sb) { if (mutex_trylock(&sb->s_lock) != 0) BUG(); sb->s_dirt = 0; } -static int ext3_sync_fs(struct super_block *sb, int wait) +static int ext4_sync_fs(struct super_block *sb, int wait) { tid_t target; sb->s_dirt = 0; - if (journal_start_commit(EXT3_SB(sb)->s_journal, &target)) { + if (journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { if (wait) - log_wait_commit(EXT3_SB(sb)->s_journal, target); + log_wait_commit(EXT4_SB(sb)->s_journal, target); } return 0; } @@ -2228,20 +2228,20 @@ static int ext3_sync_fs(struct super_block *sb, int wait) * LVM calls this function before a (read-only) snapshot is created. This * gives us a chance to flush the journal completely and mark the fs clean. */ -static void ext3_write_super_lockfs(struct super_block *sb) +static void ext4_write_super_lockfs(struct super_block *sb) { sb->s_dirt = 0; if (!(sb->s_flags & MS_RDONLY)) { - journal_t *journal = EXT3_SB(sb)->s_journal; + journal_t *journal = EXT4_SB(sb)->s_journal; /* Now we set up the journal barrier. */ journal_lock_updates(journal); journal_flush(journal); /* Journal blocked and flushed, clear needs_recovery flag. */ - EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); - ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1); + EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1); } } @@ -2249,25 +2249,25 @@ static void ext3_write_super_lockfs(struct super_block *sb) * Called by LVM after the snapshot is done. We need to reset the RECOVER * flag here, even though the filesystem is not technically dirty yet. */ -static void ext3_unlockfs(struct super_block *sb) +static void ext4_unlockfs(struct super_block *sb) { if (!(sb->s_flags & MS_RDONLY)) { lock_super(sb); /* Reser the needs_recovery flag before the fs is unlocked. */ - EXT3_SET_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); - ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1); + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1); unlock_super(sb); - journal_unlock_updates(EXT3_SB(sb)->s_journal); + journal_unlock_updates(EXT4_SB(sb)->s_journal); } } -static int ext3_remount (struct super_block * sb, int * flags, char * data) +static int ext4_remount (struct super_block * sb, int * flags, char * data) { - struct ext3_super_block * es; - struct ext3_sb_info *sbi = EXT3_SB(sb); - ext3_fsblk_t n_blocks_count = 0; + struct ext4_super_block * es; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_fsblk_t n_blocks_count = 0; unsigned long old_sb_flags; - struct ext3_mount_options old_opts; + struct ext4_mount_options old_opts; int err; #ifdef CONFIG_QUOTA int i; @@ -2293,19 +2293,19 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) goto restore_opts; } - if (sbi->s_mount_opt & EXT3_MOUNT_ABORT) - ext3_abort(sb, __FUNCTION__, "Abort forced by user"); + if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) + ext4_abort(sb, __FUNCTION__, "Abort forced by user"); sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | - ((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + ((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); es = sbi->s_es; - ext3_init_journal_params(sb, sbi->s_journal); + ext4_init_journal_params(sb, sbi->s_journal); if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY) || n_blocks_count > le32_to_cpu(es->s_blocks_count)) { - if (sbi->s_mount_opt & EXT3_MOUNT_ABORT) { + if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) { err = -EROFS; goto restore_opts; } @@ -2322,16 +2322,16 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) * readonly, and if so set the rdonly flag and then * mark the partition as valid again. */ - if (!(es->s_state & cpu_to_le16(EXT3_VALID_FS)) && - (sbi->s_mount_state & EXT3_VALID_FS)) + if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) && + (sbi->s_mount_state & EXT4_VALID_FS)) es->s_state = cpu_to_le16(sbi->s_mount_state); - ext3_mark_recovery_complete(sb, es); + ext4_mark_recovery_complete(sb, es); } else { __le32 ret; - if ((ret = EXT3_HAS_RO_COMPAT_FEATURE(sb, - ~EXT3_FEATURE_RO_COMPAT_SUPP))) { - printk(KERN_WARNING "EXT3-fs: %s: couldn't " + if ((ret = EXT4_HAS_RO_COMPAT_FEATURE(sb, + ~EXT4_FEATURE_RO_COMPAT_SUPP))) { + printk(KERN_WARNING "EXT4-fs: %s: couldn't " "remount RDWR because of unsupported " "optional features (%x).\n", sb->s_id, le32_to_cpu(ret)); @@ -2344,11 +2344,11 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) * been changed by e2fsck since we originally mounted * the partition.) */ - ext3_clear_journal_err(sb, es); + ext4_clear_journal_err(sb, es); sbi->s_mount_state = le16_to_cpu(es->s_state); - if ((err = ext3_group_extend(sb, es, n_blocks_count))) + if ((err = ext4_group_extend(sb, es, n_blocks_count))) goto restore_opts; - if (!ext3_setup_super (sb, es, 0)) + if (!ext4_setup_super (sb, es, 0)) sb->s_flags &= ~MS_RDONLY; } } @@ -2378,19 +2378,19 @@ restore_opts: return err; } -static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) +static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) { struct super_block *sb = dentry->d_sb; - struct ext3_sb_info *sbi = EXT3_SB(sb); - struct ext3_super_block *es = sbi->s_es; - ext3_fsblk_t overhead; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_fsblk_t overhead; int i; if (test_opt (sb, MINIX_DF)) overhead = 0; else { unsigned long ngroups; - ngroups = EXT3_SB(sb)->s_groups_count; + ngroups = EXT4_SB(sb)->s_groups_count; smp_rmb(); /* @@ -2409,8 +2409,8 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) * feature is turned on, then not all groups have this. */ for (i = 0; i < ngroups; i++) { - overhead += ext3_bg_has_super(sb, i) + - ext3_bg_num_gdb(sb, i); + overhead += ext4_bg_has_super(sb, i) + + ext4_bg_num_gdb(sb, i); cond_resched(); } @@ -2418,10 +2418,10 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) * Every block group has an inode bitmap, a block * bitmap, and an inode table. */ - overhead += (ngroups * (2 + EXT3_SB(sb)->s_itb_per_group)); + overhead += (ngroups * (2 + EXT4_SB(sb)->s_itb_per_group)); } - buf->f_type = EXT3_SUPER_MAGIC; + buf->f_type = EXT4_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; buf->f_blocks = le32_to_cpu(es->s_blocks_count) - overhead; buf->f_bfree = percpu_counter_sum(&sbi->s_freeblocks_counter); @@ -2430,14 +2430,14 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) buf->f_bavail = 0; buf->f_files = le32_to_cpu(es->s_inodes_count); buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter); - buf->f_namelen = EXT3_NAME_LEN; + buf->f_namelen = EXT4_NAME_LEN; return 0; } /* Helper function for writing quotas on sync - we need to start transaction before quota file * is locked for write. Otherwise the are possible deadlocks: * Process 1 Process 2 - * ext3_create() quota_sync() + * ext4_create() quota_sync() * journal_start() write_dquot() * DQUOT_INIT() down(dqio_mutex) * down(dqio_mutex) journal_start() @@ -2451,111 +2451,111 @@ static inline struct inode *dquot_to_inode(struct dquot *dquot) return sb_dqopt(dquot->dq_sb)->files[dquot->dq_type]; } -static int ext3_dquot_initialize(struct inode *inode, int type) +static int ext4_dquot_initialize(struct inode *inode, int type) { handle_t *handle; int ret, err; /* We may create quota structure so we need to reserve enough blocks */ - handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)); + handle = ext4_journal_start(inode, 2*EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_initialize(inode, type); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; } -static int ext3_dquot_drop(struct inode *inode) +static int ext4_dquot_drop(struct inode *inode) { handle_t *handle; int ret, err; /* We may delete quota structure so we need to reserve enough blocks */ - handle = ext3_journal_start(inode, 2*EXT3_QUOTA_DEL_BLOCKS(inode->i_sb)); + handle = ext4_journal_start(inode, 2*EXT4_QUOTA_DEL_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_drop(inode); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; } -static int ext3_write_dquot(struct dquot *dquot) +static int ext4_write_dquot(struct dquot *dquot) { int ret, err; handle_t *handle; struct inode *inode; inode = dquot_to_inode(dquot); - handle = ext3_journal_start(inode, - EXT3_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); + handle = ext4_journal_start(inode, + EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_commit(dquot); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; } -static int ext3_acquire_dquot(struct dquot *dquot) +static int ext4_acquire_dquot(struct dquot *dquot) { int ret, err; handle_t *handle; - handle = ext3_journal_start(dquot_to_inode(dquot), - EXT3_QUOTA_INIT_BLOCKS(dquot->dq_sb)); + handle = ext4_journal_start(dquot_to_inode(dquot), + EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_acquire(dquot); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; } -static int ext3_release_dquot(struct dquot *dquot) +static int ext4_release_dquot(struct dquot *dquot) { int ret, err; handle_t *handle; - handle = ext3_journal_start(dquot_to_inode(dquot), - EXT3_QUOTA_DEL_BLOCKS(dquot->dq_sb)); + handle = ext4_journal_start(dquot_to_inode(dquot), + EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_release(dquot); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; } -static int ext3_mark_dquot_dirty(struct dquot *dquot) +static int ext4_mark_dquot_dirty(struct dquot *dquot) { /* Are we journalling quotas? */ - if (EXT3_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] || - EXT3_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) { + if (EXT4_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] || + EXT4_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) { dquot_mark_dquot_dirty(dquot); - return ext3_write_dquot(dquot); + return ext4_write_dquot(dquot); } else { return dquot_mark_dquot_dirty(dquot); } } -static int ext3_write_info(struct super_block *sb, int type) +static int ext4_write_info(struct super_block *sb, int type) { int ret, err; handle_t *handle; /* Data block + inode block */ - handle = ext3_journal_start(sb->s_root->d_inode, 2); + handle = ext4_journal_start(sb->s_root->d_inode, 2); if (IS_ERR(handle)) return PTR_ERR(handle); ret = dquot_commit_info(sb, type); - err = ext3_journal_stop(handle); + err = ext4_journal_stop(handle); if (!ret) ret = err; return ret; @@ -2565,16 +2565,16 @@ static int ext3_write_info(struct super_block *sb, int type) * Turn on quotas during mount time - we need to find * the quota file and such... */ -static int ext3_quota_on_mount(struct super_block *sb, int type) +static int ext4_quota_on_mount(struct super_block *sb, int type) { - return vfs_quota_on_mount(sb, EXT3_SB(sb)->s_qf_names[type], - EXT3_SB(sb)->s_jquota_fmt, type); + return vfs_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type], + EXT4_SB(sb)->s_jquota_fmt, type); } /* * Standard function to be called on quota_on */ -static int ext3_quota_on(struct super_block *sb, int type, int format_id, +static int ext4_quota_on(struct super_block *sb, int type, int format_id, char *path) { int err; @@ -2583,8 +2583,8 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, if (!test_opt(sb, QUOTA)) return -EINVAL; /* Not journalling quota? */ - if (!EXT3_SB(sb)->s_qf_names[USRQUOTA] && - !EXT3_SB(sb)->s_qf_names[GRPQUOTA]) + if (!EXT4_SB(sb)->s_qf_names[USRQUOTA] && + !EXT4_SB(sb)->s_qf_names[GRPQUOTA]) return vfs_quota_on(sb, type, format_id, path); err = path_lookup(path, LOOKUP_FOLLOW, &nd); if (err) @@ -2597,7 +2597,7 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, /* Quotafile not of fs root? */ if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode) printk(KERN_WARNING - "EXT3-fs: Quota file not on filesystem root. " + "EXT4-fs: Quota file not on filesystem root. " "Journalled quota will not work.\n"); path_release(&nd); return vfs_quota_on(sb, type, format_id, path); @@ -2607,11 +2607,11 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, * acquiring the locks... As quota files are never truncated and quota code * itself serializes the operations (and noone else should touch the files) * we don't have to be afraid of races */ -static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, +static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) { struct inode *inode = sb_dqopt(sb)->files[type]; - sector_t blk = off >> EXT3_BLOCK_SIZE_BITS(sb); + sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); int err = 0; int offset = off & (sb->s_blocksize - 1); int tocopy; @@ -2627,7 +2627,7 @@ static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, while (toread > 0) { tocopy = sb->s_blocksize - offset < toread ? sb->s_blocksize - offset : toread; - bh = ext3_bread(NULL, inode, blk, 0, &err); + bh = ext4_bread(NULL, inode, blk, 0, &err); if (err) return err; if (!bh) /* A hole? */ @@ -2645,15 +2645,15 @@ static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, /* Write to quotafile (we know the transaction is already started and has * enough credits) */ -static ssize_t ext3_quota_write(struct super_block *sb, int type, +static ssize_t ext4_quota_write(struct super_block *sb, int type, const char *data, size_t len, loff_t off) { struct inode *inode = sb_dqopt(sb)->files[type]; - sector_t blk = off >> EXT3_BLOCK_SIZE_BITS(sb); + sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); int err = 0; int offset = off & (sb->s_blocksize - 1); int tocopy; - int journal_quota = EXT3_SB(sb)->s_qf_names[type] != NULL; + int journal_quota = EXT4_SB(sb)->s_qf_names[type] != NULL; size_t towrite = len; struct buffer_head *bh; handle_t *handle = journal_current_handle(); @@ -2662,11 +2662,11 @@ static ssize_t ext3_quota_write(struct super_block *sb, int type, while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; - bh = ext3_bread(handle, inode, blk, 1, &err); + bh = ext4_bread(handle, inode, blk, 1, &err); if (!bh) goto out; if (journal_quota) { - err = ext3_journal_get_write_access(handle, bh); + err = ext4_journal_get_write_access(handle, bh); if (err) { brelse(bh); goto out; @@ -2677,10 +2677,10 @@ static ssize_t ext3_quota_write(struct super_block *sb, int type, flush_dcache_page(bh->b_page); unlock_buffer(bh); if (journal_quota) - err = ext3_journal_dirty_metadata(handle, bh); + err = ext4_journal_dirty_metadata(handle, bh); else { /* Always do at least ordered writes for quotas */ - err = ext3_journal_dirty_data(handle, bh); + err = ext4_journal_dirty_data(handle, bh); mark_buffer_dirty(bh); } brelse(bh); @@ -2696,59 +2696,59 @@ out: return err; if (inode->i_size < off+len-towrite) { i_size_write(inode, off+len-towrite); - EXT3_I(inode)->i_disksize = inode->i_size; + EXT4_I(inode)->i_disksize = inode->i_size; } inode->i_version++; inode->i_mtime = inode->i_ctime = CURRENT_TIME; - ext3_mark_inode_dirty(handle, inode); + ext4_mark_inode_dirty(handle, inode); mutex_unlock(&inode->i_mutex); return len - towrite; } #endif -static int ext3_get_sb(struct file_system_type *fs_type, +static int ext4_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, struct vfsmount *mnt) { - return get_sb_bdev(fs_type, flags, dev_name, data, ext3_fill_super, mnt); + return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt); } -static struct file_system_type ext3_fs_type = { +static struct file_system_type ext4dev_fs_type = { .owner = THIS_MODULE, - .name = "ext3", - .get_sb = ext3_get_sb, + .name = "ext4dev", + .get_sb = ext4_get_sb, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; -static int __init init_ext3_fs(void) +static int __init init_ext4_fs(void) { - int err = init_ext3_xattr(); + int err = init_ext4_xattr(); if (err) return err; err = init_inodecache(); if (err) goto out1; - err = register_filesystem(&ext3_fs_type); + err = register_filesystem(&ext4dev_fs_type); if (err) goto out; return 0; out: destroy_inodecache(); out1: - exit_ext3_xattr(); + exit_ext4_xattr(); return err; } -static void __exit exit_ext3_fs(void) +static void __exit exit_ext4_fs(void) { - unregister_filesystem(&ext3_fs_type); + unregister_filesystem(&ext4dev_fs_type); destroy_inodecache(); - exit_ext3_xattr(); + exit_ext4_xattr(); } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); -MODULE_DESCRIPTION("Second Extended Filesystem with journaling extensions"); +MODULE_DESCRIPTION("Fourth Extended Filesystem with extents"); MODULE_LICENSE("GPL"); -module_init(init_ext3_fs) -module_exit(exit_ext3_fs) +module_init(init_ext4_fs) +module_exit(exit_ext4_fs) diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c index 4f79122..9e4c75f 100644 --- a/fs/ext4/symlink.c +++ b/fs/ext4/symlink.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/symlink.c + * linux/fs/ext4/symlink.c * * Only fast symlinks left here - the rest is done by generic code. AV, 1999 * @@ -14,41 +14,41 @@ * * Copyright (C) 1991, 1992 Linus Torvalds * - * ext3 symlink handling code + * ext4 symlink handling code */ #include #include -#include +#include #include #include "xattr.h" -static void * ext3_follow_link(struct dentry *dentry, struct nameidata *nd) +static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd) { - struct ext3_inode_info *ei = EXT3_I(dentry->d_inode); + struct ext4_inode_info *ei = EXT4_I(dentry->d_inode); nd_set_link(nd, (char*)ei->i_data); return NULL; } -struct inode_operations ext3_symlink_inode_operations = { +struct inode_operations ext4_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = page_follow_link_light, .put_link = page_put_link, -#ifdef CONFIG_EXT3_FS_XATTR +#ifdef CONFIG_EXT4DEV_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, - .listxattr = ext3_listxattr, + .listxattr = ext4_listxattr, .removexattr = generic_removexattr, #endif }; -struct inode_operations ext3_fast_symlink_inode_operations = { +struct inode_operations ext4_fast_symlink_inode_operations = { .readlink = generic_readlink, - .follow_link = ext3_follow_link, -#ifdef CONFIG_EXT3_FS_XATTR + .follow_link = ext4_follow_link, +#ifdef CONFIG_EXT4DEV_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, - .listxattr = ext3_listxattr, + .listxattr = ext4_listxattr, .removexattr = generic_removexattr, #endif }; diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index f86f248..d3a4081 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1,10 +1,10 @@ /* - * linux/fs/ext3/xattr.c + * linux/fs/ext4/xattr.c * * Copyright (C) 2001-2003 Andreas Gruenbacher, * * Fix by Harrison Xing . - * Ext3 code with a lot of help from Eric Jarman . + * Ext4 code with a lot of help from Eric Jarman . * Extended attributes for symlinks and special files added per * suggestion of Luka Renko . * xattr consolidation Copyright (c) 2004 James Morris , @@ -43,7 +43,7 @@ * * Locking strategy * ---------------- - * EXT3_I(inode)->i_file_acl is protected by EXT3_I(inode)->xattr_sem. + * EXT4_I(inode)->i_file_acl is protected by EXT4_I(inode)->xattr_sem. * EA blocks are only changed if they are exclusive to an inode, so * holding xattr_sem also means that nothing but the EA block's reference * count can change. Multiple writers to the same block are synchronized @@ -53,27 +53,27 @@ #include #include #include -#include -#include +#include +#include #include #include #include #include "xattr.h" #include "acl.h" -#define BHDR(bh) ((struct ext3_xattr_header *)((bh)->b_data)) -#define ENTRY(ptr) ((struct ext3_xattr_entry *)(ptr)) +#define BHDR(bh) ((struct ext4_xattr_header *)((bh)->b_data)) +#define ENTRY(ptr) ((struct ext4_xattr_entry *)(ptr)) #define BFIRST(bh) ENTRY(BHDR(bh)+1) #define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0) #define IHDR(inode, raw_inode) \ - ((struct ext3_xattr_ibody_header *) \ + ((struct ext4_xattr_ibody_header *) \ ((void *)raw_inode + \ - EXT3_GOOD_OLD_INODE_SIZE + \ - EXT3_I(inode)->i_extra_isize)) -#define IFIRST(hdr) ((struct ext3_xattr_entry *)((hdr)+1)) + EXT4_GOOD_OLD_INODE_SIZE + \ + EXT4_I(inode)->i_extra_isize)) +#define IFIRST(hdr) ((struct ext4_xattr_entry *)((hdr)+1)) -#ifdef EXT3_XATTR_DEBUG +#ifdef EXT4_XATTR_DEBUG # define ea_idebug(inode, f...) do { \ printk(KERN_DEBUG "inode %s:%lu: ", \ inode->i_sb->s_id, inode->i_ino); \ @@ -93,47 +93,47 @@ # define ea_bdebug(f...) #endif -static void ext3_xattr_cache_insert(struct buffer_head *); -static struct buffer_head *ext3_xattr_cache_find(struct inode *, - struct ext3_xattr_header *, +static void ext4_xattr_cache_insert(struct buffer_head *); +static struct buffer_head *ext4_xattr_cache_find(struct inode *, + struct ext4_xattr_header *, struct mb_cache_entry **); -static void ext3_xattr_rehash(struct ext3_xattr_header *, - struct ext3_xattr_entry *); +static void ext4_xattr_rehash(struct ext4_xattr_header *, + struct ext4_xattr_entry *); -static struct mb_cache *ext3_xattr_cache; +static struct mb_cache *ext4_xattr_cache; -static struct xattr_handler *ext3_xattr_handler_map[] = { - [EXT3_XATTR_INDEX_USER] = &ext3_xattr_user_handler, -#ifdef CONFIG_EXT3_FS_POSIX_ACL - [EXT3_XATTR_INDEX_POSIX_ACL_ACCESS] = &ext3_xattr_acl_access_handler, - [EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext3_xattr_acl_default_handler, +static struct xattr_handler *ext4_xattr_handler_map[] = { + [EXT4_XATTR_INDEX_USER] = &ext4_xattr_user_handler, +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + [EXT4_XATTR_INDEX_POSIX_ACL_ACCESS] = &ext4_xattr_acl_access_handler, + [EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext4_xattr_acl_default_handler, #endif - [EXT3_XATTR_INDEX_TRUSTED] = &ext3_xattr_trusted_handler, -#ifdef CONFIG_EXT3_FS_SECURITY - [EXT3_XATTR_INDEX_SECURITY] = &ext3_xattr_security_handler, + [EXT4_XATTR_INDEX_TRUSTED] = &ext4_xattr_trusted_handler, +#ifdef CONFIG_EXT4DEV_FS_SECURITY + [EXT4_XATTR_INDEX_SECURITY] = &ext4_xattr_security_handler, #endif }; -struct xattr_handler *ext3_xattr_handlers[] = { - &ext3_xattr_user_handler, - &ext3_xattr_trusted_handler, -#ifdef CONFIG_EXT3_FS_POSIX_ACL - &ext3_xattr_acl_access_handler, - &ext3_xattr_acl_default_handler, +struct xattr_handler *ext4_xattr_handlers[] = { + &ext4_xattr_user_handler, + &ext4_xattr_trusted_handler, +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + &ext4_xattr_acl_access_handler, + &ext4_xattr_acl_default_handler, #endif -#ifdef CONFIG_EXT3_FS_SECURITY - &ext3_xattr_security_handler, +#ifdef CONFIG_EXT4DEV_FS_SECURITY + &ext4_xattr_security_handler, #endif NULL }; static inline struct xattr_handler * -ext3_xattr_handler(int name_index) +ext4_xattr_handler(int name_index) { struct xattr_handler *handler = NULL; - if (name_index > 0 && name_index < ARRAY_SIZE(ext3_xattr_handler_map)) - handler = ext3_xattr_handler_map[name_index]; + if (name_index > 0 && name_index < ARRAY_SIZE(ext4_xattr_handler_map)) + handler = ext4_xattr_handler_map[name_index]; return handler; } @@ -143,16 +143,16 @@ ext3_xattr_handler(int name_index) * dentry->d_inode->i_mutex: don't care */ ssize_t -ext3_listxattr(struct dentry *dentry, char *buffer, size_t size) +ext4_listxattr(struct dentry *dentry, char *buffer, size_t size) { - return ext3_xattr_list(dentry->d_inode, buffer, size); + return ext4_xattr_list(dentry->d_inode, buffer, size); } static int -ext3_xattr_check_names(struct ext3_xattr_entry *entry, void *end) +ext4_xattr_check_names(struct ext4_xattr_entry *entry, void *end) { while (!IS_LAST_ENTRY(entry)) { - struct ext3_xattr_entry *next = EXT3_XATTR_NEXT(entry); + struct ext4_xattr_entry *next = EXT4_XATTR_NEXT(entry); if ((void *)next >= end) return -EIO; entry = next; @@ -161,19 +161,19 @@ ext3_xattr_check_names(struct ext3_xattr_entry *entry, void *end) } static inline int -ext3_xattr_check_block(struct buffer_head *bh) +ext4_xattr_check_block(struct buffer_head *bh) { int error; - if (BHDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || BHDR(bh)->h_blocks != cpu_to_le32(1)) return -EIO; - error = ext3_xattr_check_names(BFIRST(bh), bh->b_data + bh->b_size); + error = ext4_xattr_check_names(BFIRST(bh), bh->b_data + bh->b_size); return error; } static inline int -ext3_xattr_check_entry(struct ext3_xattr_entry *entry, size_t size) +ext4_xattr_check_entry(struct ext4_xattr_entry *entry, size_t size) { size_t value_size = le32_to_cpu(entry->e_value_size); @@ -184,10 +184,10 @@ ext3_xattr_check_entry(struct ext3_xattr_entry *entry, size_t size) } static int -ext3_xattr_find_entry(struct ext3_xattr_entry **pentry, int name_index, +ext4_xattr_find_entry(struct ext4_xattr_entry **pentry, int name_index, const char *name, size_t size, int sorted) { - struct ext3_xattr_entry *entry; + struct ext4_xattr_entry *entry; size_t name_len; int cmp = 1; @@ -195,7 +195,7 @@ ext3_xattr_find_entry(struct ext3_xattr_entry **pentry, int name_index, return -EINVAL; name_len = strlen(name); entry = *pentry; - for (; !IS_LAST_ENTRY(entry); entry = EXT3_XATTR_NEXT(entry)) { + for (; !IS_LAST_ENTRY(entry); entry = EXT4_XATTR_NEXT(entry)) { cmp = name_index - entry->e_name_index; if (!cmp) cmp = name_len - entry->e_name_len; @@ -205,17 +205,17 @@ ext3_xattr_find_entry(struct ext3_xattr_entry **pentry, int name_index, break; } *pentry = entry; - if (!cmp && ext3_xattr_check_entry(entry, size)) + if (!cmp && ext4_xattr_check_entry(entry, size)) return -EIO; return cmp ? -ENODATA : 0; } static int -ext3_xattr_block_get(struct inode *inode, int name_index, const char *name, +ext4_xattr_block_get(struct inode *inode, int name_index, const char *name, void *buffer, size_t buffer_size) { struct buffer_head *bh = NULL; - struct ext3_xattr_entry *entry; + struct ext4_xattr_entry *entry; size_t size; int error; @@ -223,24 +223,24 @@ ext3_xattr_block_get(struct inode *inode, int name_index, const char *name, name_index, name, buffer, (long)buffer_size); error = -ENODATA; - if (!EXT3_I(inode)->i_file_acl) + if (!EXT4_I(inode)->i_file_acl) goto cleanup; - ea_idebug(inode, "reading block %u", EXT3_I(inode)->i_file_acl); - bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + ea_idebug(inode, "reading block %u", EXT4_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); if (!bh) goto cleanup; ea_bdebug(bh, "b_count=%d, refcount=%d", atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); - if (ext3_xattr_check_block(bh)) { -bad_block: ext3_error(inode->i_sb, __FUNCTION__, + if (ext4_xattr_check_block(bh)) { +bad_block: ext4_error(inode->i_sb, __FUNCTION__, "inode %lu: bad block "E3FSBLK, inode->i_ino, - EXT3_I(inode)->i_file_acl); + EXT4_I(inode)->i_file_acl); error = -EIO; goto cleanup; } - ext3_xattr_cache_insert(bh); + ext4_xattr_cache_insert(bh); entry = BFIRST(bh); - error = ext3_xattr_find_entry(&entry, name_index, name, bh->b_size, 1); + error = ext4_xattr_find_entry(&entry, name_index, name, bh->b_size, 1); if (error == -EIO) goto bad_block; if (error) @@ -261,30 +261,30 @@ cleanup: } static int -ext3_xattr_ibody_get(struct inode *inode, int name_index, const char *name, +ext4_xattr_ibody_get(struct inode *inode, int name_index, const char *name, void *buffer, size_t buffer_size) { - struct ext3_xattr_ibody_header *header; - struct ext3_xattr_entry *entry; - struct ext3_inode *raw_inode; - struct ext3_iloc iloc; + struct ext4_xattr_ibody_header *header; + struct ext4_xattr_entry *entry; + struct ext4_inode *raw_inode; + struct ext4_iloc iloc; size_t size; void *end; int error; - if (!(EXT3_I(inode)->i_state & EXT3_STATE_XATTR)) + if (!(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)) return -ENODATA; - error = ext3_get_inode_loc(inode, &iloc); + error = ext4_get_inode_loc(inode, &iloc); if (error) return error; - raw_inode = ext3_raw_inode(&iloc); + raw_inode = ext4_raw_inode(&iloc); header = IHDR(inode, raw_inode); entry = IFIRST(header); - end = (void *)raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; - error = ext3_xattr_check_names(entry, end); + end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size; + error = ext4_xattr_check_names(entry, end); if (error) goto cleanup; - error = ext3_xattr_find_entry(&entry, name_index, name, + error = ext4_xattr_find_entry(&entry, name_index, name, end - (void *)entry, 0); if (error) goto cleanup; @@ -304,7 +304,7 @@ cleanup: } /* - * ext3_xattr_get() + * ext4_xattr_get() * * Copy an extended attribute into the buffer * provided, or compute the buffer size required. @@ -314,30 +314,30 @@ cleanup: * used / required on success. */ int -ext3_xattr_get(struct inode *inode, int name_index, const char *name, +ext4_xattr_get(struct inode *inode, int name_index, const char *name, void *buffer, size_t buffer_size) { int error; - down_read(&EXT3_I(inode)->xattr_sem); - error = ext3_xattr_ibody_get(inode, name_index, name, buffer, + down_read(&EXT4_I(inode)->xattr_sem); + error = ext4_xattr_ibody_get(inode, name_index, name, buffer, buffer_size); if (error == -ENODATA) - error = ext3_xattr_block_get(inode, name_index, name, buffer, + error = ext4_xattr_block_get(inode, name_index, name, buffer, buffer_size); - up_read(&EXT3_I(inode)->xattr_sem); + up_read(&EXT4_I(inode)->xattr_sem); return error; } static int -ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry, +ext4_xattr_list_entries(struct inode *inode, struct ext4_xattr_entry *entry, char *buffer, size_t buffer_size) { size_t rest = buffer_size; - for (; !IS_LAST_ENTRY(entry); entry = EXT3_XATTR_NEXT(entry)) { + for (; !IS_LAST_ENTRY(entry); entry = EXT4_XATTR_NEXT(entry)) { struct xattr_handler *handler = - ext3_xattr_handler(entry->e_name_index); + ext4_xattr_handler(entry->e_name_index); if (handler) { size_t size = handler->list(inode, buffer, rest, @@ -355,7 +355,7 @@ ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry, } static int -ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) +ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) { struct buffer_head *bh = NULL; int error; @@ -364,24 +364,24 @@ ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) buffer, (long)buffer_size); error = 0; - if (!EXT3_I(inode)->i_file_acl) + if (!EXT4_I(inode)->i_file_acl) goto cleanup; - ea_idebug(inode, "reading block %u", EXT3_I(inode)->i_file_acl); - bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + ea_idebug(inode, "reading block %u", EXT4_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); error = -EIO; if (!bh) goto cleanup; ea_bdebug(bh, "b_count=%d, refcount=%d", atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); - if (ext3_xattr_check_block(bh)) { - ext3_error(inode->i_sb, __FUNCTION__, + if (ext4_xattr_check_block(bh)) { + ext4_error(inode->i_sb, __FUNCTION__, "inode %lu: bad block "E3FSBLK, inode->i_ino, - EXT3_I(inode)->i_file_acl); + EXT4_I(inode)->i_file_acl); error = -EIO; goto cleanup; } - ext3_xattr_cache_insert(bh); - error = ext3_xattr_list_entries(inode, BFIRST(bh), buffer, buffer_size); + ext4_xattr_cache_insert(bh); + error = ext4_xattr_list_entries(inode, BFIRST(bh), buffer, buffer_size); cleanup: brelse(bh); @@ -390,26 +390,26 @@ cleanup: } static int -ext3_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size) +ext4_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size) { - struct ext3_xattr_ibody_header *header; - struct ext3_inode *raw_inode; - struct ext3_iloc iloc; + struct ext4_xattr_ibody_header *header; + struct ext4_inode *raw_inode; + struct ext4_iloc iloc; void *end; int error; - if (!(EXT3_I(inode)->i_state & EXT3_STATE_XATTR)) + if (!(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)) return 0; - error = ext3_get_inode_loc(inode, &iloc); + error = ext4_get_inode_loc(inode, &iloc); if (error) return error; - raw_inode = ext3_raw_inode(&iloc); + raw_inode = ext4_raw_inode(&iloc); header = IHDR(inode, raw_inode); - end = (void *)raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; - error = ext3_xattr_check_names(IFIRST(header), end); + end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size; + error = ext4_xattr_check_names(IFIRST(header), end); if (error) goto cleanup; - error = ext3_xattr_list_entries(inode, IFIRST(header), + error = ext4_xattr_list_entries(inode, IFIRST(header), buffer, buffer_size); cleanup: @@ -418,7 +418,7 @@ cleanup: } /* - * ext3_xattr_list() + * ext4_xattr_list() * * Copy a list of attribute names into the buffer * provided, or compute the buffer size required. @@ -428,12 +428,12 @@ cleanup: * used / required on success. */ int -ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) +ext4_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) { int i_error, b_error; - down_read(&EXT3_I(inode)->xattr_sem); - i_error = ext3_xattr_ibody_list(inode, buffer, buffer_size); + down_read(&EXT4_I(inode)->xattr_sem); + i_error = ext4_xattr_ibody_list(inode, buffer, buffer_size); if (i_error < 0) { b_error = 0; } else { @@ -441,30 +441,30 @@ ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) buffer += i_error; buffer_size -= i_error; } - b_error = ext3_xattr_block_list(inode, buffer, buffer_size); + b_error = ext4_xattr_block_list(inode, buffer, buffer_size); if (b_error < 0) i_error = 0; } - up_read(&EXT3_I(inode)->xattr_sem); + up_read(&EXT4_I(inode)->xattr_sem); return i_error + b_error; } /* - * If the EXT3_FEATURE_COMPAT_EXT_ATTR feature of this file system is + * If the EXT4_FEATURE_COMPAT_EXT_ATTR feature of this file system is * not set, set it. */ -static void ext3_xattr_update_super_block(handle_t *handle, +static void ext4_xattr_update_super_block(handle_t *handle, struct super_block *sb) { - if (EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_EXT_ATTR)) + if (EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_EXT_ATTR)) return; lock_super(sb); - if (ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh) == 0) { - EXT3_SB(sb)->s_es->s_feature_compat |= - cpu_to_le32(EXT3_FEATURE_COMPAT_EXT_ATTR); + if (ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh) == 0) { + EXT4_SB(sb)->s_es->s_feature_compat |= + cpu_to_le32(EXT4_FEATURE_COMPAT_EXT_ATTR); sb->s_dirt = 1; - ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); + ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); } unlock_super(sb); } @@ -474,25 +474,25 @@ static void ext3_xattr_update_super_block(handle_t *handle, * it; otherwise free the block. */ static void -ext3_xattr_release_block(handle_t *handle, struct inode *inode, +ext4_xattr_release_block(handle_t *handle, struct inode *inode, struct buffer_head *bh) { struct mb_cache_entry *ce = NULL; - ce = mb_cache_entry_get(ext3_xattr_cache, bh->b_bdev, bh->b_blocknr); + ce = mb_cache_entry_get(ext4_xattr_cache, bh->b_bdev, bh->b_blocknr); if (BHDR(bh)->h_refcount == cpu_to_le32(1)) { ea_bdebug(bh, "refcount now=0; freeing"); if (ce) mb_cache_entry_free(ce); - ext3_free_blocks(handle, inode, bh->b_blocknr, 1); + ext4_free_blocks(handle, inode, bh->b_blocknr, 1); get_bh(bh); - ext3_forget(handle, 1, inode, bh, bh->b_blocknr); + ext4_forget(handle, 1, inode, bh, bh->b_blocknr); } else { - if (ext3_journal_get_write_access(handle, bh) == 0) { + if (ext4_journal_get_write_access(handle, bh) == 0) { lock_buffer(bh); BHDR(bh)->h_refcount = cpu_to_le32( le32_to_cpu(BHDR(bh)->h_refcount) - 1); - ext3_journal_dirty_metadata(handle, bh); + ext4_journal_dirty_metadata(handle, bh); if (IS_SYNC(inode)) handle->h_sync = 1; DQUOT_FREE_BLOCK(inode, 1); @@ -505,30 +505,30 @@ ext3_xattr_release_block(handle_t *handle, struct inode *inode, } } -struct ext3_xattr_info { +struct ext4_xattr_info { int name_index; const char *name; const void *value; size_t value_len; }; -struct ext3_xattr_search { - struct ext3_xattr_entry *first; +struct ext4_xattr_search { + struct ext4_xattr_entry *first; void *base; void *end; - struct ext3_xattr_entry *here; + struct ext4_xattr_entry *here; int not_found; }; static int -ext3_xattr_set_entry(struct ext3_xattr_info *i, struct ext3_xattr_search *s) +ext4_xattr_set_entry(struct ext4_xattr_info *i, struct ext4_xattr_search *s) { - struct ext3_xattr_entry *last; + struct ext4_xattr_entry *last; size_t free, min_offs = s->end - s->base, name_len = strlen(i->name); /* Compute min_offs and last. */ last = s->first; - for (; !IS_LAST_ENTRY(last); last = EXT3_XATTR_NEXT(last)) { + for (; !IS_LAST_ENTRY(last); last = EXT4_XATTR_NEXT(last)) { if (!last->e_value_block && last->e_value_size) { size_t offs = le16_to_cpu(last->e_value_offs); if (offs < min_offs) @@ -539,20 +539,20 @@ ext3_xattr_set_entry(struct ext3_xattr_info *i, struct ext3_xattr_search *s) if (!s->not_found) { if (!s->here->e_value_block && s->here->e_value_size) { size_t size = le32_to_cpu(s->here->e_value_size); - free += EXT3_XATTR_SIZE(size); + free += EXT4_XATTR_SIZE(size); } - free += EXT3_XATTR_LEN(name_len); + free += EXT4_XATTR_LEN(name_len); } if (i->value) { - if (free < EXT3_XATTR_SIZE(i->value_len) || - free < EXT3_XATTR_LEN(name_len) + - EXT3_XATTR_SIZE(i->value_len)) + if (free < EXT4_XATTR_SIZE(i->value_len) || + free < EXT4_XATTR_LEN(name_len) + + EXT4_XATTR_SIZE(i->value_len)) return -ENOSPC; } if (i->value && s->not_found) { /* Insert the new name. */ - size_t size = EXT3_XATTR_LEN(name_len); + size_t size = EXT4_XATTR_LEN(name_len); size_t rest = (void *)last - (void *)s->here + sizeof(__u32); memmove((void *)s->here + size, s->here, rest); memset(s->here, 0, size); @@ -564,16 +564,16 @@ ext3_xattr_set_entry(struct ext3_xattr_info *i, struct ext3_xattr_search *s) void *first_val = s->base + min_offs; size_t offs = le16_to_cpu(s->here->e_value_offs); void *val = s->base + offs; - size_t size = EXT3_XATTR_SIZE( + size_t size = EXT4_XATTR_SIZE( le32_to_cpu(s->here->e_value_size)); - if (i->value && size == EXT3_XATTR_SIZE(i->value_len)) { + if (i->value && size == EXT4_XATTR_SIZE(i->value_len)) { /* The old and the new value have the same size. Just replace. */ s->here->e_value_size = cpu_to_le32(i->value_len); - memset(val + size - EXT3_XATTR_PAD, 0, - EXT3_XATTR_PAD); /* Clear pad bytes. */ + memset(val + size - EXT4_XATTR_PAD, 0, + EXT4_XATTR_PAD); /* Clear pad bytes. */ memcpy(val, i->value, i->value_len); return 0; } @@ -593,12 +593,12 @@ ext3_xattr_set_entry(struct ext3_xattr_info *i, struct ext3_xattr_search *s) last->e_value_size && o < offs) last->e_value_offs = cpu_to_le16(o + size); - last = EXT3_XATTR_NEXT(last); + last = EXT4_XATTR_NEXT(last); } } if (!i->value) { /* Remove the old name. */ - size_t size = EXT3_XATTR_LEN(name_len); + size_t size = EXT4_XATTR_LEN(name_len); last = ENTRY((void *)last - size); memmove(s->here, (void *)s->here + size, (void *)last - (void *)s->here + sizeof(__u32)); @@ -610,25 +610,25 @@ ext3_xattr_set_entry(struct ext3_xattr_info *i, struct ext3_xattr_search *s) /* Insert the new value. */ s->here->e_value_size = cpu_to_le32(i->value_len); if (i->value_len) { - size_t size = EXT3_XATTR_SIZE(i->value_len); + size_t size = EXT4_XATTR_SIZE(i->value_len); void *val = s->base + min_offs - size; s->here->e_value_offs = cpu_to_le16(min_offs - size); - memset(val + size - EXT3_XATTR_PAD, 0, - EXT3_XATTR_PAD); /* Clear the pad bytes. */ + memset(val + size - EXT4_XATTR_PAD, 0, + EXT4_XATTR_PAD); /* Clear the pad bytes. */ memcpy(val, i->value, i->value_len); } } return 0; } -struct ext3_xattr_block_find { - struct ext3_xattr_search s; +struct ext4_xattr_block_find { + struct ext4_xattr_search s; struct buffer_head *bh; }; static int -ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i, - struct ext3_xattr_block_find *bs) +ext4_xattr_block_find(struct inode *inode, struct ext4_xattr_info *i, + struct ext4_xattr_block_find *bs) { struct super_block *sb = inode->i_sb; int error; @@ -636,19 +636,19 @@ ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i, ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld", i->name_index, i->name, i->value, (long)i->value_len); - if (EXT3_I(inode)->i_file_acl) { + if (EXT4_I(inode)->i_file_acl) { /* The inode already has an extended attribute block. */ - bs->bh = sb_bread(sb, EXT3_I(inode)->i_file_acl); + bs->bh = sb_bread(sb, EXT4_I(inode)->i_file_acl); error = -EIO; if (!bs->bh) goto cleanup; ea_bdebug(bs->bh, "b_count=%d, refcount=%d", atomic_read(&(bs->bh->b_count)), le32_to_cpu(BHDR(bs->bh)->h_refcount)); - if (ext3_xattr_check_block(bs->bh)) { - ext3_error(sb, __FUNCTION__, + if (ext4_xattr_check_block(bs->bh)) { + ext4_error(sb, __FUNCTION__, "inode %lu: bad block "E3FSBLK, inode->i_ino, - EXT3_I(inode)->i_file_acl); + EXT4_I(inode)->i_file_acl); error = -EIO; goto cleanup; } @@ -657,7 +657,7 @@ ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i, bs->s.first = BFIRST(bs->bh); bs->s.end = bs->bh->b_data + bs->bh->b_size; bs->s.here = bs->s.first; - error = ext3_xattr_find_entry(&bs->s.here, i->name_index, + error = ext4_xattr_find_entry(&bs->s.here, i->name_index, i->name, bs->bh->b_size, 1); if (error && error != -ENODATA) goto cleanup; @@ -670,22 +670,22 @@ cleanup: } static int -ext3_xattr_block_set(handle_t *handle, struct inode *inode, - struct ext3_xattr_info *i, - struct ext3_xattr_block_find *bs) +ext4_xattr_block_set(handle_t *handle, struct inode *inode, + struct ext4_xattr_info *i, + struct ext4_xattr_block_find *bs) { struct super_block *sb = inode->i_sb; struct buffer_head *new_bh = NULL; - struct ext3_xattr_search *s = &bs->s; + struct ext4_xattr_search *s = &bs->s; struct mb_cache_entry *ce = NULL; int error; -#define header(x) ((struct ext3_xattr_header *)(x)) +#define header(x) ((struct ext4_xattr_header *)(x)) if (i->value && i->value_len > sb->s_blocksize) return -ENOSPC; if (s->base) { - ce = mb_cache_entry_get(ext3_xattr_cache, bs->bh->b_bdev, + ce = mb_cache_entry_get(ext4_xattr_cache, bs->bh->b_bdev, bs->bh->b_blocknr); if (header(s->base)->h_refcount == cpu_to_le32(1)) { if (ce) { @@ -693,22 +693,22 @@ ext3_xattr_block_set(handle_t *handle, struct inode *inode, ce = NULL; } ea_bdebug(bs->bh, "modifying in-place"); - error = ext3_journal_get_write_access(handle, bs->bh); + error = ext4_journal_get_write_access(handle, bs->bh); if (error) goto cleanup; lock_buffer(bs->bh); - error = ext3_xattr_set_entry(i, s); + error = ext4_xattr_set_entry(i, s); if (!error) { if (!IS_LAST_ENTRY(s->first)) - ext3_xattr_rehash(header(s->base), + ext4_xattr_rehash(header(s->base), s->here); - ext3_xattr_cache_insert(bs->bh); + ext4_xattr_cache_insert(bs->bh); } unlock_buffer(bs->bh); if (error == -EIO) goto bad_block; if (!error) - error = ext3_journal_dirty_metadata(handle, + error = ext4_journal_dirty_metadata(handle, bs->bh); if (error) goto cleanup; @@ -739,7 +739,7 @@ ext3_xattr_block_set(handle_t *handle, struct inode *inode, if (s->base == NULL) goto cleanup; memset(s->base, 0, sb->s_blocksize); - header(s->base)->h_magic = cpu_to_le32(EXT3_XATTR_MAGIC); + header(s->base)->h_magic = cpu_to_le32(EXT4_XATTR_MAGIC); header(s->base)->h_blocks = cpu_to_le32(1); header(s->base)->h_refcount = cpu_to_le32(1); s->first = ENTRY(header(s->base)+1); @@ -747,17 +747,17 @@ ext3_xattr_block_set(handle_t *handle, struct inode *inode, s->end = s->base + sb->s_blocksize; } - error = ext3_xattr_set_entry(i, s); + error = ext4_xattr_set_entry(i, s); if (error == -EIO) goto bad_block; if (error) goto cleanup; if (!IS_LAST_ENTRY(s->first)) - ext3_xattr_rehash(header(s->base), s->here); + ext4_xattr_rehash(header(s->base), s->here); inserted: if (!IS_LAST_ENTRY(s->first)) { - new_bh = ext3_xattr_cache_find(inode, header(s->base), &ce); + new_bh = ext4_xattr_cache_find(inode, header(s->base), &ce); if (new_bh) { /* We found an identical block in the cache. */ if (new_bh == bs->bh) @@ -768,7 +768,7 @@ inserted: error = -EDQUOT; if (DQUOT_ALLOC_BLOCK(inode, 1)) goto cleanup; - error = ext3_journal_get_write_access(handle, + error = ext4_journal_get_write_access(handle, new_bh); if (error) goto cleanup_dquot; @@ -778,7 +778,7 @@ inserted: ea_bdebug(new_bh, "reusing; refcount now=%d", le32_to_cpu(BHDR(new_bh)->h_refcount)); unlock_buffer(new_bh); - error = ext3_journal_dirty_metadata(handle, + error = ext4_journal_dirty_metadata(handle, new_bh); if (error) goto cleanup_dquot; @@ -792,11 +792,11 @@ inserted: get_bh(new_bh); } else { /* We need to allocate a new block */ - ext3_fsblk_t goal = le32_to_cpu( - EXT3_SB(sb)->s_es->s_first_data_block) + - (ext3_fsblk_t)EXT3_I(inode)->i_block_group * - EXT3_BLOCKS_PER_GROUP(sb); - ext3_fsblk_t block = ext3_new_block(handle, inode, + ext4_fsblk_t goal = le32_to_cpu( + EXT4_SB(sb)->s_es->s_first_data_block) + + (ext4_fsblk_t)EXT4_I(inode)->i_block_group * + EXT4_BLOCKS_PER_GROUP(sb); + ext4_fsblk_t block = ext4_new_block(handle, inode, goal, &error); if (error) goto cleanup; @@ -805,12 +805,12 @@ inserted: new_bh = sb_getblk(sb, block); if (!new_bh) { getblk_failed: - ext3_free_blocks(handle, inode, block, 1); + ext4_free_blocks(handle, inode, block, 1); error = -EIO; goto cleanup; } lock_buffer(new_bh); - error = ext3_journal_get_create_access(handle, new_bh); + error = ext4_journal_get_create_access(handle, new_bh); if (error) { unlock_buffer(new_bh); goto getblk_failed; @@ -818,19 +818,19 @@ getblk_failed: memcpy(new_bh->b_data, s->base, new_bh->b_size); set_buffer_uptodate(new_bh); unlock_buffer(new_bh); - ext3_xattr_cache_insert(new_bh); - error = ext3_journal_dirty_metadata(handle, new_bh); + ext4_xattr_cache_insert(new_bh); + error = ext4_journal_dirty_metadata(handle, new_bh); if (error) goto cleanup; } } /* Update the inode. */ - EXT3_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0; + EXT4_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0; /* Drop the previous xattr block. */ if (bs->bh && bs->bh != new_bh) - ext3_xattr_release_block(handle, inode, bs->bh); + ext4_xattr_release_block(handle, inode, bs->bh); error = 0; cleanup: @@ -847,40 +847,40 @@ cleanup_dquot: goto cleanup; bad_block: - ext3_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __FUNCTION__, "inode %lu: bad block "E3FSBLK, inode->i_ino, - EXT3_I(inode)->i_file_acl); + EXT4_I(inode)->i_file_acl); goto cleanup; #undef header } -struct ext3_xattr_ibody_find { - struct ext3_xattr_search s; - struct ext3_iloc iloc; +struct ext4_xattr_ibody_find { + struct ext4_xattr_search s; + struct ext4_iloc iloc; }; static int -ext3_xattr_ibody_find(struct inode *inode, struct ext3_xattr_info *i, - struct ext3_xattr_ibody_find *is) +ext4_xattr_ibody_find(struct inode *inode, struct ext4_xattr_info *i, + struct ext4_xattr_ibody_find *is) { - struct ext3_xattr_ibody_header *header; - struct ext3_inode *raw_inode; + struct ext4_xattr_ibody_header *header; + struct ext4_inode *raw_inode; int error; - if (EXT3_I(inode)->i_extra_isize == 0) + if (EXT4_I(inode)->i_extra_isize == 0) return 0; - raw_inode = ext3_raw_inode(&is->iloc); + raw_inode = ext4_raw_inode(&is->iloc); header = IHDR(inode, raw_inode); is->s.base = is->s.first = IFIRST(header); is->s.here = is->s.first; - is->s.end = (void *)raw_inode + EXT3_SB(inode->i_sb)->s_inode_size; - if (EXT3_I(inode)->i_state & EXT3_STATE_XATTR) { - error = ext3_xattr_check_names(IFIRST(header), is->s.end); + is->s.end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size; + if (EXT4_I(inode)->i_state & EXT4_STATE_XATTR) { + error = ext4_xattr_check_names(IFIRST(header), is->s.end); if (error) return error; /* Find the named attribute. */ - error = ext3_xattr_find_entry(&is->s.here, i->name_index, + error = ext4_xattr_find_entry(&is->s.here, i->name_index, i->name, is->s.end - (void *)is->s.base, 0); if (error && error != -ENODATA) @@ -891,32 +891,32 @@ ext3_xattr_ibody_find(struct inode *inode, struct ext3_xattr_info *i, } static int -ext3_xattr_ibody_set(handle_t *handle, struct inode *inode, - struct ext3_xattr_info *i, - struct ext3_xattr_ibody_find *is) +ext4_xattr_ibody_set(handle_t *handle, struct inode *inode, + struct ext4_xattr_info *i, + struct ext4_xattr_ibody_find *is) { - struct ext3_xattr_ibody_header *header; - struct ext3_xattr_search *s = &is->s; + struct ext4_xattr_ibody_header *header; + struct ext4_xattr_search *s = &is->s; int error; - if (EXT3_I(inode)->i_extra_isize == 0) + if (EXT4_I(inode)->i_extra_isize == 0) return -ENOSPC; - error = ext3_xattr_set_entry(i, s); + error = ext4_xattr_set_entry(i, s); if (error) return error; - header = IHDR(inode, ext3_raw_inode(&is->iloc)); + header = IHDR(inode, ext4_raw_inode(&is->iloc)); if (!IS_LAST_ENTRY(s->first)) { - header->h_magic = cpu_to_le32(EXT3_XATTR_MAGIC); - EXT3_I(inode)->i_state |= EXT3_STATE_XATTR; + header->h_magic = cpu_to_le32(EXT4_XATTR_MAGIC); + EXT4_I(inode)->i_state |= EXT4_STATE_XATTR; } else { header->h_magic = cpu_to_le32(0); - EXT3_I(inode)->i_state &= ~EXT3_STATE_XATTR; + EXT4_I(inode)->i_state &= ~EXT4_STATE_XATTR; } return 0; } /* - * ext3_xattr_set_handle() + * ext4_xattr_set_handle() * * Create, replace or remove an extended attribute for this inode. Buffer * is NULL to remove an existing extended attribute, and non-NULL to @@ -928,21 +928,21 @@ ext3_xattr_ibody_set(handle_t *handle, struct inode *inode, * Returns 0, or a negative error number on failure. */ int -ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, +ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, const char *name, const void *value, size_t value_len, int flags) { - struct ext3_xattr_info i = { + struct ext4_xattr_info i = { .name_index = name_index, .name = name, .value = value, .value_len = value_len, }; - struct ext3_xattr_ibody_find is = { + struct ext4_xattr_ibody_find is = { .s = { .not_found = -ENODATA, }, }; - struct ext3_xattr_block_find bs = { + struct ext4_xattr_block_find bs = { .s = { .not_found = -ENODATA, }, }; int error; @@ -951,22 +951,22 @@ ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, return -EINVAL; if (strlen(name) > 255) return -ERANGE; - down_write(&EXT3_I(inode)->xattr_sem); - error = ext3_get_inode_loc(inode, &is.iloc); + down_write(&EXT4_I(inode)->xattr_sem); + error = ext4_get_inode_loc(inode, &is.iloc); if (error) goto cleanup; - if (EXT3_I(inode)->i_state & EXT3_STATE_NEW) { - struct ext3_inode *raw_inode = ext3_raw_inode(&is.iloc); - memset(raw_inode, 0, EXT3_SB(inode->i_sb)->s_inode_size); - EXT3_I(inode)->i_state &= ~EXT3_STATE_NEW; + if (EXT4_I(inode)->i_state & EXT4_STATE_NEW) { + struct ext4_inode *raw_inode = ext4_raw_inode(&is.iloc); + memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size); + EXT4_I(inode)->i_state &= ~EXT4_STATE_NEW; } - error = ext3_xattr_ibody_find(inode, &i, &is); + error = ext4_xattr_ibody_find(inode, &i, &is); if (error) goto cleanup; if (is.s.not_found) - error = ext3_xattr_block_find(inode, &i, &bs); + error = ext4_xattr_block_find(inode, &i, &bs); if (error) goto cleanup; if (is.s.not_found && bs.s.not_found) { @@ -981,36 +981,36 @@ ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, if (flags & XATTR_CREATE) goto cleanup; } - error = ext3_journal_get_write_access(handle, is.iloc.bh); + error = ext4_journal_get_write_access(handle, is.iloc.bh); if (error) goto cleanup; if (!value) { if (!is.s.not_found) - error = ext3_xattr_ibody_set(handle, inode, &i, &is); + error = ext4_xattr_ibody_set(handle, inode, &i, &is); else if (!bs.s.not_found) - error = ext3_xattr_block_set(handle, inode, &i, &bs); + error = ext4_xattr_block_set(handle, inode, &i, &bs); } else { - error = ext3_xattr_ibody_set(handle, inode, &i, &is); + error = ext4_xattr_ibody_set(handle, inode, &i, &is); if (!error && !bs.s.not_found) { i.value = NULL; - error = ext3_xattr_block_set(handle, inode, &i, &bs); + error = ext4_xattr_block_set(handle, inode, &i, &bs); } else if (error == -ENOSPC) { - error = ext3_xattr_block_set(handle, inode, &i, &bs); + error = ext4_xattr_block_set(handle, inode, &i, &bs); if (error) goto cleanup; if (!is.s.not_found) { i.value = NULL; - error = ext3_xattr_ibody_set(handle, inode, &i, + error = ext4_xattr_ibody_set(handle, inode, &i, &is); } } } if (!error) { - ext3_xattr_update_super_block(handle, inode->i_sb); + ext4_xattr_update_super_block(handle, inode->i_sb); inode->i_ctime = CURRENT_TIME_SEC; - error = ext3_mark_iloc_dirty(handle, inode, &is.iloc); + error = ext4_mark_iloc_dirty(handle, inode, &is.iloc); /* - * The bh is consumed by ext3_mark_iloc_dirty, even with + * The bh is consumed by ext4_mark_iloc_dirty, even with * error != 0. */ is.iloc.bh = NULL; @@ -1021,37 +1021,37 @@ ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, cleanup: brelse(is.iloc.bh); brelse(bs.bh); - up_write(&EXT3_I(inode)->xattr_sem); + up_write(&EXT4_I(inode)->xattr_sem); return error; } /* - * ext3_xattr_set() + * ext4_xattr_set() * - * Like ext3_xattr_set_handle, but start from an inode. This extended + * Like ext4_xattr_set_handle, but start from an inode. This extended * attribute modification is a filesystem transaction by itself. * * Returns 0, or a negative error number on failure. */ int -ext3_xattr_set(struct inode *inode, int name_index, const char *name, +ext4_xattr_set(struct inode *inode, int name_index, const char *name, const void *value, size_t value_len, int flags) { handle_t *handle; int error, retries = 0; retry: - handle = ext3_journal_start(inode, EXT3_DATA_TRANS_BLOCKS(inode->i_sb)); + handle = ext4_journal_start(inode, EXT4_DATA_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) { error = PTR_ERR(handle); } else { int error2; - error = ext3_xattr_set_handle(handle, inode, name_index, name, + error = ext4_xattr_set_handle(handle, inode, name_index, name, value, value_len, flags); - error2 = ext3_journal_stop(handle); + error2 = ext4_journal_stop(handle); if (error == -ENOSPC && - ext3_should_retry_alloc(inode->i_sb, &retries)) + ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; if (error == 0) error = error2; @@ -1061,53 +1061,53 @@ retry: } /* - * ext3_xattr_delete_inode() + * ext4_xattr_delete_inode() * * Free extended attribute resources associated with this inode. This * is called immediately before an inode is freed. We have exclusive * access to the inode. */ void -ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) +ext4_xattr_delete_inode(handle_t *handle, struct inode *inode) { struct buffer_head *bh = NULL; - if (!EXT3_I(inode)->i_file_acl) + if (!EXT4_I(inode)->i_file_acl) goto cleanup; - bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); if (!bh) { - ext3_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __FUNCTION__, "inode %lu: block "E3FSBLK" read error", inode->i_ino, - EXT3_I(inode)->i_file_acl); + EXT4_I(inode)->i_file_acl); goto cleanup; } - if (BHDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || + if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || BHDR(bh)->h_blocks != cpu_to_le32(1)) { - ext3_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __FUNCTION__, "inode %lu: bad block "E3FSBLK, inode->i_ino, - EXT3_I(inode)->i_file_acl); + EXT4_I(inode)->i_file_acl); goto cleanup; } - ext3_xattr_release_block(handle, inode, bh); - EXT3_I(inode)->i_file_acl = 0; + ext4_xattr_release_block(handle, inode, bh); + EXT4_I(inode)->i_file_acl = 0; cleanup: brelse(bh); } /* - * ext3_xattr_put_super() + * ext4_xattr_put_super() * * This is called when a file system is unmounted. */ void -ext3_xattr_put_super(struct super_block *sb) +ext4_xattr_put_super(struct super_block *sb) { mb_cache_shrink(sb->s_bdev); } /* - * ext3_xattr_cache_insert() + * ext4_xattr_cache_insert() * * Create a new entry in the extended attribute cache, and insert * it unless such an entry is already in the cache. @@ -1115,13 +1115,13 @@ ext3_xattr_put_super(struct super_block *sb) * Returns 0, or a negative error number on failure. */ static void -ext3_xattr_cache_insert(struct buffer_head *bh) +ext4_xattr_cache_insert(struct buffer_head *bh) { __u32 hash = le32_to_cpu(BHDR(bh)->h_hash); struct mb_cache_entry *ce; int error; - ce = mb_cache_entry_alloc(ext3_xattr_cache); + ce = mb_cache_entry_alloc(ext4_xattr_cache); if (!ce) { ea_bdebug(bh, "out of memory"); return; @@ -1140,7 +1140,7 @@ ext3_xattr_cache_insert(struct buffer_head *bh) } /* - * ext3_xattr_cmp() + * ext4_xattr_cmp() * * Compare two extended attribute blocks for equality. * @@ -1148,10 +1148,10 @@ ext3_xattr_cache_insert(struct buffer_head *bh) * a negative error number on errors. */ static int -ext3_xattr_cmp(struct ext3_xattr_header *header1, - struct ext3_xattr_header *header2) +ext4_xattr_cmp(struct ext4_xattr_header *header1, + struct ext4_xattr_header *header2) { - struct ext3_xattr_entry *entry1, *entry2; + struct ext4_xattr_entry *entry1, *entry2; entry1 = ENTRY(header1+1); entry2 = ENTRY(header2+1); @@ -1171,8 +1171,8 @@ ext3_xattr_cmp(struct ext3_xattr_header *header1, le32_to_cpu(entry1->e_value_size))) return 1; - entry1 = EXT3_XATTR_NEXT(entry1); - entry2 = EXT3_XATTR_NEXT(entry2); + entry1 = EXT4_XATTR_NEXT(entry1); + entry2 = EXT4_XATTR_NEXT(entry2); } if (!IS_LAST_ENTRY(entry2)) return 1; @@ -1180,7 +1180,7 @@ ext3_xattr_cmp(struct ext3_xattr_header *header1, } /* - * ext3_xattr_cache_find() + * ext4_xattr_cache_find() * * Find an identical extended attribute block. * @@ -1188,7 +1188,7 @@ ext3_xattr_cmp(struct ext3_xattr_header *header1, * not found or an error occurred. */ static struct buffer_head * -ext3_xattr_cache_find(struct inode *inode, struct ext3_xattr_header *header, +ext4_xattr_cache_find(struct inode *inode, struct ext4_xattr_header *header, struct mb_cache_entry **pce) { __u32 hash = le32_to_cpu(header->h_hash); @@ -1198,7 +1198,7 @@ ext3_xattr_cache_find(struct inode *inode, struct ext3_xattr_header *header, return NULL; /* never share */ ea_idebug(inode, "looking for cached blocks [%x]", (int)hash); again: - ce = mb_cache_entry_find_first(ext3_xattr_cache, 0, + ce = mb_cache_entry_find_first(ext4_xattr_cache, 0, inode->i_sb->s_bdev, hash); while (ce) { struct buffer_head *bh; @@ -1210,16 +1210,16 @@ again: } bh = sb_bread(inode->i_sb, ce->e_block); if (!bh) { - ext3_error(inode->i_sb, __FUNCTION__, + ext4_error(inode->i_sb, __FUNCTION__, "inode %lu: block %lu read error", inode->i_ino, (unsigned long) ce->e_block); } else if (le32_to_cpu(BHDR(bh)->h_refcount) >= - EXT3_XATTR_REFCOUNT_MAX) { + EXT4_XATTR_REFCOUNT_MAX) { ea_idebug(inode, "block %lu refcount %d>=%d", (unsigned long) ce->e_block, le32_to_cpu(BHDR(bh)->h_refcount), - EXT3_XATTR_REFCOUNT_MAX); - } else if (ext3_xattr_cmp(header, BHDR(bh)) == 0) { + EXT4_XATTR_REFCOUNT_MAX); + } else if (ext4_xattr_cmp(header, BHDR(bh)) == 0) { *pce = ce; return bh; } @@ -1233,12 +1233,12 @@ again: #define VALUE_HASH_SHIFT 16 /* - * ext3_xattr_hash_entry() + * ext4_xattr_hash_entry() * * Compute the hash of an extended attribute. */ -static inline void ext3_xattr_hash_entry(struct ext3_xattr_header *header, - struct ext3_xattr_entry *entry) +static inline void ext4_xattr_hash_entry(struct ext4_xattr_header *header, + struct ext4_xattr_entry *entry) { __u32 hash = 0; char *name = entry->e_name; @@ -1254,7 +1254,7 @@ static inline void ext3_xattr_hash_entry(struct ext3_xattr_header *header, __le32 *value = (__le32 *)((char *)header + le16_to_cpu(entry->e_value_offs)); for (n = (le32_to_cpu(entry->e_value_size) + - EXT3_XATTR_ROUND) >> EXT3_XATTR_PAD_BITS; n; n--) { + EXT4_XATTR_ROUND) >> EXT4_XATTR_PAD_BITS; n; n--) { hash = (hash << VALUE_HASH_SHIFT) ^ (hash >> (8*sizeof(hash) - VALUE_HASH_SHIFT)) ^ le32_to_cpu(*value++); @@ -1269,17 +1269,17 @@ static inline void ext3_xattr_hash_entry(struct ext3_xattr_header *header, #define BLOCK_HASH_SHIFT 16 /* - * ext3_xattr_rehash() + * ext4_xattr_rehash() * * Re-compute the extended attribute hash value after an entry has changed. */ -static void ext3_xattr_rehash(struct ext3_xattr_header *header, - struct ext3_xattr_entry *entry) +static void ext4_xattr_rehash(struct ext4_xattr_header *header, + struct ext4_xattr_entry *entry) { - struct ext3_xattr_entry *here; + struct ext4_xattr_entry *here; __u32 hash = 0; - ext3_xattr_hash_entry(header, entry); + ext4_xattr_hash_entry(header, entry); here = ENTRY(header+1); while (!IS_LAST_ENTRY(here)) { if (!here->e_hash) { @@ -1290,7 +1290,7 @@ static void ext3_xattr_rehash(struct ext3_xattr_header *header, hash = (hash << BLOCK_HASH_SHIFT) ^ (hash >> (8*sizeof(hash) - BLOCK_HASH_SHIFT)) ^ le32_to_cpu(here->e_hash); - here = EXT3_XATTR_NEXT(here); + here = EXT4_XATTR_NEXT(here); } header->h_hash = cpu_to_le32(hash); } @@ -1298,20 +1298,20 @@ static void ext3_xattr_rehash(struct ext3_xattr_header *header, #undef BLOCK_HASH_SHIFT int __init -init_ext3_xattr(void) +init_ext4_xattr(void) { - ext3_xattr_cache = mb_cache_create("ext3_xattr", NULL, + ext4_xattr_cache = mb_cache_create("ext4_xattr", NULL, sizeof(struct mb_cache_entry) + sizeof(((struct mb_cache_entry *) 0)->e_indexes[0]), 1, 6); - if (!ext3_xattr_cache) + if (!ext4_xattr_cache) return -ENOMEM; return 0; } void -exit_ext3_xattr(void) +exit_ext4_xattr(void) { - if (ext3_xattr_cache) - mb_cache_destroy(ext3_xattr_cache); - ext3_xattr_cache = NULL; + if (ext4_xattr_cache) + mb_cache_destroy(ext4_xattr_cache); + ext4_xattr_cache = NULL; } diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h index 6b1ae1c..79432b3 100644 --- a/fs/ext4/xattr.h +++ b/fs/ext4/xattr.h @@ -1,7 +1,7 @@ /* - File: fs/ext3/xattr.h + File: fs/ext4/xattr.h - On-disk format of extended attributes for the ext3 filesystem. + On-disk format of extended attributes for the ext4 filesystem. (C) 2001 Andreas Gruenbacher, */ @@ -9,20 +9,20 @@ #include /* Magic value in attribute blocks */ -#define EXT3_XATTR_MAGIC 0xEA020000 +#define EXT4_XATTR_MAGIC 0xEA020000 /* Maximum number of references to one attribute block */ -#define EXT3_XATTR_REFCOUNT_MAX 1024 +#define EXT4_XATTR_REFCOUNT_MAX 1024 /* Name indexes */ -#define EXT3_XATTR_INDEX_USER 1 -#define EXT3_XATTR_INDEX_POSIX_ACL_ACCESS 2 -#define EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT 3 -#define EXT3_XATTR_INDEX_TRUSTED 4 -#define EXT3_XATTR_INDEX_LUSTRE 5 -#define EXT3_XATTR_INDEX_SECURITY 6 - -struct ext3_xattr_header { +#define EXT4_XATTR_INDEX_USER 1 +#define EXT4_XATTR_INDEX_POSIX_ACL_ACCESS 2 +#define EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT 3 +#define EXT4_XATTR_INDEX_TRUSTED 4 +#define EXT4_XATTR_INDEX_LUSTRE 5 +#define EXT4_XATTR_INDEX_SECURITY 6 + +struct ext4_xattr_header { __le32 h_magic; /* magic number for identification */ __le32 h_refcount; /* reference count */ __le32 h_blocks; /* number of disk blocks used */ @@ -30,11 +30,11 @@ struct ext3_xattr_header { __u32 h_reserved[4]; /* zero right now */ }; -struct ext3_xattr_ibody_header { +struct ext4_xattr_ibody_header { __le32 h_magic; /* magic number for identification */ }; -struct ext3_xattr_entry { +struct ext4_xattr_entry { __u8 e_name_len; /* length of name */ __u8 e_name_index; /* attribute name index */ __le16 e_value_offs; /* offset in disk block of value */ @@ -44,100 +44,100 @@ struct ext3_xattr_entry { char e_name[0]; /* attribute name */ }; -#define EXT3_XATTR_PAD_BITS 2 -#define EXT3_XATTR_PAD (1<e_name_len)) ) -#define EXT3_XATTR_SIZE(size) \ - (((size) + EXT3_XATTR_ROUND) & ~EXT3_XATTR_ROUND) +#define EXT4_XATTR_PAD_BITS 2 +#define EXT4_XATTR_PAD (1<e_name_len)) ) +#define EXT4_XATTR_SIZE(size) \ + (((size) + EXT4_XATTR_ROUND) & ~EXT4_XATTR_ROUND) -# ifdef CONFIG_EXT3_FS_XATTR +# ifdef CONFIG_EXT4DEV_FS_XATTR -extern struct xattr_handler ext3_xattr_user_handler; -extern struct xattr_handler ext3_xattr_trusted_handler; -extern struct xattr_handler ext3_xattr_acl_access_handler; -extern struct xattr_handler ext3_xattr_acl_default_handler; -extern struct xattr_handler ext3_xattr_security_handler; +extern struct xattr_handler ext4_xattr_user_handler; +extern struct xattr_handler ext4_xattr_trusted_handler; +extern struct xattr_handler ext4_xattr_acl_access_handler; +extern struct xattr_handler ext4_xattr_acl_default_handler; +extern struct xattr_handler ext4_xattr_security_handler; -extern ssize_t ext3_listxattr(struct dentry *, char *, size_t); +extern ssize_t ext4_listxattr(struct dentry *, char *, size_t); -extern int ext3_xattr_get(struct inode *, int, const char *, void *, size_t); -extern int ext3_xattr_list(struct inode *, char *, size_t); -extern int ext3_xattr_set(struct inode *, int, const char *, const void *, size_t, int); -extern int ext3_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); +extern int ext4_xattr_get(struct inode *, int, const char *, void *, size_t); +extern int ext4_xattr_list(struct inode *, char *, size_t); +extern int ext4_xattr_set(struct inode *, int, const char *, const void *, size_t, int); +extern int ext4_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); -extern void ext3_xattr_delete_inode(handle_t *, struct inode *); -extern void ext3_xattr_put_super(struct super_block *); +extern void ext4_xattr_delete_inode(handle_t *, struct inode *); +extern void ext4_xattr_put_super(struct super_block *); -extern int init_ext3_xattr(void); -extern void exit_ext3_xattr(void); +extern int init_ext4_xattr(void); +extern void exit_ext4_xattr(void); -extern struct xattr_handler *ext3_xattr_handlers[]; +extern struct xattr_handler *ext4_xattr_handlers[]; -# else /* CONFIG_EXT3_FS_XATTR */ +# else /* CONFIG_EXT4DEV_FS_XATTR */ static inline int -ext3_xattr_get(struct inode *inode, int name_index, const char *name, +ext4_xattr_get(struct inode *inode, int name_index, const char *name, void *buffer, size_t size, int flags) { return -EOPNOTSUPP; } static inline int -ext3_xattr_list(struct inode *inode, void *buffer, size_t size) +ext4_xattr_list(struct inode *inode, void *buffer, size_t size) { return -EOPNOTSUPP; } static inline int -ext3_xattr_set(struct inode *inode, int name_index, const char *name, +ext4_xattr_set(struct inode *inode, int name_index, const char *name, const void *value, size_t size, int flags) { return -EOPNOTSUPP; } static inline int -ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, +ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, const char *name, const void *value, size_t size, int flags) { return -EOPNOTSUPP; } static inline void -ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) +ext4_xattr_delete_inode(handle_t *handle, struct inode *inode) { } static inline void -ext3_xattr_put_super(struct super_block *sb) +ext4_xattr_put_super(struct super_block *sb) { } static inline int -init_ext3_xattr(void) +init_ext4_xattr(void) { return 0; } static inline void -exit_ext3_xattr(void) +exit_ext4_xattr(void) { } -#define ext3_xattr_handlers NULL +#define ext4_xattr_handlers NULL -# endif /* CONFIG_EXT3_FS_XATTR */ +# endif /* CONFIG_EXT4DEV_FS_XATTR */ -#ifdef CONFIG_EXT3_FS_SECURITY -extern int ext3_init_security(handle_t *handle, struct inode *inode, +#ifdef CONFIG_EXT4DEV_FS_SECURITY +extern int ext4_init_security(handle_t *handle, struct inode *inode, struct inode *dir); #else -static inline int ext3_init_security(handle_t *handle, struct inode *inode, +static inline int ext4_init_security(handle_t *handle, struct inode *inode, struct inode *dir) { return 0; diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c index b9c40c1..d84b1da 100644 --- a/fs/ext4/xattr_security.c +++ b/fs/ext4/xattr_security.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/xattr_security.c + * linux/fs/ext4/xattr_security.c * Handler for storing security labels as extended attributes. */ @@ -7,13 +7,13 @@ #include #include #include -#include -#include +#include +#include #include #include "xattr.h" static size_t -ext3_xattr_security_list(struct inode *inode, char *list, size_t list_size, +ext4_xattr_security_list(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { const size_t prefix_len = sizeof(XATTR_SECURITY_PREFIX)-1; @@ -29,27 +29,27 @@ ext3_xattr_security_list(struct inode *inode, char *list, size_t list_size, } static int -ext3_xattr_security_get(struct inode *inode, const char *name, +ext4_xattr_security_get(struct inode *inode, const char *name, void *buffer, size_t size) { if (strcmp(name, "") == 0) return -EINVAL; - return ext3_xattr_get(inode, EXT3_XATTR_INDEX_SECURITY, name, + return ext4_xattr_get(inode, EXT4_XATTR_INDEX_SECURITY, name, buffer, size); } static int -ext3_xattr_security_set(struct inode *inode, const char *name, +ext4_xattr_security_set(struct inode *inode, const char *name, const void *value, size_t size, int flags) { if (strcmp(name, "") == 0) return -EINVAL; - return ext3_xattr_set(inode, EXT3_XATTR_INDEX_SECURITY, name, + return ext4_xattr_set(inode, EXT4_XATTR_INDEX_SECURITY, name, value, size, flags); } int -ext3_init_security(handle_t *handle, struct inode *inode, struct inode *dir) +ext4_init_security(handle_t *handle, struct inode *inode, struct inode *dir) { int err; size_t len; @@ -62,16 +62,16 @@ ext3_init_security(handle_t *handle, struct inode *inode, struct inode *dir) return 0; return err; } - err = ext3_xattr_set_handle(handle, inode, EXT3_XATTR_INDEX_SECURITY, + err = ext4_xattr_set_handle(handle, inode, EXT4_XATTR_INDEX_SECURITY, name, value, len, 0); kfree(name); kfree(value); return err; } -struct xattr_handler ext3_xattr_security_handler = { +struct xattr_handler ext4_xattr_security_handler = { .prefix = XATTR_SECURITY_PREFIX, - .list = ext3_xattr_security_list, - .get = ext3_xattr_security_get, - .set = ext3_xattr_security_set, + .list = ext4_xattr_security_list, + .get = ext4_xattr_security_get, + .set = ext4_xattr_security_set, }; diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c index 86d91f1..11bd58c 100644 --- a/fs/ext4/xattr_trusted.c +++ b/fs/ext4/xattr_trusted.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/xattr_trusted.c + * linux/fs/ext4/xattr_trusted.c * Handler for trusted extended attributes. * * Copyright (C) 2003 by Andreas Gruenbacher, @@ -10,14 +10,14 @@ #include #include #include -#include -#include +#include +#include #include "xattr.h" #define XATTR_TRUSTED_PREFIX "trusted." static size_t -ext3_xattr_trusted_list(struct inode *inode, char *list, size_t list_size, +ext4_xattr_trusted_list(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { const size_t prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1; @@ -35,28 +35,28 @@ ext3_xattr_trusted_list(struct inode *inode, char *list, size_t list_size, } static int -ext3_xattr_trusted_get(struct inode *inode, const char *name, +ext4_xattr_trusted_get(struct inode *inode, const char *name, void *buffer, size_t size) { if (strcmp(name, "") == 0) return -EINVAL; - return ext3_xattr_get(inode, EXT3_XATTR_INDEX_TRUSTED, name, + return ext4_xattr_get(inode, EXT4_XATTR_INDEX_TRUSTED, name, buffer, size); } static int -ext3_xattr_trusted_set(struct inode *inode, const char *name, +ext4_xattr_trusted_set(struct inode *inode, const char *name, const void *value, size_t size, int flags) { if (strcmp(name, "") == 0) return -EINVAL; - return ext3_xattr_set(inode, EXT3_XATTR_INDEX_TRUSTED, name, + return ext4_xattr_set(inode, EXT4_XATTR_INDEX_TRUSTED, name, value, size, flags); } -struct xattr_handler ext3_xattr_trusted_handler = { +struct xattr_handler ext4_xattr_trusted_handler = { .prefix = XATTR_TRUSTED_PREFIX, - .list = ext3_xattr_trusted_list, - .get = ext3_xattr_trusted_get, - .set = ext3_xattr_trusted_set, + .list = ext4_xattr_trusted_list, + .get = ext4_xattr_trusted_get, + .set = ext4_xattr_trusted_set, }; diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c index a85a0a1..9c5a665 100644 --- a/fs/ext4/xattr_user.c +++ b/fs/ext4/xattr_user.c @@ -1,5 +1,5 @@ /* - * linux/fs/ext3/xattr_user.c + * linux/fs/ext4/xattr_user.c * Handler for extended user attributes. * * Copyright (C) 2001 by Andreas Gruenbacher, @@ -9,14 +9,14 @@ #include #include #include -#include -#include +#include +#include #include "xattr.h" #define XATTR_USER_PREFIX "user." static size_t -ext3_xattr_user_list(struct inode *inode, char *list, size_t list_size, +ext4_xattr_user_list(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { const size_t prefix_len = sizeof(XATTR_USER_PREFIX)-1; @@ -34,31 +34,31 @@ ext3_xattr_user_list(struct inode *inode, char *list, size_t list_size, } static int -ext3_xattr_user_get(struct inode *inode, const char *name, +ext4_xattr_user_get(struct inode *inode, const char *name, void *buffer, size_t size) { if (strcmp(name, "") == 0) return -EINVAL; if (!test_opt(inode->i_sb, XATTR_USER)) return -EOPNOTSUPP; - return ext3_xattr_get(inode, EXT3_XATTR_INDEX_USER, name, buffer, size); + return ext4_xattr_get(inode, EXT4_XATTR_INDEX_USER, name, buffer, size); } static int -ext3_xattr_user_set(struct inode *inode, const char *name, +ext4_xattr_user_set(struct inode *inode, const char *name, const void *value, size_t size, int flags) { if (strcmp(name, "") == 0) return -EINVAL; if (!test_opt(inode->i_sb, XATTR_USER)) return -EOPNOTSUPP; - return ext3_xattr_set(inode, EXT3_XATTR_INDEX_USER, name, + return ext4_xattr_set(inode, EXT4_XATTR_INDEX_USER, name, value, size, flags); } -struct xattr_handler ext3_xattr_user_handler = { +struct xattr_handler ext4_xattr_user_handler = { .prefix = XATTR_USER_PREFIX, - .list = ext3_xattr_user_list, - .get = ext3_xattr_user_get, - .set = ext3_xattr_user_set, + .list = ext4_xattr_user_list, + .get = ext4_xattr_user_get, + .set = ext4_xattr_user_set, }; diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 11cca1b..f582cd7 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/ext3_fs.h + * linux/include/linux/ext4_fs.h * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -13,8 +13,8 @@ * Copyright (C) 1991, 1992 Linus Torvalds */ -#ifndef _LINUX_EXT3_FS_H -#define _LINUX_EXT3_FS_H +#ifndef _LINUX_EXT4_FS_H +#define _LINUX_EXT4_FS_H #include #include @@ -24,102 +24,102 @@ */ /* - * Define EXT3FS_DEBUG to produce debug messages + * Define EXT4FS_DEBUG to produce debug messages */ -#undef EXT3FS_DEBUG +#undef EXT4FS_DEBUG /* - * Define EXT3_RESERVATION to reserve data blocks for expanding files + * Define EXT4_RESERVATION to reserve data blocks for expanding files */ -#define EXT3_DEFAULT_RESERVE_BLOCKS 8 +#define EXT4_DEFAULT_RESERVE_BLOCKS 8 /*max window size: 1024(direct blocks) + 3([t,d]indirect blocks) */ -#define EXT3_MAX_RESERVE_BLOCKS 1027 -#define EXT3_RESERVE_WINDOW_NOT_ALLOCATED 0 +#define EXT4_MAX_RESERVE_BLOCKS 1027 +#define EXT4_RESERVE_WINDOW_NOT_ALLOCATED 0 /* * Always enable hashed directories */ -#define CONFIG_EXT3_INDEX +#define CONFIG_EXT4_INDEX /* * Debug code */ -#ifdef EXT3FS_DEBUG -#define ext3_debug(f, a...) \ +#ifdef EXT4FS_DEBUG +#define ext4_debug(f, a...) \ do { \ - printk (KERN_DEBUG "EXT3-fs DEBUG (%s, %d): %s:", \ + printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ __FILE__, __LINE__, __FUNCTION__); \ printk (KERN_DEBUG f, ## a); \ } while (0) #else -#define ext3_debug(f, a...) do {} while (0) +#define ext4_debug(f, a...) do {} while (0) #endif /* * Special inodes numbers */ -#define EXT3_BAD_INO 1 /* Bad blocks inode */ -#define EXT3_ROOT_INO 2 /* Root inode */ -#define EXT3_BOOT_LOADER_INO 5 /* Boot loader inode */ -#define EXT3_UNDEL_DIR_INO 6 /* Undelete directory inode */ -#define EXT3_RESIZE_INO 7 /* Reserved group descriptors inode */ -#define EXT3_JOURNAL_INO 8 /* Journal inode */ +#define EXT4_BAD_INO 1 /* Bad blocks inode */ +#define EXT4_ROOT_INO 2 /* Root inode */ +#define EXT4_BOOT_LOADER_INO 5 /* Boot loader inode */ +#define EXT4_UNDEL_DIR_INO 6 /* Undelete directory inode */ +#define EXT4_RESIZE_INO 7 /* Reserved group descriptors inode */ +#define EXT4_JOURNAL_INO 8 /* Journal inode */ -/* First non-reserved inode for old ext3 filesystems */ -#define EXT3_GOOD_OLD_FIRST_INO 11 +/* First non-reserved inode for old ext4 filesystems */ +#define EXT4_GOOD_OLD_FIRST_INO 11 /* * Maximal count of links to a file */ -#define EXT3_LINK_MAX 32000 +#define EXT4_LINK_MAX 32000 /* * Macro-instructions used to manage several block sizes */ -#define EXT3_MIN_BLOCK_SIZE 1024 -#define EXT3_MAX_BLOCK_SIZE 4096 -#define EXT3_MIN_BLOCK_LOG_SIZE 10 +#define EXT4_MIN_BLOCK_SIZE 1024 +#define EXT4_MAX_BLOCK_SIZE 4096 +#define EXT4_MIN_BLOCK_LOG_SIZE 10 #ifdef __KERNEL__ -# define EXT3_BLOCK_SIZE(s) ((s)->s_blocksize) +# define EXT4_BLOCK_SIZE(s) ((s)->s_blocksize) #else -# define EXT3_BLOCK_SIZE(s) (EXT3_MIN_BLOCK_SIZE << (s)->s_log_block_size) +# define EXT4_BLOCK_SIZE(s) (EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size) #endif -#define EXT3_ADDR_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / sizeof (__u32)) +#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (__u32)) #ifdef __KERNEL__ -# define EXT3_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) +# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) #else -# define EXT3_BLOCK_SIZE_BITS(s) ((s)->s_log_block_size + 10) +# define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_log_block_size + 10) #endif #ifdef __KERNEL__ -#define EXT3_ADDR_PER_BLOCK_BITS(s) (EXT3_SB(s)->s_addr_per_block_bits) -#define EXT3_INODE_SIZE(s) (EXT3_SB(s)->s_inode_size) -#define EXT3_FIRST_INO(s) (EXT3_SB(s)->s_first_ino) +#define EXT4_ADDR_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_addr_per_block_bits) +#define EXT4_INODE_SIZE(s) (EXT4_SB(s)->s_inode_size) +#define EXT4_FIRST_INO(s) (EXT4_SB(s)->s_first_ino) #else -#define EXT3_INODE_SIZE(s) (((s)->s_rev_level == EXT3_GOOD_OLD_REV) ? \ - EXT3_GOOD_OLD_INODE_SIZE : \ +#define EXT4_INODE_SIZE(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \ + EXT4_GOOD_OLD_INODE_SIZE : \ (s)->s_inode_size) -#define EXT3_FIRST_INO(s) (((s)->s_rev_level == EXT3_GOOD_OLD_REV) ? \ - EXT3_GOOD_OLD_FIRST_INO : \ +#define EXT4_FIRST_INO(s) (((s)->s_rev_level == EXT4_GOOD_OLD_REV) ? \ + EXT4_GOOD_OLD_FIRST_INO : \ (s)->s_first_ino) #endif /* * Macro-instructions used to manage fragments */ -#define EXT3_MIN_FRAG_SIZE 1024 -#define EXT3_MAX_FRAG_SIZE 4096 -#define EXT3_MIN_FRAG_LOG_SIZE 10 +#define EXT4_MIN_FRAG_SIZE 1024 +#define EXT4_MAX_FRAG_SIZE 4096 +#define EXT4_MIN_FRAG_LOG_SIZE 10 #ifdef __KERNEL__ -# define EXT3_FRAG_SIZE(s) (EXT3_SB(s)->s_frag_size) -# define EXT3_FRAGS_PER_BLOCK(s) (EXT3_SB(s)->s_frags_per_block) +# define EXT4_FRAG_SIZE(s) (EXT4_SB(s)->s_frag_size) +# define EXT4_FRAGS_PER_BLOCK(s) (EXT4_SB(s)->s_frags_per_block) #else -# define EXT3_FRAG_SIZE(s) (EXT3_MIN_FRAG_SIZE << (s)->s_log_frag_size) -# define EXT3_FRAGS_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / EXT3_FRAG_SIZE(s)) +# define EXT4_FRAG_SIZE(s) (EXT4_MIN_FRAG_SIZE << (s)->s_log_frag_size) +# define EXT4_FRAGS_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / EXT4_FRAG_SIZE(s)) #endif /* * Structure of a blocks group descriptor */ -struct ext3_group_desc +struct ext4_group_desc { __le32 bg_block_bitmap; /* Blocks bitmap block */ __le32 bg_inode_bitmap; /* Inodes bitmap block */ @@ -135,62 +135,62 @@ struct ext3_group_desc * Macro-instructions used to manage group descriptors */ #ifdef __KERNEL__ -# define EXT3_BLOCKS_PER_GROUP(s) (EXT3_SB(s)->s_blocks_per_group) -# define EXT3_DESC_PER_BLOCK(s) (EXT3_SB(s)->s_desc_per_block) -# define EXT3_INODES_PER_GROUP(s) (EXT3_SB(s)->s_inodes_per_group) -# define EXT3_DESC_PER_BLOCK_BITS(s) (EXT3_SB(s)->s_desc_per_block_bits) +# define EXT4_BLOCKS_PER_GROUP(s) (EXT4_SB(s)->s_blocks_per_group) +# define EXT4_DESC_PER_BLOCK(s) (EXT4_SB(s)->s_desc_per_block) +# define EXT4_INODES_PER_GROUP(s) (EXT4_SB(s)->s_inodes_per_group) +# define EXT4_DESC_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_desc_per_block_bits) #else -# define EXT3_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group) -# define EXT3_DESC_PER_BLOCK(s) (EXT3_BLOCK_SIZE(s) / sizeof (struct ext3_group_desc)) -# define EXT3_INODES_PER_GROUP(s) ((s)->s_inodes_per_group) +# define EXT4_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group) +# define EXT4_DESC_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (struct ext4_group_desc)) +# define EXT4_INODES_PER_GROUP(s) ((s)->s_inodes_per_group) #endif /* * Constants relative to the data blocks */ -#define EXT3_NDIR_BLOCKS 12 -#define EXT3_IND_BLOCK EXT3_NDIR_BLOCKS -#define EXT3_DIND_BLOCK (EXT3_IND_BLOCK + 1) -#define EXT3_TIND_BLOCK (EXT3_DIND_BLOCK + 1) -#define EXT3_N_BLOCKS (EXT3_TIND_BLOCK + 1) +#define EXT4_NDIR_BLOCKS 12 +#define EXT4_IND_BLOCK EXT4_NDIR_BLOCKS +#define EXT4_DIND_BLOCK (EXT4_IND_BLOCK + 1) +#define EXT4_TIND_BLOCK (EXT4_DIND_BLOCK + 1) +#define EXT4_N_BLOCKS (EXT4_TIND_BLOCK + 1) /* * Inode flags */ -#define EXT3_SECRM_FL 0x00000001 /* Secure deletion */ -#define EXT3_UNRM_FL 0x00000002 /* Undelete */ -#define EXT3_COMPR_FL 0x00000004 /* Compress file */ -#define EXT3_SYNC_FL 0x00000008 /* Synchronous updates */ -#define EXT3_IMMUTABLE_FL 0x00000010 /* Immutable file */ -#define EXT3_APPEND_FL 0x00000020 /* writes to file may only append */ -#define EXT3_NODUMP_FL 0x00000040 /* do not dump file */ -#define EXT3_NOATIME_FL 0x00000080 /* do not update atime */ +#define EXT4_SECRM_FL 0x00000001 /* Secure deletion */ +#define EXT4_UNRM_FL 0x00000002 /* Undelete */ +#define EXT4_COMPR_FL 0x00000004 /* Compress file */ +#define EXT4_SYNC_FL 0x00000008 /* Synchronous updates */ +#define EXT4_IMMUTABLE_FL 0x00000010 /* Immutable file */ +#define EXT4_APPEND_FL 0x00000020 /* writes to file may only append */ +#define EXT4_NODUMP_FL 0x00000040 /* do not dump file */ +#define EXT4_NOATIME_FL 0x00000080 /* do not update atime */ /* Reserved for compression usage... */ -#define EXT3_DIRTY_FL 0x00000100 -#define EXT3_COMPRBLK_FL 0x00000200 /* One or more compressed clusters */ -#define EXT3_NOCOMPR_FL 0x00000400 /* Don't compress */ -#define EXT3_ECOMPR_FL 0x00000800 /* Compression error */ +#define EXT4_DIRTY_FL 0x00000100 +#define EXT4_COMPRBLK_FL 0x00000200 /* One or more compressed clusters */ +#define EXT4_NOCOMPR_FL 0x00000400 /* Don't compress */ +#define EXT4_ECOMPR_FL 0x00000800 /* Compression error */ /* End compression flags --- maybe not all used */ -#define EXT3_INDEX_FL 0x00001000 /* hash-indexed directory */ -#define EXT3_IMAGIC_FL 0x00002000 /* AFS directory */ -#define EXT3_JOURNAL_DATA_FL 0x00004000 /* file data should be journaled */ -#define EXT3_NOTAIL_FL 0x00008000 /* file tail should not be merged */ -#define EXT3_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ -#define EXT3_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ -#define EXT3_RESERVED_FL 0x80000000 /* reserved for ext3 lib */ +#define EXT4_INDEX_FL 0x00001000 /* hash-indexed directory */ +#define EXT4_IMAGIC_FL 0x00002000 /* AFS directory */ +#define EXT4_JOURNAL_DATA_FL 0x00004000 /* file data should be journaled */ +#define EXT4_NOTAIL_FL 0x00008000 /* file tail should not be merged */ +#define EXT4_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ +#define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ +#define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ -#define EXT3_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */ -#define EXT3_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ +#define EXT4_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */ +#define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ /* * Inode dynamic state flags */ -#define EXT3_STATE_JDATA 0x00000001 /* journaled data exists */ -#define EXT3_STATE_NEW 0x00000002 /* inode is newly created */ -#define EXT3_STATE_XATTR 0x00000004 /* has in-inode xattrs */ +#define EXT4_STATE_JDATA 0x00000001 /* journaled data exists */ +#define EXT4_STATE_NEW 0x00000002 /* inode is newly created */ +#define EXT4_STATE_XATTR 0x00000004 /* has in-inode xattrs */ /* Used to pass group descriptor data when online resize is done */ -struct ext3_new_group_input { +struct ext4_new_group_input { __u32 group; /* Group number for this data */ __u32 block_bitmap; /* Absolute block number of block bitmap */ __u32 inode_bitmap; /* Absolute block number of inode bitmap */ @@ -200,8 +200,8 @@ struct ext3_new_group_input { __u16 unused; }; -/* The struct ext3_new_group_input in kernel space, with free_blocks_count */ -struct ext3_new_group_data { +/* The struct ext4_new_group_input in kernel space, with free_blocks_count */ +struct ext4_new_group_data { __u32 group; __u32 block_bitmap; __u32 inode_bitmap; @@ -216,41 +216,41 @@ struct ext3_new_group_data { /* * ioctl commands */ -#define EXT3_IOC_GETFLAGS FS_IOC_GETFLAGS -#define EXT3_IOC_SETFLAGS FS_IOC_SETFLAGS -#define EXT3_IOC_GETVERSION _IOR('f', 3, long) -#define EXT3_IOC_SETVERSION _IOW('f', 4, long) -#define EXT3_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) -#define EXT3_IOC_GROUP_ADD _IOW('f', 8,struct ext3_new_group_input) -#define EXT3_IOC_GETVERSION_OLD FS_IOC_GETVERSION -#define EXT3_IOC_SETVERSION_OLD FS_IOC_SETVERSION +#define EXT4_IOC_GETFLAGS FS_IOC_GETFLAGS +#define EXT4_IOC_SETFLAGS FS_IOC_SETFLAGS +#define EXT4_IOC_GETVERSION _IOR('f', 3, long) +#define EXT4_IOC_SETVERSION _IOW('f', 4, long) +#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) +#define EXT4_IOC_GROUP_ADD _IOW('f', 8,struct ext4_new_group_input) +#define EXT4_IOC_GETVERSION_OLD FS_IOC_GETVERSION +#define EXT4_IOC_SETVERSION_OLD FS_IOC_SETVERSION #ifdef CONFIG_JBD_DEBUG -#define EXT3_IOC_WAIT_FOR_READONLY _IOR('f', 99, long) +#define EXT4_IOC_WAIT_FOR_READONLY _IOR('f', 99, long) #endif -#define EXT3_IOC_GETRSVSZ _IOR('f', 5, long) -#define EXT3_IOC_SETRSVSZ _IOW('f', 6, long) +#define EXT4_IOC_GETRSVSZ _IOR('f', 5, long) +#define EXT4_IOC_SETRSVSZ _IOW('f', 6, long) /* * ioctl commands in 32 bit emulation */ -#define EXT3_IOC32_GETFLAGS FS_IOC32_GETFLAGS -#define EXT3_IOC32_SETFLAGS FS_IOC32_SETFLAGS -#define EXT3_IOC32_GETVERSION _IOR('f', 3, int) -#define EXT3_IOC32_SETVERSION _IOW('f', 4, int) -#define EXT3_IOC32_GETRSVSZ _IOR('f', 5, int) -#define EXT3_IOC32_SETRSVSZ _IOW('f', 6, int) -#define EXT3_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int) +#define EXT4_IOC32_GETFLAGS FS_IOC32_GETFLAGS +#define EXT4_IOC32_SETFLAGS FS_IOC32_SETFLAGS +#define EXT4_IOC32_GETVERSION _IOR('f', 3, int) +#define EXT4_IOC32_SETVERSION _IOW('f', 4, int) +#define EXT4_IOC32_GETRSVSZ _IOR('f', 5, int) +#define EXT4_IOC32_SETRSVSZ _IOW('f', 6, int) +#define EXT4_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int) #ifdef CONFIG_JBD_DEBUG -#define EXT3_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int) +#define EXT4_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int) #endif -#define EXT3_IOC32_GETVERSION_OLD FS_IOC32_GETVERSION -#define EXT3_IOC32_SETVERSION_OLD FS_IOC32_SETVERSION +#define EXT4_IOC32_GETVERSION_OLD FS_IOC32_GETVERSION +#define EXT4_IOC32_SETVERSION_OLD FS_IOC32_SETVERSION /* * Mount options */ -struct ext3_mount_options { +struct ext4_mount_options { unsigned long s_mount_opt; uid_t s_resuid; gid_t s_resgid; @@ -264,7 +264,7 @@ struct ext3_mount_options { /* * Structure of an inode on the disk */ -struct ext3_inode { +struct ext4_inode { __le16 i_mode; /* File mode */ __le16 i_uid; /* Low 16 bits of Owner Uid */ __le32 i_size; /* Size in bytes */ @@ -287,7 +287,7 @@ struct ext3_inode { __u32 m_i_reserved1; } masix1; } osd1; /* OS dependent 1 */ - __le32 i_block[EXT3_N_BLOCKS];/* Pointers to blocks */ + __le32 i_block[EXT4_N_BLOCKS];/* Pointers to blocks */ __le32 i_generation; /* File version (for NFS) */ __le32 i_file_acl; /* File ACL */ __le32 i_dir_acl; /* Directory ACL */ @@ -353,76 +353,76 @@ struct ext3_inode { /* * File system states */ -#define EXT3_VALID_FS 0x0001 /* Unmounted cleanly */ -#define EXT3_ERROR_FS 0x0002 /* Errors detected */ -#define EXT3_ORPHAN_FS 0x0004 /* Orphans being recovered */ +#define EXT4_VALID_FS 0x0001 /* Unmounted cleanly */ +#define EXT4_ERROR_FS 0x0002 /* Errors detected */ +#define EXT4_ORPHAN_FS 0x0004 /* Orphans being recovered */ /* * Mount flags */ -#define EXT3_MOUNT_CHECK 0x00001 /* Do mount-time checks */ -#define EXT3_MOUNT_OLDALLOC 0x00002 /* Don't use the new Orlov allocator */ -#define EXT3_MOUNT_GRPID 0x00004 /* Create files with directory's group */ -#define EXT3_MOUNT_DEBUG 0x00008 /* Some debugging messages */ -#define EXT3_MOUNT_ERRORS_CONT 0x00010 /* Continue on errors */ -#define EXT3_MOUNT_ERRORS_RO 0x00020 /* Remount fs ro on errors */ -#define EXT3_MOUNT_ERRORS_PANIC 0x00040 /* Panic on errors */ -#define EXT3_MOUNT_MINIX_DF 0x00080 /* Mimics the Minix statfs */ -#define EXT3_MOUNT_NOLOAD 0x00100 /* Don't use existing journal*/ -#define EXT3_MOUNT_ABORT 0x00200 /* Fatal error detected */ -#define EXT3_MOUNT_DATA_FLAGS 0x00C00 /* Mode for data writes: */ -#define EXT3_MOUNT_JOURNAL_DATA 0x00400 /* Write data to journal */ -#define EXT3_MOUNT_ORDERED_DATA 0x00800 /* Flush data before commit */ -#define EXT3_MOUNT_WRITEBACK_DATA 0x00C00 /* No data ordering */ -#define EXT3_MOUNT_UPDATE_JOURNAL 0x01000 /* Update the journal format */ -#define EXT3_MOUNT_NO_UID32 0x02000 /* Disable 32-bit UIDs */ -#define EXT3_MOUNT_XATTR_USER 0x04000 /* Extended user attributes */ -#define EXT3_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */ -#define EXT3_MOUNT_RESERVATION 0x10000 /* Preallocation */ -#define EXT3_MOUNT_BARRIER 0x20000 /* Use block barriers */ -#define EXT3_MOUNT_NOBH 0x40000 /* No bufferheads */ -#define EXT3_MOUNT_QUOTA 0x80000 /* Some quota option set */ -#define EXT3_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ -#define EXT3_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ - -/* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ +#define EXT4_MOUNT_CHECK 0x00001 /* Do mount-time checks */ +#define EXT4_MOUNT_OLDALLOC 0x00002 /* Don't use the new Orlov allocator */ +#define EXT4_MOUNT_GRPID 0x00004 /* Create files with directory's group */ +#define EXT4_MOUNT_DEBUG 0x00008 /* Some debugging messages */ +#define EXT4_MOUNT_ERRORS_CONT 0x00010 /* Continue on errors */ +#define EXT4_MOUNT_ERRORS_RO 0x00020 /* Remount fs ro on errors */ +#define EXT4_MOUNT_ERRORS_PANIC 0x00040 /* Panic on errors */ +#define EXT4_MOUNT_MINIX_DF 0x00080 /* Mimics the Minix statfs */ +#define EXT4_MOUNT_NOLOAD 0x00100 /* Don't use existing journal*/ +#define EXT4_MOUNT_ABORT 0x00200 /* Fatal error detected */ +#define EXT4_MOUNT_DATA_FLAGS 0x00C00 /* Mode for data writes: */ +#define EXT4_MOUNT_JOURNAL_DATA 0x00400 /* Write data to journal */ +#define EXT4_MOUNT_ORDERED_DATA 0x00800 /* Flush data before commit */ +#define EXT4_MOUNT_WRITEBACK_DATA 0x00C00 /* No data ordering */ +#define EXT4_MOUNT_UPDATE_JOURNAL 0x01000 /* Update the journal format */ +#define EXT4_MOUNT_NO_UID32 0x02000 /* Disable 32-bit UIDs */ +#define EXT4_MOUNT_XATTR_USER 0x04000 /* Extended user attributes */ +#define EXT4_MOUNT_POSIX_ACL 0x08000 /* POSIX Access Control Lists */ +#define EXT4_MOUNT_RESERVATION 0x10000 /* Preallocation */ +#define EXT4_MOUNT_BARRIER 0x20000 /* Use block barriers */ +#define EXT4_MOUNT_NOBH 0x40000 /* No bufferheads */ +#define EXT4_MOUNT_QUOTA 0x80000 /* Some quota option set */ +#define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ +#define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ + +/* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H -#define clear_opt(o, opt) o &= ~EXT3_MOUNT_##opt -#define set_opt(o, opt) o |= EXT3_MOUNT_##opt -#define test_opt(sb, opt) (EXT3_SB(sb)->s_mount_opt & \ - EXT3_MOUNT_##opt) +#define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt +#define set_opt(o, opt) o |= EXT4_MOUNT_##opt +#define test_opt(sb, opt) (EXT4_SB(sb)->s_mount_opt & \ + EXT4_MOUNT_##opt) #else -#define EXT2_MOUNT_NOLOAD EXT3_MOUNT_NOLOAD -#define EXT2_MOUNT_ABORT EXT3_MOUNT_ABORT -#define EXT2_MOUNT_DATA_FLAGS EXT3_MOUNT_DATA_FLAGS +#define EXT2_MOUNT_NOLOAD EXT4_MOUNT_NOLOAD +#define EXT2_MOUNT_ABORT EXT4_MOUNT_ABORT +#define EXT2_MOUNT_DATA_FLAGS EXT4_MOUNT_DATA_FLAGS #endif -#define ext3_set_bit ext2_set_bit -#define ext3_set_bit_atomic ext2_set_bit_atomic -#define ext3_clear_bit ext2_clear_bit -#define ext3_clear_bit_atomic ext2_clear_bit_atomic -#define ext3_test_bit ext2_test_bit -#define ext3_find_first_zero_bit ext2_find_first_zero_bit -#define ext3_find_next_zero_bit ext2_find_next_zero_bit +#define ext4_set_bit ext2_set_bit +#define ext4_set_bit_atomic ext2_set_bit_atomic +#define ext4_clear_bit ext2_clear_bit +#define ext4_clear_bit_atomic ext2_clear_bit_atomic +#define ext4_test_bit ext2_test_bit +#define ext4_find_first_zero_bit ext2_find_first_zero_bit +#define ext4_find_next_zero_bit ext2_find_next_zero_bit /* * Maximal mount counts between two filesystem checks */ -#define EXT3_DFL_MAX_MNT_COUNT 20 /* Allow 20 mounts */ -#define EXT3_DFL_CHECKINTERVAL 0 /* Don't use interval check */ +#define EXT4_DFL_MAX_MNT_COUNT 20 /* Allow 20 mounts */ +#define EXT4_DFL_CHECKINTERVAL 0 /* Don't use interval check */ /* * Behaviour when detecting errors */ -#define EXT3_ERRORS_CONTINUE 1 /* Continue execution */ -#define EXT3_ERRORS_RO 2 /* Remount fs read-only */ -#define EXT3_ERRORS_PANIC 3 /* Panic */ -#define EXT3_ERRORS_DEFAULT EXT3_ERRORS_CONTINUE +#define EXT4_ERRORS_CONTINUE 1 /* Continue execution */ +#define EXT4_ERRORS_RO 2 /* Remount fs read-only */ +#define EXT4_ERRORS_PANIC 3 /* Panic */ +#define EXT4_ERRORS_DEFAULT EXT4_ERRORS_CONTINUE /* * Structure of the super block */ -struct ext3_super_block { +struct ext4_super_block { /*00*/ __le32 s_inodes_count; /* Inodes count */ __le32 s_blocks_count; /* Blocks count */ __le32 s_r_blocks_count; /* Reserved blocks count */ @@ -449,7 +449,7 @@ struct ext3_super_block { /*50*/ __le16 s_def_resuid; /* Default uid for reserved blocks */ __le16 s_def_resgid; /* Default gid for reserved blocks */ /* - * These fields are for EXT3_DYNAMIC_REV superblocks only. + * These fields are for EXT4_DYNAMIC_REV superblocks only. * * Note: the difference between the compatible feature set and * the incompatible feature set is that if there is a bit set @@ -473,13 +473,13 @@ struct ext3_super_block { /*C8*/ __le32 s_algorithm_usage_bitmap; /* For compression */ /* * Performance hints. Directory preallocation should only - * happen if the EXT3_FEATURE_COMPAT_DIR_PREALLOC flag is on. + * happen if the EXT4_FEATURE_COMPAT_DIR_PREALLOC flag is on. */ __u8 s_prealloc_blocks; /* Nr of blocks to try to preallocate*/ __u8 s_prealloc_dir_blocks; /* Nr to preallocate for dirs */ __le16 s_reserved_gdt_blocks; /* Per group desc for online growth */ /* - * Journaling support valid if EXT3_FEATURE_COMPAT_HAS_JOURNAL set. + * Journaling support valid if EXT4_FEATURE_COMPAT_HAS_JOURNAL set. */ /*D0*/ __u8 s_journal_uuid[16]; /* uuid of journal superblock */ /*E0*/ __le32 s_journal_inum; /* inode number of journal file */ @@ -495,186 +495,186 @@ struct ext3_super_block { }; #ifdef __KERNEL__ -#include -#include -static inline struct ext3_sb_info * EXT3_SB(struct super_block *sb) +#include +#include +static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb) { return sb->s_fs_info; } -static inline struct ext3_inode_info *EXT3_I(struct inode *inode) +static inline struct ext4_inode_info *EXT4_I(struct inode *inode) { - return container_of(inode, struct ext3_inode_info, vfs_inode); + return container_of(inode, struct ext4_inode_info, vfs_inode); } -static inline int ext3_valid_inum(struct super_block *sb, unsigned long ino) +static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) { - return ino == EXT3_ROOT_INO || - ino == EXT3_JOURNAL_INO || - ino == EXT3_RESIZE_INO || - (ino >= EXT3_FIRST_INO(sb) && - ino <= le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)); + return ino == EXT4_ROOT_INO || + ino == EXT4_JOURNAL_INO || + ino == EXT4_RESIZE_INO || + (ino >= EXT4_FIRST_INO(sb) && + ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count)); } #else -/* Assume that user mode programs are passing in an ext3fs superblock, not +/* Assume that user mode programs are passing in an ext4fs superblock, not * a kernel struct super_block. This will allow us to call the feature-test * macros from user land. */ -#define EXT3_SB(sb) (sb) +#define EXT4_SB(sb) (sb) #endif -#define NEXT_ORPHAN(inode) EXT3_I(inode)->i_dtime +#define NEXT_ORPHAN(inode) EXT4_I(inode)->i_dtime /* * Codes for operating systems */ -#define EXT3_OS_LINUX 0 -#define EXT3_OS_HURD 1 -#define EXT3_OS_MASIX 2 -#define EXT3_OS_FREEBSD 3 -#define EXT3_OS_LITES 4 +#define EXT4_OS_LINUX 0 +#define EXT4_OS_HURD 1 +#define EXT4_OS_MASIX 2 +#define EXT4_OS_FREEBSD 3 +#define EXT4_OS_LITES 4 /* * Revision levels */ -#define EXT3_GOOD_OLD_REV 0 /* The good old (original) format */ -#define EXT3_DYNAMIC_REV 1 /* V2 format w/ dynamic inode sizes */ +#define EXT4_GOOD_OLD_REV 0 /* The good old (original) format */ +#define EXT4_DYNAMIC_REV 1 /* V2 format w/ dynamic inode sizes */ -#define EXT3_CURRENT_REV EXT3_GOOD_OLD_REV -#define EXT3_MAX_SUPP_REV EXT3_DYNAMIC_REV +#define EXT4_CURRENT_REV EXT4_GOOD_OLD_REV +#define EXT4_MAX_SUPP_REV EXT4_DYNAMIC_REV -#define EXT3_GOOD_OLD_INODE_SIZE 128 +#define EXT4_GOOD_OLD_INODE_SIZE 128 /* * Feature set definitions */ -#define EXT3_HAS_COMPAT_FEATURE(sb,mask) \ - ( EXT3_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) -#define EXT3_HAS_RO_COMPAT_FEATURE(sb,mask) \ - ( EXT3_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) -#define EXT3_HAS_INCOMPAT_FEATURE(sb,mask) \ - ( EXT3_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) -#define EXT3_SET_COMPAT_FEATURE(sb,mask) \ - EXT3_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) -#define EXT3_SET_RO_COMPAT_FEATURE(sb,mask) \ - EXT3_SB(sb)->s_es->s_feature_ro_compat |= cpu_to_le32(mask) -#define EXT3_SET_INCOMPAT_FEATURE(sb,mask) \ - EXT3_SB(sb)->s_es->s_feature_incompat |= cpu_to_le32(mask) -#define EXT3_CLEAR_COMPAT_FEATURE(sb,mask) \ - EXT3_SB(sb)->s_es->s_feature_compat &= ~cpu_to_le32(mask) -#define EXT3_CLEAR_RO_COMPAT_FEATURE(sb,mask) \ - EXT3_SB(sb)->s_es->s_feature_ro_compat &= ~cpu_to_le32(mask) -#define EXT3_CLEAR_INCOMPAT_FEATURE(sb,mask) \ - EXT3_SB(sb)->s_es->s_feature_incompat &= ~cpu_to_le32(mask) - -#define EXT3_FEATURE_COMPAT_DIR_PREALLOC 0x0001 -#define EXT3_FEATURE_COMPAT_IMAGIC_INODES 0x0002 -#define EXT3_FEATURE_COMPAT_HAS_JOURNAL 0x0004 -#define EXT3_FEATURE_COMPAT_EXT_ATTR 0x0008 -#define EXT3_FEATURE_COMPAT_RESIZE_INODE 0x0010 -#define EXT3_FEATURE_COMPAT_DIR_INDEX 0x0020 - -#define EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001 -#define EXT3_FEATURE_RO_COMPAT_LARGE_FILE 0x0002 -#define EXT3_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 - -#define EXT3_FEATURE_INCOMPAT_COMPRESSION 0x0001 -#define EXT3_FEATURE_INCOMPAT_FILETYPE 0x0002 -#define EXT3_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ -#define EXT3_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ -#define EXT3_FEATURE_INCOMPAT_META_BG 0x0010 - -#define EXT3_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR -#define EXT3_FEATURE_INCOMPAT_SUPP (EXT3_FEATURE_INCOMPAT_FILETYPE| \ - EXT3_FEATURE_INCOMPAT_RECOVER| \ - EXT3_FEATURE_INCOMPAT_META_BG) -#define EXT3_FEATURE_RO_COMPAT_SUPP (EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER| \ - EXT3_FEATURE_RO_COMPAT_LARGE_FILE| \ - EXT3_FEATURE_RO_COMPAT_BTREE_DIR) +#define EXT4_HAS_COMPAT_FEATURE(sb,mask) \ + ( EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) +#define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask) \ + ( EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) +#define EXT4_HAS_INCOMPAT_FEATURE(sb,mask) \ + ( EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) +#define EXT4_SET_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) +#define EXT4_SET_RO_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_ro_compat |= cpu_to_le32(mask) +#define EXT4_SET_INCOMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_incompat |= cpu_to_le32(mask) +#define EXT4_CLEAR_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_compat &= ~cpu_to_le32(mask) +#define EXT4_CLEAR_RO_COMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_ro_compat &= ~cpu_to_le32(mask) +#define EXT4_CLEAR_INCOMPAT_FEATURE(sb,mask) \ + EXT4_SB(sb)->s_es->s_feature_incompat &= ~cpu_to_le32(mask) + +#define EXT4_FEATURE_COMPAT_DIR_PREALLOC 0x0001 +#define EXT4_FEATURE_COMPAT_IMAGIC_INODES 0x0002 +#define EXT4_FEATURE_COMPAT_HAS_JOURNAL 0x0004 +#define EXT4_FEATURE_COMPAT_EXT_ATTR 0x0008 +#define EXT4_FEATURE_COMPAT_RESIZE_INODE 0x0010 +#define EXT4_FEATURE_COMPAT_DIR_INDEX 0x0020 + +#define EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER 0x0001 +#define EXT4_FEATURE_RO_COMPAT_LARGE_FILE 0x0002 +#define EXT4_FEATURE_RO_COMPAT_BTREE_DIR 0x0004 + +#define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001 +#define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002 +#define EXT4_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ +#define EXT4_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ +#define EXT4_FEATURE_INCOMPAT_META_BG 0x0010 + +#define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR +#define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ + EXT4_FEATURE_INCOMPAT_RECOVER| \ + EXT4_FEATURE_INCOMPAT_META_BG) +#define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ + EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ + EXT4_FEATURE_RO_COMPAT_BTREE_DIR) /* * Default values for user and/or group using reserved blocks */ -#define EXT3_DEF_RESUID 0 -#define EXT3_DEF_RESGID 0 +#define EXT4_DEF_RESUID 0 +#define EXT4_DEF_RESGID 0 /* * Default mount options */ -#define EXT3_DEFM_DEBUG 0x0001 -#define EXT3_DEFM_BSDGROUPS 0x0002 -#define EXT3_DEFM_XATTR_USER 0x0004 -#define EXT3_DEFM_ACL 0x0008 -#define EXT3_DEFM_UID16 0x0010 -#define EXT3_DEFM_JMODE 0x0060 -#define EXT3_DEFM_JMODE_DATA 0x0020 -#define EXT3_DEFM_JMODE_ORDERED 0x0040 -#define EXT3_DEFM_JMODE_WBACK 0x0060 +#define EXT4_DEFM_DEBUG 0x0001 +#define EXT4_DEFM_BSDGROUPS 0x0002 +#define EXT4_DEFM_XATTR_USER 0x0004 +#define EXT4_DEFM_ACL 0x0008 +#define EXT4_DEFM_UID16 0x0010 +#define EXT4_DEFM_JMODE 0x0060 +#define EXT4_DEFM_JMODE_DATA 0x0020 +#define EXT4_DEFM_JMODE_ORDERED 0x0040 +#define EXT4_DEFM_JMODE_WBACK 0x0060 /* * Structure of a directory entry */ -#define EXT3_NAME_LEN 255 +#define EXT4_NAME_LEN 255 -struct ext3_dir_entry { +struct ext4_dir_entry { __le32 inode; /* Inode number */ __le16 rec_len; /* Directory entry length */ __le16 name_len; /* Name length */ - char name[EXT3_NAME_LEN]; /* File name */ + char name[EXT4_NAME_LEN]; /* File name */ }; /* - * The new version of the directory entry. Since EXT3 structures are + * The new version of the directory entry. Since EXT4 structures are * stored in intel byte order, and the name_len field could never be * bigger than 255 chars, it's safe to reclaim the extra byte for the * file_type field. */ -struct ext3_dir_entry_2 { +struct ext4_dir_entry_2 { __le32 inode; /* Inode number */ __le16 rec_len; /* Directory entry length */ __u8 name_len; /* Name length */ __u8 file_type; - char name[EXT3_NAME_LEN]; /* File name */ + char name[EXT4_NAME_LEN]; /* File name */ }; /* - * Ext3 directory file types. Only the low 3 bits are used. The + * Ext4 directory file types. Only the low 3 bits are used. The * other bits are reserved for now. */ -#define EXT3_FT_UNKNOWN 0 -#define EXT3_FT_REG_FILE 1 -#define EXT3_FT_DIR 2 -#define EXT3_FT_CHRDEV 3 -#define EXT3_FT_BLKDEV 4 -#define EXT3_FT_FIFO 5 -#define EXT3_FT_SOCK 6 -#define EXT3_FT_SYMLINK 7 +#define EXT4_FT_UNKNOWN 0 +#define EXT4_FT_REG_FILE 1 +#define EXT4_FT_DIR 2 +#define EXT4_FT_CHRDEV 3 +#define EXT4_FT_BLKDEV 4 +#define EXT4_FT_FIFO 5 +#define EXT4_FT_SOCK 6 +#define EXT4_FT_SYMLINK 7 -#define EXT3_FT_MAX 8 +#define EXT4_FT_MAX 8 /* - * EXT3_DIR_PAD defines the directory entries boundaries + * EXT4_DIR_PAD defines the directory entries boundaries * * NOTE: It must be a multiple of 4 */ -#define EXT3_DIR_PAD 4 -#define EXT3_DIR_ROUND (EXT3_DIR_PAD - 1) -#define EXT3_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT3_DIR_ROUND) & \ - ~EXT3_DIR_ROUND) +#define EXT4_DIR_PAD 4 +#define EXT4_DIR_ROUND (EXT4_DIR_PAD - 1) +#define EXT4_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT4_DIR_ROUND) & \ + ~EXT4_DIR_ROUND) /* * Hash Tree Directory indexing * (c) Daniel Phillips, 2001 */ -#ifdef CONFIG_EXT3_INDEX - #define is_dx(dir) (EXT3_HAS_COMPAT_FEATURE(dir->i_sb, \ - EXT3_FEATURE_COMPAT_DIR_INDEX) && \ - (EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) -#define EXT3_DIR_LINK_MAX(dir) (!is_dx(dir) && (dir)->i_nlink >= EXT3_LINK_MAX) -#define EXT3_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2 || (dir)->i_nlink == 1) +#ifdef CONFIG_EXT4_INDEX + #define is_dx(dir) (EXT4_HAS_COMPAT_FEATURE(dir->i_sb, \ + EXT4_FEATURE_COMPAT_DIR_INDEX) && \ + (EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) +#define EXT4_DIR_LINK_MAX(dir) (!is_dx(dir) && (dir)->i_nlink >= EXT4_LINK_MAX) +#define EXT4_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2 || (dir)->i_nlink == 1) #else #define is_dx(dir) 0 -#define EXT3_DIR_LINK_MAX(dir) ((dir)->i_nlink >= EXT3_LINK_MAX) -#define EXT3_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2) +#define EXT4_DIR_LINK_MAX(dir) ((dir)->i_nlink >= EXT4_LINK_MAX) +#define EXT4_DIR_LINK_EMPTY(dir) ((dir)->i_nlink == 2) #endif /* Legal values for the dx_root hash_version field: */ @@ -694,10 +694,10 @@ struct dx_hash_info u32 *seed; }; -#define EXT3_HTREE_EOF 0x7fffffff +#define EXT4_HTREE_EOF 0x7fffffff /* - * Control parameters used by ext3_htree_next_block + * Control parameters used by ext4_htree_next_block */ #define HASH_NB_ALWAYS 1 @@ -705,16 +705,16 @@ struct dx_hash_info /* * Describe an inode's exact location on disk and in memory */ -struct ext3_iloc +struct ext4_iloc { struct buffer_head *bh; unsigned long offset; unsigned long block_group; }; -static inline struct ext3_inode *ext3_raw_inode(struct ext3_iloc *iloc) +static inline struct ext4_inode *ext4_raw_inode(struct ext4_iloc *iloc) { - return (struct ext3_inode *) (iloc->bh->b_data + iloc->offset); + return (struct ext4_inode *) (iloc->bh->b_data + iloc->offset); } /* @@ -733,11 +733,11 @@ struct dir_private_info { }; /* calculate the first block number of the group */ -static inline ext3_fsblk_t -ext3_group_first_block_no(struct super_block *sb, unsigned long group_no) +static inline ext4_fsblk_t +ext4_group_first_block_no(struct super_block *sb, unsigned long group_no) { - return group_no * (ext3_fsblk_t)EXT3_BLOCKS_PER_GROUP(sb) + - le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block); + return group_no * (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) + + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); } /* @@ -751,113 +751,113 @@ ext3_group_first_block_no(struct super_block *sb, unsigned long group_no) /* * Ok, these declarations are also in but none of the - * ext3 source programs needs to include it so they are duplicated here. + * ext4 source programs needs to include it so they are duplicated here. */ # define NORET_TYPE /**/ # define ATTRIB_NORET __attribute__((noreturn)) # define NORET_AND noreturn, /* balloc.c */ -extern int ext3_bg_has_super(struct super_block *sb, int group); -extern unsigned long ext3_bg_num_gdb(struct super_block *sb, int group); -extern ext3_fsblk_t ext3_new_block (handle_t *handle, struct inode *inode, - ext3_fsblk_t goal, int *errp); -extern ext3_fsblk_t ext3_new_blocks (handle_t *handle, struct inode *inode, - ext3_fsblk_t goal, unsigned long *count, int *errp); -extern void ext3_free_blocks (handle_t *handle, struct inode *inode, - ext3_fsblk_t block, unsigned long count); -extern void ext3_free_blocks_sb (handle_t *handle, struct super_block *sb, - ext3_fsblk_t block, unsigned long count, +extern int ext4_bg_has_super(struct super_block *sb, int group); +extern unsigned long ext4_bg_num_gdb(struct super_block *sb, int group); +extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int *errp); +extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp); +extern void ext4_free_blocks (handle_t *handle, struct inode *inode, + ext4_fsblk_t block, unsigned long count); +extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count, unsigned long *pdquot_freed_blocks); -extern ext3_fsblk_t ext3_count_free_blocks (struct super_block *); -extern void ext3_check_blocks_bitmap (struct super_block *); -extern struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb, +extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *); +extern void ext4_check_blocks_bitmap (struct super_block *); +extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, unsigned int block_group, struct buffer_head ** bh); -extern int ext3_should_retry_alloc(struct super_block *sb, int *retries); -extern void ext3_init_block_alloc_info(struct inode *); -extern void ext3_rsv_window_add(struct super_block *sb, struct ext3_reserve_window_node *rsv); +extern int ext4_should_retry_alloc(struct super_block *sb, int *retries); +extern void ext4_init_block_alloc_info(struct inode *); +extern void ext4_rsv_window_add(struct super_block *sb, struct ext4_reserve_window_node *rsv); /* dir.c */ -extern int ext3_check_dir_entry(const char *, struct inode *, - struct ext3_dir_entry_2 *, +extern int ext4_check_dir_entry(const char *, struct inode *, + struct ext4_dir_entry_2 *, struct buffer_head *, unsigned long); -extern int ext3_htree_store_dirent(struct file *dir_file, __u32 hash, +extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, __u32 minor_hash, - struct ext3_dir_entry_2 *dirent); -extern void ext3_htree_free_dir_info(struct dir_private_info *p); + struct ext4_dir_entry_2 *dirent); +extern void ext4_htree_free_dir_info(struct dir_private_info *p); /* fsync.c */ -extern int ext3_sync_file (struct file *, struct dentry *, int); +extern int ext4_sync_file (struct file *, struct dentry *, int); /* hash.c */ -extern int ext3fs_dirhash(const char *name, int len, struct +extern int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo); /* ialloc.c */ -extern struct inode * ext3_new_inode (handle_t *, struct inode *, int); -extern void ext3_free_inode (handle_t *, struct inode *); -extern struct inode * ext3_orphan_get (struct super_block *, unsigned long); -extern unsigned long ext3_count_free_inodes (struct super_block *); -extern unsigned long ext3_count_dirs (struct super_block *); -extern void ext3_check_inodes_bitmap (struct super_block *); -extern unsigned long ext3_count_free (struct buffer_head *, unsigned); +extern struct inode * ext4_new_inode (handle_t *, struct inode *, int); +extern void ext4_free_inode (handle_t *, struct inode *); +extern struct inode * ext4_orphan_get (struct super_block *, unsigned long); +extern unsigned long ext4_count_free_inodes (struct super_block *); +extern unsigned long ext4_count_dirs (struct super_block *); +extern void ext4_check_inodes_bitmap (struct super_block *); +extern unsigned long ext4_count_free (struct buffer_head *, unsigned); /* inode.c */ -int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, - struct buffer_head *bh, ext3_fsblk_t blocknr); -struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *); -struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *); -int ext3_get_blocks_handle(handle_t *handle, struct inode *inode, +int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, + struct buffer_head *bh, ext4_fsblk_t blocknr); +struct buffer_head * ext4_getblk (handle_t *, struct inode *, long, int, int *); +struct buffer_head * ext4_bread (handle_t *, struct inode *, int, int, int *); +int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, sector_t iblock, unsigned long maxblocks, struct buffer_head *bh_result, int create, int extend_disksize); -extern void ext3_read_inode (struct inode *); -extern int ext3_write_inode (struct inode *, int); -extern int ext3_setattr (struct dentry *, struct iattr *); -extern void ext3_delete_inode (struct inode *); -extern int ext3_sync_inode (handle_t *, struct inode *); -extern void ext3_discard_reservation (struct inode *); -extern void ext3_dirty_inode(struct inode *); -extern int ext3_change_inode_journal_flag(struct inode *, int); -extern int ext3_get_inode_loc(struct inode *, struct ext3_iloc *); -extern void ext3_truncate (struct inode *); -extern void ext3_set_inode_flags(struct inode *); -extern void ext3_set_aops(struct inode *inode); +extern void ext4_read_inode (struct inode *); +extern int ext4_write_inode (struct inode *, int); +extern int ext4_setattr (struct dentry *, struct iattr *); +extern void ext4_delete_inode (struct inode *); +extern int ext4_sync_inode (handle_t *, struct inode *); +extern void ext4_discard_reservation (struct inode *); +extern void ext4_dirty_inode(struct inode *); +extern int ext4_change_inode_journal_flag(struct inode *, int); +extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); +extern void ext4_truncate (struct inode *); +extern void ext4_set_inode_flags(struct inode *); +extern void ext4_set_aops(struct inode *inode); /* ioctl.c */ -extern int ext3_ioctl (struct inode *, struct file *, unsigned int, +extern int ext4_ioctl (struct inode *, struct file *, unsigned int, unsigned long); -extern long ext3_compat_ioctl (struct file *, unsigned int, unsigned long); +extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long); /* namei.c */ -extern int ext3_orphan_add(handle_t *, struct inode *); -extern int ext3_orphan_del(handle_t *, struct inode *); -extern int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, +extern int ext4_orphan_add(handle_t *, struct inode *); +extern int ext4_orphan_del(handle_t *, struct inode *); +extern int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, __u32 start_minor_hash, __u32 *next_hash); /* resize.c */ -extern int ext3_group_add(struct super_block *sb, - struct ext3_new_group_data *input); -extern int ext3_group_extend(struct super_block *sb, - struct ext3_super_block *es, - ext3_fsblk_t n_blocks_count); +extern int ext4_group_add(struct super_block *sb, + struct ext4_new_group_data *input); +extern int ext4_group_extend(struct super_block *sb, + struct ext4_super_block *es, + ext4_fsblk_t n_blocks_count); /* super.c */ -extern void ext3_error (struct super_block *, const char *, const char *, ...) +extern void ext4_error (struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void __ext3_std_error (struct super_block *, const char *, int); -extern void ext3_abort (struct super_block *, const char *, const char *, ...) +extern void __ext4_std_error (struct super_block *, const char *, int); +extern void ext4_abort (struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void ext3_warning (struct super_block *, const char *, const char *, ...) +extern void ext4_warning (struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void ext3_update_dynamic_rev (struct super_block *sb); +extern void ext4_update_dynamic_rev (struct super_block *sb); -#define ext3_std_error(sb, errno) \ +#define ext4_std_error(sb, errno) \ do { \ if ((errno)) \ - __ext3_std_error((sb), __FUNCTION__, (errno)); \ + __ext4_std_error((sb), __FUNCTION__, (errno)); \ } while (0) /* @@ -865,21 +865,21 @@ do { \ */ /* dir.c */ -extern const struct file_operations ext3_dir_operations; +extern const struct file_operations ext4_dir_operations; /* file.c */ -extern struct inode_operations ext3_file_inode_operations; -extern const struct file_operations ext3_file_operations; +extern struct inode_operations ext4_file_inode_operations; +extern const struct file_operations ext4_file_operations; /* namei.c */ -extern struct inode_operations ext3_dir_inode_operations; -extern struct inode_operations ext3_special_inode_operations; +extern struct inode_operations ext4_dir_inode_operations; +extern struct inode_operations ext4_special_inode_operations; /* symlink.c */ -extern struct inode_operations ext3_symlink_inode_operations; -extern struct inode_operations ext3_fast_symlink_inode_operations; +extern struct inode_operations ext4_symlink_inode_operations; +extern struct inode_operations ext4_fast_symlink_inode_operations; #endif /* __KERNEL__ */ -#endif /* _LINUX_EXT3_FS_H */ +#endif /* _LINUX_EXT4_FS_H */ diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h index 4395e52..18a6ce9 100644 --- a/include/linux/ext4_fs_i.h +++ b/include/linux/ext4_fs_i.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/ext3_fs_i.h + * linux/include/linux/ext4_fs_i.h * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -13,8 +13,8 @@ * Copyright (C) 1991, 1992 Linus Torvalds */ -#ifndef _LINUX_EXT3_FS_I -#define _LINUX_EXT3_FS_I +#ifndef _LINUX_EXT4_FS_I +#define _LINUX_EXT4_FS_I #include #include @@ -22,43 +22,43 @@ #include /* data type for block offset of block group */ -typedef int ext3_grpblk_t; +typedef int ext4_grpblk_t; /* data type for filesystem-wide blocks number */ -typedef unsigned long ext3_fsblk_t; +typedef unsigned long ext4_fsblk_t; #define E3FSBLK "%lu" -struct ext3_reserve_window { - ext3_fsblk_t _rsv_start; /* First byte reserved */ - ext3_fsblk_t _rsv_end; /* Last byte reserved or 0 */ +struct ext4_reserve_window { + ext4_fsblk_t _rsv_start; /* First byte reserved */ + ext4_fsblk_t _rsv_end; /* Last byte reserved or 0 */ }; -struct ext3_reserve_window_node { +struct ext4_reserve_window_node { struct rb_node rsv_node; __u32 rsv_goal_size; __u32 rsv_alloc_hit; - struct ext3_reserve_window rsv_window; + struct ext4_reserve_window rsv_window; }; -struct ext3_block_alloc_info { +struct ext4_block_alloc_info { /* information about reservation window */ - struct ext3_reserve_window_node rsv_window_node; + struct ext4_reserve_window_node rsv_window_node; /* - * was i_next_alloc_block in ext3_inode_info + * was i_next_alloc_block in ext4_inode_info * is the logical (file-relative) number of the * most-recently-allocated block in this file. * We use this for detecting linearly ascending allocation requests. */ __u32 last_alloc_logical_block; /* - * Was i_next_alloc_goal in ext3_inode_info + * Was i_next_alloc_goal in ext4_inode_info * is the *physical* companion to i_next_alloc_block. * it the the physical block number of the block which was most-recentl * allocated to this file. This give us the goal (target) for the next * allocation when we detect linearly ascending requests. */ - ext3_fsblk_t last_alloc_physical_block; + ext4_fsblk_t last_alloc_physical_block; }; #define rsv_start rsv_window._rsv_start @@ -67,15 +67,15 @@ struct ext3_block_alloc_info { /* * third extended file system inode data in memory */ -struct ext3_inode_info { +struct ext4_inode_info { __le32 i_data[15]; /* unconverted */ __u32 i_flags; -#ifdef EXT3_FRAGMENTS +#ifdef EXT4_FRAGMENTS __u32 i_faddr; __u8 i_frag_no; __u8 i_frag_size; #endif - ext3_fsblk_t i_file_acl; + ext4_fsblk_t i_file_acl; __u32 i_dir_acl; __u32 i_dtime; @@ -87,13 +87,13 @@ struct ext3_inode_info { * near to their parent directory's inode. */ __u32 i_block_group; - __u32 i_state; /* Dynamic state flags for ext3 */ + __u32 i_state; /* Dynamic state flags for ext4 */ /* block reservation info */ - struct ext3_block_alloc_info *i_block_alloc_info; + struct ext4_block_alloc_info *i_block_alloc_info; __u32 i_dir_start_lookup; -#ifdef CONFIG_EXT3_FS_XATTR +#ifdef CONFIG_EXT4DEV_FS_XATTR /* * Extended attributes can be read independently of the main file * data. Taking i_mutex even when reading would cause contention @@ -103,7 +103,7 @@ struct ext3_inode_info { */ struct rw_semaphore xattr_sem; #endif -#ifdef CONFIG_EXT3_FS_POSIX_ACL +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL struct posix_acl *i_acl; struct posix_acl *i_default_acl; #endif @@ -113,7 +113,7 @@ struct ext3_inode_info { /* * i_disksize keeps track of what the inode size is ON DISK, not * in memory. During truncate, i_size is set to the new size by - * the VFS prior to calling ext3_truncate(), but the filesystem won't + * the VFS prior to calling ext4_truncate(), but the filesystem won't * set i_disksize to 0 until the truncate is actually under way. * * The intent is that i_disksize always represents the blocks which @@ -123,7 +123,7 @@ struct ext3_inode_info { * * The only time when i_disksize and i_size may be different is when * a truncate is in progress. The only things which change i_disksize - * are ext3_get_block (growth) and ext3_truncate (shrinkth). + * are ext4_get_block (growth) and ext4_truncate (shrinkth). */ loff_t i_disksize; @@ -131,10 +131,10 @@ struct ext3_inode_info { __u16 i_extra_isize; /* - * truncate_mutex is for serialising ext3_truncate() against - * ext3_getblock(). In the 2.4 ext2 design, great chunks of inode's + * truncate_mutex is for serialising ext4_truncate() against + * ext4_getblock(). In the 2.4 ext2 design, great chunks of inode's * data tree are chopped off during truncate. We can't do that in - * ext3 because whenever we perform intermediate commits during + * ext4 because whenever we perform intermediate commits during * truncate, the inode and all the metadata blocks *must* be in a * consistent state which allows truncation of the orphans to restart * during recovery. Hence we must fix the get_block-vs-truncate race @@ -144,4 +144,4 @@ struct ext3_inode_info { struct inode vfs_inode; }; -#endif /* _LINUX_EXT3_FS_I */ +#endif /* _LINUX_EXT4_FS_I */ diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h index f61309c..ce4856d 100644 --- a/include/linux/ext4_fs_sb.h +++ b/include/linux/ext4_fs_sb.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/ext3_fs_sb.h + * linux/include/linux/ext4_fs_sb.h * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card (card@masi.ibp.fr) @@ -13,8 +13,8 @@ * Copyright (C) 1991, 1992 Linus Torvalds */ -#ifndef _LINUX_EXT3_FS_SB -#define _LINUX_EXT3_FS_SB +#ifndef _LINUX_EXT4_FS_SB +#define _LINUX_EXT4_FS_SB #ifdef __KERNEL__ #include @@ -27,7 +27,7 @@ /* * third extended-fs super-block data in memory */ -struct ext3_sb_info { +struct ext4_sb_info { unsigned long s_frag_size; /* Size of a fragment in bytes */ unsigned long s_frags_per_block;/* Number of fragments per block */ unsigned long s_inodes_per_block;/* Number of inodes per block */ @@ -39,7 +39,7 @@ struct ext3_sb_info { unsigned long s_desc_per_block; /* Number of group descriptors per block */ unsigned long s_groups_count; /* Number of groups in the fs */ struct buffer_head * s_sbh; /* Buffer containing the super block */ - struct ext3_super_block * s_es; /* Pointer to the super block in the buffer */ + struct ext4_super_block * s_es; /* Pointer to the super block in the buffer */ struct buffer_head ** s_group_desc; unsigned long s_mount_opt; uid_t s_resuid; @@ -62,7 +62,7 @@ struct ext3_sb_info { /* root of the per fs reservation window tree */ spinlock_t s_rsv_window_lock; struct rb_root s_rsv_window_root; - struct ext3_reserve_window_node s_rsv_window_head; + struct ext4_reserve_window_node s_rsv_window_head; /* Journaling */ struct inode * s_journal_inode; @@ -80,4 +80,4 @@ struct ext3_sb_info { #endif }; -#endif /* _LINUX_EXT3_FS_SB */ +#endif /* _LINUX_EXT4_FS_SB */ diff --git a/include/linux/ext4_jbd.h b/include/linux/ext4_jbd.h index ce0e610..3dbf6c7 100644 --- a/include/linux/ext4_jbd.h +++ b/include/linux/ext4_jbd.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/ext3_jbd.h + * linux/include/linux/ext4_jbd.h * * Written by Stephen C. Tweedie , 1999 * @@ -9,17 +9,17 @@ * the terms of the GNU General Public License, version 2, or at your * option, any later version, incorporated herein by reference. * - * Ext3-specific journaling extensions. + * Ext4-specific journaling extensions. */ -#ifndef _LINUX_EXT3_JBD_H -#define _LINUX_EXT3_JBD_H +#ifndef _LINUX_EXT4_JBD_H +#define _LINUX_EXT4_JBD_H #include #include -#include +#include -#define EXT3_JOURNAL(inode) (EXT3_SB((inode)->i_sb)->s_journal) +#define EXT4_JOURNAL(inode) (EXT4_SB((inode)->i_sb)->s_journal) /* Define the number of blocks we need to account to a transaction to * modify one block of data. @@ -28,13 +28,13 @@ * indirection blocks, the group and superblock summaries, and the data * block to complete the transaction. */ -#define EXT3_SINGLEDATA_TRANS_BLOCKS 8U +#define EXT4_SINGLEDATA_TRANS_BLOCKS 8U /* Extended attribute operations touch at most two data buffers, * two bitmap buffers, and two group summaries, in addition to the inode * and the superblock, which are already accounted for. */ -#define EXT3_XATTR_TRANS_BLOCKS 6U +#define EXT4_XATTR_TRANS_BLOCKS 6U /* Define the minimum size for a transaction which modifies data. This * needs to take into account the fact that we may end up modifying two @@ -42,15 +42,15 @@ * superblock only gets updated once, of course, so don't bother * counting that again for the quota updates. */ -#define EXT3_DATA_TRANS_BLOCKS(sb) (EXT3_SINGLEDATA_TRANS_BLOCKS + \ - EXT3_XATTR_TRANS_BLOCKS - 2 + \ - 2*EXT3_QUOTA_TRANS_BLOCKS(sb)) +#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS + \ + EXT4_XATTR_TRANS_BLOCKS - 2 + \ + 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) /* Delete operations potentially hit one directory's namespace plus an * entire inode, plus arbitrary amounts of bitmap/indirection data. Be * generous. We can grow the delete transaction later if necessary. */ -#define EXT3_DELETE_TRANS_BLOCKS(sb) (2 * EXT3_DATA_TRANS_BLOCKS(sb) + 64) +#define EXT4_DELETE_TRANS_BLOCKS(sb) (2 * EXT4_DATA_TRANS_BLOCKS(sb) + 64) /* Define an arbitrary limit for the amount of data we will anticipate * writing to any given transaction. For unbounded transactions such as @@ -58,7 +58,7 @@ * start off at the maximum transaction size and grow the transaction * optimistically as we go. */ -#define EXT3_MAX_TRANS_DATA 64U +#define EXT4_MAX_TRANS_DATA 64U /* We break up a large truncate or write transaction once the handle's * buffer credits gets this low, we need either to extend the @@ -67,202 +67,202 @@ * one block, plus two quota updates. Quota allocations are not * needed. */ -#define EXT3_RESERVE_TRANS_BLOCKS 12U +#define EXT4_RESERVE_TRANS_BLOCKS 12U -#define EXT3_INDEX_EXTRA_TRANS_BLOCKS 8 +#define EXT4_INDEX_EXTRA_TRANS_BLOCKS 8 #ifdef CONFIG_QUOTA /* Amount of blocks needed for quota update - we know that the structure was * allocated so we need to update only inode+data */ -#define EXT3_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) +#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) /* Amount of blocks needed for quota insert/delete - we do some block writes * but inode, sb and group updates are done only once */ -#define EXT3_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ - (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) -#define EXT3_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ - (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) +#define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ + (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) +#define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ + (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) #else -#define EXT3_QUOTA_TRANS_BLOCKS(sb) 0 -#define EXT3_QUOTA_INIT_BLOCKS(sb) 0 -#define EXT3_QUOTA_DEL_BLOCKS(sb) 0 +#define EXT4_QUOTA_TRANS_BLOCKS(sb) 0 +#define EXT4_QUOTA_INIT_BLOCKS(sb) 0 +#define EXT4_QUOTA_DEL_BLOCKS(sb) 0 #endif int -ext3_mark_iloc_dirty(handle_t *handle, +ext4_mark_iloc_dirty(handle_t *handle, struct inode *inode, - struct ext3_iloc *iloc); + struct ext4_iloc *iloc); /* * On success, We end up with an outstanding reference count against * iloc->bh. This _must_ be cleaned up later. */ -int ext3_reserve_inode_write(handle_t *handle, struct inode *inode, - struct ext3_iloc *iloc); +int ext4_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext4_iloc *iloc); -int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode); +int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode); /* - * Wrapper functions with which ext3 calls into JBD. The intent here is - * to allow these to be turned into appropriate stubs so ext3 can control - * ext2 filesystems, so ext2+ext3 systems only nee one fs. This work hasn't + * Wrapper functions with which ext4 calls into JBD. The intent here is + * to allow these to be turned into appropriate stubs so ext4 can control + * ext2 filesystems, so ext2+ext4 systems only nee one fs. This work hasn't * been done yet. */ -void ext3_journal_abort_handle(const char *caller, const char *err_fn, +void ext4_journal_abort_handle(const char *caller, const char *err_fn, struct buffer_head *bh, handle_t *handle, int err); static inline int -__ext3_journal_get_undo_access(const char *where, handle_t *handle, +__ext4_journal_get_undo_access(const char *where, handle_t *handle, struct buffer_head *bh) { int err = journal_get_undo_access(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; } static inline int -__ext3_journal_get_write_access(const char *where, handle_t *handle, +__ext4_journal_get_write_access(const char *where, handle_t *handle, struct buffer_head *bh) { int err = journal_get_write_access(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; } static inline void -ext3_journal_release_buffer(handle_t *handle, struct buffer_head *bh) +ext4_journal_release_buffer(handle_t *handle, struct buffer_head *bh) { journal_release_buffer(handle, bh); } static inline int -__ext3_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) +__ext4_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) { int err = journal_forget(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; } static inline int -__ext3_journal_revoke(const char *where, handle_t *handle, +__ext4_journal_revoke(const char *where, handle_t *handle, unsigned long blocknr, struct buffer_head *bh) { int err = journal_revoke(handle, blocknr, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; } static inline int -__ext3_journal_get_create_access(const char *where, +__ext4_journal_get_create_access(const char *where, handle_t *handle, struct buffer_head *bh) { int err = journal_get_create_access(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; } static inline int -__ext3_journal_dirty_metadata(const char *where, +__ext4_journal_dirty_metadata(const char *where, handle_t *handle, struct buffer_head *bh) { int err = journal_dirty_metadata(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; } -#define ext3_journal_get_undo_access(handle, bh) \ - __ext3_journal_get_undo_access(__FUNCTION__, (handle), (bh)) -#define ext3_journal_get_write_access(handle, bh) \ - __ext3_journal_get_write_access(__FUNCTION__, (handle), (bh)) -#define ext3_journal_revoke(handle, blocknr, bh) \ - __ext3_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) -#define ext3_journal_get_create_access(handle, bh) \ - __ext3_journal_get_create_access(__FUNCTION__, (handle), (bh)) -#define ext3_journal_dirty_metadata(handle, bh) \ - __ext3_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) -#define ext3_journal_forget(handle, bh) \ - __ext3_journal_forget(__FUNCTION__, (handle), (bh)) +#define ext4_journal_get_undo_access(handle, bh) \ + __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_get_write_access(handle, bh) \ + __ext4_journal_get_write_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_revoke(handle, blocknr, bh) \ + __ext4_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) +#define ext4_journal_get_create_access(handle, bh) \ + __ext4_journal_get_create_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_dirty_metadata(handle, bh) \ + __ext4_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) +#define ext4_journal_forget(handle, bh) \ + __ext4_journal_forget(__FUNCTION__, (handle), (bh)) -int ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh); +int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh); -handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks); -int __ext3_journal_stop(const char *where, handle_t *handle); +handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); +int __ext4_journal_stop(const char *where, handle_t *handle); -static inline handle_t *ext3_journal_start(struct inode *inode, int nblocks) +static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks) { - return ext3_journal_start_sb(inode->i_sb, nblocks); + return ext4_journal_start_sb(inode->i_sb, nblocks); } -#define ext3_journal_stop(handle) \ - __ext3_journal_stop(__FUNCTION__, (handle)) +#define ext4_journal_stop(handle) \ + __ext4_journal_stop(__FUNCTION__, (handle)) -static inline handle_t *ext3_journal_current_handle(void) +static inline handle_t *ext4_journal_current_handle(void) { return journal_current_handle(); } -static inline int ext3_journal_extend(handle_t *handle, int nblocks) +static inline int ext4_journal_extend(handle_t *handle, int nblocks) { return journal_extend(handle, nblocks); } -static inline int ext3_journal_restart(handle_t *handle, int nblocks) +static inline int ext4_journal_restart(handle_t *handle, int nblocks) { return journal_restart(handle, nblocks); } -static inline int ext3_journal_blocks_per_page(struct inode *inode) +static inline int ext4_journal_blocks_per_page(struct inode *inode) { return journal_blocks_per_page(inode); } -static inline int ext3_journal_force_commit(journal_t *journal) +static inline int ext4_journal_force_commit(journal_t *journal) { return journal_force_commit(journal); } /* super.c */ -int ext3_force_commit(struct super_block *sb); +int ext4_force_commit(struct super_block *sb); -static inline int ext3_should_journal_data(struct inode *inode) +static inline int ext4_should_journal_data(struct inode *inode) { if (!S_ISREG(inode->i_mode)) return 1; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA) + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) return 1; - if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL) + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) return 1; return 0; } -static inline int ext3_should_order_data(struct inode *inode) +static inline int ext4_should_order_data(struct inode *inode) { if (!S_ISREG(inode->i_mode)) return 0; - if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL) + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) return 0; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA) + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) return 1; return 0; } -static inline int ext3_should_writeback_data(struct inode *inode) +static inline int ext4_should_writeback_data(struct inode *inode) { if (!S_ISREG(inode->i_mode)) return 0; - if (EXT3_I(inode)->i_flags & EXT3_JOURNAL_DATA_FL) + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) return 0; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA) + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) return 1; return 0; } -#endif /* _LINUX_EXT3_JBD_H */ +#endif /* _LINUX_EXT4_JBD_H */ -- cgit v0.10.2 From 02ea2104c55b625cf5b5d9ba8586a4fc17920f5c Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:20:56 -0700 Subject: [PATCH] ext4: enable building of ext4 Originally part of a patch from Mingming Cao and Randy Dunlap. Reorganized by Shaggy. Signed-off-by: Randy Dunlap Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/Kconfig b/fs/Kconfig index 599de54..ac9ba1c 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -140,6 +140,73 @@ config EXT3_FS_SECURITY If you are not using a security module that requires using extended attributes for file security labels, say N. +config EXT4DEV_FS + tristate "Ext4dev/ext4 extended fs support development (EXPERIMENTAL)" + depends on EXPERIMENTAL + select JBD + help + Ext4dev is a predecessor filesystem of the next generation + extended fs ext4, based on ext3 filesystem code. It will be + renamed ext4 fs later, once ext4dev is mature and stabilized. + + Unlike the change from ext2 filesystem to ext3 filesystem, + the on-disk format of ext4dev is not the same as ext3 any more: + it is based on extent maps and it supports 48-bit physical block + numbers. These combined on-disk format changes will allow + ext4dev/ext4 to handle more than 16 TB filesystem volumes -- + a hard limit that ext3 cannot overcome without changing the + on-disk format. + + Other than extent maps and 48-bit block numbers, ext4dev also is + likely to have other new features such as persistent preallocation, + high resolution time stamps, and larger file support etc. These + features will be added to ext4dev gradually. + + To compile this file system support as a module, choose M here. The + module will be called ext4dev. Be aware, however, that the filesystem + of your root partition (the one containing the directory /) cannot + be compiled as a module, and so this could be dangerous. + + If unsure, say N. + +config EXT4DEV_FS_XATTR + bool "Ext4dev extended attributes" + depends on EXT4DEV_FS + default y + help + Extended attributes are name:value pairs associated with inodes by + the kernel or by users (see the attr(5) manual page, or visit + for details). + + If unsure, say N. + + You need this for POSIX ACL support on ext4dev/ext4. + +config EXT4DEV_FS_POSIX_ACL + bool "Ext4dev POSIX Access Control Lists" + depends on EXT4DEV_FS_XATTR + select FS_POSIX_ACL + help + POSIX Access Control Lists (ACLs) support permissions for users and + groups beyond the owner/group/world scheme. + + To learn more about Access Control Lists, visit the POSIX ACLs for + Linux website . + + If you don't know what Access Control Lists are, say N + +config EXT4DEV_FS_SECURITY + bool "Ext4dev Security Labels" + depends on EXT4DEV_FS_XATTR + help + Security labels support alternative access control models + implemented by security modules like SELinux. This option + enables an extended attribute handler for file security + labels in the ext4dev/ext4 filesystem. + + If you are not using a security module that requires using + extended attributes for file security labels, say N. + config JBD tristate help @@ -173,11 +240,11 @@ config JBD_DEBUG "echo 0 > /proc/sys/fs/jbd-debug". config FS_MBCACHE -# Meta block cache for Extended Attributes (ext2/ext3) +# Meta block cache for Extended Attributes (ext2/ext3/ext4) tristate - depends on EXT2_FS_XATTR || EXT3_FS_XATTR - default y if EXT2_FS=y || EXT3_FS=y - default m if EXT2_FS=m || EXT3_FS=m + depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4DEV_FS_XATTR + default y if EXT2_FS=y || EXT3_FS=y || EXT4DEV_FS=y + default m if EXT2_FS=m || EXT3_FS=m || EXT4DEV_FS=m config REISERFS_FS tristate "Reiserfs support" diff --git a/fs/Makefile b/fs/Makefile index df614ea..64396af 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -62,6 +62,7 @@ obj-$(CONFIG_DLM) += dlm/ # Do not add any filesystems before this line obj-$(CONFIG_REISERFS_FS) += reiserfs/ obj-$(CONFIG_EXT3_FS) += ext3/ # Before ext2 so root fs can be ext3 +obj-$(CONFIG_EXT4DEV_FS) += ext4/ # Before ext2 so root fs can be ext4dev obj-$(CONFIG_JBD) += jbd/ obj-$(CONFIG_EXT2_FS) += ext2/ obj-$(CONFIG_CRAMFS) += cramfs/ diff --git a/include/linux/magic.h b/include/linux/magic.h index 22036dd..156c40f 100644 --- a/include/linux/magic.h +++ b/include/linux/magic.h @@ -8,6 +8,7 @@ #define EFS_SUPER_MAGIC 0x414A53 #define EXT2_SUPER_MAGIC 0xEF53 #define EXT3_SUPER_MAGIC 0xEF53 +#define EXT4_SUPER_MAGIC 0xEF53 #define HPFS_SUPER_MAGIC 0xf995e849 #define ISOFS_SUPER_MAGIC 0x9660 #define JFFS2_SUPER_MAGIC 0x72b6 -- cgit v0.10.2 From 470decc613ab2048b619a01028072d932d9086ee Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Wed, 11 Oct 2006 01:20:57 -0700 Subject: [PATCH] jbd2: initial copy of files from jbd This is a simple copy of the files in fs/jbd to fs/jbd2 and /usr/incude/linux/[ext4_]jbd.h to /usr/include/[ext4_]jbd2.h Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/Makefile b/fs/jbd2/Makefile new file mode 100644 index 0000000..54aca48 --- /dev/null +++ b/fs/jbd2/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the linux journaling routines. +# + +obj-$(CONFIG_JBD) += jbd.o + +jbd-objs := transaction.o commit.o recovery.o checkpoint.o revoke.o journal.o diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c new file mode 100644 index 0000000..0208cc7 --- /dev/null +++ b/fs/jbd2/checkpoint.c @@ -0,0 +1,697 @@ +/* + * linux/fs/checkpoint.c + * + * Written by Stephen C. Tweedie , 1999 + * + * Copyright 1999 Red Hat Software --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Checkpoint routines for the generic filesystem journaling code. + * Part of the ext2fs journaling system. + * + * Checkpointing is the process of ensuring that a section of the log is + * committed fully to disk, so that that portion of the log can be + * reused. + */ + +#include +#include +#include +#include +#include + +/* + * Unlink a buffer from a transaction checkpoint list. + * + * Called with j_list_lock held. + */ +static inline void __buffer_unlink_first(struct journal_head *jh) +{ + transaction_t *transaction = jh->b_cp_transaction; + + jh->b_cpnext->b_cpprev = jh->b_cpprev; + jh->b_cpprev->b_cpnext = jh->b_cpnext; + if (transaction->t_checkpoint_list == jh) { + transaction->t_checkpoint_list = jh->b_cpnext; + if (transaction->t_checkpoint_list == jh) + transaction->t_checkpoint_list = NULL; + } +} + +/* + * Unlink a buffer from a transaction checkpoint(io) list. + * + * Called with j_list_lock held. + */ +static inline void __buffer_unlink(struct journal_head *jh) +{ + transaction_t *transaction = jh->b_cp_transaction; + + __buffer_unlink_first(jh); + if (transaction->t_checkpoint_io_list == jh) { + transaction->t_checkpoint_io_list = jh->b_cpnext; + if (transaction->t_checkpoint_io_list == jh) + transaction->t_checkpoint_io_list = NULL; + } +} + +/* + * Move a buffer from the checkpoint list to the checkpoint io list + * + * Called with j_list_lock held + */ +static inline void __buffer_relink_io(struct journal_head *jh) +{ + transaction_t *transaction = jh->b_cp_transaction; + + __buffer_unlink_first(jh); + + if (!transaction->t_checkpoint_io_list) { + jh->b_cpnext = jh->b_cpprev = jh; + } else { + jh->b_cpnext = transaction->t_checkpoint_io_list; + jh->b_cpprev = transaction->t_checkpoint_io_list->b_cpprev; + jh->b_cpprev->b_cpnext = jh; + jh->b_cpnext->b_cpprev = jh; + } + transaction->t_checkpoint_io_list = jh; +} + +/* + * Try to release a checkpointed buffer from its transaction. + * Returns 1 if we released it and 2 if we also released the + * whole transaction. + * + * Requires j_list_lock + * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it + */ +static int __try_to_free_cp_buf(struct journal_head *jh) +{ + int ret = 0; + struct buffer_head *bh = jh2bh(jh); + + if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) { + JBUFFER_TRACE(jh, "remove from checkpoint list"); + ret = __journal_remove_checkpoint(jh) + 1; + jbd_unlock_bh_state(bh); + journal_remove_journal_head(bh); + BUFFER_TRACE(bh, "release"); + __brelse(bh); + } else { + jbd_unlock_bh_state(bh); + } + return ret; +} + +/* + * __log_wait_for_space: wait until there is space in the journal. + * + * Called under j-state_lock *only*. It will be unlocked if we have to wait + * for a checkpoint to free up some space in the log. + */ +void __log_wait_for_space(journal_t *journal) +{ + int nblocks; + assert_spin_locked(&journal->j_state_lock); + + nblocks = jbd_space_needed(journal); + while (__log_space_left(journal) < nblocks) { + if (journal->j_flags & JFS_ABORT) + return; + spin_unlock(&journal->j_state_lock); + mutex_lock(&journal->j_checkpoint_mutex); + + /* + * Test again, another process may have checkpointed while we + * were waiting for the checkpoint lock + */ + spin_lock(&journal->j_state_lock); + nblocks = jbd_space_needed(journal); + if (__log_space_left(journal) < nblocks) { + spin_unlock(&journal->j_state_lock); + log_do_checkpoint(journal); + spin_lock(&journal->j_state_lock); + } + mutex_unlock(&journal->j_checkpoint_mutex); + } +} + +/* + * We were unable to perform jbd_trylock_bh_state() inside j_list_lock. + * The caller must restart a list walk. Wait for someone else to run + * jbd_unlock_bh_state(). + */ +static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh) + __releases(journal->j_list_lock) +{ + get_bh(bh); + spin_unlock(&journal->j_list_lock); + jbd_lock_bh_state(bh); + jbd_unlock_bh_state(bh); + put_bh(bh); +} + +/* + * Clean up transaction's list of buffers submitted for io. + * We wait for any pending IO to complete and remove any clean + * buffers. Note that we take the buffers in the opposite ordering + * from the one in which they were submitted for IO. + * + * Called with j_list_lock held. + */ +static void __wait_cp_io(journal_t *journal, transaction_t *transaction) +{ + struct journal_head *jh; + struct buffer_head *bh; + tid_t this_tid; + int released = 0; + + this_tid = transaction->t_tid; +restart: + /* Did somebody clean up the transaction in the meanwhile? */ + if (journal->j_checkpoint_transactions != transaction || + transaction->t_tid != this_tid) + return; + while (!released && transaction->t_checkpoint_io_list) { + jh = transaction->t_checkpoint_io_list; + bh = jh2bh(jh); + if (!jbd_trylock_bh_state(bh)) { + jbd_sync_bh(journal, bh); + spin_lock(&journal->j_list_lock); + goto restart; + } + if (buffer_locked(bh)) { + atomic_inc(&bh->b_count); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + wait_on_buffer(bh); + /* the journal_head may have gone by now */ + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + spin_lock(&journal->j_list_lock); + goto restart; + } + /* + * Now in whatever state the buffer currently is, we know that + * it has been written out and so we can drop it from the list + */ + released = __journal_remove_checkpoint(jh); + jbd_unlock_bh_state(bh); + journal_remove_journal_head(bh); + __brelse(bh); + } +} + +#define NR_BATCH 64 + +static void +__flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) +{ + int i; + + ll_rw_block(SWRITE, *batch_count, bhs); + for (i = 0; i < *batch_count; i++) { + struct buffer_head *bh = bhs[i]; + clear_buffer_jwrite(bh); + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + } + *batch_count = 0; +} + +/* + * Try to flush one buffer from the checkpoint list to disk. + * + * Return 1 if something happened which requires us to abort the current + * scan of the checkpoint list. + * + * Called with j_list_lock held and drops it if 1 is returned + * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it + */ +static int __process_buffer(journal_t *journal, struct journal_head *jh, + struct buffer_head **bhs, int *batch_count) +{ + struct buffer_head *bh = jh2bh(jh); + int ret = 0; + + if (buffer_locked(bh)) { + atomic_inc(&bh->b_count); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + wait_on_buffer(bh); + /* the journal_head may have gone by now */ + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + ret = 1; + } else if (jh->b_transaction != NULL) { + transaction_t *t = jh->b_transaction; + tid_t tid = t->t_tid; + + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + log_start_commit(journal, tid); + log_wait_commit(journal, tid); + ret = 1; + } else if (!buffer_dirty(bh)) { + J_ASSERT_JH(jh, !buffer_jbddirty(bh)); + BUFFER_TRACE(bh, "remove from checkpoint"); + __journal_remove_checkpoint(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + journal_remove_journal_head(bh); + __brelse(bh); + ret = 1; + } else { + /* + * Important: we are about to write the buffer, and + * possibly block, while still holding the journal lock. + * We cannot afford to let the transaction logic start + * messing around with this buffer before we write it to + * disk, as that would break recoverability. + */ + BUFFER_TRACE(bh, "queue"); + get_bh(bh); + J_ASSERT_BH(bh, !buffer_jwrite(bh)); + set_buffer_jwrite(bh); + bhs[*batch_count] = bh; + __buffer_relink_io(jh); + jbd_unlock_bh_state(bh); + (*batch_count)++; + if (*batch_count == NR_BATCH) { + spin_unlock(&journal->j_list_lock); + __flush_batch(journal, bhs, batch_count); + ret = 1; + } + } + return ret; +} + +/* + * Perform an actual checkpoint. We take the first transaction on the + * list of transactions to be checkpointed and send all its buffers + * to disk. We submit larger chunks of data at once. + * + * The journal should be locked before calling this function. + */ +int log_do_checkpoint(journal_t *journal) +{ + transaction_t *transaction; + tid_t this_tid; + int result; + + jbd_debug(1, "Start checkpoint\n"); + + /* + * First thing: if there are any transactions in the log which + * don't need checkpointing, just eliminate them from the + * journal straight away. + */ + result = cleanup_journal_tail(journal); + jbd_debug(1, "cleanup_journal_tail returned %d\n", result); + if (result <= 0) + return result; + + /* + * OK, we need to start writing disk blocks. Take one transaction + * and write it. + */ + spin_lock(&journal->j_list_lock); + if (!journal->j_checkpoint_transactions) + goto out; + transaction = journal->j_checkpoint_transactions; + this_tid = transaction->t_tid; +restart: + /* + * If someone cleaned up this transaction while we slept, we're + * done (maybe it's a new transaction, but it fell at the same + * address). + */ + if (journal->j_checkpoint_transactions == transaction && + transaction->t_tid == this_tid) { + int batch_count = 0; + struct buffer_head *bhs[NR_BATCH]; + struct journal_head *jh; + int retry = 0; + + while (!retry && transaction->t_checkpoint_list) { + struct buffer_head *bh; + + jh = transaction->t_checkpoint_list; + bh = jh2bh(jh); + if (!jbd_trylock_bh_state(bh)) { + jbd_sync_bh(journal, bh); + retry = 1; + break; + } + retry = __process_buffer(journal, jh, bhs,&batch_count); + if (!retry && lock_need_resched(&journal->j_list_lock)){ + spin_unlock(&journal->j_list_lock); + retry = 1; + break; + } + } + + if (batch_count) { + if (!retry) { + spin_unlock(&journal->j_list_lock); + retry = 1; + } + __flush_batch(journal, bhs, &batch_count); + } + + if (retry) { + spin_lock(&journal->j_list_lock); + goto restart; + } + /* + * Now we have cleaned up the first transaction's checkpoint + * list. Let's clean up the second one + */ + __wait_cp_io(journal, transaction); + } +out: + spin_unlock(&journal->j_list_lock); + result = cleanup_journal_tail(journal); + if (result < 0) + return result; + return 0; +} + +/* + * Check the list of checkpoint transactions for the journal to see if + * we have already got rid of any since the last update of the log tail + * in the journal superblock. If so, we can instantly roll the + * superblock forward to remove those transactions from the log. + * + * Return <0 on error, 0 on success, 1 if there was nothing to clean up. + * + * Called with the journal lock held. + * + * This is the only part of the journaling code which really needs to be + * aware of transaction aborts. Checkpointing involves writing to the + * main filesystem area rather than to the journal, so it can proceed + * even in abort state, but we must not update the journal superblock if + * we have an abort error outstanding. + */ + +int cleanup_journal_tail(journal_t *journal) +{ + transaction_t * transaction; + tid_t first_tid; + unsigned long blocknr, freed; + + /* OK, work out the oldest transaction remaining in the log, and + * the log block it starts at. + * + * If the log is now empty, we need to work out which is the + * next transaction ID we will write, and where it will + * start. */ + + spin_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); + transaction = journal->j_checkpoint_transactions; + if (transaction) { + first_tid = transaction->t_tid; + blocknr = transaction->t_log_start; + } else if ((transaction = journal->j_committing_transaction) != NULL) { + first_tid = transaction->t_tid; + blocknr = transaction->t_log_start; + } else if ((transaction = journal->j_running_transaction) != NULL) { + first_tid = transaction->t_tid; + blocknr = journal->j_head; + } else { + first_tid = journal->j_transaction_sequence; + blocknr = journal->j_head; + } + spin_unlock(&journal->j_list_lock); + J_ASSERT(blocknr != 0); + + /* If the oldest pinned transaction is at the tail of the log + already then there's not much we can do right now. */ + if (journal->j_tail_sequence == first_tid) { + spin_unlock(&journal->j_state_lock); + return 1; + } + + /* OK, update the superblock to recover the freed space. + * Physical blocks come first: have we wrapped beyond the end of + * the log? */ + freed = blocknr - journal->j_tail; + if (blocknr < journal->j_tail) + freed = freed + journal->j_last - journal->j_first; + + jbd_debug(1, + "Cleaning journal tail from %d to %d (offset %lu), " + "freeing %lu\n", + journal->j_tail_sequence, first_tid, blocknr, freed); + + journal->j_free += freed; + journal->j_tail_sequence = first_tid; + journal->j_tail = blocknr; + spin_unlock(&journal->j_state_lock); + if (!(journal->j_flags & JFS_ABORT)) + journal_update_superblock(journal, 1); + return 0; +} + + +/* Checkpoint list management */ + +/* + * journal_clean_one_cp_list + * + * Find all the written-back checkpoint buffers in the given list and release them. + * + * Called with the journal locked. + * Called with j_list_lock held. + * Returns number of bufers reaped (for debug) + */ + +static int journal_clean_one_cp_list(struct journal_head *jh, int *released) +{ + struct journal_head *last_jh; + struct journal_head *next_jh = jh; + int ret, freed = 0; + + *released = 0; + if (!jh) + return 0; + + last_jh = jh->b_cpprev; + do { + jh = next_jh; + next_jh = jh->b_cpnext; + /* Use trylock because of the ranking */ + if (jbd_trylock_bh_state(jh2bh(jh))) { + ret = __try_to_free_cp_buf(jh); + if (ret) { + freed++; + if (ret == 2) { + *released = 1; + return freed; + } + } + } + /* + * This function only frees up some memory + * if possible so we dont have an obligation + * to finish processing. Bail out if preemption + * requested: + */ + if (need_resched()) + return freed; + } while (jh != last_jh); + + return freed; +} + +/* + * journal_clean_checkpoint_list + * + * Find all the written-back checkpoint buffers in the journal and release them. + * + * Called with the journal locked. + * Called with j_list_lock held. + * Returns number of buffers reaped (for debug) + */ + +int __journal_clean_checkpoint_list(journal_t *journal) +{ + transaction_t *transaction, *last_transaction, *next_transaction; + int ret = 0; + int released; + + transaction = journal->j_checkpoint_transactions; + if (!transaction) + goto out; + + last_transaction = transaction->t_cpprev; + next_transaction = transaction; + do { + transaction = next_transaction; + next_transaction = transaction->t_cpnext; + ret += journal_clean_one_cp_list(transaction-> + t_checkpoint_list, &released); + /* + * This function only frees up some memory if possible so we + * dont have an obligation to finish processing. Bail out if + * preemption requested: + */ + if (need_resched()) + goto out; + if (released) + continue; + /* + * It is essential that we are as careful as in the case of + * t_checkpoint_list with removing the buffer from the list as + * we can possibly see not yet submitted buffers on io_list + */ + ret += journal_clean_one_cp_list(transaction-> + t_checkpoint_io_list, &released); + if (need_resched()) + goto out; + } while (transaction != last_transaction); +out: + return ret; +} + +/* + * journal_remove_checkpoint: called after a buffer has been committed + * to disk (either by being write-back flushed to disk, or being + * committed to the log). + * + * We cannot safely clean a transaction out of the log until all of the + * buffer updates committed in that transaction have safely been stored + * elsewhere on disk. To achieve this, all of the buffers in a + * transaction need to be maintained on the transaction's checkpoint + * lists until they have been rewritten, at which point this function is + * called to remove the buffer from the existing transaction's + * checkpoint lists. + * + * The function returns 1 if it frees the transaction, 0 otherwise. + * + * This function is called with the journal locked. + * This function is called with j_list_lock held. + * This function is called with jbd_lock_bh_state(jh2bh(jh)) + */ + +int __journal_remove_checkpoint(struct journal_head *jh) +{ + transaction_t *transaction; + journal_t *journal; + int ret = 0; + + JBUFFER_TRACE(jh, "entry"); + + if ((transaction = jh->b_cp_transaction) == NULL) { + JBUFFER_TRACE(jh, "not on transaction"); + goto out; + } + journal = transaction->t_journal; + + __buffer_unlink(jh); + jh->b_cp_transaction = NULL; + + if (transaction->t_checkpoint_list != NULL || + transaction->t_checkpoint_io_list != NULL) + goto out; + JBUFFER_TRACE(jh, "transaction has no more buffers"); + + /* + * There is one special case to worry about: if we have just pulled the + * buffer off a committing transaction's forget list, then even if the + * checkpoint list is empty, the transaction obviously cannot be + * dropped! + * + * The locking here around j_committing_transaction is a bit sleazy. + * See the comment at the end of journal_commit_transaction(). + */ + if (transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "belongs to committing transaction"); + goto out; + } + + /* OK, that was the last buffer for the transaction: we can now + safely remove this transaction from the log */ + + __journal_drop_transaction(journal, transaction); + + /* Just in case anybody was waiting for more transactions to be + checkpointed... */ + wake_up(&journal->j_wait_logspace); + ret = 1; +out: + JBUFFER_TRACE(jh, "exit"); + return ret; +} + +/* + * journal_insert_checkpoint: put a committed buffer onto a checkpoint + * list so that we know when it is safe to clean the transaction out of + * the log. + * + * Called with the journal locked. + * Called with j_list_lock held. + */ +void __journal_insert_checkpoint(struct journal_head *jh, + transaction_t *transaction) +{ + JBUFFER_TRACE(jh, "entry"); + J_ASSERT_JH(jh, buffer_dirty(jh2bh(jh)) || buffer_jbddirty(jh2bh(jh))); + J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + + jh->b_cp_transaction = transaction; + + if (!transaction->t_checkpoint_list) { + jh->b_cpnext = jh->b_cpprev = jh; + } else { + jh->b_cpnext = transaction->t_checkpoint_list; + jh->b_cpprev = transaction->t_checkpoint_list->b_cpprev; + jh->b_cpprev->b_cpnext = jh; + jh->b_cpnext->b_cpprev = jh; + } + transaction->t_checkpoint_list = jh; +} + +/* + * We've finished with this transaction structure: adios... + * + * The transaction must have no links except for the checkpoint by this + * point. + * + * Called with the journal locked. + * Called with j_list_lock held. + */ + +void __journal_drop_transaction(journal_t *journal, transaction_t *transaction) +{ + assert_spin_locked(&journal->j_list_lock); + if (transaction->t_cpnext) { + transaction->t_cpnext->t_cpprev = transaction->t_cpprev; + transaction->t_cpprev->t_cpnext = transaction->t_cpnext; + if (journal->j_checkpoint_transactions == transaction) + journal->j_checkpoint_transactions = + transaction->t_cpnext; + if (journal->j_checkpoint_transactions == transaction) + journal->j_checkpoint_transactions = NULL; + } + + J_ASSERT(transaction->t_state == T_FINISHED); + J_ASSERT(transaction->t_buffers == NULL); + J_ASSERT(transaction->t_sync_datalist == NULL); + J_ASSERT(transaction->t_forget == NULL); + J_ASSERT(transaction->t_iobuf_list == NULL); + J_ASSERT(transaction->t_shadow_list == NULL); + J_ASSERT(transaction->t_log_list == NULL); + J_ASSERT(transaction->t_checkpoint_list == NULL); + J_ASSERT(transaction->t_checkpoint_io_list == NULL); + J_ASSERT(transaction->t_updates == 0); + J_ASSERT(journal->j_committing_transaction != transaction); + J_ASSERT(journal->j_running_transaction != transaction); + + jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid); + kfree(transaction); +} diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c new file mode 100644 index 0000000..10be512 --- /dev/null +++ b/fs/jbd2/commit.c @@ -0,0 +1,911 @@ +/* + * linux/fs/jbd/commit.c + * + * Written by Stephen C. Tweedie , 1998 + * + * Copyright 1998 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Journal commit routines for the generic filesystem journaling code; + * part of the ext2fs journaling system. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Default IO end handler for temporary BJ_IO buffer_heads. + */ +static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate) +{ + BUFFER_TRACE(bh, ""); + if (uptodate) + set_buffer_uptodate(bh); + else + clear_buffer_uptodate(bh); + unlock_buffer(bh); +} + +/* + * When an ext3-ordered file is truncated, it is possible that many pages are + * not sucessfully freed, because they are attached to a committing transaction. + * After the transaction commits, these pages are left on the LRU, with no + * ->mapping, and with attached buffers. These pages are trivially reclaimable + * by the VM, but their apparent absence upsets the VM accounting, and it makes + * the numbers in /proc/meminfo look odd. + * + * So here, we have a buffer which has just come off the forget list. Look to + * see if we can strip all buffers from the backing page. + * + * Called under lock_journal(), and possibly under journal_datalist_lock. The + * caller provided us with a ref against the buffer, and we drop that here. + */ +static void release_buffer_page(struct buffer_head *bh) +{ + struct page *page; + + if (buffer_dirty(bh)) + goto nope; + if (atomic_read(&bh->b_count) != 1) + goto nope; + page = bh->b_page; + if (!page) + goto nope; + if (page->mapping) + goto nope; + + /* OK, it's a truncated page */ + if (TestSetPageLocked(page)) + goto nope; + + page_cache_get(page); + __brelse(bh); + try_to_free_buffers(page); + unlock_page(page); + page_cache_release(page); + return; + +nope: + __brelse(bh); +} + +/* + * Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is + * held. For ranking reasons we must trylock. If we lose, schedule away and + * return 0. j_list_lock is dropped in this case. + */ +static int inverted_lock(journal_t *journal, struct buffer_head *bh) +{ + if (!jbd_trylock_bh_state(bh)) { + spin_unlock(&journal->j_list_lock); + schedule(); + return 0; + } + return 1; +} + +/* Done it all: now write the commit record. We should have + * cleaned up our previous buffers by now, so if we are in abort + * mode we can now just skip the rest of the journal write + * entirely. + * + * Returns 1 if the journal needs to be aborted or 0 on success + */ +static int journal_write_commit_record(journal_t *journal, + transaction_t *commit_transaction) +{ + struct journal_head *descriptor; + struct buffer_head *bh; + int i, ret; + int barrier_done = 0; + + if (is_journal_aborted(journal)) + return 0; + + descriptor = journal_get_descriptor_buffer(journal); + if (!descriptor) + return 1; + + bh = jh2bh(descriptor); + + /* AKPM: buglet - add `i' to tmp! */ + for (i = 0; i < bh->b_size; i += 512) { + journal_header_t *tmp = (journal_header_t*)bh->b_data; + tmp->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); + tmp->h_blocktype = cpu_to_be32(JFS_COMMIT_BLOCK); + tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + } + + JBUFFER_TRACE(descriptor, "write commit block"); + set_buffer_dirty(bh); + if (journal->j_flags & JFS_BARRIER) { + set_buffer_ordered(bh); + barrier_done = 1; + } + ret = sync_dirty_buffer(bh); + /* is it possible for another commit to fail at roughly + * the same time as this one? If so, we don't want to + * trust the barrier flag in the super, but instead want + * to remember if we sent a barrier request + */ + if (ret == -EOPNOTSUPP && barrier_done) { + char b[BDEVNAME_SIZE]; + + printk(KERN_WARNING + "JBD: barrier-based sync failed on %s - " + "disabling barriers\n", + bdevname(journal->j_dev, b)); + spin_lock(&journal->j_state_lock); + journal->j_flags &= ~JFS_BARRIER; + spin_unlock(&journal->j_state_lock); + + /* And try again, without the barrier */ + clear_buffer_ordered(bh); + set_buffer_uptodate(bh); + set_buffer_dirty(bh); + ret = sync_dirty_buffer(bh); + } + put_bh(bh); /* One for getblk() */ + journal_put_journal_head(descriptor); + + return (ret == -EIO); +} + +static void journal_do_submit_data(struct buffer_head **wbuf, int bufs) +{ + int i; + + for (i = 0; i < bufs; i++) { + wbuf[i]->b_end_io = end_buffer_write_sync; + /* We use-up our safety reference in submit_bh() */ + submit_bh(WRITE, wbuf[i]); + } +} + +/* + * Submit all the data buffers to disk + */ +static void journal_submit_data_buffers(journal_t *journal, + transaction_t *commit_transaction) +{ + struct journal_head *jh; + struct buffer_head *bh; + int locked; + int bufs = 0; + struct buffer_head **wbuf = journal->j_wbuf; + + /* + * Whenever we unlock the journal and sleep, things can get added + * onto ->t_sync_datalist, so we have to keep looping back to + * write_out_data until we *know* that the list is empty. + * + * Cleanup any flushed data buffers from the data list. Even in + * abort mode, we want to flush this out as soon as possible. + */ +write_out_data: + cond_resched(); + spin_lock(&journal->j_list_lock); + + while (commit_transaction->t_sync_datalist) { + jh = commit_transaction->t_sync_datalist; + bh = jh2bh(jh); + locked = 0; + + /* Get reference just to make sure buffer does not disappear + * when we are forced to drop various locks */ + get_bh(bh); + /* If the buffer is dirty, we need to submit IO and hence + * we need the buffer lock. We try to lock the buffer without + * blocking. If we fail, we need to drop j_list_lock and do + * blocking lock_buffer(). + */ + if (buffer_dirty(bh)) { + if (test_set_buffer_locked(bh)) { + BUFFER_TRACE(bh, "needs blocking lock"); + spin_unlock(&journal->j_list_lock); + /* Write out all data to prevent deadlocks */ + journal_do_submit_data(wbuf, bufs); + bufs = 0; + lock_buffer(bh); + spin_lock(&journal->j_list_lock); + } + locked = 1; + } + /* We have to get bh_state lock. Again out of order, sigh. */ + if (!inverted_lock(journal, bh)) { + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + } + /* Someone already cleaned up the buffer? */ + if (!buffer_jbd(bh) + || jh->b_transaction != commit_transaction + || jh->b_jlist != BJ_SyncData) { + jbd_unlock_bh_state(bh); + if (locked) + unlock_buffer(bh); + BUFFER_TRACE(bh, "already cleaned up"); + put_bh(bh); + continue; + } + if (locked && test_clear_buffer_dirty(bh)) { + BUFFER_TRACE(bh, "needs writeout, adding to array"); + wbuf[bufs++] = bh; + __journal_file_buffer(jh, commit_transaction, + BJ_Locked); + jbd_unlock_bh_state(bh); + if (bufs == journal->j_wbufsize) { + spin_unlock(&journal->j_list_lock); + journal_do_submit_data(wbuf, bufs); + bufs = 0; + goto write_out_data; + } + } + else { + BUFFER_TRACE(bh, "writeout complete: unfile"); + __journal_unfile_buffer(jh); + jbd_unlock_bh_state(bh); + if (locked) + unlock_buffer(bh); + journal_remove_journal_head(bh); + /* Once for our safety reference, once for + * journal_remove_journal_head() */ + put_bh(bh); + put_bh(bh); + } + + if (lock_need_resched(&journal->j_list_lock)) { + spin_unlock(&journal->j_list_lock); + goto write_out_data; + } + } + spin_unlock(&journal->j_list_lock); + journal_do_submit_data(wbuf, bufs); +} + +/* + * journal_commit_transaction + * + * The primary function for committing a transaction to the log. This + * function is called by the journal thread to begin a complete commit. + */ +void journal_commit_transaction(journal_t *journal) +{ + transaction_t *commit_transaction; + struct journal_head *jh, *new_jh, *descriptor; + struct buffer_head **wbuf = journal->j_wbuf; + int bufs; + int flags; + int err; + unsigned long blocknr; + char *tagp = NULL; + journal_header_t *header; + journal_block_tag_t *tag = NULL; + int space_left = 0; + int first_tag = 0; + int tag_flag; + int i; + + /* + * First job: lock down the current transaction and wait for + * all outstanding updates to complete. + */ + +#ifdef COMMIT_STATS + spin_lock(&journal->j_list_lock); + summarise_journal_usage(journal); + spin_unlock(&journal->j_list_lock); +#endif + + /* Do we need to erase the effects of a prior journal_flush? */ + if (journal->j_flags & JFS_FLUSHED) { + jbd_debug(3, "super block updated\n"); + journal_update_superblock(journal, 1); + } else { + jbd_debug(3, "superblock not updated\n"); + } + + J_ASSERT(journal->j_running_transaction != NULL); + J_ASSERT(journal->j_committing_transaction == NULL); + + commit_transaction = journal->j_running_transaction; + J_ASSERT(commit_transaction->t_state == T_RUNNING); + + jbd_debug(1, "JBD: starting commit of transaction %d\n", + commit_transaction->t_tid); + + spin_lock(&journal->j_state_lock); + commit_transaction->t_state = T_LOCKED; + + spin_lock(&commit_transaction->t_handle_lock); + while (commit_transaction->t_updates) { + DEFINE_WAIT(wait); + + prepare_to_wait(&journal->j_wait_updates, &wait, + TASK_UNINTERRUPTIBLE); + if (commit_transaction->t_updates) { + spin_unlock(&commit_transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); + schedule(); + spin_lock(&journal->j_state_lock); + spin_lock(&commit_transaction->t_handle_lock); + } + finish_wait(&journal->j_wait_updates, &wait); + } + spin_unlock(&commit_transaction->t_handle_lock); + + J_ASSERT (commit_transaction->t_outstanding_credits <= + journal->j_max_transaction_buffers); + + /* + * First thing we are allowed to do is to discard any remaining + * BJ_Reserved buffers. Note, it is _not_ permissible to assume + * that there are no such buffers: if a large filesystem + * operation like a truncate needs to split itself over multiple + * transactions, then it may try to do a journal_restart() while + * there are still BJ_Reserved buffers outstanding. These must + * be released cleanly from the current transaction. + * + * In this case, the filesystem must still reserve write access + * again before modifying the buffer in the new transaction, but + * we do not require it to remember exactly which old buffers it + * has reserved. This is consistent with the existing behaviour + * that multiple journal_get_write_access() calls to the same + * buffer are perfectly permissable. + */ + while (commit_transaction->t_reserved_list) { + jh = commit_transaction->t_reserved_list; + JBUFFER_TRACE(jh, "reserved, unused: refile"); + /* + * A journal_get_undo_access()+journal_release_buffer() may + * leave undo-committed data. + */ + if (jh->b_committed_data) { + struct buffer_head *bh = jh2bh(jh); + + jbd_lock_bh_state(bh); + jbd_slab_free(jh->b_committed_data, bh->b_size); + jh->b_committed_data = NULL; + jbd_unlock_bh_state(bh); + } + journal_refile_buffer(journal, jh); + } + + /* + * Now try to drop any written-back buffers from the journal's + * checkpoint lists. We do this *before* commit because it potentially + * frees some memory + */ + spin_lock(&journal->j_list_lock); + __journal_clean_checkpoint_list(journal); + spin_unlock(&journal->j_list_lock); + + jbd_debug (3, "JBD: commit phase 1\n"); + + /* + * Switch to a new revoke table. + */ + journal_switch_revoke_table(journal); + + commit_transaction->t_state = T_FLUSH; + journal->j_committing_transaction = commit_transaction; + journal->j_running_transaction = NULL; + commit_transaction->t_log_start = journal->j_head; + wake_up(&journal->j_wait_transaction_locked); + spin_unlock(&journal->j_state_lock); + + jbd_debug (3, "JBD: commit phase 2\n"); + + /* + * First, drop modified flag: all accesses to the buffers + * will be tracked for a new trasaction only -bzzz + */ + spin_lock(&journal->j_list_lock); + if (commit_transaction->t_buffers) { + new_jh = jh = commit_transaction->t_buffers->b_tnext; + do { + J_ASSERT_JH(new_jh, new_jh->b_modified == 1 || + new_jh->b_modified == 0); + new_jh->b_modified = 0; + new_jh = new_jh->b_tnext; + } while (new_jh != jh); + } + spin_unlock(&journal->j_list_lock); + + /* + * Now start flushing things to disk, in the order they appear + * on the transaction lists. Data blocks go first. + */ + err = 0; + journal_submit_data_buffers(journal, commit_transaction); + + /* + * Wait for all previously submitted IO to complete. + */ + spin_lock(&journal->j_list_lock); + while (commit_transaction->t_locked_list) { + struct buffer_head *bh; + + jh = commit_transaction->t_locked_list->b_tprev; + bh = jh2bh(jh); + get_bh(bh); + if (buffer_locked(bh)) { + spin_unlock(&journal->j_list_lock); + wait_on_buffer(bh); + if (unlikely(!buffer_uptodate(bh))) + err = -EIO; + spin_lock(&journal->j_list_lock); + } + if (!inverted_lock(journal, bh)) { + put_bh(bh); + spin_lock(&journal->j_list_lock); + continue; + } + if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { + __journal_unfile_buffer(jh); + jbd_unlock_bh_state(bh); + journal_remove_journal_head(bh); + put_bh(bh); + } else { + jbd_unlock_bh_state(bh); + } + put_bh(bh); + cond_resched_lock(&journal->j_list_lock); + } + spin_unlock(&journal->j_list_lock); + + if (err) + __journal_abort_hard(journal); + + journal_write_revoke_records(journal, commit_transaction); + + jbd_debug(3, "JBD: commit phase 2\n"); + + /* + * If we found any dirty or locked buffers, then we should have + * looped back up to the write_out_data label. If there weren't + * any then journal_clean_data_list should have wiped the list + * clean by now, so check that it is in fact empty. + */ + J_ASSERT (commit_transaction->t_sync_datalist == NULL); + + jbd_debug (3, "JBD: commit phase 3\n"); + + /* + * Way to go: we have now written out all of the data for a + * transaction! Now comes the tricky part: we need to write out + * metadata. Loop over the transaction's entire buffer list: + */ + commit_transaction->t_state = T_COMMIT; + + descriptor = NULL; + bufs = 0; + while (commit_transaction->t_buffers) { + + /* Find the next buffer to be journaled... */ + + jh = commit_transaction->t_buffers; + + /* If we're in abort mode, we just un-journal the buffer and + release it for background writing. */ + + if (is_journal_aborted(journal)) { + JBUFFER_TRACE(jh, "journal is aborting: refile"); + journal_refile_buffer(journal, jh); + /* If that was the last one, we need to clean up + * any descriptor buffers which may have been + * already allocated, even if we are now + * aborting. */ + if (!commit_transaction->t_buffers) + goto start_journal_io; + continue; + } + + /* Make sure we have a descriptor block in which to + record the metadata buffer. */ + + if (!descriptor) { + struct buffer_head *bh; + + J_ASSERT (bufs == 0); + + jbd_debug(4, "JBD: get descriptor\n"); + + descriptor = journal_get_descriptor_buffer(journal); + if (!descriptor) { + __journal_abort_hard(journal); + continue; + } + + bh = jh2bh(descriptor); + jbd_debug(4, "JBD: got buffer %llu (%p)\n", + (unsigned long long)bh->b_blocknr, bh->b_data); + header = (journal_header_t *)&bh->b_data[0]; + header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JFS_DESCRIPTOR_BLOCK); + header->h_sequence = cpu_to_be32(commit_transaction->t_tid); + + tagp = &bh->b_data[sizeof(journal_header_t)]; + space_left = bh->b_size - sizeof(journal_header_t); + first_tag = 1; + set_buffer_jwrite(bh); + set_buffer_dirty(bh); + wbuf[bufs++] = bh; + + /* Record it so that we can wait for IO + completion later */ + BUFFER_TRACE(bh, "ph3: file as descriptor"); + journal_file_buffer(descriptor, commit_transaction, + BJ_LogCtl); + } + + /* Where is the buffer to be written? */ + + err = journal_next_log_block(journal, &blocknr); + /* If the block mapping failed, just abandon the buffer + and repeat this loop: we'll fall into the + refile-on-abort condition above. */ + if (err) { + __journal_abort_hard(journal); + continue; + } + + /* + * start_this_handle() uses t_outstanding_credits to determine + * the free space in the log, but this counter is changed + * by journal_next_log_block() also. + */ + commit_transaction->t_outstanding_credits--; + + /* Bump b_count to prevent truncate from stumbling over + the shadowed buffer! @@@ This can go if we ever get + rid of the BJ_IO/BJ_Shadow pairing of buffers. */ + atomic_inc(&jh2bh(jh)->b_count); + + /* Make a temporary IO buffer with which to write it out + (this will requeue both the metadata buffer and the + temporary IO buffer). new_bh goes on BJ_IO*/ + + set_bit(BH_JWrite, &jh2bh(jh)->b_state); + /* + * akpm: journal_write_metadata_buffer() sets + * new_bh->b_transaction to commit_transaction. + * We need to clean this up before we release new_bh + * (which is of type BJ_IO) + */ + JBUFFER_TRACE(jh, "ph3: write metadata"); + flags = journal_write_metadata_buffer(commit_transaction, + jh, &new_jh, blocknr); + set_bit(BH_JWrite, &jh2bh(new_jh)->b_state); + wbuf[bufs++] = jh2bh(new_jh); + + /* Record the new block's tag in the current descriptor + buffer */ + + tag_flag = 0; + if (flags & 1) + tag_flag |= JFS_FLAG_ESCAPE; + if (!first_tag) + tag_flag |= JFS_FLAG_SAME_UUID; + + tag = (journal_block_tag_t *) tagp; + tag->t_blocknr = cpu_to_be32(jh2bh(jh)->b_blocknr); + tag->t_flags = cpu_to_be32(tag_flag); + tagp += sizeof(journal_block_tag_t); + space_left -= sizeof(journal_block_tag_t); + + if (first_tag) { + memcpy (tagp, journal->j_uuid, 16); + tagp += 16; + space_left -= 16; + first_tag = 0; + } + + /* If there's no more to do, or if the descriptor is full, + let the IO rip! */ + + if (bufs == journal->j_wbufsize || + commit_transaction->t_buffers == NULL || + space_left < sizeof(journal_block_tag_t) + 16) { + + jbd_debug(4, "JBD: Submit %d IOs\n", bufs); + + /* Write an end-of-descriptor marker before + submitting the IOs. "tag" still points to + the last tag we set up. */ + + tag->t_flags |= cpu_to_be32(JFS_FLAG_LAST_TAG); + +start_journal_io: + for (i = 0; i < bufs; i++) { + struct buffer_head *bh = wbuf[i]; + lock_buffer(bh); + clear_buffer_dirty(bh); + set_buffer_uptodate(bh); + bh->b_end_io = journal_end_buffer_io_sync; + submit_bh(WRITE, bh); + } + cond_resched(); + + /* Force a new descriptor to be generated next + time round the loop. */ + descriptor = NULL; + bufs = 0; + } + } + + /* Lo and behold: we have just managed to send a transaction to + the log. Before we can commit it, wait for the IO so far to + complete. Control buffers being written are on the + transaction's t_log_list queue, and metadata buffers are on + the t_iobuf_list queue. + + Wait for the buffers in reverse order. That way we are + less likely to be woken up until all IOs have completed, and + so we incur less scheduling load. + */ + + jbd_debug(3, "JBD: commit phase 4\n"); + + /* + * akpm: these are BJ_IO, and j_list_lock is not needed. + * See __journal_try_to_free_buffer. + */ +wait_for_iobuf: + while (commit_transaction->t_iobuf_list != NULL) { + struct buffer_head *bh; + + jh = commit_transaction->t_iobuf_list->b_tprev; + bh = jh2bh(jh); + if (buffer_locked(bh)) { + wait_on_buffer(bh); + goto wait_for_iobuf; + } + if (cond_resched()) + goto wait_for_iobuf; + + if (unlikely(!buffer_uptodate(bh))) + err = -EIO; + + clear_buffer_jwrite(bh); + + JBUFFER_TRACE(jh, "ph4: unfile after journal write"); + journal_unfile_buffer(journal, jh); + + /* + * ->t_iobuf_list should contain only dummy buffer_heads + * which were created by journal_write_metadata_buffer(). + */ + BUFFER_TRACE(bh, "dumping temporary bh"); + journal_put_journal_head(jh); + __brelse(bh); + J_ASSERT_BH(bh, atomic_read(&bh->b_count) == 0); + free_buffer_head(bh); + + /* We also have to unlock and free the corresponding + shadowed buffer */ + jh = commit_transaction->t_shadow_list->b_tprev; + bh = jh2bh(jh); + clear_bit(BH_JWrite, &bh->b_state); + J_ASSERT_BH(bh, buffer_jbddirty(bh)); + + /* The metadata is now released for reuse, but we need + to remember it against this transaction so that when + we finally commit, we can do any checkpointing + required. */ + JBUFFER_TRACE(jh, "file as BJ_Forget"); + journal_file_buffer(jh, commit_transaction, BJ_Forget); + /* Wake up any transactions which were waiting for this + IO to complete */ + wake_up_bit(&bh->b_state, BH_Unshadow); + JBUFFER_TRACE(jh, "brelse shadowed buffer"); + __brelse(bh); + } + + J_ASSERT (commit_transaction->t_shadow_list == NULL); + + jbd_debug(3, "JBD: commit phase 5\n"); + + /* Here we wait for the revoke record and descriptor record buffers */ + wait_for_ctlbuf: + while (commit_transaction->t_log_list != NULL) { + struct buffer_head *bh; + + jh = commit_transaction->t_log_list->b_tprev; + bh = jh2bh(jh); + if (buffer_locked(bh)) { + wait_on_buffer(bh); + goto wait_for_ctlbuf; + } + if (cond_resched()) + goto wait_for_ctlbuf; + + if (unlikely(!buffer_uptodate(bh))) + err = -EIO; + + BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile"); + clear_buffer_jwrite(bh); + journal_unfile_buffer(journal, jh); + journal_put_journal_head(jh); + __brelse(bh); /* One for getblk */ + /* AKPM: bforget here */ + } + + jbd_debug(3, "JBD: commit phase 6\n"); + + if (journal_write_commit_record(journal, commit_transaction)) + err = -EIO; + + if (err) + __journal_abort_hard(journal); + + /* End of a transaction! Finally, we can do checkpoint + processing: any buffers committed as a result of this + transaction can be removed from any checkpoint list it was on + before. */ + + jbd_debug(3, "JBD: commit phase 7\n"); + + J_ASSERT(commit_transaction->t_sync_datalist == NULL); + J_ASSERT(commit_transaction->t_buffers == NULL); + J_ASSERT(commit_transaction->t_checkpoint_list == NULL); + J_ASSERT(commit_transaction->t_iobuf_list == NULL); + J_ASSERT(commit_transaction->t_shadow_list == NULL); + J_ASSERT(commit_transaction->t_log_list == NULL); + +restart_loop: + /* + * As there are other places (journal_unmap_buffer()) adding buffers + * to this list we have to be careful and hold the j_list_lock. + */ + spin_lock(&journal->j_list_lock); + while (commit_transaction->t_forget) { + transaction_t *cp_transaction; + struct buffer_head *bh; + + jh = commit_transaction->t_forget; + spin_unlock(&journal->j_list_lock); + bh = jh2bh(jh); + jbd_lock_bh_state(bh); + J_ASSERT_JH(jh, jh->b_transaction == commit_transaction || + jh->b_transaction == journal->j_running_transaction); + + /* + * If there is undo-protected committed data against + * this buffer, then we can remove it now. If it is a + * buffer needing such protection, the old frozen_data + * field now points to a committed version of the + * buffer, so rotate that field to the new committed + * data. + * + * Otherwise, we can just throw away the frozen data now. + */ + if (jh->b_committed_data) { + jbd_slab_free(jh->b_committed_data, bh->b_size); + jh->b_committed_data = NULL; + if (jh->b_frozen_data) { + jh->b_committed_data = jh->b_frozen_data; + jh->b_frozen_data = NULL; + } + } else if (jh->b_frozen_data) { + jbd_slab_free(jh->b_frozen_data, bh->b_size); + jh->b_frozen_data = NULL; + } + + spin_lock(&journal->j_list_lock); + cp_transaction = jh->b_cp_transaction; + if (cp_transaction) { + JBUFFER_TRACE(jh, "remove from old cp transaction"); + __journal_remove_checkpoint(jh); + } + + /* Only re-checkpoint the buffer_head if it is marked + * dirty. If the buffer was added to the BJ_Forget list + * by journal_forget, it may no longer be dirty and + * there's no point in keeping a checkpoint record for + * it. */ + + /* A buffer which has been freed while still being + * journaled by a previous transaction may end up still + * being dirty here, but we want to avoid writing back + * that buffer in the future now that the last use has + * been committed. That's not only a performance gain, + * it also stops aliasing problems if the buffer is left + * behind for writeback and gets reallocated for another + * use in a different page. */ + if (buffer_freed(bh)) { + clear_buffer_freed(bh); + clear_buffer_jbddirty(bh); + } + + if (buffer_jbddirty(bh)) { + JBUFFER_TRACE(jh, "add to new checkpointing trans"); + __journal_insert_checkpoint(jh, commit_transaction); + JBUFFER_TRACE(jh, "refile for checkpoint writeback"); + __journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + } else { + J_ASSERT_BH(bh, !buffer_dirty(bh)); + /* The buffer on BJ_Forget list and not jbddirty means + * it has been freed by this transaction and hence it + * could not have been reallocated until this + * transaction has committed. *BUT* it could be + * reallocated once we have written all the data to + * disk and before we process the buffer on BJ_Forget + * list. */ + JBUFFER_TRACE(jh, "refile or unfile freed buffer"); + __journal_refile_buffer(jh); + if (!jh->b_transaction) { + jbd_unlock_bh_state(bh); + /* needs a brelse */ + journal_remove_journal_head(bh); + release_buffer_page(bh); + } else + jbd_unlock_bh_state(bh); + } + cond_resched_lock(&journal->j_list_lock); + } + spin_unlock(&journal->j_list_lock); + /* + * This is a bit sleazy. We borrow j_list_lock to protect + * journal->j_committing_transaction in __journal_remove_checkpoint. + * Really, __journal_remove_checkpoint should be using j_state_lock but + * it's a bit hassle to hold that across __journal_remove_checkpoint + */ + spin_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); + /* + * Now recheck if some buffers did not get attached to the transaction + * while the lock was dropped... + */ + if (commit_transaction->t_forget) { + spin_unlock(&journal->j_list_lock); + spin_unlock(&journal->j_state_lock); + goto restart_loop; + } + + /* Done with this transaction! */ + + jbd_debug(3, "JBD: commit phase 8\n"); + + J_ASSERT(commit_transaction->t_state == T_COMMIT); + + commit_transaction->t_state = T_FINISHED; + J_ASSERT(commit_transaction == journal->j_committing_transaction); + journal->j_commit_sequence = commit_transaction->t_tid; + journal->j_committing_transaction = NULL; + spin_unlock(&journal->j_state_lock); + + if (commit_transaction->t_checkpoint_list == NULL) { + __journal_drop_transaction(journal, commit_transaction); + } else { + if (journal->j_checkpoint_transactions == NULL) { + journal->j_checkpoint_transactions = commit_transaction; + commit_transaction->t_cpnext = commit_transaction; + commit_transaction->t_cpprev = commit_transaction; + } else { + commit_transaction->t_cpnext = + journal->j_checkpoint_transactions; + commit_transaction->t_cpprev = + commit_transaction->t_cpnext->t_cpprev; + commit_transaction->t_cpnext->t_cpprev = + commit_transaction; + commit_transaction->t_cpprev->t_cpnext = + commit_transaction; + } + } + spin_unlock(&journal->j_list_lock); + + jbd_debug(1, "JBD: commit %d complete, head %d\n", + journal->j_commit_sequence, journal->j_tail_sequence); + + wake_up(&journal->j_wait_done_commit); +} diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c new file mode 100644 index 0000000..c518dd8 --- /dev/null +++ b/fs/jbd2/journal.c @@ -0,0 +1,2072 @@ +/* + * linux/fs/jbd/journal.c + * + * Written by Stephen C. Tweedie , 1998 + * + * Copyright 1998 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Generic filesystem journal-writing code; part of the ext2fs + * journaling system. + * + * This file manages journals: areas of disk reserved for logging + * transactional updates. This includes the kernel journaling thread + * which is responsible for scheduling updates to the log. + * + * We do not actually manage the physical storage of the journal in this + * file: that is left to a per-journal policy function, which allows us + * to store the journal within a filesystem-specified area for ext2 + * journaling (ext2 can use a reserved inode for storing the log). + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +EXPORT_SYMBOL(journal_start); +EXPORT_SYMBOL(journal_restart); +EXPORT_SYMBOL(journal_extend); +EXPORT_SYMBOL(journal_stop); +EXPORT_SYMBOL(journal_lock_updates); +EXPORT_SYMBOL(journal_unlock_updates); +EXPORT_SYMBOL(journal_get_write_access); +EXPORT_SYMBOL(journal_get_create_access); +EXPORT_SYMBOL(journal_get_undo_access); +EXPORT_SYMBOL(journal_dirty_data); +EXPORT_SYMBOL(journal_dirty_metadata); +EXPORT_SYMBOL(journal_release_buffer); +EXPORT_SYMBOL(journal_forget); +#if 0 +EXPORT_SYMBOL(journal_sync_buffer); +#endif +EXPORT_SYMBOL(journal_flush); +EXPORT_SYMBOL(journal_revoke); + +EXPORT_SYMBOL(journal_init_dev); +EXPORT_SYMBOL(journal_init_inode); +EXPORT_SYMBOL(journal_update_format); +EXPORT_SYMBOL(journal_check_used_features); +EXPORT_SYMBOL(journal_check_available_features); +EXPORT_SYMBOL(journal_set_features); +EXPORT_SYMBOL(journal_create); +EXPORT_SYMBOL(journal_load); +EXPORT_SYMBOL(journal_destroy); +EXPORT_SYMBOL(journal_update_superblock); +EXPORT_SYMBOL(journal_abort); +EXPORT_SYMBOL(journal_errno); +EXPORT_SYMBOL(journal_ack_err); +EXPORT_SYMBOL(journal_clear_err); +EXPORT_SYMBOL(log_wait_commit); +EXPORT_SYMBOL(journal_start_commit); +EXPORT_SYMBOL(journal_force_commit_nested); +EXPORT_SYMBOL(journal_wipe); +EXPORT_SYMBOL(journal_blocks_per_page); +EXPORT_SYMBOL(journal_invalidatepage); +EXPORT_SYMBOL(journal_try_to_free_buffers); +EXPORT_SYMBOL(journal_force_commit); + +static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *); +static void __journal_abort_soft (journal_t *journal, int errno); +static int journal_create_jbd_slab(size_t slab_size); + +/* + * Helper function used to manage commit timeouts + */ + +static void commit_timeout(unsigned long __data) +{ + struct task_struct * p = (struct task_struct *) __data; + + wake_up_process(p); +} + +/* + * kjournald: The main thread function used to manage a logging device + * journal. + * + * This kernel thread is responsible for two things: + * + * 1) COMMIT: Every so often we need to commit the current state of the + * filesystem to disk. The journal thread is responsible for writing + * all of the metadata buffers to disk. + * + * 2) CHECKPOINT: We cannot reuse a used section of the log file until all + * of the data in that part of the log has been rewritten elsewhere on + * the disk. Flushing these old buffers to reclaim space in the log is + * known as checkpointing, and this thread is responsible for that job. + */ + +static int kjournald(void *arg) +{ + journal_t *journal = arg; + transaction_t *transaction; + + /* + * Set up an interval timer which can be used to trigger a commit wakeup + * after the commit interval expires + */ + setup_timer(&journal->j_commit_timer, commit_timeout, + (unsigned long)current); + + /* Record that the journal thread is running */ + journal->j_task = current; + wake_up(&journal->j_wait_done_commit); + + printk(KERN_INFO "kjournald starting. Commit interval %ld seconds\n", + journal->j_commit_interval / HZ); + + /* + * And now, wait forever for commit wakeup events. + */ + spin_lock(&journal->j_state_lock); + +loop: + if (journal->j_flags & JFS_UNMOUNT) + goto end_loop; + + jbd_debug(1, "commit_sequence=%d, commit_request=%d\n", + journal->j_commit_sequence, journal->j_commit_request); + + if (journal->j_commit_sequence != journal->j_commit_request) { + jbd_debug(1, "OK, requests differ\n"); + spin_unlock(&journal->j_state_lock); + del_timer_sync(&journal->j_commit_timer); + journal_commit_transaction(journal); + spin_lock(&journal->j_state_lock); + goto loop; + } + + wake_up(&journal->j_wait_done_commit); + if (freezing(current)) { + /* + * The simpler the better. Flushing journal isn't a + * good idea, because that depends on threads that may + * be already stopped. + */ + jbd_debug(1, "Now suspending kjournald\n"); + spin_unlock(&journal->j_state_lock); + refrigerator(); + spin_lock(&journal->j_state_lock); + } else { + /* + * We assume on resume that commits are already there, + * so we don't sleep + */ + DEFINE_WAIT(wait); + int should_sleep = 1; + + prepare_to_wait(&journal->j_wait_commit, &wait, + TASK_INTERRUPTIBLE); + if (journal->j_commit_sequence != journal->j_commit_request) + should_sleep = 0; + transaction = journal->j_running_transaction; + if (transaction && time_after_eq(jiffies, + transaction->t_expires)) + should_sleep = 0; + if (journal->j_flags & JFS_UNMOUNT) + should_sleep = 0; + if (should_sleep) { + spin_unlock(&journal->j_state_lock); + schedule(); + spin_lock(&journal->j_state_lock); + } + finish_wait(&journal->j_wait_commit, &wait); + } + + jbd_debug(1, "kjournald wakes\n"); + + /* + * Were we woken up by a commit wakeup event? + */ + transaction = journal->j_running_transaction; + if (transaction && time_after_eq(jiffies, transaction->t_expires)) { + journal->j_commit_request = transaction->t_tid; + jbd_debug(1, "woke because of timeout\n"); + } + goto loop; + +end_loop: + spin_unlock(&journal->j_state_lock); + del_timer_sync(&journal->j_commit_timer); + journal->j_task = NULL; + wake_up(&journal->j_wait_done_commit); + jbd_debug(1, "Journal thread exiting.\n"); + return 0; +} + +static void journal_start_thread(journal_t *journal) +{ + kthread_run(kjournald, journal, "kjournald"); + wait_event(journal->j_wait_done_commit, journal->j_task != 0); +} + +static void journal_kill_thread(journal_t *journal) +{ + spin_lock(&journal->j_state_lock); + journal->j_flags |= JFS_UNMOUNT; + + while (journal->j_task) { + wake_up(&journal->j_wait_commit); + spin_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_done_commit, journal->j_task == 0); + spin_lock(&journal->j_state_lock); + } + spin_unlock(&journal->j_state_lock); +} + +/* + * journal_write_metadata_buffer: write a metadata buffer to the journal. + * + * Writes a metadata buffer to a given disk block. The actual IO is not + * performed but a new buffer_head is constructed which labels the data + * to be written with the correct destination disk block. + * + * Any magic-number escaping which needs to be done will cause a + * copy-out here. If the buffer happens to start with the + * JFS_MAGIC_NUMBER, then we can't write it to the log directly: the + * magic number is only written to the log for descripter blocks. In + * this case, we copy the data and replace the first word with 0, and we + * return a result code which indicates that this buffer needs to be + * marked as an escaped buffer in the corresponding log descriptor + * block. The missing word can then be restored when the block is read + * during recovery. + * + * If the source buffer has already been modified by a new transaction + * since we took the last commit snapshot, we use the frozen copy of + * that data for IO. If we end up using the existing buffer_head's data + * for the write, then we *have* to lock the buffer to prevent anyone + * else from using and possibly modifying it while the IO is in + * progress. + * + * The function returns a pointer to the buffer_heads to be used for IO. + * + * We assume that the journal has already been locked in this function. + * + * Return value: + * <0: Error + * >=0: Finished OK + * + * On success: + * Bit 0 set == escape performed on the data + * Bit 1 set == buffer copy-out performed (kfree the data after IO) + */ + +int journal_write_metadata_buffer(transaction_t *transaction, + struct journal_head *jh_in, + struct journal_head **jh_out, + unsigned long blocknr) +{ + int need_copy_out = 0; + int done_copy_out = 0; + int do_escape = 0; + char *mapped_data; + struct buffer_head *new_bh; + struct journal_head *new_jh; + struct page *new_page; + unsigned int new_offset; + struct buffer_head *bh_in = jh2bh(jh_in); + + /* + * The buffer really shouldn't be locked: only the current committing + * transaction is allowed to write it, so nobody else is allowed + * to do any IO. + * + * akpm: except if we're journalling data, and write() output is + * also part of a shared mapping, and another thread has + * decided to launch a writepage() against this buffer. + */ + J_ASSERT_BH(bh_in, buffer_jbddirty(bh_in)); + + new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL); + + /* + * If a new transaction has already done a buffer copy-out, then + * we use that version of the data for the commit. + */ + jbd_lock_bh_state(bh_in); +repeat: + if (jh_in->b_frozen_data) { + done_copy_out = 1; + new_page = virt_to_page(jh_in->b_frozen_data); + new_offset = offset_in_page(jh_in->b_frozen_data); + } else { + new_page = jh2bh(jh_in)->b_page; + new_offset = offset_in_page(jh2bh(jh_in)->b_data); + } + + mapped_data = kmap_atomic(new_page, KM_USER0); + /* + * Check for escaping + */ + if (*((__be32 *)(mapped_data + new_offset)) == + cpu_to_be32(JFS_MAGIC_NUMBER)) { + need_copy_out = 1; + do_escape = 1; + } + kunmap_atomic(mapped_data, KM_USER0); + + /* + * Do we need to do a data copy? + */ + if (need_copy_out && !done_copy_out) { + char *tmp; + + jbd_unlock_bh_state(bh_in); + tmp = jbd_slab_alloc(bh_in->b_size, GFP_NOFS); + jbd_lock_bh_state(bh_in); + if (jh_in->b_frozen_data) { + jbd_slab_free(tmp, bh_in->b_size); + goto repeat; + } + + jh_in->b_frozen_data = tmp; + mapped_data = kmap_atomic(new_page, KM_USER0); + memcpy(tmp, mapped_data + new_offset, jh2bh(jh_in)->b_size); + kunmap_atomic(mapped_data, KM_USER0); + + new_page = virt_to_page(tmp); + new_offset = offset_in_page(tmp); + done_copy_out = 1; + } + + /* + * Did we need to do an escaping? Now we've done all the + * copying, we can finally do so. + */ + if (do_escape) { + mapped_data = kmap_atomic(new_page, KM_USER0); + *((unsigned int *)(mapped_data + new_offset)) = 0; + kunmap_atomic(mapped_data, KM_USER0); + } + + /* keep subsequent assertions sane */ + new_bh->b_state = 0; + init_buffer(new_bh, NULL, NULL); + atomic_set(&new_bh->b_count, 1); + jbd_unlock_bh_state(bh_in); + + new_jh = journal_add_journal_head(new_bh); /* This sleeps */ + + set_bh_page(new_bh, new_page, new_offset); + new_jh->b_transaction = NULL; + new_bh->b_size = jh2bh(jh_in)->b_size; + new_bh->b_bdev = transaction->t_journal->j_dev; + new_bh->b_blocknr = blocknr; + set_buffer_mapped(new_bh); + set_buffer_dirty(new_bh); + + *jh_out = new_jh; + + /* + * The to-be-written buffer needs to get moved to the io queue, + * and the original buffer whose contents we are shadowing or + * copying is moved to the transaction's shadow queue. + */ + JBUFFER_TRACE(jh_in, "file as BJ_Shadow"); + journal_file_buffer(jh_in, transaction, BJ_Shadow); + JBUFFER_TRACE(new_jh, "file as BJ_IO"); + journal_file_buffer(new_jh, transaction, BJ_IO); + + return do_escape | (done_copy_out << 1); +} + +/* + * Allocation code for the journal file. Manage the space left in the + * journal, so that we can begin checkpointing when appropriate. + */ + +/* + * __log_space_left: Return the number of free blocks left in the journal. + * + * Called with the journal already locked. + * + * Called under j_state_lock + */ + +int __log_space_left(journal_t *journal) +{ + int left = journal->j_free; + + assert_spin_locked(&journal->j_state_lock); + + /* + * Be pessimistic here about the number of those free blocks which + * might be required for log descriptor control blocks. + */ + +#define MIN_LOG_RESERVED_BLOCKS 32 /* Allow for rounding errors */ + + left -= MIN_LOG_RESERVED_BLOCKS; + + if (left <= 0) + return 0; + left -= (left >> 3); + return left; +} + +/* + * Called under j_state_lock. Returns true if a transaction was started. + */ +int __log_start_commit(journal_t *journal, tid_t target) +{ + /* + * Are we already doing a recent enough commit? + */ + if (!tid_geq(journal->j_commit_request, target)) { + /* + * We want a new commit: OK, mark the request and wakup the + * commit thread. We do _not_ do the commit ourselves. + */ + + journal->j_commit_request = target; + jbd_debug(1, "JBD: requesting commit %d/%d\n", + journal->j_commit_request, + journal->j_commit_sequence); + wake_up(&journal->j_wait_commit); + return 1; + } + return 0; +} + +int log_start_commit(journal_t *journal, tid_t tid) +{ + int ret; + + spin_lock(&journal->j_state_lock); + ret = __log_start_commit(journal, tid); + spin_unlock(&journal->j_state_lock); + return ret; +} + +/* + * Force and wait upon a commit if the calling process is not within + * transaction. This is used for forcing out undo-protected data which contains + * bitmaps, when the fs is running out of space. + * + * We can only force the running transaction if we don't have an active handle; + * otherwise, we will deadlock. + * + * Returns true if a transaction was started. + */ +int journal_force_commit_nested(journal_t *journal) +{ + transaction_t *transaction = NULL; + tid_t tid; + + spin_lock(&journal->j_state_lock); + if (journal->j_running_transaction && !current->journal_info) { + transaction = journal->j_running_transaction; + __log_start_commit(journal, transaction->t_tid); + } else if (journal->j_committing_transaction) + transaction = journal->j_committing_transaction; + + if (!transaction) { + spin_unlock(&journal->j_state_lock); + return 0; /* Nothing to retry */ + } + + tid = transaction->t_tid; + spin_unlock(&journal->j_state_lock); + log_wait_commit(journal, tid); + return 1; +} + +/* + * Start a commit of the current running transaction (if any). Returns true + * if a transaction was started, and fills its tid in at *ptid + */ +int journal_start_commit(journal_t *journal, tid_t *ptid) +{ + int ret = 0; + + spin_lock(&journal->j_state_lock); + if (journal->j_running_transaction) { + tid_t tid = journal->j_running_transaction->t_tid; + + ret = __log_start_commit(journal, tid); + if (ret && ptid) + *ptid = tid; + } else if (journal->j_committing_transaction && ptid) { + /* + * If ext3_write_super() recently started a commit, then we + * have to wait for completion of that transaction + */ + *ptid = journal->j_committing_transaction->t_tid; + ret = 1; + } + spin_unlock(&journal->j_state_lock); + return ret; +} + +/* + * Wait for a specified commit to complete. + * The caller may not hold the journal lock. + */ +int log_wait_commit(journal_t *journal, tid_t tid) +{ + int err = 0; + +#ifdef CONFIG_JBD_DEBUG + spin_lock(&journal->j_state_lock); + if (!tid_geq(journal->j_commit_request, tid)) { + printk(KERN_EMERG + "%s: error: j_commit_request=%d, tid=%d\n", + __FUNCTION__, journal->j_commit_request, tid); + } + spin_unlock(&journal->j_state_lock); +#endif + spin_lock(&journal->j_state_lock); + while (tid_gt(tid, journal->j_commit_sequence)) { + jbd_debug(1, "JBD: want %d, j_commit_sequence=%d\n", + tid, journal->j_commit_sequence); + wake_up(&journal->j_wait_commit); + spin_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_done_commit, + !tid_gt(tid, journal->j_commit_sequence)); + spin_lock(&journal->j_state_lock); + } + spin_unlock(&journal->j_state_lock); + + if (unlikely(is_journal_aborted(journal))) { + printk(KERN_EMERG "journal commit I/O error\n"); + err = -EIO; + } + return err; +} + +/* + * Log buffer allocation routines: + */ + +int journal_next_log_block(journal_t *journal, unsigned long *retp) +{ + unsigned long blocknr; + + spin_lock(&journal->j_state_lock); + J_ASSERT(journal->j_free > 1); + + blocknr = journal->j_head; + journal->j_head++; + journal->j_free--; + if (journal->j_head == journal->j_last) + journal->j_head = journal->j_first; + spin_unlock(&journal->j_state_lock); + return journal_bmap(journal, blocknr, retp); +} + +/* + * Conversion of logical to physical block numbers for the journal + * + * On external journals the journal blocks are identity-mapped, so + * this is a no-op. If needed, we can use j_blk_offset - everything is + * ready. + */ +int journal_bmap(journal_t *journal, unsigned long blocknr, + unsigned long *retp) +{ + int err = 0; + unsigned long ret; + + if (journal->j_inode) { + ret = bmap(journal->j_inode, blocknr); + if (ret) + *retp = ret; + else { + char b[BDEVNAME_SIZE]; + + printk(KERN_ALERT "%s: journal block not found " + "at offset %lu on %s\n", + __FUNCTION__, + blocknr, + bdevname(journal->j_dev, b)); + err = -EIO; + __journal_abort_soft(journal, err); + } + } else { + *retp = blocknr; /* +journal->j_blk_offset */ + } + return err; +} + +/* + * We play buffer_head aliasing tricks to write data/metadata blocks to + * the journal without copying their contents, but for journal + * descriptor blocks we do need to generate bona fide buffers. + * + * After the caller of journal_get_descriptor_buffer() has finished modifying + * the buffer's contents they really should run flush_dcache_page(bh->b_page). + * But we don't bother doing that, so there will be coherency problems with + * mmaps of blockdevs which hold live JBD-controlled filesystems. + */ +struct journal_head *journal_get_descriptor_buffer(journal_t *journal) +{ + struct buffer_head *bh; + unsigned long blocknr; + int err; + + err = journal_next_log_block(journal, &blocknr); + + if (err) + return NULL; + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + lock_buffer(bh); + memset(bh->b_data, 0, journal->j_blocksize); + set_buffer_uptodate(bh); + unlock_buffer(bh); + BUFFER_TRACE(bh, "return this buffer"); + return journal_add_journal_head(bh); +} + +/* + * Management for journal control blocks: functions to create and + * destroy journal_t structures, and to initialise and read existing + * journal blocks from disk. */ + +/* First: create and setup a journal_t object in memory. We initialise + * very few fields yet: that has to wait until we have created the + * journal structures from from scratch, or loaded them from disk. */ + +static journal_t * journal_init_common (void) +{ + journal_t *journal; + int err; + + journal = jbd_kmalloc(sizeof(*journal), GFP_KERNEL); + if (!journal) + goto fail; + memset(journal, 0, sizeof(*journal)); + + init_waitqueue_head(&journal->j_wait_transaction_locked); + init_waitqueue_head(&journal->j_wait_logspace); + init_waitqueue_head(&journal->j_wait_done_commit); + init_waitqueue_head(&journal->j_wait_checkpoint); + init_waitqueue_head(&journal->j_wait_commit); + init_waitqueue_head(&journal->j_wait_updates); + mutex_init(&journal->j_barrier); + mutex_init(&journal->j_checkpoint_mutex); + spin_lock_init(&journal->j_revoke_lock); + spin_lock_init(&journal->j_list_lock); + spin_lock_init(&journal->j_state_lock); + + journal->j_commit_interval = (HZ * JBD_DEFAULT_MAX_COMMIT_AGE); + + /* The journal is marked for error until we succeed with recovery! */ + journal->j_flags = JFS_ABORT; + + /* Set up a default-sized revoke table for the new mount. */ + err = journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH); + if (err) { + kfree(journal); + goto fail; + } + return journal; +fail: + return NULL; +} + +/* journal_init_dev and journal_init_inode: + * + * Create a journal structure assigned some fixed set of disk blocks to + * the journal. We don't actually touch those disk blocks yet, but we + * need to set up all of the mapping information to tell the journaling + * system where the journal blocks are. + * + */ + +/** + * journal_t * journal_init_dev() - creates an initialises a journal structure + * @bdev: Block device on which to create the journal + * @fs_dev: Device which hold journalled filesystem for this journal. + * @start: Block nr Start of journal. + * @len: Length of the journal in blocks. + * @blocksize: blocksize of journalling device + * @returns: a newly created journal_t * + * + * journal_init_dev creates a journal which maps a fixed contiguous + * range of blocks on an arbitrary block device. + * + */ +journal_t * journal_init_dev(struct block_device *bdev, + struct block_device *fs_dev, + int start, int len, int blocksize) +{ + journal_t *journal = journal_init_common(); + struct buffer_head *bh; + int n; + + if (!journal) + return NULL; + + /* journal descriptor can store up to n blocks -bzzz */ + journal->j_blocksize = blocksize; + n = journal->j_blocksize / sizeof(journal_block_tag_t); + journal->j_wbufsize = n; + journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); + if (!journal->j_wbuf) { + printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + __FUNCTION__); + kfree(journal); + journal = NULL; + } + journal->j_dev = bdev; + journal->j_fs_dev = fs_dev; + journal->j_blk_offset = start; + journal->j_maxlen = len; + + bh = __getblk(journal->j_dev, start, journal->j_blocksize); + J_ASSERT(bh != NULL); + journal->j_sb_buffer = bh; + journal->j_superblock = (journal_superblock_t *)bh->b_data; + + return journal; +} + +/** + * journal_t * journal_init_inode () - creates a journal which maps to a inode. + * @inode: An inode to create the journal in + * + * journal_init_inode creates a journal which maps an on-disk inode as + * the journal. The inode must exist already, must support bmap() and + * must have all data blocks preallocated. + */ +journal_t * journal_init_inode (struct inode *inode) +{ + struct buffer_head *bh; + journal_t *journal = journal_init_common(); + int err; + int n; + unsigned long blocknr; + + if (!journal) + return NULL; + + journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev; + journal->j_inode = inode; + jbd_debug(1, + "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n", + journal, inode->i_sb->s_id, inode->i_ino, + (long long) inode->i_size, + inode->i_sb->s_blocksize_bits, inode->i_sb->s_blocksize); + + journal->j_maxlen = inode->i_size >> inode->i_sb->s_blocksize_bits; + journal->j_blocksize = inode->i_sb->s_blocksize; + + /* journal descriptor can store up to n blocks -bzzz */ + n = journal->j_blocksize / sizeof(journal_block_tag_t); + journal->j_wbufsize = n; + journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); + if (!journal->j_wbuf) { + printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + __FUNCTION__); + kfree(journal); + return NULL; + } + + err = journal_bmap(journal, 0, &blocknr); + /* If that failed, give up */ + if (err) { + printk(KERN_ERR "%s: Cannnot locate journal superblock\n", + __FUNCTION__); + kfree(journal); + return NULL; + } + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + J_ASSERT(bh != NULL); + journal->j_sb_buffer = bh; + journal->j_superblock = (journal_superblock_t *)bh->b_data; + + return journal; +} + +/* + * If the journal init or create aborts, we need to mark the journal + * superblock as being NULL to prevent the journal destroy from writing + * back a bogus superblock. + */ +static void journal_fail_superblock (journal_t *journal) +{ + struct buffer_head *bh = journal->j_sb_buffer; + brelse(bh); + journal->j_sb_buffer = NULL; +} + +/* + * Given a journal_t structure, initialise the various fields for + * startup of a new journaling session. We use this both when creating + * a journal, and after recovering an old journal to reset it for + * subsequent use. + */ + +static int journal_reset(journal_t *journal) +{ + journal_superblock_t *sb = journal->j_superblock; + unsigned long first, last; + + first = be32_to_cpu(sb->s_first); + last = be32_to_cpu(sb->s_maxlen); + + journal->j_first = first; + journal->j_last = last; + + journal->j_head = first; + journal->j_tail = first; + journal->j_free = last - first; + + journal->j_tail_sequence = journal->j_transaction_sequence; + journal->j_commit_sequence = journal->j_transaction_sequence - 1; + journal->j_commit_request = journal->j_commit_sequence; + + journal->j_max_transaction_buffers = journal->j_maxlen / 4; + + /* Add the dynamic fields and write it to disk. */ + journal_update_superblock(journal, 1); + journal_start_thread(journal); + return 0; +} + +/** + * int journal_create() - Initialise the new journal file + * @journal: Journal to create. This structure must have been initialised + * + * Given a journal_t structure which tells us which disk blocks we can + * use, create a new journal superblock and initialise all of the + * journal fields from scratch. + **/ +int journal_create(journal_t *journal) +{ + unsigned long blocknr; + struct buffer_head *bh; + journal_superblock_t *sb; + int i, err; + + if (journal->j_maxlen < JFS_MIN_JOURNAL_BLOCKS) { + printk (KERN_ERR "Journal length (%d blocks) too short.\n", + journal->j_maxlen); + journal_fail_superblock(journal); + return -EINVAL; + } + + if (journal->j_inode == NULL) { + /* + * We don't know what block to start at! + */ + printk(KERN_EMERG + "%s: creation of journal on external device!\n", + __FUNCTION__); + BUG(); + } + + /* Zero out the entire journal on disk. We cannot afford to + have any blocks on disk beginning with JFS_MAGIC_NUMBER. */ + jbd_debug(1, "JBD: Zeroing out journal blocks...\n"); + for (i = 0; i < journal->j_maxlen; i++) { + err = journal_bmap(journal, i, &blocknr); + if (err) + return err; + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + lock_buffer(bh); + memset (bh->b_data, 0, journal->j_blocksize); + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + BUFFER_TRACE(bh, "marking uptodate"); + set_buffer_uptodate(bh); + unlock_buffer(bh); + __brelse(bh); + } + + sync_blockdev(journal->j_dev); + jbd_debug(1, "JBD: journal cleared.\n"); + + /* OK, fill in the initial static fields in the new superblock */ + sb = journal->j_superblock; + + sb->s_header.h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); + sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2); + + sb->s_blocksize = cpu_to_be32(journal->j_blocksize); + sb->s_maxlen = cpu_to_be32(journal->j_maxlen); + sb->s_first = cpu_to_be32(1); + + journal->j_transaction_sequence = 1; + + journal->j_flags &= ~JFS_ABORT; + journal->j_format_version = 2; + + return journal_reset(journal); +} + +/** + * void journal_update_superblock() - Update journal sb on disk. + * @journal: The journal to update. + * @wait: Set to '0' if you don't want to wait for IO completion. + * + * Update a journal's dynamic superblock fields and write it to disk, + * optionally waiting for the IO to complete. + */ +void journal_update_superblock(journal_t *journal, int wait) +{ + journal_superblock_t *sb = journal->j_superblock; + struct buffer_head *bh = journal->j_sb_buffer; + + /* + * As a special case, if the on-disk copy is already marked as needing + * no recovery (s_start == 0) and there are no outstanding transactions + * in the filesystem, then we can safely defer the superblock update + * until the next commit by setting JFS_FLUSHED. This avoids + * attempting a write to a potential-readonly device. + */ + if (sb->s_start == 0 && journal->j_tail_sequence == + journal->j_transaction_sequence) { + jbd_debug(1,"JBD: Skipping superblock update on recovered sb " + "(start %ld, seq %d, errno %d)\n", + journal->j_tail, journal->j_tail_sequence, + journal->j_errno); + goto out; + } + + spin_lock(&journal->j_state_lock); + jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n", + journal->j_tail, journal->j_tail_sequence, journal->j_errno); + + sb->s_sequence = cpu_to_be32(journal->j_tail_sequence); + sb->s_start = cpu_to_be32(journal->j_tail); + sb->s_errno = cpu_to_be32(journal->j_errno); + spin_unlock(&journal->j_state_lock); + + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + if (wait) + sync_dirty_buffer(bh); + else + ll_rw_block(SWRITE, 1, &bh); + +out: + /* If we have just flushed the log (by marking s_start==0), then + * any future commit will have to be careful to update the + * superblock again to re-record the true start of the log. */ + + spin_lock(&journal->j_state_lock); + if (sb->s_start) + journal->j_flags &= ~JFS_FLUSHED; + else + journal->j_flags |= JFS_FLUSHED; + spin_unlock(&journal->j_state_lock); +} + +/* + * Read the superblock for a given journal, performing initial + * validation of the format. + */ + +static int journal_get_superblock(journal_t *journal) +{ + struct buffer_head *bh; + journal_superblock_t *sb; + int err = -EIO; + + bh = journal->j_sb_buffer; + + J_ASSERT(bh != NULL); + if (!buffer_uptodate(bh)) { + ll_rw_block(READ, 1, &bh); + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) { + printk (KERN_ERR + "JBD: IO error reading journal superblock\n"); + goto out; + } + } + + sb = journal->j_superblock; + + err = -EINVAL; + + if (sb->s_header.h_magic != cpu_to_be32(JFS_MAGIC_NUMBER) || + sb->s_blocksize != cpu_to_be32(journal->j_blocksize)) { + printk(KERN_WARNING "JBD: no valid journal superblock found\n"); + goto out; + } + + switch(be32_to_cpu(sb->s_header.h_blocktype)) { + case JFS_SUPERBLOCK_V1: + journal->j_format_version = 1; + break; + case JFS_SUPERBLOCK_V2: + journal->j_format_version = 2; + break; + default: + printk(KERN_WARNING "JBD: unrecognised superblock format ID\n"); + goto out; + } + + if (be32_to_cpu(sb->s_maxlen) < journal->j_maxlen) + journal->j_maxlen = be32_to_cpu(sb->s_maxlen); + else if (be32_to_cpu(sb->s_maxlen) > journal->j_maxlen) { + printk (KERN_WARNING "JBD: journal file too short\n"); + goto out; + } + + return 0; + +out: + journal_fail_superblock(journal); + return err; +} + +/* + * Load the on-disk journal superblock and read the key fields into the + * journal_t. + */ + +static int load_superblock(journal_t *journal) +{ + int err; + journal_superblock_t *sb; + + err = journal_get_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + + journal->j_tail_sequence = be32_to_cpu(sb->s_sequence); + journal->j_tail = be32_to_cpu(sb->s_start); + journal->j_first = be32_to_cpu(sb->s_first); + journal->j_last = be32_to_cpu(sb->s_maxlen); + journal->j_errno = be32_to_cpu(sb->s_errno); + + return 0; +} + + +/** + * int journal_load() - Read journal from disk. + * @journal: Journal to act on. + * + * Given a journal_t structure which tells us which disk blocks contain + * a journal, read the journal from disk to initialise the in-memory + * structures. + */ +int journal_load(journal_t *journal) +{ + int err; + journal_superblock_t *sb; + + err = load_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + /* If this is a V2 superblock, then we have to check the + * features flags on it. */ + + if (journal->j_format_version >= 2) { + if ((sb->s_feature_ro_compat & + ~cpu_to_be32(JFS_KNOWN_ROCOMPAT_FEATURES)) || + (sb->s_feature_incompat & + ~cpu_to_be32(JFS_KNOWN_INCOMPAT_FEATURES))) { + printk (KERN_WARNING + "JBD: Unrecognised features on journal\n"); + return -EINVAL; + } + } + + /* + * Create a slab for this blocksize + */ + err = journal_create_jbd_slab(be32_to_cpu(sb->s_blocksize)); + if (err) + return err; + + /* Let the recovery code check whether it needs to recover any + * data from the journal. */ + if (journal_recover(journal)) + goto recovery_error; + + /* OK, we've finished with the dynamic journal bits: + * reinitialise the dynamic contents of the superblock in memory + * and reset them on disk. */ + if (journal_reset(journal)) + goto recovery_error; + + journal->j_flags &= ~JFS_ABORT; + journal->j_flags |= JFS_LOADED; + return 0; + +recovery_error: + printk (KERN_WARNING "JBD: recovery failed\n"); + return -EIO; +} + +/** + * void journal_destroy() - Release a journal_t structure. + * @journal: Journal to act on. + * + * Release a journal_t structure once it is no longer in use by the + * journaled object. + */ +void journal_destroy(journal_t *journal) +{ + /* Wait for the commit thread to wake up and die. */ + journal_kill_thread(journal); + + /* Force a final log commit */ + if (journal->j_running_transaction) + journal_commit_transaction(journal); + + /* Force any old transactions to disk */ + + /* Totally anal locking here... */ + spin_lock(&journal->j_list_lock); + while (journal->j_checkpoint_transactions != NULL) { + spin_unlock(&journal->j_list_lock); + log_do_checkpoint(journal); + spin_lock(&journal->j_list_lock); + } + + J_ASSERT(journal->j_running_transaction == NULL); + J_ASSERT(journal->j_committing_transaction == NULL); + J_ASSERT(journal->j_checkpoint_transactions == NULL); + spin_unlock(&journal->j_list_lock); + + /* We can now mark the journal as empty. */ + journal->j_tail = 0; + journal->j_tail_sequence = ++journal->j_transaction_sequence; + if (journal->j_sb_buffer) { + journal_update_superblock(journal, 1); + brelse(journal->j_sb_buffer); + } + + if (journal->j_inode) + iput(journal->j_inode); + if (journal->j_revoke) + journal_destroy_revoke(journal); + kfree(journal->j_wbuf); + kfree(journal); +} + + +/** + *int journal_check_used_features () - Check if features specified are used. + * @journal: Journal to check. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Check whether the journal uses all of a given set of + * features. Return true (non-zero) if it does. + **/ + +int journal_check_used_features (journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + if (!compat && !ro && !incompat) + return 1; + if (journal->j_format_version == 1) + return 0; + + sb = journal->j_superblock; + + if (((be32_to_cpu(sb->s_feature_compat) & compat) == compat) && + ((be32_to_cpu(sb->s_feature_ro_compat) & ro) == ro) && + ((be32_to_cpu(sb->s_feature_incompat) & incompat) == incompat)) + return 1; + + return 0; +} + +/** + * int journal_check_available_features() - Check feature set in journalling layer + * @journal: Journal to check. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Check whether the journaling code supports the use of + * all of a given set of features on this journal. Return true + * (non-zero) if it can. */ + +int journal_check_available_features (journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + if (!compat && !ro && !incompat) + return 1; + + sb = journal->j_superblock; + + /* We can support any known requested features iff the + * superblock is in version 2. Otherwise we fail to support any + * extended sb features. */ + + if (journal->j_format_version != 2) + return 0; + + if ((compat & JFS_KNOWN_COMPAT_FEATURES) == compat && + (ro & JFS_KNOWN_ROCOMPAT_FEATURES) == ro && + (incompat & JFS_KNOWN_INCOMPAT_FEATURES) == incompat) + return 1; + + return 0; +} + +/** + * int journal_set_features () - Mark a given journal feature in the superblock + * @journal: Journal to act on. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Mark a given journal feature as present on the + * superblock. Returns true if the requested features could be set. + * + */ + +int journal_set_features (journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + if (journal_check_used_features(journal, compat, ro, incompat)) + return 1; + + if (!journal_check_available_features(journal, compat, ro, incompat)) + return 0; + + jbd_debug(1, "Setting new features 0x%lx/0x%lx/0x%lx\n", + compat, ro, incompat); + + sb = journal->j_superblock; + + sb->s_feature_compat |= cpu_to_be32(compat); + sb->s_feature_ro_compat |= cpu_to_be32(ro); + sb->s_feature_incompat |= cpu_to_be32(incompat); + + return 1; +} + + +/** + * int journal_update_format () - Update on-disk journal structure. + * @journal: Journal to act on. + * + * Given an initialised but unloaded journal struct, poke about in the + * on-disk structure to update it to the most recent supported version. + */ +int journal_update_format (journal_t *journal) +{ + journal_superblock_t *sb; + int err; + + err = journal_get_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + + switch (be32_to_cpu(sb->s_header.h_blocktype)) { + case JFS_SUPERBLOCK_V2: + return 0; + case JFS_SUPERBLOCK_V1: + return journal_convert_superblock_v1(journal, sb); + default: + break; + } + return -EINVAL; +} + +static int journal_convert_superblock_v1(journal_t *journal, + journal_superblock_t *sb) +{ + int offset, blocksize; + struct buffer_head *bh; + + printk(KERN_WARNING + "JBD: Converting superblock from version 1 to 2.\n"); + + /* Pre-initialise new fields to zero */ + offset = ((char *) &(sb->s_feature_compat)) - ((char *) sb); + blocksize = be32_to_cpu(sb->s_blocksize); + memset(&sb->s_feature_compat, 0, blocksize-offset); + + sb->s_nr_users = cpu_to_be32(1); + sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2); + journal->j_format_version = 2; + + bh = journal->j_sb_buffer; + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + sync_dirty_buffer(bh); + return 0; +} + + +/** + * int journal_flush () - Flush journal + * @journal: Journal to act on. + * + * Flush all data for a given journal to disk and empty the journal. + * Filesystems can use this when remounting readonly to ensure that + * recovery does not need to happen on remount. + */ + +int journal_flush(journal_t *journal) +{ + int err = 0; + transaction_t *transaction = NULL; + unsigned long old_tail; + + spin_lock(&journal->j_state_lock); + + /* Force everything buffered to the log... */ + if (journal->j_running_transaction) { + transaction = journal->j_running_transaction; + __log_start_commit(journal, transaction->t_tid); + } else if (journal->j_committing_transaction) + transaction = journal->j_committing_transaction; + + /* Wait for the log commit to complete... */ + if (transaction) { + tid_t tid = transaction->t_tid; + + spin_unlock(&journal->j_state_lock); + log_wait_commit(journal, tid); + } else { + spin_unlock(&journal->j_state_lock); + } + + /* ...and flush everything in the log out to disk. */ + spin_lock(&journal->j_list_lock); + while (!err && journal->j_checkpoint_transactions != NULL) { + spin_unlock(&journal->j_list_lock); + err = log_do_checkpoint(journal); + spin_lock(&journal->j_list_lock); + } + spin_unlock(&journal->j_list_lock); + cleanup_journal_tail(journal); + + /* Finally, mark the journal as really needing no recovery. + * This sets s_start==0 in the underlying superblock, which is + * the magic code for a fully-recovered superblock. Any future + * commits of data to the journal will restore the current + * s_start value. */ + spin_lock(&journal->j_state_lock); + old_tail = journal->j_tail; + journal->j_tail = 0; + spin_unlock(&journal->j_state_lock); + journal_update_superblock(journal, 1); + spin_lock(&journal->j_state_lock); + journal->j_tail = old_tail; + + J_ASSERT(!journal->j_running_transaction); + J_ASSERT(!journal->j_committing_transaction); + J_ASSERT(!journal->j_checkpoint_transactions); + J_ASSERT(journal->j_head == journal->j_tail); + J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence); + spin_unlock(&journal->j_state_lock); + return err; +} + +/** + * int journal_wipe() - Wipe journal contents + * @journal: Journal to act on. + * @write: flag (see below) + * + * Wipe out all of the contents of a journal, safely. This will produce + * a warning if the journal contains any valid recovery information. + * Must be called between journal_init_*() and journal_load(). + * + * If 'write' is non-zero, then we wipe out the journal on disk; otherwise + * we merely suppress recovery. + */ + +int journal_wipe(journal_t *journal, int write) +{ + journal_superblock_t *sb; + int err = 0; + + J_ASSERT (!(journal->j_flags & JFS_LOADED)); + + err = load_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + + if (!journal->j_tail) + goto no_recovery; + + printk (KERN_WARNING "JBD: %s recovery information on journal\n", + write ? "Clearing" : "Ignoring"); + + err = journal_skip_recovery(journal); + if (write) + journal_update_superblock(journal, 1); + + no_recovery: + return err; +} + +/* + * journal_dev_name: format a character string to describe on what + * device this journal is present. + */ + +static const char *journal_dev_name(journal_t *journal, char *buffer) +{ + struct block_device *bdev; + + if (journal->j_inode) + bdev = journal->j_inode->i_sb->s_bdev; + else + bdev = journal->j_dev; + + return bdevname(bdev, buffer); +} + +/* + * Journal abort has very specific semantics, which we describe + * for journal abort. + * + * Two internal function, which provide abort to te jbd layer + * itself are here. + */ + +/* + * Quick version for internal journal use (doesn't lock the journal). + * Aborts hard --- we mark the abort as occurred, but do _nothing_ else, + * and don't attempt to make any other journal updates. + */ +void __journal_abort_hard(journal_t *journal) +{ + transaction_t *transaction; + char b[BDEVNAME_SIZE]; + + if (journal->j_flags & JFS_ABORT) + return; + + printk(KERN_ERR "Aborting journal on device %s.\n", + journal_dev_name(journal, b)); + + spin_lock(&journal->j_state_lock); + journal->j_flags |= JFS_ABORT; + transaction = journal->j_running_transaction; + if (transaction) + __log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); +} + +/* Soft abort: record the abort error status in the journal superblock, + * but don't do any other IO. */ +static void __journal_abort_soft (journal_t *journal, int errno) +{ + if (journal->j_flags & JFS_ABORT) + return; + + if (!journal->j_errno) + journal->j_errno = errno; + + __journal_abort_hard(journal); + + if (errno) + journal_update_superblock(journal, 1); +} + +/** + * void journal_abort () - Shutdown the journal immediately. + * @journal: the journal to shutdown. + * @errno: an error number to record in the journal indicating + * the reason for the shutdown. + * + * Perform a complete, immediate shutdown of the ENTIRE + * journal (not of a single transaction). This operation cannot be + * undone without closing and reopening the journal. + * + * The journal_abort function is intended to support higher level error + * recovery mechanisms such as the ext2/ext3 remount-readonly error + * mode. + * + * Journal abort has very specific semantics. Any existing dirty, + * unjournaled buffers in the main filesystem will still be written to + * disk by bdflush, but the journaling mechanism will be suspended + * immediately and no further transaction commits will be honoured. + * + * Any dirty, journaled buffers will be written back to disk without + * hitting the journal. Atomicity cannot be guaranteed on an aborted + * filesystem, but we _do_ attempt to leave as much data as possible + * behind for fsck to use for cleanup. + * + * Any attempt to get a new transaction handle on a journal which is in + * ABORT state will just result in an -EROFS error return. A + * journal_stop on an existing handle will return -EIO if we have + * entered abort state during the update. + * + * Recursive transactions are not disturbed by journal abort until the + * final journal_stop, which will receive the -EIO error. + * + * Finally, the journal_abort call allows the caller to supply an errno + * which will be recorded (if possible) in the journal superblock. This + * allows a client to record failure conditions in the middle of a + * transaction without having to complete the transaction to record the + * failure to disk. ext3_error, for example, now uses this + * functionality. + * + * Errors which originate from within the journaling layer will NOT + * supply an errno; a null errno implies that absolutely no further + * writes are done to the journal (unless there are any already in + * progress). + * + */ + +void journal_abort(journal_t *journal, int errno) +{ + __journal_abort_soft(journal, errno); +} + +/** + * int journal_errno () - returns the journal's error state. + * @journal: journal to examine. + * + * This is the errno numbet set with journal_abort(), the last + * time the journal was mounted - if the journal was stopped + * without calling abort this will be 0. + * + * If the journal has been aborted on this mount time -EROFS will + * be returned. + */ +int journal_errno(journal_t *journal) +{ + int err; + + spin_lock(&journal->j_state_lock); + if (journal->j_flags & JFS_ABORT) + err = -EROFS; + else + err = journal->j_errno; + spin_unlock(&journal->j_state_lock); + return err; +} + +/** + * int journal_clear_err () - clears the journal's error state + * @journal: journal to act on. + * + * An error must be cleared or Acked to take a FS out of readonly + * mode. + */ +int journal_clear_err(journal_t *journal) +{ + int err = 0; + + spin_lock(&journal->j_state_lock); + if (journal->j_flags & JFS_ABORT) + err = -EROFS; + else + journal->j_errno = 0; + spin_unlock(&journal->j_state_lock); + return err; +} + +/** + * void journal_ack_err() - Ack journal err. + * @journal: journal to act on. + * + * An error must be cleared or Acked to take a FS out of readonly + * mode. + */ +void journal_ack_err(journal_t *journal) +{ + spin_lock(&journal->j_state_lock); + if (journal->j_errno) + journal->j_flags |= JFS_ACK_ERR; + spin_unlock(&journal->j_state_lock); +} + +int journal_blocks_per_page(struct inode *inode) +{ + return 1 << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); +} + +/* + * Simple support for retrying memory allocations. Introduced to help to + * debug different VM deadlock avoidance strategies. + */ +void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) +{ + return kmalloc(size, flags | (retry ? __GFP_NOFAIL : 0)); +} + +/* + * jbd slab management: create 1k, 2k, 4k, 8k slabs as needed + * and allocate frozen and commit buffers from these slabs. + * + * Reason for doing this is to avoid, SLAB_DEBUG - since it could + * cause bh to cross page boundary. + */ + +#define JBD_MAX_SLABS 5 +#define JBD_SLAB_INDEX(size) (size >> 11) + +static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static const char *jbd_slab_names[JBD_MAX_SLABS] = { + "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" +}; + +static void journal_destroy_jbd_slabs(void) +{ + int i; + + for (i = 0; i < JBD_MAX_SLABS; i++) { + if (jbd_slab[i]) + kmem_cache_destroy(jbd_slab[i]); + jbd_slab[i] = NULL; + } +} + +static int journal_create_jbd_slab(size_t slab_size) +{ + int i = JBD_SLAB_INDEX(slab_size); + + BUG_ON(i >= JBD_MAX_SLABS); + + /* + * Check if we already have a slab created for this size + */ + if (jbd_slab[i]) + return 0; + + /* + * Create a slab and force alignment to be same as slabsize - + * this will make sure that allocations won't cross the page + * boundary. + */ + jbd_slab[i] = kmem_cache_create(jbd_slab_names[i], + slab_size, slab_size, 0, NULL, NULL); + if (!jbd_slab[i]) { + printk(KERN_EMERG "JBD: no memory for jbd_slab cache\n"); + return -ENOMEM; + } + return 0; +} + +void * jbd_slab_alloc(size_t size, gfp_t flags) +{ + int idx; + + idx = JBD_SLAB_INDEX(size); + BUG_ON(jbd_slab[idx] == NULL); + return kmem_cache_alloc(jbd_slab[idx], flags | __GFP_NOFAIL); +} + +void jbd_slab_free(void *ptr, size_t size) +{ + int idx; + + idx = JBD_SLAB_INDEX(size); + BUG_ON(jbd_slab[idx] == NULL); + kmem_cache_free(jbd_slab[idx], ptr); +} + +/* + * Journal_head storage management + */ +static kmem_cache_t *journal_head_cache; +#ifdef CONFIG_JBD_DEBUG +static atomic_t nr_journal_heads = ATOMIC_INIT(0); +#endif + +static int journal_init_journal_head_cache(void) +{ + int retval; + + J_ASSERT(journal_head_cache == 0); + journal_head_cache = kmem_cache_create("journal_head", + sizeof(struct journal_head), + 0, /* offset */ + 0, /* flags */ + NULL, /* ctor */ + NULL); /* dtor */ + retval = 0; + if (journal_head_cache == 0) { + retval = -ENOMEM; + printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); + } + return retval; +} + +static void journal_destroy_journal_head_cache(void) +{ + J_ASSERT(journal_head_cache != NULL); + kmem_cache_destroy(journal_head_cache); + journal_head_cache = NULL; +} + +/* + * journal_head splicing and dicing + */ +static struct journal_head *journal_alloc_journal_head(void) +{ + struct journal_head *ret; + static unsigned long last_warning; + +#ifdef CONFIG_JBD_DEBUG + atomic_inc(&nr_journal_heads); +#endif + ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS); + if (ret == 0) { + jbd_debug(1, "out of memory for journal_head\n"); + if (time_after(jiffies, last_warning + 5*HZ)) { + printk(KERN_NOTICE "ENOMEM in %s, retrying.\n", + __FUNCTION__); + last_warning = jiffies; + } + while (ret == 0) { + yield(); + ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS); + } + } + return ret; +} + +static void journal_free_journal_head(struct journal_head *jh) +{ +#ifdef CONFIG_JBD_DEBUG + atomic_dec(&nr_journal_heads); + memset(jh, JBD_POISON_FREE, sizeof(*jh)); +#endif + kmem_cache_free(journal_head_cache, jh); +} + +/* + * A journal_head is attached to a buffer_head whenever JBD has an + * interest in the buffer. + * + * Whenever a buffer has an attached journal_head, its ->b_state:BH_JBD bit + * is set. This bit is tested in core kernel code where we need to take + * JBD-specific actions. Testing the zeroness of ->b_private is not reliable + * there. + * + * When a buffer has its BH_JBD bit set, its ->b_count is elevated by one. + * + * When a buffer has its BH_JBD bit set it is immune from being released by + * core kernel code, mainly via ->b_count. + * + * A journal_head may be detached from its buffer_head when the journal_head's + * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL. + * Various places in JBD call journal_remove_journal_head() to indicate that the + * journal_head can be dropped if needed. + * + * Various places in the kernel want to attach a journal_head to a buffer_head + * _before_ attaching the journal_head to a transaction. To protect the + * journal_head in this situation, journal_add_journal_head elevates the + * journal_head's b_jcount refcount by one. The caller must call + * journal_put_journal_head() to undo this. + * + * So the typical usage would be: + * + * (Attach a journal_head if needed. Increments b_jcount) + * struct journal_head *jh = journal_add_journal_head(bh); + * ... + * jh->b_transaction = xxx; + * journal_put_journal_head(jh); + * + * Now, the journal_head's b_jcount is zero, but it is safe from being released + * because it has a non-zero b_transaction. + */ + +/* + * Give a buffer_head a journal_head. + * + * Doesn't need the journal lock. + * May sleep. + */ +struct journal_head *journal_add_journal_head(struct buffer_head *bh) +{ + struct journal_head *jh; + struct journal_head *new_jh = NULL; + +repeat: + if (!buffer_jbd(bh)) { + new_jh = journal_alloc_journal_head(); + memset(new_jh, 0, sizeof(*new_jh)); + } + + jbd_lock_bh_journal_head(bh); + if (buffer_jbd(bh)) { + jh = bh2jh(bh); + } else { + J_ASSERT_BH(bh, + (atomic_read(&bh->b_count) > 0) || + (bh->b_page && bh->b_page->mapping)); + + if (!new_jh) { + jbd_unlock_bh_journal_head(bh); + goto repeat; + } + + jh = new_jh; + new_jh = NULL; /* We consumed it */ + set_buffer_jbd(bh); + bh->b_private = jh; + jh->b_bh = bh; + get_bh(bh); + BUFFER_TRACE(bh, "added journal_head"); + } + jh->b_jcount++; + jbd_unlock_bh_journal_head(bh); + if (new_jh) + journal_free_journal_head(new_jh); + return bh->b_private; +} + +/* + * Grab a ref against this buffer_head's journal_head. If it ended up not + * having a journal_head, return NULL + */ +struct journal_head *journal_grab_journal_head(struct buffer_head *bh) +{ + struct journal_head *jh = NULL; + + jbd_lock_bh_journal_head(bh); + if (buffer_jbd(bh)) { + jh = bh2jh(bh); + jh->b_jcount++; + } + jbd_unlock_bh_journal_head(bh); + return jh; +} + +static void __journal_remove_journal_head(struct buffer_head *bh) +{ + struct journal_head *jh = bh2jh(bh); + + J_ASSERT_JH(jh, jh->b_jcount >= 0); + + get_bh(bh); + if (jh->b_jcount == 0) { + if (jh->b_transaction == NULL && + jh->b_next_transaction == NULL && + jh->b_cp_transaction == NULL) { + J_ASSERT_JH(jh, jh->b_jlist == BJ_None); + J_ASSERT_BH(bh, buffer_jbd(bh)); + J_ASSERT_BH(bh, jh2bh(jh) == bh); + BUFFER_TRACE(bh, "remove journal_head"); + if (jh->b_frozen_data) { + printk(KERN_WARNING "%s: freeing " + "b_frozen_data\n", + __FUNCTION__); + jbd_slab_free(jh->b_frozen_data, bh->b_size); + } + if (jh->b_committed_data) { + printk(KERN_WARNING "%s: freeing " + "b_committed_data\n", + __FUNCTION__); + jbd_slab_free(jh->b_committed_data, bh->b_size); + } + bh->b_private = NULL; + jh->b_bh = NULL; /* debug, really */ + clear_buffer_jbd(bh); + __brelse(bh); + journal_free_journal_head(jh); + } else { + BUFFER_TRACE(bh, "journal_head was locked"); + } + } +} + +/* + * journal_remove_journal_head(): if the buffer isn't attached to a transaction + * and has a zero b_jcount then remove and release its journal_head. If we did + * see that the buffer is not used by any transaction we also "logically" + * decrement ->b_count. + * + * We in fact take an additional increment on ->b_count as a convenience, + * because the caller usually wants to do additional things with the bh + * after calling here. + * The caller of journal_remove_journal_head() *must* run __brelse(bh) at some + * time. Once the caller has run __brelse(), the buffer is eligible for + * reaping by try_to_free_buffers(). + */ +void journal_remove_journal_head(struct buffer_head *bh) +{ + jbd_lock_bh_journal_head(bh); + __journal_remove_journal_head(bh); + jbd_unlock_bh_journal_head(bh); +} + +/* + * Drop a reference on the passed journal_head. If it fell to zero then try to + * release the journal_head from the buffer_head. + */ +void journal_put_journal_head(struct journal_head *jh) +{ + struct buffer_head *bh = jh2bh(jh); + + jbd_lock_bh_journal_head(bh); + J_ASSERT_JH(jh, jh->b_jcount > 0); + --jh->b_jcount; + if (!jh->b_jcount && !jh->b_transaction) { + __journal_remove_journal_head(bh); + __brelse(bh); + } + jbd_unlock_bh_journal_head(bh); +} + +/* + * /proc tunables + */ +#if defined(CONFIG_JBD_DEBUG) +int journal_enable_debug; +EXPORT_SYMBOL(journal_enable_debug); +#endif + +#if defined(CONFIG_JBD_DEBUG) && defined(CONFIG_PROC_FS) + +static struct proc_dir_entry *proc_jbd_debug; + +static int read_jbd_debug(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + int ret; + + ret = sprintf(page + off, "%d\n", journal_enable_debug); + *eof = 1; + return ret; +} + +static int write_jbd_debug(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char buf[32]; + + if (count > ARRAY_SIZE(buf) - 1) + count = ARRAY_SIZE(buf) - 1; + if (copy_from_user(buf, buffer, count)) + return -EFAULT; + buf[ARRAY_SIZE(buf) - 1] = '\0'; + journal_enable_debug = simple_strtoul(buf, NULL, 10); + return count; +} + +#define JBD_PROC_NAME "sys/fs/jbd-debug" + +static void __init create_jbd_proc_entry(void) +{ + proc_jbd_debug = create_proc_entry(JBD_PROC_NAME, 0644, NULL); + if (proc_jbd_debug) { + /* Why is this so hard? */ + proc_jbd_debug->read_proc = read_jbd_debug; + proc_jbd_debug->write_proc = write_jbd_debug; + } +} + +static void __exit remove_jbd_proc_entry(void) +{ + if (proc_jbd_debug) + remove_proc_entry(JBD_PROC_NAME, NULL); +} + +#else + +#define create_jbd_proc_entry() do {} while (0) +#define remove_jbd_proc_entry() do {} while (0) + +#endif + +kmem_cache_t *jbd_handle_cache; + +static int __init journal_init_handle_cache(void) +{ + jbd_handle_cache = kmem_cache_create("journal_handle", + sizeof(handle_t), + 0, /* offset */ + 0, /* flags */ + NULL, /* ctor */ + NULL); /* dtor */ + if (jbd_handle_cache == NULL) { + printk(KERN_EMERG "JBD: failed to create handle cache\n"); + return -ENOMEM; + } + return 0; +} + +static void journal_destroy_handle_cache(void) +{ + if (jbd_handle_cache) + kmem_cache_destroy(jbd_handle_cache); +} + +/* + * Module startup and shutdown + */ + +static int __init journal_init_caches(void) +{ + int ret; + + ret = journal_init_revoke_caches(); + if (ret == 0) + ret = journal_init_journal_head_cache(); + if (ret == 0) + ret = journal_init_handle_cache(); + return ret; +} + +static void journal_destroy_caches(void) +{ + journal_destroy_revoke_caches(); + journal_destroy_journal_head_cache(); + journal_destroy_handle_cache(); + journal_destroy_jbd_slabs(); +} + +static int __init journal_init(void) +{ + int ret; + + BUILD_BUG_ON(sizeof(struct journal_superblock_s) != 1024); + + ret = journal_init_caches(); + if (ret != 0) + journal_destroy_caches(); + create_jbd_proc_entry(); + return ret; +} + +static void __exit journal_exit(void) +{ +#ifdef CONFIG_JBD_DEBUG + int n = atomic_read(&nr_journal_heads); + if (n) + printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n); +#endif + remove_jbd_proc_entry(); + journal_destroy_caches(); +} + +MODULE_LICENSE("GPL"); +module_init(journal_init); +module_exit(journal_exit); + diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c new file mode 100644 index 0000000..11563fe --- /dev/null +++ b/fs/jbd2/recovery.c @@ -0,0 +1,592 @@ +/* + * linux/fs/recovery.c + * + * Written by Stephen C. Tweedie , 1999 + * + * Copyright 1999-2000 Red Hat Software --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Journal recovery routines for the generic filesystem journaling code; + * part of the ext2fs journaling system. + */ + +#ifndef __KERNEL__ +#include "jfs_user.h" +#else +#include +#include +#include +#include +#include +#endif + +/* + * Maintain information about the progress of the recovery job, so that + * the different passes can carry information between them. + */ +struct recovery_info +{ + tid_t start_transaction; + tid_t end_transaction; + + int nr_replays; + int nr_revokes; + int nr_revoke_hits; +}; + +enum passtype {PASS_SCAN, PASS_REVOKE, PASS_REPLAY}; +static int do_one_pass(journal_t *journal, + struct recovery_info *info, enum passtype pass); +static int scan_revoke_records(journal_t *, struct buffer_head *, + tid_t, struct recovery_info *); + +#ifdef __KERNEL__ + +/* Release readahead buffers after use */ +static void journal_brelse_array(struct buffer_head *b[], int n) +{ + while (--n >= 0) + brelse (b[n]); +} + + +/* + * When reading from the journal, we are going through the block device + * layer directly and so there is no readahead being done for us. We + * need to implement any readahead ourselves if we want it to happen at + * all. Recovery is basically one long sequential read, so make sure we + * do the IO in reasonably large chunks. + * + * This is not so critical that we need to be enormously clever about + * the readahead size, though. 128K is a purely arbitrary, good-enough + * fixed value. + */ + +#define MAXBUF 8 +static int do_readahead(journal_t *journal, unsigned int start) +{ + int err; + unsigned int max, nbufs, next; + unsigned long blocknr; + struct buffer_head *bh; + + struct buffer_head * bufs[MAXBUF]; + + /* Do up to 128K of readahead */ + max = start + (128 * 1024 / journal->j_blocksize); + if (max > journal->j_maxlen) + max = journal->j_maxlen; + + /* Do the readahead itself. We'll submit MAXBUF buffer_heads at + * a time to the block device IO layer. */ + + nbufs = 0; + + for (next = start; next < max; next++) { + err = journal_bmap(journal, next, &blocknr); + + if (err) { + printk (KERN_ERR "JBD: bad block at offset %u\n", + next); + goto failed; + } + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + if (!bh) { + err = -ENOMEM; + goto failed; + } + + if (!buffer_uptodate(bh) && !buffer_locked(bh)) { + bufs[nbufs++] = bh; + if (nbufs == MAXBUF) { + ll_rw_block(READ, nbufs, bufs); + journal_brelse_array(bufs, nbufs); + nbufs = 0; + } + } else + brelse(bh); + } + + if (nbufs) + ll_rw_block(READ, nbufs, bufs); + err = 0; + +failed: + if (nbufs) + journal_brelse_array(bufs, nbufs); + return err; +} + +#endif /* __KERNEL__ */ + + +/* + * Read a block from the journal + */ + +static int jread(struct buffer_head **bhp, journal_t *journal, + unsigned int offset) +{ + int err; + unsigned long blocknr; + struct buffer_head *bh; + + *bhp = NULL; + + if (offset >= journal->j_maxlen) { + printk(KERN_ERR "JBD: corrupted journal superblock\n"); + return -EIO; + } + + err = journal_bmap(journal, offset, &blocknr); + + if (err) { + printk (KERN_ERR "JBD: bad block at offset %u\n", + offset); + return err; + } + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + if (!bh) + return -ENOMEM; + + if (!buffer_uptodate(bh)) { + /* If this is a brand new buffer, start readahead. + Otherwise, we assume we are already reading it. */ + if (!buffer_req(bh)) + do_readahead(journal, offset); + wait_on_buffer(bh); + } + + if (!buffer_uptodate(bh)) { + printk (KERN_ERR "JBD: Failed to read block at offset %u\n", + offset); + brelse(bh); + return -EIO; + } + + *bhp = bh; + return 0; +} + + +/* + * Count the number of in-use tags in a journal descriptor block. + */ + +static int count_tags(struct buffer_head *bh, int size) +{ + char * tagp; + journal_block_tag_t * tag; + int nr = 0; + + tagp = &bh->b_data[sizeof(journal_header_t)]; + + while ((tagp - bh->b_data + sizeof(journal_block_tag_t)) <= size) { + tag = (journal_block_tag_t *) tagp; + + nr++; + tagp += sizeof(journal_block_tag_t); + if (!(tag->t_flags & cpu_to_be32(JFS_FLAG_SAME_UUID))) + tagp += 16; + + if (tag->t_flags & cpu_to_be32(JFS_FLAG_LAST_TAG)) + break; + } + + return nr; +} + + +/* Make sure we wrap around the log correctly! */ +#define wrap(journal, var) \ +do { \ + if (var >= (journal)->j_last) \ + var -= ((journal)->j_last - (journal)->j_first); \ +} while (0) + +/** + * journal_recover - recovers a on-disk journal + * @journal: the journal to recover + * + * The primary function for recovering the log contents when mounting a + * journaled device. + * + * Recovery is done in three passes. In the first pass, we look for the + * end of the log. In the second, we assemble the list of revoke + * blocks. In the third and final pass, we replay any un-revoked blocks + * in the log. + */ +int journal_recover(journal_t *journal) +{ + int err; + journal_superblock_t * sb; + + struct recovery_info info; + + memset(&info, 0, sizeof(info)); + sb = journal->j_superblock; + + /* + * The journal superblock's s_start field (the current log head) + * is always zero if, and only if, the journal was cleanly + * unmounted. + */ + + if (!sb->s_start) { + jbd_debug(1, "No recovery required, last transaction %d\n", + be32_to_cpu(sb->s_sequence)); + journal->j_transaction_sequence = be32_to_cpu(sb->s_sequence) + 1; + return 0; + } + + err = do_one_pass(journal, &info, PASS_SCAN); + if (!err) + err = do_one_pass(journal, &info, PASS_REVOKE); + if (!err) + err = do_one_pass(journal, &info, PASS_REPLAY); + + jbd_debug(0, "JBD: recovery, exit status %d, " + "recovered transactions %u to %u\n", + err, info.start_transaction, info.end_transaction); + jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n", + info.nr_replays, info.nr_revoke_hits, info.nr_revokes); + + /* Restart the log at the next transaction ID, thus invalidating + * any existing commit records in the log. */ + journal->j_transaction_sequence = ++info.end_transaction; + + journal_clear_revoke(journal); + sync_blockdev(journal->j_fs_dev); + return err; +} + +/** + * journal_skip_recovery - Start journal and wipe exiting records + * @journal: journal to startup + * + * Locate any valid recovery information from the journal and set up the + * journal structures in memory to ignore it (presumably because the + * caller has evidence that it is out of date). + * This function does'nt appear to be exorted.. + * + * We perform one pass over the journal to allow us to tell the user how + * much recovery information is being erased, and to let us initialise + * the journal transaction sequence numbers to the next unused ID. + */ +int journal_skip_recovery(journal_t *journal) +{ + int err; + journal_superblock_t * sb; + + struct recovery_info info; + + memset (&info, 0, sizeof(info)); + sb = journal->j_superblock; + + err = do_one_pass(journal, &info, PASS_SCAN); + + if (err) { + printk(KERN_ERR "JBD: error %d scanning journal\n", err); + ++journal->j_transaction_sequence; + } else { +#ifdef CONFIG_JBD_DEBUG + int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence); +#endif + jbd_debug(0, + "JBD: ignoring %d transaction%s from the journal.\n", + dropped, (dropped == 1) ? "" : "s"); + journal->j_transaction_sequence = ++info.end_transaction; + } + + journal->j_tail = 0; + return err; +} + +static int do_one_pass(journal_t *journal, + struct recovery_info *info, enum passtype pass) +{ + unsigned int first_commit_ID, next_commit_ID; + unsigned long next_log_block; + int err, success = 0; + journal_superblock_t * sb; + journal_header_t * tmp; + struct buffer_head * bh; + unsigned int sequence; + int blocktype; + + /* Precompute the maximum metadata descriptors in a descriptor block */ + int MAX_BLOCKS_PER_DESC; + MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t)) + / sizeof(journal_block_tag_t)); + + /* + * First thing is to establish what we expect to find in the log + * (in terms of transaction IDs), and where (in terms of log + * block offsets): query the superblock. + */ + + sb = journal->j_superblock; + next_commit_ID = be32_to_cpu(sb->s_sequence); + next_log_block = be32_to_cpu(sb->s_start); + + first_commit_ID = next_commit_ID; + if (pass == PASS_SCAN) + info->start_transaction = first_commit_ID; + + jbd_debug(1, "Starting recovery pass %d\n", pass); + + /* + * Now we walk through the log, transaction by transaction, + * making sure that each transaction has a commit block in the + * expected place. Each complete transaction gets replayed back + * into the main filesystem. + */ + + while (1) { + int flags; + char * tagp; + journal_block_tag_t * tag; + struct buffer_head * obh; + struct buffer_head * nbh; + + cond_resched(); /* We're under lock_kernel() */ + + /* If we already know where to stop the log traversal, + * check right now that we haven't gone past the end of + * the log. */ + + if (pass != PASS_SCAN) + if (tid_geq(next_commit_ID, info->end_transaction)) + break; + + jbd_debug(2, "Scanning for sequence ID %u at %lu/%lu\n", + next_commit_ID, next_log_block, journal->j_last); + + /* Skip over each chunk of the transaction looking + * either the next descriptor block or the final commit + * record. */ + + jbd_debug(3, "JBD: checking block %ld\n", next_log_block); + err = jread(&bh, journal, next_log_block); + if (err) + goto failed; + + next_log_block++; + wrap(journal, next_log_block); + + /* What kind of buffer is it? + * + * If it is a descriptor block, check that it has the + * expected sequence number. Otherwise, we're all done + * here. */ + + tmp = (journal_header_t *)bh->b_data; + + if (tmp->h_magic != cpu_to_be32(JFS_MAGIC_NUMBER)) { + brelse(bh); + break; + } + + blocktype = be32_to_cpu(tmp->h_blocktype); + sequence = be32_to_cpu(tmp->h_sequence); + jbd_debug(3, "Found magic %d, sequence %d\n", + blocktype, sequence); + + if (sequence != next_commit_ID) { + brelse(bh); + break; + } + + /* OK, we have a valid descriptor block which matches + * all of the sequence number checks. What are we going + * to do with it? That depends on the pass... */ + + switch(blocktype) { + case JFS_DESCRIPTOR_BLOCK: + /* If it is a valid descriptor block, replay it + * in pass REPLAY; otherwise, just skip over the + * blocks it describes. */ + if (pass != PASS_REPLAY) { + next_log_block += + count_tags(bh, journal->j_blocksize); + wrap(journal, next_log_block); + brelse(bh); + continue; + } + + /* A descriptor block: we can now write all of + * the data blocks. Yay, useful work is finally + * getting done here! */ + + tagp = &bh->b_data[sizeof(journal_header_t)]; + while ((tagp - bh->b_data +sizeof(journal_block_tag_t)) + <= journal->j_blocksize) { + unsigned long io_block; + + tag = (journal_block_tag_t *) tagp; + flags = be32_to_cpu(tag->t_flags); + + io_block = next_log_block++; + wrap(journal, next_log_block); + err = jread(&obh, journal, io_block); + if (err) { + /* Recover what we can, but + * report failure at the end. */ + success = err; + printk (KERN_ERR + "JBD: IO error %d recovering " + "block %ld in log\n", + err, io_block); + } else { + unsigned long blocknr; + + J_ASSERT(obh != NULL); + blocknr = be32_to_cpu(tag->t_blocknr); + + /* If the block has been + * revoked, then we're all done + * here. */ + if (journal_test_revoke + (journal, blocknr, + next_commit_ID)) { + brelse(obh); + ++info->nr_revoke_hits; + goto skip_write; + } + + /* Find a buffer for the new + * data being restored */ + nbh = __getblk(journal->j_fs_dev, + blocknr, + journal->j_blocksize); + if (nbh == NULL) { + printk(KERN_ERR + "JBD: Out of memory " + "during recovery.\n"); + err = -ENOMEM; + brelse(bh); + brelse(obh); + goto failed; + } + + lock_buffer(nbh); + memcpy(nbh->b_data, obh->b_data, + journal->j_blocksize); + if (flags & JFS_FLAG_ESCAPE) { + *((__be32 *)bh->b_data) = + cpu_to_be32(JFS_MAGIC_NUMBER); + } + + BUFFER_TRACE(nbh, "marking dirty"); + set_buffer_uptodate(nbh); + mark_buffer_dirty(nbh); + BUFFER_TRACE(nbh, "marking uptodate"); + ++info->nr_replays; + /* ll_rw_block(WRITE, 1, &nbh); */ + unlock_buffer(nbh); + brelse(obh); + brelse(nbh); + } + + skip_write: + tagp += sizeof(journal_block_tag_t); + if (!(flags & JFS_FLAG_SAME_UUID)) + tagp += 16; + + if (flags & JFS_FLAG_LAST_TAG) + break; + } + + brelse(bh); + continue; + + case JFS_COMMIT_BLOCK: + /* Found an expected commit block: not much to + * do other than move on to the next sequence + * number. */ + brelse(bh); + next_commit_ID++; + continue; + + case JFS_REVOKE_BLOCK: + /* If we aren't in the REVOKE pass, then we can + * just skip over this block. */ + if (pass != PASS_REVOKE) { + brelse(bh); + continue; + } + + err = scan_revoke_records(journal, bh, + next_commit_ID, info); + brelse(bh); + if (err) + goto failed; + continue; + + default: + jbd_debug(3, "Unrecognised magic %d, end of scan.\n", + blocktype); + brelse(bh); + goto done; + } + } + + done: + /* + * We broke out of the log scan loop: either we came to the + * known end of the log or we found an unexpected block in the + * log. If the latter happened, then we know that the "current" + * transaction marks the end of the valid log. + */ + + if (pass == PASS_SCAN) + info->end_transaction = next_commit_ID; + else { + /* It's really bad news if different passes end up at + * different places (but possible due to IO errors). */ + if (info->end_transaction != next_commit_ID) { + printk (KERN_ERR "JBD: recovery pass %d ended at " + "transaction %u, expected %u\n", + pass, next_commit_ID, info->end_transaction); + if (!success) + success = -EIO; + } + } + + return success; + + failed: + return err; +} + + +/* Scan a revoke record, marking all blocks mentioned as revoked. */ + +static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, + tid_t sequence, struct recovery_info *info) +{ + journal_revoke_header_t *header; + int offset, max; + + header = (journal_revoke_header_t *) bh->b_data; + offset = sizeof(journal_revoke_header_t); + max = be32_to_cpu(header->r_count); + + while (offset < max) { + unsigned long blocknr; + int err; + + blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset))); + offset += 4; + err = journal_set_revoke(journal, blocknr, sequence); + if (err) + return err; + ++info->nr_revokes; + } + return 0; +} diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c new file mode 100644 index 0000000..c532429 --- /dev/null +++ b/fs/jbd2/revoke.c @@ -0,0 +1,703 @@ +/* + * linux/fs/revoke.c + * + * Written by Stephen C. Tweedie , 2000 + * + * Copyright 2000 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Journal revoke routines for the generic filesystem journaling code; + * part of the ext2fs journaling system. + * + * Revoke is the mechanism used to prevent old log records for deleted + * metadata from being replayed on top of newer data using the same + * blocks. The revoke mechanism is used in two separate places: + * + * + Commit: during commit we write the entire list of the current + * transaction's revoked blocks to the journal + * + * + Recovery: during recovery we record the transaction ID of all + * revoked blocks. If there are multiple revoke records in the log + * for a single block, only the last one counts, and if there is a log + * entry for a block beyond the last revoke, then that log entry still + * gets replayed. + * + * We can get interactions between revokes and new log data within a + * single transaction: + * + * Block is revoked and then journaled: + * The desired end result is the journaling of the new block, so we + * cancel the revoke before the transaction commits. + * + * Block is journaled and then revoked: + * The revoke must take precedence over the write of the block, so we + * need either to cancel the journal entry or to write the revoke + * later in the log than the log block. In this case, we choose the + * latter: journaling a block cancels any revoke record for that block + * in the current transaction, so any revoke for that block in the + * transaction must have happened after the block was journaled and so + * the revoke must take precedence. + * + * Block is revoked and then written as data: + * The data write is allowed to succeed, but the revoke is _not_ + * cancelled. We still need to prevent old log records from + * overwriting the new data. We don't even need to clear the revoke + * bit here. + * + * Revoke information on buffers is a tri-state value: + * + * RevokeValid clear: no cached revoke status, need to look it up + * RevokeValid set, Revoked clear: + * buffer has not been revoked, and cancel_revoke + * need do nothing. + * RevokeValid set, Revoked set: + * buffer has been revoked. + */ + +#ifndef __KERNEL__ +#include "jfs_user.h" +#else +#include +#include +#include +#include +#include +#include +#include +#include +#endif + +static kmem_cache_t *revoke_record_cache; +static kmem_cache_t *revoke_table_cache; + +/* Each revoke record represents one single revoked block. During + journal replay, this involves recording the transaction ID of the + last transaction to revoke this block. */ + +struct jbd_revoke_record_s +{ + struct list_head hash; + tid_t sequence; /* Used for recovery only */ + unsigned long blocknr; +}; + + +/* The revoke table is just a simple hash table of revoke records. */ +struct jbd_revoke_table_s +{ + /* It is conceivable that we might want a larger hash table + * for recovery. Must be a power of two. */ + int hash_size; + int hash_shift; + struct list_head *hash_table; +}; + + +#ifdef __KERNEL__ +static void write_one_revoke_record(journal_t *, transaction_t *, + struct journal_head **, int *, + struct jbd_revoke_record_s *); +static void flush_descriptor(journal_t *, struct journal_head *, int); +#endif + +/* Utility functions to maintain the revoke table */ + +/* Borrowed from buffer.c: this is a tried and tested block hash function */ +static inline int hash(journal_t *journal, unsigned long block) +{ + struct jbd_revoke_table_s *table = journal->j_revoke; + int hash_shift = table->hash_shift; + + return ((block << (hash_shift - 6)) ^ + (block >> 13) ^ + (block << (hash_shift - 12))) & (table->hash_size - 1); +} + +static int insert_revoke_hash(journal_t *journal, unsigned long blocknr, + tid_t seq) +{ + struct list_head *hash_list; + struct jbd_revoke_record_s *record; + +repeat: + record = kmem_cache_alloc(revoke_record_cache, GFP_NOFS); + if (!record) + goto oom; + + record->sequence = seq; + record->blocknr = blocknr; + hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)]; + spin_lock(&journal->j_revoke_lock); + list_add(&record->hash, hash_list); + spin_unlock(&journal->j_revoke_lock); + return 0; + +oom: + if (!journal_oom_retry) + return -ENOMEM; + jbd_debug(1, "ENOMEM in %s, retrying\n", __FUNCTION__); + yield(); + goto repeat; +} + +/* Find a revoke record in the journal's hash table. */ + +static struct jbd_revoke_record_s *find_revoke_record(journal_t *journal, + unsigned long blocknr) +{ + struct list_head *hash_list; + struct jbd_revoke_record_s *record; + + hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)]; + + spin_lock(&journal->j_revoke_lock); + record = (struct jbd_revoke_record_s *) hash_list->next; + while (&(record->hash) != hash_list) { + if (record->blocknr == blocknr) { + spin_unlock(&journal->j_revoke_lock); + return record; + } + record = (struct jbd_revoke_record_s *) record->hash.next; + } + spin_unlock(&journal->j_revoke_lock); + return NULL; +} + +int __init journal_init_revoke_caches(void) +{ + revoke_record_cache = kmem_cache_create("revoke_record", + sizeof(struct jbd_revoke_record_s), + 0, SLAB_HWCACHE_ALIGN, NULL, NULL); + if (revoke_record_cache == 0) + return -ENOMEM; + + revoke_table_cache = kmem_cache_create("revoke_table", + sizeof(struct jbd_revoke_table_s), + 0, 0, NULL, NULL); + if (revoke_table_cache == 0) { + kmem_cache_destroy(revoke_record_cache); + revoke_record_cache = NULL; + return -ENOMEM; + } + return 0; +} + +void journal_destroy_revoke_caches(void) +{ + kmem_cache_destroy(revoke_record_cache); + revoke_record_cache = NULL; + kmem_cache_destroy(revoke_table_cache); + revoke_table_cache = NULL; +} + +/* Initialise the revoke table for a given journal to a given size. */ + +int journal_init_revoke(journal_t *journal, int hash_size) +{ + int shift, tmp; + + J_ASSERT (journal->j_revoke_table[0] == NULL); + + shift = 0; + tmp = hash_size; + while((tmp >>= 1UL) != 0UL) + shift++; + + journal->j_revoke_table[0] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL); + if (!journal->j_revoke_table[0]) + return -ENOMEM; + journal->j_revoke = journal->j_revoke_table[0]; + + /* Check that the hash_size is a power of two */ + J_ASSERT ((hash_size & (hash_size-1)) == 0); + + journal->j_revoke->hash_size = hash_size; + + journal->j_revoke->hash_shift = shift; + + journal->j_revoke->hash_table = + kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); + if (!journal->j_revoke->hash_table) { + kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]); + journal->j_revoke = NULL; + return -ENOMEM; + } + + for (tmp = 0; tmp < hash_size; tmp++) + INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); + + journal->j_revoke_table[1] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL); + if (!journal->j_revoke_table[1]) { + kfree(journal->j_revoke_table[0]->hash_table); + kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]); + return -ENOMEM; + } + + journal->j_revoke = journal->j_revoke_table[1]; + + /* Check that the hash_size is a power of two */ + J_ASSERT ((hash_size & (hash_size-1)) == 0); + + journal->j_revoke->hash_size = hash_size; + + journal->j_revoke->hash_shift = shift; + + journal->j_revoke->hash_table = + kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); + if (!journal->j_revoke->hash_table) { + kfree(journal->j_revoke_table[0]->hash_table); + kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]); + kmem_cache_free(revoke_table_cache, journal->j_revoke_table[1]); + journal->j_revoke = NULL; + return -ENOMEM; + } + + for (tmp = 0; tmp < hash_size; tmp++) + INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); + + spin_lock_init(&journal->j_revoke_lock); + + return 0; +} + +/* Destoy a journal's revoke table. The table must already be empty! */ + +void journal_destroy_revoke(journal_t *journal) +{ + struct jbd_revoke_table_s *table; + struct list_head *hash_list; + int i; + + table = journal->j_revoke_table[0]; + if (!table) + return; + + for (i=0; ihash_size; i++) { + hash_list = &table->hash_table[i]; + J_ASSERT (list_empty(hash_list)); + } + + kfree(table->hash_table); + kmem_cache_free(revoke_table_cache, table); + journal->j_revoke = NULL; + + table = journal->j_revoke_table[1]; + if (!table) + return; + + for (i=0; ihash_size; i++) { + hash_list = &table->hash_table[i]; + J_ASSERT (list_empty(hash_list)); + } + + kfree(table->hash_table); + kmem_cache_free(revoke_table_cache, table); + journal->j_revoke = NULL; +} + + +#ifdef __KERNEL__ + +/* + * journal_revoke: revoke a given buffer_head from the journal. This + * prevents the block from being replayed during recovery if we take a + * crash after this current transaction commits. Any subsequent + * metadata writes of the buffer in this transaction cancel the + * revoke. + * + * Note that this call may block --- it is up to the caller to make + * sure that there are no further calls to journal_write_metadata + * before the revoke is complete. In ext3, this implies calling the + * revoke before clearing the block bitmap when we are deleting + * metadata. + * + * Revoke performs a journal_forget on any buffer_head passed in as a + * parameter, but does _not_ forget the buffer_head if the bh was only + * found implicitly. + * + * bh_in may not be a journalled buffer - it may have come off + * the hash tables without an attached journal_head. + * + * If bh_in is non-zero, journal_revoke() will decrement its b_count + * by one. + */ + +int journal_revoke(handle_t *handle, unsigned long blocknr, + struct buffer_head *bh_in) +{ + struct buffer_head *bh = NULL; + journal_t *journal; + struct block_device *bdev; + int err; + + might_sleep(); + if (bh_in) + BUFFER_TRACE(bh_in, "enter"); + + journal = handle->h_transaction->t_journal; + if (!journal_set_features(journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)){ + J_ASSERT (!"Cannot set revoke feature!"); + return -EINVAL; + } + + bdev = journal->j_fs_dev; + bh = bh_in; + + if (!bh) { + bh = __find_get_block(bdev, blocknr, journal->j_blocksize); + if (bh) + BUFFER_TRACE(bh, "found on hash"); + } +#ifdef JBD_EXPENSIVE_CHECKING + else { + struct buffer_head *bh2; + + /* If there is a different buffer_head lying around in + * memory anywhere... */ + bh2 = __find_get_block(bdev, blocknr, journal->j_blocksize); + if (bh2) { + /* ... and it has RevokeValid status... */ + if (bh2 != bh && buffer_revokevalid(bh2)) + /* ...then it better be revoked too, + * since it's illegal to create a revoke + * record against a buffer_head which is + * not marked revoked --- that would + * risk missing a subsequent revoke + * cancel. */ + J_ASSERT_BH(bh2, buffer_revoked(bh2)); + put_bh(bh2); + } + } +#endif + + /* We really ought not ever to revoke twice in a row without + first having the revoke cancelled: it's illegal to free a + block twice without allocating it in between! */ + if (bh) { + if (!J_EXPECT_BH(bh, !buffer_revoked(bh), + "inconsistent data on disk")) { + if (!bh_in) + brelse(bh); + return -EIO; + } + set_buffer_revoked(bh); + set_buffer_revokevalid(bh); + if (bh_in) { + BUFFER_TRACE(bh_in, "call journal_forget"); + journal_forget(handle, bh_in); + } else { + BUFFER_TRACE(bh, "call brelse"); + __brelse(bh); + } + } + + jbd_debug(2, "insert revoke for block %lu, bh_in=%p\n", blocknr, bh_in); + err = insert_revoke_hash(journal, blocknr, + handle->h_transaction->t_tid); + BUFFER_TRACE(bh_in, "exit"); + return err; +} + +/* + * Cancel an outstanding revoke. For use only internally by the + * journaling code (called from journal_get_write_access). + * + * We trust buffer_revoked() on the buffer if the buffer is already + * being journaled: if there is no revoke pending on the buffer, then we + * don't do anything here. + * + * This would break if it were possible for a buffer to be revoked and + * discarded, and then reallocated within the same transaction. In such + * a case we would have lost the revoked bit, but when we arrived here + * the second time we would still have a pending revoke to cancel. So, + * do not trust the Revoked bit on buffers unless RevokeValid is also + * set. + * + * The caller must have the journal locked. + */ +int journal_cancel_revoke(handle_t *handle, struct journal_head *jh) +{ + struct jbd_revoke_record_s *record; + journal_t *journal = handle->h_transaction->t_journal; + int need_cancel; + int did_revoke = 0; /* akpm: debug */ + struct buffer_head *bh = jh2bh(jh); + + jbd_debug(4, "journal_head %p, cancelling revoke\n", jh); + + /* Is the existing Revoke bit valid? If so, we trust it, and + * only perform the full cancel if the revoke bit is set. If + * not, we can't trust the revoke bit, and we need to do the + * full search for a revoke record. */ + if (test_set_buffer_revokevalid(bh)) { + need_cancel = test_clear_buffer_revoked(bh); + } else { + need_cancel = 1; + clear_buffer_revoked(bh); + } + + if (need_cancel) { + record = find_revoke_record(journal, bh->b_blocknr); + if (record) { + jbd_debug(4, "cancelled existing revoke on " + "blocknr %llu\n", (unsigned long long)bh->b_blocknr); + spin_lock(&journal->j_revoke_lock); + list_del(&record->hash); + spin_unlock(&journal->j_revoke_lock); + kmem_cache_free(revoke_record_cache, record); + did_revoke = 1; + } + } + +#ifdef JBD_EXPENSIVE_CHECKING + /* There better not be one left behind by now! */ + record = find_revoke_record(journal, bh->b_blocknr); + J_ASSERT_JH(jh, record == NULL); +#endif + + /* Finally, have we just cleared revoke on an unhashed + * buffer_head? If so, we'd better make sure we clear the + * revoked status on any hashed alias too, otherwise the revoke + * state machine will get very upset later on. */ + if (need_cancel) { + struct buffer_head *bh2; + bh2 = __find_get_block(bh->b_bdev, bh->b_blocknr, bh->b_size); + if (bh2) { + if (bh2 != bh) + clear_buffer_revoked(bh2); + __brelse(bh2); + } + } + return did_revoke; +} + +/* journal_switch_revoke table select j_revoke for next transaction + * we do not want to suspend any processing until all revokes are + * written -bzzz + */ +void journal_switch_revoke_table(journal_t *journal) +{ + int i; + + if (journal->j_revoke == journal->j_revoke_table[0]) + journal->j_revoke = journal->j_revoke_table[1]; + else + journal->j_revoke = journal->j_revoke_table[0]; + + for (i = 0; i < journal->j_revoke->hash_size; i++) + INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]); +} + +/* + * Write revoke records to the journal for all entries in the current + * revoke hash, deleting the entries as we go. + * + * Called with the journal lock held. + */ + +void journal_write_revoke_records(journal_t *journal, + transaction_t *transaction) +{ + struct journal_head *descriptor; + struct jbd_revoke_record_s *record; + struct jbd_revoke_table_s *revoke; + struct list_head *hash_list; + int i, offset, count; + + descriptor = NULL; + offset = 0; + count = 0; + + /* select revoke table for committing transaction */ + revoke = journal->j_revoke == journal->j_revoke_table[0] ? + journal->j_revoke_table[1] : journal->j_revoke_table[0]; + + for (i = 0; i < revoke->hash_size; i++) { + hash_list = &revoke->hash_table[i]; + + while (!list_empty(hash_list)) { + record = (struct jbd_revoke_record_s *) + hash_list->next; + write_one_revoke_record(journal, transaction, + &descriptor, &offset, + record); + count++; + list_del(&record->hash); + kmem_cache_free(revoke_record_cache, record); + } + } + if (descriptor) + flush_descriptor(journal, descriptor, offset); + jbd_debug(1, "Wrote %d revoke records\n", count); +} + +/* + * Write out one revoke record. We need to create a new descriptor + * block if the old one is full or if we have not already created one. + */ + +static void write_one_revoke_record(journal_t *journal, + transaction_t *transaction, + struct journal_head **descriptorp, + int *offsetp, + struct jbd_revoke_record_s *record) +{ + struct journal_head *descriptor; + int offset; + journal_header_t *header; + + /* If we are already aborting, this all becomes a noop. We + still need to go round the loop in + journal_write_revoke_records in order to free all of the + revoke records: only the IO to the journal is omitted. */ + if (is_journal_aborted(journal)) + return; + + descriptor = *descriptorp; + offset = *offsetp; + + /* Make sure we have a descriptor with space left for the record */ + if (descriptor) { + if (offset == journal->j_blocksize) { + flush_descriptor(journal, descriptor, offset); + descriptor = NULL; + } + } + + if (!descriptor) { + descriptor = journal_get_descriptor_buffer(journal); + if (!descriptor) + return; + header = (journal_header_t *) &jh2bh(descriptor)->b_data[0]; + header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JFS_REVOKE_BLOCK); + header->h_sequence = cpu_to_be32(transaction->t_tid); + + /* Record it so that we can wait for IO completion later */ + JBUFFER_TRACE(descriptor, "file as BJ_LogCtl"); + journal_file_buffer(descriptor, transaction, BJ_LogCtl); + + offset = sizeof(journal_revoke_header_t); + *descriptorp = descriptor; + } + + * ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = + cpu_to_be32(record->blocknr); + offset += 4; + *offsetp = offset; +} + +/* + * Flush a revoke descriptor out to the journal. If we are aborting, + * this is a noop; otherwise we are generating a buffer which needs to + * be waited for during commit, so it has to go onto the appropriate + * journal buffer list. + */ + +static void flush_descriptor(journal_t *journal, + struct journal_head *descriptor, + int offset) +{ + journal_revoke_header_t *header; + struct buffer_head *bh = jh2bh(descriptor); + + if (is_journal_aborted(journal)) { + put_bh(bh); + return; + } + + header = (journal_revoke_header_t *) jh2bh(descriptor)->b_data; + header->r_count = cpu_to_be32(offset); + set_buffer_jwrite(bh); + BUFFER_TRACE(bh, "write"); + set_buffer_dirty(bh); + ll_rw_block(SWRITE, 1, &bh); +} +#endif + +/* + * Revoke support for recovery. + * + * Recovery needs to be able to: + * + * record all revoke records, including the tid of the latest instance + * of each revoke in the journal + * + * check whether a given block in a given transaction should be replayed + * (ie. has not been revoked by a revoke record in that or a subsequent + * transaction) + * + * empty the revoke table after recovery. + */ + +/* + * First, setting revoke records. We create a new revoke record for + * every block ever revoked in the log as we scan it for recovery, and + * we update the existing records if we find multiple revokes for a + * single block. + */ + +int journal_set_revoke(journal_t *journal, + unsigned long blocknr, + tid_t sequence) +{ + struct jbd_revoke_record_s *record; + + record = find_revoke_record(journal, blocknr); + if (record) { + /* If we have multiple occurrences, only record the + * latest sequence number in the hashed record */ + if (tid_gt(sequence, record->sequence)) + record->sequence = sequence; + return 0; + } + return insert_revoke_hash(journal, blocknr, sequence); +} + +/* + * Test revoke records. For a given block referenced in the log, has + * that block been revoked? A revoke record with a given transaction + * sequence number revokes all blocks in that transaction and earlier + * ones, but later transactions still need replayed. + */ + +int journal_test_revoke(journal_t *journal, + unsigned long blocknr, + tid_t sequence) +{ + struct jbd_revoke_record_s *record; + + record = find_revoke_record(journal, blocknr); + if (!record) + return 0; + if (tid_gt(sequence, record->sequence)) + return 0; + return 1; +} + +/* + * Finally, once recovery is over, we need to clear the revoke table so + * that it can be reused by the running filesystem. + */ + +void journal_clear_revoke(journal_t *journal) +{ + int i; + struct list_head *hash_list; + struct jbd_revoke_record_s *record; + struct jbd_revoke_table_s *revoke; + + revoke = journal->j_revoke; + + for (i = 0; i < revoke->hash_size; i++) { + hash_list = &revoke->hash_table[i]; + while (!list_empty(hash_list)) { + record = (struct jbd_revoke_record_s*) hash_list->next; + list_del(&record->hash); + kmem_cache_free(revoke_record_cache, record); + } + } +} diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c new file mode 100644 index 0000000..e1b3c8a --- /dev/null +++ b/fs/jbd2/transaction.c @@ -0,0 +1,2080 @@ +/* + * linux/fs/transaction.c + * + * Written by Stephen C. Tweedie , 1998 + * + * Copyright 1998 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Generic filesystem transaction handling code; part of the ext2fs + * journaling system. + * + * This file manages transactions (compound commits managed by the + * journaling code) and handles (individual atomic operations by the + * filesystem). + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * get_transaction: obtain a new transaction_t object. + * + * Simply allocate and initialise a new transaction. Create it in + * RUNNING state and add it to the current journal (which should not + * have an existing running transaction: we only make a new transaction + * once we have started to commit the old one). + * + * Preconditions: + * The journal MUST be locked. We don't perform atomic mallocs on the + * new transaction and we can't block without protecting against other + * processes trying to touch the journal while it is in transition. + * + * Called under j_state_lock + */ + +static transaction_t * +get_transaction(journal_t *journal, transaction_t *transaction) +{ + transaction->t_journal = journal; + transaction->t_state = T_RUNNING; + transaction->t_tid = journal->j_transaction_sequence++; + transaction->t_expires = jiffies + journal->j_commit_interval; + spin_lock_init(&transaction->t_handle_lock); + + /* Set up the commit timer for the new transaction. */ + journal->j_commit_timer.expires = transaction->t_expires; + add_timer(&journal->j_commit_timer); + + J_ASSERT(journal->j_running_transaction == NULL); + journal->j_running_transaction = transaction; + + return transaction; +} + +/* + * Handle management. + * + * A handle_t is an object which represents a single atomic update to a + * filesystem, and which tracks all of the modifications which form part + * of that one update. + */ + +/* + * start_this_handle: Given a handle, deal with any locking or stalling + * needed to make sure that there is enough journal space for the handle + * to begin. Attach the handle to a transaction and set up the + * transaction's buffer credits. + */ + +static int start_this_handle(journal_t *journal, handle_t *handle) +{ + transaction_t *transaction; + int needed; + int nblocks = handle->h_buffer_credits; + transaction_t *new_transaction = NULL; + int ret = 0; + + if (nblocks > journal->j_max_transaction_buffers) { + printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n", + current->comm, nblocks, + journal->j_max_transaction_buffers); + ret = -ENOSPC; + goto out; + } + +alloc_transaction: + if (!journal->j_running_transaction) { + new_transaction = jbd_kmalloc(sizeof(*new_transaction), + GFP_NOFS); + if (!new_transaction) { + ret = -ENOMEM; + goto out; + } + memset(new_transaction, 0, sizeof(*new_transaction)); + } + + jbd_debug(3, "New handle %p going live.\n", handle); + +repeat: + + /* + * We need to hold j_state_lock until t_updates has been incremented, + * for proper journal barrier handling + */ + spin_lock(&journal->j_state_lock); +repeat_locked: + if (is_journal_aborted(journal) || + (journal->j_errno != 0 && !(journal->j_flags & JFS_ACK_ERR))) { + spin_unlock(&journal->j_state_lock); + ret = -EROFS; + goto out; + } + + /* Wait on the journal's transaction barrier if necessary */ + if (journal->j_barrier_count) { + spin_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_transaction_locked, + journal->j_barrier_count == 0); + goto repeat; + } + + if (!journal->j_running_transaction) { + if (!new_transaction) { + spin_unlock(&journal->j_state_lock); + goto alloc_transaction; + } + get_transaction(journal, new_transaction); + new_transaction = NULL; + } + + transaction = journal->j_running_transaction; + + /* + * If the current transaction is locked down for commit, wait for the + * lock to be released. + */ + if (transaction->t_state == T_LOCKED) { + DEFINE_WAIT(wait); + + prepare_to_wait(&journal->j_wait_transaction_locked, + &wait, TASK_UNINTERRUPTIBLE); + spin_unlock(&journal->j_state_lock); + schedule(); + finish_wait(&journal->j_wait_transaction_locked, &wait); + goto repeat; + } + + /* + * If there is not enough space left in the log to write all potential + * buffers requested by this operation, we need to stall pending a log + * checkpoint to free some more log space. + */ + spin_lock(&transaction->t_handle_lock); + needed = transaction->t_outstanding_credits + nblocks; + + if (needed > journal->j_max_transaction_buffers) { + /* + * If the current transaction is already too large, then start + * to commit it: we can then go back and attach this handle to + * a new transaction. + */ + DEFINE_WAIT(wait); + + jbd_debug(2, "Handle %p starting new commit...\n", handle); + spin_unlock(&transaction->t_handle_lock); + prepare_to_wait(&journal->j_wait_transaction_locked, &wait, + TASK_UNINTERRUPTIBLE); + __log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); + schedule(); + finish_wait(&journal->j_wait_transaction_locked, &wait); + goto repeat; + } + + /* + * The commit code assumes that it can get enough log space + * without forcing a checkpoint. This is *critical* for + * correctness: a checkpoint of a buffer which is also + * associated with a committing transaction creates a deadlock, + * so commit simply cannot force through checkpoints. + * + * We must therefore ensure the necessary space in the journal + * *before* starting to dirty potentially checkpointed buffers + * in the new transaction. + * + * The worst part is, any transaction currently committing can + * reduce the free space arbitrarily. Be careful to account for + * those buffers when checkpointing. + */ + + /* + * @@@ AKPM: This seems rather over-defensive. We're giving commit + * a _lot_ of headroom: 1/4 of the journal plus the size of + * the committing transaction. Really, we only need to give it + * committing_transaction->t_outstanding_credits plus "enough" for + * the log control blocks. + * Also, this test is inconsitent with the matching one in + * journal_extend(). + */ + if (__log_space_left(journal) < jbd_space_needed(journal)) { + jbd_debug(2, "Handle %p waiting for checkpoint...\n", handle); + spin_unlock(&transaction->t_handle_lock); + __log_wait_for_space(journal); + goto repeat_locked; + } + + /* OK, account for the buffers that this operation expects to + * use and add the handle to the running transaction. */ + + handle->h_transaction = transaction; + transaction->t_outstanding_credits += nblocks; + transaction->t_updates++; + transaction->t_handle_count++; + jbd_debug(4, "Handle %p given %d credits (total %d, free %d)\n", + handle, nblocks, transaction->t_outstanding_credits, + __log_space_left(journal)); + spin_unlock(&transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); +out: + if (unlikely(new_transaction)) /* It's usually NULL */ + kfree(new_transaction); + return ret; +} + +/* Allocate a new handle. This should probably be in a slab... */ +static handle_t *new_handle(int nblocks) +{ + handle_t *handle = jbd_alloc_handle(GFP_NOFS); + if (!handle) + return NULL; + memset(handle, 0, sizeof(*handle)); + handle->h_buffer_credits = nblocks; + handle->h_ref = 1; + + return handle; +} + +/** + * handle_t *journal_start() - Obtain a new handle. + * @journal: Journal to start transaction on. + * @nblocks: number of block buffer we might modify + * + * We make sure that the transaction can guarantee at least nblocks of + * modified buffers in the log. We block until the log can guarantee + * that much space. + * + * This function is visible to journal users (like ext3fs), so is not + * called with the journal already locked. + * + * Return a pointer to a newly allocated handle, or NULL on failure + */ +handle_t *journal_start(journal_t *journal, int nblocks) +{ + handle_t *handle = journal_current_handle(); + int err; + + if (!journal) + return ERR_PTR(-EROFS); + + if (handle) { + J_ASSERT(handle->h_transaction->t_journal == journal); + handle->h_ref++; + return handle; + } + + handle = new_handle(nblocks); + if (!handle) + return ERR_PTR(-ENOMEM); + + current->journal_info = handle; + + err = start_this_handle(journal, handle); + if (err < 0) { + jbd_free_handle(handle); + current->journal_info = NULL; + handle = ERR_PTR(err); + } + return handle; +} + +/** + * int journal_extend() - extend buffer credits. + * @handle: handle to 'extend' + * @nblocks: nr blocks to try to extend by. + * + * Some transactions, such as large extends and truncates, can be done + * atomically all at once or in several stages. The operation requests + * a credit for a number of buffer modications in advance, but can + * extend its credit if it needs more. + * + * journal_extend tries to give the running handle more buffer credits. + * It does not guarantee that allocation - this is a best-effort only. + * The calling process MUST be able to deal cleanly with a failure to + * extend here. + * + * Return 0 on success, non-zero on failure. + * + * return code < 0 implies an error + * return code > 0 implies normal transaction-full status. + */ +int journal_extend(handle_t *handle, int nblocks) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + int result; + int wanted; + + result = -EIO; + if (is_handle_aborted(handle)) + goto out; + + result = 1; + + spin_lock(&journal->j_state_lock); + + /* Don't extend a locked-down transaction! */ + if (handle->h_transaction->t_state != T_RUNNING) { + jbd_debug(3, "denied handle %p %d blocks: " + "transaction not running\n", handle, nblocks); + goto error_out; + } + + spin_lock(&transaction->t_handle_lock); + wanted = transaction->t_outstanding_credits + nblocks; + + if (wanted > journal->j_max_transaction_buffers) { + jbd_debug(3, "denied handle %p %d blocks: " + "transaction too large\n", handle, nblocks); + goto unlock; + } + + if (wanted > __log_space_left(journal)) { + jbd_debug(3, "denied handle %p %d blocks: " + "insufficient log space\n", handle, nblocks); + goto unlock; + } + + handle->h_buffer_credits += nblocks; + transaction->t_outstanding_credits += nblocks; + result = 0; + + jbd_debug(3, "extended handle %p by %d\n", handle, nblocks); +unlock: + spin_unlock(&transaction->t_handle_lock); +error_out: + spin_unlock(&journal->j_state_lock); +out: + return result; +} + + +/** + * int journal_restart() - restart a handle . + * @handle: handle to restart + * @nblocks: nr credits requested + * + * Restart a handle for a multi-transaction filesystem + * operation. + * + * If the journal_extend() call above fails to grant new buffer credits + * to a running handle, a call to journal_restart will commit the + * handle's transaction so far and reattach the handle to a new + * transaction capabable of guaranteeing the requested number of + * credits. + */ + +int journal_restart(handle_t *handle, int nblocks) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + int ret; + + /* If we've had an abort of any type, don't even think about + * actually doing the restart! */ + if (is_handle_aborted(handle)) + return 0; + + /* + * First unlink the handle from its current transaction, and start the + * commit on that. + */ + J_ASSERT(transaction->t_updates > 0); + J_ASSERT(journal_current_handle() == handle); + + spin_lock(&journal->j_state_lock); + spin_lock(&transaction->t_handle_lock); + transaction->t_outstanding_credits -= handle->h_buffer_credits; + transaction->t_updates--; + + if (!transaction->t_updates) + wake_up(&journal->j_wait_updates); + spin_unlock(&transaction->t_handle_lock); + + jbd_debug(2, "restarting handle %p\n", handle); + __log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); + + handle->h_buffer_credits = nblocks; + ret = start_this_handle(journal, handle); + return ret; +} + + +/** + * void journal_lock_updates () - establish a transaction barrier. + * @journal: Journal to establish a barrier on. + * + * This locks out any further updates from being started, and blocks + * until all existing updates have completed, returning only once the + * journal is in a quiescent state with no updates running. + * + * The journal lock should not be held on entry. + */ +void journal_lock_updates(journal_t *journal) +{ + DEFINE_WAIT(wait); + + spin_lock(&journal->j_state_lock); + ++journal->j_barrier_count; + + /* Wait until there are no running updates */ + while (1) { + transaction_t *transaction = journal->j_running_transaction; + + if (!transaction) + break; + + spin_lock(&transaction->t_handle_lock); + if (!transaction->t_updates) { + spin_unlock(&transaction->t_handle_lock); + break; + } + prepare_to_wait(&journal->j_wait_updates, &wait, + TASK_UNINTERRUPTIBLE); + spin_unlock(&transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); + schedule(); + finish_wait(&journal->j_wait_updates, &wait); + spin_lock(&journal->j_state_lock); + } + spin_unlock(&journal->j_state_lock); + + /* + * We have now established a barrier against other normal updates, but + * we also need to barrier against other journal_lock_updates() calls + * to make sure that we serialise special journal-locked operations + * too. + */ + mutex_lock(&journal->j_barrier); +} + +/** + * void journal_unlock_updates (journal_t* journal) - release barrier + * @journal: Journal to release the barrier on. + * + * Release a transaction barrier obtained with journal_lock_updates(). + * + * Should be called without the journal lock held. + */ +void journal_unlock_updates (journal_t *journal) +{ + J_ASSERT(journal->j_barrier_count != 0); + + mutex_unlock(&journal->j_barrier); + spin_lock(&journal->j_state_lock); + --journal->j_barrier_count; + spin_unlock(&journal->j_state_lock); + wake_up(&journal->j_wait_transaction_locked); +} + +/* + * Report any unexpected dirty buffers which turn up. Normally those + * indicate an error, but they can occur if the user is running (say) + * tune2fs to modify the live filesystem, so we need the option of + * continuing as gracefully as possible. # + * + * The caller should already hold the journal lock and + * j_list_lock spinlock: most callers will need those anyway + * in order to probe the buffer's journaling state safely. + */ +static void jbd_unexpected_dirty_buffer(struct journal_head *jh) +{ + int jlist; + + /* If this buffer is one which might reasonably be dirty + * --- ie. data, or not part of this journal --- then + * we're OK to leave it alone, but otherwise we need to + * move the dirty bit to the journal's own internal + * JBDDirty bit. */ + jlist = jh->b_jlist; + + if (jlist == BJ_Metadata || jlist == BJ_Reserved || + jlist == BJ_Shadow || jlist == BJ_Forget) { + struct buffer_head *bh = jh2bh(jh); + + if (test_clear_buffer_dirty(bh)) + set_buffer_jbddirty(bh); + } +} + +/* + * If the buffer is already part of the current transaction, then there + * is nothing we need to do. If it is already part of a prior + * transaction which we are still committing to disk, then we need to + * make sure that we do not overwrite the old copy: we do copy-out to + * preserve the copy going to disk. We also account the buffer against + * the handle's metadata buffer credits (unless the buffer is already + * part of the transaction, that is). + * + */ +static int +do_get_write_access(handle_t *handle, struct journal_head *jh, + int force_copy) +{ + struct buffer_head *bh; + transaction_t *transaction; + journal_t *journal; + int error; + char *frozen_buffer = NULL; + int need_copy = 0; + + if (is_handle_aborted(handle)) + return -EROFS; + + transaction = handle->h_transaction; + journal = transaction->t_journal; + + jbd_debug(5, "buffer_head %p, force_copy %d\n", jh, force_copy); + + JBUFFER_TRACE(jh, "entry"); +repeat: + bh = jh2bh(jh); + + /* @@@ Need to check for errors here at some point. */ + + lock_buffer(bh); + jbd_lock_bh_state(bh); + + /* We now hold the buffer lock so it is safe to query the buffer + * state. Is the buffer dirty? + * + * If so, there are two possibilities. The buffer may be + * non-journaled, and undergoing a quite legitimate writeback. + * Otherwise, it is journaled, and we don't expect dirty buffers + * in that state (the buffers should be marked JBD_Dirty + * instead.) So either the IO is being done under our own + * control and this is a bug, or it's a third party IO such as + * dump(8) (which may leave the buffer scheduled for read --- + * ie. locked but not dirty) or tune2fs (which may actually have + * the buffer dirtied, ugh.) */ + + if (buffer_dirty(bh)) { + /* + * First question: is this buffer already part of the current + * transaction or the existing committing transaction? + */ + if (jh->b_transaction) { + J_ASSERT_JH(jh, + jh->b_transaction == transaction || + jh->b_transaction == + journal->j_committing_transaction); + if (jh->b_next_transaction) + J_ASSERT_JH(jh, jh->b_next_transaction == + transaction); + } + /* + * In any case we need to clean the dirty flag and we must + * do it under the buffer lock to be sure we don't race + * with running write-out. + */ + JBUFFER_TRACE(jh, "Unexpected dirty buffer"); + jbd_unexpected_dirty_buffer(jh); + } + + unlock_buffer(bh); + + error = -EROFS; + if (is_handle_aborted(handle)) { + jbd_unlock_bh_state(bh); + goto out; + } + error = 0; + + /* + * The buffer is already part of this transaction if b_transaction or + * b_next_transaction points to it + */ + if (jh->b_transaction == transaction || + jh->b_next_transaction == transaction) + goto done; + + /* + * If there is already a copy-out version of this buffer, then we don't + * need to make another one + */ + if (jh->b_frozen_data) { + JBUFFER_TRACE(jh, "has frozen data"); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + jh->b_next_transaction = transaction; + goto done; + } + + /* Is there data here we need to preserve? */ + + if (jh->b_transaction && jh->b_transaction != transaction) { + JBUFFER_TRACE(jh, "owned by older transaction"); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_committing_transaction); + + /* There is one case we have to be very careful about. + * If the committing transaction is currently writing + * this buffer out to disk and has NOT made a copy-out, + * then we cannot modify the buffer contents at all + * right now. The essence of copy-out is that it is the + * extra copy, not the primary copy, which gets + * journaled. If the primary copy is already going to + * disk then we cannot do copy-out here. */ + + if (jh->b_jlist == BJ_Shadow) { + DEFINE_WAIT_BIT(wait, &bh->b_state, BH_Unshadow); + wait_queue_head_t *wqh; + + wqh = bit_waitqueue(&bh->b_state, BH_Unshadow); + + JBUFFER_TRACE(jh, "on shadow: sleep"); + jbd_unlock_bh_state(bh); + /* commit wakes up all shadow buffers after IO */ + for ( ; ; ) { + prepare_to_wait(wqh, &wait.wait, + TASK_UNINTERRUPTIBLE); + if (jh->b_jlist != BJ_Shadow) + break; + schedule(); + } + finish_wait(wqh, &wait.wait); + goto repeat; + } + + /* Only do the copy if the currently-owning transaction + * still needs it. If it is on the Forget list, the + * committing transaction is past that stage. The + * buffer had better remain locked during the kmalloc, + * but that should be true --- we hold the journal lock + * still and the buffer is already on the BUF_JOURNAL + * list so won't be flushed. + * + * Subtle point, though: if this is a get_undo_access, + * then we will be relying on the frozen_data to contain + * the new value of the committed_data record after the + * transaction, so we HAVE to force the frozen_data copy + * in that case. */ + + if (jh->b_jlist != BJ_Forget || force_copy) { + JBUFFER_TRACE(jh, "generate frozen data"); + if (!frozen_buffer) { + JBUFFER_TRACE(jh, "allocate memory for buffer"); + jbd_unlock_bh_state(bh); + frozen_buffer = + jbd_slab_alloc(jh2bh(jh)->b_size, + GFP_NOFS); + if (!frozen_buffer) { + printk(KERN_EMERG + "%s: OOM for frozen_buffer\n", + __FUNCTION__); + JBUFFER_TRACE(jh, "oom!"); + error = -ENOMEM; + jbd_lock_bh_state(bh); + goto done; + } + goto repeat; + } + jh->b_frozen_data = frozen_buffer; + frozen_buffer = NULL; + need_copy = 1; + } + jh->b_next_transaction = transaction; + } + + + /* + * Finally, if the buffer is not journaled right now, we need to make + * sure it doesn't get written to disk before the caller actually + * commits the new data + */ + if (!jh->b_transaction) { + JBUFFER_TRACE(jh, "no transaction"); + J_ASSERT_JH(jh, !jh->b_next_transaction); + jh->b_transaction = transaction; + JBUFFER_TRACE(jh, "file as BJ_Reserved"); + spin_lock(&journal->j_list_lock); + __journal_file_buffer(jh, transaction, BJ_Reserved); + spin_unlock(&journal->j_list_lock); + } + +done: + if (need_copy) { + struct page *page; + int offset; + char *source; + + J_EXPECT_JH(jh, buffer_uptodate(jh2bh(jh)), + "Possible IO failure.\n"); + page = jh2bh(jh)->b_page; + offset = ((unsigned long) jh2bh(jh)->b_data) & ~PAGE_MASK; + source = kmap_atomic(page, KM_USER0); + memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size); + kunmap_atomic(source, KM_USER0); + } + jbd_unlock_bh_state(bh); + + /* + * If we are about to journal a buffer, then any revoke pending on it is + * no longer valid + */ + journal_cancel_revoke(handle, jh); + +out: + if (unlikely(frozen_buffer)) /* It's usually NULL */ + jbd_slab_free(frozen_buffer, bh->b_size); + + JBUFFER_TRACE(jh, "exit"); + return error; +} + +/** + * int journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. + * @handle: transaction to add buffer modifications to + * @bh: bh to be used for metadata writes + * @credits: variable that will receive credits for the buffer + * + * Returns an error code or 0 on success. + * + * In full data journalling mode the buffer may be of type BJ_AsyncData, + * because we're write()ing a buffer which is also part of a shared mapping. + */ + +int journal_get_write_access(handle_t *handle, struct buffer_head *bh) +{ + struct journal_head *jh = journal_add_journal_head(bh); + int rc; + + /* We do not want to get caught playing with fields which the + * log thread also manipulates. Make sure that the buffer + * completes any outstanding IO before proceeding. */ + rc = do_get_write_access(handle, jh, 0); + journal_put_journal_head(jh); + return rc; +} + + +/* + * When the user wants to journal a newly created buffer_head + * (ie. getblk() returned a new buffer and we are going to populate it + * manually rather than reading off disk), then we need to keep the + * buffer_head locked until it has been completely filled with new + * data. In this case, we should be able to make the assertion that + * the bh is not already part of an existing transaction. + * + * The buffer should already be locked by the caller by this point. + * There is no lock ranking violation: it was a newly created, + * unlocked buffer beforehand. */ + +/** + * int journal_get_create_access () - notify intent to use newly created bh + * @handle: transaction to new buffer to + * @bh: new buffer. + * + * Call this if you create a new bh. + */ +int journal_get_create_access(handle_t *handle, struct buffer_head *bh) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + struct journal_head *jh = journal_add_journal_head(bh); + int err; + + jbd_debug(5, "journal_head %p\n", jh); + err = -EROFS; + if (is_handle_aborted(handle)) + goto out; + err = 0; + + JBUFFER_TRACE(jh, "entry"); + /* + * The buffer may already belong to this transaction due to pre-zeroing + * in the filesystem's new_block code. It may also be on the previous, + * committing transaction's lists, but it HAS to be in Forget state in + * that case: the transaction must have deleted the buffer for it to be + * reused here. + */ + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + J_ASSERT_JH(jh, (jh->b_transaction == transaction || + jh->b_transaction == NULL || + (jh->b_transaction == journal->j_committing_transaction && + jh->b_jlist == BJ_Forget))); + + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, buffer_locked(jh2bh(jh))); + + if (jh->b_transaction == NULL) { + jh->b_transaction = transaction; + JBUFFER_TRACE(jh, "file as BJ_Reserved"); + __journal_file_buffer(jh, transaction, BJ_Reserved); + } else if (jh->b_transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "set next transaction"); + jh->b_next_transaction = transaction; + } + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + + /* + * akpm: I added this. ext3_alloc_branch can pick up new indirect + * blocks which contain freed but then revoked metadata. We need + * to cancel the revoke in case we end up freeing it yet again + * and the reallocating as data - this would cause a second revoke, + * which hits an assertion error. + */ + JBUFFER_TRACE(jh, "cancelling revoke"); + journal_cancel_revoke(handle, jh); + journal_put_journal_head(jh); +out: + return err; +} + +/** + * int journal_get_undo_access() - Notify intent to modify metadata with + * non-rewindable consequences + * @handle: transaction + * @bh: buffer to undo + * @credits: store the number of taken credits here (if not NULL) + * + * Sometimes there is a need to distinguish between metadata which has + * been committed to disk and that which has not. The ext3fs code uses + * this for freeing and allocating space, we have to make sure that we + * do not reuse freed space until the deallocation has been committed, + * since if we overwrote that space we would make the delete + * un-rewindable in case of a crash. + * + * To deal with that, journal_get_undo_access requests write access to a + * buffer for parts of non-rewindable operations such as delete + * operations on the bitmaps. The journaling code must keep a copy of + * the buffer's contents prior to the undo_access call until such time + * as we know that the buffer has definitely been committed to disk. + * + * We never need to know which transaction the committed data is part + * of, buffers touched here are guaranteed to be dirtied later and so + * will be committed to a new transaction in due course, at which point + * we can discard the old committed data pointer. + * + * Returns error number or 0 on success. + */ +int journal_get_undo_access(handle_t *handle, struct buffer_head *bh) +{ + int err; + struct journal_head *jh = journal_add_journal_head(bh); + char *committed_data = NULL; + + JBUFFER_TRACE(jh, "entry"); + + /* + * Do this first --- it can drop the journal lock, so we want to + * make sure that obtaining the committed_data is done + * atomically wrt. completion of any outstanding commits. + */ + err = do_get_write_access(handle, jh, 1); + if (err) + goto out; + +repeat: + if (!jh->b_committed_data) { + committed_data = jbd_slab_alloc(jh2bh(jh)->b_size, GFP_NOFS); + if (!committed_data) { + printk(KERN_EMERG "%s: No memory for committed data\n", + __FUNCTION__); + err = -ENOMEM; + goto out; + } + } + + jbd_lock_bh_state(bh); + if (!jh->b_committed_data) { + /* Copy out the current buffer contents into the + * preserved, committed copy. */ + JBUFFER_TRACE(jh, "generate b_committed data"); + if (!committed_data) { + jbd_unlock_bh_state(bh); + goto repeat; + } + + jh->b_committed_data = committed_data; + committed_data = NULL; + memcpy(jh->b_committed_data, bh->b_data, bh->b_size); + } + jbd_unlock_bh_state(bh); +out: + journal_put_journal_head(jh); + if (unlikely(committed_data)) + jbd_slab_free(committed_data, bh->b_size); + return err; +} + +/** + * int journal_dirty_data() - mark a buffer as containing dirty data which + * needs to be flushed before we can commit the + * current transaction. + * @handle: transaction + * @bh: bufferhead to mark + * + * The buffer is placed on the transaction's data list and is marked as + * belonging to the transaction. + * + * Returns error number or 0 on success. + * + * journal_dirty_data() can be called via page_launder->ext3_writepage + * by kswapd. + */ +int journal_dirty_data(handle_t *handle, struct buffer_head *bh) +{ + journal_t *journal = handle->h_transaction->t_journal; + int need_brelse = 0; + struct journal_head *jh; + + if (is_handle_aborted(handle)) + return 0; + + jh = journal_add_journal_head(bh); + JBUFFER_TRACE(jh, "entry"); + + /* + * The buffer could *already* be dirty. Writeout can start + * at any time. + */ + jbd_debug(4, "jh: %p, tid:%d\n", jh, handle->h_transaction->t_tid); + + /* + * What if the buffer is already part of a running transaction? + * + * There are two cases: + * 1) It is part of the current running transaction. Refile it, + * just in case we have allocated it as metadata, deallocated + * it, then reallocated it as data. + * 2) It is part of the previous, still-committing transaction. + * If all we want to do is to guarantee that the buffer will be + * written to disk before this new transaction commits, then + * being sure that the *previous* transaction has this same + * property is sufficient for us! Just leave it on its old + * transaction. + * + * In case (2), the buffer must not already exist as metadata + * --- that would violate write ordering (a transaction is free + * to write its data at any point, even before the previous + * committing transaction has committed). The caller must + * never, ever allow this to happen: there's nothing we can do + * about it in this layer. + */ + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + if (jh->b_transaction) { + JBUFFER_TRACE(jh, "has transaction"); + if (jh->b_transaction != handle->h_transaction) { + JBUFFER_TRACE(jh, "belongs to older transaction"); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_committing_transaction); + + /* @@@ IS THIS TRUE ? */ + /* + * Not any more. Scenario: someone does a write() + * in data=journal mode. The buffer's transaction has + * moved into commit. Then someone does another + * write() to the file. We do the frozen data copyout + * and set b_next_transaction to point to j_running_t. + * And while we're in that state, someone does a + * writepage() in an attempt to pageout the same area + * of the file via a shared mapping. At present that + * calls journal_dirty_data(), and we get right here. + * It may be too late to journal the data. Simply + * falling through to the next test will suffice: the + * data will be dirty and wil be checkpointed. The + * ordering comments in the next comment block still + * apply. + */ + //J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + + /* + * If we're journalling data, and this buffer was + * subject to a write(), it could be metadata, forget + * or shadow against the committing transaction. Now, + * someone has dirtied the same darn page via a mapping + * and it is being writepage()'d. + * We *could* just steal the page from commit, with some + * fancy locking there. Instead, we just skip it - + * don't tie the page's buffers to the new transaction + * at all. + * Implication: if we crash before the writepage() data + * is written into the filesystem, recovery will replay + * the write() data. + */ + if (jh->b_jlist != BJ_None && + jh->b_jlist != BJ_SyncData && + jh->b_jlist != BJ_Locked) { + JBUFFER_TRACE(jh, "Not stealing"); + goto no_journal; + } + + /* + * This buffer may be undergoing writeout in commit. We + * can't return from here and let the caller dirty it + * again because that can cause the write-out loop in + * commit to never terminate. + */ + if (buffer_dirty(bh)) { + get_bh(bh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + need_brelse = 1; + sync_dirty_buffer(bh); + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + /* The buffer may become locked again at any + time if it is redirtied */ + } + + /* journal_clean_data_list() may have got there first */ + if (jh->b_transaction != NULL) { + JBUFFER_TRACE(jh, "unfile from commit"); + __journal_temp_unlink_buffer(jh); + /* It still points to the committing + * transaction; move it to this one so + * that the refile assert checks are + * happy. */ + jh->b_transaction = handle->h_transaction; + } + /* The buffer will be refiled below */ + + } + /* + * Special case --- the buffer might actually have been + * allocated and then immediately deallocated in the previous, + * committing transaction, so might still be left on that + * transaction's metadata lists. + */ + if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) { + JBUFFER_TRACE(jh, "not on correct data list: unfile"); + J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow); + __journal_temp_unlink_buffer(jh); + jh->b_transaction = handle->h_transaction; + JBUFFER_TRACE(jh, "file as data"); + __journal_file_buffer(jh, handle->h_transaction, + BJ_SyncData); + } + } else { + JBUFFER_TRACE(jh, "not on a transaction"); + __journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); + } +no_journal: + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + if (need_brelse) { + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + } + JBUFFER_TRACE(jh, "exit"); + journal_put_journal_head(jh); + return 0; +} + +/** + * int journal_dirty_metadata() - mark a buffer as containing dirty metadata + * @handle: transaction to add buffer to. + * @bh: buffer to mark + * + * mark dirty metadata which needs to be journaled as part of the current + * transaction. + * + * The buffer is placed on the transaction's metadata list and is marked + * as belonging to the transaction. + * + * Returns error number or 0 on success. + * + * Special care needs to be taken if the buffer already belongs to the + * current committing transaction (in which case we should have frozen + * data present for that commit). In that case, we don't relink the + * buffer: that only gets done when the old transaction finally + * completes its commit. + */ +int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + struct journal_head *jh = bh2jh(bh); + + jbd_debug(5, "journal_head %p\n", jh); + JBUFFER_TRACE(jh, "entry"); + if (is_handle_aborted(handle)) + goto out; + + jbd_lock_bh_state(bh); + + if (jh->b_modified == 0) { + /* + * This buffer's got modified and becoming part + * of the transaction. This needs to be done + * once a transaction -bzzz + */ + jh->b_modified = 1; + J_ASSERT_JH(jh, handle->h_buffer_credits > 0); + handle->h_buffer_credits--; + } + + /* + * fastpath, to avoid expensive locking. If this buffer is already + * on the running transaction's metadata list there is nothing to do. + * Nobody can take it off again because there is a handle open. + * I _think_ we're OK here with SMP barriers - a mistaken decision will + * result in this test being false, so we go in and take the locks. + */ + if (jh->b_transaction == transaction && jh->b_jlist == BJ_Metadata) { + JBUFFER_TRACE(jh, "fastpath"); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_running_transaction); + goto out_unlock_bh; + } + + set_buffer_jbddirty(bh); + + /* + * Metadata already on the current transaction list doesn't + * need to be filed. Metadata on another transaction's list must + * be committing, and will be refiled once the commit completes: + * leave it alone for now. + */ + if (jh->b_transaction != transaction) { + JBUFFER_TRACE(jh, "already on other transaction"); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_committing_transaction); + J_ASSERT_JH(jh, jh->b_next_transaction == transaction); + /* And this case is illegal: we can't reuse another + * transaction's data buffer, ever. */ + goto out_unlock_bh; + } + + /* That test should have eliminated the following case: */ + J_ASSERT_JH(jh, jh->b_frozen_data == 0); + + JBUFFER_TRACE(jh, "file as BJ_Metadata"); + spin_lock(&journal->j_list_lock); + __journal_file_buffer(jh, handle->h_transaction, BJ_Metadata); + spin_unlock(&journal->j_list_lock); +out_unlock_bh: + jbd_unlock_bh_state(bh); +out: + JBUFFER_TRACE(jh, "exit"); + return 0; +} + +/* + * journal_release_buffer: undo a get_write_access without any buffer + * updates, if the update decided in the end that it didn't need access. + * + */ +void +journal_release_buffer(handle_t *handle, struct buffer_head *bh) +{ + BUFFER_TRACE(bh, "entry"); +} + +/** + * void journal_forget() - bforget() for potentially-journaled buffers. + * @handle: transaction handle + * @bh: bh to 'forget' + * + * We can only do the bforget if there are no commits pending against the + * buffer. If the buffer is dirty in the current running transaction we + * can safely unlink it. + * + * bh may not be a journalled buffer at all - it may be a non-JBD + * buffer which came off the hashtable. Check for this. + * + * Decrements bh->b_count by one. + * + * Allow this call even if the handle has aborted --- it may be part of + * the caller's cleanup after an abort. + */ +int journal_forget (handle_t *handle, struct buffer_head *bh) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + struct journal_head *jh; + int drop_reserve = 0; + int err = 0; + + BUFFER_TRACE(bh, "entry"); + + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + if (!buffer_jbd(bh)) + goto not_jbd; + jh = bh2jh(bh); + + /* Critical error: attempting to delete a bitmap buffer, maybe? + * Don't do any jbd operations, and return an error. */ + if (!J_EXPECT_JH(jh, !jh->b_committed_data, + "inconsistent data on disk")) { + err = -EIO; + goto not_jbd; + } + + /* + * The buffer's going from the transaction, we must drop + * all references -bzzz + */ + jh->b_modified = 0; + + if (jh->b_transaction == handle->h_transaction) { + J_ASSERT_JH(jh, !jh->b_frozen_data); + + /* If we are forgetting a buffer which is already part + * of this transaction, then we can just drop it from + * the transaction immediately. */ + clear_buffer_dirty(bh); + clear_buffer_jbddirty(bh); + + JBUFFER_TRACE(jh, "belongs to current transaction: unfile"); + + drop_reserve = 1; + + /* + * We are no longer going to journal this buffer. + * However, the commit of this transaction is still + * important to the buffer: the delete that we are now + * processing might obsolete an old log entry, so by + * committing, we can satisfy the buffer's checkpoint. + * + * So, if we have a checkpoint on the buffer, we should + * now refile the buffer on our BJ_Forget list so that + * we know to remove the checkpoint after we commit. + */ + + if (jh->b_cp_transaction) { + __journal_temp_unlink_buffer(jh); + __journal_file_buffer(jh, transaction, BJ_Forget); + } else { + __journal_unfile_buffer(jh); + journal_remove_journal_head(bh); + __brelse(bh); + if (!buffer_jbd(bh)) { + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + __bforget(bh); + goto drop; + } + } + } else if (jh->b_transaction) { + J_ASSERT_JH(jh, (jh->b_transaction == + journal->j_committing_transaction)); + /* However, if the buffer is still owned by a prior + * (committing) transaction, we can't drop it yet... */ + JBUFFER_TRACE(jh, "belongs to older transaction"); + /* ... but we CAN drop it from the new transaction if we + * have also modified it since the original commit. */ + + if (jh->b_next_transaction) { + J_ASSERT(jh->b_next_transaction == transaction); + jh->b_next_transaction = NULL; + drop_reserve = 1; + } + } + +not_jbd: + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + __brelse(bh); +drop: + if (drop_reserve) { + /* no need to reserve log space for this block -bzzz */ + handle->h_buffer_credits++; + } + return err; +} + +/** + * int journal_stop() - complete a transaction + * @handle: tranaction to complete. + * + * All done for a particular handle. + * + * There is not much action needed here. We just return any remaining + * buffer credits to the transaction and remove the handle. The only + * complication is that we need to start a commit operation if the + * filesystem is marked for synchronous update. + * + * journal_stop itself will not usually return an error, but it may + * do so in unusual circumstances. In particular, expect it to + * return -EIO if a journal_abort has been executed since the + * transaction began. + */ +int journal_stop(handle_t *handle) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + int old_handle_count, err; + pid_t pid; + + J_ASSERT(transaction->t_updates > 0); + J_ASSERT(journal_current_handle() == handle); + + if (is_handle_aborted(handle)) + err = -EIO; + else + err = 0; + + if (--handle->h_ref > 0) { + jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1, + handle->h_ref); + return err; + } + + jbd_debug(4, "Handle %p going down\n", handle); + + /* + * Implement synchronous transaction batching. If the handle + * was synchronous, don't force a commit immediately. Let's + * yield and let another thread piggyback onto this transaction. + * Keep doing that while new threads continue to arrive. + * It doesn't cost much - we're about to run a commit and sleep + * on IO anyway. Speeds up many-threaded, many-dir operations + * by 30x or more... + * + * But don't do this if this process was the most recent one to + * perform a synchronous write. We do this to detect the case where a + * single process is doing a stream of sync writes. No point in waiting + * for joiners in that case. + */ + pid = current->pid; + if (handle->h_sync && journal->j_last_sync_writer != pid) { + journal->j_last_sync_writer = pid; + do { + old_handle_count = transaction->t_handle_count; + schedule_timeout_uninterruptible(1); + } while (old_handle_count != transaction->t_handle_count); + } + + current->journal_info = NULL; + spin_lock(&journal->j_state_lock); + spin_lock(&transaction->t_handle_lock); + transaction->t_outstanding_credits -= handle->h_buffer_credits; + transaction->t_updates--; + if (!transaction->t_updates) { + wake_up(&journal->j_wait_updates); + if (journal->j_barrier_count) + wake_up(&journal->j_wait_transaction_locked); + } + + /* + * If the handle is marked SYNC, we need to set another commit + * going! We also want to force a commit if the current + * transaction is occupying too much of the log, or if the + * transaction is too old now. + */ + if (handle->h_sync || + transaction->t_outstanding_credits > + journal->j_max_transaction_buffers || + time_after_eq(jiffies, transaction->t_expires)) { + /* Do this even for aborted journals: an abort still + * completes the commit thread, it just doesn't write + * anything to disk. */ + tid_t tid = transaction->t_tid; + + spin_unlock(&transaction->t_handle_lock); + jbd_debug(2, "transaction too old, requesting commit for " + "handle %p\n", handle); + /* This is non-blocking */ + __log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); + + /* + * Special case: JFS_SYNC synchronous updates require us + * to wait for the commit to complete. + */ + if (handle->h_sync && !(current->flags & PF_MEMALLOC)) + err = log_wait_commit(journal, tid); + } else { + spin_unlock(&transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); + } + + jbd_free_handle(handle); + return err; +} + +/**int journal_force_commit() - force any uncommitted transactions + * @journal: journal to force + * + * For synchronous operations: force any uncommitted transactions + * to disk. May seem kludgy, but it reuses all the handle batching + * code in a very simple manner. + */ +int journal_force_commit(journal_t *journal) +{ + handle_t *handle; + int ret; + + handle = journal_start(journal, 1); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + } else { + handle->h_sync = 1; + ret = journal_stop(handle); + } + return ret; +} + +/* + * + * List management code snippets: various functions for manipulating the + * transaction buffer lists. + * + */ + +/* + * Append a buffer to a transaction list, given the transaction's list head + * pointer. + * + * j_list_lock is held. + * + * jbd_lock_bh_state(jh2bh(jh)) is held. + */ + +static inline void +__blist_add_buffer(struct journal_head **list, struct journal_head *jh) +{ + if (!*list) { + jh->b_tnext = jh->b_tprev = jh; + *list = jh; + } else { + /* Insert at the tail of the list to preserve order */ + struct journal_head *first = *list, *last = first->b_tprev; + jh->b_tprev = last; + jh->b_tnext = first; + last->b_tnext = first->b_tprev = jh; + } +} + +/* + * Remove a buffer from a transaction list, given the transaction's list + * head pointer. + * + * Called with j_list_lock held, and the journal may not be locked. + * + * jbd_lock_bh_state(jh2bh(jh)) is held. + */ + +static inline void +__blist_del_buffer(struct journal_head **list, struct journal_head *jh) +{ + if (*list == jh) { + *list = jh->b_tnext; + if (*list == jh) + *list = NULL; + } + jh->b_tprev->b_tnext = jh->b_tnext; + jh->b_tnext->b_tprev = jh->b_tprev; +} + +/* + * Remove a buffer from the appropriate transaction list. + * + * Note that this function can *change* the value of + * bh->b_transaction->t_sync_datalist, t_buffers, t_forget, + * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list. If the caller + * is holding onto a copy of one of thee pointers, it could go bad. + * Generally the caller needs to re-read the pointer from the transaction_t. + * + * Called under j_list_lock. The journal may not be locked. + */ +void __journal_temp_unlink_buffer(struct journal_head *jh) +{ + struct journal_head **list = NULL; + transaction_t *transaction; + struct buffer_head *bh = jh2bh(jh); + + J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); + transaction = jh->b_transaction; + if (transaction) + assert_spin_locked(&transaction->t_journal->j_list_lock); + + J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); + if (jh->b_jlist != BJ_None) + J_ASSERT_JH(jh, transaction != 0); + + switch (jh->b_jlist) { + case BJ_None: + return; + case BJ_SyncData: + list = &transaction->t_sync_datalist; + break; + case BJ_Metadata: + transaction->t_nr_buffers--; + J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0); + list = &transaction->t_buffers; + break; + case BJ_Forget: + list = &transaction->t_forget; + break; + case BJ_IO: + list = &transaction->t_iobuf_list; + break; + case BJ_Shadow: + list = &transaction->t_shadow_list; + break; + case BJ_LogCtl: + list = &transaction->t_log_list; + break; + case BJ_Reserved: + list = &transaction->t_reserved_list; + break; + case BJ_Locked: + list = &transaction->t_locked_list; + break; + } + + __blist_del_buffer(list, jh); + jh->b_jlist = BJ_None; + if (test_clear_buffer_jbddirty(bh)) + mark_buffer_dirty(bh); /* Expose it to the VM */ +} + +void __journal_unfile_buffer(struct journal_head *jh) +{ + __journal_temp_unlink_buffer(jh); + jh->b_transaction = NULL; +} + +void journal_unfile_buffer(journal_t *journal, struct journal_head *jh) +{ + jbd_lock_bh_state(jh2bh(jh)); + spin_lock(&journal->j_list_lock); + __journal_unfile_buffer(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(jh2bh(jh)); +} + +/* + * Called from journal_try_to_free_buffers(). + * + * Called under jbd_lock_bh_state(bh) + */ +static void +__journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) +{ + struct journal_head *jh; + + jh = bh2jh(bh); + + if (buffer_locked(bh) || buffer_dirty(bh)) + goto out; + + if (jh->b_next_transaction != 0) + goto out; + + spin_lock(&journal->j_list_lock); + if (jh->b_transaction != 0 && jh->b_cp_transaction == 0) { + if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) { + /* A written-back ordered data buffer */ + JBUFFER_TRACE(jh, "release data"); + __journal_unfile_buffer(jh); + journal_remove_journal_head(bh); + __brelse(bh); + } + } else if (jh->b_cp_transaction != 0 && jh->b_transaction == 0) { + /* written-back checkpointed metadata buffer */ + if (jh->b_jlist == BJ_None) { + JBUFFER_TRACE(jh, "remove from checkpoint list"); + __journal_remove_checkpoint(jh); + journal_remove_journal_head(bh); + __brelse(bh); + } + } + spin_unlock(&journal->j_list_lock); +out: + return; +} + + +/** + * int journal_try_to_free_buffers() - try to free page buffers. + * @journal: journal for operation + * @page: to try and free + * @unused_gfp_mask: unused + * + * + * For all the buffers on this page, + * if they are fully written out ordered data, move them onto BUF_CLEAN + * so try_to_free_buffers() can reap them. + * + * This function returns non-zero if we wish try_to_free_buffers() + * to be called. We do this if the page is releasable by try_to_free_buffers(). + * We also do it if the page has locked or dirty buffers and the caller wants + * us to perform sync or async writeout. + * + * This complicates JBD locking somewhat. We aren't protected by the + * BKL here. We wish to remove the buffer from its committing or + * running transaction's ->t_datalist via __journal_unfile_buffer. + * + * This may *change* the value of transaction_t->t_datalist, so anyone + * who looks at t_datalist needs to lock against this function. + * + * Even worse, someone may be doing a journal_dirty_data on this + * buffer. So we need to lock against that. journal_dirty_data() + * will come out of the lock with the buffer dirty, which makes it + * ineligible for release here. + * + * Who else is affected by this? hmm... Really the only contender + * is do_get_write_access() - it could be looking at the buffer while + * journal_try_to_free_buffer() is changing its state. But that + * cannot happen because we never reallocate freed data as metadata + * while the data is part of a transaction. Yes? + */ +int journal_try_to_free_buffers(journal_t *journal, + struct page *page, gfp_t unused_gfp_mask) +{ + struct buffer_head *head; + struct buffer_head *bh; + int ret = 0; + + J_ASSERT(PageLocked(page)); + + head = page_buffers(page); + bh = head; + do { + struct journal_head *jh; + + /* + * We take our own ref against the journal_head here to avoid + * having to add tons of locking around each instance of + * journal_remove_journal_head() and journal_put_journal_head(). + */ + jh = journal_grab_journal_head(bh); + if (!jh) + continue; + + jbd_lock_bh_state(bh); + __journal_try_to_free_buffer(journal, bh); + journal_put_journal_head(jh); + jbd_unlock_bh_state(bh); + if (buffer_jbd(bh)) + goto busy; + } while ((bh = bh->b_this_page) != head); + ret = try_to_free_buffers(page); +busy: + return ret; +} + +/* + * This buffer is no longer needed. If it is on an older transaction's + * checkpoint list we need to record it on this transaction's forget list + * to pin this buffer (and hence its checkpointing transaction) down until + * this transaction commits. If the buffer isn't on a checkpoint list, we + * release it. + * Returns non-zero if JBD no longer has an interest in the buffer. + * + * Called under j_list_lock. + * + * Called under jbd_lock_bh_state(bh). + */ +static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) +{ + int may_free = 1; + struct buffer_head *bh = jh2bh(jh); + + __journal_unfile_buffer(jh); + + if (jh->b_cp_transaction) { + JBUFFER_TRACE(jh, "on running+cp transaction"); + __journal_file_buffer(jh, transaction, BJ_Forget); + clear_buffer_jbddirty(bh); + may_free = 0; + } else { + JBUFFER_TRACE(jh, "on running transaction"); + journal_remove_journal_head(bh); + __brelse(bh); + } + return may_free; +} + +/* + * journal_invalidatepage + * + * This code is tricky. It has a number of cases to deal with. + * + * There are two invariants which this code relies on: + * + * i_size must be updated on disk before we start calling invalidatepage on the + * data. + * + * This is done in ext3 by defining an ext3_setattr method which + * updates i_size before truncate gets going. By maintaining this + * invariant, we can be sure that it is safe to throw away any buffers + * attached to the current transaction: once the transaction commits, + * we know that the data will not be needed. + * + * Note however that we can *not* throw away data belonging to the + * previous, committing transaction! + * + * Any disk blocks which *are* part of the previous, committing + * transaction (and which therefore cannot be discarded immediately) are + * not going to be reused in the new running transaction + * + * The bitmap committed_data images guarantee this: any block which is + * allocated in one transaction and removed in the next will be marked + * as in-use in the committed_data bitmap, so cannot be reused until + * the next transaction to delete the block commits. This means that + * leaving committing buffers dirty is quite safe: the disk blocks + * cannot be reallocated to a different file and so buffer aliasing is + * not possible. + * + * + * The above applies mainly to ordered data mode. In writeback mode we + * don't make guarantees about the order in which data hits disk --- in + * particular we don't guarantee that new dirty data is flushed before + * transaction commit --- so it is always safe just to discard data + * immediately in that mode. --sct + */ + +/* + * The journal_unmap_buffer helper function returns zero if the buffer + * concerned remains pinned as an anonymous buffer belonging to an older + * transaction. + * + * We're outside-transaction here. Either or both of j_running_transaction + * and j_committing_transaction may be NULL. + */ +static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) +{ + transaction_t *transaction; + struct journal_head *jh; + int may_free = 1; + int ret; + + BUFFER_TRACE(bh, "entry"); + + /* + * It is safe to proceed here without the j_list_lock because the + * buffers cannot be stolen by try_to_free_buffers as long as we are + * holding the page lock. --sct + */ + + if (!buffer_jbd(bh)) + goto zap_buffer_unlocked; + + spin_lock(&journal->j_state_lock); + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + jh = journal_grab_journal_head(bh); + if (!jh) + goto zap_buffer_no_jh; + + transaction = jh->b_transaction; + if (transaction == NULL) { + /* First case: not on any transaction. If it + * has no checkpoint link, then we can zap it: + * it's a writeback-mode buffer so we don't care + * if it hits disk safely. */ + if (!jh->b_cp_transaction) { + JBUFFER_TRACE(jh, "not on any transaction: zap"); + goto zap_buffer; + } + + if (!buffer_dirty(bh)) { + /* bdflush has written it. We can drop it now */ + goto zap_buffer; + } + + /* OK, it must be in the journal but still not + * written fully to disk: it's metadata or + * journaled data... */ + + if (journal->j_running_transaction) { + /* ... and once the current transaction has + * committed, the buffer won't be needed any + * longer. */ + JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget"); + ret = __dispose_buffer(jh, + journal->j_running_transaction); + journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); + return ret; + } else { + /* There is no currently-running transaction. So the + * orphan record which we wrote for this file must have + * passed into commit. We must attach this buffer to + * the committing transaction, if it exists. */ + if (journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "give to committing trans"); + ret = __dispose_buffer(jh, + journal->j_committing_transaction); + journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); + return ret; + } else { + /* The orphan record's transaction has + * committed. We can cleanse this buffer */ + clear_buffer_jbddirty(bh); + goto zap_buffer; + } + } + } else if (transaction == journal->j_committing_transaction) { + if (jh->b_jlist == BJ_Locked) { + /* + * The buffer is on the committing transaction's locked + * list. We have the buffer locked, so I/O has + * completed. So we can nail the buffer now. + */ + may_free = __dispose_buffer(jh, transaction); + goto zap_buffer; + } + /* + * If it is committing, we simply cannot touch it. We + * can remove it's next_transaction pointer from the + * running transaction if that is set, but nothing + * else. */ + JBUFFER_TRACE(jh, "on committing transaction"); + set_buffer_freed(bh); + if (jh->b_next_transaction) { + J_ASSERT(jh->b_next_transaction == + journal->j_running_transaction); + jh->b_next_transaction = NULL; + } + journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); + return 0; + } else { + /* Good, the buffer belongs to the running transaction. + * We are writing our own transaction's data, not any + * previous one's, so it is safe to throw it away + * (remember that we expect the filesystem to have set + * i_size already for this truncate so recovery will not + * expose the disk blocks we are discarding here.) */ + J_ASSERT_JH(jh, transaction == journal->j_running_transaction); + may_free = __dispose_buffer(jh, transaction); + } + +zap_buffer: + journal_put_journal_head(jh); +zap_buffer_no_jh: + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); +zap_buffer_unlocked: + clear_buffer_dirty(bh); + J_ASSERT_BH(bh, !buffer_jbddirty(bh)); + clear_buffer_mapped(bh); + clear_buffer_req(bh); + clear_buffer_new(bh); + bh->b_bdev = NULL; + return may_free; +} + +/** + * void journal_invalidatepage() + * @journal: journal to use for flush... + * @page: page to flush + * @offset: length of page to invalidate. + * + * Reap page buffers containing data after offset in page. + * + */ +void journal_invalidatepage(journal_t *journal, + struct page *page, + unsigned long offset) +{ + struct buffer_head *head, *bh, *next; + unsigned int curr_off = 0; + int may_free = 1; + + if (!PageLocked(page)) + BUG(); + if (!page_has_buffers(page)) + return; + + /* We will potentially be playing with lists other than just the + * data lists (especially for journaled data mode), so be + * cautious in our locking. */ + + head = bh = page_buffers(page); + do { + unsigned int next_off = curr_off + bh->b_size; + next = bh->b_this_page; + + if (offset <= curr_off) { + /* This block is wholly outside the truncation point */ + lock_buffer(bh); + may_free &= journal_unmap_buffer(journal, bh); + unlock_buffer(bh); + } + curr_off = next_off; + bh = next; + + } while (bh != head); + + if (!offset) { + if (may_free && try_to_free_buffers(page)) + J_ASSERT(!page_has_buffers(page)); + } +} + +/* + * File a buffer on the given transaction list. + */ +void __journal_file_buffer(struct journal_head *jh, + transaction_t *transaction, int jlist) +{ + struct journal_head **list = NULL; + int was_dirty = 0; + struct buffer_head *bh = jh2bh(jh); + + J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); + assert_spin_locked(&transaction->t_journal->j_list_lock); + + J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); + J_ASSERT_JH(jh, jh->b_transaction == transaction || + jh->b_transaction == 0); + + if (jh->b_transaction && jh->b_jlist == jlist) + return; + + /* The following list of buffer states needs to be consistent + * with __jbd_unexpected_dirty_buffer()'s handling of dirty + * state. */ + + if (jlist == BJ_Metadata || jlist == BJ_Reserved || + jlist == BJ_Shadow || jlist == BJ_Forget) { + if (test_clear_buffer_dirty(bh) || + test_clear_buffer_jbddirty(bh)) + was_dirty = 1; + } + + if (jh->b_transaction) + __journal_temp_unlink_buffer(jh); + jh->b_transaction = transaction; + + switch (jlist) { + case BJ_None: + J_ASSERT_JH(jh, !jh->b_committed_data); + J_ASSERT_JH(jh, !jh->b_frozen_data); + return; + case BJ_SyncData: + list = &transaction->t_sync_datalist; + break; + case BJ_Metadata: + transaction->t_nr_buffers++; + list = &transaction->t_buffers; + break; + case BJ_Forget: + list = &transaction->t_forget; + break; + case BJ_IO: + list = &transaction->t_iobuf_list; + break; + case BJ_Shadow: + list = &transaction->t_shadow_list; + break; + case BJ_LogCtl: + list = &transaction->t_log_list; + break; + case BJ_Reserved: + list = &transaction->t_reserved_list; + break; + case BJ_Locked: + list = &transaction->t_locked_list; + break; + } + + __blist_add_buffer(list, jh); + jh->b_jlist = jlist; + + if (was_dirty) + set_buffer_jbddirty(bh); +} + +void journal_file_buffer(struct journal_head *jh, + transaction_t *transaction, int jlist) +{ + jbd_lock_bh_state(jh2bh(jh)); + spin_lock(&transaction->t_journal->j_list_lock); + __journal_file_buffer(jh, transaction, jlist); + spin_unlock(&transaction->t_journal->j_list_lock); + jbd_unlock_bh_state(jh2bh(jh)); +} + +/* + * Remove a buffer from its current buffer list in preparation for + * dropping it from its current transaction entirely. If the buffer has + * already started to be used by a subsequent transaction, refile the + * buffer on that transaction's metadata list. + * + * Called under journal->j_list_lock + * + * Called under jbd_lock_bh_state(jh2bh(jh)) + */ +void __journal_refile_buffer(struct journal_head *jh) +{ + int was_dirty; + struct buffer_head *bh = jh2bh(jh); + + J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); + if (jh->b_transaction) + assert_spin_locked(&jh->b_transaction->t_journal->j_list_lock); + + /* If the buffer is now unused, just drop it. */ + if (jh->b_next_transaction == NULL) { + __journal_unfile_buffer(jh); + return; + } + + /* + * It has been modified by a later transaction: add it to the new + * transaction's metadata list. + */ + + was_dirty = test_clear_buffer_jbddirty(bh); + __journal_temp_unlink_buffer(jh); + jh->b_transaction = jh->b_next_transaction; + jh->b_next_transaction = NULL; + __journal_file_buffer(jh, jh->b_transaction, + was_dirty ? BJ_Metadata : BJ_Reserved); + J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING); + + if (was_dirty) + set_buffer_jbddirty(bh); +} + +/* + * For the unlocked version of this call, also make sure that any + * hanging journal_head is cleaned up if necessary. + * + * __journal_refile_buffer is usually called as part of a single locked + * operation on a buffer_head, in which the caller is probably going to + * be hooking the journal_head onto other lists. In that case it is up + * to the caller to remove the journal_head if necessary. For the + * unlocked journal_refile_buffer call, the caller isn't going to be + * doing anything else to the buffer so we need to do the cleanup + * ourselves to avoid a jh leak. + * + * *** The journal_head may be freed by this call! *** + */ +void journal_refile_buffer(journal_t *journal, struct journal_head *jh) +{ + struct buffer_head *bh = jh2bh(jh); + + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + __journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + journal_remove_journal_head(bh); + + spin_unlock(&journal->j_list_lock); + __brelse(bh); +} diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h new file mode 100644 index 0000000..3dbf6c7 --- /dev/null +++ b/include/linux/ext4_jbd2.h @@ -0,0 +1,268 @@ +/* + * linux/include/linux/ext4_jbd.h + * + * Written by Stephen C. Tweedie , 1999 + * + * Copyright 1998--1999 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Ext4-specific journaling extensions. + */ + +#ifndef _LINUX_EXT4_JBD_H +#define _LINUX_EXT4_JBD_H + +#include +#include +#include + +#define EXT4_JOURNAL(inode) (EXT4_SB((inode)->i_sb)->s_journal) + +/* Define the number of blocks we need to account to a transaction to + * modify one block of data. + * + * We may have to touch one inode, one bitmap buffer, up to three + * indirection blocks, the group and superblock summaries, and the data + * block to complete the transaction. */ + +#define EXT4_SINGLEDATA_TRANS_BLOCKS 8U + +/* Extended attribute operations touch at most two data buffers, + * two bitmap buffers, and two group summaries, in addition to the inode + * and the superblock, which are already accounted for. */ + +#define EXT4_XATTR_TRANS_BLOCKS 6U + +/* Define the minimum size for a transaction which modifies data. This + * needs to take into account the fact that we may end up modifying two + * quota files too (one for the group, one for the user quota). The + * superblock only gets updated once, of course, so don't bother + * counting that again for the quota updates. */ + +#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS + \ + EXT4_XATTR_TRANS_BLOCKS - 2 + \ + 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) + +/* Delete operations potentially hit one directory's namespace plus an + * entire inode, plus arbitrary amounts of bitmap/indirection data. Be + * generous. We can grow the delete transaction later if necessary. */ + +#define EXT4_DELETE_TRANS_BLOCKS(sb) (2 * EXT4_DATA_TRANS_BLOCKS(sb) + 64) + +/* Define an arbitrary limit for the amount of data we will anticipate + * writing to any given transaction. For unbounded transactions such as + * write(2) and truncate(2) we can write more than this, but we always + * start off at the maximum transaction size and grow the transaction + * optimistically as we go. */ + +#define EXT4_MAX_TRANS_DATA 64U + +/* We break up a large truncate or write transaction once the handle's + * buffer credits gets this low, we need either to extend the + * transaction or to start a new one. Reserve enough space here for + * inode, bitmap, superblock, group and indirection updates for at least + * one block, plus two quota updates. Quota allocations are not + * needed. */ + +#define EXT4_RESERVE_TRANS_BLOCKS 12U + +#define EXT4_INDEX_EXTRA_TRANS_BLOCKS 8 + +#ifdef CONFIG_QUOTA +/* Amount of blocks needed for quota update - we know that the structure was + * allocated so we need to update only inode+data */ +#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) +/* Amount of blocks needed for quota insert/delete - we do some block writes + * but inode, sb and group updates are done only once */ +#define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ + (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) +#define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ + (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) +#else +#define EXT4_QUOTA_TRANS_BLOCKS(sb) 0 +#define EXT4_QUOTA_INIT_BLOCKS(sb) 0 +#define EXT4_QUOTA_DEL_BLOCKS(sb) 0 +#endif + +int +ext4_mark_iloc_dirty(handle_t *handle, + struct inode *inode, + struct ext4_iloc *iloc); + +/* + * On success, We end up with an outstanding reference count against + * iloc->bh. This _must_ be cleaned up later. + */ + +int ext4_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext4_iloc *iloc); + +int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode); + +/* + * Wrapper functions with which ext4 calls into JBD. The intent here is + * to allow these to be turned into appropriate stubs so ext4 can control + * ext2 filesystems, so ext2+ext4 systems only nee one fs. This work hasn't + * been done yet. + */ + +void ext4_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err); + +static inline int +__ext4_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_undo_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext4_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_write_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline void +ext4_journal_release_buffer(handle_t *handle, struct buffer_head *bh) +{ + journal_release_buffer(handle, bh); +} + +static inline int +__ext4_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) +{ + int err = journal_forget(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext4_journal_revoke(const char *where, handle_t *handle, + unsigned long blocknr, struct buffer_head *bh) +{ + int err = journal_revoke(handle, blocknr, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext4_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_get_create_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +static inline int +__ext4_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_dirty_metadata(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + + +#define ext4_journal_get_undo_access(handle, bh) \ + __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_get_write_access(handle, bh) \ + __ext4_journal_get_write_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_revoke(handle, blocknr, bh) \ + __ext4_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) +#define ext4_journal_get_create_access(handle, bh) \ + __ext4_journal_get_create_access(__FUNCTION__, (handle), (bh)) +#define ext4_journal_dirty_metadata(handle, bh) \ + __ext4_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) +#define ext4_journal_forget(handle, bh) \ + __ext4_journal_forget(__FUNCTION__, (handle), (bh)) + +int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh); + +handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); +int __ext4_journal_stop(const char *where, handle_t *handle); + +static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks) +{ + return ext4_journal_start_sb(inode->i_sb, nblocks); +} + +#define ext4_journal_stop(handle) \ + __ext4_journal_stop(__FUNCTION__, (handle)) + +static inline handle_t *ext4_journal_current_handle(void) +{ + return journal_current_handle(); +} + +static inline int ext4_journal_extend(handle_t *handle, int nblocks) +{ + return journal_extend(handle, nblocks); +} + +static inline int ext4_journal_restart(handle_t *handle, int nblocks) +{ + return journal_restart(handle, nblocks); +} + +static inline int ext4_journal_blocks_per_page(struct inode *inode) +{ + return journal_blocks_per_page(inode); +} + +static inline int ext4_journal_force_commit(journal_t *journal) +{ + return journal_force_commit(journal); +} + +/* super.c */ +int ext4_force_commit(struct super_block *sb); + +static inline int ext4_should_journal_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 1; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) + return 1; + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) + return 1; + return 0; +} + +static inline int ext4_should_order_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) + return 0; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) + return 1; + return 0; +} + +static inline int ext4_should_writeback_data(struct inode *inode) +{ + if (!S_ISREG(inode->i_mode)) + return 0; + if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) + return 0; + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) + return 1; + return 0; +} + +#endif /* _LINUX_EXT4_JBD_H */ diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h new file mode 100644 index 0000000..fe89444 --- /dev/null +++ b/include/linux/jbd2.h @@ -0,0 +1,1098 @@ +/* + * linux/include/linux/jbd.h + * + * Written by Stephen C. Tweedie + * + * Copyright 1998-2000 Red Hat, Inc --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Definitions for transaction data structures for the buffer cache + * filesystem journaling support. + */ + +#ifndef _LINUX_JBD_H +#define _LINUX_JBD_H + +/* Allow this file to be included directly into e2fsprogs */ +#ifndef __KERNEL__ +#include "jfs_compat.h" +#define JFS_DEBUG +#define jfs_debug jbd_debug +#else + +#include +#include +#include +#include +#include +#include +#include + +#include +#endif + +#define journal_oom_retry 1 + +/* + * Define JBD_PARANIOD_IOFAIL to cause a kernel BUG() if ext3 finds + * certain classes of error which can occur due to failed IOs. Under + * normal use we want ext3 to continue after such errors, because + * hardware _can_ fail, but for debugging purposes when running tests on + * known-good hardware we may want to trap these errors. + */ +#undef JBD_PARANOID_IOFAIL + +/* + * The default maximum commit age, in seconds. + */ +#define JBD_DEFAULT_MAX_COMMIT_AGE 5 + +#ifdef CONFIG_JBD_DEBUG +/* + * Define JBD_EXPENSIVE_CHECKING to enable more expensive internal + * consistency checks. By default we don't do this unless + * CONFIG_JBD_DEBUG is on. + */ +#define JBD_EXPENSIVE_CHECKING +extern int journal_enable_debug; + +#define jbd_debug(n, f, a...) \ + do { \ + if ((n) <= journal_enable_debug) { \ + printk (KERN_DEBUG "(%s, %d): %s: ", \ + __FILE__, __LINE__, __FUNCTION__); \ + printk (f, ## a); \ + } \ + } while (0) +#else +#define jbd_debug(f, a...) /**/ +#endif + +extern void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry); +extern void * jbd_slab_alloc(size_t size, gfp_t flags); +extern void jbd_slab_free(void *ptr, size_t size); + +#define jbd_kmalloc(size, flags) \ + __jbd_kmalloc(__FUNCTION__, (size), (flags), journal_oom_retry) +#define jbd_rep_kmalloc(size, flags) \ + __jbd_kmalloc(__FUNCTION__, (size), (flags), 1) + +#define JFS_MIN_JOURNAL_BLOCKS 1024 + +#ifdef __KERNEL__ + +/** + * typedef handle_t - The handle_t type represents a single atomic update being performed by some process. + * + * All filesystem modifications made by the process go + * through this handle. Recursive operations (such as quota operations) + * are gathered into a single update. + * + * The buffer credits field is used to account for journaled buffers + * being modified by the running process. To ensure that there is + * enough log space for all outstanding operations, we need to limit the + * number of outstanding buffers possible at any time. When the + * operation completes, any buffer credits not used are credited back to + * the transaction, so that at all times we know how many buffers the + * outstanding updates on a transaction might possibly touch. + * + * This is an opaque datatype. + **/ +typedef struct handle_s handle_t; /* Atomic operation type */ + + +/** + * typedef journal_t - The journal_t maintains all of the journaling state information for a single filesystem. + * + * journal_t is linked to from the fs superblock structure. + * + * We use the journal_t to keep track of all outstanding transaction + * activity on the filesystem, and to manage the state of the log + * writing process. + * + * This is an opaque datatype. + **/ +typedef struct journal_s journal_t; /* Journal control structure */ +#endif + +/* + * Internal structures used by the logging mechanism: + */ + +#define JFS_MAGIC_NUMBER 0xc03b3998U /* The first 4 bytes of /dev/random! */ + +/* + * On-disk structures + */ + +/* + * Descriptor block types: + */ + +#define JFS_DESCRIPTOR_BLOCK 1 +#define JFS_COMMIT_BLOCK 2 +#define JFS_SUPERBLOCK_V1 3 +#define JFS_SUPERBLOCK_V2 4 +#define JFS_REVOKE_BLOCK 5 + +/* + * Standard header for all descriptor blocks: + */ +typedef struct journal_header_s +{ + __be32 h_magic; + __be32 h_blocktype; + __be32 h_sequence; +} journal_header_t; + + +/* + * The block tag: used to describe a single buffer in the journal + */ +typedef struct journal_block_tag_s +{ + __be32 t_blocknr; /* The on-disk block number */ + __be32 t_flags; /* See below */ +} journal_block_tag_t; + +/* + * The revoke descriptor: used on disk to describe a series of blocks to + * be revoked from the log + */ +typedef struct journal_revoke_header_s +{ + journal_header_t r_header; + __be32 r_count; /* Count of bytes used in the block */ +} journal_revoke_header_t; + + +/* Definitions for the journal tag flags word: */ +#define JFS_FLAG_ESCAPE 1 /* on-disk block is escaped */ +#define JFS_FLAG_SAME_UUID 2 /* block has same uuid as previous */ +#define JFS_FLAG_DELETED 4 /* block deleted by this transaction */ +#define JFS_FLAG_LAST_TAG 8 /* last tag in this descriptor block */ + + +/* + * The journal superblock. All fields are in big-endian byte order. + */ +typedef struct journal_superblock_s +{ +/* 0x0000 */ + journal_header_t s_header; + +/* 0x000C */ + /* Static information describing the journal */ + __be32 s_blocksize; /* journal device blocksize */ + __be32 s_maxlen; /* total blocks in journal file */ + __be32 s_first; /* first block of log information */ + +/* 0x0018 */ + /* Dynamic information describing the current state of the log */ + __be32 s_sequence; /* first commit ID expected in log */ + __be32 s_start; /* blocknr of start of log */ + +/* 0x0020 */ + /* Error value, as set by journal_abort(). */ + __be32 s_errno; + +/* 0x0024 */ + /* Remaining fields are only valid in a version-2 superblock */ + __be32 s_feature_compat; /* compatible feature set */ + __be32 s_feature_incompat; /* incompatible feature set */ + __be32 s_feature_ro_compat; /* readonly-compatible feature set */ +/* 0x0030 */ + __u8 s_uuid[16]; /* 128-bit uuid for journal */ + +/* 0x0040 */ + __be32 s_nr_users; /* Nr of filesystems sharing log */ + + __be32 s_dynsuper; /* Blocknr of dynamic superblock copy*/ + +/* 0x0048 */ + __be32 s_max_transaction; /* Limit of journal blocks per trans.*/ + __be32 s_max_trans_data; /* Limit of data blocks per trans. */ + +/* 0x0050 */ + __u32 s_padding[44]; + +/* 0x0100 */ + __u8 s_users[16*48]; /* ids of all fs'es sharing the log */ +/* 0x0400 */ +} journal_superblock_t; + +#define JFS_HAS_COMPAT_FEATURE(j,mask) \ + ((j)->j_format_version >= 2 && \ + ((j)->j_superblock->s_feature_compat & cpu_to_be32((mask)))) +#define JFS_HAS_RO_COMPAT_FEATURE(j,mask) \ + ((j)->j_format_version >= 2 && \ + ((j)->j_superblock->s_feature_ro_compat & cpu_to_be32((mask)))) +#define JFS_HAS_INCOMPAT_FEATURE(j,mask) \ + ((j)->j_format_version >= 2 && \ + ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask)))) + +#define JFS_FEATURE_INCOMPAT_REVOKE 0x00000001 + +/* Features known to this kernel version: */ +#define JFS_KNOWN_COMPAT_FEATURES 0 +#define JFS_KNOWN_ROCOMPAT_FEATURES 0 +#define JFS_KNOWN_INCOMPAT_FEATURES JFS_FEATURE_INCOMPAT_REVOKE + +#ifdef __KERNEL__ + +#include +#include + +#define JBD_ASSERTIONS +#ifdef JBD_ASSERTIONS +#define J_ASSERT(assert) \ +do { \ + if (!(assert)) { \ + printk (KERN_EMERG \ + "Assertion failure in %s() at %s:%d: \"%s\"\n", \ + __FUNCTION__, __FILE__, __LINE__, # assert); \ + BUG(); \ + } \ +} while (0) + +#if defined(CONFIG_BUFFER_DEBUG) +void buffer_assertion_failure(struct buffer_head *bh); +#define J_ASSERT_BH(bh, expr) \ + do { \ + if (!(expr)) \ + buffer_assertion_failure(bh); \ + J_ASSERT(expr); \ + } while (0) +#define J_ASSERT_JH(jh, expr) J_ASSERT_BH(jh2bh(jh), expr) +#else +#define J_ASSERT_BH(bh, expr) J_ASSERT(expr) +#define J_ASSERT_JH(jh, expr) J_ASSERT(expr) +#endif + +#else +#define J_ASSERT(assert) do { } while (0) +#endif /* JBD_ASSERTIONS */ + +#if defined(JBD_PARANOID_IOFAIL) +#define J_EXPECT(expr, why...) J_ASSERT(expr) +#define J_EXPECT_BH(bh, expr, why...) J_ASSERT_BH(bh, expr) +#define J_EXPECT_JH(jh, expr, why...) J_ASSERT_JH(jh, expr) +#else +#define __journal_expect(expr, why...) \ + ({ \ + int val = (expr); \ + if (!val) { \ + printk(KERN_ERR \ + "EXT3-fs unexpected failure: %s;\n",# expr); \ + printk(KERN_ERR why "\n"); \ + } \ + val; \ + }) +#define J_EXPECT(expr, why...) __journal_expect(expr, ## why) +#define J_EXPECT_BH(bh, expr, why...) __journal_expect(expr, ## why) +#define J_EXPECT_JH(jh, expr, why...) __journal_expect(expr, ## why) +#endif + +enum jbd_state_bits { + BH_JBD /* Has an attached ext3 journal_head */ + = BH_PrivateStart, + BH_JWrite, /* Being written to log (@@@ DEBUGGING) */ + BH_Freed, /* Has been freed (truncated) */ + BH_Revoked, /* Has been revoked from the log */ + BH_RevokeValid, /* Revoked flag is valid */ + BH_JBDDirty, /* Is dirty but journaled */ + BH_State, /* Pins most journal_head state */ + BH_JournalHead, /* Pins bh->b_private and jh->b_bh */ + BH_Unshadow, /* Dummy bit, for BJ_Shadow wakeup filtering */ +}; + +BUFFER_FNS(JBD, jbd) +BUFFER_FNS(JWrite, jwrite) +BUFFER_FNS(JBDDirty, jbddirty) +TAS_BUFFER_FNS(JBDDirty, jbddirty) +BUFFER_FNS(Revoked, revoked) +TAS_BUFFER_FNS(Revoked, revoked) +BUFFER_FNS(RevokeValid, revokevalid) +TAS_BUFFER_FNS(RevokeValid, revokevalid) +BUFFER_FNS(Freed, freed) + +static inline struct buffer_head *jh2bh(struct journal_head *jh) +{ + return jh->b_bh; +} + +static inline struct journal_head *bh2jh(struct buffer_head *bh) +{ + return bh->b_private; +} + +static inline void jbd_lock_bh_state(struct buffer_head *bh) +{ + bit_spin_lock(BH_State, &bh->b_state); +} + +static inline int jbd_trylock_bh_state(struct buffer_head *bh) +{ + return bit_spin_trylock(BH_State, &bh->b_state); +} + +static inline int jbd_is_locked_bh_state(struct buffer_head *bh) +{ + return bit_spin_is_locked(BH_State, &bh->b_state); +} + +static inline void jbd_unlock_bh_state(struct buffer_head *bh) +{ + bit_spin_unlock(BH_State, &bh->b_state); +} + +static inline void jbd_lock_bh_journal_head(struct buffer_head *bh) +{ + bit_spin_lock(BH_JournalHead, &bh->b_state); +} + +static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh) +{ + bit_spin_unlock(BH_JournalHead, &bh->b_state); +} + +struct jbd_revoke_table_s; + +/** + * struct handle_s - The handle_s type is the concrete type associated with + * handle_t. + * @h_transaction: Which compound transaction is this update a part of? + * @h_buffer_credits: Number of remaining buffers we are allowed to dirty. + * @h_ref: Reference count on this handle + * @h_err: Field for caller's use to track errors through large fs operations + * @h_sync: flag for sync-on-close + * @h_jdata: flag to force data journaling + * @h_aborted: flag indicating fatal error on handle + **/ + +/* Docbook can't yet cope with the bit fields, but will leave the documentation + * in so it can be fixed later. + */ + +struct handle_s +{ + /* Which compound transaction is this update a part of? */ + transaction_t *h_transaction; + + /* Number of remaining buffers we are allowed to dirty: */ + int h_buffer_credits; + + /* Reference count on this handle */ + int h_ref; + + /* Field for caller's use to track errors through large fs */ + /* operations */ + int h_err; + + /* Flags [no locking] */ + unsigned int h_sync: 1; /* sync-on-close */ + unsigned int h_jdata: 1; /* force data journaling */ + unsigned int h_aborted: 1; /* fatal error on handle */ +}; + + +/* The transaction_t type is the guts of the journaling mechanism. It + * tracks a compound transaction through its various states: + * + * RUNNING: accepting new updates + * LOCKED: Updates still running but we don't accept new ones + * RUNDOWN: Updates are tidying up but have finished requesting + * new buffers to modify (state not used for now) + * FLUSH: All updates complete, but we are still writing to disk + * COMMIT: All data on disk, writing commit record + * FINISHED: We still have to keep the transaction for checkpointing. + * + * The transaction keeps track of all of the buffers modified by a + * running transaction, and all of the buffers committed but not yet + * flushed to home for finished transactions. + */ + +/* + * Lock ranking: + * + * j_list_lock + * ->jbd_lock_bh_journal_head() (This is "innermost") + * + * j_state_lock + * ->jbd_lock_bh_state() + * + * jbd_lock_bh_state() + * ->j_list_lock + * + * j_state_lock + * ->t_handle_lock + * + * j_state_lock + * ->j_list_lock (journal_unmap_buffer) + * + */ + +struct transaction_s +{ + /* Pointer to the journal for this transaction. [no locking] */ + journal_t *t_journal; + + /* Sequence number for this transaction [no locking] */ + tid_t t_tid; + + /* + * Transaction's current state + * [no locking - only kjournald alters this] + * FIXME: needs barriers + * KLUDGE: [use j_state_lock] + */ + enum { + T_RUNNING, + T_LOCKED, + T_RUNDOWN, + T_FLUSH, + T_COMMIT, + T_FINISHED + } t_state; + + /* + * Where in the log does this transaction's commit start? [no locking] + */ + unsigned long t_log_start; + + /* Number of buffers on the t_buffers list [j_list_lock] */ + int t_nr_buffers; + + /* + * Doubly-linked circular list of all buffers reserved but not yet + * modified by this transaction [j_list_lock] + */ + struct journal_head *t_reserved_list; + + /* + * Doubly-linked circular list of all buffers under writeout during + * commit [j_list_lock] + */ + struct journal_head *t_locked_list; + + /* + * Doubly-linked circular list of all metadata buffers owned by this + * transaction [j_list_lock] + */ + struct journal_head *t_buffers; + + /* + * Doubly-linked circular list of all data buffers still to be + * flushed before this transaction can be committed [j_list_lock] + */ + struct journal_head *t_sync_datalist; + + /* + * Doubly-linked circular list of all forget buffers (superseded + * buffers which we can un-checkpoint once this transaction commits) + * [j_list_lock] + */ + struct journal_head *t_forget; + + /* + * Doubly-linked circular list of all buffers still to be flushed before + * this transaction can be checkpointed. [j_list_lock] + */ + struct journal_head *t_checkpoint_list; + + /* + * Doubly-linked circular list of all buffers submitted for IO while + * checkpointing. [j_list_lock] + */ + struct journal_head *t_checkpoint_io_list; + + /* + * Doubly-linked circular list of temporary buffers currently undergoing + * IO in the log [j_list_lock] + */ + struct journal_head *t_iobuf_list; + + /* + * Doubly-linked circular list of metadata buffers being shadowed by log + * IO. The IO buffers on the iobuf list and the shadow buffers on this + * list match each other one for one at all times. [j_list_lock] + */ + struct journal_head *t_shadow_list; + + /* + * Doubly-linked circular list of control buffers being written to the + * log. [j_list_lock] + */ + struct journal_head *t_log_list; + + /* + * Protects info related to handles + */ + spinlock_t t_handle_lock; + + /* + * Number of outstanding updates running on this transaction + * [t_handle_lock] + */ + int t_updates; + + /* + * Number of buffers reserved for use by all handles in this transaction + * handle but not yet modified. [t_handle_lock] + */ + int t_outstanding_credits; + + /* + * Forward and backward links for the circular list of all transactions + * awaiting checkpoint. [j_list_lock] + */ + transaction_t *t_cpnext, *t_cpprev; + + /* + * When will the transaction expire (become due for commit), in jiffies? + * [no locking] + */ + unsigned long t_expires; + + /* + * How many handles used this transaction? [t_handle_lock] + */ + int t_handle_count; + +}; + +/** + * struct journal_s - The journal_s type is the concrete type associated with + * journal_t. + * @j_flags: General journaling state flags + * @j_errno: Is there an outstanding uncleared error on the journal (from a + * prior abort)? + * @j_sb_buffer: First part of superblock buffer + * @j_superblock: Second part of superblock buffer + * @j_format_version: Version of the superblock format + * @j_state_lock: Protect the various scalars in the journal + * @j_barrier_count: Number of processes waiting to create a barrier lock + * @j_barrier: The barrier lock itself + * @j_running_transaction: The current running transaction.. + * @j_committing_transaction: the transaction we are pushing to disk + * @j_checkpoint_transactions: a linked circular list of all transactions + * waiting for checkpointing + * @j_wait_transaction_locked: Wait queue for waiting for a locked transaction + * to start committing, or for a barrier lock to be released + * @j_wait_logspace: Wait queue for waiting for checkpointing to complete + * @j_wait_done_commit: Wait queue for waiting for commit to complete + * @j_wait_checkpoint: Wait queue to trigger checkpointing + * @j_wait_commit: Wait queue to trigger commit + * @j_wait_updates: Wait queue to wait for updates to complete + * @j_checkpoint_mutex: Mutex for locking against concurrent checkpoints + * @j_head: Journal head - identifies the first unused block in the journal + * @j_tail: Journal tail - identifies the oldest still-used block in the + * journal. + * @j_free: Journal free - how many free blocks are there in the journal? + * @j_first: The block number of the first usable block + * @j_last: The block number one beyond the last usable block + * @j_dev: Device where we store the journal + * @j_blocksize: blocksize for the location where we store the journal. + * @j_blk_offset: starting block offset for into the device where we store the + * journal + * @j_fs_dev: Device which holds the client fs. For internal journal this will + * be equal to j_dev + * @j_maxlen: Total maximum capacity of the journal region on disk. + * @j_list_lock: Protects the buffer lists and internal buffer state. + * @j_inode: Optional inode where we store the journal. If present, all journal + * block numbers are mapped into this inode via bmap(). + * @j_tail_sequence: Sequence number of the oldest transaction in the log + * @j_transaction_sequence: Sequence number of the next transaction to grant + * @j_commit_sequence: Sequence number of the most recently committed + * transaction + * @j_commit_request: Sequence number of the most recent transaction wanting + * commit + * @j_uuid: Uuid of client object. + * @j_task: Pointer to the current commit thread for this journal + * @j_max_transaction_buffers: Maximum number of metadata buffers to allow in a + * single compound commit transaction + * @j_commit_interval: What is the maximum transaction lifetime before we begin + * a commit? + * @j_commit_timer: The timer used to wakeup the commit thread + * @j_revoke_lock: Protect the revoke table + * @j_revoke: The revoke table - maintains the list of revoked blocks in the + * current transaction. + * @j_revoke_table: alternate revoke tables for j_revoke + * @j_wbuf: array of buffer_heads for journal_commit_transaction + * @j_wbufsize: maximum number of buffer_heads allowed in j_wbuf, the + * number that will fit in j_blocksize + * @j_last_sync_writer: most recent pid which did a synchronous write + * @j_private: An opaque pointer to fs-private information. + */ + +struct journal_s +{ + /* General journaling state flags [j_state_lock] */ + unsigned long j_flags; + + /* + * Is there an outstanding uncleared error on the journal (from a prior + * abort)? [j_state_lock] + */ + int j_errno; + + /* The superblock buffer */ + struct buffer_head *j_sb_buffer; + journal_superblock_t *j_superblock; + + /* Version of the superblock format */ + int j_format_version; + + /* + * Protect the various scalars in the journal + */ + spinlock_t j_state_lock; + + /* + * Number of processes waiting to create a barrier lock [j_state_lock] + */ + int j_barrier_count; + + /* The barrier lock itself */ + struct mutex j_barrier; + + /* + * Transactions: The current running transaction... + * [j_state_lock] [caller holding open handle] + */ + transaction_t *j_running_transaction; + + /* + * the transaction we are pushing to disk + * [j_state_lock] [caller holding open handle] + */ + transaction_t *j_committing_transaction; + + /* + * ... and a linked circular list of all transactions waiting for + * checkpointing. [j_list_lock] + */ + transaction_t *j_checkpoint_transactions; + + /* + * Wait queue for waiting for a locked transaction to start committing, + * or for a barrier lock to be released + */ + wait_queue_head_t j_wait_transaction_locked; + + /* Wait queue for waiting for checkpointing to complete */ + wait_queue_head_t j_wait_logspace; + + /* Wait queue for waiting for commit to complete */ + wait_queue_head_t j_wait_done_commit; + + /* Wait queue to trigger checkpointing */ + wait_queue_head_t j_wait_checkpoint; + + /* Wait queue to trigger commit */ + wait_queue_head_t j_wait_commit; + + /* Wait queue to wait for updates to complete */ + wait_queue_head_t j_wait_updates; + + /* Semaphore for locking against concurrent checkpoints */ + struct mutex j_checkpoint_mutex; + + /* + * Journal head: identifies the first unused block in the journal. + * [j_state_lock] + */ + unsigned long j_head; + + /* + * Journal tail: identifies the oldest still-used block in the journal. + * [j_state_lock] + */ + unsigned long j_tail; + + /* + * Journal free: how many free blocks are there in the journal? + * [j_state_lock] + */ + unsigned long j_free; + + /* + * Journal start and end: the block numbers of the first usable block + * and one beyond the last usable block in the journal. [j_state_lock] + */ + unsigned long j_first; + unsigned long j_last; + + /* + * Device, blocksize and starting block offset for the location where we + * store the journal. + */ + struct block_device *j_dev; + int j_blocksize; + unsigned long j_blk_offset; + + /* + * Device which holds the client fs. For internal journal this will be + * equal to j_dev. + */ + struct block_device *j_fs_dev; + + /* Total maximum capacity of the journal region on disk. */ + unsigned int j_maxlen; + + /* + * Protects the buffer lists and internal buffer state. + */ + spinlock_t j_list_lock; + + /* Optional inode where we store the journal. If present, all */ + /* journal block numbers are mapped into this inode via */ + /* bmap(). */ + struct inode *j_inode; + + /* + * Sequence number of the oldest transaction in the log [j_state_lock] + */ + tid_t j_tail_sequence; + + /* + * Sequence number of the next transaction to grant [j_state_lock] + */ + tid_t j_transaction_sequence; + + /* + * Sequence number of the most recently committed transaction + * [j_state_lock]. + */ + tid_t j_commit_sequence; + + /* + * Sequence number of the most recent transaction wanting commit + * [j_state_lock] + */ + tid_t j_commit_request; + + /* + * Journal uuid: identifies the object (filesystem, LVM volume etc) + * backed by this journal. This will eventually be replaced by an array + * of uuids, allowing us to index multiple devices within a single + * journal and to perform atomic updates across them. + */ + __u8 j_uuid[16]; + + /* Pointer to the current commit thread for this journal */ + struct task_struct *j_task; + + /* + * Maximum number of metadata buffers to allow in a single compound + * commit transaction + */ + int j_max_transaction_buffers; + + /* + * What is the maximum transaction lifetime before we begin a commit? + */ + unsigned long j_commit_interval; + + /* The timer used to wakeup the commit thread: */ + struct timer_list j_commit_timer; + + /* + * The revoke table: maintains the list of revoked blocks in the + * current transaction. [j_revoke_lock] + */ + spinlock_t j_revoke_lock; + struct jbd_revoke_table_s *j_revoke; + struct jbd_revoke_table_s *j_revoke_table[2]; + + /* + * array of bhs for journal_commit_transaction + */ + struct buffer_head **j_wbuf; + int j_wbufsize; + + pid_t j_last_sync_writer; + + /* + * An opaque pointer to fs-private information. ext3 puts its + * superblock pointer here + */ + void *j_private; +}; + +/* + * Journal flag definitions + */ +#define JFS_UNMOUNT 0x001 /* Journal thread is being destroyed */ +#define JFS_ABORT 0x002 /* Journaling has been aborted for errors. */ +#define JFS_ACK_ERR 0x004 /* The errno in the sb has been acked */ +#define JFS_FLUSHED 0x008 /* The journal superblock has been flushed */ +#define JFS_LOADED 0x010 /* The journal superblock has been loaded */ +#define JFS_BARRIER 0x020 /* Use IDE barriers */ + +/* + * Function declarations for the journaling transaction and buffer + * management + */ + +/* Filing buffers */ +extern void __journal_temp_unlink_buffer(struct journal_head *jh); +extern void journal_unfile_buffer(journal_t *, struct journal_head *); +extern void __journal_unfile_buffer(struct journal_head *); +extern void __journal_refile_buffer(struct journal_head *); +extern void journal_refile_buffer(journal_t *, struct journal_head *); +extern void __journal_file_buffer(struct journal_head *, transaction_t *, int); +extern void __journal_free_buffer(struct journal_head *bh); +extern void journal_file_buffer(struct journal_head *, transaction_t *, int); +extern void __journal_clean_data_list(transaction_t *transaction); + +/* Log buffer allocation */ +extern struct journal_head * journal_get_descriptor_buffer(journal_t *); +int journal_next_log_block(journal_t *, unsigned long *); + +/* Commit management */ +extern void journal_commit_transaction(journal_t *); + +/* Checkpoint list management */ +int __journal_clean_checkpoint_list(journal_t *journal); +int __journal_remove_checkpoint(struct journal_head *); +void __journal_insert_checkpoint(struct journal_head *, transaction_t *); + +/* Buffer IO */ +extern int +journal_write_metadata_buffer(transaction_t *transaction, + struct journal_head *jh_in, + struct journal_head **jh_out, + unsigned long blocknr); + +/* Transaction locking */ +extern void __wait_on_journal (journal_t *); + +/* + * Journal locking. + * + * We need to lock the journal during transaction state changes so that nobody + * ever tries to take a handle on the running transaction while we are in the + * middle of moving it to the commit phase. j_state_lock does this. + * + * Note that the locking is completely interrupt unsafe. We never touch + * journal structures from interrupts. + */ + +static inline handle_t *journal_current_handle(void) +{ + return current->journal_info; +} + +/* The journaling code user interface: + * + * Create and destroy handles + * Register buffer modifications against the current transaction. + */ + +extern handle_t *journal_start(journal_t *, int nblocks); +extern int journal_restart (handle_t *, int nblocks); +extern int journal_extend (handle_t *, int nblocks); +extern int journal_get_write_access(handle_t *, struct buffer_head *); +extern int journal_get_create_access (handle_t *, struct buffer_head *); +extern int journal_get_undo_access(handle_t *, struct buffer_head *); +extern int journal_dirty_data (handle_t *, struct buffer_head *); +extern int journal_dirty_metadata (handle_t *, struct buffer_head *); +extern void journal_release_buffer (handle_t *, struct buffer_head *); +extern int journal_forget (handle_t *, struct buffer_head *); +extern void journal_sync_buffer (struct buffer_head *); +extern void journal_invalidatepage(journal_t *, + struct page *, unsigned long); +extern int journal_try_to_free_buffers(journal_t *, struct page *, gfp_t); +extern int journal_stop(handle_t *); +extern int journal_flush (journal_t *); +extern void journal_lock_updates (journal_t *); +extern void journal_unlock_updates (journal_t *); + +extern journal_t * journal_init_dev(struct block_device *bdev, + struct block_device *fs_dev, + int start, int len, int bsize); +extern journal_t * journal_init_inode (struct inode *); +extern int journal_update_format (journal_t *); +extern int journal_check_used_features + (journal_t *, unsigned long, unsigned long, unsigned long); +extern int journal_check_available_features + (journal_t *, unsigned long, unsigned long, unsigned long); +extern int journal_set_features + (journal_t *, unsigned long, unsigned long, unsigned long); +extern int journal_create (journal_t *); +extern int journal_load (journal_t *journal); +extern void journal_destroy (journal_t *); +extern int journal_recover (journal_t *journal); +extern int journal_wipe (journal_t *, int); +extern int journal_skip_recovery (journal_t *); +extern void journal_update_superblock (journal_t *, int); +extern void __journal_abort_hard (journal_t *); +extern void journal_abort (journal_t *, int); +extern int journal_errno (journal_t *); +extern void journal_ack_err (journal_t *); +extern int journal_clear_err (journal_t *); +extern int journal_bmap(journal_t *, unsigned long, unsigned long *); +extern int journal_force_commit(journal_t *); + +/* + * journal_head management + */ +struct journal_head *journal_add_journal_head(struct buffer_head *bh); +struct journal_head *journal_grab_journal_head(struct buffer_head *bh); +void journal_remove_journal_head(struct buffer_head *bh); +void journal_put_journal_head(struct journal_head *jh); + +/* + * handle management + */ +extern kmem_cache_t *jbd_handle_cache; + +static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags) +{ + return kmem_cache_alloc(jbd_handle_cache, gfp_flags); +} + +static inline void jbd_free_handle(handle_t *handle) +{ + kmem_cache_free(jbd_handle_cache, handle); +} + +/* Primary revoke support */ +#define JOURNAL_REVOKE_DEFAULT_HASH 256 +extern int journal_init_revoke(journal_t *, int); +extern void journal_destroy_revoke_caches(void); +extern int journal_init_revoke_caches(void); + +extern void journal_destroy_revoke(journal_t *); +extern int journal_revoke (handle_t *, + unsigned long, struct buffer_head *); +extern int journal_cancel_revoke(handle_t *, struct journal_head *); +extern void journal_write_revoke_records(journal_t *, transaction_t *); + +/* Recovery revoke support */ +extern int journal_set_revoke(journal_t *, unsigned long, tid_t); +extern int journal_test_revoke(journal_t *, unsigned long, tid_t); +extern void journal_clear_revoke(journal_t *); +extern void journal_switch_revoke_table(journal_t *journal); + +/* + * The log thread user interface: + * + * Request space in the current transaction, and force transaction commit + * transitions on demand. + */ + +int __log_space_left(journal_t *); /* Called with journal locked */ +int log_start_commit(journal_t *journal, tid_t tid); +int __log_start_commit(journal_t *journal, tid_t tid); +int journal_start_commit(journal_t *journal, tid_t *tid); +int journal_force_commit_nested(journal_t *journal); +int log_wait_commit(journal_t *journal, tid_t tid); +int log_do_checkpoint(journal_t *journal); + +void __log_wait_for_space(journal_t *journal); +extern void __journal_drop_transaction(journal_t *, transaction_t *); +extern int cleanup_journal_tail(journal_t *); + +/* Debugging code only: */ + +#define jbd_ENOSYS() \ +do { \ + printk (KERN_ERR "JBD unimplemented function %s\n", __FUNCTION__); \ + current->state = TASK_UNINTERRUPTIBLE; \ + schedule(); \ +} while (1) + +/* + * is_journal_abort + * + * Simple test wrapper function to test the JFS_ABORT state flag. This + * bit, when set, indicates that we have had a fatal error somewhere, + * either inside the journaling layer or indicated to us by the client + * (eg. ext3), and that we and should not commit any further + * transactions. + */ + +static inline int is_journal_aborted(journal_t *journal) +{ + return journal->j_flags & JFS_ABORT; +} + +static inline int is_handle_aborted(handle_t *handle) +{ + if (handle->h_aborted) + return 1; + return is_journal_aborted(handle->h_transaction->t_journal); +} + +static inline void journal_abort_handle(handle_t *handle) +{ + handle->h_aborted = 1; +} + +#endif /* __KERNEL__ */ + +/* Comparison functions for transaction IDs: perform comparisons using + * modulo arithmetic so that they work over sequence number wraps. */ + +static inline int tid_gt(tid_t x, tid_t y) +{ + int difference = (x - y); + return (difference > 0); +} + +static inline int tid_geq(tid_t x, tid_t y) +{ + int difference = (x - y); + return (difference >= 0); +} + +extern int journal_blocks_per_page(struct inode *inode); + +/* + * Return the minimum number of blocks which must be free in the journal + * before a new transaction may be started. Must be called under j_state_lock. + */ +static inline int jbd_space_needed(journal_t *journal) +{ + int nblocks = journal->j_max_transaction_buffers; + if (journal->j_committing_transaction) + nblocks += journal->j_committing_transaction-> + t_outstanding_credits; + return nblocks; +} + +/* + * Definitions which augment the buffer_head layer + */ + +/* journaling buffer types */ +#define BJ_None 0 /* Not journaled */ +#define BJ_SyncData 1 /* Normal data: flush before commit */ +#define BJ_Metadata 2 /* Normal journaled metadata */ +#define BJ_Forget 3 /* Buffer superseded by this transaction */ +#define BJ_IO 4 /* Buffer is for temporary IO use */ +#define BJ_Shadow 5 /* Buffer contents being shadowed to the log */ +#define BJ_LogCtl 6 /* Buffer contains log descriptors */ +#define BJ_Reserved 7 /* Buffer is reserved for access by journal */ +#define BJ_Locked 8 /* Locked for I/O during commit */ +#define BJ_Types 9 + +extern int jbd_blocks_per_page(struct inode *inode); + +#ifdef __KERNEL__ + +#define buffer_trace_init(bh) do {} while (0) +#define print_buffer_fields(bh) do {} while (0) +#define print_buffer_trace(bh) do {} while (0) +#define BUFFER_TRACE(bh, info) do {} while (0) +#define BUFFER_TRACE2(bh, bh2, info) do {} while (0) +#define JBUFFER_TRACE(jh, info) do {} while (0) + +#endif /* __KERNEL__ */ + +#endif /* _LINUX_JBD_H */ -- cgit v0.10.2 From f7f4bccb729844a0fa873e224e3a6f7eeed095bb Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:20:59 -0700 Subject: [PATCH] jbd2: rename jbd2 symbols to avoid duplication of jbd symbols Mingming Cao originally did this work, and Shaggy reproduced it using some scripts from her. Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/Makefile b/fs/jbd2/Makefile index 54aca48..802a341 100644 --- a/fs/jbd2/Makefile +++ b/fs/jbd2/Makefile @@ -2,6 +2,6 @@ # Makefile for the linux journaling routines. # -obj-$(CONFIG_JBD) += jbd.o +obj-$(CONFIG_JBD2) += jbd2.o -jbd-objs := transaction.o commit.o recovery.o checkpoint.o revoke.o journal.o +jbd2-objs := transaction.o commit.o recovery.o checkpoint.o revoke.o journal.o diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 0208cc7..68039fa 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -19,7 +19,7 @@ #include #include -#include +#include #include #include @@ -95,9 +95,9 @@ static int __try_to_free_cp_buf(struct journal_head *jh) if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) { JBUFFER_TRACE(jh, "remove from checkpoint list"); - ret = __journal_remove_checkpoint(jh) + 1; + ret = __jbd2_journal_remove_checkpoint(jh) + 1; jbd_unlock_bh_state(bh); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); BUFFER_TRACE(bh, "release"); __brelse(bh); } else { @@ -107,19 +107,19 @@ static int __try_to_free_cp_buf(struct journal_head *jh) } /* - * __log_wait_for_space: wait until there is space in the journal. + * __jbd2_log_wait_for_space: wait until there is space in the journal. * * Called under j-state_lock *only*. It will be unlocked if we have to wait * for a checkpoint to free up some space in the log. */ -void __log_wait_for_space(journal_t *journal) +void __jbd2_log_wait_for_space(journal_t *journal) { int nblocks; assert_spin_locked(&journal->j_state_lock); nblocks = jbd_space_needed(journal); - while (__log_space_left(journal) < nblocks) { - if (journal->j_flags & JFS_ABORT) + while (__jbd2_log_space_left(journal) < nblocks) { + if (journal->j_flags & JBD2_ABORT) return; spin_unlock(&journal->j_state_lock); mutex_lock(&journal->j_checkpoint_mutex); @@ -130,9 +130,9 @@ void __log_wait_for_space(journal_t *journal) */ spin_lock(&journal->j_state_lock); nblocks = jbd_space_needed(journal); - if (__log_space_left(journal) < nblocks) { + if (__jbd2_log_space_left(journal) < nblocks) { spin_unlock(&journal->j_state_lock); - log_do_checkpoint(journal); + jbd2_log_do_checkpoint(journal); spin_lock(&journal->j_state_lock); } mutex_unlock(&journal->j_checkpoint_mutex); @@ -198,9 +198,9 @@ restart: * Now in whatever state the buffer currently is, we know that * it has been written out and so we can drop it from the list */ - released = __journal_remove_checkpoint(jh); + released = __jbd2_journal_remove_checkpoint(jh); jbd_unlock_bh_state(bh); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); __brelse(bh); } } @@ -252,16 +252,16 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); - log_start_commit(journal, tid); - log_wait_commit(journal, tid); + jbd2_log_start_commit(journal, tid); + jbd2_log_wait_commit(journal, tid); ret = 1; } else if (!buffer_dirty(bh)) { J_ASSERT_JH(jh, !buffer_jbddirty(bh)); BUFFER_TRACE(bh, "remove from checkpoint"); - __journal_remove_checkpoint(jh); + __jbd2_journal_remove_checkpoint(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); __brelse(bh); ret = 1; } else { @@ -296,7 +296,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, * * The journal should be locked before calling this function. */ -int log_do_checkpoint(journal_t *journal) +int jbd2_log_do_checkpoint(journal_t *journal) { transaction_t *transaction; tid_t this_tid; @@ -309,7 +309,7 @@ int log_do_checkpoint(journal_t *journal) * don't need checkpointing, just eliminate them from the * journal straight away. */ - result = cleanup_journal_tail(journal); + result = jbd2_cleanup_journal_tail(journal); jbd_debug(1, "cleanup_journal_tail returned %d\n", result); if (result <= 0) return result; @@ -374,7 +374,7 @@ restart: } out: spin_unlock(&journal->j_list_lock); - result = cleanup_journal_tail(journal); + result = jbd2_cleanup_journal_tail(journal); if (result < 0) return result; return 0; @@ -397,7 +397,7 @@ out: * we have an abort error outstanding. */ -int cleanup_journal_tail(journal_t *journal) +int jbd2_cleanup_journal_tail(journal_t *journal) { transaction_t * transaction; tid_t first_tid; @@ -452,8 +452,8 @@ int cleanup_journal_tail(journal_t *journal) journal->j_tail_sequence = first_tid; journal->j_tail = blocknr; spin_unlock(&journal->j_state_lock); - if (!(journal->j_flags & JFS_ABORT)) - journal_update_superblock(journal, 1); + if (!(journal->j_flags & JBD2_ABORT)) + jbd2_journal_update_superblock(journal, 1); return 0; } @@ -518,7 +518,7 @@ static int journal_clean_one_cp_list(struct journal_head *jh, int *released) * Returns number of buffers reaped (for debug) */ -int __journal_clean_checkpoint_list(journal_t *journal) +int __jbd2_journal_clean_checkpoint_list(journal_t *journal) { transaction_t *transaction, *last_transaction, *next_transaction; int ret = 0; @@ -578,7 +578,7 @@ out: * This function is called with jbd_lock_bh_state(jh2bh(jh)) */ -int __journal_remove_checkpoint(struct journal_head *jh) +int __jbd2_journal_remove_checkpoint(struct journal_head *jh) { transaction_t *transaction; journal_t *journal; @@ -607,7 +607,7 @@ int __journal_remove_checkpoint(struct journal_head *jh) * dropped! * * The locking here around j_committing_transaction is a bit sleazy. - * See the comment at the end of journal_commit_transaction(). + * See the comment at the end of jbd2_journal_commit_transaction(). */ if (transaction == journal->j_committing_transaction) { JBUFFER_TRACE(jh, "belongs to committing transaction"); @@ -617,7 +617,7 @@ int __journal_remove_checkpoint(struct journal_head *jh) /* OK, that was the last buffer for the transaction: we can now safely remove this transaction from the log */ - __journal_drop_transaction(journal, transaction); + __jbd2_journal_drop_transaction(journal, transaction); /* Just in case anybody was waiting for more transactions to be checkpointed... */ @@ -636,7 +636,7 @@ out: * Called with the journal locked. * Called with j_list_lock held. */ -void __journal_insert_checkpoint(struct journal_head *jh, +void __jbd2_journal_insert_checkpoint(struct journal_head *jh, transaction_t *transaction) { JBUFFER_TRACE(jh, "entry"); @@ -666,7 +666,7 @@ void __journal_insert_checkpoint(struct journal_head *jh, * Called with j_list_lock held. */ -void __journal_drop_transaction(journal_t *journal, transaction_t *transaction) +void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transaction) { assert_spin_locked(&journal->j_list_lock); if (transaction->t_cpnext) { diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 10be512..b1a4eaf 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -1,5 +1,5 @@ /* - * linux/fs/jbd/commit.c + * linux/fs/jbd2/commit.c * * Written by Stephen C. Tweedie , 1998 * @@ -15,7 +15,7 @@ #include #include -#include +#include #include #include #include @@ -111,7 +111,7 @@ static int journal_write_commit_record(journal_t *journal, if (is_journal_aborted(journal)) return 0; - descriptor = journal_get_descriptor_buffer(journal); + descriptor = jbd2_journal_get_descriptor_buffer(journal); if (!descriptor) return 1; @@ -120,14 +120,14 @@ static int journal_write_commit_record(journal_t *journal, /* AKPM: buglet - add `i' to tmp! */ for (i = 0; i < bh->b_size; i += 512) { journal_header_t *tmp = (journal_header_t*)bh->b_data; - tmp->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); - tmp->h_blocktype = cpu_to_be32(JFS_COMMIT_BLOCK); + tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); } JBUFFER_TRACE(descriptor, "write commit block"); set_buffer_dirty(bh); - if (journal->j_flags & JFS_BARRIER) { + if (journal->j_flags & JBD2_BARRIER) { set_buffer_ordered(bh); barrier_done = 1; } @@ -145,7 +145,7 @@ static int journal_write_commit_record(journal_t *journal, "disabling barriers\n", bdevname(journal->j_dev, b)); spin_lock(&journal->j_state_lock); - journal->j_flags &= ~JFS_BARRIER; + journal->j_flags &= ~JBD2_BARRIER; spin_unlock(&journal->j_state_lock); /* And try again, without the barrier */ @@ -155,7 +155,7 @@ static int journal_write_commit_record(journal_t *journal, ret = sync_dirty_buffer(bh); } put_bh(bh); /* One for getblk() */ - journal_put_journal_head(descriptor); + jbd2_journal_put_journal_head(descriptor); return (ret == -EIO); } @@ -239,7 +239,7 @@ write_out_data: if (locked && test_clear_buffer_dirty(bh)) { BUFFER_TRACE(bh, "needs writeout, adding to array"); wbuf[bufs++] = bh; - __journal_file_buffer(jh, commit_transaction, + __jbd2_journal_file_buffer(jh, commit_transaction, BJ_Locked); jbd_unlock_bh_state(bh); if (bufs == journal->j_wbufsize) { @@ -251,13 +251,13 @@ write_out_data: } else { BUFFER_TRACE(bh, "writeout complete: unfile"); - __journal_unfile_buffer(jh); + __jbd2_journal_unfile_buffer(jh); jbd_unlock_bh_state(bh); if (locked) unlock_buffer(bh); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); /* Once for our safety reference, once for - * journal_remove_journal_head() */ + * jbd2_journal_remove_journal_head() */ put_bh(bh); put_bh(bh); } @@ -272,12 +272,12 @@ write_out_data: } /* - * journal_commit_transaction + * jbd2_journal_commit_transaction * * The primary function for committing a transaction to the log. This * function is called by the journal thread to begin a complete commit. */ -void journal_commit_transaction(journal_t *journal) +void jbd2_journal_commit_transaction(journal_t *journal) { transaction_t *commit_transaction; struct journal_head *jh, *new_jh, *descriptor; @@ -305,10 +305,10 @@ void journal_commit_transaction(journal_t *journal) spin_unlock(&journal->j_list_lock); #endif - /* Do we need to erase the effects of a prior journal_flush? */ - if (journal->j_flags & JFS_FLUSHED) { + /* Do we need to erase the effects of a prior jbd2_journal_flush? */ + if (journal->j_flags & JBD2_FLUSHED) { jbd_debug(3, "super block updated\n"); - journal_update_superblock(journal, 1); + jbd2_journal_update_superblock(journal, 1); } else { jbd_debug(3, "superblock not updated\n"); } @@ -350,7 +350,7 @@ void journal_commit_transaction(journal_t *journal) * BJ_Reserved buffers. Note, it is _not_ permissible to assume * that there are no such buffers: if a large filesystem * operation like a truncate needs to split itself over multiple - * transactions, then it may try to do a journal_restart() while + * transactions, then it may try to do a jbd2_journal_restart() while * there are still BJ_Reserved buffers outstanding. These must * be released cleanly from the current transaction. * @@ -358,25 +358,25 @@ void journal_commit_transaction(journal_t *journal) * again before modifying the buffer in the new transaction, but * we do not require it to remember exactly which old buffers it * has reserved. This is consistent with the existing behaviour - * that multiple journal_get_write_access() calls to the same + * that multiple jbd2_journal_get_write_access() calls to the same * buffer are perfectly permissable. */ while (commit_transaction->t_reserved_list) { jh = commit_transaction->t_reserved_list; JBUFFER_TRACE(jh, "reserved, unused: refile"); /* - * A journal_get_undo_access()+journal_release_buffer() may + * A jbd2_journal_get_undo_access()+jbd2_journal_release_buffer() may * leave undo-committed data. */ if (jh->b_committed_data) { struct buffer_head *bh = jh2bh(jh); jbd_lock_bh_state(bh); - jbd_slab_free(jh->b_committed_data, bh->b_size); + jbd2_slab_free(jh->b_committed_data, bh->b_size); jh->b_committed_data = NULL; jbd_unlock_bh_state(bh); } - journal_refile_buffer(journal, jh); + jbd2_journal_refile_buffer(journal, jh); } /* @@ -385,7 +385,7 @@ void journal_commit_transaction(journal_t *journal) * frees some memory */ spin_lock(&journal->j_list_lock); - __journal_clean_checkpoint_list(journal); + __jbd2_journal_clean_checkpoint_list(journal); spin_unlock(&journal->j_list_lock); jbd_debug (3, "JBD: commit phase 1\n"); @@ -393,7 +393,7 @@ void journal_commit_transaction(journal_t *journal) /* * Switch to a new revoke table. */ - journal_switch_revoke_table(journal); + jbd2_journal_switch_revoke_table(journal); commit_transaction->t_state = T_FLUSH; journal->j_committing_transaction = commit_transaction; @@ -450,9 +450,9 @@ void journal_commit_transaction(journal_t *journal) continue; } if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { - __journal_unfile_buffer(jh); + __jbd2_journal_unfile_buffer(jh); jbd_unlock_bh_state(bh); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); put_bh(bh); } else { jbd_unlock_bh_state(bh); @@ -463,9 +463,9 @@ void journal_commit_transaction(journal_t *journal) spin_unlock(&journal->j_list_lock); if (err) - __journal_abort_hard(journal); + __jbd2_journal_abort_hard(journal); - journal_write_revoke_records(journal, commit_transaction); + jbd2_journal_write_revoke_records(journal, commit_transaction); jbd_debug(3, "JBD: commit phase 2\n"); @@ -499,7 +499,7 @@ void journal_commit_transaction(journal_t *journal) if (is_journal_aborted(journal)) { JBUFFER_TRACE(jh, "journal is aborting: refile"); - journal_refile_buffer(journal, jh); + jbd2_journal_refile_buffer(journal, jh); /* If that was the last one, we need to clean up * any descriptor buffers which may have been * already allocated, even if we are now @@ -519,9 +519,9 @@ void journal_commit_transaction(journal_t *journal) jbd_debug(4, "JBD: get descriptor\n"); - descriptor = journal_get_descriptor_buffer(journal); + descriptor = jbd2_journal_get_descriptor_buffer(journal); if (!descriptor) { - __journal_abort_hard(journal); + __jbd2_journal_abort_hard(journal); continue; } @@ -529,8 +529,8 @@ void journal_commit_transaction(journal_t *journal) jbd_debug(4, "JBD: got buffer %llu (%p)\n", (unsigned long long)bh->b_blocknr, bh->b_data); header = (journal_header_t *)&bh->b_data[0]; - header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); - header->h_blocktype = cpu_to_be32(JFS_DESCRIPTOR_BLOCK); + header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JBD2_DESCRIPTOR_BLOCK); header->h_sequence = cpu_to_be32(commit_transaction->t_tid); tagp = &bh->b_data[sizeof(journal_header_t)]; @@ -543,25 +543,25 @@ void journal_commit_transaction(journal_t *journal) /* Record it so that we can wait for IO completion later */ BUFFER_TRACE(bh, "ph3: file as descriptor"); - journal_file_buffer(descriptor, commit_transaction, + jbd2_journal_file_buffer(descriptor, commit_transaction, BJ_LogCtl); } /* Where is the buffer to be written? */ - err = journal_next_log_block(journal, &blocknr); + err = jbd2_journal_next_log_block(journal, &blocknr); /* If the block mapping failed, just abandon the buffer and repeat this loop: we'll fall into the refile-on-abort condition above. */ if (err) { - __journal_abort_hard(journal); + __jbd2_journal_abort_hard(journal); continue; } /* * start_this_handle() uses t_outstanding_credits to determine * the free space in the log, but this counter is changed - * by journal_next_log_block() also. + * by jbd2_journal_next_log_block() also. */ commit_transaction->t_outstanding_credits--; @@ -576,13 +576,13 @@ void journal_commit_transaction(journal_t *journal) set_bit(BH_JWrite, &jh2bh(jh)->b_state); /* - * akpm: journal_write_metadata_buffer() sets + * akpm: jbd2_journal_write_metadata_buffer() sets * new_bh->b_transaction to commit_transaction. * We need to clean this up before we release new_bh * (which is of type BJ_IO) */ JBUFFER_TRACE(jh, "ph3: write metadata"); - flags = journal_write_metadata_buffer(commit_transaction, + flags = jbd2_journal_write_metadata_buffer(commit_transaction, jh, &new_jh, blocknr); set_bit(BH_JWrite, &jh2bh(new_jh)->b_state); wbuf[bufs++] = jh2bh(new_jh); @@ -592,9 +592,9 @@ void journal_commit_transaction(journal_t *journal) tag_flag = 0; if (flags & 1) - tag_flag |= JFS_FLAG_ESCAPE; + tag_flag |= JBD2_FLAG_ESCAPE; if (!first_tag) - tag_flag |= JFS_FLAG_SAME_UUID; + tag_flag |= JBD2_FLAG_SAME_UUID; tag = (journal_block_tag_t *) tagp; tag->t_blocknr = cpu_to_be32(jh2bh(jh)->b_blocknr); @@ -622,7 +622,7 @@ void journal_commit_transaction(journal_t *journal) submitting the IOs. "tag" still points to the last tag we set up. */ - tag->t_flags |= cpu_to_be32(JFS_FLAG_LAST_TAG); + tag->t_flags |= cpu_to_be32(JBD2_FLAG_LAST_TAG); start_journal_io: for (i = 0; i < bufs; i++) { @@ -678,14 +678,14 @@ wait_for_iobuf: clear_buffer_jwrite(bh); JBUFFER_TRACE(jh, "ph4: unfile after journal write"); - journal_unfile_buffer(journal, jh); + jbd2_journal_unfile_buffer(journal, jh); /* * ->t_iobuf_list should contain only dummy buffer_heads - * which were created by journal_write_metadata_buffer(). + * which were created by jbd2_journal_write_metadata_buffer(). */ BUFFER_TRACE(bh, "dumping temporary bh"); - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); __brelse(bh); J_ASSERT_BH(bh, atomic_read(&bh->b_count) == 0); free_buffer_head(bh); @@ -702,7 +702,7 @@ wait_for_iobuf: we finally commit, we can do any checkpointing required. */ JBUFFER_TRACE(jh, "file as BJ_Forget"); - journal_file_buffer(jh, commit_transaction, BJ_Forget); + jbd2_journal_file_buffer(jh, commit_transaction, BJ_Forget); /* Wake up any transactions which were waiting for this IO to complete */ wake_up_bit(&bh->b_state, BH_Unshadow); @@ -733,8 +733,8 @@ wait_for_iobuf: BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile"); clear_buffer_jwrite(bh); - journal_unfile_buffer(journal, jh); - journal_put_journal_head(jh); + jbd2_journal_unfile_buffer(journal, jh); + jbd2_journal_put_journal_head(jh); __brelse(bh); /* One for getblk */ /* AKPM: bforget here */ } @@ -745,7 +745,7 @@ wait_for_iobuf: err = -EIO; if (err) - __journal_abort_hard(journal); + __jbd2_journal_abort_hard(journal); /* End of a transaction! Finally, we can do checkpoint processing: any buffers committed as a result of this @@ -789,14 +789,14 @@ restart_loop: * Otherwise, we can just throw away the frozen data now. */ if (jh->b_committed_data) { - jbd_slab_free(jh->b_committed_data, bh->b_size); + jbd2_slab_free(jh->b_committed_data, bh->b_size); jh->b_committed_data = NULL; if (jh->b_frozen_data) { jh->b_committed_data = jh->b_frozen_data; jh->b_frozen_data = NULL; } } else if (jh->b_frozen_data) { - jbd_slab_free(jh->b_frozen_data, bh->b_size); + jbd2_slab_free(jh->b_frozen_data, bh->b_size); jh->b_frozen_data = NULL; } @@ -804,12 +804,12 @@ restart_loop: cp_transaction = jh->b_cp_transaction; if (cp_transaction) { JBUFFER_TRACE(jh, "remove from old cp transaction"); - __journal_remove_checkpoint(jh); + __jbd2_journal_remove_checkpoint(jh); } /* Only re-checkpoint the buffer_head if it is marked * dirty. If the buffer was added to the BJ_Forget list - * by journal_forget, it may no longer be dirty and + * by jbd2_journal_forget, it may no longer be dirty and * there's no point in keeping a checkpoint record for * it. */ @@ -828,9 +828,9 @@ restart_loop: if (buffer_jbddirty(bh)) { JBUFFER_TRACE(jh, "add to new checkpointing trans"); - __journal_insert_checkpoint(jh, commit_transaction); + __jbd2_journal_insert_checkpoint(jh, commit_transaction); JBUFFER_TRACE(jh, "refile for checkpoint writeback"); - __journal_refile_buffer(jh); + __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); } else { J_ASSERT_BH(bh, !buffer_dirty(bh)); @@ -842,11 +842,11 @@ restart_loop: * disk and before we process the buffer on BJ_Forget * list. */ JBUFFER_TRACE(jh, "refile or unfile freed buffer"); - __journal_refile_buffer(jh); + __jbd2_journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); /* needs a brelse */ - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); release_buffer_page(bh); } else jbd_unlock_bh_state(bh); @@ -856,9 +856,9 @@ restart_loop: spin_unlock(&journal->j_list_lock); /* * This is a bit sleazy. We borrow j_list_lock to protect - * journal->j_committing_transaction in __journal_remove_checkpoint. - * Really, __journal_remove_checkpoint should be using j_state_lock but - * it's a bit hassle to hold that across __journal_remove_checkpoint + * journal->j_committing_transaction in __jbd2_journal_remove_checkpoint. + * Really, __jbd2_journal_remove_checkpoint should be using j_state_lock but + * it's a bit hassle to hold that across __jbd2_journal_remove_checkpoint */ spin_lock(&journal->j_state_lock); spin_lock(&journal->j_list_lock); @@ -885,7 +885,7 @@ restart_loop: spin_unlock(&journal->j_state_lock); if (commit_transaction->t_checkpoint_list == NULL) { - __journal_drop_transaction(journal, commit_transaction); + __jbd2_journal_drop_transaction(journal, commit_transaction); } else { if (journal->j_checkpoint_transactions == NULL) { journal->j_checkpoint_transactions = commit_transaction; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c518dd8..3fbbba2 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1,5 +1,5 @@ /* - * linux/fs/jbd/journal.c + * linux/fs/jbd2/journal.c * * Written by Stephen C. Tweedie , 1998 * @@ -25,7 +25,7 @@ #include #include #include -#include +#include #include #include #include @@ -40,51 +40,51 @@ #include #include -EXPORT_SYMBOL(journal_start); -EXPORT_SYMBOL(journal_restart); -EXPORT_SYMBOL(journal_extend); -EXPORT_SYMBOL(journal_stop); -EXPORT_SYMBOL(journal_lock_updates); -EXPORT_SYMBOL(journal_unlock_updates); -EXPORT_SYMBOL(journal_get_write_access); -EXPORT_SYMBOL(journal_get_create_access); -EXPORT_SYMBOL(journal_get_undo_access); -EXPORT_SYMBOL(journal_dirty_data); -EXPORT_SYMBOL(journal_dirty_metadata); -EXPORT_SYMBOL(journal_release_buffer); -EXPORT_SYMBOL(journal_forget); +EXPORT_SYMBOL(jbd2_journal_start); +EXPORT_SYMBOL(jbd2_journal_restart); +EXPORT_SYMBOL(jbd2_journal_extend); +EXPORT_SYMBOL(jbd2_journal_stop); +EXPORT_SYMBOL(jbd2_journal_lock_updates); +EXPORT_SYMBOL(jbd2_journal_unlock_updates); +EXPORT_SYMBOL(jbd2_journal_get_write_access); +EXPORT_SYMBOL(jbd2_journal_get_create_access); +EXPORT_SYMBOL(jbd2_journal_get_undo_access); +EXPORT_SYMBOL(jbd2_journal_dirty_data); +EXPORT_SYMBOL(jbd2_journal_dirty_metadata); +EXPORT_SYMBOL(jbd2_journal_release_buffer); +EXPORT_SYMBOL(jbd2_journal_forget); #if 0 EXPORT_SYMBOL(journal_sync_buffer); #endif -EXPORT_SYMBOL(journal_flush); -EXPORT_SYMBOL(journal_revoke); - -EXPORT_SYMBOL(journal_init_dev); -EXPORT_SYMBOL(journal_init_inode); -EXPORT_SYMBOL(journal_update_format); -EXPORT_SYMBOL(journal_check_used_features); -EXPORT_SYMBOL(journal_check_available_features); -EXPORT_SYMBOL(journal_set_features); -EXPORT_SYMBOL(journal_create); -EXPORT_SYMBOL(journal_load); -EXPORT_SYMBOL(journal_destroy); -EXPORT_SYMBOL(journal_update_superblock); -EXPORT_SYMBOL(journal_abort); -EXPORT_SYMBOL(journal_errno); -EXPORT_SYMBOL(journal_ack_err); -EXPORT_SYMBOL(journal_clear_err); -EXPORT_SYMBOL(log_wait_commit); -EXPORT_SYMBOL(journal_start_commit); -EXPORT_SYMBOL(journal_force_commit_nested); -EXPORT_SYMBOL(journal_wipe); -EXPORT_SYMBOL(journal_blocks_per_page); -EXPORT_SYMBOL(journal_invalidatepage); -EXPORT_SYMBOL(journal_try_to_free_buffers); -EXPORT_SYMBOL(journal_force_commit); +EXPORT_SYMBOL(jbd2_journal_flush); +EXPORT_SYMBOL(jbd2_journal_revoke); + +EXPORT_SYMBOL(jbd2_journal_init_dev); +EXPORT_SYMBOL(jbd2_journal_init_inode); +EXPORT_SYMBOL(jbd2_journal_update_format); +EXPORT_SYMBOL(jbd2_journal_check_used_features); +EXPORT_SYMBOL(jbd2_journal_check_available_features); +EXPORT_SYMBOL(jbd2_journal_set_features); +EXPORT_SYMBOL(jbd2_journal_create); +EXPORT_SYMBOL(jbd2_journal_load); +EXPORT_SYMBOL(jbd2_journal_destroy); +EXPORT_SYMBOL(jbd2_journal_update_superblock); +EXPORT_SYMBOL(jbd2_journal_abort); +EXPORT_SYMBOL(jbd2_journal_errno); +EXPORT_SYMBOL(jbd2_journal_ack_err); +EXPORT_SYMBOL(jbd2_journal_clear_err); +EXPORT_SYMBOL(jbd2_log_wait_commit); +EXPORT_SYMBOL(jbd2_journal_start_commit); +EXPORT_SYMBOL(jbd2_journal_force_commit_nested); +EXPORT_SYMBOL(jbd2_journal_wipe); +EXPORT_SYMBOL(jbd2_journal_blocks_per_page); +EXPORT_SYMBOL(jbd2_journal_invalidatepage); +EXPORT_SYMBOL(jbd2_journal_try_to_free_buffers); +EXPORT_SYMBOL(jbd2_journal_force_commit); static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *); static void __journal_abort_soft (journal_t *journal, int errno); -static int journal_create_jbd_slab(size_t slab_size); +static int jbd2_journal_create_jbd_slab(size_t slab_size); /* * Helper function used to manage commit timeouts @@ -98,7 +98,7 @@ static void commit_timeout(unsigned long __data) } /* - * kjournald: The main thread function used to manage a logging device + * kjournald2: The main thread function used to manage a logging device * journal. * * This kernel thread is responsible for two things: @@ -113,7 +113,7 @@ static void commit_timeout(unsigned long __data) * known as checkpointing, and this thread is responsible for that job. */ -static int kjournald(void *arg) +static int kjournald2(void *arg) { journal_t *journal = arg; transaction_t *transaction; @@ -129,7 +129,7 @@ static int kjournald(void *arg) journal->j_task = current; wake_up(&journal->j_wait_done_commit); - printk(KERN_INFO "kjournald starting. Commit interval %ld seconds\n", + printk(KERN_INFO "kjournald2 starting. Commit interval %ld seconds\n", journal->j_commit_interval / HZ); /* @@ -138,7 +138,7 @@ static int kjournald(void *arg) spin_lock(&journal->j_state_lock); loop: - if (journal->j_flags & JFS_UNMOUNT) + if (journal->j_flags & JBD2_UNMOUNT) goto end_loop; jbd_debug(1, "commit_sequence=%d, commit_request=%d\n", @@ -148,7 +148,7 @@ loop: jbd_debug(1, "OK, requests differ\n"); spin_unlock(&journal->j_state_lock); del_timer_sync(&journal->j_commit_timer); - journal_commit_transaction(journal); + jbd2_journal_commit_transaction(journal); spin_lock(&journal->j_state_lock); goto loop; } @@ -160,7 +160,7 @@ loop: * good idea, because that depends on threads that may * be already stopped. */ - jbd_debug(1, "Now suspending kjournald\n"); + jbd_debug(1, "Now suspending kjournald2\n"); spin_unlock(&journal->j_state_lock); refrigerator(); spin_lock(&journal->j_state_lock); @@ -180,7 +180,7 @@ loop: if (transaction && time_after_eq(jiffies, transaction->t_expires)) should_sleep = 0; - if (journal->j_flags & JFS_UNMOUNT) + if (journal->j_flags & JBD2_UNMOUNT) should_sleep = 0; if (should_sleep) { spin_unlock(&journal->j_state_lock); @@ -190,7 +190,7 @@ loop: finish_wait(&journal->j_wait_commit, &wait); } - jbd_debug(1, "kjournald wakes\n"); + jbd_debug(1, "kjournald2 wakes\n"); /* * Were we woken up by a commit wakeup event? @@ -211,16 +211,16 @@ end_loop: return 0; } -static void journal_start_thread(journal_t *journal) +static void jbd2_journal_start_thread(journal_t *journal) { - kthread_run(kjournald, journal, "kjournald"); + kthread_run(kjournald2, journal, "kjournald2"); wait_event(journal->j_wait_done_commit, journal->j_task != 0); } static void journal_kill_thread(journal_t *journal) { spin_lock(&journal->j_state_lock); - journal->j_flags |= JFS_UNMOUNT; + journal->j_flags |= JBD2_UNMOUNT; while (journal->j_task) { wake_up(&journal->j_wait_commit); @@ -232,7 +232,7 @@ static void journal_kill_thread(journal_t *journal) } /* - * journal_write_metadata_buffer: write a metadata buffer to the journal. + * jbd2_journal_write_metadata_buffer: write a metadata buffer to the journal. * * Writes a metadata buffer to a given disk block. The actual IO is not * performed but a new buffer_head is constructed which labels the data @@ -240,7 +240,7 @@ static void journal_kill_thread(journal_t *journal) * * Any magic-number escaping which needs to be done will cause a * copy-out here. If the buffer happens to start with the - * JFS_MAGIC_NUMBER, then we can't write it to the log directly: the + * JBD2_MAGIC_NUMBER, then we can't write it to the log directly: the * magic number is only written to the log for descripter blocks. In * this case, we copy the data and replace the first word with 0, and we * return a result code which indicates that this buffer needs to be @@ -268,7 +268,7 @@ static void journal_kill_thread(journal_t *journal) * Bit 1 set == buffer copy-out performed (kfree the data after IO) */ -int journal_write_metadata_buffer(transaction_t *transaction, +int jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, struct journal_head **jh_out, unsigned long blocknr) @@ -316,7 +316,7 @@ repeat: * Check for escaping */ if (*((__be32 *)(mapped_data + new_offset)) == - cpu_to_be32(JFS_MAGIC_NUMBER)) { + cpu_to_be32(JBD2_MAGIC_NUMBER)) { need_copy_out = 1; do_escape = 1; } @@ -329,10 +329,10 @@ repeat: char *tmp; jbd_unlock_bh_state(bh_in); - tmp = jbd_slab_alloc(bh_in->b_size, GFP_NOFS); + tmp = jbd2_slab_alloc(bh_in->b_size, GFP_NOFS); jbd_lock_bh_state(bh_in); if (jh_in->b_frozen_data) { - jbd_slab_free(tmp, bh_in->b_size); + jbd2_slab_free(tmp, bh_in->b_size); goto repeat; } @@ -362,7 +362,7 @@ repeat: atomic_set(&new_bh->b_count, 1); jbd_unlock_bh_state(bh_in); - new_jh = journal_add_journal_head(new_bh); /* This sleeps */ + new_jh = jbd2_journal_add_journal_head(new_bh); /* This sleeps */ set_bh_page(new_bh, new_page, new_offset); new_jh->b_transaction = NULL; @@ -380,9 +380,9 @@ repeat: * copying is moved to the transaction's shadow queue. */ JBUFFER_TRACE(jh_in, "file as BJ_Shadow"); - journal_file_buffer(jh_in, transaction, BJ_Shadow); + jbd2_journal_file_buffer(jh_in, transaction, BJ_Shadow); JBUFFER_TRACE(new_jh, "file as BJ_IO"); - journal_file_buffer(new_jh, transaction, BJ_IO); + jbd2_journal_file_buffer(new_jh, transaction, BJ_IO); return do_escape | (done_copy_out << 1); } @@ -393,14 +393,14 @@ repeat: */ /* - * __log_space_left: Return the number of free blocks left in the journal. + * __jbd2_log_space_left: Return the number of free blocks left in the journal. * * Called with the journal already locked. * * Called under j_state_lock */ -int __log_space_left(journal_t *journal) +int __jbd2_log_space_left(journal_t *journal) { int left = journal->j_free; @@ -424,7 +424,7 @@ int __log_space_left(journal_t *journal) /* * Called under j_state_lock. Returns true if a transaction was started. */ -int __log_start_commit(journal_t *journal, tid_t target) +int __jbd2_log_start_commit(journal_t *journal, tid_t target) { /* * Are we already doing a recent enough commit? @@ -445,12 +445,12 @@ int __log_start_commit(journal_t *journal, tid_t target) return 0; } -int log_start_commit(journal_t *journal, tid_t tid) +int jbd2_log_start_commit(journal_t *journal, tid_t tid) { int ret; spin_lock(&journal->j_state_lock); - ret = __log_start_commit(journal, tid); + ret = __jbd2_log_start_commit(journal, tid); spin_unlock(&journal->j_state_lock); return ret; } @@ -465,7 +465,7 @@ int log_start_commit(journal_t *journal, tid_t tid) * * Returns true if a transaction was started. */ -int journal_force_commit_nested(journal_t *journal) +int jbd2_journal_force_commit_nested(journal_t *journal) { transaction_t *transaction = NULL; tid_t tid; @@ -473,7 +473,7 @@ int journal_force_commit_nested(journal_t *journal) spin_lock(&journal->j_state_lock); if (journal->j_running_transaction && !current->journal_info) { transaction = journal->j_running_transaction; - __log_start_commit(journal, transaction->t_tid); + __jbd2_log_start_commit(journal, transaction->t_tid); } else if (journal->j_committing_transaction) transaction = journal->j_committing_transaction; @@ -484,7 +484,7 @@ int journal_force_commit_nested(journal_t *journal) tid = transaction->t_tid; spin_unlock(&journal->j_state_lock); - log_wait_commit(journal, tid); + jbd2_log_wait_commit(journal, tid); return 1; } @@ -492,7 +492,7 @@ int journal_force_commit_nested(journal_t *journal) * Start a commit of the current running transaction (if any). Returns true * if a transaction was started, and fills its tid in at *ptid */ -int journal_start_commit(journal_t *journal, tid_t *ptid) +int jbd2_journal_start_commit(journal_t *journal, tid_t *ptid) { int ret = 0; @@ -500,7 +500,7 @@ int journal_start_commit(journal_t *journal, tid_t *ptid) if (journal->j_running_transaction) { tid_t tid = journal->j_running_transaction->t_tid; - ret = __log_start_commit(journal, tid); + ret = __jbd2_log_start_commit(journal, tid); if (ret && ptid) *ptid = tid; } else if (journal->j_committing_transaction && ptid) { @@ -519,7 +519,7 @@ int journal_start_commit(journal_t *journal, tid_t *ptid) * Wait for a specified commit to complete. * The caller may not hold the journal lock. */ -int log_wait_commit(journal_t *journal, tid_t tid) +int jbd2_log_wait_commit(journal_t *journal, tid_t tid) { int err = 0; @@ -555,7 +555,7 @@ int log_wait_commit(journal_t *journal, tid_t tid) * Log buffer allocation routines: */ -int journal_next_log_block(journal_t *journal, unsigned long *retp) +int jbd2_journal_next_log_block(journal_t *journal, unsigned long *retp) { unsigned long blocknr; @@ -568,7 +568,7 @@ int journal_next_log_block(journal_t *journal, unsigned long *retp) if (journal->j_head == journal->j_last) journal->j_head = journal->j_first; spin_unlock(&journal->j_state_lock); - return journal_bmap(journal, blocknr, retp); + return jbd2_journal_bmap(journal, blocknr, retp); } /* @@ -578,7 +578,7 @@ int journal_next_log_block(journal_t *journal, unsigned long *retp) * this is a no-op. If needed, we can use j_blk_offset - everything is * ready. */ -int journal_bmap(journal_t *journal, unsigned long blocknr, +int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, unsigned long *retp) { int err = 0; @@ -610,18 +610,18 @@ int journal_bmap(journal_t *journal, unsigned long blocknr, * the journal without copying their contents, but for journal * descriptor blocks we do need to generate bona fide buffers. * - * After the caller of journal_get_descriptor_buffer() has finished modifying + * After the caller of jbd2_journal_get_descriptor_buffer() has finished modifying * the buffer's contents they really should run flush_dcache_page(bh->b_page). * But we don't bother doing that, so there will be coherency problems with * mmaps of blockdevs which hold live JBD-controlled filesystems. */ -struct journal_head *journal_get_descriptor_buffer(journal_t *journal) +struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) { struct buffer_head *bh; unsigned long blocknr; int err; - err = journal_next_log_block(journal, &blocknr); + err = jbd2_journal_next_log_block(journal, &blocknr); if (err) return NULL; @@ -632,7 +632,7 @@ struct journal_head *journal_get_descriptor_buffer(journal_t *journal) set_buffer_uptodate(bh); unlock_buffer(bh); BUFFER_TRACE(bh, "return this buffer"); - return journal_add_journal_head(bh); + return jbd2_journal_add_journal_head(bh); } /* @@ -669,10 +669,10 @@ static journal_t * journal_init_common (void) journal->j_commit_interval = (HZ * JBD_DEFAULT_MAX_COMMIT_AGE); /* The journal is marked for error until we succeed with recovery! */ - journal->j_flags = JFS_ABORT; + journal->j_flags = JBD2_ABORT; /* Set up a default-sized revoke table for the new mount. */ - err = journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH); + err = jbd2_journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH); if (err) { kfree(journal); goto fail; @@ -682,7 +682,7 @@ fail: return NULL; } -/* journal_init_dev and journal_init_inode: +/* jbd2_journal_init_dev and jbd2_journal_init_inode: * * Create a journal structure assigned some fixed set of disk blocks to * the journal. We don't actually touch those disk blocks yet, but we @@ -692,7 +692,7 @@ fail: */ /** - * journal_t * journal_init_dev() - creates an initialises a journal structure + * journal_t * jbd2_journal_init_dev() - creates an initialises a journal structure * @bdev: Block device on which to create the journal * @fs_dev: Device which hold journalled filesystem for this journal. * @start: Block nr Start of journal. @@ -700,11 +700,11 @@ fail: * @blocksize: blocksize of journalling device * @returns: a newly created journal_t * * - * journal_init_dev creates a journal which maps a fixed contiguous + * jbd2_journal_init_dev creates a journal which maps a fixed contiguous * range of blocks on an arbitrary block device. * */ -journal_t * journal_init_dev(struct block_device *bdev, +journal_t * jbd2_journal_init_dev(struct block_device *bdev, struct block_device *fs_dev, int start, int len, int blocksize) { @@ -740,14 +740,14 @@ journal_t * journal_init_dev(struct block_device *bdev, } /** - * journal_t * journal_init_inode () - creates a journal which maps to a inode. + * journal_t * jbd2_journal_init_inode () - creates a journal which maps to a inode. * @inode: An inode to create the journal in * - * journal_init_inode creates a journal which maps an on-disk inode as + * jbd2_journal_init_inode creates a journal which maps an on-disk inode as * the journal. The inode must exist already, must support bmap() and * must have all data blocks preallocated. */ -journal_t * journal_init_inode (struct inode *inode) +journal_t * jbd2_journal_init_inode (struct inode *inode) { struct buffer_head *bh; journal_t *journal = journal_init_common(); @@ -780,7 +780,7 @@ journal_t * journal_init_inode (struct inode *inode) return NULL; } - err = journal_bmap(journal, 0, &blocknr); + err = jbd2_journal_bmap(journal, 0, &blocknr); /* If that failed, give up */ if (err) { printk(KERN_ERR "%s: Cannnot locate journal superblock\n", @@ -838,27 +838,27 @@ static int journal_reset(journal_t *journal) journal->j_max_transaction_buffers = journal->j_maxlen / 4; /* Add the dynamic fields and write it to disk. */ - journal_update_superblock(journal, 1); - journal_start_thread(journal); + jbd2_journal_update_superblock(journal, 1); + jbd2_journal_start_thread(journal); return 0; } /** - * int journal_create() - Initialise the new journal file + * int jbd2_journal_create() - Initialise the new journal file * @journal: Journal to create. This structure must have been initialised * * Given a journal_t structure which tells us which disk blocks we can * use, create a new journal superblock and initialise all of the * journal fields from scratch. **/ -int journal_create(journal_t *journal) +int jbd2_journal_create(journal_t *journal) { unsigned long blocknr; struct buffer_head *bh; journal_superblock_t *sb; int i, err; - if (journal->j_maxlen < JFS_MIN_JOURNAL_BLOCKS) { + if (journal->j_maxlen < JBD2_MIN_JOURNAL_BLOCKS) { printk (KERN_ERR "Journal length (%d blocks) too short.\n", journal->j_maxlen); journal_fail_superblock(journal); @@ -876,10 +876,10 @@ int journal_create(journal_t *journal) } /* Zero out the entire journal on disk. We cannot afford to - have any blocks on disk beginning with JFS_MAGIC_NUMBER. */ + have any blocks on disk beginning with JBD2_MAGIC_NUMBER. */ jbd_debug(1, "JBD: Zeroing out journal blocks...\n"); for (i = 0; i < journal->j_maxlen; i++) { - err = journal_bmap(journal, i, &blocknr); + err = jbd2_journal_bmap(journal, i, &blocknr); if (err) return err; bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); @@ -899,8 +899,8 @@ int journal_create(journal_t *journal) /* OK, fill in the initial static fields in the new superblock */ sb = journal->j_superblock; - sb->s_header.h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); - sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2); + sb->s_header.h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2); sb->s_blocksize = cpu_to_be32(journal->j_blocksize); sb->s_maxlen = cpu_to_be32(journal->j_maxlen); @@ -908,21 +908,21 @@ int journal_create(journal_t *journal) journal->j_transaction_sequence = 1; - journal->j_flags &= ~JFS_ABORT; + journal->j_flags &= ~JBD2_ABORT; journal->j_format_version = 2; return journal_reset(journal); } /** - * void journal_update_superblock() - Update journal sb on disk. + * void jbd2_journal_update_superblock() - Update journal sb on disk. * @journal: The journal to update. * @wait: Set to '0' if you don't want to wait for IO completion. * * Update a journal's dynamic superblock fields and write it to disk, * optionally waiting for the IO to complete. */ -void journal_update_superblock(journal_t *journal, int wait) +void jbd2_journal_update_superblock(journal_t *journal, int wait) { journal_superblock_t *sb = journal->j_superblock; struct buffer_head *bh = journal->j_sb_buffer; @@ -931,7 +931,7 @@ void journal_update_superblock(journal_t *journal, int wait) * As a special case, if the on-disk copy is already marked as needing * no recovery (s_start == 0) and there are no outstanding transactions * in the filesystem, then we can safely defer the superblock update - * until the next commit by setting JFS_FLUSHED. This avoids + * until the next commit by setting JBD2_FLUSHED. This avoids * attempting a write to a potential-readonly device. */ if (sb->s_start == 0 && journal->j_tail_sequence == @@ -966,9 +966,9 @@ out: spin_lock(&journal->j_state_lock); if (sb->s_start) - journal->j_flags &= ~JFS_FLUSHED; + journal->j_flags &= ~JBD2_FLUSHED; else - journal->j_flags |= JFS_FLUSHED; + journal->j_flags |= JBD2_FLUSHED; spin_unlock(&journal->j_state_lock); } @@ -1000,17 +1000,17 @@ static int journal_get_superblock(journal_t *journal) err = -EINVAL; - if (sb->s_header.h_magic != cpu_to_be32(JFS_MAGIC_NUMBER) || + if (sb->s_header.h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER) || sb->s_blocksize != cpu_to_be32(journal->j_blocksize)) { printk(KERN_WARNING "JBD: no valid journal superblock found\n"); goto out; } switch(be32_to_cpu(sb->s_header.h_blocktype)) { - case JFS_SUPERBLOCK_V1: + case JBD2_SUPERBLOCK_V1: journal->j_format_version = 1; break; - case JFS_SUPERBLOCK_V2: + case JBD2_SUPERBLOCK_V2: journal->j_format_version = 2; break; default: @@ -1059,14 +1059,14 @@ static int load_superblock(journal_t *journal) /** - * int journal_load() - Read journal from disk. + * int jbd2_journal_load() - Read journal from disk. * @journal: Journal to act on. * * Given a journal_t structure which tells us which disk blocks contain * a journal, read the journal from disk to initialise the in-memory * structures. */ -int journal_load(journal_t *journal) +int jbd2_journal_load(journal_t *journal) { int err; journal_superblock_t *sb; @@ -1081,9 +1081,9 @@ int journal_load(journal_t *journal) if (journal->j_format_version >= 2) { if ((sb->s_feature_ro_compat & - ~cpu_to_be32(JFS_KNOWN_ROCOMPAT_FEATURES)) || + ~cpu_to_be32(JBD2_KNOWN_ROCOMPAT_FEATURES)) || (sb->s_feature_incompat & - ~cpu_to_be32(JFS_KNOWN_INCOMPAT_FEATURES))) { + ~cpu_to_be32(JBD2_KNOWN_INCOMPAT_FEATURES))) { printk (KERN_WARNING "JBD: Unrecognised features on journal\n"); return -EINVAL; @@ -1093,13 +1093,13 @@ int journal_load(journal_t *journal) /* * Create a slab for this blocksize */ - err = journal_create_jbd_slab(be32_to_cpu(sb->s_blocksize)); + err = jbd2_journal_create_jbd_slab(be32_to_cpu(sb->s_blocksize)); if (err) return err; /* Let the recovery code check whether it needs to recover any * data from the journal. */ - if (journal_recover(journal)) + if (jbd2_journal_recover(journal)) goto recovery_error; /* OK, we've finished with the dynamic journal bits: @@ -1108,8 +1108,8 @@ int journal_load(journal_t *journal) if (journal_reset(journal)) goto recovery_error; - journal->j_flags &= ~JFS_ABORT; - journal->j_flags |= JFS_LOADED; + journal->j_flags &= ~JBD2_ABORT; + journal->j_flags |= JBD2_LOADED; return 0; recovery_error: @@ -1118,20 +1118,20 @@ recovery_error: } /** - * void journal_destroy() - Release a journal_t structure. + * void jbd2_journal_destroy() - Release a journal_t structure. * @journal: Journal to act on. * * Release a journal_t structure once it is no longer in use by the * journaled object. */ -void journal_destroy(journal_t *journal) +void jbd2_journal_destroy(journal_t *journal) { /* Wait for the commit thread to wake up and die. */ journal_kill_thread(journal); /* Force a final log commit */ if (journal->j_running_transaction) - journal_commit_transaction(journal); + jbd2_journal_commit_transaction(journal); /* Force any old transactions to disk */ @@ -1139,7 +1139,7 @@ void journal_destroy(journal_t *journal) spin_lock(&journal->j_list_lock); while (journal->j_checkpoint_transactions != NULL) { spin_unlock(&journal->j_list_lock); - log_do_checkpoint(journal); + jbd2_log_do_checkpoint(journal); spin_lock(&journal->j_list_lock); } @@ -1152,21 +1152,21 @@ void journal_destroy(journal_t *journal) journal->j_tail = 0; journal->j_tail_sequence = ++journal->j_transaction_sequence; if (journal->j_sb_buffer) { - journal_update_superblock(journal, 1); + jbd2_journal_update_superblock(journal, 1); brelse(journal->j_sb_buffer); } if (journal->j_inode) iput(journal->j_inode); if (journal->j_revoke) - journal_destroy_revoke(journal); + jbd2_journal_destroy_revoke(journal); kfree(journal->j_wbuf); kfree(journal); } /** - *int journal_check_used_features () - Check if features specified are used. + *int jbd2_journal_check_used_features () - Check if features specified are used. * @journal: Journal to check. * @compat: bitmask of compatible features * @ro: bitmask of features that force read-only mount @@ -1176,7 +1176,7 @@ void journal_destroy(journal_t *journal) * features. Return true (non-zero) if it does. **/ -int journal_check_used_features (journal_t *journal, unsigned long compat, +int jbd2_journal_check_used_features (journal_t *journal, unsigned long compat, unsigned long ro, unsigned long incompat) { journal_superblock_t *sb; @@ -1197,7 +1197,7 @@ int journal_check_used_features (journal_t *journal, unsigned long compat, } /** - * int journal_check_available_features() - Check feature set in journalling layer + * int jbd2_journal_check_available_features() - Check feature set in journalling layer * @journal: Journal to check. * @compat: bitmask of compatible features * @ro: bitmask of features that force read-only mount @@ -1207,7 +1207,7 @@ int journal_check_used_features (journal_t *journal, unsigned long compat, * all of a given set of features on this journal. Return true * (non-zero) if it can. */ -int journal_check_available_features (journal_t *journal, unsigned long compat, +int jbd2_journal_check_available_features (journal_t *journal, unsigned long compat, unsigned long ro, unsigned long incompat) { journal_superblock_t *sb; @@ -1224,16 +1224,16 @@ int journal_check_available_features (journal_t *journal, unsigned long compat, if (journal->j_format_version != 2) return 0; - if ((compat & JFS_KNOWN_COMPAT_FEATURES) == compat && - (ro & JFS_KNOWN_ROCOMPAT_FEATURES) == ro && - (incompat & JFS_KNOWN_INCOMPAT_FEATURES) == incompat) + if ((compat & JBD2_KNOWN_COMPAT_FEATURES) == compat && + (ro & JBD2_KNOWN_ROCOMPAT_FEATURES) == ro && + (incompat & JBD2_KNOWN_INCOMPAT_FEATURES) == incompat) return 1; return 0; } /** - * int journal_set_features () - Mark a given journal feature in the superblock + * int jbd2_journal_set_features () - Mark a given journal feature in the superblock * @journal: Journal to act on. * @compat: bitmask of compatible features * @ro: bitmask of features that force read-only mount @@ -1244,15 +1244,15 @@ int journal_check_available_features (journal_t *journal, unsigned long compat, * */ -int journal_set_features (journal_t *journal, unsigned long compat, +int jbd2_journal_set_features (journal_t *journal, unsigned long compat, unsigned long ro, unsigned long incompat) { journal_superblock_t *sb; - if (journal_check_used_features(journal, compat, ro, incompat)) + if (jbd2_journal_check_used_features(journal, compat, ro, incompat)) return 1; - if (!journal_check_available_features(journal, compat, ro, incompat)) + if (!jbd2_journal_check_available_features(journal, compat, ro, incompat)) return 0; jbd_debug(1, "Setting new features 0x%lx/0x%lx/0x%lx\n", @@ -1269,13 +1269,13 @@ int journal_set_features (journal_t *journal, unsigned long compat, /** - * int journal_update_format () - Update on-disk journal structure. + * int jbd2_journal_update_format () - Update on-disk journal structure. * @journal: Journal to act on. * * Given an initialised but unloaded journal struct, poke about in the * on-disk structure to update it to the most recent supported version. */ -int journal_update_format (journal_t *journal) +int jbd2_journal_update_format (journal_t *journal) { journal_superblock_t *sb; int err; @@ -1287,9 +1287,9 @@ int journal_update_format (journal_t *journal) sb = journal->j_superblock; switch (be32_to_cpu(sb->s_header.h_blocktype)) { - case JFS_SUPERBLOCK_V2: + case JBD2_SUPERBLOCK_V2: return 0; - case JFS_SUPERBLOCK_V1: + case JBD2_SUPERBLOCK_V1: return journal_convert_superblock_v1(journal, sb); default: break; @@ -1312,7 +1312,7 @@ static int journal_convert_superblock_v1(journal_t *journal, memset(&sb->s_feature_compat, 0, blocksize-offset); sb->s_nr_users = cpu_to_be32(1); - sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2); + sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2); journal->j_format_version = 2; bh = journal->j_sb_buffer; @@ -1324,7 +1324,7 @@ static int journal_convert_superblock_v1(journal_t *journal, /** - * int journal_flush () - Flush journal + * int jbd2_journal_flush () - Flush journal * @journal: Journal to act on. * * Flush all data for a given journal to disk and empty the journal. @@ -1332,7 +1332,7 @@ static int journal_convert_superblock_v1(journal_t *journal, * recovery does not need to happen on remount. */ -int journal_flush(journal_t *journal) +int jbd2_journal_flush(journal_t *journal) { int err = 0; transaction_t *transaction = NULL; @@ -1343,7 +1343,7 @@ int journal_flush(journal_t *journal) /* Force everything buffered to the log... */ if (journal->j_running_transaction) { transaction = journal->j_running_transaction; - __log_start_commit(journal, transaction->t_tid); + __jbd2_log_start_commit(journal, transaction->t_tid); } else if (journal->j_committing_transaction) transaction = journal->j_committing_transaction; @@ -1352,7 +1352,7 @@ int journal_flush(journal_t *journal) tid_t tid = transaction->t_tid; spin_unlock(&journal->j_state_lock); - log_wait_commit(journal, tid); + jbd2_log_wait_commit(journal, tid); } else { spin_unlock(&journal->j_state_lock); } @@ -1361,11 +1361,11 @@ int journal_flush(journal_t *journal) spin_lock(&journal->j_list_lock); while (!err && journal->j_checkpoint_transactions != NULL) { spin_unlock(&journal->j_list_lock); - err = log_do_checkpoint(journal); + err = jbd2_log_do_checkpoint(journal); spin_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); - cleanup_journal_tail(journal); + jbd2_cleanup_journal_tail(journal); /* Finally, mark the journal as really needing no recovery. * This sets s_start==0 in the underlying superblock, which is @@ -1376,7 +1376,7 @@ int journal_flush(journal_t *journal) old_tail = journal->j_tail; journal->j_tail = 0; spin_unlock(&journal->j_state_lock); - journal_update_superblock(journal, 1); + jbd2_journal_update_superblock(journal, 1); spin_lock(&journal->j_state_lock); journal->j_tail = old_tail; @@ -1390,24 +1390,24 @@ int journal_flush(journal_t *journal) } /** - * int journal_wipe() - Wipe journal contents + * int jbd2_journal_wipe() - Wipe journal contents * @journal: Journal to act on. * @write: flag (see below) * * Wipe out all of the contents of a journal, safely. This will produce * a warning if the journal contains any valid recovery information. - * Must be called between journal_init_*() and journal_load(). + * Must be called between journal_init_*() and jbd2_journal_load(). * * If 'write' is non-zero, then we wipe out the journal on disk; otherwise * we merely suppress recovery. */ -int journal_wipe(journal_t *journal, int write) +int jbd2_journal_wipe(journal_t *journal, int write) { journal_superblock_t *sb; int err = 0; - J_ASSERT (!(journal->j_flags & JFS_LOADED)); + J_ASSERT (!(journal->j_flags & JBD2_LOADED)); err = load_superblock(journal); if (err) @@ -1421,9 +1421,9 @@ int journal_wipe(journal_t *journal, int write) printk (KERN_WARNING "JBD: %s recovery information on journal\n", write ? "Clearing" : "Ignoring"); - err = journal_skip_recovery(journal); + err = jbd2_journal_skip_recovery(journal); if (write) - journal_update_superblock(journal, 1); + jbd2_journal_update_superblock(journal, 1); no_recovery: return err; @@ -1459,22 +1459,22 @@ static const char *journal_dev_name(journal_t *journal, char *buffer) * Aborts hard --- we mark the abort as occurred, but do _nothing_ else, * and don't attempt to make any other journal updates. */ -void __journal_abort_hard(journal_t *journal) +void __jbd2_journal_abort_hard(journal_t *journal) { transaction_t *transaction; char b[BDEVNAME_SIZE]; - if (journal->j_flags & JFS_ABORT) + if (journal->j_flags & JBD2_ABORT) return; printk(KERN_ERR "Aborting journal on device %s.\n", journal_dev_name(journal, b)); spin_lock(&journal->j_state_lock); - journal->j_flags |= JFS_ABORT; + journal->j_flags |= JBD2_ABORT; transaction = journal->j_running_transaction; if (transaction) - __log_start_commit(journal, transaction->t_tid); + __jbd2_log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); } @@ -1482,20 +1482,20 @@ void __journal_abort_hard(journal_t *journal) * but don't do any other IO. */ static void __journal_abort_soft (journal_t *journal, int errno) { - if (journal->j_flags & JFS_ABORT) + if (journal->j_flags & JBD2_ABORT) return; if (!journal->j_errno) journal->j_errno = errno; - __journal_abort_hard(journal); + __jbd2_journal_abort_hard(journal); if (errno) - journal_update_superblock(journal, 1); + jbd2_journal_update_superblock(journal, 1); } /** - * void journal_abort () - Shutdown the journal immediately. + * void jbd2_journal_abort () - Shutdown the journal immediately. * @journal: the journal to shutdown. * @errno: an error number to record in the journal indicating * the reason for the shutdown. @@ -1504,7 +1504,7 @@ static void __journal_abort_soft (journal_t *journal, int errno) * journal (not of a single transaction). This operation cannot be * undone without closing and reopening the journal. * - * The journal_abort function is intended to support higher level error + * The jbd2_journal_abort function is intended to support higher level error * recovery mechanisms such as the ext2/ext3 remount-readonly error * mode. * @@ -1520,13 +1520,13 @@ static void __journal_abort_soft (journal_t *journal, int errno) * * Any attempt to get a new transaction handle on a journal which is in * ABORT state will just result in an -EROFS error return. A - * journal_stop on an existing handle will return -EIO if we have + * jbd2_journal_stop on an existing handle will return -EIO if we have * entered abort state during the update. * * Recursive transactions are not disturbed by journal abort until the - * final journal_stop, which will receive the -EIO error. + * final jbd2_journal_stop, which will receive the -EIO error. * - * Finally, the journal_abort call allows the caller to supply an errno + * Finally, the jbd2_journal_abort call allows the caller to supply an errno * which will be recorded (if possible) in the journal superblock. This * allows a client to record failure conditions in the middle of a * transaction without having to complete the transaction to record the @@ -1540,28 +1540,28 @@ static void __journal_abort_soft (journal_t *journal, int errno) * */ -void journal_abort(journal_t *journal, int errno) +void jbd2_journal_abort(journal_t *journal, int errno) { __journal_abort_soft(journal, errno); } /** - * int journal_errno () - returns the journal's error state. + * int jbd2_journal_errno () - returns the journal's error state. * @journal: journal to examine. * - * This is the errno numbet set with journal_abort(), the last + * This is the errno numbet set with jbd2_journal_abort(), the last * time the journal was mounted - if the journal was stopped * without calling abort this will be 0. * * If the journal has been aborted on this mount time -EROFS will * be returned. */ -int journal_errno(journal_t *journal) +int jbd2_journal_errno(journal_t *journal) { int err; spin_lock(&journal->j_state_lock); - if (journal->j_flags & JFS_ABORT) + if (journal->j_flags & JBD2_ABORT) err = -EROFS; else err = journal->j_errno; @@ -1570,18 +1570,18 @@ int journal_errno(journal_t *journal) } /** - * int journal_clear_err () - clears the journal's error state + * int jbd2_journal_clear_err () - clears the journal's error state * @journal: journal to act on. * * An error must be cleared or Acked to take a FS out of readonly * mode. */ -int journal_clear_err(journal_t *journal) +int jbd2_journal_clear_err(journal_t *journal) { int err = 0; spin_lock(&journal->j_state_lock); - if (journal->j_flags & JFS_ABORT) + if (journal->j_flags & JBD2_ABORT) err = -EROFS; else journal->j_errno = 0; @@ -1590,21 +1590,21 @@ int journal_clear_err(journal_t *journal) } /** - * void journal_ack_err() - Ack journal err. + * void jbd2_journal_ack_err() - Ack journal err. * @journal: journal to act on. * * An error must be cleared or Acked to take a FS out of readonly * mode. */ -void journal_ack_err(journal_t *journal) +void jbd2_journal_ack_err(journal_t *journal) { spin_lock(&journal->j_state_lock); if (journal->j_errno) - journal->j_flags |= JFS_ACK_ERR; + journal->j_flags |= JBD2_ACK_ERR; spin_unlock(&journal->j_state_lock); } -int journal_blocks_per_page(struct inode *inode) +int jbd2_journal_blocks_per_page(struct inode *inode) { return 1 << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); } @@ -1613,7 +1613,7 @@ int journal_blocks_per_page(struct inode *inode) * Simple support for retrying memory allocations. Introduced to help to * debug different VM deadlock avoidance strategies. */ -void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) +void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry) { return kmalloc(size, flags | (retry ? __GFP_NOFAIL : 0)); } @@ -1634,7 +1634,7 @@ static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" }; -static void journal_destroy_jbd_slabs(void) +static void jbd2_journal_destroy_jbd_slabs(void) { int i; @@ -1645,7 +1645,7 @@ static void journal_destroy_jbd_slabs(void) } } -static int journal_create_jbd_slab(size_t slab_size) +static int jbd2_journal_create_jbd_slab(size_t slab_size) { int i = JBD_SLAB_INDEX(slab_size); @@ -1671,7 +1671,7 @@ static int journal_create_jbd_slab(size_t slab_size) return 0; } -void * jbd_slab_alloc(size_t size, gfp_t flags) +void * jbd2_slab_alloc(size_t size, gfp_t flags) { int idx; @@ -1680,7 +1680,7 @@ void * jbd_slab_alloc(size_t size, gfp_t flags) return kmem_cache_alloc(jbd_slab[idx], flags | __GFP_NOFAIL); } -void jbd_slab_free(void *ptr, size_t size) +void jbd2_slab_free(void *ptr, size_t size) { int idx; @@ -1692,35 +1692,35 @@ void jbd_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *journal_head_cache; +static kmem_cache_t *jbd2_journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif -static int journal_init_journal_head_cache(void) +static int journal_init_jbd2_journal_head_cache(void) { int retval; - J_ASSERT(journal_head_cache == 0); - journal_head_cache = kmem_cache_create("journal_head", + J_ASSERT(jbd2_journal_head_cache == 0); + jbd2_journal_head_cache = kmem_cache_create("journal_head", sizeof(struct journal_head), 0, /* offset */ 0, /* flags */ NULL, /* ctor */ NULL); /* dtor */ retval = 0; - if (journal_head_cache == 0) { + if (jbd2_journal_head_cache == 0) { retval = -ENOMEM; printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); } return retval; } -static void journal_destroy_journal_head_cache(void) +static void jbd2_journal_destroy_jbd2_journal_head_cache(void) { - J_ASSERT(journal_head_cache != NULL); - kmem_cache_destroy(journal_head_cache); - journal_head_cache = NULL; + J_ASSERT(jbd2_journal_head_cache != NULL); + kmem_cache_destroy(jbd2_journal_head_cache); + jbd2_journal_head_cache = NULL; } /* @@ -1734,7 +1734,7 @@ static struct journal_head *journal_alloc_journal_head(void) #ifdef CONFIG_JBD_DEBUG atomic_inc(&nr_journal_heads); #endif - ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS); + ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); if (ret == 0) { jbd_debug(1, "out of memory for journal_head\n"); if (time_after(jiffies, last_warning + 5*HZ)) { @@ -1744,7 +1744,7 @@ static struct journal_head *journal_alloc_journal_head(void) } while (ret == 0) { yield(); - ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS); + ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); } } return ret; @@ -1756,7 +1756,7 @@ static void journal_free_journal_head(struct journal_head *jh) atomic_dec(&nr_journal_heads); memset(jh, JBD_POISON_FREE, sizeof(*jh)); #endif - kmem_cache_free(journal_head_cache, jh); + kmem_cache_free(jbd2_journal_head_cache, jh); } /* @@ -1775,22 +1775,22 @@ static void journal_free_journal_head(struct journal_head *jh) * * A journal_head may be detached from its buffer_head when the journal_head's * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL. - * Various places in JBD call journal_remove_journal_head() to indicate that the + * Various places in JBD call jbd2_journal_remove_journal_head() to indicate that the * journal_head can be dropped if needed. * * Various places in the kernel want to attach a journal_head to a buffer_head * _before_ attaching the journal_head to a transaction. To protect the - * journal_head in this situation, journal_add_journal_head elevates the + * journal_head in this situation, jbd2_journal_add_journal_head elevates the * journal_head's b_jcount refcount by one. The caller must call - * journal_put_journal_head() to undo this. + * jbd2_journal_put_journal_head() to undo this. * * So the typical usage would be: * * (Attach a journal_head if needed. Increments b_jcount) - * struct journal_head *jh = journal_add_journal_head(bh); + * struct journal_head *jh = jbd2_journal_add_journal_head(bh); * ... * jh->b_transaction = xxx; - * journal_put_journal_head(jh); + * jbd2_journal_put_journal_head(jh); * * Now, the journal_head's b_jcount is zero, but it is safe from being released * because it has a non-zero b_transaction. @@ -1802,7 +1802,7 @@ static void journal_free_journal_head(struct journal_head *jh) * Doesn't need the journal lock. * May sleep. */ -struct journal_head *journal_add_journal_head(struct buffer_head *bh) +struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh) { struct journal_head *jh; struct journal_head *new_jh = NULL; @@ -1845,7 +1845,7 @@ repeat: * Grab a ref against this buffer_head's journal_head. If it ended up not * having a journal_head, return NULL */ -struct journal_head *journal_grab_journal_head(struct buffer_head *bh) +struct journal_head *jbd2_journal_grab_journal_head(struct buffer_head *bh) { struct journal_head *jh = NULL; @@ -1877,13 +1877,13 @@ static void __journal_remove_journal_head(struct buffer_head *bh) printk(KERN_WARNING "%s: freeing " "b_frozen_data\n", __FUNCTION__); - jbd_slab_free(jh->b_frozen_data, bh->b_size); + jbd2_slab_free(jh->b_frozen_data, bh->b_size); } if (jh->b_committed_data) { printk(KERN_WARNING "%s: freeing " "b_committed_data\n", __FUNCTION__); - jbd_slab_free(jh->b_committed_data, bh->b_size); + jbd2_slab_free(jh->b_committed_data, bh->b_size); } bh->b_private = NULL; jh->b_bh = NULL; /* debug, really */ @@ -1897,7 +1897,7 @@ static void __journal_remove_journal_head(struct buffer_head *bh) } /* - * journal_remove_journal_head(): if the buffer isn't attached to a transaction + * jbd2_journal_remove_journal_head(): if the buffer isn't attached to a transaction * and has a zero b_jcount then remove and release its journal_head. If we did * see that the buffer is not used by any transaction we also "logically" * decrement ->b_count. @@ -1905,11 +1905,11 @@ static void __journal_remove_journal_head(struct buffer_head *bh) * We in fact take an additional increment on ->b_count as a convenience, * because the caller usually wants to do additional things with the bh * after calling here. - * The caller of journal_remove_journal_head() *must* run __brelse(bh) at some + * The caller of jbd2_journal_remove_journal_head() *must* run __brelse(bh) at some * time. Once the caller has run __brelse(), the buffer is eligible for * reaping by try_to_free_buffers(). */ -void journal_remove_journal_head(struct buffer_head *bh) +void jbd2_journal_remove_journal_head(struct buffer_head *bh) { jbd_lock_bh_journal_head(bh); __journal_remove_journal_head(bh); @@ -1920,7 +1920,7 @@ void journal_remove_journal_head(struct buffer_head *bh) * Drop a reference on the passed journal_head. If it fell to zero then try to * release the journal_head from the buffer_head. */ -void journal_put_journal_head(struct journal_head *jh) +void jbd2_journal_put_journal_head(struct journal_head *jh) { struct buffer_head *bh = jh2bh(jh); @@ -1938,8 +1938,8 @@ void journal_put_journal_head(struct journal_head *jh) * /proc tunables */ #if defined(CONFIG_JBD_DEBUG) -int journal_enable_debug; -EXPORT_SYMBOL(journal_enable_debug); +int jbd2_journal_enable_debug; +EXPORT_SYMBOL(jbd2_journal_enable_debug); #endif #if defined(CONFIG_JBD_DEBUG) && defined(CONFIG_PROC_FS) @@ -1951,7 +1951,7 @@ static int read_jbd_debug(char *page, char **start, off_t off, { int ret; - ret = sprintf(page + off, "%d\n", journal_enable_debug); + ret = sprintf(page + off, "%d\n", jbd2_journal_enable_debug); *eof = 1; return ret; } @@ -1966,11 +1966,11 @@ static int write_jbd_debug(struct file *file, const char __user *buffer, if (copy_from_user(buf, buffer, count)) return -EFAULT; buf[ARRAY_SIZE(buf) - 1] = '\0'; - journal_enable_debug = simple_strtoul(buf, NULL, 10); + jbd2_journal_enable_debug = simple_strtoul(buf, NULL, 10); return count; } -#define JBD_PROC_NAME "sys/fs/jbd-debug" +#define JBD_PROC_NAME "sys/fs/jbd2-debug" static void __init create_jbd_proc_entry(void) { @@ -1982,7 +1982,7 @@ static void __init create_jbd_proc_entry(void) } } -static void __exit remove_jbd_proc_entry(void) +static void __exit jbd2_remove_jbd_proc_entry(void) { if (proc_jbd_debug) remove_proc_entry(JBD_PROC_NAME, NULL); @@ -1991,31 +1991,31 @@ static void __exit remove_jbd_proc_entry(void) #else #define create_jbd_proc_entry() do {} while (0) -#define remove_jbd_proc_entry() do {} while (0) +#define jbd2_remove_jbd_proc_entry() do {} while (0) #endif -kmem_cache_t *jbd_handle_cache; +kmem_cache_t *jbd2_handle_cache; static int __init journal_init_handle_cache(void) { - jbd_handle_cache = kmem_cache_create("journal_handle", + jbd2_handle_cache = kmem_cache_create("journal_handle", sizeof(handle_t), 0, /* offset */ 0, /* flags */ NULL, /* ctor */ NULL); /* dtor */ - if (jbd_handle_cache == NULL) { + if (jbd2_handle_cache == NULL) { printk(KERN_EMERG "JBD: failed to create handle cache\n"); return -ENOMEM; } return 0; } -static void journal_destroy_handle_cache(void) +static void jbd2_journal_destroy_handle_cache(void) { - if (jbd_handle_cache) - kmem_cache_destroy(jbd_handle_cache); + if (jbd2_handle_cache) + kmem_cache_destroy(jbd2_handle_cache); } /* @@ -2026,20 +2026,20 @@ static int __init journal_init_caches(void) { int ret; - ret = journal_init_revoke_caches(); + ret = jbd2_journal_init_revoke_caches(); if (ret == 0) - ret = journal_init_journal_head_cache(); + ret = journal_init_jbd2_journal_head_cache(); if (ret == 0) ret = journal_init_handle_cache(); return ret; } -static void journal_destroy_caches(void) +static void jbd2_journal_destroy_caches(void) { - journal_destroy_revoke_caches(); - journal_destroy_journal_head_cache(); - journal_destroy_handle_cache(); - journal_destroy_jbd_slabs(); + jbd2_journal_destroy_revoke_caches(); + jbd2_journal_destroy_jbd2_journal_head_cache(); + jbd2_journal_destroy_handle_cache(); + jbd2_journal_destroy_jbd_slabs(); } static int __init journal_init(void) @@ -2050,7 +2050,7 @@ static int __init journal_init(void) ret = journal_init_caches(); if (ret != 0) - journal_destroy_caches(); + jbd2_journal_destroy_caches(); create_jbd_proc_entry(); return ret; } @@ -2062,8 +2062,8 @@ static void __exit journal_exit(void) if (n) printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n); #endif - remove_jbd_proc_entry(); - journal_destroy_caches(); + jbd2_remove_jbd_proc_entry(); + jbd2_journal_destroy_caches(); } MODULE_LICENSE("GPL"); diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 11563fe..b2012d1 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -18,7 +18,7 @@ #else #include #include -#include +#include #include #include #endif @@ -86,7 +86,7 @@ static int do_readahead(journal_t *journal, unsigned int start) nbufs = 0; for (next = start; next < max; next++) { - err = journal_bmap(journal, next, &blocknr); + err = jbd2_journal_bmap(journal, next, &blocknr); if (err) { printk (KERN_ERR "JBD: bad block at offset %u\n", @@ -142,7 +142,7 @@ static int jread(struct buffer_head **bhp, journal_t *journal, return -EIO; } - err = journal_bmap(journal, offset, &blocknr); + err = jbd2_journal_bmap(journal, offset, &blocknr); if (err) { printk (KERN_ERR "JBD: bad block at offset %u\n", @@ -191,10 +191,10 @@ static int count_tags(struct buffer_head *bh, int size) nr++; tagp += sizeof(journal_block_tag_t); - if (!(tag->t_flags & cpu_to_be32(JFS_FLAG_SAME_UUID))) + if (!(tag->t_flags & cpu_to_be32(JBD2_FLAG_SAME_UUID))) tagp += 16; - if (tag->t_flags & cpu_to_be32(JFS_FLAG_LAST_TAG)) + if (tag->t_flags & cpu_to_be32(JBD2_FLAG_LAST_TAG)) break; } @@ -210,7 +210,7 @@ do { \ } while (0) /** - * journal_recover - recovers a on-disk journal + * jbd2_journal_recover - recovers a on-disk journal * @journal: the journal to recover * * The primary function for recovering the log contents when mounting a @@ -221,7 +221,7 @@ do { \ * blocks. In the third and final pass, we replay any un-revoked blocks * in the log. */ -int journal_recover(journal_t *journal) +int jbd2_journal_recover(journal_t *journal) { int err; journal_superblock_t * sb; @@ -260,13 +260,13 @@ int journal_recover(journal_t *journal) * any existing commit records in the log. */ journal->j_transaction_sequence = ++info.end_transaction; - journal_clear_revoke(journal); + jbd2_journal_clear_revoke(journal); sync_blockdev(journal->j_fs_dev); return err; } /** - * journal_skip_recovery - Start journal and wipe exiting records + * jbd2_journal_skip_recovery - Start journal and wipe exiting records * @journal: journal to startup * * Locate any valid recovery information from the journal and set up the @@ -278,7 +278,7 @@ int journal_recover(journal_t *journal) * much recovery information is being erased, and to let us initialise * the journal transaction sequence numbers to the next unused ID. */ -int journal_skip_recovery(journal_t *journal) +int jbd2_journal_skip_recovery(journal_t *journal) { int err; journal_superblock_t * sb; @@ -387,7 +387,7 @@ static int do_one_pass(journal_t *journal, tmp = (journal_header_t *)bh->b_data; - if (tmp->h_magic != cpu_to_be32(JFS_MAGIC_NUMBER)) { + if (tmp->h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER)) { brelse(bh); break; } @@ -407,7 +407,7 @@ static int do_one_pass(journal_t *journal, * to do with it? That depends on the pass... */ switch(blocktype) { - case JFS_DESCRIPTOR_BLOCK: + case JBD2_DESCRIPTOR_BLOCK: /* If it is a valid descriptor block, replay it * in pass REPLAY; otherwise, just skip over the * blocks it describes. */ @@ -451,7 +451,7 @@ static int do_one_pass(journal_t *journal, /* If the block has been * revoked, then we're all done * here. */ - if (journal_test_revoke + if (jbd2_journal_test_revoke (journal, blocknr, next_commit_ID)) { brelse(obh); @@ -477,9 +477,9 @@ static int do_one_pass(journal_t *journal, lock_buffer(nbh); memcpy(nbh->b_data, obh->b_data, journal->j_blocksize); - if (flags & JFS_FLAG_ESCAPE) { + if (flags & JBD2_FLAG_ESCAPE) { *((__be32 *)bh->b_data) = - cpu_to_be32(JFS_MAGIC_NUMBER); + cpu_to_be32(JBD2_MAGIC_NUMBER); } BUFFER_TRACE(nbh, "marking dirty"); @@ -495,17 +495,17 @@ static int do_one_pass(journal_t *journal, skip_write: tagp += sizeof(journal_block_tag_t); - if (!(flags & JFS_FLAG_SAME_UUID)) + if (!(flags & JBD2_FLAG_SAME_UUID)) tagp += 16; - if (flags & JFS_FLAG_LAST_TAG) + if (flags & JBD2_FLAG_LAST_TAG) break; } brelse(bh); continue; - case JFS_COMMIT_BLOCK: + case JBD2_COMMIT_BLOCK: /* Found an expected commit block: not much to * do other than move on to the next sequence * number. */ @@ -513,7 +513,7 @@ static int do_one_pass(journal_t *journal, next_commit_ID++; continue; - case JFS_REVOKE_BLOCK: + case JBD2_REVOKE_BLOCK: /* If we aren't in the REVOKE pass, then we can * just skip over this block. */ if (pass != PASS_REVOKE) { @@ -570,11 +570,11 @@ static int do_one_pass(journal_t *journal, static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, tid_t sequence, struct recovery_info *info) { - journal_revoke_header_t *header; + jbd2_journal_revoke_header_t *header; int offset, max; - header = (journal_revoke_header_t *) bh->b_data; - offset = sizeof(journal_revoke_header_t); + header = (jbd2_journal_revoke_header_t *) bh->b_data; + offset = sizeof(jbd2_journal_revoke_header_t); max = be32_to_cpu(header->r_count); while (offset < max) { @@ -583,7 +583,7 @@ static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset))); offset += 4; - err = journal_set_revoke(journal, blocknr, sequence); + err = jbd2_journal_set_revoke(journal, blocknr, sequence); if (err) return err; ++info->nr_revokes; diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index c532429..2fccddc 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -62,7 +62,7 @@ #else #include #include -#include +#include #include #include #include @@ -70,14 +70,14 @@ #include #endif -static kmem_cache_t *revoke_record_cache; -static kmem_cache_t *revoke_table_cache; +static kmem_cache_t *jbd2_revoke_record_cache; +static kmem_cache_t *jbd2_revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the last transaction to revoke this block. */ -struct jbd_revoke_record_s +struct jbd2_revoke_record_s { struct list_head hash; tid_t sequence; /* Used for recovery only */ @@ -86,7 +86,7 @@ struct jbd_revoke_record_s /* The revoke table is just a simple hash table of revoke records. */ -struct jbd_revoke_table_s +struct jbd2_revoke_table_s { /* It is conceivable that we might want a larger hash table * for recovery. Must be a power of two. */ @@ -99,7 +99,7 @@ struct jbd_revoke_table_s #ifdef __KERNEL__ static void write_one_revoke_record(journal_t *, transaction_t *, struct journal_head **, int *, - struct jbd_revoke_record_s *); + struct jbd2_revoke_record_s *); static void flush_descriptor(journal_t *, struct journal_head *, int); #endif @@ -108,7 +108,7 @@ static void flush_descriptor(journal_t *, struct journal_head *, int); /* Borrowed from buffer.c: this is a tried and tested block hash function */ static inline int hash(journal_t *journal, unsigned long block) { - struct jbd_revoke_table_s *table = journal->j_revoke; + struct jbd2_revoke_table_s *table = journal->j_revoke; int hash_shift = table->hash_shift; return ((block << (hash_shift - 6)) ^ @@ -120,10 +120,10 @@ static int insert_revoke_hash(journal_t *journal, unsigned long blocknr, tid_t seq) { struct list_head *hash_list; - struct jbd_revoke_record_s *record; + struct jbd2_revoke_record_s *record; repeat: - record = kmem_cache_alloc(revoke_record_cache, GFP_NOFS); + record = kmem_cache_alloc(jbd2_revoke_record_cache, GFP_NOFS); if (!record) goto oom; @@ -145,57 +145,57 @@ oom: /* Find a revoke record in the journal's hash table. */ -static struct jbd_revoke_record_s *find_revoke_record(journal_t *journal, +static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, unsigned long blocknr) { struct list_head *hash_list; - struct jbd_revoke_record_s *record; + struct jbd2_revoke_record_s *record; hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)]; spin_lock(&journal->j_revoke_lock); - record = (struct jbd_revoke_record_s *) hash_list->next; + record = (struct jbd2_revoke_record_s *) hash_list->next; while (&(record->hash) != hash_list) { if (record->blocknr == blocknr) { spin_unlock(&journal->j_revoke_lock); return record; } - record = (struct jbd_revoke_record_s *) record->hash.next; + record = (struct jbd2_revoke_record_s *) record->hash.next; } spin_unlock(&journal->j_revoke_lock); return NULL; } -int __init journal_init_revoke_caches(void) +int __init jbd2_journal_init_revoke_caches(void) { - revoke_record_cache = kmem_cache_create("revoke_record", - sizeof(struct jbd_revoke_record_s), + jbd2_revoke_record_cache = kmem_cache_create("revoke_record", + sizeof(struct jbd2_revoke_record_s), 0, SLAB_HWCACHE_ALIGN, NULL, NULL); - if (revoke_record_cache == 0) + if (jbd2_revoke_record_cache == 0) return -ENOMEM; - revoke_table_cache = kmem_cache_create("revoke_table", - sizeof(struct jbd_revoke_table_s), + jbd2_revoke_table_cache = kmem_cache_create("revoke_table", + sizeof(struct jbd2_revoke_table_s), 0, 0, NULL, NULL); - if (revoke_table_cache == 0) { - kmem_cache_destroy(revoke_record_cache); - revoke_record_cache = NULL; + if (jbd2_revoke_table_cache == 0) { + kmem_cache_destroy(jbd2_revoke_record_cache); + jbd2_revoke_record_cache = NULL; return -ENOMEM; } return 0; } -void journal_destroy_revoke_caches(void) +void jbd2_journal_destroy_revoke_caches(void) { - kmem_cache_destroy(revoke_record_cache); - revoke_record_cache = NULL; - kmem_cache_destroy(revoke_table_cache); - revoke_table_cache = NULL; + kmem_cache_destroy(jbd2_revoke_record_cache); + jbd2_revoke_record_cache = NULL; + kmem_cache_destroy(jbd2_revoke_table_cache); + jbd2_revoke_table_cache = NULL; } /* Initialise the revoke table for a given journal to a given size. */ -int journal_init_revoke(journal_t *journal, int hash_size) +int jbd2_journal_init_revoke(journal_t *journal, int hash_size) { int shift, tmp; @@ -206,7 +206,7 @@ int journal_init_revoke(journal_t *journal, int hash_size) while((tmp >>= 1UL) != 0UL) shift++; - journal->j_revoke_table[0] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL); + journal->j_revoke_table[0] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); if (!journal->j_revoke_table[0]) return -ENOMEM; journal->j_revoke = journal->j_revoke_table[0]; @@ -221,7 +221,7 @@ int journal_init_revoke(journal_t *journal, int hash_size) journal->j_revoke->hash_table = kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); if (!journal->j_revoke->hash_table) { - kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); journal->j_revoke = NULL; return -ENOMEM; } @@ -229,10 +229,10 @@ int journal_init_revoke(journal_t *journal, int hash_size) for (tmp = 0; tmp < hash_size; tmp++) INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); - journal->j_revoke_table[1] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL); + journal->j_revoke_table[1] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); if (!journal->j_revoke_table[1]) { kfree(journal->j_revoke_table[0]->hash_table); - kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); return -ENOMEM; } @@ -249,8 +249,8 @@ int journal_init_revoke(journal_t *journal, int hash_size) kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); if (!journal->j_revoke->hash_table) { kfree(journal->j_revoke_table[0]->hash_table); - kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]); - kmem_cache_free(revoke_table_cache, journal->j_revoke_table[1]); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[1]); journal->j_revoke = NULL; return -ENOMEM; } @@ -265,9 +265,9 @@ int journal_init_revoke(journal_t *journal, int hash_size) /* Destoy a journal's revoke table. The table must already be empty! */ -void journal_destroy_revoke(journal_t *journal) +void jbd2_journal_destroy_revoke(journal_t *journal) { - struct jbd_revoke_table_s *table; + struct jbd2_revoke_table_s *table; struct list_head *hash_list; int i; @@ -281,7 +281,7 @@ void journal_destroy_revoke(journal_t *journal) } kfree(table->hash_table); - kmem_cache_free(revoke_table_cache, table); + kmem_cache_free(jbd2_revoke_table_cache, table); journal->j_revoke = NULL; table = journal->j_revoke_table[1]; @@ -294,7 +294,7 @@ void journal_destroy_revoke(journal_t *journal) } kfree(table->hash_table); - kmem_cache_free(revoke_table_cache, table); + kmem_cache_free(jbd2_revoke_table_cache, table); journal->j_revoke = NULL; } @@ -302,7 +302,7 @@ void journal_destroy_revoke(journal_t *journal) #ifdef __KERNEL__ /* - * journal_revoke: revoke a given buffer_head from the journal. This + * jbd2_journal_revoke: revoke a given buffer_head from the journal. This * prevents the block from being replayed during recovery if we take a * crash after this current transaction commits. Any subsequent * metadata writes of the buffer in this transaction cancel the @@ -314,18 +314,18 @@ void journal_destroy_revoke(journal_t *journal) * revoke before clearing the block bitmap when we are deleting * metadata. * - * Revoke performs a journal_forget on any buffer_head passed in as a + * Revoke performs a jbd2_journal_forget on any buffer_head passed in as a * parameter, but does _not_ forget the buffer_head if the bh was only * found implicitly. * * bh_in may not be a journalled buffer - it may have come off * the hash tables without an attached journal_head. * - * If bh_in is non-zero, journal_revoke() will decrement its b_count + * If bh_in is non-zero, jbd2_journal_revoke() will decrement its b_count * by one. */ -int journal_revoke(handle_t *handle, unsigned long blocknr, +int jbd2_journal_revoke(handle_t *handle, unsigned long blocknr, struct buffer_head *bh_in) { struct buffer_head *bh = NULL; @@ -338,7 +338,7 @@ int journal_revoke(handle_t *handle, unsigned long blocknr, BUFFER_TRACE(bh_in, "enter"); journal = handle->h_transaction->t_journal; - if (!journal_set_features(journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)){ + if (!jbd2_journal_set_features(journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)){ J_ASSERT (!"Cannot set revoke feature!"); return -EINVAL; } @@ -386,8 +386,8 @@ int journal_revoke(handle_t *handle, unsigned long blocknr, set_buffer_revoked(bh); set_buffer_revokevalid(bh); if (bh_in) { - BUFFER_TRACE(bh_in, "call journal_forget"); - journal_forget(handle, bh_in); + BUFFER_TRACE(bh_in, "call jbd2_journal_forget"); + jbd2_journal_forget(handle, bh_in); } else { BUFFER_TRACE(bh, "call brelse"); __brelse(bh); @@ -403,7 +403,7 @@ int journal_revoke(handle_t *handle, unsigned long blocknr, /* * Cancel an outstanding revoke. For use only internally by the - * journaling code (called from journal_get_write_access). + * journaling code (called from jbd2_journal_get_write_access). * * We trust buffer_revoked() on the buffer if the buffer is already * being journaled: if there is no revoke pending on the buffer, then we @@ -418,9 +418,9 @@ int journal_revoke(handle_t *handle, unsigned long blocknr, * * The caller must have the journal locked. */ -int journal_cancel_revoke(handle_t *handle, struct journal_head *jh) +int jbd2_journal_cancel_revoke(handle_t *handle, struct journal_head *jh) { - struct jbd_revoke_record_s *record; + struct jbd2_revoke_record_s *record; journal_t *journal = handle->h_transaction->t_journal; int need_cancel; int did_revoke = 0; /* akpm: debug */ @@ -447,7 +447,7 @@ int journal_cancel_revoke(handle_t *handle, struct journal_head *jh) spin_lock(&journal->j_revoke_lock); list_del(&record->hash); spin_unlock(&journal->j_revoke_lock); - kmem_cache_free(revoke_record_cache, record); + kmem_cache_free(jbd2_revoke_record_cache, record); did_revoke = 1; } } @@ -478,7 +478,7 @@ int journal_cancel_revoke(handle_t *handle, struct journal_head *jh) * we do not want to suspend any processing until all revokes are * written -bzzz */ -void journal_switch_revoke_table(journal_t *journal) +void jbd2_journal_switch_revoke_table(journal_t *journal) { int i; @@ -498,12 +498,12 @@ void journal_switch_revoke_table(journal_t *journal) * Called with the journal lock held. */ -void journal_write_revoke_records(journal_t *journal, +void jbd2_journal_write_revoke_records(journal_t *journal, transaction_t *transaction) { struct journal_head *descriptor; - struct jbd_revoke_record_s *record; - struct jbd_revoke_table_s *revoke; + struct jbd2_revoke_record_s *record; + struct jbd2_revoke_table_s *revoke; struct list_head *hash_list; int i, offset, count; @@ -519,14 +519,14 @@ void journal_write_revoke_records(journal_t *journal, hash_list = &revoke->hash_table[i]; while (!list_empty(hash_list)) { - record = (struct jbd_revoke_record_s *) + record = (struct jbd2_revoke_record_s *) hash_list->next; write_one_revoke_record(journal, transaction, &descriptor, &offset, record); count++; list_del(&record->hash); - kmem_cache_free(revoke_record_cache, record); + kmem_cache_free(jbd2_revoke_record_cache, record); } } if (descriptor) @@ -543,7 +543,7 @@ static void write_one_revoke_record(journal_t *journal, transaction_t *transaction, struct journal_head **descriptorp, int *offsetp, - struct jbd_revoke_record_s *record) + struct jbd2_revoke_record_s *record) { struct journal_head *descriptor; int offset; @@ -551,7 +551,7 @@ static void write_one_revoke_record(journal_t *journal, /* If we are already aborting, this all becomes a noop. We still need to go round the loop in - journal_write_revoke_records in order to free all of the + jbd2_journal_write_revoke_records in order to free all of the revoke records: only the IO to the journal is omitted. */ if (is_journal_aborted(journal)) return; @@ -568,19 +568,19 @@ static void write_one_revoke_record(journal_t *journal, } if (!descriptor) { - descriptor = journal_get_descriptor_buffer(journal); + descriptor = jbd2_journal_get_descriptor_buffer(journal); if (!descriptor) return; header = (journal_header_t *) &jh2bh(descriptor)->b_data[0]; - header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER); - header->h_blocktype = cpu_to_be32(JFS_REVOKE_BLOCK); + header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JBD2_REVOKE_BLOCK); header->h_sequence = cpu_to_be32(transaction->t_tid); /* Record it so that we can wait for IO completion later */ JBUFFER_TRACE(descriptor, "file as BJ_LogCtl"); - journal_file_buffer(descriptor, transaction, BJ_LogCtl); + jbd2_journal_file_buffer(descriptor, transaction, BJ_LogCtl); - offset = sizeof(journal_revoke_header_t); + offset = sizeof(jbd2_journal_revoke_header_t); *descriptorp = descriptor; } @@ -601,7 +601,7 @@ static void flush_descriptor(journal_t *journal, struct journal_head *descriptor, int offset) { - journal_revoke_header_t *header; + jbd2_journal_revoke_header_t *header; struct buffer_head *bh = jh2bh(descriptor); if (is_journal_aborted(journal)) { @@ -609,7 +609,7 @@ static void flush_descriptor(journal_t *journal, return; } - header = (journal_revoke_header_t *) jh2bh(descriptor)->b_data; + header = (jbd2_journal_revoke_header_t *) jh2bh(descriptor)->b_data; header->r_count = cpu_to_be32(offset); set_buffer_jwrite(bh); BUFFER_TRACE(bh, "write"); @@ -640,11 +640,11 @@ static void flush_descriptor(journal_t *journal, * single block. */ -int journal_set_revoke(journal_t *journal, +int jbd2_journal_set_revoke(journal_t *journal, unsigned long blocknr, tid_t sequence) { - struct jbd_revoke_record_s *record; + struct jbd2_revoke_record_s *record; record = find_revoke_record(journal, blocknr); if (record) { @@ -664,11 +664,11 @@ int journal_set_revoke(journal_t *journal, * ones, but later transactions still need replayed. */ -int journal_test_revoke(journal_t *journal, +int jbd2_journal_test_revoke(journal_t *journal, unsigned long blocknr, tid_t sequence) { - struct jbd_revoke_record_s *record; + struct jbd2_revoke_record_s *record; record = find_revoke_record(journal, blocknr); if (!record) @@ -683,21 +683,21 @@ int journal_test_revoke(journal_t *journal, * that it can be reused by the running filesystem. */ -void journal_clear_revoke(journal_t *journal) +void jbd2_journal_clear_revoke(journal_t *journal) { int i; struct list_head *hash_list; - struct jbd_revoke_record_s *record; - struct jbd_revoke_table_s *revoke; + struct jbd2_revoke_record_s *record; + struct jbd2_revoke_table_s *revoke; revoke = journal->j_revoke; for (i = 0; i < revoke->hash_size; i++) { hash_list = &revoke->hash_table[i]; while (!list_empty(hash_list)) { - record = (struct jbd_revoke_record_s*) hash_list->next; + record = (struct jbd2_revoke_record_s*) hash_list->next; list_del(&record->hash); - kmem_cache_free(revoke_record_cache, record); + kmem_cache_free(jbd2_revoke_record_cache, record); } } } diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index e1b3c8a..149957b 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -19,7 +19,7 @@ #include #include -#include +#include #include #include #include @@ -28,7 +28,7 @@ #include /* - * get_transaction: obtain a new transaction_t object. + * jbd2_get_transaction: obtain a new transaction_t object. * * Simply allocate and initialise a new transaction. Create it in * RUNNING state and add it to the current journal (which should not @@ -44,7 +44,7 @@ */ static transaction_t * -get_transaction(journal_t *journal, transaction_t *transaction) +jbd2_get_transaction(journal_t *journal, transaction_t *transaction) { transaction->t_journal = journal; transaction->t_state = T_RUNNING; @@ -115,7 +115,7 @@ repeat: spin_lock(&journal->j_state_lock); repeat_locked: if (is_journal_aborted(journal) || - (journal->j_errno != 0 && !(journal->j_flags & JFS_ACK_ERR))) { + (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) { spin_unlock(&journal->j_state_lock); ret = -EROFS; goto out; @@ -134,7 +134,7 @@ repeat_locked: spin_unlock(&journal->j_state_lock); goto alloc_transaction; } - get_transaction(journal, new_transaction); + jbd2_get_transaction(journal, new_transaction); new_transaction = NULL; } @@ -175,7 +175,7 @@ repeat_locked: spin_unlock(&transaction->t_handle_lock); prepare_to_wait(&journal->j_wait_transaction_locked, &wait, TASK_UNINTERRUPTIBLE); - __log_start_commit(journal, transaction->t_tid); + __jbd2_log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); schedule(); finish_wait(&journal->j_wait_transaction_locked, &wait); @@ -205,12 +205,12 @@ repeat_locked: * committing_transaction->t_outstanding_credits plus "enough" for * the log control blocks. * Also, this test is inconsitent with the matching one in - * journal_extend(). + * jbd2_journal_extend(). */ - if (__log_space_left(journal) < jbd_space_needed(journal)) { + if (__jbd2_log_space_left(journal) < jbd_space_needed(journal)) { jbd_debug(2, "Handle %p waiting for checkpoint...\n", handle); spin_unlock(&transaction->t_handle_lock); - __log_wait_for_space(journal); + __jbd2_log_wait_for_space(journal); goto repeat_locked; } @@ -223,7 +223,7 @@ repeat_locked: transaction->t_handle_count++; jbd_debug(4, "Handle %p given %d credits (total %d, free %d)\n", handle, nblocks, transaction->t_outstanding_credits, - __log_space_left(journal)); + __jbd2_log_space_left(journal)); spin_unlock(&transaction->t_handle_lock); spin_unlock(&journal->j_state_lock); out: @@ -246,7 +246,7 @@ static handle_t *new_handle(int nblocks) } /** - * handle_t *journal_start() - Obtain a new handle. + * handle_t *jbd2_journal_start() - Obtain a new handle. * @journal: Journal to start transaction on. * @nblocks: number of block buffer we might modify * @@ -259,7 +259,7 @@ static handle_t *new_handle(int nblocks) * * Return a pointer to a newly allocated handle, or NULL on failure */ -handle_t *journal_start(journal_t *journal, int nblocks) +handle_t *jbd2_journal_start(journal_t *journal, int nblocks) { handle_t *handle = journal_current_handle(); int err; @@ -289,7 +289,7 @@ handle_t *journal_start(journal_t *journal, int nblocks) } /** - * int journal_extend() - extend buffer credits. + * int jbd2_journal_extend() - extend buffer credits. * @handle: handle to 'extend' * @nblocks: nr blocks to try to extend by. * @@ -298,7 +298,7 @@ handle_t *journal_start(journal_t *journal, int nblocks) * a credit for a number of buffer modications in advance, but can * extend its credit if it needs more. * - * journal_extend tries to give the running handle more buffer credits. + * jbd2_journal_extend tries to give the running handle more buffer credits. * It does not guarantee that allocation - this is a best-effort only. * The calling process MUST be able to deal cleanly with a failure to * extend here. @@ -308,7 +308,7 @@ handle_t *journal_start(journal_t *journal, int nblocks) * return code < 0 implies an error * return code > 0 implies normal transaction-full status. */ -int journal_extend(handle_t *handle, int nblocks) +int jbd2_journal_extend(handle_t *handle, int nblocks) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; @@ -339,7 +339,7 @@ int journal_extend(handle_t *handle, int nblocks) goto unlock; } - if (wanted > __log_space_left(journal)) { + if (wanted > __jbd2_log_space_left(journal)) { jbd_debug(3, "denied handle %p %d blocks: " "insufficient log space\n", handle, nblocks); goto unlock; @@ -360,21 +360,21 @@ out: /** - * int journal_restart() - restart a handle . + * int jbd2_journal_restart() - restart a handle . * @handle: handle to restart * @nblocks: nr credits requested * * Restart a handle for a multi-transaction filesystem * operation. * - * If the journal_extend() call above fails to grant new buffer credits - * to a running handle, a call to journal_restart will commit the + * If the jbd2_journal_extend() call above fails to grant new buffer credits + * to a running handle, a call to jbd2_journal_restart will commit the * handle's transaction so far and reattach the handle to a new * transaction capabable of guaranteeing the requested number of * credits. */ -int journal_restart(handle_t *handle, int nblocks) +int jbd2_journal_restart(handle_t *handle, int nblocks) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; @@ -402,7 +402,7 @@ int journal_restart(handle_t *handle, int nblocks) spin_unlock(&transaction->t_handle_lock); jbd_debug(2, "restarting handle %p\n", handle); - __log_start_commit(journal, transaction->t_tid); + __jbd2_log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); handle->h_buffer_credits = nblocks; @@ -412,7 +412,7 @@ int journal_restart(handle_t *handle, int nblocks) /** - * void journal_lock_updates () - establish a transaction barrier. + * void jbd2_journal_lock_updates () - establish a transaction barrier. * @journal: Journal to establish a barrier on. * * This locks out any further updates from being started, and blocks @@ -421,7 +421,7 @@ int journal_restart(handle_t *handle, int nblocks) * * The journal lock should not be held on entry. */ -void journal_lock_updates(journal_t *journal) +void jbd2_journal_lock_updates(journal_t *journal) { DEFINE_WAIT(wait); @@ -452,7 +452,7 @@ void journal_lock_updates(journal_t *journal) /* * We have now established a barrier against other normal updates, but - * we also need to barrier against other journal_lock_updates() calls + * we also need to barrier against other jbd2_journal_lock_updates() calls * to make sure that we serialise special journal-locked operations * too. */ @@ -460,14 +460,14 @@ void journal_lock_updates(journal_t *journal) } /** - * void journal_unlock_updates (journal_t* journal) - release barrier + * void jbd2_journal_unlock_updates (journal_t* journal) - release barrier * @journal: Journal to release the barrier on. * - * Release a transaction barrier obtained with journal_lock_updates(). + * Release a transaction barrier obtained with jbd2_journal_lock_updates(). * * Should be called without the journal lock held. */ -void journal_unlock_updates (journal_t *journal) +void jbd2_journal_unlock_updates (journal_t *journal) { J_ASSERT(journal->j_barrier_count != 0); @@ -667,7 +667,7 @@ repeat: JBUFFER_TRACE(jh, "allocate memory for buffer"); jbd_unlock_bh_state(bh); frozen_buffer = - jbd_slab_alloc(jh2bh(jh)->b_size, + jbd2_slab_alloc(jh2bh(jh)->b_size, GFP_NOFS); if (!frozen_buffer) { printk(KERN_EMERG @@ -699,7 +699,7 @@ repeat: jh->b_transaction = transaction; JBUFFER_TRACE(jh, "file as BJ_Reserved"); spin_lock(&journal->j_list_lock); - __journal_file_buffer(jh, transaction, BJ_Reserved); + __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); spin_unlock(&journal->j_list_lock); } @@ -723,18 +723,18 @@ done: * If we are about to journal a buffer, then any revoke pending on it is * no longer valid */ - journal_cancel_revoke(handle, jh); + jbd2_journal_cancel_revoke(handle, jh); out: if (unlikely(frozen_buffer)) /* It's usually NULL */ - jbd_slab_free(frozen_buffer, bh->b_size); + jbd2_slab_free(frozen_buffer, bh->b_size); JBUFFER_TRACE(jh, "exit"); return error; } /** - * int journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. + * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. * @handle: transaction to add buffer modifications to * @bh: bh to be used for metadata writes * @credits: variable that will receive credits for the buffer @@ -745,16 +745,16 @@ out: * because we're write()ing a buffer which is also part of a shared mapping. */ -int journal_get_write_access(handle_t *handle, struct buffer_head *bh) +int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh) { - struct journal_head *jh = journal_add_journal_head(bh); + struct journal_head *jh = jbd2_journal_add_journal_head(bh); int rc; /* We do not want to get caught playing with fields which the * log thread also manipulates. Make sure that the buffer * completes any outstanding IO before proceeding. */ rc = do_get_write_access(handle, jh, 0); - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); return rc; } @@ -772,17 +772,17 @@ int journal_get_write_access(handle_t *handle, struct buffer_head *bh) * unlocked buffer beforehand. */ /** - * int journal_get_create_access () - notify intent to use newly created bh + * int jbd2_journal_get_create_access () - notify intent to use newly created bh * @handle: transaction to new buffer to * @bh: new buffer. * * Call this if you create a new bh. */ -int journal_get_create_access(handle_t *handle, struct buffer_head *bh) +int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; - struct journal_head *jh = journal_add_journal_head(bh); + struct journal_head *jh = jbd2_journal_add_journal_head(bh); int err; jbd_debug(5, "journal_head %p\n", jh); @@ -812,7 +812,7 @@ int journal_get_create_access(handle_t *handle, struct buffer_head *bh) if (jh->b_transaction == NULL) { jh->b_transaction = transaction; JBUFFER_TRACE(jh, "file as BJ_Reserved"); - __journal_file_buffer(jh, transaction, BJ_Reserved); + __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); } else if (jh->b_transaction == journal->j_committing_transaction) { JBUFFER_TRACE(jh, "set next transaction"); jh->b_next_transaction = transaction; @@ -828,14 +828,14 @@ int journal_get_create_access(handle_t *handle, struct buffer_head *bh) * which hits an assertion error. */ JBUFFER_TRACE(jh, "cancelling revoke"); - journal_cancel_revoke(handle, jh); - journal_put_journal_head(jh); + jbd2_journal_cancel_revoke(handle, jh); + jbd2_journal_put_journal_head(jh); out: return err; } /** - * int journal_get_undo_access() - Notify intent to modify metadata with + * int jbd2_journal_get_undo_access() - Notify intent to modify metadata with * non-rewindable consequences * @handle: transaction * @bh: buffer to undo @@ -848,7 +848,7 @@ out: * since if we overwrote that space we would make the delete * un-rewindable in case of a crash. * - * To deal with that, journal_get_undo_access requests write access to a + * To deal with that, jbd2_journal_get_undo_access requests write access to a * buffer for parts of non-rewindable operations such as delete * operations on the bitmaps. The journaling code must keep a copy of * the buffer's contents prior to the undo_access call until such time @@ -861,10 +861,10 @@ out: * * Returns error number or 0 on success. */ -int journal_get_undo_access(handle_t *handle, struct buffer_head *bh) +int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh) { int err; - struct journal_head *jh = journal_add_journal_head(bh); + struct journal_head *jh = jbd2_journal_add_journal_head(bh); char *committed_data = NULL; JBUFFER_TRACE(jh, "entry"); @@ -880,7 +880,7 @@ int journal_get_undo_access(handle_t *handle, struct buffer_head *bh) repeat: if (!jh->b_committed_data) { - committed_data = jbd_slab_alloc(jh2bh(jh)->b_size, GFP_NOFS); + committed_data = jbd2_slab_alloc(jh2bh(jh)->b_size, GFP_NOFS); if (!committed_data) { printk(KERN_EMERG "%s: No memory for committed data\n", __FUNCTION__); @@ -905,14 +905,14 @@ repeat: } jbd_unlock_bh_state(bh); out: - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); if (unlikely(committed_data)) - jbd_slab_free(committed_data, bh->b_size); + jbd2_slab_free(committed_data, bh->b_size); return err; } /** - * int journal_dirty_data() - mark a buffer as containing dirty data which + * int jbd2_journal_dirty_data() - mark a buffer as containing dirty data which * needs to be flushed before we can commit the * current transaction. * @handle: transaction @@ -923,10 +923,10 @@ out: * * Returns error number or 0 on success. * - * journal_dirty_data() can be called via page_launder->ext3_writepage + * jbd2_journal_dirty_data() can be called via page_launder->ext3_writepage * by kswapd. */ -int journal_dirty_data(handle_t *handle, struct buffer_head *bh) +int jbd2_journal_dirty_data(handle_t *handle, struct buffer_head *bh) { journal_t *journal = handle->h_transaction->t_journal; int need_brelse = 0; @@ -935,7 +935,7 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) if (is_handle_aborted(handle)) return 0; - jh = journal_add_journal_head(bh); + jh = jbd2_journal_add_journal_head(bh); JBUFFER_TRACE(jh, "entry"); /* @@ -984,7 +984,7 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) * And while we're in that state, someone does a * writepage() in an attempt to pageout the same area * of the file via a shared mapping. At present that - * calls journal_dirty_data(), and we get right here. + * calls jbd2_journal_dirty_data(), and we get right here. * It may be too late to journal the data. Simply * falling through to the next test will suffice: the * data will be dirty and wil be checkpointed. The @@ -1035,7 +1035,7 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) /* journal_clean_data_list() may have got there first */ if (jh->b_transaction != NULL) { JBUFFER_TRACE(jh, "unfile from commit"); - __journal_temp_unlink_buffer(jh); + __jbd2_journal_temp_unlink_buffer(jh); /* It still points to the committing * transaction; move it to this one so * that the refile assert checks are @@ -1054,15 +1054,15 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) { JBUFFER_TRACE(jh, "not on correct data list: unfile"); J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow); - __journal_temp_unlink_buffer(jh); + __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = handle->h_transaction; JBUFFER_TRACE(jh, "file as data"); - __journal_file_buffer(jh, handle->h_transaction, + __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); } } else { JBUFFER_TRACE(jh, "not on a transaction"); - __journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); + __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); } no_journal: spin_unlock(&journal->j_list_lock); @@ -1072,12 +1072,12 @@ no_journal: __brelse(bh); } JBUFFER_TRACE(jh, "exit"); - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); return 0; } /** - * int journal_dirty_metadata() - mark a buffer as containing dirty metadata + * int jbd2_journal_dirty_metadata() - mark a buffer as containing dirty metadata * @handle: transaction to add buffer to. * @bh: buffer to mark * @@ -1095,7 +1095,7 @@ no_journal: * buffer: that only gets done when the old transaction finally * completes its commit. */ -int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) +int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; @@ -1156,7 +1156,7 @@ int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) JBUFFER_TRACE(jh, "file as BJ_Metadata"); spin_lock(&journal->j_list_lock); - __journal_file_buffer(jh, handle->h_transaction, BJ_Metadata); + __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_Metadata); spin_unlock(&journal->j_list_lock); out_unlock_bh: jbd_unlock_bh_state(bh); @@ -1166,18 +1166,18 @@ out: } /* - * journal_release_buffer: undo a get_write_access without any buffer + * jbd2_journal_release_buffer: undo a get_write_access without any buffer * updates, if the update decided in the end that it didn't need access. * */ void -journal_release_buffer(handle_t *handle, struct buffer_head *bh) +jbd2_journal_release_buffer(handle_t *handle, struct buffer_head *bh) { BUFFER_TRACE(bh, "entry"); } /** - * void journal_forget() - bforget() for potentially-journaled buffers. + * void jbd2_journal_forget() - bforget() for potentially-journaled buffers. * @handle: transaction handle * @bh: bh to 'forget' * @@ -1193,7 +1193,7 @@ journal_release_buffer(handle_t *handle, struct buffer_head *bh) * Allow this call even if the handle has aborted --- it may be part of * the caller's cleanup after an abort. */ -int journal_forget (handle_t *handle, struct buffer_head *bh) +int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; @@ -1250,11 +1250,11 @@ int journal_forget (handle_t *handle, struct buffer_head *bh) */ if (jh->b_cp_transaction) { - __journal_temp_unlink_buffer(jh); - __journal_file_buffer(jh, transaction, BJ_Forget); + __jbd2_journal_temp_unlink_buffer(jh); + __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); } else { - __journal_unfile_buffer(jh); - journal_remove_journal_head(bh); + __jbd2_journal_unfile_buffer(jh); + jbd2_journal_remove_journal_head(bh); __brelse(bh); if (!buffer_jbd(bh)) { spin_unlock(&journal->j_list_lock); @@ -1292,7 +1292,7 @@ drop: } /** - * int journal_stop() - complete a transaction + * int jbd2_journal_stop() - complete a transaction * @handle: tranaction to complete. * * All done for a particular handle. @@ -1302,12 +1302,12 @@ drop: * complication is that we need to start a commit operation if the * filesystem is marked for synchronous update. * - * journal_stop itself will not usually return an error, but it may + * jbd2_journal_stop itself will not usually return an error, but it may * do so in unusual circumstances. In particular, expect it to - * return -EIO if a journal_abort has been executed since the + * return -EIO if a jbd2_journal_abort has been executed since the * transaction began. */ -int journal_stop(handle_t *handle) +int jbd2_journal_stop(handle_t *handle) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; @@ -1383,15 +1383,15 @@ int journal_stop(handle_t *handle) jbd_debug(2, "transaction too old, requesting commit for " "handle %p\n", handle); /* This is non-blocking */ - __log_start_commit(journal, transaction->t_tid); + __jbd2_log_start_commit(journal, transaction->t_tid); spin_unlock(&journal->j_state_lock); /* - * Special case: JFS_SYNC synchronous updates require us + * Special case: JBD2_SYNC synchronous updates require us * to wait for the commit to complete. */ if (handle->h_sync && !(current->flags & PF_MEMALLOC)) - err = log_wait_commit(journal, tid); + err = jbd2_log_wait_commit(journal, tid); } else { spin_unlock(&transaction->t_handle_lock); spin_unlock(&journal->j_state_lock); @@ -1401,24 +1401,24 @@ int journal_stop(handle_t *handle) return err; } -/**int journal_force_commit() - force any uncommitted transactions +/**int jbd2_journal_force_commit() - force any uncommitted transactions * @journal: journal to force * * For synchronous operations: force any uncommitted transactions * to disk. May seem kludgy, but it reuses all the handle batching * code in a very simple manner. */ -int journal_force_commit(journal_t *journal) +int jbd2_journal_force_commit(journal_t *journal) { handle_t *handle; int ret; - handle = journal_start(journal, 1); + handle = jbd2_journal_start(journal, 1); if (IS_ERR(handle)) { ret = PTR_ERR(handle); } else { handle->h_sync = 1; - ret = journal_stop(handle); + ret = jbd2_journal_stop(handle); } return ret; } @@ -1486,7 +1486,7 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh) * * Called under j_list_lock. The journal may not be locked. */ -void __journal_temp_unlink_buffer(struct journal_head *jh) +void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) { struct journal_head **list = NULL; transaction_t *transaction; @@ -1538,23 +1538,23 @@ void __journal_temp_unlink_buffer(struct journal_head *jh) mark_buffer_dirty(bh); /* Expose it to the VM */ } -void __journal_unfile_buffer(struct journal_head *jh) +void __jbd2_journal_unfile_buffer(struct journal_head *jh) { - __journal_temp_unlink_buffer(jh); + __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = NULL; } -void journal_unfile_buffer(journal_t *journal, struct journal_head *jh) +void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh) { jbd_lock_bh_state(jh2bh(jh)); spin_lock(&journal->j_list_lock); - __journal_unfile_buffer(jh); + __jbd2_journal_unfile_buffer(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(jh2bh(jh)); } /* - * Called from journal_try_to_free_buffers(). + * Called from jbd2_journal_try_to_free_buffers(). * * Called under jbd_lock_bh_state(bh) */ @@ -1576,16 +1576,16 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) { /* A written-back ordered data buffer */ JBUFFER_TRACE(jh, "release data"); - __journal_unfile_buffer(jh); - journal_remove_journal_head(bh); + __jbd2_journal_unfile_buffer(jh); + jbd2_journal_remove_journal_head(bh); __brelse(bh); } } else if (jh->b_cp_transaction != 0 && jh->b_transaction == 0) { /* written-back checkpointed metadata buffer */ if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); - __journal_remove_checkpoint(jh); - journal_remove_journal_head(bh); + __jbd2_journal_remove_checkpoint(jh); + jbd2_journal_remove_journal_head(bh); __brelse(bh); } } @@ -1596,7 +1596,7 @@ out: /** - * int journal_try_to_free_buffers() - try to free page buffers. + * int jbd2_journal_try_to_free_buffers() - try to free page buffers. * @journal: journal for operation * @page: to try and free * @unused_gfp_mask: unused @@ -1613,13 +1613,13 @@ out: * * This complicates JBD locking somewhat. We aren't protected by the * BKL here. We wish to remove the buffer from its committing or - * running transaction's ->t_datalist via __journal_unfile_buffer. + * running transaction's ->t_datalist via __jbd2_journal_unfile_buffer. * * This may *change* the value of transaction_t->t_datalist, so anyone * who looks at t_datalist needs to lock against this function. * - * Even worse, someone may be doing a journal_dirty_data on this - * buffer. So we need to lock against that. journal_dirty_data() + * Even worse, someone may be doing a jbd2_journal_dirty_data on this + * buffer. So we need to lock against that. jbd2_journal_dirty_data() * will come out of the lock with the buffer dirty, which makes it * ineligible for release here. * @@ -1629,7 +1629,7 @@ out: * cannot happen because we never reallocate freed data as metadata * while the data is part of a transaction. Yes? */ -int journal_try_to_free_buffers(journal_t *journal, +int jbd2_journal_try_to_free_buffers(journal_t *journal, struct page *page, gfp_t unused_gfp_mask) { struct buffer_head *head; @@ -1646,15 +1646,15 @@ int journal_try_to_free_buffers(journal_t *journal, /* * We take our own ref against the journal_head here to avoid * having to add tons of locking around each instance of - * journal_remove_journal_head() and journal_put_journal_head(). + * jbd2_journal_remove_journal_head() and jbd2_journal_put_journal_head(). */ - jh = journal_grab_journal_head(bh); + jh = jbd2_journal_grab_journal_head(bh); if (!jh) continue; jbd_lock_bh_state(bh); __journal_try_to_free_buffer(journal, bh); - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); jbd_unlock_bh_state(bh); if (buffer_jbd(bh)) goto busy; @@ -1681,23 +1681,23 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) int may_free = 1; struct buffer_head *bh = jh2bh(jh); - __journal_unfile_buffer(jh); + __jbd2_journal_unfile_buffer(jh); if (jh->b_cp_transaction) { JBUFFER_TRACE(jh, "on running+cp transaction"); - __journal_file_buffer(jh, transaction, BJ_Forget); + __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); clear_buffer_jbddirty(bh); may_free = 0; } else { JBUFFER_TRACE(jh, "on running transaction"); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); __brelse(bh); } return may_free; } /* - * journal_invalidatepage + * jbd2_journal_invalidatepage * * This code is tricky. It has a number of cases to deal with. * @@ -1765,7 +1765,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); - jh = journal_grab_journal_head(bh); + jh = jbd2_journal_grab_journal_head(bh); if (!jh) goto zap_buffer_no_jh; @@ -1796,7 +1796,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget"); ret = __dispose_buffer(jh, journal->j_running_transaction); - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); @@ -1810,7 +1810,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) JBUFFER_TRACE(jh, "give to committing trans"); ret = __dispose_buffer(jh, journal->j_committing_transaction); - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); @@ -1844,7 +1844,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) journal->j_running_transaction); jh->b_next_transaction = NULL; } - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); spin_unlock(&journal->j_state_lock); @@ -1861,7 +1861,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) } zap_buffer: - journal_put_journal_head(jh); + jbd2_journal_put_journal_head(jh); zap_buffer_no_jh: spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); @@ -1877,7 +1877,7 @@ zap_buffer_unlocked: } /** - * void journal_invalidatepage() + * void jbd2_journal_invalidatepage() * @journal: journal to use for flush... * @page: page to flush * @offset: length of page to invalidate. @@ -1885,7 +1885,7 @@ zap_buffer_unlocked: * Reap page buffers containing data after offset in page. * */ -void journal_invalidatepage(journal_t *journal, +void jbd2_journal_invalidatepage(journal_t *journal, struct page *page, unsigned long offset) { @@ -1927,7 +1927,7 @@ void journal_invalidatepage(journal_t *journal, /* * File a buffer on the given transaction list. */ -void __journal_file_buffer(struct journal_head *jh, +void __jbd2_journal_file_buffer(struct journal_head *jh, transaction_t *transaction, int jlist) { struct journal_head **list = NULL; @@ -1956,7 +1956,7 @@ void __journal_file_buffer(struct journal_head *jh, } if (jh->b_transaction) - __journal_temp_unlink_buffer(jh); + __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = transaction; switch (jlist) { @@ -1998,12 +1998,12 @@ void __journal_file_buffer(struct journal_head *jh, set_buffer_jbddirty(bh); } -void journal_file_buffer(struct journal_head *jh, +void jbd2_journal_file_buffer(struct journal_head *jh, transaction_t *transaction, int jlist) { jbd_lock_bh_state(jh2bh(jh)); spin_lock(&transaction->t_journal->j_list_lock); - __journal_file_buffer(jh, transaction, jlist); + __jbd2_journal_file_buffer(jh, transaction, jlist); spin_unlock(&transaction->t_journal->j_list_lock); jbd_unlock_bh_state(jh2bh(jh)); } @@ -2018,7 +2018,7 @@ void journal_file_buffer(struct journal_head *jh, * * Called under jbd_lock_bh_state(jh2bh(jh)) */ -void __journal_refile_buffer(struct journal_head *jh) +void __jbd2_journal_refile_buffer(struct journal_head *jh) { int was_dirty; struct buffer_head *bh = jh2bh(jh); @@ -2029,7 +2029,7 @@ void __journal_refile_buffer(struct journal_head *jh) /* If the buffer is now unused, just drop it. */ if (jh->b_next_transaction == NULL) { - __journal_unfile_buffer(jh); + __jbd2_journal_unfile_buffer(jh); return; } @@ -2039,10 +2039,10 @@ void __journal_refile_buffer(struct journal_head *jh) */ was_dirty = test_clear_buffer_jbddirty(bh); - __journal_temp_unlink_buffer(jh); + __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; - __journal_file_buffer(jh, jh->b_transaction, + __jbd2_journal_file_buffer(jh, jh->b_transaction, was_dirty ? BJ_Metadata : BJ_Reserved); J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING); @@ -2054,26 +2054,26 @@ void __journal_refile_buffer(struct journal_head *jh) * For the unlocked version of this call, also make sure that any * hanging journal_head is cleaned up if necessary. * - * __journal_refile_buffer is usually called as part of a single locked + * __jbd2_journal_refile_buffer is usually called as part of a single locked * operation on a buffer_head, in which the caller is probably going to * be hooking the journal_head onto other lists. In that case it is up * to the caller to remove the journal_head if necessary. For the - * unlocked journal_refile_buffer call, the caller isn't going to be + * unlocked jbd2_journal_refile_buffer call, the caller isn't going to be * doing anything else to the buffer so we need to do the cleanup * ourselves to avoid a jh leak. * * *** The journal_head may be freed by this call! *** */ -void journal_refile_buffer(journal_t *journal, struct journal_head *jh) +void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) { struct buffer_head *bh = jh2bh(jh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); - __journal_refile_buffer(jh); + __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); - journal_remove_journal_head(bh); + jbd2_journal_remove_journal_head(bh); spin_unlock(&journal->j_list_lock); __brelse(bh); diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h index 3dbf6c7..99d3755 100644 --- a/include/linux/ext4_jbd2.h +++ b/include/linux/ext4_jbd2.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/ext4_jbd.h + * linux/include/linux/ext4_jbd2.h * * Written by Stephen C. Tweedie , 1999 * @@ -16,7 +16,7 @@ #define _LINUX_EXT4_JBD_H #include -#include +#include #include #define EXT4_JOURNAL(inode) (EXT4_SB((inode)->i_sb)->s_journal) @@ -116,7 +116,7 @@ static inline int __ext4_journal_get_undo_access(const char *where, handle_t *handle, struct buffer_head *bh) { - int err = journal_get_undo_access(handle, bh); + int err = jbd2_journal_get_undo_access(handle, bh); if (err) ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; @@ -126,7 +126,7 @@ static inline int __ext4_journal_get_write_access(const char *where, handle_t *handle, struct buffer_head *bh) { - int err = journal_get_write_access(handle, bh); + int err = jbd2_journal_get_write_access(handle, bh); if (err) ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; @@ -135,13 +135,13 @@ __ext4_journal_get_write_access(const char *where, handle_t *handle, static inline void ext4_journal_release_buffer(handle_t *handle, struct buffer_head *bh) { - journal_release_buffer(handle, bh); + jbd2_journal_release_buffer(handle, bh); } static inline int __ext4_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) { - int err = journal_forget(handle, bh); + int err = jbd2_journal_forget(handle, bh); if (err) ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; @@ -151,7 +151,7 @@ static inline int __ext4_journal_revoke(const char *where, handle_t *handle, unsigned long blocknr, struct buffer_head *bh) { - int err = journal_revoke(handle, blocknr, bh); + int err = jbd2_journal_revoke(handle, blocknr, bh); if (err) ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; @@ -161,7 +161,7 @@ static inline int __ext4_journal_get_create_access(const char *where, handle_t *handle, struct buffer_head *bh) { - int err = journal_get_create_access(handle, bh); + int err = jbd2_journal_get_create_access(handle, bh); if (err) ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; @@ -171,7 +171,7 @@ static inline int __ext4_journal_dirty_metadata(const char *where, handle_t *handle, struct buffer_head *bh) { - int err = journal_dirty_metadata(handle, bh); + int err = jbd2_journal_dirty_metadata(handle, bh); if (err) ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); return err; @@ -211,22 +211,22 @@ static inline handle_t *ext4_journal_current_handle(void) static inline int ext4_journal_extend(handle_t *handle, int nblocks) { - return journal_extend(handle, nblocks); + return jbd2_journal_extend(handle, nblocks); } static inline int ext4_journal_restart(handle_t *handle, int nblocks) { - return journal_restart(handle, nblocks); + return jbd2_journal_restart(handle, nblocks); } static inline int ext4_journal_blocks_per_page(struct inode *inode) { - return journal_blocks_per_page(inode); + return jbd2_journal_blocks_per_page(inode); } static inline int ext4_journal_force_commit(journal_t *journal) { - return journal_force_commit(journal); + return jbd2_journal_force_commit(journal); } /* super.c */ diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index fe89444..3251f7a 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1,5 +1,5 @@ /* - * linux/include/linux/jbd.h + * linux/include/linux/jbd2.h * * Written by Stephen C. Tweedie * @@ -19,7 +19,7 @@ /* Allow this file to be included directly into e2fsprogs */ #ifndef __KERNEL__ #include "jfs_compat.h" -#define JFS_DEBUG +#define JBD2_DEBUG #define jfs_debug jbd_debug #else @@ -57,11 +57,11 @@ * CONFIG_JBD_DEBUG is on. */ #define JBD_EXPENSIVE_CHECKING -extern int journal_enable_debug; +extern int jbd2_journal_enable_debug; #define jbd_debug(n, f, a...) \ do { \ - if ((n) <= journal_enable_debug) { \ + if ((n) <= jbd2_journal_enable_debug) { \ printk (KERN_DEBUG "(%s, %d): %s: ", \ __FILE__, __LINE__, __FUNCTION__); \ printk (f, ## a); \ @@ -71,16 +71,16 @@ extern int journal_enable_debug; #define jbd_debug(f, a...) /**/ #endif -extern void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry); -extern void * jbd_slab_alloc(size_t size, gfp_t flags); -extern void jbd_slab_free(void *ptr, size_t size); +extern void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry); +extern void * jbd2_slab_alloc(size_t size, gfp_t flags); +extern void jbd2_slab_free(void *ptr, size_t size); #define jbd_kmalloc(size, flags) \ - __jbd_kmalloc(__FUNCTION__, (size), (flags), journal_oom_retry) + __jbd2_kmalloc(__FUNCTION__, (size), (flags), journal_oom_retry) #define jbd_rep_kmalloc(size, flags) \ - __jbd_kmalloc(__FUNCTION__, (size), (flags), 1) + __jbd2_kmalloc(__FUNCTION__, (size), (flags), 1) -#define JFS_MIN_JOURNAL_BLOCKS 1024 +#define JBD2_MIN_JOURNAL_BLOCKS 1024 #ifdef __KERNEL__ @@ -122,7 +122,7 @@ typedef struct journal_s journal_t; /* Journal control structure */ * Internal structures used by the logging mechanism: */ -#define JFS_MAGIC_NUMBER 0xc03b3998U /* The first 4 bytes of /dev/random! */ +#define JBD2_MAGIC_NUMBER 0xc03b3998U /* The first 4 bytes of /dev/random! */ /* * On-disk structures @@ -132,11 +132,11 @@ typedef struct journal_s journal_t; /* Journal control structure */ * Descriptor block types: */ -#define JFS_DESCRIPTOR_BLOCK 1 -#define JFS_COMMIT_BLOCK 2 -#define JFS_SUPERBLOCK_V1 3 -#define JFS_SUPERBLOCK_V2 4 -#define JFS_REVOKE_BLOCK 5 +#define JBD2_DESCRIPTOR_BLOCK 1 +#define JBD2_COMMIT_BLOCK 2 +#define JBD2_SUPERBLOCK_V1 3 +#define JBD2_SUPERBLOCK_V2 4 +#define JBD2_REVOKE_BLOCK 5 /* * Standard header for all descriptor blocks: @@ -162,18 +162,18 @@ typedef struct journal_block_tag_s * The revoke descriptor: used on disk to describe a series of blocks to * be revoked from the log */ -typedef struct journal_revoke_header_s +typedef struct jbd2_journal_revoke_header_s { journal_header_t r_header; __be32 r_count; /* Count of bytes used in the block */ -} journal_revoke_header_t; +} jbd2_journal_revoke_header_t; /* Definitions for the journal tag flags word: */ -#define JFS_FLAG_ESCAPE 1 /* on-disk block is escaped */ -#define JFS_FLAG_SAME_UUID 2 /* block has same uuid as previous */ -#define JFS_FLAG_DELETED 4 /* block deleted by this transaction */ -#define JFS_FLAG_LAST_TAG 8 /* last tag in this descriptor block */ +#define JBD2_FLAG_ESCAPE 1 /* on-disk block is escaped */ +#define JBD2_FLAG_SAME_UUID 2 /* block has same uuid as previous */ +#define JBD2_FLAG_DELETED 4 /* block deleted by this transaction */ +#define JBD2_FLAG_LAST_TAG 8 /* last tag in this descriptor block */ /* @@ -196,7 +196,7 @@ typedef struct journal_superblock_s __be32 s_start; /* blocknr of start of log */ /* 0x0020 */ - /* Error value, as set by journal_abort(). */ + /* Error value, as set by jbd2_journal_abort(). */ __be32 s_errno; /* 0x0024 */ @@ -224,22 +224,22 @@ typedef struct journal_superblock_s /* 0x0400 */ } journal_superblock_t; -#define JFS_HAS_COMPAT_FEATURE(j,mask) \ +#define JBD2_HAS_COMPAT_FEATURE(j,mask) \ ((j)->j_format_version >= 2 && \ ((j)->j_superblock->s_feature_compat & cpu_to_be32((mask)))) -#define JFS_HAS_RO_COMPAT_FEATURE(j,mask) \ +#define JBD2_HAS_RO_COMPAT_FEATURE(j,mask) \ ((j)->j_format_version >= 2 && \ ((j)->j_superblock->s_feature_ro_compat & cpu_to_be32((mask)))) -#define JFS_HAS_INCOMPAT_FEATURE(j,mask) \ +#define JBD2_HAS_INCOMPAT_FEATURE(j,mask) \ ((j)->j_format_version >= 2 && \ ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask)))) -#define JFS_FEATURE_INCOMPAT_REVOKE 0x00000001 +#define JBD2_FEATURE_INCOMPAT_REVOKE 0x00000001 /* Features known to this kernel version: */ -#define JFS_KNOWN_COMPAT_FEATURES 0 -#define JFS_KNOWN_ROCOMPAT_FEATURES 0 -#define JFS_KNOWN_INCOMPAT_FEATURES JFS_FEATURE_INCOMPAT_REVOKE +#define JBD2_KNOWN_COMPAT_FEATURES 0 +#define JBD2_KNOWN_ROCOMPAT_FEATURES 0 +#define JBD2_KNOWN_INCOMPAT_FEATURES JBD2_FEATURE_INCOMPAT_REVOKE #ifdef __KERNEL__ @@ -359,7 +359,7 @@ static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh) bit_spin_unlock(BH_JournalHead, &bh->b_state); } -struct jbd_revoke_table_s; +struct jbd2_revoke_table_s; /** * struct handle_s - The handle_s type is the concrete type associated with @@ -445,7 +445,7 @@ struct transaction_s /* * Transaction's current state - * [no locking - only kjournald alters this] + * [no locking - only kjournald2 alters this] * FIXME: needs barriers * KLUDGE: [use j_state_lock] */ @@ -621,7 +621,7 @@ struct transaction_s * @j_revoke: The revoke table - maintains the list of revoked blocks in the * current transaction. * @j_revoke_table: alternate revoke tables for j_revoke - * @j_wbuf: array of buffer_heads for journal_commit_transaction + * @j_wbuf: array of buffer_heads for jbd2_journal_commit_transaction * @j_wbufsize: maximum number of buffer_heads allowed in j_wbuf, the * number that will fit in j_blocksize * @j_last_sync_writer: most recent pid which did a synchronous write @@ -805,11 +805,11 @@ struct journal_s * current transaction. [j_revoke_lock] */ spinlock_t j_revoke_lock; - struct jbd_revoke_table_s *j_revoke; - struct jbd_revoke_table_s *j_revoke_table[2]; + struct jbd2_revoke_table_s *j_revoke; + struct jbd2_revoke_table_s *j_revoke_table[2]; /* - * array of bhs for journal_commit_transaction + * array of bhs for jbd2_journal_commit_transaction */ struct buffer_head **j_wbuf; int j_wbufsize; @@ -826,12 +826,12 @@ struct journal_s /* * Journal flag definitions */ -#define JFS_UNMOUNT 0x001 /* Journal thread is being destroyed */ -#define JFS_ABORT 0x002 /* Journaling has been aborted for errors. */ -#define JFS_ACK_ERR 0x004 /* The errno in the sb has been acked */ -#define JFS_FLUSHED 0x008 /* The journal superblock has been flushed */ -#define JFS_LOADED 0x010 /* The journal superblock has been loaded */ -#define JFS_BARRIER 0x020 /* Use IDE barriers */ +#define JBD2_UNMOUNT 0x001 /* Journal thread is being destroyed */ +#define JBD2_ABORT 0x002 /* Journaling has been aborted for errors. */ +#define JBD2_ACK_ERR 0x004 /* The errno in the sb has been acked */ +#define JBD2_FLUSHED 0x008 /* The journal superblock has been flushed */ +#define JBD2_LOADED 0x010 /* The journal superblock has been loaded */ +#define JBD2_BARRIER 0x020 /* Use IDE barriers */ /* * Function declarations for the journaling transaction and buffer @@ -839,31 +839,31 @@ struct journal_s */ /* Filing buffers */ -extern void __journal_temp_unlink_buffer(struct journal_head *jh); -extern void journal_unfile_buffer(journal_t *, struct journal_head *); -extern void __journal_unfile_buffer(struct journal_head *); -extern void __journal_refile_buffer(struct journal_head *); -extern void journal_refile_buffer(journal_t *, struct journal_head *); -extern void __journal_file_buffer(struct journal_head *, transaction_t *, int); +extern void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); +extern void jbd2_journal_unfile_buffer(journal_t *, struct journal_head *); +extern void __jbd2_journal_unfile_buffer(struct journal_head *); +extern void __jbd2_journal_refile_buffer(struct journal_head *); +extern void jbd2_journal_refile_buffer(journal_t *, struct journal_head *); +extern void __jbd2_journal_file_buffer(struct journal_head *, transaction_t *, int); extern void __journal_free_buffer(struct journal_head *bh); -extern void journal_file_buffer(struct journal_head *, transaction_t *, int); +extern void jbd2_journal_file_buffer(struct journal_head *, transaction_t *, int); extern void __journal_clean_data_list(transaction_t *transaction); /* Log buffer allocation */ -extern struct journal_head * journal_get_descriptor_buffer(journal_t *); -int journal_next_log_block(journal_t *, unsigned long *); +extern struct journal_head * jbd2_journal_get_descriptor_buffer(journal_t *); +int jbd2_journal_next_log_block(journal_t *, unsigned long *); /* Commit management */ -extern void journal_commit_transaction(journal_t *); +extern void jbd2_journal_commit_transaction(journal_t *); /* Checkpoint list management */ -int __journal_clean_checkpoint_list(journal_t *journal); -int __journal_remove_checkpoint(struct journal_head *); -void __journal_insert_checkpoint(struct journal_head *, transaction_t *); +int __jbd2_journal_clean_checkpoint_list(journal_t *journal); +int __jbd2_journal_remove_checkpoint(struct journal_head *); +void __jbd2_journal_insert_checkpoint(struct journal_head *, transaction_t *); /* Buffer IO */ extern int -journal_write_metadata_buffer(transaction_t *transaction, +jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, struct journal_head **jh_out, unsigned long blocknr); @@ -893,91 +893,91 @@ static inline handle_t *journal_current_handle(void) * Register buffer modifications against the current transaction. */ -extern handle_t *journal_start(journal_t *, int nblocks); -extern int journal_restart (handle_t *, int nblocks); -extern int journal_extend (handle_t *, int nblocks); -extern int journal_get_write_access(handle_t *, struct buffer_head *); -extern int journal_get_create_access (handle_t *, struct buffer_head *); -extern int journal_get_undo_access(handle_t *, struct buffer_head *); -extern int journal_dirty_data (handle_t *, struct buffer_head *); -extern int journal_dirty_metadata (handle_t *, struct buffer_head *); -extern void journal_release_buffer (handle_t *, struct buffer_head *); -extern int journal_forget (handle_t *, struct buffer_head *); +extern handle_t *jbd2_journal_start(journal_t *, int nblocks); +extern int jbd2_journal_restart (handle_t *, int nblocks); +extern int jbd2_journal_extend (handle_t *, int nblocks); +extern int jbd2_journal_get_write_access(handle_t *, struct buffer_head *); +extern int jbd2_journal_get_create_access (handle_t *, struct buffer_head *); +extern int jbd2_journal_get_undo_access(handle_t *, struct buffer_head *); +extern int jbd2_journal_dirty_data (handle_t *, struct buffer_head *); +extern int jbd2_journal_dirty_metadata (handle_t *, struct buffer_head *); +extern void jbd2_journal_release_buffer (handle_t *, struct buffer_head *); +extern int jbd2_journal_forget (handle_t *, struct buffer_head *); extern void journal_sync_buffer (struct buffer_head *); -extern void journal_invalidatepage(journal_t *, +extern void jbd2_journal_invalidatepage(journal_t *, struct page *, unsigned long); -extern int journal_try_to_free_buffers(journal_t *, struct page *, gfp_t); -extern int journal_stop(handle_t *); -extern int journal_flush (journal_t *); -extern void journal_lock_updates (journal_t *); -extern void journal_unlock_updates (journal_t *); +extern int jbd2_journal_try_to_free_buffers(journal_t *, struct page *, gfp_t); +extern int jbd2_journal_stop(handle_t *); +extern int jbd2_journal_flush (journal_t *); +extern void jbd2_journal_lock_updates (journal_t *); +extern void jbd2_journal_unlock_updates (journal_t *); -extern journal_t * journal_init_dev(struct block_device *bdev, +extern journal_t * jbd2_journal_init_dev(struct block_device *bdev, struct block_device *fs_dev, int start, int len, int bsize); -extern journal_t * journal_init_inode (struct inode *); -extern int journal_update_format (journal_t *); -extern int journal_check_used_features +extern journal_t * jbd2_journal_init_inode (struct inode *); +extern int jbd2_journal_update_format (journal_t *); +extern int jbd2_journal_check_used_features (journal_t *, unsigned long, unsigned long, unsigned long); -extern int journal_check_available_features +extern int jbd2_journal_check_available_features (journal_t *, unsigned long, unsigned long, unsigned long); -extern int journal_set_features +extern int jbd2_journal_set_features (journal_t *, unsigned long, unsigned long, unsigned long); -extern int journal_create (journal_t *); -extern int journal_load (journal_t *journal); -extern void journal_destroy (journal_t *); -extern int journal_recover (journal_t *journal); -extern int journal_wipe (journal_t *, int); -extern int journal_skip_recovery (journal_t *); -extern void journal_update_superblock (journal_t *, int); -extern void __journal_abort_hard (journal_t *); -extern void journal_abort (journal_t *, int); -extern int journal_errno (journal_t *); -extern void journal_ack_err (journal_t *); -extern int journal_clear_err (journal_t *); -extern int journal_bmap(journal_t *, unsigned long, unsigned long *); -extern int journal_force_commit(journal_t *); +extern int jbd2_journal_create (journal_t *); +extern int jbd2_journal_load (journal_t *journal); +extern void jbd2_journal_destroy (journal_t *); +extern int jbd2_journal_recover (journal_t *journal); +extern int jbd2_journal_wipe (journal_t *, int); +extern int jbd2_journal_skip_recovery (journal_t *); +extern void jbd2_journal_update_superblock (journal_t *, int); +extern void __jbd2_journal_abort_hard (journal_t *); +extern void jbd2_journal_abort (journal_t *, int); +extern int jbd2_journal_errno (journal_t *); +extern void jbd2_journal_ack_err (journal_t *); +extern int jbd2_journal_clear_err (journal_t *); +extern int jbd2_journal_bmap(journal_t *, unsigned long, unsigned long *); +extern int jbd2_journal_force_commit(journal_t *); /* * journal_head management */ -struct journal_head *journal_add_journal_head(struct buffer_head *bh); -struct journal_head *journal_grab_journal_head(struct buffer_head *bh); -void journal_remove_journal_head(struct buffer_head *bh); -void journal_put_journal_head(struct journal_head *jh); +struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh); +struct journal_head *jbd2_journal_grab_journal_head(struct buffer_head *bh); +void jbd2_journal_remove_journal_head(struct buffer_head *bh); +void jbd2_journal_put_journal_head(struct journal_head *jh); /* * handle management */ -extern kmem_cache_t *jbd_handle_cache; +extern kmem_cache_t *jbd2_handle_cache; static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags) { - return kmem_cache_alloc(jbd_handle_cache, gfp_flags); + return kmem_cache_alloc(jbd2_handle_cache, gfp_flags); } static inline void jbd_free_handle(handle_t *handle) { - kmem_cache_free(jbd_handle_cache, handle); + kmem_cache_free(jbd2_handle_cache, handle); } /* Primary revoke support */ #define JOURNAL_REVOKE_DEFAULT_HASH 256 -extern int journal_init_revoke(journal_t *, int); -extern void journal_destroy_revoke_caches(void); -extern int journal_init_revoke_caches(void); +extern int jbd2_journal_init_revoke(journal_t *, int); +extern void jbd2_journal_destroy_revoke_caches(void); +extern int jbd2_journal_init_revoke_caches(void); -extern void journal_destroy_revoke(journal_t *); -extern int journal_revoke (handle_t *, +extern void jbd2_journal_destroy_revoke(journal_t *); +extern int jbd2_journal_revoke (handle_t *, unsigned long, struct buffer_head *); -extern int journal_cancel_revoke(handle_t *, struct journal_head *); -extern void journal_write_revoke_records(journal_t *, transaction_t *); +extern int jbd2_journal_cancel_revoke(handle_t *, struct journal_head *); +extern void jbd2_journal_write_revoke_records(journal_t *, transaction_t *); /* Recovery revoke support */ -extern int journal_set_revoke(journal_t *, unsigned long, tid_t); -extern int journal_test_revoke(journal_t *, unsigned long, tid_t); -extern void journal_clear_revoke(journal_t *); -extern void journal_switch_revoke_table(journal_t *journal); +extern int jbd2_journal_set_revoke(journal_t *, unsigned long, tid_t); +extern int jbd2_journal_test_revoke(journal_t *, unsigned long, tid_t); +extern void jbd2_journal_clear_revoke(journal_t *); +extern void jbd2_journal_switch_revoke_table(journal_t *journal); /* * The log thread user interface: @@ -986,17 +986,17 @@ extern void journal_switch_revoke_table(journal_t *journal); * transitions on demand. */ -int __log_space_left(journal_t *); /* Called with journal locked */ -int log_start_commit(journal_t *journal, tid_t tid); -int __log_start_commit(journal_t *journal, tid_t tid); -int journal_start_commit(journal_t *journal, tid_t *tid); -int journal_force_commit_nested(journal_t *journal); -int log_wait_commit(journal_t *journal, tid_t tid); -int log_do_checkpoint(journal_t *journal); +int __jbd2_log_space_left(journal_t *); /* Called with journal locked */ +int jbd2_log_start_commit(journal_t *journal, tid_t tid); +int __jbd2_log_start_commit(journal_t *journal, tid_t tid); +int jbd2_journal_start_commit(journal_t *journal, tid_t *tid); +int jbd2_journal_force_commit_nested(journal_t *journal); +int jbd2_log_wait_commit(journal_t *journal, tid_t tid); +int jbd2_log_do_checkpoint(journal_t *journal); -void __log_wait_for_space(journal_t *journal); -extern void __journal_drop_transaction(journal_t *, transaction_t *); -extern int cleanup_journal_tail(journal_t *); +void __jbd2_log_wait_for_space(journal_t *journal); +extern void __jbd2_journal_drop_transaction(journal_t *, transaction_t *); +extern int jbd2_cleanup_journal_tail(journal_t *); /* Debugging code only: */ @@ -1010,7 +1010,7 @@ do { \ /* * is_journal_abort * - * Simple test wrapper function to test the JFS_ABORT state flag. This + * Simple test wrapper function to test the JBD2_ABORT state flag. This * bit, when set, indicates that we have had a fatal error somewhere, * either inside the journaling layer or indicated to us by the client * (eg. ext3), and that we and should not commit any further @@ -1019,7 +1019,7 @@ do { \ static inline int is_journal_aborted(journal_t *journal) { - return journal->j_flags & JFS_ABORT; + return journal->j_flags & JBD2_ABORT; } static inline int is_handle_aborted(handle_t *handle) @@ -1029,7 +1029,7 @@ static inline int is_handle_aborted(handle_t *handle) return is_journal_aborted(handle->h_transaction->t_journal); } -static inline void journal_abort_handle(handle_t *handle) +static inline void jbd2_journal_abort_handle(handle_t *handle) { handle->h_aborted = 1; } @@ -1051,7 +1051,7 @@ static inline int tid_geq(tid_t x, tid_t y) return (difference >= 0); } -extern int journal_blocks_per_page(struct inode *inode); +extern int jbd2_journal_blocks_per_page(struct inode *inode); /* * Return the minimum number of blocks which must be free in the journal -- cgit v0.10.2 From a920e9416b3469994860ab552dfd7fd5a5aff162 Mon Sep 17 00:00:00 2001 From: Johann Lombardi Date: Wed, 11 Oct 2006 01:21:00 -0700 Subject: [PATCH] jbd2: rename slab jbd and jbd2 currently use the same slab names which must be unique. The patch below just renames jbd2's slabs. Signed-off-by: Johann Lombardi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 3fbbba2..8d0f71e 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1631,7 +1631,7 @@ void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry) static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { - "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" + "jbd2_1k", "jbd2_2k", "jbd2_4k", NULL, "jbd2_8k" }; static void jbd2_journal_destroy_jbd_slabs(void) @@ -1702,7 +1702,7 @@ static int journal_init_jbd2_journal_head_cache(void) int retval; J_ASSERT(jbd2_journal_head_cache == 0); - jbd2_journal_head_cache = kmem_cache_create("journal_head", + jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head", sizeof(struct journal_head), 0, /* offset */ 0, /* flags */ @@ -1999,7 +1999,7 @@ kmem_cache_t *jbd2_handle_cache; static int __init journal_init_handle_cache(void) { - jbd2_handle_cache = kmem_cache_create("journal_handle", + jbd2_handle_cache = kmem_cache_create("jbd2_journal_handle", sizeof(handle_t), 0, /* offset */ 0, /* flags */ diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 2fccddc..5820a0c 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -168,13 +168,13 @@ static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, int __init jbd2_journal_init_revoke_caches(void) { - jbd2_revoke_record_cache = kmem_cache_create("revoke_record", + jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record", sizeof(struct jbd2_revoke_record_s), 0, SLAB_HWCACHE_ALIGN, NULL, NULL); if (jbd2_revoke_record_cache == 0) return -ENOMEM; - jbd2_revoke_table_cache = kmem_cache_create("revoke_table", + jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table", sizeof(struct jbd2_revoke_table_s), 0, 0, NULL, NULL); if (jbd2_revoke_table_cache == 0) { -- cgit v0.10.2 From dab291af8d6307a3075c3d67d0cc8f98e646cb94 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:21:01 -0700 Subject: [PATCH] jbd2: enable building of jbd2 and have ext4 use it rather than jbd Reworked from a patch by Mingming Cao and Randy Dunlap Signed-off-By: Randy Dunlap Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/Kconfig b/fs/Kconfig index ac9ba1c..db4d133 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -143,24 +143,24 @@ config EXT3_FS_SECURITY config EXT4DEV_FS tristate "Ext4dev/ext4 extended fs support development (EXPERIMENTAL)" depends on EXPERIMENTAL - select JBD - help + select JBD2 + help Ext4dev is a predecessor filesystem of the next generation extended fs ext4, based on ext3 filesystem code. It will be renamed ext4 fs later, once ext4dev is mature and stabilized. - Unlike the change from ext2 filesystem to ext3 filesystem, - the on-disk format of ext4dev is not the same as ext3 any more: + Unlike the change from ext2 filesystem to ext3 filesystem, + the on-disk format of ext4dev is not the same as ext3 any more: it is based on extent maps and it supports 48-bit physical block - numbers. These combined on-disk format changes will allow + numbers. These combined on-disk format changes will allow ext4dev/ext4 to handle more than 16 TB filesystem volumes -- a hard limit that ext3 cannot overcome without changing the - on-disk format. + on-disk format. Other than extent maps and 48-bit block numbers, ext4dev also is - likely to have other new features such as persistent preallocation, + likely to have other new features such as persistent preallocation, high resolution time stamps, and larger file support etc. These - features will be added to ext4dev gradually. + features will be added to ext4dev gradually. To compile this file system support as a module, choose M here. The module will be called ext4dev. Be aware, however, that the filesystem @@ -239,6 +239,38 @@ config JBD_DEBUG generated. To turn debugging off again, do "echo 0 > /proc/sys/fs/jbd-debug". +config JBD2 + tristate + help + This is a generic journaling layer for block devices that support + both 32-bit and 64-bit block numbers. It is currently used by + the ext4dev/ext4 filesystem, but it could also be used to add + journal support to other file systems or block devices such + as RAID or LVM. + + If you are using ext4dev/ext4, you need to say Y here. If you are not + using ext4dev/ext4 then you will probably want to say N. + + To compile this device as a module, choose M here. The module will be + called jbd2. If you are compiling ext4dev/ext4 into the kernel, + you cannot compile this code as a module. + +config JBD2_DEBUG + bool "JBD2 (ext4dev/ext4) debugging support" + depends on JBD2 + help + If you are using the ext4dev/ext4 journaled file system (or + potentially any other filesystem/device using JBD2), this option + allows you to enable debugging output while the system is running, + in order to help track down any problems you are having. + By default, the debugging output will be turned off. + + If you select Y here, then you will be able to turn on debugging + with "echo N > /proc/sys/fs/jbd2-debug", where N is a number between + 1 and 5. The higher the number, the more debugging output is + generated. To turn debugging off again, do + "echo 0 > /proc/sys/fs/jbd2-debug". + config FS_MBCACHE # Meta block cache for Extended Attributes (ext2/ext3/ext4) tristate diff --git a/fs/Makefile b/fs/Makefile index 64396af..9a5ce93 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -64,6 +64,7 @@ obj-$(CONFIG_REISERFS_FS) += reiserfs/ obj-$(CONFIG_EXT3_FS) += ext3/ # Before ext2 so root fs can be ext3 obj-$(CONFIG_EXT4DEV_FS) += ext4/ # Before ext2 so root fs can be ext4dev obj-$(CONFIG_JBD) += jbd/ +obj-$(CONFIG_JBD2) += jbd2/ obj-$(CONFIG_EXT2_FS) += ext2/ obj-$(CONFIG_CRAMFS) += cramfs/ obj-$(CONFIG_RAMFS) += ramfs/ diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index d143489..0a965dd 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #include #include "xattr.h" #include "acl.h" diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 357e4e5..e9e9844 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -14,9 +14,9 @@ #include #include #include -#include +#include #include -#include +#include #include #include @@ -526,12 +526,12 @@ do_more: * transaction. * * Ideally we would want to allow that to happen, but to - * do so requires making journal_forget() capable of + * do so requires making jbd2_journal_forget() capable of * revoking the queued write of a data block, which * implies blocking on the journal lock. *forget() * cannot block due to truncate races. * - * Eventually we can fix this by making journal_forget() + * Eventually we can fix this by making jbd2_journal_forget() * return a status indicating whether or not it was able * to revoke the buffer. On successful revoke, it is * safe not to set the allocation bit in the committed @@ -1382,7 +1382,7 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); - return journal_force_commit_nested(EXT4_SB(sb)->s_journal); + return jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal); } /** diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index f4b3570..11e93c1 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -8,7 +8,7 @@ */ #include -#include +#include #include #ifdef EXT4FS_DEBUG diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index ec114d7..9833d5d 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -22,7 +22,7 @@ */ #include -#include +#include #include #include #include diff --git a/fs/ext4/file.c b/fs/ext4/file.c index d938fbe..0b622c0 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -20,9 +20,9 @@ #include #include -#include +#include #include -#include +#include #include "xattr.h" #include "acl.h" diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 272faa2..2a167d7 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -26,9 +26,9 @@ #include #include #include -#include +#include #include -#include +#include /* * akpm: A new design for ext4_sync_file(). diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c index d15bb42..a679663 100644 --- a/fs/ext4/hash.c +++ b/fs/ext4/hash.c @@ -10,7 +10,7 @@ */ #include -#include +#include #include #include #include diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 4b92066..34d39ae 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -14,9 +14,9 @@ #include #include -#include +#include #include -#include +#include #include #include #include @@ -497,7 +497,7 @@ repeat_in_this_group: goto got; } /* we lost it */ - journal_release_buffer(handle, bitmap_bh); + jbd2_journal_release_buffer(handle, bitmap_bh); if (++ino < EXT4_INODES_PER_GROUP(sb)) goto repeat_in_this_group; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 7275d60..0d96c7d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -25,8 +25,8 @@ #include #include #include -#include -#include +#include +#include #include #include #include @@ -84,7 +84,7 @@ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA || (!is_metadata && !ext4_should_journal_data(inode))) { if (bh) { - BUFFER_TRACE(bh, "call journal_forget"); + BUFFER_TRACE(bh, "call jbd2_journal_forget"); return ext4_journal_forget(handle, bh); } return 0; @@ -657,7 +657,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, failed: /* Allocation failed, free what we already allocated */ for (i = 1; i <= n ; i++) { - BUFFER_TRACE(branch[i].bh, "call journal_forget"); + BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget"); ext4_journal_forget(handle, branch[i].bh); } for (i = 0; i @@ -1135,7 +1135,7 @@ static int walk_page_buffers( handle_t *handle, * transaction open and was blocking on the quota lock - a ranking * violation. * - * So what we do is to rely on the fact that journal_stop/journal_start + * So what we do is to rely on the fact that jbd2_journal_stop/journal_start * will _not_ run commit under these circumstances because handle->h_ref * is elevated. We'll still have enough credits for the tiny quotafile * write. @@ -1184,7 +1184,7 @@ out: int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh) { - int err = journal_dirty_data(handle, bh); + int err = jbd2_journal_dirty_data(handle, bh); if (err) ext4_journal_abort_handle(__FUNCTION__, __FUNCTION__, bh, handle,err); @@ -1333,9 +1333,9 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) EXT4_I(inode)->i_state &= ~EXT4_STATE_JDATA; journal = EXT4_JOURNAL(inode); - journal_lock_updates(journal); - err = journal_flush(journal); - journal_unlock_updates(journal); + jbd2_journal_lock_updates(journal); + err = jbd2_journal_flush(journal); + jbd2_journal_unlock_updates(journal); if (err) return 0; @@ -1356,7 +1356,7 @@ static int bput_one(handle_t *handle, struct buffer_head *bh) return 0; } -static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) +static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) { if (buffer_mapped(bh)) return ext4_journal_dirty_data(handle, bh); @@ -1464,7 +1464,7 @@ static int ext4_ordered_writepage(struct page *page, */ if (ret == 0) { err = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, - NULL, journal_dirty_data_fn); + NULL, jbd2_journal_dirty_data_fn); if (!ret) ret = err; } @@ -1595,7 +1595,7 @@ static void ext4_invalidatepage(struct page *page, unsigned long offset) if (offset == 0) ClearPageChecked(page); - journal_invalidatepage(journal, page, offset); + jbd2_journal_invalidatepage(journal, page, offset); } static int ext4_releasepage(struct page *page, gfp_t wait) @@ -1605,7 +1605,7 @@ static int ext4_releasepage(struct page *page, gfp_t wait) WARN_ON(PageChecked(page)); if (!page_has_buffers(page)) return 0; - return journal_try_to_free_buffers(journal, page, wait); + return jbd2_journal_try_to_free_buffers(journal, page, wait); } /* @@ -1982,11 +1982,11 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode, /* * Any buffers which are on the journal will be in memory. We find - * them on the hash table so journal_revoke() will run journal_forget() + * them on the hash table so jbd2_journal_revoke() will run jbd2_journal_forget() * on them. We've already detached each block from the file, so - * bforget() in journal_forget() should be safe. + * bforget() in jbd2_journal_forget() should be safe. * - * AKPM: turn on bforget in journal_forget()!!! + * AKPM: turn on bforget in jbd2_journal_forget()!!! */ for (p = first; p < last; p++) { u32 nr = le32_to_cpu(*p); @@ -2132,11 +2132,11 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, * We've probably journalled the indirect block several * times during the truncate. But it's no longer * needed and we now drop it from the transaction via - * journal_revoke(). + * jbd2_journal_revoke(). * * That's easy if it's exclusively part of this * transaction. But if it's part of the committing - * transaction then journal_forget() will simply + * transaction then jbd2_journal_forget() will simply * brelse() it. That means that if the underlying * block is reallocated in ext4_get_block(), * unmap_underlying_metadata() will find this block @@ -2251,7 +2251,7 @@ void ext4_truncate(struct inode *inode) /* * We have to lock the EOF page here, because lock_page() nests - * outside journal_start(). + * outside jbd2_journal_start(). */ if ((inode->i_size & (blocksize - 1)) == 0) { /* Block boundary? Nothing to do */ @@ -3035,7 +3035,7 @@ int ext4_mark_iloc_dirty(handle_t *handle, /* the do_update_inode consumes one bh->b_count */ get_bh(iloc->bh); - /* ext4_do_update_inode() does journal_dirty_metadata */ + /* ext4_do_update_inode() does jbd2_journal_dirty_metadata */ err = ext4_do_update_inode(handle, inode, iloc); put_bh(iloc->bh); return err; @@ -3153,7 +3153,7 @@ static int ext4_pin_inode(handle_t *handle, struct inode *inode) err = ext4_get_inode_loc(inode, &iloc); if (!err) { BUFFER_TRACE(iloc.bh, "get_write_access"); - err = journal_get_write_access(handle, iloc.bh); + err = jbd2_journal_get_write_access(handle, iloc.bh); if (!err) err = ext4_journal_dirty_metadata(handle, iloc.bh); @@ -3185,8 +3185,8 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) if (is_journal_aborted(journal) || IS_RDONLY(inode)) return -EROFS; - journal_lock_updates(journal); - journal_flush(journal); + jbd2_journal_lock_updates(journal); + jbd2_journal_flush(journal); /* * OK, there are no updates running now, and all cached data is @@ -3202,7 +3202,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) EXT4_I(inode)->i_flags &= ~EXT4_JOURNAL_DATA_FL; ext4_set_aops(inode); - journal_unlock_updates(journal); + jbd2_journal_unlock_updates(journal); /* Finally we can mark the inode as dirty. */ diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index a567af1..a63dce2 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -8,10 +8,10 @@ */ #include -#include +#include #include #include -#include +#include #include #include #include @@ -219,9 +219,9 @@ flags_err: return -EFAULT; err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count); - journal_lock_updates(EXT4_SB(sb)->s_journal); - journal_flush(EXT4_SB(sb)->s_journal); - journal_unlock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_flush(EXT4_SB(sb)->s_journal); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); return err; } @@ -241,9 +241,9 @@ flags_err: return -EFAULT; err = ext4_group_add(sb, &input); - journal_lock_updates(EXT4_SB(sb)->s_journal); - journal_flush(EXT4_SB(sb)->s_journal); - journal_unlock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_flush(EXT4_SB(sb)->s_journal); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); return err; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 956b381..f98b999 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -26,10 +26,10 @@ #include #include -#include +#include #include #include -#include +#include #include #include #include diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 4a47895..5b2828d 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -13,7 +13,7 @@ #include #include -#include +#include #include #include diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9e32a2a..f131bb6 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -20,9 +20,9 @@ #include #include #include -#include +#include #include -#include +#include #include #include #include @@ -63,7 +63,7 @@ static void ext4_write_super (struct super_block * sb); static void ext4_write_super_lockfs(struct super_block *sb); /* - * Wrappers for journal_start/end. + * Wrappers for jbd2_journal_start/end. * * The only special thing we need to do here is to make sure that all * journal_end calls result in the superblock being marked dirty, so @@ -87,12 +87,12 @@ handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks) return ERR_PTR(-EROFS); } - return journal_start(journal, nblocks); + return jbd2_journal_start(journal, nblocks); } /* * The only special thing we need to do here is to make sure that all - * journal_stop calls result in the superblock being marked dirty, so + * jbd2_journal_stop calls result in the superblock being marked dirty, so * that sync() will call the filesystem's write_super callback if * appropriate. */ @@ -104,7 +104,7 @@ int __ext4_journal_stop(const char *where, handle_t *handle) sb = handle->h_transaction->t_journal->j_private; err = handle->h_err; - rc = journal_stop(handle); + rc = jbd2_journal_stop(handle); if (!err) err = rc; @@ -131,7 +131,7 @@ void ext4_journal_abort_handle(const char *caller, const char *err_fn, printk(KERN_ERR "%s: aborting transaction: %s in %s\n", caller, errstr, err_fn); - journal_abort_handle(handle); + jbd2_journal_abort_handle(handle); } /* Deal with the reporting of failure conditions on a filesystem such as @@ -144,7 +144,7 @@ void ext4_journal_abort_handle(const char *caller, const char *err_fn, * be aborted, we can't rely on the current, or future, transactions to * write out the superblock safely. * - * We'll just use the journal_abort() error code to record an error in + * We'll just use the jbd2_journal_abort() error code to record an error in * the journal instead. On recovery, the journal will compain about * that error until we've noted it down and cleared it. */ @@ -164,7 +164,7 @@ static void ext4_handle_error(struct super_block *sb) EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; if (journal) - journal_abort(journal, -EIO); + jbd2_journal_abort(journal, -EIO); } if (test_opt (sb, ERRORS_RO)) { printk (KERN_CRIT "Remounting filesystem read-only\n"); @@ -203,7 +203,7 @@ static const char *ext4_decode_error(struct super_block * sb, int errno, errstr = "Out of memory"; break; case -EROFS: - if (!sb || EXT4_SB(sb)->s_journal->j_flags & JFS_ABORT) + if (!sb || EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT) errstr = "Journal has aborted"; else errstr = "Readonly filesystem"; @@ -279,7 +279,7 @@ void ext4_abort (struct super_block * sb, const char * function, EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; sb->s_flags |= MS_RDONLY; EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; - journal_abort(EXT4_SB(sb)->s_journal, -EIO); + jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO); } void ext4_warning (struct super_block * sb, const char * function, @@ -391,7 +391,7 @@ static void ext4_put_super (struct super_block * sb) int i; ext4_xattr_put_super(sb); - journal_destroy(sbi->s_journal); + jbd2_journal_destroy(sbi->s_journal); if (!(sb->s_flags & MS_RDONLY)) { EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); es->s_state = cpu_to_le16(sbi->s_mount_state); @@ -1722,8 +1722,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) /* No mode set, assume a default based on the journal capabilities: ORDERED_DATA if the journal can cope, else JOURNAL_DATA */ - if (journal_check_available_features - (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) + if (jbd2_journal_check_available_features + (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) set_opt(sbi->s_mount_opt, ORDERED_DATA); else set_opt(sbi->s_mount_opt, JOURNAL_DATA); @@ -1731,8 +1731,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) case EXT4_MOUNT_ORDERED_DATA: case EXT4_MOUNT_WRITEBACK_DATA: - if (!journal_check_available_features - (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) { + if (!jbd2_journal_check_available_features + (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) { printk(KERN_ERR "EXT4-fs: Journal does not support " "requested data journaling mode\n"); goto failed_mount4; @@ -1749,7 +1749,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) } } /* - * The journal_load will have done any necessary log recovery, + * The jbd2_journal_load will have done any necessary log recovery, * so we can safely mount the rest of the filesystem now. */ @@ -1797,7 +1797,7 @@ cantfind_ext4: goto failed_mount; failed_mount4: - journal_destroy(sbi->s_journal); + jbd2_journal_destroy(sbi->s_journal); failed_mount3: percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); @@ -1837,9 +1837,9 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) spin_lock(&journal->j_state_lock); if (test_opt(sb, BARRIER)) - journal->j_flags |= JFS_BARRIER; + journal->j_flags |= JBD2_BARRIER; else - journal->j_flags &= ~JFS_BARRIER; + journal->j_flags &= ~JBD2_BARRIER; spin_unlock(&journal->j_state_lock); } @@ -1873,7 +1873,7 @@ static journal_t *ext4_get_journal(struct super_block *sb, return NULL; } - journal = journal_init_inode(journal_inode); + journal = jbd2_journal_init_inode(journal_inode); if (!journal) { printk(KERN_ERR "EXT4-fs: Could not load journal inode\n"); iput(journal_inode); @@ -1945,7 +1945,7 @@ static journal_t *ext4_get_dev_journal(struct super_block *sb, start = sb_block + 1; brelse(bh); /* we're done with the superblock */ - journal = journal_init_dev(bdev, sb->s_bdev, + journal = jbd2_journal_init_dev(bdev, sb->s_bdev, start, len, blocksize); if (!journal) { printk(KERN_ERR "EXT4-fs: failed to create device journal\n"); @@ -1968,7 +1968,7 @@ static journal_t *ext4_get_dev_journal(struct super_block *sb, ext4_init_journal_params(sb, journal); return journal; out_journal: - journal_destroy(journal); + jbd2_journal_destroy(journal); out_bdev: ext4_blkdev_put(bdev); return NULL; @@ -2029,22 +2029,22 @@ static int ext4_load_journal(struct super_block *sb, } if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { - err = journal_update_format(journal); + err = jbd2_journal_update_format(journal); if (err) { printk(KERN_ERR "EXT4-fs: error updating journal.\n"); - journal_destroy(journal); + jbd2_journal_destroy(journal); return err; } } if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) - err = journal_wipe(journal, !really_read_only); + err = jbd2_journal_wipe(journal, !really_read_only); if (!err) - err = journal_load(journal); + err = jbd2_journal_load(journal); if (err) { printk(KERN_ERR "EXT4-fs: error loading journal.\n"); - journal_destroy(journal); + jbd2_journal_destroy(journal); return err; } @@ -2081,9 +2081,9 @@ static int ext4_create_journal(struct super_block * sb, printk(KERN_INFO "EXT4-fs: creating new journal on inode %u\n", journal_inum); - if (journal_create(journal)) { + if (jbd2_journal_create(journal)) { printk(KERN_ERR "EXT4-fs: error creating journal.\n"); - journal_destroy(journal); + jbd2_journal_destroy(journal); return -EIO; } @@ -2130,15 +2130,15 @@ static void ext4_mark_recovery_complete(struct super_block * sb, { journal_t *journal = EXT4_SB(sb)->s_journal; - journal_lock_updates(journal); - journal_flush(journal); + jbd2_journal_lock_updates(journal); + jbd2_journal_flush(journal); if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) && sb->s_flags & MS_RDONLY) { EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); sb->s_dirt = 0; ext4_commit_super(sb, es, 1); } - journal_unlock_updates(journal); + jbd2_journal_unlock_updates(journal); } /* @@ -2160,7 +2160,7 @@ static void ext4_clear_journal_err(struct super_block * sb, * journal by a prior ext4_error() or ext4_abort() */ - j_errno = journal_errno(journal); + j_errno = jbd2_journal_errno(journal); if (j_errno) { char nbuf[16]; @@ -2174,7 +2174,7 @@ static void ext4_clear_journal_err(struct super_block * sb, es->s_state |= cpu_to_le16(EXT4_ERROR_FS); ext4_commit_super (sb, es, 1); - journal_clear_err(journal); + jbd2_journal_clear_err(journal); } } @@ -2217,9 +2217,9 @@ static int ext4_sync_fs(struct super_block *sb, int wait) tid_t target; sb->s_dirt = 0; - if (journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { + if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { if (wait) - log_wait_commit(EXT4_SB(sb)->s_journal, target); + jbd2_log_wait_commit(EXT4_SB(sb)->s_journal, target); } return 0; } @@ -2236,8 +2236,8 @@ static void ext4_write_super_lockfs(struct super_block *sb) journal_t *journal = EXT4_SB(sb)->s_journal; /* Now we set up the journal barrier. */ - journal_lock_updates(journal); - journal_flush(journal); + jbd2_journal_lock_updates(journal); + jbd2_journal_flush(journal); /* Journal blocked and flushed, clear needs_recovery flag. */ EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); @@ -2257,7 +2257,7 @@ static void ext4_unlockfs(struct super_block *sb) EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1); unlock_super(sb); - journal_unlock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); } } @@ -2438,9 +2438,9 @@ static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) * is locked for write. Otherwise the are possible deadlocks: * Process 1 Process 2 * ext4_create() quota_sync() - * journal_start() write_dquot() + * jbd2_journal_start() write_dquot() * DQUOT_INIT() down(dqio_mutex) - * down(dqio_mutex) journal_start() + * down(dqio_mutex) jbd2_journal_start() * */ diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c index 9e4c75f..fcf5272 100644 --- a/fs/ext4/symlink.c +++ b/fs/ext4/symlink.c @@ -18,7 +18,7 @@ */ #include -#include +#include #include #include #include "xattr.h" diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index d3a4081..90f7d5c 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -53,7 +53,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c index d84b1da..b6a6861 100644 --- a/fs/ext4/xattr_security.c +++ b/fs/ext4/xattr_security.c @@ -7,7 +7,7 @@ #include #include #include -#include +#include #include #include #include "xattr.h" diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c index 11bd58c..b76f2db 100644 --- a/fs/ext4/xattr_trusted.c +++ b/fs/ext4/xattr_trusted.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include #include "xattr.h" diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c index 9c5a665..c53cded 100644 --- a/fs/ext4/xattr_user.c +++ b/fs/ext4/xattr_user.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #include #include "xattr.h" -- cgit v0.10.2 From c3fcc8137ce4296ad6ab94f88bd60cbe03d21527 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Wed, 11 Oct 2006 01:21:02 -0700 Subject: [PATCH] jbd2: cleanup ext4_jbd.h To allow ext4 to build during the transition from jbd to jbd2, we have both ext4_jbd.h and ext4_jbd2.h in the tree. We no longer need the former. Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/ext4_jbd.h b/include/linux/ext4_jbd.h deleted file mode 100644 index 3dbf6c7..0000000 --- a/include/linux/ext4_jbd.h +++ /dev/null @@ -1,268 +0,0 @@ -/* - * linux/include/linux/ext4_jbd.h - * - * Written by Stephen C. Tweedie , 1999 - * - * Copyright 1998--1999 Red Hat corp --- All Rights Reserved - * - * This file is part of the Linux kernel and is made available under - * the terms of the GNU General Public License, version 2, or at your - * option, any later version, incorporated herein by reference. - * - * Ext4-specific journaling extensions. - */ - -#ifndef _LINUX_EXT4_JBD_H -#define _LINUX_EXT4_JBD_H - -#include -#include -#include - -#define EXT4_JOURNAL(inode) (EXT4_SB((inode)->i_sb)->s_journal) - -/* Define the number of blocks we need to account to a transaction to - * modify one block of data. - * - * We may have to touch one inode, one bitmap buffer, up to three - * indirection blocks, the group and superblock summaries, and the data - * block to complete the transaction. */ - -#define EXT4_SINGLEDATA_TRANS_BLOCKS 8U - -/* Extended attribute operations touch at most two data buffers, - * two bitmap buffers, and two group summaries, in addition to the inode - * and the superblock, which are already accounted for. */ - -#define EXT4_XATTR_TRANS_BLOCKS 6U - -/* Define the minimum size for a transaction which modifies data. This - * needs to take into account the fact that we may end up modifying two - * quota files too (one for the group, one for the user quota). The - * superblock only gets updated once, of course, so don't bother - * counting that again for the quota updates. */ - -#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS + \ - EXT4_XATTR_TRANS_BLOCKS - 2 + \ - 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) - -/* Delete operations potentially hit one directory's namespace plus an - * entire inode, plus arbitrary amounts of bitmap/indirection data. Be - * generous. We can grow the delete transaction later if necessary. */ - -#define EXT4_DELETE_TRANS_BLOCKS(sb) (2 * EXT4_DATA_TRANS_BLOCKS(sb) + 64) - -/* Define an arbitrary limit for the amount of data we will anticipate - * writing to any given transaction. For unbounded transactions such as - * write(2) and truncate(2) we can write more than this, but we always - * start off at the maximum transaction size and grow the transaction - * optimistically as we go. */ - -#define EXT4_MAX_TRANS_DATA 64U - -/* We break up a large truncate or write transaction once the handle's - * buffer credits gets this low, we need either to extend the - * transaction or to start a new one. Reserve enough space here for - * inode, bitmap, superblock, group and indirection updates for at least - * one block, plus two quota updates. Quota allocations are not - * needed. */ - -#define EXT4_RESERVE_TRANS_BLOCKS 12U - -#define EXT4_INDEX_EXTRA_TRANS_BLOCKS 8 - -#ifdef CONFIG_QUOTA -/* Amount of blocks needed for quota update - we know that the structure was - * allocated so we need to update only inode+data */ -#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) -/* Amount of blocks needed for quota insert/delete - we do some block writes - * but inode, sb and group updates are done only once */ -#define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ - (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) -#define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ - (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) -#else -#define EXT4_QUOTA_TRANS_BLOCKS(sb) 0 -#define EXT4_QUOTA_INIT_BLOCKS(sb) 0 -#define EXT4_QUOTA_DEL_BLOCKS(sb) 0 -#endif - -int -ext4_mark_iloc_dirty(handle_t *handle, - struct inode *inode, - struct ext4_iloc *iloc); - -/* - * On success, We end up with an outstanding reference count against - * iloc->bh. This _must_ be cleaned up later. - */ - -int ext4_reserve_inode_write(handle_t *handle, struct inode *inode, - struct ext4_iloc *iloc); - -int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode); - -/* - * Wrapper functions with which ext4 calls into JBD. The intent here is - * to allow these to be turned into appropriate stubs so ext4 can control - * ext2 filesystems, so ext2+ext4 systems only nee one fs. This work hasn't - * been done yet. - */ - -void ext4_journal_abort_handle(const char *caller, const char *err_fn, - struct buffer_head *bh, handle_t *handle, int err); - -static inline int -__ext4_journal_get_undo_access(const char *where, handle_t *handle, - struct buffer_head *bh) -{ - int err = journal_get_undo_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} - -static inline int -__ext4_journal_get_write_access(const char *where, handle_t *handle, - struct buffer_head *bh) -{ - int err = journal_get_write_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} - -static inline void -ext4_journal_release_buffer(handle_t *handle, struct buffer_head *bh) -{ - journal_release_buffer(handle, bh); -} - -static inline int -__ext4_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) -{ - int err = journal_forget(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} - -static inline int -__ext4_journal_revoke(const char *where, handle_t *handle, - unsigned long blocknr, struct buffer_head *bh) -{ - int err = journal_revoke(handle, blocknr, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} - -static inline int -__ext4_journal_get_create_access(const char *where, - handle_t *handle, struct buffer_head *bh) -{ - int err = journal_get_create_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} - -static inline int -__ext4_journal_dirty_metadata(const char *where, - handle_t *handle, struct buffer_head *bh) -{ - int err = journal_dirty_metadata(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} - - -#define ext4_journal_get_undo_access(handle, bh) \ - __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) -#define ext4_journal_get_write_access(handle, bh) \ - __ext4_journal_get_write_access(__FUNCTION__, (handle), (bh)) -#define ext4_journal_revoke(handle, blocknr, bh) \ - __ext4_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) -#define ext4_journal_get_create_access(handle, bh) \ - __ext4_journal_get_create_access(__FUNCTION__, (handle), (bh)) -#define ext4_journal_dirty_metadata(handle, bh) \ - __ext4_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) -#define ext4_journal_forget(handle, bh) \ - __ext4_journal_forget(__FUNCTION__, (handle), (bh)) - -int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh); - -handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); -int __ext4_journal_stop(const char *where, handle_t *handle); - -static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks) -{ - return ext4_journal_start_sb(inode->i_sb, nblocks); -} - -#define ext4_journal_stop(handle) \ - __ext4_journal_stop(__FUNCTION__, (handle)) - -static inline handle_t *ext4_journal_current_handle(void) -{ - return journal_current_handle(); -} - -static inline int ext4_journal_extend(handle_t *handle, int nblocks) -{ - return journal_extend(handle, nblocks); -} - -static inline int ext4_journal_restart(handle_t *handle, int nblocks) -{ - return journal_restart(handle, nblocks); -} - -static inline int ext4_journal_blocks_per_page(struct inode *inode) -{ - return journal_blocks_per_page(inode); -} - -static inline int ext4_journal_force_commit(journal_t *journal) -{ - return journal_force_commit(journal); -} - -/* super.c */ -int ext4_force_commit(struct super_block *sb); - -static inline int ext4_should_journal_data(struct inode *inode) -{ - if (!S_ISREG(inode->i_mode)) - return 1; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) - return 1; - if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) - return 1; - return 0; -} - -static inline int ext4_should_order_data(struct inode *inode) -{ - if (!S_ISREG(inode->i_mode)) - return 0; - if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) - return 0; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) - return 1; - return 0; -} - -static inline int ext4_should_writeback_data(struct inode *inode) -{ - if (!S_ISREG(inode->i_mode)) - return 0; - if (EXT4_I(inode)->i_flags & EXT4_JOURNAL_DATA_FL) - return 0; - if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) - return 1; - return 0; -} - -#endif /* _LINUX_EXT4_JBD_H */ -- cgit v0.10.2 From a86c61812637c7dd0c57e29880cffd477b62f2e7 Mon Sep 17 00:00:00 2001 From: Alex Tomas Date: Wed, 11 Oct 2006 01:21:03 -0700 Subject: [PATCH] ext3: add extent map support On disk extents format: /* * this is extent on-disk structure * it's used at the bottom of the tree */ struct ext3_extent { __le32 ee_block; /* first logical block extent covers */ __le16 ee_len; /* number of blocks covered by extent */ __le16 ee_start_hi; /* high 16 bits of physical block */ __le32 ee_start; /* low 32 bigs of physical block */ }; Signed-off-by: Alex Tomas Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index 09c4878..a6acb96 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -5,7 +5,7 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ - ioctl.o namei.o super.o symlink.o hash.o resize.o + ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 9833d5d..18ac173a 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -134,8 +134,7 @@ static int ext4_readdir(struct file * filp, struct buffer_head *bh = NULL; map_bh.b_state = 0; - err = ext4_get_blocks_handle(NULL, inode, blk, 1, - &map_bh, 0, 0); + err = ext4_get_blocks_wrap(NULL, inode, blk, 1, &map_bh, 0, 0); if (err > 0) { page_cache_readahead(sb->s_bdev->bd_inode->i_mapping, &filp->f_ra, diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c new file mode 100644 index 0000000..f67b2ef --- /dev/null +++ b/fs/ext4/extents.c @@ -0,0 +1,2075 @@ +/* + * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com + * Written by Alex Tomas + * + * Architecture independence: + * Copyright (c) 2005, Bull S.A. + * Written by Pierre Peiffer + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + */ + +/* + * Extents support for EXT4 + * + * TODO: + * - ext4*_error() should be used in some situations + * - analyze all BUG()/BUG_ON(), use -EIO where appropriate + * - smart tree reduction + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +static int ext4_ext_check_header(const char *function, struct inode *inode, + struct ext4_extent_header *eh) +{ + const char *error_msg = NULL; + + if (unlikely(eh->eh_magic != EXT4_EXT_MAGIC)) { + error_msg = "invalid magic"; + goto corrupted; + } + if (unlikely(eh->eh_max == 0)) { + error_msg = "invalid eh_max"; + goto corrupted; + } + if (unlikely(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max))) { + error_msg = "invalid eh_entries"; + goto corrupted; + } + return 0; + +corrupted: + ext4_error(inode->i_sb, function, + "bad header in inode #%lu: %s - magic %x, " + "entries %u, max %u, depth %u", + inode->i_ino, error_msg, le16_to_cpu(eh->eh_magic), + le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max), + le16_to_cpu(eh->eh_depth)); + + return -EIO; +} + +static handle_t *ext4_ext_journal_restart(handle_t *handle, int needed) +{ + int err; + + if (handle->h_buffer_credits > needed) + return handle; + if (!ext4_journal_extend(handle, needed)) + return handle; + err = ext4_journal_restart(handle, needed); + + return handle; +} + +/* + * could return: + * - EROFS + * - ENOMEM + */ +static int ext4_ext_get_access(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + if (path->p_bh) { + /* path points to block */ + return ext4_journal_get_write_access(handle, path->p_bh); + } + /* path points to leaf/index in inode body */ + /* we use in-core data, no need to protect them */ + return 0; +} + +/* + * could return: + * - EROFS + * - ENOMEM + * - EIO + */ +static int ext4_ext_dirty(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + int err; + if (path->p_bh) { + /* path points to block */ + err = ext4_journal_dirty_metadata(handle, path->p_bh); + } else { + /* path points to leaf/index in inode body */ + err = ext4_mark_inode_dirty(handle, inode); + } + return err; +} + +static int ext4_ext_find_goal(struct inode *inode, + struct ext4_ext_path *path, + unsigned long block) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + unsigned long bg_start; + unsigned long colour; + int depth; + + if (path) { + struct ext4_extent *ex; + depth = path->p_depth; + + /* try to predict block placement */ + if ((ex = path[depth].p_ext)) + return le32_to_cpu(ex->ee_start) + + (block - le32_to_cpu(ex->ee_block)); + + /* it looks index is empty + * try to find starting from index itself */ + if (path[depth].p_bh) + return path[depth].p_bh->b_blocknr; + } + + /* OK. use inode's group */ + bg_start = (ei->i_block_group * EXT4_BLOCKS_PER_GROUP(inode->i_sb)) + + le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_first_data_block); + colour = (current->pid % 16) * + (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); + return bg_start + colour + block; +} + +static int +ext4_ext_new_block(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *ex, int *err) +{ + int goal, newblock; + + goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block)); + newblock = ext4_new_block(handle, inode, goal, err); + return newblock; +} + +static inline int ext4_ext_space_block(struct inode *inode) +{ + int size; + + size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) + / sizeof(struct ext4_extent); +#ifdef AGRESSIVE_TEST + if (size > 6) + size = 6; +#endif + return size; +} + +static inline int ext4_ext_space_block_idx(struct inode *inode) +{ + int size; + + size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) + / sizeof(struct ext4_extent_idx); +#ifdef AGRESSIVE_TEST + if (size > 5) + size = 5; +#endif + return size; +} + +static inline int ext4_ext_space_root(struct inode *inode) +{ + int size; + + size = sizeof(EXT4_I(inode)->i_data); + size -= sizeof(struct ext4_extent_header); + size /= sizeof(struct ext4_extent); +#ifdef AGRESSIVE_TEST + if (size > 3) + size = 3; +#endif + return size; +} + +static inline int ext4_ext_space_root_idx(struct inode *inode) +{ + int size; + + size = sizeof(EXT4_I(inode)->i_data); + size -= sizeof(struct ext4_extent_header); + size /= sizeof(struct ext4_extent_idx); +#ifdef AGRESSIVE_TEST + if (size > 4) + size = 4; +#endif + return size; +} + +#ifdef EXT_DEBUG +static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) +{ + int k, l = path->p_depth; + + ext_debug("path:"); + for (k = 0; k <= l; k++, path++) { + if (path->p_idx) { + ext_debug(" %d->%d", le32_to_cpu(path->p_idx->ei_block), + le32_to_cpu(path->p_idx->ei_leaf)); + } else if (path->p_ext) { + ext_debug(" %d:%d:%d", + le32_to_cpu(path->p_ext->ee_block), + le16_to_cpu(path->p_ext->ee_len), + le32_to_cpu(path->p_ext->ee_start)); + } else + ext_debug(" []"); + } + ext_debug("\n"); +} + +static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) +{ + int depth = ext_depth(inode); + struct ext4_extent_header *eh; + struct ext4_extent *ex; + int i; + + if (!path) + return; + + eh = path[depth].p_hdr; + ex = EXT_FIRST_EXTENT(eh); + + for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) { + ext_debug("%d:%d:%d ", le32_to_cpu(ex->ee_block), + le16_to_cpu(ex->ee_len), + le32_to_cpu(ex->ee_start)); + } + ext_debug("\n"); +} +#else +#define ext4_ext_show_path(inode,path) +#define ext4_ext_show_leaf(inode,path) +#endif + +static void ext4_ext_drop_refs(struct ext4_ext_path *path) +{ + int depth = path->p_depth; + int i; + + for (i = 0; i <= depth; i++, path++) + if (path->p_bh) { + brelse(path->p_bh); + path->p_bh = NULL; + } +} + +/* + * binary search for closest index by given block + */ +static void +ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int block) +{ + struct ext4_extent_header *eh = path->p_hdr; + struct ext4_extent_idx *r, *l, *m; + + BUG_ON(eh->eh_magic != EXT4_EXT_MAGIC); + BUG_ON(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max)); + BUG_ON(le16_to_cpu(eh->eh_entries) <= 0); + + ext_debug("binsearch for %d(idx): ", block); + + l = EXT_FIRST_INDEX(eh) + 1; + r = EXT_FIRST_INDEX(eh) + le16_to_cpu(eh->eh_entries) - 1; + while (l <= r) { + m = l + (r - l) / 2; + if (block < le32_to_cpu(m->ei_block)) + r = m - 1; + else + l = m + 1; + ext_debug("%p(%u):%p(%u):%p(%u) ", l, l->ei_block, + m, m->ei_block, r, r->ei_block); + } + + path->p_idx = l - 1; + ext_debug(" -> %d->%d ", le32_to_cpu(path->p_idx->ei_block), + le32_to_cpu(path->p_idx->ei_leaf)); + +#ifdef CHECK_BINSEARCH + { + struct ext4_extent_idx *chix, *ix; + int k; + + chix = ix = EXT_FIRST_INDEX(eh); + for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) { + if (k != 0 && + le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) { + printk("k=%d, ix=0x%p, first=0x%p\n", k, + ix, EXT_FIRST_INDEX(eh)); + printk("%u <= %u\n", + le32_to_cpu(ix->ei_block), + le32_to_cpu(ix[-1].ei_block)); + } + BUG_ON(k && le32_to_cpu(ix->ei_block) + <= le32_to_cpu(ix[-1].ei_block)); + if (block < le32_to_cpu(ix->ei_block)) + break; + chix = ix; + } + BUG_ON(chix != path->p_idx); + } +#endif + +} + +/* + * binary search for closest extent by given block + */ +static void +ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) +{ + struct ext4_extent_header *eh = path->p_hdr; + struct ext4_extent *r, *l, *m; + + BUG_ON(eh->eh_magic != EXT4_EXT_MAGIC); + BUG_ON(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max)); + + if (eh->eh_entries == 0) { + /* + * this leaf is empty yet: + * we get such a leaf in split/add case + */ + return; + } + + ext_debug("binsearch for %d: ", block); + + l = EXT_FIRST_EXTENT(eh) + 1; + r = EXT_FIRST_EXTENT(eh) + le16_to_cpu(eh->eh_entries) - 1; + + while (l <= r) { + m = l + (r - l) / 2; + if (block < le32_to_cpu(m->ee_block)) + r = m - 1; + else + l = m + 1; + ext_debug("%p(%u):%p(%u):%p(%u) ", l, l->ee_block, + m, m->ee_block, r, r->ee_block); + } + + path->p_ext = l - 1; + ext_debug(" -> %d:%d:%d ", + le32_to_cpu(path->p_ext->ee_block), + le32_to_cpu(path->p_ext->ee_start), + le16_to_cpu(path->p_ext->ee_len)); + +#ifdef CHECK_BINSEARCH + { + struct ext4_extent *chex, *ex; + int k; + + chex = ex = EXT_FIRST_EXTENT(eh); + for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ex++) { + BUG_ON(k && le32_to_cpu(ex->ee_block) + <= le32_to_cpu(ex[-1].ee_block)); + if (block < le32_to_cpu(ex->ee_block)) + break; + chex = ex; + } + BUG_ON(chex != path->p_ext); + } +#endif + +} + +int ext4_ext_tree_init(handle_t *handle, struct inode *inode) +{ + struct ext4_extent_header *eh; + + eh = ext_inode_hdr(inode); + eh->eh_depth = 0; + eh->eh_entries = 0; + eh->eh_magic = EXT4_EXT_MAGIC; + eh->eh_max = cpu_to_le16(ext4_ext_space_root(inode)); + ext4_mark_inode_dirty(handle, inode); + ext4_ext_invalidate_cache(inode); + return 0; +} + +struct ext4_ext_path * +ext4_ext_find_extent(struct inode *inode, int block, struct ext4_ext_path *path) +{ + struct ext4_extent_header *eh; + struct buffer_head *bh; + short int depth, i, ppos = 0, alloc = 0; + + eh = ext_inode_hdr(inode); + BUG_ON(eh == NULL); + if (ext4_ext_check_header(__FUNCTION__, inode, eh)) + return ERR_PTR(-EIO); + + i = depth = ext_depth(inode); + + /* account possible depth increase */ + if (!path) { + path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 2), + GFP_NOFS); + if (!path) + return ERR_PTR(-ENOMEM); + alloc = 1; + } + memset(path, 0, sizeof(struct ext4_ext_path) * (depth + 1)); + path[0].p_hdr = eh; + + /* walk through the tree */ + while (i) { + ext_debug("depth %d: num %d, max %d\n", + ppos, le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); + ext4_ext_binsearch_idx(inode, path + ppos, block); + path[ppos].p_block = le32_to_cpu(path[ppos].p_idx->ei_leaf); + path[ppos].p_depth = i; + path[ppos].p_ext = NULL; + + bh = sb_bread(inode->i_sb, path[ppos].p_block); + if (!bh) + goto err; + + eh = ext_block_hdr(bh); + ppos++; + BUG_ON(ppos > depth); + path[ppos].p_bh = bh; + path[ppos].p_hdr = eh; + i--; + + if (ext4_ext_check_header(__FUNCTION__, inode, eh)) + goto err; + } + + path[ppos].p_depth = i; + path[ppos].p_hdr = eh; + path[ppos].p_ext = NULL; + path[ppos].p_idx = NULL; + + if (ext4_ext_check_header(__FUNCTION__, inode, eh)) + goto err; + + /* find extent */ + ext4_ext_binsearch(inode, path + ppos, block); + + ext4_ext_show_path(inode, path); + + return path; + +err: + ext4_ext_drop_refs(path); + if (alloc) + kfree(path); + return ERR_PTR(-EIO); +} + +/* + * insert new index [logical;ptr] into the block at cupr + * it check where to insert: before curp or after curp + */ +static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, + struct ext4_ext_path *curp, + int logical, int ptr) +{ + struct ext4_extent_idx *ix; + int len, err; + + if ((err = ext4_ext_get_access(handle, inode, curp))) + return err; + + BUG_ON(logical == le32_to_cpu(curp->p_idx->ei_block)); + len = EXT_MAX_INDEX(curp->p_hdr) - curp->p_idx; + if (logical > le32_to_cpu(curp->p_idx->ei_block)) { + /* insert after */ + if (curp->p_idx != EXT_LAST_INDEX(curp->p_hdr)) { + len = (len - 1) * sizeof(struct ext4_extent_idx); + len = len < 0 ? 0 : len; + ext_debug("insert new index %d after: %d. " + "move %d from 0x%p to 0x%p\n", + logical, ptr, len, + (curp->p_idx + 1), (curp->p_idx + 2)); + memmove(curp->p_idx + 2, curp->p_idx + 1, len); + } + ix = curp->p_idx + 1; + } else { + /* insert before */ + len = len * sizeof(struct ext4_extent_idx); + len = len < 0 ? 0 : len; + ext_debug("insert new index %d before: %d. " + "move %d from 0x%p to 0x%p\n", + logical, ptr, len, + curp->p_idx, (curp->p_idx + 1)); + memmove(curp->p_idx + 1, curp->p_idx, len); + ix = curp->p_idx; + } + + ix->ei_block = cpu_to_le32(logical); + ix->ei_leaf = cpu_to_le32(ptr); + curp->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(curp->p_hdr->eh_entries)+1); + + BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries) + > le16_to_cpu(curp->p_hdr->eh_max)); + BUG_ON(ix > EXT_LAST_INDEX(curp->p_hdr)); + + err = ext4_ext_dirty(handle, inode, curp); + ext4_std_error(inode->i_sb, err); + + return err; +} + +/* + * routine inserts new subtree into the path, using free index entry + * at depth 'at: + * - allocates all needed blocks (new leaf and all intermediate index blocks) + * - makes decision where to split + * - moves remaining extens and index entries (right to the split point) + * into the newly allocated blocks + * - initialize subtree + */ +static int ext4_ext_split(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext, int at) +{ + struct buffer_head *bh = NULL; + int depth = ext_depth(inode); + struct ext4_extent_header *neh; + struct ext4_extent_idx *fidx; + struct ext4_extent *ex; + int i = at, k, m, a; + unsigned long newblock, oldblock; + __le32 border; + int *ablocks = NULL; /* array of allocated blocks */ + int err = 0; + + /* make decision: where to split? */ + /* FIXME: now desicion is simplest: at current extent */ + + /* if current leaf will be splitted, then we should use + * border from split point */ + BUG_ON(path[depth].p_ext > EXT_MAX_EXTENT(path[depth].p_hdr)); + if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) { + border = path[depth].p_ext[1].ee_block; + ext_debug("leaf will be splitted." + " next leaf starts at %d\n", + le32_to_cpu(border)); + } else { + border = newext->ee_block; + ext_debug("leaf will be added." + " next leaf starts at %d\n", + le32_to_cpu(border)); + } + + /* + * if error occurs, then we break processing + * and turn filesystem read-only. so, index won't + * be inserted and tree will be in consistent + * state. next mount will repair buffers too + */ + + /* + * get array to track all allocated blocks + * we need this to handle errors and free blocks + * upon them + */ + ablocks = kmalloc(sizeof(unsigned long) * depth, GFP_NOFS); + if (!ablocks) + return -ENOMEM; + memset(ablocks, 0, sizeof(unsigned long) * depth); + + /* allocate all needed blocks */ + ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); + for (a = 0; a < depth - at; a++) { + newblock = ext4_ext_new_block(handle, inode, path, newext, &err); + if (newblock == 0) + goto cleanup; + ablocks[a] = newblock; + } + + /* initialize new leaf */ + newblock = ablocks[--a]; + BUG_ON(newblock == 0); + bh = sb_getblk(inode->i_sb, newblock); + if (!bh) { + err = -EIO; + goto cleanup; + } + lock_buffer(bh); + + if ((err = ext4_journal_get_create_access(handle, bh))) + goto cleanup; + + neh = ext_block_hdr(bh); + neh->eh_entries = 0; + neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode)); + neh->eh_magic = EXT4_EXT_MAGIC; + neh->eh_depth = 0; + ex = EXT_FIRST_EXTENT(neh); + + /* move remain of path[depth] to the new leaf */ + BUG_ON(path[depth].p_hdr->eh_entries != path[depth].p_hdr->eh_max); + /* start copy from next extent */ + /* TODO: we could do it by single memmove */ + m = 0; + path[depth].p_ext++; + while (path[depth].p_ext <= + EXT_MAX_EXTENT(path[depth].p_hdr)) { + ext_debug("move %d:%d:%d in new leaf %lu\n", + le32_to_cpu(path[depth].p_ext->ee_block), + le32_to_cpu(path[depth].p_ext->ee_start), + le16_to_cpu(path[depth].p_ext->ee_len), + newblock); + /*memmove(ex++, path[depth].p_ext++, + sizeof(struct ext4_extent)); + neh->eh_entries++;*/ + path[depth].p_ext++; + m++; + } + if (m) { + memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m); + neh->eh_entries = cpu_to_le16(le16_to_cpu(neh->eh_entries)+m); + } + + set_buffer_uptodate(bh); + unlock_buffer(bh); + + if ((err = ext4_journal_dirty_metadata(handle, bh))) + goto cleanup; + brelse(bh); + bh = NULL; + + /* correct old leaf */ + if (m) { + if ((err = ext4_ext_get_access(handle, inode, path + depth))) + goto cleanup; + path[depth].p_hdr->eh_entries = + cpu_to_le16(le16_to_cpu(path[depth].p_hdr->eh_entries)-m); + if ((err = ext4_ext_dirty(handle, inode, path + depth))) + goto cleanup; + + } + + /* create intermediate indexes */ + k = depth - at - 1; + BUG_ON(k < 0); + if (k) + ext_debug("create %d intermediate indices\n", k); + /* insert new index into current index block */ + /* current depth stored in i var */ + i = depth - 1; + while (k--) { + oldblock = newblock; + newblock = ablocks[--a]; + bh = sb_getblk(inode->i_sb, newblock); + if (!bh) { + err = -EIO; + goto cleanup; + } + lock_buffer(bh); + + if ((err = ext4_journal_get_create_access(handle, bh))) + goto cleanup; + + neh = ext_block_hdr(bh); + neh->eh_entries = cpu_to_le16(1); + neh->eh_magic = EXT4_EXT_MAGIC; + neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode)); + neh->eh_depth = cpu_to_le16(depth - i); + fidx = EXT_FIRST_INDEX(neh); + fidx->ei_block = border; + fidx->ei_leaf = cpu_to_le32(oldblock); + + ext_debug("int.index at %d (block %lu): %lu -> %lu\n", i, + newblock, (unsigned long) le32_to_cpu(border), + oldblock); + /* copy indexes */ + m = 0; + path[i].p_idx++; + + ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx, + EXT_MAX_INDEX(path[i].p_hdr)); + BUG_ON(EXT_MAX_INDEX(path[i].p_hdr) != + EXT_LAST_INDEX(path[i].p_hdr)); + while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) { + ext_debug("%d: move %d:%d in new index %lu\n", i, + le32_to_cpu(path[i].p_idx->ei_block), + le32_to_cpu(path[i].p_idx->ei_leaf), + newblock); + /*memmove(++fidx, path[i].p_idx++, + sizeof(struct ext4_extent_idx)); + neh->eh_entries++; + BUG_ON(neh->eh_entries > neh->eh_max);*/ + path[i].p_idx++; + m++; + } + if (m) { + memmove(++fidx, path[i].p_idx - m, + sizeof(struct ext4_extent_idx) * m); + neh->eh_entries = + cpu_to_le16(le16_to_cpu(neh->eh_entries) + m); + } + set_buffer_uptodate(bh); + unlock_buffer(bh); + + if ((err = ext4_journal_dirty_metadata(handle, bh))) + goto cleanup; + brelse(bh); + bh = NULL; + + /* correct old index */ + if (m) { + err = ext4_ext_get_access(handle, inode, path + i); + if (err) + goto cleanup; + path[i].p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path[i].p_hdr->eh_entries)-m); + err = ext4_ext_dirty(handle, inode, path + i); + if (err) + goto cleanup; + } + + i--; + } + + /* insert new index */ + if (err) + goto cleanup; + + err = ext4_ext_insert_index(handle, inode, path + at, + le32_to_cpu(border), newblock); + +cleanup: + if (bh) { + if (buffer_locked(bh)) + unlock_buffer(bh); + brelse(bh); + } + + if (err) { + /* free all allocated blocks in error case */ + for (i = 0; i < depth; i++) { + if (!ablocks[i]) + continue; + ext4_free_blocks(handle, inode, ablocks[i], 1); + } + } + kfree(ablocks); + + return err; +} + +/* + * routine implements tree growing procedure: + * - allocates new block + * - moves top-level data (index block or leaf) into the new block + * - initialize new top-level, creating index that points to the + * just created block + */ +static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext) +{ + struct ext4_ext_path *curp = path; + struct ext4_extent_header *neh; + struct ext4_extent_idx *fidx; + struct buffer_head *bh; + unsigned long newblock; + int err = 0; + + newblock = ext4_ext_new_block(handle, inode, path, newext, &err); + if (newblock == 0) + return err; + + bh = sb_getblk(inode->i_sb, newblock); + if (!bh) { + err = -EIO; + ext4_std_error(inode->i_sb, err); + return err; + } + lock_buffer(bh); + + if ((err = ext4_journal_get_create_access(handle, bh))) { + unlock_buffer(bh); + goto out; + } + + /* move top-level index/leaf into new block */ + memmove(bh->b_data, curp->p_hdr, sizeof(EXT4_I(inode)->i_data)); + + /* set size of new block */ + neh = ext_block_hdr(bh); + /* old root could have indexes or leaves + * so calculate e_max right way */ + if (ext_depth(inode)) + neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode)); + else + neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode)); + neh->eh_magic = EXT4_EXT_MAGIC; + set_buffer_uptodate(bh); + unlock_buffer(bh); + + if ((err = ext4_journal_dirty_metadata(handle, bh))) + goto out; + + /* create index in new top-level index: num,max,pointer */ + if ((err = ext4_ext_get_access(handle, inode, curp))) + goto out; + + curp->p_hdr->eh_magic = EXT4_EXT_MAGIC; + curp->p_hdr->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode)); + curp->p_hdr->eh_entries = cpu_to_le16(1); + curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr); + /* FIXME: it works, but actually path[0] can be index */ + curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block; + curp->p_idx->ei_leaf = cpu_to_le32(newblock); + + neh = ext_inode_hdr(inode); + fidx = EXT_FIRST_INDEX(neh); + ext_debug("new root: num %d(%d), lblock %d, ptr %d\n", + le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max), + le32_to_cpu(fidx->ei_block), le32_to_cpu(fidx->ei_leaf)); + + neh->eh_depth = cpu_to_le16(path->p_depth + 1); + err = ext4_ext_dirty(handle, inode, curp); +out: + brelse(bh); + + return err; +} + +/* + * routine finds empty index and adds new leaf. if no free index found + * then it requests in-depth growing + */ +static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext) +{ + struct ext4_ext_path *curp; + int depth, i, err = 0; + +repeat: + i = depth = ext_depth(inode); + + /* walk up to the tree and look for free index entry */ + curp = path + depth; + while (i > 0 && !EXT_HAS_FREE_INDEX(curp)) { + i--; + curp--; + } + + /* we use already allocated block for index block + * so, subsequent data blocks should be contigoues */ + if (EXT_HAS_FREE_INDEX(curp)) { + /* if we found index with free entry, then use that + * entry: create all needed subtree and add new leaf */ + err = ext4_ext_split(handle, inode, path, newext, i); + + /* refill path */ + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, + le32_to_cpu(newext->ee_block), + path); + if (IS_ERR(path)) + err = PTR_ERR(path); + } else { + /* tree is full, time to grow in depth */ + err = ext4_ext_grow_indepth(handle, inode, path, newext); + if (err) + goto out; + + /* refill path */ + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, + le32_to_cpu(newext->ee_block), + path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + goto out; + } + + /* + * only first (depth 0 -> 1) produces free space + * in all other cases we have to split growed tree + */ + depth = ext_depth(inode); + if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) { + /* now we need split */ + goto repeat; + } + } + +out: + return err; +} + +/* + * returns allocated block in subsequent extent or EXT_MAX_BLOCK + * NOTE: it consider block number from index entry as + * allocated block. thus, index entries have to be consistent + * with leafs + */ +static unsigned long +ext4_ext_next_allocated_block(struct ext4_ext_path *path) +{ + int depth; + + BUG_ON(path == NULL); + depth = path->p_depth; + + if (depth == 0 && path->p_ext == NULL) + return EXT_MAX_BLOCK; + + while (depth >= 0) { + if (depth == path->p_depth) { + /* leaf */ + if (path[depth].p_ext != + EXT_LAST_EXTENT(path[depth].p_hdr)) + return le32_to_cpu(path[depth].p_ext[1].ee_block); + } else { + /* index */ + if (path[depth].p_idx != + EXT_LAST_INDEX(path[depth].p_hdr)) + return le32_to_cpu(path[depth].p_idx[1].ei_block); + } + depth--; + } + + return EXT_MAX_BLOCK; +} + +/* + * returns first allocated block from next leaf or EXT_MAX_BLOCK + */ +static unsigned ext4_ext_next_leaf_block(struct inode *inode, + struct ext4_ext_path *path) +{ + int depth; + + BUG_ON(path == NULL); + depth = path->p_depth; + + /* zero-tree has no leaf blocks at all */ + if (depth == 0) + return EXT_MAX_BLOCK; + + /* go to index block */ + depth--; + + while (depth >= 0) { + if (path[depth].p_idx != + EXT_LAST_INDEX(path[depth].p_hdr)) + return le32_to_cpu(path[depth].p_idx[1].ei_block); + depth--; + } + + return EXT_MAX_BLOCK; +} + +/* + * if leaf gets modified and modified extent is first in the leaf + * then we have to correct all indexes above + * TODO: do we need to correct tree in all cases? + */ +int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + struct ext4_extent_header *eh; + int depth = ext_depth(inode); + struct ext4_extent *ex; + __le32 border; + int k, err = 0; + + eh = path[depth].p_hdr; + ex = path[depth].p_ext; + BUG_ON(ex == NULL); + BUG_ON(eh == NULL); + + if (depth == 0) { + /* there is no tree at all */ + return 0; + } + + if (ex != EXT_FIRST_EXTENT(eh)) { + /* we correct tree if first leaf got modified only */ + return 0; + } + + /* + * TODO: we need correction if border is smaller then current one + */ + k = depth - 1; + border = path[depth].p_ext->ee_block; + if ((err = ext4_ext_get_access(handle, inode, path + k))) + return err; + path[k].p_idx->ei_block = border; + if ((err = ext4_ext_dirty(handle, inode, path + k))) + return err; + + while (k--) { + /* change all left-side indexes */ + if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr)) + break; + if ((err = ext4_ext_get_access(handle, inode, path + k))) + break; + path[k].p_idx->ei_block = border; + if ((err = ext4_ext_dirty(handle, inode, path + k))) + break; + } + + return err; +} + +static int inline +ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, + struct ext4_extent *ex2) +{ + /* FIXME: 48bit support */ + if (le32_to_cpu(ex1->ee_block) + le16_to_cpu(ex1->ee_len) + != le32_to_cpu(ex2->ee_block)) + return 0; + +#ifdef AGRESSIVE_TEST + if (le16_to_cpu(ex1->ee_len) >= 4) + return 0; +#endif + + if (le32_to_cpu(ex1->ee_start) + le16_to_cpu(ex1->ee_len) + == le32_to_cpu(ex2->ee_start)) + return 1; + return 0; +} + +/* + * this routine tries to merge requsted extent into the existing + * extent or inserts requested extent as new one into the tree, + * creating new leaf in no-space case + */ +int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext) +{ + struct ext4_extent_header * eh; + struct ext4_extent *ex, *fex; + struct ext4_extent *nearex; /* nearest extent */ + struct ext4_ext_path *npath = NULL; + int depth, len, err, next; + + BUG_ON(newext->ee_len == 0); + depth = ext_depth(inode); + ex = path[depth].p_ext; + BUG_ON(path[depth].p_hdr == NULL); + + /* try to insert block into found extent and return */ + if (ex && ext4_can_extents_be_merged(inode, ex, newext)) { + ext_debug("append %d block to %d:%d (from %d)\n", + le16_to_cpu(newext->ee_len), + le32_to_cpu(ex->ee_block), + le16_to_cpu(ex->ee_len), + le32_to_cpu(ex->ee_start)); + if ((err = ext4_ext_get_access(handle, inode, path + depth))) + return err; + ex->ee_len = cpu_to_le16(le16_to_cpu(ex->ee_len) + + le16_to_cpu(newext->ee_len)); + eh = path[depth].p_hdr; + nearex = ex; + goto merge; + } + +repeat: + depth = ext_depth(inode); + eh = path[depth].p_hdr; + if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) + goto has_space; + + /* probably next leaf has space for us? */ + fex = EXT_LAST_EXTENT(eh); + next = ext4_ext_next_leaf_block(inode, path); + if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block) + && next != EXT_MAX_BLOCK) { + ext_debug("next leaf block - %d\n", next); + BUG_ON(npath != NULL); + npath = ext4_ext_find_extent(inode, next, NULL); + if (IS_ERR(npath)) + return PTR_ERR(npath); + BUG_ON(npath->p_depth != path->p_depth); + eh = npath[depth].p_hdr; + if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) { + ext_debug("next leaf isnt full(%d)\n", + le16_to_cpu(eh->eh_entries)); + path = npath; + goto repeat; + } + ext_debug("next leaf has no free space(%d,%d)\n", + le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); + } + + /* + * there is no free space in found leaf + * we're gonna add new leaf in the tree + */ + err = ext4_ext_create_new_leaf(handle, inode, path, newext); + if (err) + goto cleanup; + depth = ext_depth(inode); + eh = path[depth].p_hdr; + +has_space: + nearex = path[depth].p_ext; + + if ((err = ext4_ext_get_access(handle, inode, path + depth))) + goto cleanup; + + if (!nearex) { + /* there is no extent in this leaf, create first one */ + ext_debug("first extent in the leaf: %d:%d:%d\n", + le32_to_cpu(newext->ee_block), + le32_to_cpu(newext->ee_start), + le16_to_cpu(newext->ee_len)); + path[depth].p_ext = EXT_FIRST_EXTENT(eh); + } else if (le32_to_cpu(newext->ee_block) + > le32_to_cpu(nearex->ee_block)) { +/* BUG_ON(newext->ee_block == nearex->ee_block); */ + if (nearex != EXT_LAST_EXTENT(eh)) { + len = EXT_MAX_EXTENT(eh) - nearex; + len = (len - 1) * sizeof(struct ext4_extent); + len = len < 0 ? 0 : len; + ext_debug("insert %d:%d:%d after: nearest 0x%p, " + "move %d from 0x%p to 0x%p\n", + le32_to_cpu(newext->ee_block), + le32_to_cpu(newext->ee_start), + le16_to_cpu(newext->ee_len), + nearex, len, nearex + 1, nearex + 2); + memmove(nearex + 2, nearex + 1, len); + } + path[depth].p_ext = nearex + 1; + } else { + BUG_ON(newext->ee_block == nearex->ee_block); + len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext4_extent); + len = len < 0 ? 0 : len; + ext_debug("insert %d:%d:%d before: nearest 0x%p, " + "move %d from 0x%p to 0x%p\n", + le32_to_cpu(newext->ee_block), + le32_to_cpu(newext->ee_start), + le16_to_cpu(newext->ee_len), + nearex, len, nearex + 1, nearex + 2); + memmove(nearex + 1, nearex, len); + path[depth].p_ext = nearex; + } + + eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)+1); + nearex = path[depth].p_ext; + nearex->ee_block = newext->ee_block; + nearex->ee_start = newext->ee_start; + nearex->ee_len = newext->ee_len; + /* FIXME: support for large fs */ + nearex->ee_start_hi = 0; + +merge: + /* try to merge extents to the right */ + while (nearex < EXT_LAST_EXTENT(eh)) { + if (!ext4_can_extents_be_merged(inode, nearex, nearex + 1)) + break; + /* merge with next extent! */ + nearex->ee_len = cpu_to_le16(le16_to_cpu(nearex->ee_len) + + le16_to_cpu(nearex[1].ee_len)); + if (nearex + 1 < EXT_LAST_EXTENT(eh)) { + len = (EXT_LAST_EXTENT(eh) - nearex - 1) + * sizeof(struct ext4_extent); + memmove(nearex + 1, nearex + 2, len); + } + eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); + BUG_ON(eh->eh_entries == 0); + } + + /* try to merge extents to the left */ + + /* time to correct all indexes above */ + err = ext4_ext_correct_indexes(handle, inode, path); + if (err) + goto cleanup; + + err = ext4_ext_dirty(handle, inode, path + depth); + +cleanup: + if (npath) { + ext4_ext_drop_refs(npath); + kfree(npath); + } + ext4_ext_tree_changed(inode); + ext4_ext_invalidate_cache(inode); + return err; +} + +int ext4_ext_walk_space(struct inode *inode, unsigned long block, + unsigned long num, ext_prepare_callback func, + void *cbdata) +{ + struct ext4_ext_path *path = NULL; + struct ext4_ext_cache cbex; + struct ext4_extent *ex; + unsigned long next, start = 0, end = 0; + unsigned long last = block + num; + int depth, exists, err = 0; + + BUG_ON(func == NULL); + BUG_ON(inode == NULL); + + while (block < last && block != EXT_MAX_BLOCK) { + num = last - block; + /* find extent for this block */ + path = ext4_ext_find_extent(inode, block, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + break; + } + + depth = ext_depth(inode); + BUG_ON(path[depth].p_hdr == NULL); + ex = path[depth].p_ext; + next = ext4_ext_next_allocated_block(path); + + exists = 0; + if (!ex) { + /* there is no extent yet, so try to allocate + * all requested space */ + start = block; + end = block + num; + } else if (le32_to_cpu(ex->ee_block) > block) { + /* need to allocate space before found extent */ + start = block; + end = le32_to_cpu(ex->ee_block); + if (block + num < end) + end = block + num; + } else if (block >= + le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len)) { + /* need to allocate space after found extent */ + start = block; + end = block + num; + if (end >= next) + end = next; + } else if (block >= le32_to_cpu(ex->ee_block)) { + /* + * some part of requested space is covered + * by found extent + */ + start = block; + end = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len); + if (block + num < end) + end = block + num; + exists = 1; + } else { + BUG(); + } + BUG_ON(end <= start); + + if (!exists) { + cbex.ec_block = start; + cbex.ec_len = end - start; + cbex.ec_start = 0; + cbex.ec_type = EXT4_EXT_CACHE_GAP; + } else { + cbex.ec_block = le32_to_cpu(ex->ee_block); + cbex.ec_len = le16_to_cpu(ex->ee_len); + cbex.ec_start = le32_to_cpu(ex->ee_start); + cbex.ec_type = EXT4_EXT_CACHE_EXTENT; + } + + BUG_ON(cbex.ec_len == 0); + err = func(inode, path, &cbex, cbdata); + ext4_ext_drop_refs(path); + + if (err < 0) + break; + if (err == EXT_REPEAT) + continue; + else if (err == EXT_BREAK) { + err = 0; + break; + } + + if (ext_depth(inode) != depth) { + /* depth was changed. we have to realloc path */ + kfree(path); + path = NULL; + } + + block = cbex.ec_block + cbex.ec_len; + } + + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + + return err; +} + +static inline void +ext4_ext_put_in_cache(struct inode *inode, __u32 block, + __u32 len, __u32 start, int type) +{ + struct ext4_ext_cache *cex; + BUG_ON(len == 0); + cex = &EXT4_I(inode)->i_cached_extent; + cex->ec_type = type; + cex->ec_block = block; + cex->ec_len = len; + cex->ec_start = start; +} + +/* + * this routine calculate boundaries of the gap requested block fits into + * and cache this gap + */ +static inline void +ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, + unsigned long block) +{ + int depth = ext_depth(inode); + unsigned long lblock, len; + struct ext4_extent *ex; + + ex = path[depth].p_ext; + if (ex == NULL) { + /* there is no extent yet, so gap is [0;-] */ + lblock = 0; + len = EXT_MAX_BLOCK; + ext_debug("cache gap(whole file):"); + } else if (block < le32_to_cpu(ex->ee_block)) { + lblock = block; + len = le32_to_cpu(ex->ee_block) - block; + ext_debug("cache gap(before): %lu [%lu:%lu]", + (unsigned long) block, + (unsigned long) le32_to_cpu(ex->ee_block), + (unsigned long) le16_to_cpu(ex->ee_len)); + } else if (block >= le32_to_cpu(ex->ee_block) + + le16_to_cpu(ex->ee_len)) { + lblock = le32_to_cpu(ex->ee_block) + + le16_to_cpu(ex->ee_len); + len = ext4_ext_next_allocated_block(path); + ext_debug("cache gap(after): [%lu:%lu] %lu", + (unsigned long) le32_to_cpu(ex->ee_block), + (unsigned long) le16_to_cpu(ex->ee_len), + (unsigned long) block); + BUG_ON(len == lblock); + len = len - lblock; + } else { + lblock = len = 0; + BUG(); + } + + ext_debug(" -> %lu:%lu\n", (unsigned long) lblock, len); + ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP); +} + +static inline int +ext4_ext_in_cache(struct inode *inode, unsigned long block, + struct ext4_extent *ex) +{ + struct ext4_ext_cache *cex; + + cex = &EXT4_I(inode)->i_cached_extent; + + /* has cache valid data? */ + if (cex->ec_type == EXT4_EXT_CACHE_NO) + return EXT4_EXT_CACHE_NO; + + BUG_ON(cex->ec_type != EXT4_EXT_CACHE_GAP && + cex->ec_type != EXT4_EXT_CACHE_EXTENT); + if (block >= cex->ec_block && block < cex->ec_block + cex->ec_len) { + ex->ee_block = cpu_to_le32(cex->ec_block); + ex->ee_start = cpu_to_le32(cex->ec_start); + ex->ee_len = cpu_to_le16(cex->ec_len); + ext_debug("%lu cached by %lu:%lu:%lu\n", + (unsigned long) block, + (unsigned long) cex->ec_block, + (unsigned long) cex->ec_len, + (unsigned long) cex->ec_start); + return cex->ec_type; + } + + /* not in cache */ + return EXT4_EXT_CACHE_NO; +} + +/* + * routine removes index from the index block + * it's used in truncate case only. thus all requests are for + * last index in the block only + */ +int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + struct buffer_head *bh; + int err; + unsigned long leaf; + + /* free index block */ + path--; + leaf = le32_to_cpu(path->p_idx->ei_leaf); + BUG_ON(path->p_hdr->eh_entries == 0); + if ((err = ext4_ext_get_access(handle, inode, path))) + return err; + path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1); + if ((err = ext4_ext_dirty(handle, inode, path))) + return err; + ext_debug("index is empty, remove it, free block %lu\n", leaf); + bh = sb_find_get_block(inode->i_sb, leaf); + ext4_forget(handle, 1, inode, bh, leaf); + ext4_free_blocks(handle, inode, leaf, 1); + return err; +} + +/* + * This routine returns max. credits extent tree can consume. + * It should be OK for low-performance paths like ->writepage() + * To allow many writing process to fit a single transaction, + * caller should calculate credits under truncate_mutex and + * pass actual path. + */ +int inline ext4_ext_calc_credits_for_insert(struct inode *inode, + struct ext4_ext_path *path) +{ + int depth, needed; + + if (path) { + /* probably there is space in leaf? */ + depth = ext_depth(inode); + if (le16_to_cpu(path[depth].p_hdr->eh_entries) + < le16_to_cpu(path[depth].p_hdr->eh_max)) + return 1; + } + + /* + * given 32bit logical block (4294967296 blocks), max. tree + * can be 4 levels in depth -- 4 * 340^4 == 53453440000. + * let's also add one more level for imbalance. + */ + depth = 5; + + /* allocation of new data block(s) */ + needed = 2; + + /* + * tree can be full, so it'd need to grow in depth: + * allocation + old root + new root + */ + needed += 2 + 1 + 1; + + /* + * Index split can happen, we'd need: + * allocate intermediate indexes (bitmap + group) + * + change two blocks at each level, but root (already included) + */ + needed = (depth * 2) + (depth * 2); + + /* any allocation modifies superblock */ + needed += 1; + + return needed; +} + +static int ext4_remove_blocks(handle_t *handle, struct inode *inode, + struct ext4_extent *ex, + unsigned long from, unsigned long to) +{ + struct buffer_head *bh; + int i; + +#ifdef EXTENTS_STATS + { + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + unsigned short ee_len = le16_to_cpu(ex->ee_len); + spin_lock(&sbi->s_ext_stats_lock); + sbi->s_ext_blocks += ee_len; + sbi->s_ext_extents++; + if (ee_len < sbi->s_ext_min) + sbi->s_ext_min = ee_len; + if (ee_len > sbi->s_ext_max) + sbi->s_ext_max = ee_len; + if (ext_depth(inode) > sbi->s_depth_max) + sbi->s_depth_max = ext_depth(inode); + spin_unlock(&sbi->s_ext_stats_lock); + } +#endif + if (from >= le32_to_cpu(ex->ee_block) + && to == le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - 1) { + /* tail removal */ + unsigned long num, start; + num = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - from; + start = le32_to_cpu(ex->ee_start) + le16_to_cpu(ex->ee_len) - num; + ext_debug("free last %lu blocks starting %lu\n", num, start); + for (i = 0; i < num; i++) { + bh = sb_find_get_block(inode->i_sb, start + i); + ext4_forget(handle, 0, inode, bh, start + i); + } + ext4_free_blocks(handle, inode, start, num); + } else if (from == le32_to_cpu(ex->ee_block) + && to <= le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - 1) { + printk("strange request: removal %lu-%lu from %u:%u\n", + from, to, le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len)); + } else { + printk("strange request: removal(2) %lu-%lu from %u:%u\n", + from, to, le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len)); + } + return 0; +} + +static int +ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, unsigned long start) +{ + int err = 0, correct_index = 0; + int depth = ext_depth(inode), credits; + struct ext4_extent_header *eh; + unsigned a, b, block, num; + unsigned long ex_ee_block; + unsigned short ex_ee_len; + struct ext4_extent *ex; + + ext_debug("truncate since %lu in leaf\n", start); + if (!path[depth].p_hdr) + path[depth].p_hdr = ext_block_hdr(path[depth].p_bh); + eh = path[depth].p_hdr; + BUG_ON(eh == NULL); + BUG_ON(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max)); + BUG_ON(eh->eh_magic != EXT4_EXT_MAGIC); + + /* find where to start removing */ + ex = EXT_LAST_EXTENT(eh); + + ex_ee_block = le32_to_cpu(ex->ee_block); + ex_ee_len = le16_to_cpu(ex->ee_len); + + while (ex >= EXT_FIRST_EXTENT(eh) && + ex_ee_block + ex_ee_len > start) { + ext_debug("remove ext %lu:%u\n", ex_ee_block, ex_ee_len); + path[depth].p_ext = ex; + + a = ex_ee_block > start ? ex_ee_block : start; + b = ex_ee_block + ex_ee_len - 1 < EXT_MAX_BLOCK ? + ex_ee_block + ex_ee_len - 1 : EXT_MAX_BLOCK; + + ext_debug(" border %u:%u\n", a, b); + + if (a != ex_ee_block && b != ex_ee_block + ex_ee_len - 1) { + block = 0; + num = 0; + BUG(); + } else if (a != ex_ee_block) { + /* remove tail of the extent */ + block = ex_ee_block; + num = a - block; + } else if (b != ex_ee_block + ex_ee_len - 1) { + /* remove head of the extent */ + block = a; + num = b - a; + /* there is no "make a hole" API yet */ + BUG(); + } else { + /* remove whole extent: excellent! */ + block = ex_ee_block; + num = 0; + BUG_ON(a != ex_ee_block); + BUG_ON(b != ex_ee_block + ex_ee_len - 1); + } + + /* at present, extent can't cross block group */ + /* leaf + bitmap + group desc + sb + inode */ + credits = 5; + if (ex == EXT_FIRST_EXTENT(eh)) { + correct_index = 1; + credits += (ext_depth(inode)) + 1; + } +#ifdef CONFIG_QUOTA + credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); +#endif + + handle = ext4_ext_journal_restart(handle, credits); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto out; + } + + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) + goto out; + + err = ext4_remove_blocks(handle, inode, ex, a, b); + if (err) + goto out; + + if (num == 0) { + /* this extent is removed entirely mark slot unused */ + ex->ee_start = 0; + eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); + } + + ex->ee_block = cpu_to_le32(block); + ex->ee_len = cpu_to_le16(num); + + err = ext4_ext_dirty(handle, inode, path + depth); + if (err) + goto out; + + ext_debug("new extent: %u:%u:%u\n", block, num, + le32_to_cpu(ex->ee_start)); + ex--; + ex_ee_block = le32_to_cpu(ex->ee_block); + ex_ee_len = le16_to_cpu(ex->ee_len); + } + + if (correct_index && eh->eh_entries) + err = ext4_ext_correct_indexes(handle, inode, path); + + /* if this leaf is free, then we should + * remove it from index block above */ + if (err == 0 && eh->eh_entries == 0 && path[depth].p_bh != NULL) + err = ext4_ext_rm_idx(handle, inode, path + depth); + +out: + return err; +} + +/* + * returns 1 if current index have to be freed (even partial) + */ +static int inline +ext4_ext_more_to_rm(struct ext4_ext_path *path) +{ + BUG_ON(path->p_idx == NULL); + + if (path->p_idx < EXT_FIRST_INDEX(path->p_hdr)) + return 0; + + /* + * if truncate on deeper level happened it it wasn't partial + * so we have to consider current index for truncation + */ + if (le16_to_cpu(path->p_hdr->eh_entries) == path->p_block) + return 0; + return 1; +} + +int ext4_ext_remove_space(struct inode *inode, unsigned long start) +{ + struct super_block *sb = inode->i_sb; + int depth = ext_depth(inode); + struct ext4_ext_path *path; + handle_t *handle; + int i = 0, err = 0; + + ext_debug("truncate since %lu\n", start); + + /* probably first extent we're gonna free will be last in block */ + handle = ext4_journal_start(inode, depth + 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + ext4_ext_invalidate_cache(inode); + + /* + * we start scanning from right side freeing all the blocks + * after i_size and walking into the deep + */ + path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL); + if (path == NULL) { + ext4_journal_stop(handle); + return -ENOMEM; + } + memset(path, 0, sizeof(struct ext4_ext_path) * (depth + 1)); + path[0].p_hdr = ext_inode_hdr(inode); + if (ext4_ext_check_header(__FUNCTION__, inode, path[0].p_hdr)) { + err = -EIO; + goto out; + } + path[0].p_depth = depth; + + while (i >= 0 && err == 0) { + if (i == depth) { + /* this is leaf block */ + err = ext4_ext_rm_leaf(handle, inode, path, start); + /* root level have p_bh == NULL, brelse() eats this */ + brelse(path[i].p_bh); + path[i].p_bh = NULL; + i--; + continue; + } + + /* this is index block */ + if (!path[i].p_hdr) { + ext_debug("initialize header\n"); + path[i].p_hdr = ext_block_hdr(path[i].p_bh); + if (ext4_ext_check_header(__FUNCTION__, inode, + path[i].p_hdr)) { + err = -EIO; + goto out; + } + } + + BUG_ON(le16_to_cpu(path[i].p_hdr->eh_entries) + > le16_to_cpu(path[i].p_hdr->eh_max)); + BUG_ON(path[i].p_hdr->eh_magic != EXT4_EXT_MAGIC); + + if (!path[i].p_idx) { + /* this level hasn't touched yet */ + path[i].p_idx = EXT_LAST_INDEX(path[i].p_hdr); + path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries)+1; + ext_debug("init index ptr: hdr 0x%p, num %d\n", + path[i].p_hdr, + le16_to_cpu(path[i].p_hdr->eh_entries)); + } else { + /* we've already was here, see at next index */ + path[i].p_idx--; + } + + ext_debug("level %d - index, first 0x%p, cur 0x%p\n", + i, EXT_FIRST_INDEX(path[i].p_hdr), + path[i].p_idx); + if (ext4_ext_more_to_rm(path + i)) { + /* go to the next level */ + ext_debug("move to level %d (block %d)\n", + i + 1, le32_to_cpu(path[i].p_idx->ei_leaf)); + memset(path + i + 1, 0, sizeof(*path)); + path[i+1].p_bh = + sb_bread(sb, le32_to_cpu(path[i].p_idx->ei_leaf)); + if (!path[i+1].p_bh) { + /* should we reset i_size? */ + err = -EIO; + break; + } + + /* put actual number of indexes to know is this + * number got changed at the next iteration */ + path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries); + i++; + } else { + /* we finish processing this index, go up */ + if (path[i].p_hdr->eh_entries == 0 && i > 0) { + /* index is empty, remove it + * handle must be already prepared by the + * truncatei_leaf() */ + err = ext4_ext_rm_idx(handle, inode, path + i); + } + /* root level have p_bh == NULL, brelse() eats this */ + brelse(path[i].p_bh); + path[i].p_bh = NULL; + i--; + ext_debug("return to level %d\n", i); + } + } + + /* TODO: flexible tree reduction should be here */ + if (path->p_hdr->eh_entries == 0) { + /* + * truncate to zero freed all the tree + * so, we need to correct eh_depth + */ + err = ext4_ext_get_access(handle, inode, path); + if (err == 0) { + ext_inode_hdr(inode)->eh_depth = 0; + ext_inode_hdr(inode)->eh_max = + cpu_to_le16(ext4_ext_space_root(inode)); + err = ext4_ext_dirty(handle, inode, path); + } + } +out: + ext4_ext_tree_changed(inode); + ext4_ext_drop_refs(path); + kfree(path); + ext4_journal_stop(handle); + + return err; +} + +/* + * called at mount time + */ +void ext4_ext_init(struct super_block *sb) +{ + /* + * possible initialization would be here + */ + + if (test_opt(sb, EXTENTS)) { + printk("EXT4-fs: file extents enabled"); +#ifdef AGRESSIVE_TEST + printk(", agressive tests"); +#endif +#ifdef CHECK_BINSEARCH + printk(", check binsearch"); +#endif +#ifdef EXTENTS_STATS + printk(", stats"); +#endif + printk("\n"); +#ifdef EXTENTS_STATS + spin_lock_init(&EXT4_SB(sb)->s_ext_stats_lock); + EXT4_SB(sb)->s_ext_min = 1 << 30; + EXT4_SB(sb)->s_ext_max = 0; +#endif + } +} + +/* + * called at umount time + */ +void ext4_ext_release(struct super_block *sb) +{ + if (!test_opt(sb, EXTENTS)) + return; + +#ifdef EXTENTS_STATS + if (EXT4_SB(sb)->s_ext_blocks && EXT4_SB(sb)->s_ext_extents) { + struct ext4_sb_info *sbi = EXT4_SB(sb); + printk(KERN_ERR "EXT4-fs: %lu blocks in %lu extents (%lu ave)\n", + sbi->s_ext_blocks, sbi->s_ext_extents, + sbi->s_ext_blocks / sbi->s_ext_extents); + printk(KERN_ERR "EXT4-fs: extents: %lu min, %lu max, max depth %lu\n", + sbi->s_ext_min, sbi->s_ext_max, sbi->s_depth_max); + } +#endif +} + +int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, sector_t iblock, + unsigned long max_blocks, struct buffer_head *bh_result, + int create, int extend_disksize) +{ + struct ext4_ext_path *path = NULL; + struct ext4_extent newex, *ex; + int goal, newblock, err = 0, depth; + unsigned long allocated = 0; + + __clear_bit(BH_New, &bh_result->b_state); + ext_debug("blocks %d/%lu requested for inode %u\n", (int) iblock, + max_blocks, (unsigned) inode->i_ino); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + + /* check in cache */ + if ((goal = ext4_ext_in_cache(inode, iblock, &newex))) { + if (goal == EXT4_EXT_CACHE_GAP) { + if (!create) { + /* block isn't allocated yet and + * user don't want to allocate it */ + goto out2; + } + /* we should allocate requested block */ + } else if (goal == EXT4_EXT_CACHE_EXTENT) { + /* block is already allocated */ + newblock = iblock + - le32_to_cpu(newex.ee_block) + + le32_to_cpu(newex.ee_start); + /* number of remain blocks in the extent */ + allocated = le16_to_cpu(newex.ee_len) - + (iblock - le32_to_cpu(newex.ee_block)); + goto out; + } else { + BUG(); + } + } + + /* find extent for this block */ + path = ext4_ext_find_extent(inode, iblock, NULL); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + goto out2; + } + + depth = ext_depth(inode); + + /* + * consistent leaf must not be empty + * this situations is possible, though, _during_ tree modification + * this is why assert can't be put in ext4_ext_find_extent() + */ + BUG_ON(path[depth].p_ext == NULL && depth != 0); + + if ((ex = path[depth].p_ext)) { + unsigned long ee_block = le32_to_cpu(ex->ee_block); + unsigned long ee_start = le32_to_cpu(ex->ee_start); + unsigned short ee_len = le16_to_cpu(ex->ee_len); + /* if found exent covers block, simple return it */ + if (iblock >= ee_block && iblock < ee_block + ee_len) { + newblock = iblock - ee_block + ee_start; + /* number of remain blocks in the extent */ + allocated = ee_len - (iblock - ee_block); + ext_debug("%d fit into %lu:%d -> %d\n", (int) iblock, + ee_block, ee_len, newblock); + ext4_ext_put_in_cache(inode, ee_block, ee_len, + ee_start, EXT4_EXT_CACHE_EXTENT); + goto out; + } + } + + /* + * requested block isn't allocated yet + * we couldn't try to create block if create flag is zero + */ + if (!create) { + /* put just found gap into cache to speedup subsequest reqs */ + ext4_ext_put_gap_in_cache(inode, path, iblock); + goto out2; + } + /* + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary + */ + if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) + ext4_init_block_alloc_info(inode); + + /* allocate new block */ + goal = ext4_ext_find_goal(inode, path, iblock); + allocated = max_blocks; + newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err); + if (!newblock) + goto out2; + ext_debug("allocate new block: goal %d, found %d/%lu\n", + goal, newblock, allocated); + + /* try to insert new extent into found leaf and return */ + newex.ee_block = cpu_to_le32(iblock); + newex.ee_start = cpu_to_le32(newblock); + newex.ee_len = cpu_to_le16(allocated); + err = ext4_ext_insert_extent(handle, inode, path, &newex); + if (err) + goto out2; + + if (extend_disksize && inode->i_size > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = inode->i_size; + + /* previous routine could use block we allocated */ + newblock = le32_to_cpu(newex.ee_start); + __set_bit(BH_New, &bh_result->b_state); + + ext4_ext_put_in_cache(inode, iblock, allocated, newblock, + EXT4_EXT_CACHE_EXTENT); +out: + if (allocated > max_blocks) + allocated = max_blocks; + ext4_ext_show_leaf(inode, path); + __set_bit(BH_Mapped, &bh_result->b_state); + bh_result->b_bdev = inode->i_sb->s_bdev; + bh_result->b_blocknr = newblock; +out2: + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + + return err ? err : allocated; +} + +void ext4_ext_truncate(struct inode * inode, struct page *page) +{ + struct address_space *mapping = inode->i_mapping; + struct super_block *sb = inode->i_sb; + unsigned long last_block; + handle_t *handle; + int err = 0; + + /* + * probably first extent we're gonna free will be last in block + */ + err = ext4_writepage_trans_blocks(inode) + 3; + handle = ext4_journal_start(inode, err); + if (IS_ERR(handle)) { + if (page) { + clear_highpage(page); + flush_dcache_page(page); + unlock_page(page); + page_cache_release(page); + } + return; + } + + if (page) + ext4_block_truncate_page(handle, page, mapping, inode->i_size); + + mutex_lock(&EXT4_I(inode)->truncate_mutex); + ext4_ext_invalidate_cache(inode); + + /* + * TODO: optimization is possible here + * probably we need not scaning at all, + * because page truncation is enough + */ + if (ext4_orphan_add(handle, inode)) + goto out_stop; + + /* we have to know where to truncate from in crash case */ + EXT4_I(inode)->i_disksize = inode->i_size; + ext4_mark_inode_dirty(handle, inode); + + last_block = (inode->i_size + sb->s_blocksize - 1) + >> EXT4_BLOCK_SIZE_BITS(sb); + err = ext4_ext_remove_space(inode, last_block); + + /* In a multi-transaction truncate, we only make the final + * transaction synchronous */ + if (IS_SYNC(inode)) + handle->h_sync = 1; + +out_stop: + /* + * If this was a simple ftruncate(), and the file will remain alive + * then we need to clear up the orphan record which we created above. + * However, if this was a real unlink then we were called by + * ext4_delete_inode(), and we allow that function to clean up the + * orphan info for us. + */ + if (inode->i_nlink) + ext4_orphan_del(handle, inode); + + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + ext4_journal_stop(handle); +} + +/* + * this routine calculate max number of blocks we could modify + * in order to allocate new block for an inode + */ +int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) +{ + int needed; + + needed = ext4_ext_calc_credits_for_insert(inode, NULL); + + /* caller want to allocate num blocks, but note it includes sb */ + needed = needed * num - (num - 1); + +#ifdef CONFIG_QUOTA + needed += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); +#endif + + return needed; +} + +EXPORT_SYMBOL(ext4_mark_inode_dirty); +EXPORT_SYMBOL(ext4_ext_invalidate_cache); +EXPORT_SYMBOL(ext4_ext_insert_extent); +EXPORT_SYMBOL(ext4_ext_walk_space); +EXPORT_SYMBOL(ext4_ext_find_goal); +EXPORT_SYMBOL(ext4_ext_calc_credits_for_insert); + diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 34d39ae..e17a6c9 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -615,6 +615,17 @@ got: ext4_std_error(sb, err); goto fail_free_drop; } + if (test_opt(sb, EXTENTS)) { + EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL; + ext4_ext_tree_init(handle, inode); + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) { + err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); + if (err) goto fail; + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS); + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + } + } ext4_debug("allocating inode %lu\n", inode->i_ino); goto really_out; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 0d96c7d..2b81b13 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -40,8 +40,6 @@ #include "xattr.h" #include "acl.h" -static int ext4_writepage_trans_blocks(struct inode *inode); - /* * Test whether an inode is a fast symlink. */ @@ -804,6 +802,7 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, ext4_fsblk_t first_block = 0; + J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)); J_ASSERT(handle != NULL || create == 0); depth = ext4_block_to_path(inode,iblock,offsets,&blocks_to_boundary); @@ -984,7 +983,7 @@ static int ext4_get_block(struct inode *inode, sector_t iblock, get_block: if (ret == 0) { - ret = ext4_get_blocks_handle(handle, inode, iblock, + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, bh_result, create, 0); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); @@ -1008,7 +1007,7 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, dummy.b_state = 0; dummy.b_blocknr = -1000; buffer_trace_init(&dummy.b_history); - err = ext4_get_blocks_handle(handle, inode, block, 1, + err = ext4_get_blocks_wrap(handle, inode, block, 1, &dummy, create, 1); /* * ext4_get_blocks_handle() returns number of blocks @@ -1759,7 +1758,7 @@ void ext4_set_aops(struct inode *inode) * This required during truncate. We need to physically zero the tail end * of that block so it doesn't yield old data if the file is later grown. */ -static int ext4_block_truncate_page(handle_t *handle, struct page *page, +int ext4_block_truncate_page(handle_t *handle, struct page *page, struct address_space *mapping, loff_t from) { ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; @@ -2263,6 +2262,9 @@ void ext4_truncate(struct inode *inode) return; } + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_truncate(inode, page); + handle = start_transaction(inode); if (IS_ERR(handle)) { if (page) { @@ -3003,12 +3005,15 @@ err_out: * block and work out the exact number of indirects which are touched. Pah. */ -static int ext4_writepage_trans_blocks(struct inode *inode) +int ext4_writepage_trans_blocks(struct inode *inode) { int bpp = ext4_journal_blocks_per_page(inode); int indirects = (EXT4_NDIR_BLOCKS % bpp) ? 5 : 3; int ret; + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_writepage_trans_blocks(inode, bpp); + if (ext4_should_journal_data(inode)) ret = 3 * (bpp + indirects) + 2; else diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index a63dce2..22a737c 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -248,7 +248,6 @@ flags_err: return err; } - default: return -ENOTTY; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f131bb6..69f87525 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -390,6 +390,7 @@ static void ext4_put_super (struct super_block * sb) struct ext4_super_block *es = sbi->s_es; int i; + ext4_ext_release(sb); ext4_xattr_put_super(sb); jbd2_journal_destroy(sbi->s_journal); if (!(sb->s_flags & MS_RDONLY)) { @@ -454,6 +455,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) #endif ei->i_block_alloc_info = NULL; ei->vfs_inode.i_version = 1; + memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); return &ei->vfs_inode; } @@ -677,7 +679,7 @@ enum { Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, - Opt_grpquota + Opt_grpquota, Opt_extents, }; static match_table_t tokens = { @@ -727,6 +729,7 @@ static match_table_t tokens = { {Opt_quota, "quota"}, {Opt_usrquota, "usrquota"}, {Opt_barrier, "barrier=%u"}, + {Opt_extents, "extents"}, {Opt_err, NULL}, {Opt_resize, "resize"}, }; @@ -1059,6 +1062,9 @@ clear_qf_name: case Opt_bh: clear_opt(sbi->s_mount_opt, NOBH); break; + case Opt_extents: + set_opt (sbi->s_mount_opt, EXTENTS); + break; default: printk (KERN_ERR "EXT4-fs: Unrecognized mount option \"%s\" " @@ -1787,6 +1793,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered": "writeback"); + ext4_ext_init(sb); + lock_kernel(); return 0; diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index f582cd7..b61181aad 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -178,8 +178,9 @@ struct ext4_group_desc #define EXT4_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ #define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/ #define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ +#define EXT4_EXTENTS_FL 0x00080000 /* Inode uses extents */ -#define EXT4_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */ +#define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ #define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ /* @@ -384,6 +385,7 @@ struct ext4_inode { #define EXT4_MOUNT_QUOTA 0x80000 /* Some quota option set */ #define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */ #define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */ +#define EXT4_MOUNT_EXTENTS 0x400000 /* Extents support */ /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H @@ -582,11 +584,13 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) #define EXT4_FEATURE_INCOMPAT_RECOVER 0x0004 /* Needs recovery */ #define EXT4_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ #define EXT4_FEATURE_INCOMPAT_META_BG 0x0010 +#define EXT4_FEATURE_INCOMPAT_EXTENTS 0x0040 /* extents support */ #define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR #define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ EXT4_FEATURE_INCOMPAT_RECOVER| \ - EXT4_FEATURE_INCOMPAT_META_BG) + EXT4_FEATURE_INCOMPAT_META_BG| \ + EXT4_FEATURE_INCOMPAT_EXTENTS) #define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ EXT4_FEATURE_RO_COMPAT_BTREE_DIR) @@ -825,6 +829,9 @@ extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); extern void ext4_truncate (struct inode *); extern void ext4_set_inode_flags(struct inode *); extern void ext4_set_aops(struct inode *inode); +extern int ext4_writepage_trans_blocks(struct inode *); +extern int ext4_block_truncate_page(handle_t *handle, struct page *page, + struct address_space *mapping, loff_t from); /* ioctl.c */ extern int ext4_ioctl (struct inode *, struct file *, unsigned int, @@ -879,6 +886,28 @@ extern struct inode_operations ext4_special_inode_operations; extern struct inode_operations ext4_symlink_inode_operations; extern struct inode_operations ext4_fast_symlink_inode_operations; +/* extents.c */ +extern int ext4_ext_tree_init(handle_t *handle, struct inode *); +extern int ext4_ext_writepage_trans_blocks(struct inode *, int); +extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t iblock, + unsigned long max_blocks, struct buffer_head *bh_result, + int create, int extend_disksize); +extern void ext4_ext_truncate(struct inode *, struct page *); +extern void ext4_ext_init(struct super_block *); +extern void ext4_ext_release(struct super_block *); +static inline int +ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, + unsigned long max_blocks, struct buffer_head *bh, + int create, int extend_disksize) +{ + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_get_blocks(handle, inode, block, max_blocks, + bh, create, extend_disksize); + return ext4_get_blocks_handle(handle, inode, block, max_blocks, bh, + create, extend_disksize); +} + #endif /* __KERNEL__ */ diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h new file mode 100644 index 0000000..8029879 --- /dev/null +++ b/include/linux/ext4_fs_extents.h @@ -0,0 +1,196 @@ +/* + * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com + * Written by Alex Tomas + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + */ + +#ifndef _LINUX_EXT4_EXTENTS +#define _LINUX_EXT4_EXTENTS + +#include + +/* + * with AGRESSIVE_TEST defined capacity of index/leaf blocks + * become very little, so index split, in-depth growing and + * other hard changes happens much more often + * this is for debug purposes only + */ +#define AGRESSIVE_TEST_ + +/* + * with EXTENTS_STATS defined number of blocks and extents + * are collected in truncate path. they'll be showed at + * umount time + */ +#define EXTENTS_STATS__ + +/* + * if CHECK_BINSEARCH defined, then results of binary search + * will be checked by linear search + */ +#define CHECK_BINSEARCH__ + +/* + * if EXT_DEBUG is defined you can use 'extdebug' mount option + * to get lots of info what's going on + */ +#define EXT_DEBUG__ +#ifdef EXT_DEBUG +#define ext_debug(a...) printk(a) +#else +#define ext_debug(a...) +#endif + +/* + * if EXT_STATS is defined then stats numbers are collected + * these number will be displayed at umount time + */ +#define EXT_STATS_ + + +/* + * ext4_inode has i_block array (60 bytes total) + * first 12 bytes store ext4_extent_header + * the remain stores array of ext4_extent + */ + +/* + * this is extent on-disk structure + * it's used at the bottom of the tree + */ +struct ext4_extent { + __le32 ee_block; /* first logical block extent covers */ + __le16 ee_len; /* number of blocks covered by extent */ + __le16 ee_start_hi; /* high 16 bits of physical block */ + __le32 ee_start; /* low 32 bigs of physical block */ +}; + +/* + * this is index on-disk structure + * it's used at all the levels, but the bottom + */ +struct ext4_extent_idx { + __le32 ei_block; /* index covers logical blocks from 'block' */ + __le32 ei_leaf; /* pointer to the physical block of the next * + * level. leaf or next index could bet here */ + __le16 ei_leaf_hi; /* high 16 bits of physical block */ + __u16 ei_unused; +}; + +/* + * each block (leaves and indexes), even inode-stored has header + */ +struct ext4_extent_header { + __le16 eh_magic; /* probably will support different formats */ + __le16 eh_entries; /* number of valid entries */ + __le16 eh_max; /* capacity of store in entries */ + __le16 eh_depth; /* has tree real underlaying blocks? */ + __le32 eh_generation; /* generation of the tree */ +}; + +#define EXT4_EXT_MAGIC cpu_to_le16(0xf30a) + +/* + * array of ext4_ext_path contains path to some extent + * creation/lookup routines use it for traversal/splitting/etc + * truncate uses it to simulate recursive walking + */ +struct ext4_ext_path { + __u32 p_block; + __u16 p_depth; + struct ext4_extent *p_ext; + struct ext4_extent_idx *p_idx; + struct ext4_extent_header *p_hdr; + struct buffer_head *p_bh; +}; + +/* + * structure for external API + */ + +#define EXT4_EXT_CACHE_NO 0 +#define EXT4_EXT_CACHE_GAP 1 +#define EXT4_EXT_CACHE_EXTENT 2 + +/* + * to be called by ext4_ext_walk_space() + * negative retcode - error + * positive retcode - signal for ext4_ext_walk_space(), see below + * callback must return valid extent (passed or newly created) + */ +typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, + struct ext4_ext_cache *, + void *); + +#define EXT_CONTINUE 0 +#define EXT_BREAK 1 +#define EXT_REPEAT 2 + + +#define EXT_MAX_BLOCK 0xffffffff + + +#define EXT_FIRST_EXTENT(__hdr__) \ + ((struct ext4_extent *) (((char *) (__hdr__)) + \ + sizeof(struct ext4_extent_header))) +#define EXT_FIRST_INDEX(__hdr__) \ + ((struct ext4_extent_idx *) (((char *) (__hdr__)) + \ + sizeof(struct ext4_extent_header))) +#define EXT_HAS_FREE_INDEX(__path__) \ + (le16_to_cpu((__path__)->p_hdr->eh_entries) \ + < le16_to_cpu((__path__)->p_hdr->eh_max)) +#define EXT_LAST_EXTENT(__hdr__) \ + (EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1) +#define EXT_LAST_INDEX(__hdr__) \ + (EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_entries) - 1) +#define EXT_MAX_EXTENT(__hdr__) \ + (EXT_FIRST_EXTENT((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1) +#define EXT_MAX_INDEX(__hdr__) \ + (EXT_FIRST_INDEX((__hdr__)) + le16_to_cpu((__hdr__)->eh_max) - 1) + +static inline struct ext4_extent_header *ext_inode_hdr(struct inode *inode) +{ + return (struct ext4_extent_header *) EXT4_I(inode)->i_data; +} + +static inline struct ext4_extent_header *ext_block_hdr(struct buffer_head *bh) +{ + return (struct ext4_extent_header *) bh->b_data; +} + +static inline unsigned short ext_depth(struct inode *inode) +{ + return le16_to_cpu(ext_inode_hdr(inode)->eh_depth); +} + +static inline void ext4_ext_tree_changed(struct inode *inode) +{ + EXT4_I(inode)->i_ext_generation++; +} + +static inline void +ext4_ext_invalidate_cache(struct inode *inode) +{ + EXT4_I(inode)->i_cached_extent.ec_type = EXT4_EXT_CACHE_NO; +} + +extern int ext4_extent_tree_init(handle_t *, struct inode *); +extern int ext4_ext_calc_credits_for_insert(struct inode *, struct ext4_ext_path *); +extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *); +extern int ext4_ext_walk_space(struct inode *, unsigned long, unsigned long, ext_prepare_callback, void *); +extern struct ext4_ext_path * ext4_ext_find_extent(struct inode *, int, struct ext4_ext_path *); + +#endif /* _LINUX_EXT4_EXTENTS */ + diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h index 18a6ce9..40ce04a 100644 --- a/include/linux/ext4_fs_i.h +++ b/include/linux/ext4_fs_i.h @@ -65,6 +65,16 @@ struct ext4_block_alloc_info { #define rsv_end rsv_window._rsv_end /* + * storage for cached extent + */ +struct ext4_ext_cache { + __u32 ec_start; + __u32 ec_block; + __u32 ec_len; /* must be 32bit to return holes */ + __u32 ec_type; +}; + +/* * third extended file system inode data in memory */ struct ext4_inode_info { @@ -142,6 +152,9 @@ struct ext4_inode_info { */ struct mutex truncate_mutex; struct inode vfs_inode; + + unsigned long i_ext_generation; + struct ext4_ext_cache i_cached_extent; }; #endif /* _LINUX_EXT4_FS_I */ diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h index ce4856d..ce7a844 100644 --- a/include/linux/ext4_fs_sb.h +++ b/include/linux/ext4_fs_sb.h @@ -78,6 +78,16 @@ struct ext4_sb_info { char *s_qf_names[MAXQUOTAS]; /* Names of quota files with journalled quota */ int s_jquota_fmt; /* Format of quota to use */ #endif + +#ifdef EXTENTS_STATS + /* ext4 extents stats */ + unsigned long s_ext_min; + unsigned long s_ext_max; + unsigned long s_depth_max; + spinlock_t s_ext_stats_lock; + unsigned long s_ext_blocks; + unsigned long s_ext_extents; +#endif }; #endif /* _LINUX_EXT4_FS_SB */ diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h index 99d3755..aa273f0 100644 --- a/include/linux/ext4_jbd2.h +++ b/include/linux/ext4_jbd2.h @@ -26,9 +26,14 @@ * * We may have to touch one inode, one bitmap buffer, up to three * indirection blocks, the group and superblock summaries, and the data - * block to complete the transaction. */ + * block to complete the transaction. + * + * For extents-enabled fs we may have to allocate and modify upto + * 5 levels of tree + root which is stored in inode. */ -#define EXT4_SINGLEDATA_TRANS_BLOCKS 8U +#define EXT4_SINGLEDATA_TRANS_BLOCKS(sb) \ + (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS) \ + || test_opt(sb, EXTENTS) ? 27U : 8U) /* Extended attribute operations touch at most two data buffers, * two bitmap buffers, and two group summaries, in addition to the inode @@ -42,7 +47,7 @@ * superblock only gets updated once, of course, so don't bother * counting that again for the quota updates. */ -#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS + \ +#define EXT4_DATA_TRANS_BLOCKS(sb) (EXT4_SINGLEDATA_TRANS_BLOCKS(sb) + \ EXT4_XATTR_TRANS_BLOCKS - 2 + \ 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) @@ -78,9 +83,9 @@ /* Amount of blocks needed for quota insert/delete - we do some block writes * but inode, sb and group updates are done only once */ #define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ - (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) + (EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_INIT_REWRITE) : 0) #define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ - (EXT4_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) + (EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_DEL_REWRITE) : 0) #else #define EXT4_QUOTA_TRANS_BLOCKS(sb) 0 #define EXT4_QUOTA_INIT_BLOCKS(sb) 0 -- cgit v0.10.2 From 3a5b2ecdd1fa63a8f25bd769223bc1c2564ce45d Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:21:05 -0700 Subject: [PATCH] ext4: switch fsblk to sector_t Redefine ext3 in-kernel filesystem block type (ext3_fsblk_t) from unsigned long to sector_t, to allow kernel to handle >32 bit ext3 blocks. Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index e9e9844..aa33ff2 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -147,7 +147,7 @@ restart: rsv = list_entry(n, struct ext4_reserve_window_node, rsv_node); if (verbose) printk("reservation window 0x%p " - "start: %lu, end: %lu\n", + "start: "E3FSBLK", end: "E3FSBLK"\n", rsv, rsv->rsv_start, rsv->rsv_end); if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) { printk("Bad reservation %p (start >= end)\n", @@ -443,10 +443,7 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, do_more: overflow = 0; - block_group = (block - le32_to_cpu(es->s_first_data_block)) / - EXT4_BLOCKS_PER_GROUP(sb); - bit = (block - le32_to_cpu(es->s_first_data_block)) % - EXT4_BLOCKS_PER_GROUP(sb); + ext4_get_group_no_and_offset(sb, block, &block_group, &bit); /* * Check to see if we are freeing blocks across a group * boundary. @@ -1404,7 +1401,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, { struct buffer_head *bitmap_bh = NULL; struct buffer_head *gdp_bh; - int group_no; + unsigned long group_no; int goal_group; ext4_grpblk_t grp_target_blk; /* blockgroup relative goal block */ ext4_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ @@ -1467,8 +1464,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, if (goal < le32_to_cpu(es->s_first_data_block) || goal >= le32_to_cpu(es->s_blocks_count)) goal = le32_to_cpu(es->s_first_data_block); - group_no = (goal - le32_to_cpu(es->s_first_data_block)) / - EXT4_BLOCKS_PER_GROUP(sb); + ext4_get_group_no_and_offset(sb, goal, &group_no, &grp_target_blk); goal_group = group_no; retry_alloc: gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); @@ -1485,8 +1481,6 @@ retry_alloc: my_rsv = NULL; if (free_blocks > 0) { - grp_target_blk = ((goal - le32_to_cpu(es->s_first_data_block)) % - EXT4_BLOCKS_PER_GROUP(sb)); bitmap_bh = read_block_bitmap(sb, group_no); if (!bitmap_bh) goto io_error; @@ -1613,7 +1607,7 @@ allocated: if (ret_block + num - 1 >= le32_to_cpu(es->s_blocks_count)) { ext4_error(sb, "ext4_new_block", "block("E3FSBLK") >= blocks count(%d) - " - "block_group = %d, es == %p ", ret_block, + "block_group = %lu, es == %p ", ret_block, le32_to_cpu(es->s_blocks_count), group_no, es); goto out; } @@ -1733,9 +1727,10 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) static inline int block_in_use(ext4_fsblk_t block, struct super_block *sb, unsigned char *map) { - return ext4_test_bit ((block - - le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) % - EXT4_BLOCKS_PER_GROUP(sb), map); + ext4_grpblk_t offset; + + ext4_get_group_no_and_offset(sb, block, NULL, &offset); + return ext4_test_bit (offset, map); } static inline int test_root(int a, int b) diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index e17a6c9..94e1bb4 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -23,7 +23,7 @@ #include #include #include - +#include #include #include "xattr.h" @@ -274,7 +274,8 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter); avefreei = freei / ngroups; freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter); - avefreeb = freeb / ngroups; + avefreeb = freeb; + sector_div(avefreeb, ngroups); ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter); if ((parent == sb->s_root->d_inode) || @@ -303,13 +304,15 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) goto fallback; } - blocks_per_dir = (le32_to_cpu(es->s_blocks_count) - freeb) / ndirs; + blocks_per_dir = le32_to_cpu(es->s_blocks_count) - freeb; + sector_div(blocks_per_dir, ndirs); max_dirs = ndirs / ngroups + inodes_per_group / 16; min_inodes = avefreei - inodes_per_group / 4; min_blocks = avefreeb - EXT4_BLOCKS_PER_GROUP(sb) / 4; - max_debt = EXT4_BLOCKS_PER_GROUP(sb) / max(blocks_per_dir, (ext4_fsblk_t)BLOCK_COST); + max_debt = EXT4_BLOCKS_PER_GROUP(sb); + sector_div(max_debt, max(blocks_per_dir, (ext4_fsblk_t)BLOCK_COST)); if (max_debt * INODE_COST > inodes_per_group) max_debt = inodes_per_group / INODE_COST; if (max_debt > 255) diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 5b2828d..c60bfed 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -36,7 +36,7 @@ static int verify_group_input(struct super_block *sb, le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; ext4_fsblk_t metaend = start + overhead; struct buffer_head *bh = NULL; - ext4_grpblk_t free_blocks_count; + ext4_grpblk_t free_blocks_count, offset; int err = -EINVAL; input->free_blocks_count = free_blocks_count = @@ -49,13 +49,13 @@ static int verify_group_input(struct super_block *sb, "no-super", input->group, input->blocks_count, free_blocks_count, input->reserved_blocks); + ext4_get_group_no_and_offset(sb, start, NULL, &offset); if (group != sbi->s_groups_count) ext4_warning(sb, __FUNCTION__, "Cannot add at group %u (only %lu groups)", input->group, sbi->s_groups_count); - else if ((start - le32_to_cpu(es->s_first_data_block)) % - EXT4_BLOCKS_PER_GROUP(sb)) - ext4_warning(sb, __FUNCTION__, "Last group not full"); + else if (offset != 0) + ext4_warning(sb, __FUNCTION__, "Last group not full"); else if (input->reserved_blocks > input->blocks_count / 5) ext4_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", input->reserved_blocks); @@ -945,7 +945,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, if (n_blocks_count > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { printk(KERN_ERR "EXT4-fs: filesystem on %s:" - " too large to resize to %lu blocks safely\n", + " too large to resize to "E3FSBLK" blocks safely\n", sb->s_id, n_blocks_count); if (sizeof(sector_t) < 8) ext4_warning(sb, __FUNCTION__, @@ -960,8 +960,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, } /* Handle the remaining blocks in the last group only. */ - last = (o_blocks_count - le32_to_cpu(es->s_first_data_block)) % - EXT4_BLOCKS_PER_GROUP(sb); + ext4_get_group_no_and_offset(sb, o_blocks_count, NULL, &last); if (last == 0) { ext4_warning(sb, __FUNCTION__, diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 69f87525..1d12e4f 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1433,8 +1433,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) * block sizes. We need to calculate the offset from buffer start. */ if (blocksize != EXT4_MIN_BLOCK_SIZE) { - logic_sb_block = (sb_block * EXT4_MIN_BLOCK_SIZE) / blocksize; - offset = (sb_block * EXT4_MIN_BLOCK_SIZE) % blocksize; + logic_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; + offset = sector_div(logic_sb_block, blocksize); } else { logic_sb_block = sb_block; } @@ -1539,8 +1539,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) brelse (bh); sb_set_blocksize(sb, blocksize); - logic_sb_block = (sb_block * EXT4_MIN_BLOCK_SIZE) / blocksize; - offset = (sb_block * EXT4_MIN_BLOCK_SIZE) % blocksize; + logic_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; + offset = sector_div(logic_sb_block, blocksize); bh = sb_bread(sb, logic_sb_block); if (!bh) { printk(KERN_ERR diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index b61181aad..e952c6d 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -17,6 +17,7 @@ #define _LINUX_EXT4_FS_H #include +#include #include /* @@ -750,6 +751,27 @@ ext4_group_first_block_no(struct super_block *sb, unsigned long group_no) #define ERR_BAD_DX_DIR -75000 /* + * This function calculate the block group number and offset, + * given a block number + */ + +static inline void ext4_get_group_no_and_offset(struct super_block * sb, + ext4_fsblk_t blocknr, unsigned long* blockgrpp, + ext4_grpblk_t *offsetp) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + ext4_grpblk_t offset; + + blocknr = blocknr - le32_to_cpu(es->s_first_data_block); + offset = sector_div(blocknr, EXT4_BLOCKS_PER_GROUP(sb)); + if (offsetp) + *offsetp = offset; + if (blockgrpp) + *blockgrpp = blocknr; + +} + +/* * Function prototypes */ @@ -762,6 +784,10 @@ ext4_group_first_block_no(struct super_block *sb, unsigned long group_no) # define NORET_AND noreturn, /* balloc.c */ +extern unsigned int ext4_block_group(struct super_block *sb, + ext4_fsblk_t blocknr); +extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb, + ext4_fsblk_t blocknr); extern int ext4_bg_has_super(struct super_block *sb, int group); extern unsigned long ext4_bg_num_gdb(struct super_block *sb, int group); extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode, diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h index 40ce04a..b2ccd98 100644 --- a/include/linux/ext4_fs_i.h +++ b/include/linux/ext4_fs_i.h @@ -25,9 +25,13 @@ typedef int ext4_grpblk_t; /* data type for filesystem-wide blocks number */ -typedef unsigned long ext4_fsblk_t; +typedef sector_t ext4_fsblk_t; +#if BITS_PER_LONG == 64 #define E3FSBLK "%lu" +#else +#define E3FSBLK "%llu" +#endif struct ext4_reserve_window { ext4_fsblk_t _rsv_start; /* First byte reserved */ -- cgit v0.10.2 From f65e6fba163dfd0f962efb7d8f5528b6872e2b15 Mon Sep 17 00:00:00 2001 From: Alex Tomas Date: Wed, 11 Oct 2006 01:21:05 -0700 Subject: [PATCH] ext4: 48bit physical block number support in extents Signed-off-by: Alex Tomas Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index f67b2ef..4a13b56 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -44,6 +44,44 @@ #include +/* this macro combines low and hi parts of phys. blocknr into ext4_fsblk_t */ +static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) +{ + ext4_fsblk_t block; + + block = le32_to_cpu(ex->ee_start); + if (sizeof(ext4_fsblk_t) > 4) + block |= ((ext4_fsblk_t) le16_to_cpu(ex->ee_start_hi) << 31) << 1; + return block; +} + +/* this macro combines low and hi parts of phys. blocknr into ext4_fsblk_t */ +static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) +{ + ext4_fsblk_t block; + + block = le32_to_cpu(ix->ei_leaf); + if (sizeof(ext4_fsblk_t) > 4) + block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1; + return block; +} + +/* the routine stores large phys. blocknr into extent breaking it into parts */ +static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) +{ + ex->ee_start = cpu_to_le32((unsigned long) (pb & 0xffffffff)); + if (sizeof(ext4_fsblk_t) > 4) + ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); +} + +/* the routine stores large phys. blocknr into index breaking it into parts */ +static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) +{ + ix->ei_leaf = cpu_to_le32((unsigned long) (pb & 0xffffffff)); + if (sizeof(ext4_fsblk_t) > 4) + ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); +} + static int ext4_ext_check_header(const char *function, struct inode *inode, struct ext4_extent_header *eh) { @@ -124,13 +162,13 @@ static int ext4_ext_dirty(handle_t *handle, struct inode *inode, return err; } -static int ext4_ext_find_goal(struct inode *inode, +static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, struct ext4_ext_path *path, - unsigned long block) + ext4_fsblk_t block) { struct ext4_inode_info *ei = EXT4_I(inode); - unsigned long bg_start; - unsigned long colour; + ext4_fsblk_t bg_start; + ext4_grpblk_t colour; int depth; if (path) { @@ -139,8 +177,7 @@ static int ext4_ext_find_goal(struct inode *inode, /* try to predict block placement */ if ((ex = path[depth].p_ext)) - return le32_to_cpu(ex->ee_start) - + (block - le32_to_cpu(ex->ee_block)); + return ext_pblock(ex)+(block-le32_to_cpu(ex->ee_block)); /* it looks index is empty * try to find starting from index itself */ @@ -156,12 +193,12 @@ static int ext4_ext_find_goal(struct inode *inode, return bg_start + colour + block; } -static int +static ext4_fsblk_t ext4_ext_new_block(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *ex, int *err) { - int goal, newblock; + ext4_fsblk_t goal, newblock; goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block)); newblock = ext4_new_block(handle, inode, goal, err); @@ -230,13 +267,13 @@ static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) ext_debug("path:"); for (k = 0; k <= l; k++, path++) { if (path->p_idx) { - ext_debug(" %d->%d", le32_to_cpu(path->p_idx->ei_block), - le32_to_cpu(path->p_idx->ei_leaf)); + ext_debug(" %d->"E3FSBLK, le32_to_cpu(path->p_idx->ei_block), + idx_pblock(path->p_idx)); } else if (path->p_ext) { - ext_debug(" %d:%d:%d", + ext_debug(" %d:%d:"E3FSBLK" ", le32_to_cpu(path->p_ext->ee_block), le16_to_cpu(path->p_ext->ee_len), - le32_to_cpu(path->p_ext->ee_start)); + ext_pblock(path->p_ext)); } else ext_debug(" []"); } @@ -257,9 +294,8 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) ex = EXT_FIRST_EXTENT(eh); for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) { - ext_debug("%d:%d:%d ", le32_to_cpu(ex->ee_block), - le16_to_cpu(ex->ee_len), - le32_to_cpu(ex->ee_start)); + ext_debug("%d:%d:"E3FSBLK" ", le32_to_cpu(ex->ee_block), + le16_to_cpu(ex->ee_len), ext_pblock(ex)); } ext_debug("\n"); } @@ -308,8 +344,8 @@ ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int bloc } path->p_idx = l - 1; - ext_debug(" -> %d->%d ", le32_to_cpu(path->p_idx->ei_block), - le32_to_cpu(path->p_idx->ei_leaf)); + ext_debug(" -> %d->%lld ", le32_to_cpu(path->p_idx->ei_block), + idx_block(path->p_idx)); #ifdef CHECK_BINSEARCH { @@ -374,10 +410,10 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) } path->p_ext = l - 1; - ext_debug(" -> %d:%d:%d ", + ext_debug(" -> %d:"E3FSBLK":%d ", le32_to_cpu(path->p_ext->ee_block), - le32_to_cpu(path->p_ext->ee_start), - le16_to_cpu(path->p_ext->ee_len)); + ext_pblock(path->p_ext), + le16_to_cpu(path->p_ext->ee_len)); #ifdef CHECK_BINSEARCH { @@ -442,7 +478,7 @@ ext4_ext_find_extent(struct inode *inode, int block, struct ext4_ext_path *path) ext_debug("depth %d: num %d, max %d\n", ppos, le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); ext4_ext_binsearch_idx(inode, path + ppos, block); - path[ppos].p_block = le32_to_cpu(path[ppos].p_idx->ei_leaf); + path[ppos].p_block = idx_pblock(path[ppos].p_idx); path[ppos].p_depth = i; path[ppos].p_ext = NULL; @@ -489,7 +525,7 @@ err: */ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, struct ext4_ext_path *curp, - int logical, int ptr) + int logical, ext4_fsblk_t ptr) { struct ext4_extent_idx *ix; int len, err; @@ -524,7 +560,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, } ix->ei_block = cpu_to_le32(logical); - ix->ei_leaf = cpu_to_le32(ptr); + ext4_idx_store_pblock(ix, ptr); curp->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(curp->p_hdr->eh_entries)+1); BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries) @@ -556,9 +592,9 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, struct ext4_extent_idx *fidx; struct ext4_extent *ex; int i = at, k, m, a; - unsigned long newblock, oldblock; + ext4_fsblk_t newblock, oldblock; __le32 border; - int *ablocks = NULL; /* array of allocated blocks */ + ext4_fsblk_t *ablocks = NULL; /* array of allocated blocks */ int err = 0; /* make decision: where to split? */ @@ -591,10 +627,10 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, * we need this to handle errors and free blocks * upon them */ - ablocks = kmalloc(sizeof(unsigned long) * depth, GFP_NOFS); + ablocks = kmalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS); if (!ablocks) return -ENOMEM; - memset(ablocks, 0, sizeof(unsigned long) * depth); + memset(ablocks, 0, sizeof(ext4_fsblk_t) * depth); /* allocate all needed blocks */ ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); @@ -633,9 +669,9 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, path[depth].p_ext++; while (path[depth].p_ext <= EXT_MAX_EXTENT(path[depth].p_hdr)) { - ext_debug("move %d:%d:%d in new leaf %lu\n", + ext_debug("move %d:"E3FSBLK":%d in new leaf "E3FSBLK"\n", le32_to_cpu(path[depth].p_ext->ee_block), - le32_to_cpu(path[depth].p_ext->ee_start), + ext_pblock(path[depth].p_ext), le16_to_cpu(path[depth].p_ext->ee_len), newblock); /*memmove(ex++, path[depth].p_ext++, @@ -679,7 +715,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, while (k--) { oldblock = newblock; newblock = ablocks[--a]; - bh = sb_getblk(inode->i_sb, newblock); + bh = sb_getblk(inode->i_sb, (ext4_fsblk_t)newblock); if (!bh) { err = -EIO; goto cleanup; @@ -696,9 +732,9 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, neh->eh_depth = cpu_to_le16(depth - i); fidx = EXT_FIRST_INDEX(neh); fidx->ei_block = border; - fidx->ei_leaf = cpu_to_le32(oldblock); + ext4_idx_store_pblock(fidx, oldblock); - ext_debug("int.index at %d (block %lu): %lu -> %lu\n", i, + ext_debug("int.index at %d (block "E3FSBLK"): %lu -> "E3FSBLK"\n", i, newblock, (unsigned long) le32_to_cpu(border), oldblock); /* copy indexes */ @@ -710,9 +746,9 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, BUG_ON(EXT_MAX_INDEX(path[i].p_hdr) != EXT_LAST_INDEX(path[i].p_hdr)); while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) { - ext_debug("%d: move %d:%d in new index %lu\n", i, + ext_debug("%d: move %d:%d in new index "E3FSBLK"\n", i, le32_to_cpu(path[i].p_idx->ei_block), - le32_to_cpu(path[i].p_idx->ei_leaf), + idx_pblock(path[i].p_idx), newblock); /*memmove(++fidx, path[i].p_idx++, sizeof(struct ext4_extent_idx)); @@ -791,7 +827,7 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, struct ext4_extent_header *neh; struct ext4_extent_idx *fidx; struct buffer_head *bh; - unsigned long newblock; + ext4_fsblk_t newblock; int err = 0; newblock = ext4_ext_new_block(handle, inode, path, newext, &err); @@ -839,13 +875,13 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr); /* FIXME: it works, but actually path[0] can be index */ curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block; - curp->p_idx->ei_leaf = cpu_to_le32(newblock); + ext4_idx_store_pblock(curp->p_idx, newblock); neh = ext_inode_hdr(inode); fidx = EXT_FIRST_INDEX(neh); - ext_debug("new root: num %d(%d), lblock %d, ptr %d\n", + ext_debug("new root: num %d(%d), lblock %d, ptr "E3FSBLK"\n", le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max), - le32_to_cpu(fidx->ei_block), le32_to_cpu(fidx->ei_leaf)); + le32_to_cpu(fidx->ei_block), idx_pblock(fidx)); neh->eh_depth = cpu_to_le16(path->p_depth + 1); err = ext4_ext_dirty(handle, inode, curp); @@ -1042,7 +1078,6 @@ static int inline ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, struct ext4_extent *ex2) { - /* FIXME: 48bit support */ if (le32_to_cpu(ex1->ee_block) + le16_to_cpu(ex1->ee_len) != le32_to_cpu(ex2->ee_block)) return 0; @@ -1052,8 +1087,7 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, return 0; #endif - if (le32_to_cpu(ex1->ee_start) + le16_to_cpu(ex1->ee_len) - == le32_to_cpu(ex2->ee_start)) + if (ext_pblock(ex1) + le16_to_cpu(ex1->ee_len) == ext_pblock(ex2)) return 1; return 0; } @@ -1080,11 +1114,10 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, /* try to insert block into found extent and return */ if (ex && ext4_can_extents_be_merged(inode, ex, newext)) { - ext_debug("append %d block to %d:%d (from %d)\n", + ext_debug("append %d block to %d:%d (from "E3FSBLK")\n", le16_to_cpu(newext->ee_len), le32_to_cpu(ex->ee_block), - le16_to_cpu(ex->ee_len), - le32_to_cpu(ex->ee_start)); + le16_to_cpu(ex->ee_len), ext_pblock(ex)); if ((err = ext4_ext_get_access(handle, inode, path + depth))) return err; ex->ee_len = cpu_to_le16(le16_to_cpu(ex->ee_len) @@ -1140,9 +1173,9 @@ has_space: if (!nearex) { /* there is no extent in this leaf, create first one */ - ext_debug("first extent in the leaf: %d:%d:%d\n", + ext_debug("first extent in the leaf: %d:"E3FSBLK":%d\n", le32_to_cpu(newext->ee_block), - le32_to_cpu(newext->ee_start), + ext_pblock(newext), le16_to_cpu(newext->ee_len)); path[depth].p_ext = EXT_FIRST_EXTENT(eh); } else if (le32_to_cpu(newext->ee_block) @@ -1152,10 +1185,10 @@ has_space: len = EXT_MAX_EXTENT(eh) - nearex; len = (len - 1) * sizeof(struct ext4_extent); len = len < 0 ? 0 : len; - ext_debug("insert %d:%d:%d after: nearest 0x%p, " + ext_debug("insert %d:"E3FSBLK":%d after: nearest 0x%p, " "move %d from 0x%p to 0x%p\n", le32_to_cpu(newext->ee_block), - le32_to_cpu(newext->ee_start), + ext_pblock(newext), le16_to_cpu(newext->ee_len), nearex, len, nearex + 1, nearex + 2); memmove(nearex + 2, nearex + 1, len); @@ -1165,10 +1198,10 @@ has_space: BUG_ON(newext->ee_block == nearex->ee_block); len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext4_extent); len = len < 0 ? 0 : len; - ext_debug("insert %d:%d:%d before: nearest 0x%p, " + ext_debug("insert %d:"E3FSBLK":%d before: nearest 0x%p, " "move %d from 0x%p to 0x%p\n", le32_to_cpu(newext->ee_block), - le32_to_cpu(newext->ee_start), + ext_pblock(newext), le16_to_cpu(newext->ee_len), nearex, len, nearex + 1, nearex + 2); memmove(nearex + 1, nearex, len); @@ -1179,9 +1212,8 @@ has_space: nearex = path[depth].p_ext; nearex->ee_block = newext->ee_block; nearex->ee_start = newext->ee_start; + nearex->ee_start_hi = newext->ee_start_hi; nearex->ee_len = newext->ee_len; - /* FIXME: support for large fs */ - nearex->ee_start_hi = 0; merge: /* try to merge extents to the right */ @@ -1290,7 +1322,7 @@ int ext4_ext_walk_space(struct inode *inode, unsigned long block, } else { cbex.ec_block = le32_to_cpu(ex->ee_block); cbex.ec_len = le16_to_cpu(ex->ee_len); - cbex.ec_start = le32_to_cpu(ex->ee_start); + cbex.ec_start = ext_pblock(ex); cbex.ec_type = EXT4_EXT_CACHE_EXTENT; } @@ -1398,13 +1430,13 @@ ext4_ext_in_cache(struct inode *inode, unsigned long block, cex->ec_type != EXT4_EXT_CACHE_EXTENT); if (block >= cex->ec_block && block < cex->ec_block + cex->ec_len) { ex->ee_block = cpu_to_le32(cex->ec_block); - ex->ee_start = cpu_to_le32(cex->ec_start); + ext4_ext_store_pblock(ex, cex->ec_start); ex->ee_len = cpu_to_le16(cex->ec_len); - ext_debug("%lu cached by %lu:%lu:%lu\n", + ext_debug("%lu cached by %lu:%lu:"E3FSBLK"\n", (unsigned long) block, (unsigned long) cex->ec_block, (unsigned long) cex->ec_len, - (unsigned long) cex->ec_start); + cex->ec_start); return cex->ec_type; } @@ -1422,18 +1454,18 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, { struct buffer_head *bh; int err; - unsigned long leaf; + ext4_fsblk_t leaf; /* free index block */ path--; - leaf = le32_to_cpu(path->p_idx->ei_leaf); + leaf = idx_pblock(path->p_idx); BUG_ON(path->p_hdr->eh_entries == 0); if ((err = ext4_ext_get_access(handle, inode, path))) return err; path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1); if ((err = ext4_ext_dirty(handle, inode, path))) return err; - ext_debug("index is empty, remove it, free block %lu\n", leaf); + ext_debug("index is empty, remove it, free block "E3FSBLK"\n", leaf); bh = sb_find_get_block(inode->i_sb, leaf); ext4_forget(handle, 1, inode, bh, leaf); ext4_free_blocks(handle, inode, leaf, 1); @@ -1515,10 +1547,11 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, if (from >= le32_to_cpu(ex->ee_block) && to == le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - 1) { /* tail removal */ - unsigned long num, start; + unsigned long num; + ext4_fsblk_t start; num = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - from; - start = le32_to_cpu(ex->ee_start) + le16_to_cpu(ex->ee_len) - num; - ext_debug("free last %lu blocks starting %lu\n", num, start); + start = ext_pblock(ex) + le16_to_cpu(ex->ee_len) - num; + ext_debug("free last %lu blocks starting "E3FSBLK"\n", num, start); for (i = 0; i < num; i++) { bh = sb_find_get_block(inode->i_sb, start + i); ext4_forget(handle, 0, inode, bh, start + i); @@ -1621,7 +1654,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, if (num == 0) { /* this extent is removed entirely mark slot unused */ - ex->ee_start = 0; + ext4_ext_store_pblock(ex, 0); eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); } @@ -1632,8 +1665,8 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, if (err) goto out; - ext_debug("new extent: %u:%u:%u\n", block, num, - le32_to_cpu(ex->ee_start)); + ext_debug("new extent: %u:%u:"E3FSBLK"\n", block, num, + ext_pblock(ex)); ex--; ex_ee_block = le32_to_cpu(ex->ee_block); ex_ee_len = le16_to_cpu(ex->ee_len); @@ -1748,11 +1781,11 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) path[i].p_idx); if (ext4_ext_more_to_rm(path + i)) { /* go to the next level */ - ext_debug("move to level %d (block %d)\n", - i + 1, le32_to_cpu(path[i].p_idx->ei_leaf)); + ext_debug("move to level %d (block "E3FSBLK")\n", + i + 1, idx_pblock(path[i].p_idx)); memset(path + i + 1, 0, sizeof(*path)); path[i+1].p_bh = - sb_bread(sb, le32_to_cpu(path[i].p_idx->ei_leaf)); + sb_bread(sb, idx_pblock(path[i].p_idx)); if (!path[i+1].p_bh) { /* should we reset i_size? */ err = -EIO; @@ -1851,13 +1884,15 @@ void ext4_ext_release(struct super_block *sb) #endif } -int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, sector_t iblock, +int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t iblock, unsigned long max_blocks, struct buffer_head *bh_result, int create, int extend_disksize) { struct ext4_ext_path *path = NULL; struct ext4_extent newex, *ex; - int goal, newblock, err = 0, depth; + ext4_fsblk_t goal, newblock; + int err = 0, depth; unsigned long allocated = 0; __clear_bit(BH_New, &bh_result->b_state); @@ -1878,7 +1913,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, sector_t iblock, /* block is already allocated */ newblock = iblock - le32_to_cpu(newex.ee_block) - + le32_to_cpu(newex.ee_start); + + ext_pblock(&newex); /* number of remain blocks in the extent */ allocated = le16_to_cpu(newex.ee_len) - (iblock - le32_to_cpu(newex.ee_block)); @@ -1907,14 +1942,14 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, sector_t iblock, if ((ex = path[depth].p_ext)) { unsigned long ee_block = le32_to_cpu(ex->ee_block); - unsigned long ee_start = le32_to_cpu(ex->ee_start); + ext4_fsblk_t ee_start = ext_pblock(ex); unsigned short ee_len = le16_to_cpu(ex->ee_len); /* if found exent covers block, simple return it */ if (iblock >= ee_block && iblock < ee_block + ee_len) { newblock = iblock - ee_block + ee_start; /* number of remain blocks in the extent */ allocated = ee_len - (iblock - ee_block); - ext_debug("%d fit into %lu:%d -> %d\n", (int) iblock, + ext_debug("%d fit into %lu:%d -> "E3FSBLK"\n", (int) iblock, ee_block, ee_len, newblock); ext4_ext_put_in_cache(inode, ee_block, ee_len, ee_start, EXT4_EXT_CACHE_EXTENT); @@ -1944,12 +1979,12 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, sector_t iblock, newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err); if (!newblock) goto out2; - ext_debug("allocate new block: goal %d, found %d/%lu\n", + ext_debug("allocate new block: goal "E3FSBLK", found "E3FSBLK"/%lu\n", goal, newblock, allocated); /* try to insert new extent into found leaf and return */ newex.ee_block = cpu_to_le32(iblock); - newex.ee_start = cpu_to_le32(newblock); + ext4_ext_store_pblock(&newex, newblock); newex.ee_len = cpu_to_le16(allocated); err = ext4_ext_insert_extent(handle, inode, path, &newex); if (err) @@ -1959,7 +1994,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, sector_t iblock, EXT4_I(inode)->i_disksize = inode->i_size; /* previous routine could use block we allocated */ - newblock = le32_to_cpu(newex.ee_start); + newblock = ext_pblock(&newex); __set_bit(BH_New, &bh_result->b_state); ext4_ext_put_in_cache(inode, iblock, allocated, newblock, diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h index 8029879..facc766 100644 --- a/include/linux/ext4_fs_extents.h +++ b/include/linux/ext4_fs_extents.h @@ -108,7 +108,7 @@ struct ext4_extent_header { * truncate uses it to simulate recursive walking */ struct ext4_ext_path { - __u32 p_block; + ext4_fsblk_t p_block; __u16 p_depth; struct ext4_extent *p_ext; struct ext4_extent_idx *p_idx; diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h index b2ccd98..2bed0ef 100644 --- a/include/linux/ext4_fs_i.h +++ b/include/linux/ext4_fs_i.h @@ -72,10 +72,10 @@ struct ext4_block_alloc_info { * storage for cached extent */ struct ext4_ext_cache { - __u32 ec_start; - __u32 ec_block; - __u32 ec_len; /* must be 32bit to return holes */ - __u32 ec_type; + ext4_fsblk_t ec_start; + __u32 ec_block; + __u32 ec_len; /* must be 32bit to return holes */ + __u32 ec_type; }; /* -- cgit v0.10.2 From 471d4011a9862efff02094388b8fe8cd67683c38 Mon Sep 17 00:00:00 2001 From: Suparna Bhattacharya Date: Wed, 11 Oct 2006 01:21:06 -0700 Subject: [PATCH] ext4: uninitialised extent handling Make it possible to add file preallocation support in future as an RO_COMPAT feature by recognizing uninitialized extents as holes and limiting extent length to keep the top bit of ee_len free for marking uninitialized extents. Signed-off-by: Suparna Bhattacharya Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 4a13b56..3252606 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1082,6 +1082,13 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, != le32_to_cpu(ex2->ee_block)) return 0; + /* + * To allow future support for preallocated extents to be added + * as an RO_COMPAT feature, refuse to merge to extents if + * can result in the top bit of ee_len being set + */ + if (le16_to_cpu(ex1->ee_len) + le16_to_cpu(ex2->ee_len) > EXT_MAX_LEN) + return 0; #ifdef AGRESSIVE_TEST if (le16_to_cpu(ex1->ee_len) >= 4) return 0; @@ -1944,6 +1951,15 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, unsigned long ee_block = le32_to_cpu(ex->ee_block); ext4_fsblk_t ee_start = ext_pblock(ex); unsigned short ee_len = le16_to_cpu(ex->ee_len); + + /* + * Allow future support for preallocated extents to be added + * as an RO_COMPAT feature: + * Uninitialized extents are treated as holes, except that + * we avoid (fail) allocating new blocks during a write. + */ + if (ee_len > EXT_MAX_LEN) + goto out2; /* if found exent covers block, simple return it */ if (iblock >= ee_block && iblock < ee_block + ee_len) { newblock = iblock - ee_block + ee_start; diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h index facc766..0eba0ac 100644 --- a/include/linux/ext4_fs_extents.h +++ b/include/linux/ext4_fs_extents.h @@ -141,6 +141,8 @@ typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, #define EXT_MAX_BLOCK 0xffffffff +#define EXT_MAX_LEN ((1UL << 15) - 1) + #define EXT_FIRST_EXTENT(__hdr__) \ ((struct ext4_extent *) (((char *) (__hdr__)) + \ -- cgit v0.10.2 From d0d856e8bd6e697cb44b2b4dd038f3bec576a70e Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 11 Oct 2006 01:21:07 -0700 Subject: [PATCH] ext4: clean up comments in ext4-extents patch Signed-off-by: Randy Dunlap Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 3252606..e06e937 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -44,7 +44,10 @@ #include -/* this macro combines low and hi parts of phys. blocknr into ext4_fsblk_t */ +/* + * ext_pblock: + * combine low and high parts of physical block number into ext4_fsblk_t + */ static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) { ext4_fsblk_t block; @@ -55,7 +58,10 @@ static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) return block; } -/* this macro combines low and hi parts of phys. blocknr into ext4_fsblk_t */ +/* + * idx_pblock: + * combine low and high parts of a leaf physical block number into ext4_fsblk_t + */ static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) { ext4_fsblk_t block; @@ -66,7 +72,11 @@ static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) return block; } -/* the routine stores large phys. blocknr into extent breaking it into parts */ +/* + * ext4_ext_store_pblock: + * stores a large physical block number into an extent struct, + * breaking it into parts + */ static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) { ex->ee_start = cpu_to_le32((unsigned long) (pb & 0xffffffff)); @@ -74,7 +84,11 @@ static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); } -/* the routine stores large phys. blocknr into index breaking it into parts */ +/* + * ext4_idx_store_pblock: + * stores a large physical block number into an index struct, + * breaking it into parts + */ static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) { ix->ei_leaf = cpu_to_le32((unsigned long) (pb & 0xffffffff)); @@ -179,8 +193,8 @@ static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, if ((ex = path[depth].p_ext)) return ext_pblock(ex)+(block-le32_to_cpu(ex->ee_block)); - /* it looks index is empty - * try to find starting from index itself */ + /* it looks like index is empty; + * try to find starting block from index itself */ if (path[depth].p_bh) return path[depth].p_bh->b_blocknr; } @@ -317,7 +331,8 @@ static void ext4_ext_drop_refs(struct ext4_ext_path *path) } /* - * binary search for closest index by given block + * ext4_ext_binsearch_idx: + * binary search for the closest index of the given block */ static void ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int block) @@ -375,7 +390,8 @@ ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int bloc } /* - * binary search for closest extent by given block + * ext4_ext_binsearch: + * binary search for closest extent of the given block */ static void ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) @@ -388,8 +404,8 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) if (eh->eh_entries == 0) { /* - * this leaf is empty yet: - * we get such a leaf in split/add case + * this leaf is empty: + * we get such a leaf in split/add case */ return; } @@ -520,8 +536,9 @@ err: } /* - * insert new index [logical;ptr] into the block at cupr - * it check where to insert: before curp or after curp + * ext4_ext_insert_index: + * insert new index [@logical;@ptr] into the block at @curp; + * check where to insert: before @curp or after @curp */ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, struct ext4_ext_path *curp, @@ -574,13 +591,14 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, } /* - * routine inserts new subtree into the path, using free index entry - * at depth 'at: - * - allocates all needed blocks (new leaf and all intermediate index blocks) - * - makes decision where to split - * - moves remaining extens and index entries (right to the split point) - * into the newly allocated blocks - * - initialize subtree + * ext4_ext_split: + * inserts new subtree into the path, using free index entry + * at depth @at: + * - allocates all needed blocks (new leaf and all intermediate index blocks) + * - makes decision where to split + * - moves remaining extents and index entries (right to the split point) + * into the newly allocated blocks + * - initializes subtree */ static int ext4_ext_split(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, @@ -598,14 +616,14 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, int err = 0; /* make decision: where to split? */ - /* FIXME: now desicion is simplest: at current extent */ + /* FIXME: now decision is simplest: at current extent */ - /* if current leaf will be splitted, then we should use + /* if current leaf will be split, then we should use * border from split point */ BUG_ON(path[depth].p_ext > EXT_MAX_EXTENT(path[depth].p_hdr)); if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) { border = path[depth].p_ext[1].ee_block; - ext_debug("leaf will be splitted." + ext_debug("leaf will be split." " next leaf starts at %d\n", le32_to_cpu(border)); } else { @@ -616,16 +634,16 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, } /* - * if error occurs, then we break processing - * and turn filesystem read-only. so, index won't + * If error occurs, then we break processing + * and mark filesystem read-only. index won't * be inserted and tree will be in consistent - * state. next mount will repair buffers too + * state. Next mount will repair buffers too. */ /* - * get array to track all allocated blocks - * we need this to handle errors and free blocks - * upon them + * Get array to track all allocated blocks. + * We need this to handle errors and free blocks + * upon them. */ ablocks = kmalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS); if (!ablocks) @@ -661,7 +679,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, neh->eh_depth = 0; ex = EXT_FIRST_EXTENT(neh); - /* move remain of path[depth] to the new leaf */ + /* move remainder of path[depth] to the new leaf */ BUG_ON(path[depth].p_hdr->eh_entries != path[depth].p_hdr->eh_max); /* start copy from next extent */ /* TODO: we could do it by single memmove */ @@ -813,11 +831,12 @@ cleanup: } /* - * routine implements tree growing procedure: - * - allocates new block - * - moves top-level data (index block or leaf) into the new block - * - initialize new top-level, creating index that points to the - * just created block + * ext4_ext_grow_indepth: + * implements tree growing procedure: + * - allocates new block + * - moves top-level data (index block or leaf) into the new block + * - initializes new top-level, creating index that points to the + * just created block */ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, @@ -892,8 +911,9 @@ out: } /* - * routine finds empty index and adds new leaf. if no free index found - * then it requests in-depth growing + * ext4_ext_create_new_leaf: + * finds empty index and adds new leaf. + * if no free index is found, then it requests in-depth growing. */ static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, @@ -912,8 +932,8 @@ repeat: curp--; } - /* we use already allocated block for index block - * so, subsequent data blocks should be contigoues */ + /* we use already allocated block for index block, + * so subsequent data blocks should be contiguous */ if (EXT_HAS_FREE_INDEX(curp)) { /* if we found index with free entry, then use that * entry: create all needed subtree and add new leaf */ @@ -943,12 +963,12 @@ repeat: } /* - * only first (depth 0 -> 1) produces free space - * in all other cases we have to split growed tree + * only first (depth 0 -> 1) produces free space; + * in all other cases we have to split the grown tree */ depth = ext_depth(inode); if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) { - /* now we need split */ + /* now we need to split */ goto repeat; } } @@ -958,10 +978,11 @@ out: } /* - * returns allocated block in subsequent extent or EXT_MAX_BLOCK - * NOTE: it consider block number from index entry as - * allocated block. thus, index entries have to be consistent - * with leafs + * ext4_ext_next_allocated_block: + * returns allocated block in subsequent extent or EXT_MAX_BLOCK. + * NOTE: it considers block number from index entry as + * allocated block. Thus, index entries have to be consistent + * with leaves. */ static unsigned long ext4_ext_next_allocated_block(struct ext4_ext_path *path) @@ -993,6 +1014,7 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) } /* + * ext4_ext_next_leaf_block: * returns first allocated block from next leaf or EXT_MAX_BLOCK */ static unsigned ext4_ext_next_leaf_block(struct inode *inode, @@ -1021,8 +1043,9 @@ static unsigned ext4_ext_next_leaf_block(struct inode *inode, } /* - * if leaf gets modified and modified extent is first in the leaf - * then we have to correct all indexes above + * ext4_ext_correct_indexes: + * if leaf gets modified and modified extent is first in the leaf, + * then we have to correct all indexes above. * TODO: do we need to correct tree in all cases? */ int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, @@ -1050,7 +1073,7 @@ int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, } /* - * TODO: we need correction if border is smaller then current one + * TODO: we need correction if border is smaller than current one */ k = depth - 1; border = path[depth].p_ext->ee_block; @@ -1085,7 +1108,7 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, /* * To allow future support for preallocated extents to be added * as an RO_COMPAT feature, refuse to merge to extents if - * can result in the top bit of ee_len being set + * this can result in the top bit of ee_len being set. */ if (le16_to_cpu(ex1->ee_len) + le16_to_cpu(ex2->ee_len) > EXT_MAX_LEN) return 0; @@ -1100,9 +1123,10 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, } /* - * this routine tries to merge requsted extent into the existing - * extent or inserts requested extent as new one into the tree, - * creating new leaf in no-space case + * ext4_ext_insert_extent: + * tries to merge requsted extent into the existing extent or + * inserts requested extent as new one into the tree, + * creating new leaf in the no-space case. */ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, @@ -1163,8 +1187,8 @@ repeat: } /* - * there is no free space in found leaf - * we're gonna add new leaf in the tree + * There is no free space in the found leaf. + * We're gonna add a new leaf in the tree. */ err = ext4_ext_create_new_leaf(handle, inode, path, newext); if (err) @@ -1377,7 +1401,8 @@ ext4_ext_put_in_cache(struct inode *inode, __u32 block, } /* - * this routine calculate boundaries of the gap requested block fits into + * ext4_ext_put_gap_in_cache: + * calculate boundaries of the gap that the requested block fits into * and cache this gap */ static inline void @@ -1452,9 +1477,10 @@ ext4_ext_in_cache(struct inode *inode, unsigned long block, } /* - * routine removes index from the index block - * it's used in truncate case only. thus all requests are for - * last index in the block only + * ext4_ext_rm_idx: + * removes index from the index block. + * It's used in truncate case only, thus all requests are for + * last index in the block only. */ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, struct ext4_ext_path *path) @@ -1480,11 +1506,12 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, } /* - * This routine returns max. credits extent tree can consume. + * ext4_ext_calc_credits_for_insert: + * This routine returns max. credits that the extent tree can consume. * It should be OK for low-performance paths like ->writepage() - * To allow many writing process to fit a single transaction, - * caller should calculate credits under truncate_mutex and - * pass actual path. + * To allow many writing processes to fit into a single transaction, + * the caller should calculate credits under truncate_mutex and + * pass the actual path. */ int inline ext4_ext_calc_credits_for_insert(struct inode *inode, struct ext4_ext_path *path) @@ -1500,9 +1527,9 @@ int inline ext4_ext_calc_credits_for_insert(struct inode *inode, } /* - * given 32bit logical block (4294967296 blocks), max. tree + * given 32-bit logical block (4294967296 blocks), max. tree * can be 4 levels in depth -- 4 * 340^4 == 53453440000. - * let's also add one more level for imbalance. + * Let's also add one more level for imbalance. */ depth = 5; @@ -1510,13 +1537,13 @@ int inline ext4_ext_calc_credits_for_insert(struct inode *inode, needed = 2; /* - * tree can be full, so it'd need to grow in depth: + * tree can be full, so it would need to grow in depth: * allocation + old root + new root */ needed += 2 + 1 + 1; /* - * Index split can happen, we'd need: + * Index split can happen, we would need: * allocate intermediate indexes (bitmap + group) * + change two blocks at each level, but root (already included) */ @@ -1634,7 +1661,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, BUG_ON(b != ex_ee_block + ex_ee_len - 1); } - /* at present, extent can't cross block group */ + /* at present, extent can't cross block group: */ /* leaf + bitmap + group desc + sb + inode */ credits = 5; if (ex == EXT_FIRST_EXTENT(eh)) { @@ -1660,7 +1687,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, goto out; if (num == 0) { - /* this extent is removed entirely mark slot unused */ + /* this extent is removed; mark slot entirely unused */ ext4_ext_store_pblock(ex, 0); eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); } @@ -1692,7 +1719,8 @@ out: } /* - * returns 1 if current index have to be freed (even partial) + * ext4_ext_more_to_rm: + * returns 1 if current index has to be freed (even partial) */ static int inline ext4_ext_more_to_rm(struct ext4_ext_path *path) @@ -1703,7 +1731,7 @@ ext4_ext_more_to_rm(struct ext4_ext_path *path) return 0; /* - * if truncate on deeper level happened it it wasn't partial + * if truncate on deeper level happened, it wasn't partial, * so we have to consider current index for truncation */ if (le16_to_cpu(path->p_hdr->eh_entries) == path->p_block) @@ -1729,8 +1757,8 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) ext4_ext_invalidate_cache(inode); /* - * we start scanning from right side freeing all the blocks - * after i_size and walking into the deep + * We start scanning from right side, freeing all the blocks + * after i_size and walking into the tree depth-wise. */ path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL); if (path == NULL) { @@ -1749,7 +1777,7 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) if (i == depth) { /* this is leaf block */ err = ext4_ext_rm_leaf(handle, inode, path, start); - /* root level have p_bh == NULL, brelse() eats this */ + /* root level has p_bh == NULL, brelse() eats this */ brelse(path[i].p_bh); path[i].p_bh = NULL; i--; @@ -1772,14 +1800,14 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) BUG_ON(path[i].p_hdr->eh_magic != EXT4_EXT_MAGIC); if (!path[i].p_idx) { - /* this level hasn't touched yet */ + /* this level hasn't been touched yet */ path[i].p_idx = EXT_LAST_INDEX(path[i].p_hdr); path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries)+1; ext_debug("init index ptr: hdr 0x%p, num %d\n", path[i].p_hdr, le16_to_cpu(path[i].p_hdr->eh_entries)); } else { - /* we've already was here, see at next index */ + /* we were already here, see at next index */ path[i].p_idx--; } @@ -1799,19 +1827,19 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) break; } - /* put actual number of indexes to know is this - * number got changed at the next iteration */ + /* save actual number of indexes since this + * number is changed at the next iteration */ path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries); i++; } else { - /* we finish processing this index, go up */ + /* we finished processing this index, go up */ if (path[i].p_hdr->eh_entries == 0 && i > 0) { - /* index is empty, remove it + /* index is empty, remove it; * handle must be already prepared by the * truncatei_leaf() */ err = ext4_ext_rm_idx(handle, inode, path + i); } - /* root level have p_bh == NULL, brelse() eats this */ + /* root level has p_bh == NULL, brelse() eats this */ brelse(path[i].p_bh); path[i].p_bh = NULL; i--; @@ -1822,8 +1850,8 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) /* TODO: flexible tree reduction should be here */ if (path->p_hdr->eh_entries == 0) { /* - * truncate to zero freed all the tree - * so, we need to correct eh_depth + * truncate to zero freed all the tree, + * so we need to correct eh_depth */ err = ext4_ext_get_access(handle, inode, path); if (err == 0) { @@ -1912,7 +1940,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, if (goal == EXT4_EXT_CACHE_GAP) { if (!create) { /* block isn't allocated yet and - * user don't want to allocate it */ + * user doesn't want to allocate it */ goto out2; } /* we should allocate requested block */ @@ -1921,7 +1949,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, newblock = iblock - le32_to_cpu(newex.ee_block) + ext_pblock(&newex); - /* number of remain blocks in the extent */ + /* number of remaining blocks in the extent */ allocated = le16_to_cpu(newex.ee_len) - (iblock - le32_to_cpu(newex.ee_block)); goto out; @@ -1941,8 +1969,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, depth = ext_depth(inode); /* - * consistent leaf must not be empty - * this situations is possible, though, _during_ tree modification + * consistent leaf must not be empty; + * this situation is possible, though, _during_ tree modification; * this is why assert can't be put in ext4_ext_find_extent() */ BUG_ON(path[depth].p_ext == NULL && depth != 0); @@ -1960,10 +1988,10 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, */ if (ee_len > EXT_MAX_LEN) goto out2; - /* if found exent covers block, simple return it */ + /* if found extent covers block, simply return it */ if (iblock >= ee_block && iblock < ee_block + ee_len) { newblock = iblock - ee_block + ee_start; - /* number of remain blocks in the extent */ + /* number of remaining blocks in the extent */ allocated = ee_len - (iblock - ee_block); ext_debug("%d fit into %lu:%d -> "E3FSBLK"\n", (int) iblock, ee_block, ee_len, newblock); @@ -1974,17 +2002,18 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, } /* - * requested block isn't allocated yet + * requested block isn't allocated yet; * we couldn't try to create block if create flag is zero */ if (!create) { - /* put just found gap into cache to speedup subsequest reqs */ + /* put just found gap into cache to speed up + * subsequent requests */ ext4_ext_put_gap_in_cache(inode, path, iblock); goto out2; } /* * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary + * allocation info here if necessary. */ if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) ext4_init_block_alloc_info(inode); @@ -2062,9 +2091,9 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) ext4_ext_invalidate_cache(inode); /* - * TODO: optimization is possible here - * probably we need not scaning at all, - * because page truncation is enough + * TODO: optimization is possible here. + * Probably we need not scan at all, + * because page truncation is enough. */ if (ext4_orphan_add(handle, inode)) goto out_stop; @@ -2078,13 +2107,13 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) err = ext4_ext_remove_space(inode, last_block); /* In a multi-transaction truncate, we only make the final - * transaction synchronous */ + * transaction synchronous. */ if (IS_SYNC(inode)) handle->h_sync = 1; out_stop: /* - * If this was a simple ftruncate(), and the file will remain alive + * If this was a simple ftruncate() and the file will remain alive, * then we need to clear up the orphan record which we created above. * However, if this was a real unlink then we were called by * ext4_delete_inode(), and we allow that function to clean up the @@ -2098,7 +2127,8 @@ out_stop: } /* - * this routine calculate max number of blocks we could modify + * ext4_ext_writepage_trans_blocks: + * calculate max number of blocks we could modify * in order to allocate new block for an inode */ int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) @@ -2107,7 +2137,7 @@ int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) needed = ext4_ext_calc_credits_for_insert(inode, NULL); - /* caller want to allocate num blocks, but note it includes sb */ + /* caller wants to allocate num blocks, but note it includes sb */ needed = needed * num - (num - 1); #ifdef CONFIG_QUOTA diff --git a/include/linux/ext4_fs_extents.h b/include/linux/ext4_fs_extents.h index 0eba0ac..a41cc24 100644 --- a/include/linux/ext4_fs_extents.h +++ b/include/linux/ext4_fs_extents.h @@ -22,29 +22,29 @@ #include /* - * with AGRESSIVE_TEST defined capacity of index/leaf blocks - * become very little, so index split, in-depth growing and - * other hard changes happens much more often - * this is for debug purposes only + * With AGRESSIVE_TEST defined, the capacity of index/leaf blocks + * becomes very small, so index split, in-depth growing and + * other hard changes happen much more often. + * This is for debug purposes only. */ #define AGRESSIVE_TEST_ /* - * with EXTENTS_STATS defined number of blocks and extents - * are collected in truncate path. they'll be showed at - * umount time + * With EXTENTS_STATS defined, the number of blocks and extents + * are collected in the truncate path. They'll be shown at + * umount time. */ #define EXTENTS_STATS__ /* - * if CHECK_BINSEARCH defined, then results of binary search - * will be checked by linear search + * If CHECK_BINSEARCH is defined, then the results of the binary search + * will also be checked by linear search. */ #define CHECK_BINSEARCH__ /* - * if EXT_DEBUG is defined you can use 'extdebug' mount option - * to get lots of info what's going on + * If EXT_DEBUG is defined you can use the 'extdebug' mount option + * to get lots of info about what's going on. */ #define EXT_DEBUG__ #ifdef EXT_DEBUG @@ -54,58 +54,58 @@ #endif /* - * if EXT_STATS is defined then stats numbers are collected - * these number will be displayed at umount time + * If EXT_STATS is defined then stats numbers are collected. + * These number will be displayed at umount time. */ #define EXT_STATS_ /* - * ext4_inode has i_block array (60 bytes total) - * first 12 bytes store ext4_extent_header - * the remain stores array of ext4_extent + * ext4_inode has i_block array (60 bytes total). + * The first 12 bytes store ext4_extent_header; + * the remainder stores an array of ext4_extent. */ /* - * this is extent on-disk structure - * it's used at the bottom of the tree + * This is the extent on-disk structure. + * It's used at the bottom of the tree. */ struct ext4_extent { __le32 ee_block; /* first logical block extent covers */ __le16 ee_len; /* number of blocks covered by extent */ __le16 ee_start_hi; /* high 16 bits of physical block */ - __le32 ee_start; /* low 32 bigs of physical block */ + __le32 ee_start; /* low 32 bits of physical block */ }; /* - * this is index on-disk structure - * it's used at all the levels, but the bottom + * This is index on-disk structure. + * It's used at all the levels except the bottom. */ struct ext4_extent_idx { __le32 ei_block; /* index covers logical blocks from 'block' */ __le32 ei_leaf; /* pointer to the physical block of the next * - * level. leaf or next index could bet here */ + * level. leaf or next index could be there */ __le16 ei_leaf_hi; /* high 16 bits of physical block */ __u16 ei_unused; }; /* - * each block (leaves and indexes), even inode-stored has header + * Each block (leaves and indexes), even inode-stored has header. */ struct ext4_extent_header { __le16 eh_magic; /* probably will support different formats */ __le16 eh_entries; /* number of valid entries */ __le16 eh_max; /* capacity of store in entries */ - __le16 eh_depth; /* has tree real underlaying blocks? */ + __le16 eh_depth; /* has tree real underlying blocks? */ __le32 eh_generation; /* generation of the tree */ }; #define EXT4_EXT_MAGIC cpu_to_le16(0xf30a) /* - * array of ext4_ext_path contains path to some extent - * creation/lookup routines use it for traversal/splitting/etc - * truncate uses it to simulate recursive walking + * Array of ext4_ext_path contains path to some extent. + * Creation/lookup routines use it for traversal/splitting/etc. + * Truncate uses it to simulate recursive walking. */ struct ext4_ext_path { ext4_fsblk_t p_block; diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h index aa273f0..f69af60 100644 --- a/include/linux/ext4_jbd2.h +++ b/include/linux/ext4_jbd2.h @@ -28,8 +28,8 @@ * indirection blocks, the group and superblock summaries, and the data * block to complete the transaction. * - * For extents-enabled fs we may have to allocate and modify upto - * 5 levels of tree + root which is stored in inode. */ + * For extents-enabled fs we may have to allocate and modify up to + * 5 levels of tree + root which are stored in the inode. */ #define EXT4_SINGLEDATA_TRANS_BLOCKS(sb) \ (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS) \ -- cgit v0.10.2 From b517bea1c74e4773482b3f41b3f493522a8c8e30 Mon Sep 17 00:00:00 2001 From: Zach Brown Date: Wed, 11 Oct 2006 01:21:08 -0700 Subject: [PATCH] 64-bit jbd2 core Here is the patch to JBD to handle 64 bit block numbers, originally from Zach Brown. This patch is useful only after adding support for 64-bit block numbers in the filesystem. Signed-off-by: Badari Pulavarty Signed-off-by: Zach Brown Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index b1a4eaf..44d68a1 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -271,6 +271,14 @@ write_out_data: journal_do_submit_data(wbuf, bufs); } +static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, + sector_t block) +{ + tag->t_blocknr = cpu_to_be32(block & (u32)~0); + if (tag_bytes > JBD_TAG_SIZE32) + tag->t_blocknr_high = cpu_to_be32((block >> 31) >> 1); +} + /* * jbd2_journal_commit_transaction * @@ -293,6 +301,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) int first_tag = 0; int tag_flag; int i; + int tag_bytes = journal_tag_bytes(journal); /* * First job: lock down the current transaction and wait for @@ -597,10 +606,10 @@ void jbd2_journal_commit_transaction(journal_t *journal) tag_flag |= JBD2_FLAG_SAME_UUID; tag = (journal_block_tag_t *) tagp; - tag->t_blocknr = cpu_to_be32(jh2bh(jh)->b_blocknr); + write_tag_block(tag_bytes, tag, jh2bh(jh)->b_blocknr); tag->t_flags = cpu_to_be32(tag_flag); - tagp += sizeof(journal_block_tag_t); - space_left -= sizeof(journal_block_tag_t); + tagp += tag_bytes; + space_left -= tag_bytes; if (first_tag) { memcpy (tagp, journal->j_uuid, 16); @@ -614,7 +623,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) if (bufs == journal->j_wbufsize || commit_transaction->t_buffers == NULL || - space_left < sizeof(journal_block_tag_t) + 16) { + space_left < tag_bytes + 16) { jbd_debug(4, "JBD: Submit %d IOs\n", bufs); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 8d0f71e..926ebcb 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1610,6 +1610,17 @@ int jbd2_journal_blocks_per_page(struct inode *inode) } /* + * helper functions to deal with 32 or 64bit block numbers. + */ +size_t journal_tag_bytes(journal_t *journal) +{ + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) + return JBD_TAG_SIZE64; + else + return JBD_TAG_SIZE32; +} + +/* * Simple support for retrying memory allocations. Introduced to help to * debug different VM deadlock avoidance strategies. */ diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index b2012d1..2486843 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -178,19 +178,20 @@ static int jread(struct buffer_head **bhp, journal_t *journal, * Count the number of in-use tags in a journal descriptor block. */ -static int count_tags(struct buffer_head *bh, int size) +static int count_tags(journal_t *journal, struct buffer_head *bh) { char * tagp; journal_block_tag_t * tag; - int nr = 0; + int nr = 0, size = journal->j_blocksize; + int tag_bytes = journal_tag_bytes(journal); tagp = &bh->b_data[sizeof(journal_header_t)]; - while ((tagp - bh->b_data + sizeof(journal_block_tag_t)) <= size) { + while ((tagp - bh->b_data + tag_bytes) <= size) { tag = (journal_block_tag_t *) tagp; nr++; - tagp += sizeof(journal_block_tag_t); + tagp += tag_bytes; if (!(tag->t_flags & cpu_to_be32(JBD2_FLAG_SAME_UUID))) tagp += 16; @@ -307,6 +308,14 @@ int jbd2_journal_skip_recovery(journal_t *journal) return err; } +static inline sector_t read_tag_block(int tag_bytes, journal_block_tag_t *tag) +{ + sector_t block = be32_to_cpu(tag->t_blocknr); + if (tag_bytes > JBD_TAG_SIZE32) + block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32; + return block; +} + static int do_one_pass(journal_t *journal, struct recovery_info *info, enum passtype pass) { @@ -318,11 +327,12 @@ static int do_one_pass(journal_t *journal, struct buffer_head * bh; unsigned int sequence; int blocktype; + int tag_bytes = journal_tag_bytes(journal); /* Precompute the maximum metadata descriptors in a descriptor block */ int MAX_BLOCKS_PER_DESC; MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t)) - / sizeof(journal_block_tag_t)); + / tag_bytes); /* * First thing is to establish what we expect to find in the log @@ -412,8 +422,7 @@ static int do_one_pass(journal_t *journal, * in pass REPLAY; otherwise, just skip over the * blocks it describes. */ if (pass != PASS_REPLAY) { - next_log_block += - count_tags(bh, journal->j_blocksize); + next_log_block += count_tags(journal, bh); wrap(journal, next_log_block); brelse(bh); continue; @@ -424,7 +433,7 @@ static int do_one_pass(journal_t *journal, * getting done here! */ tagp = &bh->b_data[sizeof(journal_header_t)]; - while ((tagp - bh->b_data +sizeof(journal_block_tag_t)) + while ((tagp - bh->b_data + tag_bytes) <= journal->j_blocksize) { unsigned long io_block; @@ -446,7 +455,8 @@ static int do_one_pass(journal_t *journal, unsigned long blocknr; J_ASSERT(obh != NULL); - blocknr = be32_to_cpu(tag->t_blocknr); + blocknr = read_tag_block(tag_bytes, + tag); /* If the block has been * revoked, then we're all done @@ -494,7 +504,7 @@ static int do_one_pass(journal_t *journal, } skip_write: - tagp += sizeof(journal_block_tag_t); + tagp += tag_bytes; if (!(flags & JBD2_FLAG_SAME_UUID)) tagp += 16; @@ -572,17 +582,24 @@ static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, { jbd2_journal_revoke_header_t *header; int offset, max; + int record_len = 4; header = (jbd2_journal_revoke_header_t *) bh->b_data; offset = sizeof(jbd2_journal_revoke_header_t); max = be32_to_cpu(header->r_count); - while (offset < max) { + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) + record_len = 8; + + while (offset + record_len <= max) { unsigned long blocknr; int err; - blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset))); - offset += 4; + if (record_len == 4) + blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset))); + else + blocknr = be64_to_cpu(* ((__be64 *) (bh->b_data+offset))); + offset += record_len; err = jbd2_journal_set_revoke(journal, blocknr, sequence); if (err) return err; diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 5820a0c..8aac875 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -584,9 +584,17 @@ static void write_one_revoke_record(journal_t *journal, *descriptorp = descriptor; } - * ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = - cpu_to_be32(record->blocknr); - offset += 4; + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) { + * ((__be64 *)(&jh2bh(descriptor)->b_data[offset])) = + cpu_to_be64(record->blocknr); + offset += 8; + + } else { + * ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = + cpu_to_be32(record->blocknr); + offset += 4; + } + *offsetp = offset; } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 3251f7a..5e5aa64 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -150,14 +150,21 @@ typedef struct journal_header_s /* - * The block tag: used to describe a single buffer in the journal + * The block tag: used to describe a single buffer in the journal. + * t_blocknr_high is only used if INCOMPAT_64BIT is set, so this + * raw struct shouldn't be used for pointer math or sizeof() - use + * journal_tag_bytes(journal) instead to compute this. */ typedef struct journal_block_tag_s { __be32 t_blocknr; /* The on-disk block number */ __be32 t_flags; /* See below */ + __be32 t_blocknr_high; /* most-significant high 32bits. */ } journal_block_tag_t; +#define JBD_TAG_SIZE32 (offsetof(journal_block_tag_t, t_blocknr_high)) +#define JBD_TAG_SIZE64 (sizeof(journal_block_tag_t)) + /* * The revoke descriptor: used on disk to describe a series of blocks to * be revoked from the log @@ -235,11 +242,13 @@ typedef struct journal_superblock_s ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask)))) #define JBD2_FEATURE_INCOMPAT_REVOKE 0x00000001 +#define JBD2_FEATURE_INCOMPAT_64BIT 0x00000002 /* Features known to this kernel version: */ #define JBD2_KNOWN_COMPAT_FEATURES 0 #define JBD2_KNOWN_ROCOMPAT_FEATURES 0 -#define JBD2_KNOWN_INCOMPAT_FEATURES JBD2_FEATURE_INCOMPAT_REVOKE +#define JBD2_KNOWN_INCOMPAT_FEATURES (JBD2_FEATURE_INCOMPAT_REVOKE | \ + JBD2_FEATURE_INCOMPAT_64BIT) #ifdef __KERNEL__ @@ -1052,6 +1061,7 @@ static inline int tid_geq(tid_t x, tid_t y) } extern int jbd2_journal_blocks_per_page(struct inode *inode); +extern size_t journal_tag_bytes(journal_t *journal); /* * Return the minimum number of blocks which must be free in the journal -- cgit v0.10.2 From 299717696d48531d70aeb4614c3939e4a28456c1 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:21:09 -0700 Subject: [PATCH] jbd2: sector_t conversion JBD layer in-kernel block varibles type fixes to support >32 bit block number and convert to sector_t type. Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 44d68a1..1a9ce88 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -293,7 +293,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) int bufs; int flags; int err; - unsigned long blocknr; + sector_t blocknr; char *tagp = NULL; journal_header_t *header; journal_block_tag_t *tag = NULL; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 926ebcb..259e836 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -271,7 +271,7 @@ static void journal_kill_thread(journal_t *journal) int jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, struct journal_head **jh_out, - unsigned long blocknr) + sector_t blocknr) { int need_copy_out = 0; int done_copy_out = 0; @@ -555,7 +555,7 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid) * Log buffer allocation routines: */ -int jbd2_journal_next_log_block(journal_t *journal, unsigned long *retp) +int jbd2_journal_next_log_block(journal_t *journal, sector_t *retp) { unsigned long blocknr; @@ -579,10 +579,10 @@ int jbd2_journal_next_log_block(journal_t *journal, unsigned long *retp) * ready. */ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, - unsigned long *retp) + sector_t *retp) { int err = 0; - unsigned long ret; + sector_t ret; if (journal->j_inode) { ret = bmap(journal->j_inode, blocknr); @@ -618,7 +618,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) { struct buffer_head *bh; - unsigned long blocknr; + sector_t blocknr; int err; err = jbd2_journal_next_log_block(journal, &blocknr); @@ -706,7 +706,7 @@ fail: */ journal_t * jbd2_journal_init_dev(struct block_device *bdev, struct block_device *fs_dev, - int start, int len, int blocksize) + sector_t start, int len, int blocksize) { journal_t *journal = journal_init_common(); struct buffer_head *bh; @@ -753,7 +753,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal_t *journal = journal_init_common(); int err; int n; - unsigned long blocknr; + sector_t blocknr; if (!journal) return NULL; @@ -819,7 +819,7 @@ static void journal_fail_superblock (journal_t *journal) static int journal_reset(journal_t *journal) { journal_superblock_t *sb = journal->j_superblock; - unsigned long first, last; + sector_t first, last; first = be32_to_cpu(sb->s_first); last = be32_to_cpu(sb->s_maxlen); @@ -853,7 +853,7 @@ static int journal_reset(journal_t *journal) **/ int jbd2_journal_create(journal_t *journal) { - unsigned long blocknr; + sector_t blocknr; struct buffer_head *bh; journal_superblock_t *sb; int i, err; diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 2486843..52054a83 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -70,7 +70,7 @@ static int do_readahead(journal_t *journal, unsigned int start) { int err; unsigned int max, nbufs, next; - unsigned long blocknr; + sector_t blocknr; struct buffer_head *bh; struct buffer_head * bufs[MAXBUF]; @@ -132,7 +132,7 @@ static int jread(struct buffer_head **bhp, journal_t *journal, unsigned int offset) { int err; - unsigned long blocknr; + sector_t blocknr; struct buffer_head *bh; *bhp = NULL; @@ -452,7 +452,7 @@ static int do_one_pass(journal_t *journal, "block %ld in log\n", err, io_block); } else { - unsigned long blocknr; + sector_t blocknr; J_ASSERT(obh != NULL); blocknr = read_tag_block(tag_bytes, @@ -592,7 +592,7 @@ static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, record_len = 8; while (offset + record_len <= max) { - unsigned long blocknr; + sector_t blocknr; int err; if (record_len == 4) diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 8aac875..3310a1d 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -81,7 +81,7 @@ struct jbd2_revoke_record_s { struct list_head hash; tid_t sequence; /* Used for recovery only */ - unsigned long blocknr; + sector_t blocknr; }; @@ -106,17 +106,18 @@ static void flush_descriptor(journal_t *, struct journal_head *, int); /* Utility functions to maintain the revoke table */ /* Borrowed from buffer.c: this is a tried and tested block hash function */ -static inline int hash(journal_t *journal, unsigned long block) +static inline int hash(journal_t *journal, sector_t block) { struct jbd2_revoke_table_s *table = journal->j_revoke; int hash_shift = table->hash_shift; + int hash = (int)block ^ (int)((block >> 31) >> 1); - return ((block << (hash_shift - 6)) ^ - (block >> 13) ^ - (block << (hash_shift - 12))) & (table->hash_size - 1); + return ((hash << (hash_shift - 6)) ^ + (hash >> 13) ^ + (hash << (hash_shift - 12))) & (table->hash_size - 1); } -static int insert_revoke_hash(journal_t *journal, unsigned long blocknr, +static int insert_revoke_hash(journal_t *journal, sector_t blocknr, tid_t seq) { struct list_head *hash_list; @@ -146,7 +147,7 @@ oom: /* Find a revoke record in the journal's hash table. */ static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, - unsigned long blocknr) + sector_t blocknr) { struct list_head *hash_list; struct jbd2_revoke_record_s *record; @@ -325,7 +326,7 @@ void jbd2_journal_destroy_revoke(journal_t *journal) * by one. */ -int jbd2_journal_revoke(handle_t *handle, unsigned long blocknr, +int jbd2_journal_revoke(handle_t *handle, sector_t blocknr, struct buffer_head *bh_in) { struct buffer_head *bh = NULL; @@ -394,7 +395,7 @@ int jbd2_journal_revoke(handle_t *handle, unsigned long blocknr, } } - jbd_debug(2, "insert revoke for block %lu, bh_in=%p\n", blocknr, bh_in); + jbd_debug(2, "insert revoke for block %llu, bh_in=%p\n",blocknr, bh_in); err = insert_revoke_hash(journal, blocknr, handle->h_transaction->t_tid); BUFFER_TRACE(bh_in, "exit"); @@ -649,7 +650,7 @@ static void flush_descriptor(journal_t *journal, */ int jbd2_journal_set_revoke(journal_t *journal, - unsigned long blocknr, + sector_t blocknr, tid_t sequence) { struct jbd2_revoke_record_s *record; @@ -673,7 +674,7 @@ int jbd2_journal_set_revoke(journal_t *journal, */ int jbd2_journal_test_revoke(journal_t *journal, - unsigned long blocknr, + sector_t blocknr, tid_t sequence) { struct jbd2_revoke_record_s *record; diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h index f69af60..72dd631 100644 --- a/include/linux/ext4_jbd2.h +++ b/include/linux/ext4_jbd2.h @@ -154,7 +154,7 @@ __ext4_journal_forget(const char *where, handle_t *handle, struct buffer_head *b static inline int __ext4_journal_revoke(const char *where, handle_t *handle, - unsigned long blocknr, struct buffer_head *bh) + ext4_fsblk_t blocknr, struct buffer_head *bh) { int err = jbd2_journal_revoke(handle, blocknr, bh); if (err) diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 5e5aa64..3c939c8 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -741,7 +741,7 @@ struct journal_s */ struct block_device *j_dev; int j_blocksize; - unsigned long j_blk_offset; + sector_t j_blk_offset; /* * Device which holds the client fs. For internal journal this will be @@ -860,7 +860,7 @@ extern void __journal_clean_data_list(transaction_t *transaction); /* Log buffer allocation */ extern struct journal_head * jbd2_journal_get_descriptor_buffer(journal_t *); -int jbd2_journal_next_log_block(journal_t *, unsigned long *); +int jbd2_journal_next_log_block(journal_t *, sector_t *); /* Commit management */ extern void jbd2_journal_commit_transaction(journal_t *); @@ -875,7 +875,7 @@ extern int jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, struct journal_head **jh_out, - unsigned long blocknr); + sector_t blocknr); /* Transaction locking */ extern void __wait_on_journal (journal_t *); @@ -923,7 +923,7 @@ extern void jbd2_journal_unlock_updates (journal_t *); extern journal_t * jbd2_journal_init_dev(struct block_device *bdev, struct block_device *fs_dev, - int start, int len, int bsize); + sector_t start, int len, int bsize); extern journal_t * jbd2_journal_init_inode (struct inode *); extern int jbd2_journal_update_format (journal_t *); extern int jbd2_journal_check_used_features @@ -944,7 +944,7 @@ extern void jbd2_journal_abort (journal_t *, int); extern int jbd2_journal_errno (journal_t *); extern void jbd2_journal_ack_err (journal_t *); extern int jbd2_journal_clear_err (journal_t *); -extern int jbd2_journal_bmap(journal_t *, unsigned long, unsigned long *); +extern int jbd2_journal_bmap(journal_t *, unsigned long, sector_t *); extern int jbd2_journal_force_commit(journal_t *); /* @@ -977,14 +977,13 @@ extern void jbd2_journal_destroy_revoke_caches(void); extern int jbd2_journal_init_revoke_caches(void); extern void jbd2_journal_destroy_revoke(journal_t *); -extern int jbd2_journal_revoke (handle_t *, - unsigned long, struct buffer_head *); +extern int jbd2_journal_revoke (handle_t *, sector_t, struct buffer_head *); extern int jbd2_journal_cancel_revoke(handle_t *, struct journal_head *); extern void jbd2_journal_write_revoke_records(journal_t *, transaction_t *); /* Recovery revoke support */ -extern int jbd2_journal_set_revoke(journal_t *, unsigned long, tid_t); -extern int jbd2_journal_test_revoke(journal_t *, unsigned long, tid_t); +extern int jbd2_journal_set_revoke(journal_t *, sector_t, tid_t); +extern int jbd2_journal_test_revoke(journal_t *, sector_t, tid_t); extern void jbd2_journal_clear_revoke(journal_t *); extern void jbd2_journal_switch_revoke_table(journal_t *journal); -- cgit v0.10.2 From a1ddeb7eaecea6a924e3a79aa386797020cb436f Mon Sep 17 00:00:00 2001 From: Badari Pulavarty Date: Wed, 11 Oct 2006 01:21:09 -0700 Subject: [PATCH] ext4: 48bit i_file_acl As we are planning to support 48-bit block numbers for ext4, we need to support 48-bit block numbers for extended attributes. In the short term, we can do this by reuse (on-disk) 16-bit padding (linux2.i_pad1 currently used only by "hurd") as high order bits for xattr. This patch basically does that. Signed-off-by: Badari Pulavarty Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 2b81b13..9db8cff 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2643,6 +2643,11 @@ void ext4_read_inode(struct inode * inode) ei->i_frag_size = raw_inode->i_fsize; #endif ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl); + if ((sizeof(sector_t) > 4) && + (EXT4_SB(inode->i_sb)->s_es->s_creator_os != + cpu_to_le32(EXT4_OS_HURD))) + ei->i_file_acl |= + ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; if (!S_ISREG(inode->i_mode)) { ei->i_dir_acl = le32_to_cpu(raw_inode->i_dir_acl); } else { @@ -2776,6 +2781,11 @@ static int ext4_do_update_inode(handle_t *handle, raw_inode->i_frag = ei->i_frag_no; raw_inode->i_fsize = ei->i_frag_size; #endif + if ((sizeof(sector_t) > 4) && + (EXT4_SB(inode->i_sb)->s_es->s_creator_os != + cpu_to_le32(EXT4_OS_HURD))) + raw_inode->i_file_acl_high = + cpu_to_le16(ei->i_file_acl >> 32); raw_inode->i_file_acl = cpu_to_le32(ei->i_file_acl); if (!S_ISREG(inode->i_mode)) { raw_inode->i_dir_acl = cpu_to_le32(ei->i_dir_acl); diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index e952c6d..63ed89f 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -298,7 +298,7 @@ struct ext4_inode { struct { __u8 l_i_frag; /* Fragment number */ __u8 l_i_fsize; /* Fragment size */ - __u16 i_pad1; + __le16 l_i_file_acl_high; __le16 l_i_uid_high; /* these 2 fields */ __le16 l_i_gid_high; /* were reserved2[0] */ __u32 l_i_reserved2; @@ -314,7 +314,7 @@ struct ext4_inode { struct { __u8 m_i_frag; /* Fragment number */ __u8 m_i_fsize; /* Fragment size */ - __u16 m_pad1; + __le16 m_i_file_acl_high; __u32 m_i_reserved2[2]; } masix2; } osd2; /* OS dependent 2 */ @@ -328,6 +328,7 @@ struct ext4_inode { #define i_reserved1 osd1.linux1.l_i_reserved1 #define i_frag osd2.linux2.l_i_frag #define i_fsize osd2.linux2.l_i_fsize +#define i_file_acl_high osd2.linux2.l_i_file_acl_high #define i_uid_low i_uid #define i_gid_low i_gid #define i_uid_high osd2.linux2.l_i_uid_high @@ -348,6 +349,7 @@ struct ext4_inode { #define i_reserved1 osd1.masix1.m_i_reserved1 #define i_frag osd2.masix2.m_i_frag #define i_fsize osd2.masix2.m_i_fsize +#define i_file_acl_high osd2.masix2.m_i_file_acl_high #define i_reserved2 osd2.masix2.m_i_reserved2 #endif /* defined(__KERNEL__) || defined(__linux__) */ -- cgit v0.10.2 From bd81d8eec043094d3ff729a8ff6d5b3a06d3c4b1 Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Wed, 11 Oct 2006 01:21:10 -0700 Subject: [PATCH] ext4: 64bit metadata In-kernel super block changes to support >32 bit free blocks numbers. Signed-off-by: Laurent Vivier Signed-off-by: Dave Kleikamp Signed-off-by: Alexandre Ratchov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index aa33ff2..6887151 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -99,12 +99,13 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) desc = ext4_get_group_desc (sb, block_group, NULL); if (!desc) goto error_out; - bh = sb_bread(sb, le32_to_cpu(desc->bg_block_bitmap)); + bh = sb_bread(sb, ext4_block_bitmap(desc)); if (!bh) ext4_error (sb, "read_block_bitmap", "Cannot read block bitmap - " - "block_group = %d, block_bitmap = %u", - block_group, le32_to_cpu(desc->bg_block_bitmap)); + "block_group = %d, block_bitmap = "E3FSBLK, + block_group, + ext4_block_bitmap(desc)); error_out: return bh; } @@ -432,14 +433,14 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, es = sbi->s_es; if (block < le32_to_cpu(es->s_first_data_block) || block + count < block || - block + count > le32_to_cpu(es->s_blocks_count)) { + block + count > ext4_blocks_count(es)) { ext4_error (sb, "ext4_free_blocks", "Freeing blocks not in datazone - " "block = "E3FSBLK", count = %lu", block, count); goto error_return; } - ext4_debug ("freeing block(s) %lu-%lu\n", block, block + count - 1); + ext4_debug ("freeing block(s) %llu-%llu\n", block, block + count - 1); do_more: overflow = 0; @@ -460,12 +461,11 @@ do_more: if (!desc) goto error_return; - if (in_range (le32_to_cpu(desc->bg_block_bitmap), block, count) || - in_range (le32_to_cpu(desc->bg_inode_bitmap), block, count) || - in_range (block, le32_to_cpu(desc->bg_inode_table), - sbi->s_itb_per_group) || - in_range (block + count - 1, le32_to_cpu(desc->bg_inode_table), - sbi->s_itb_per_group)) + if (in_range(ext4_block_bitmap(desc), block, count) || + in_range(ext4_inode_bitmap(desc), block, count) || + in_range(block, ext4_inode_table(desc), sbi->s_itb_per_group) || + in_range(block + count - 1, ext4_inode_table(desc), + sbi->s_itb_per_group)) ext4_error (sb, "ext4_free_blocks", "Freeing blocks in system zones - " "Block = "E3FSBLK", count = %lu", @@ -552,8 +552,8 @@ do_more: bit + i, bitmap_bh->b_data)) { jbd_unlock_bh_state(bitmap_bh); ext4_error(sb, __FUNCTION__, - "bit already cleared for block "E3FSBLK, - block + i); + "bit already cleared for block "E3FSBLK, + (ext4_fsblk_t)(block + i)); jbd_lock_bh_state(bitmap_bh); BUFFER_TRACE(bitmap_bh, "bit already cleared"); } else { @@ -1351,7 +1351,7 @@ static int ext4_has_free_blocks(struct ext4_sb_info *sbi) ext4_fsblk_t free_blocks, root_blocks; free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); - root_blocks = le32_to_cpu(sbi->s_es->s_r_blocks_count); + root_blocks = ext4_r_blocks_count(sbi->s_es); if (free_blocks < root_blocks + 1 && !capable(CAP_SYS_RESOURCE) && sbi->s_resuid != current->fsuid && (sbi->s_resgid == 0 || !in_group_p (sbi->s_resgid))) { @@ -1462,7 +1462,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, * First, test whether the goal block is free. */ if (goal < le32_to_cpu(es->s_first_data_block) || - goal >= le32_to_cpu(es->s_blocks_count)) + goal >= ext4_blocks_count(es)) goal = le32_to_cpu(es->s_first_data_block); ext4_get_group_no_and_offset(sb, goal, &group_no, &grp_target_blk); goal_group = group_no; @@ -1561,12 +1561,12 @@ allocated: ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no); - if (in_range(le32_to_cpu(gdp->bg_block_bitmap), ret_block, num) || - in_range(le32_to_cpu(gdp->bg_inode_bitmap), ret_block, num) || - in_range(ret_block, le32_to_cpu(gdp->bg_inode_table), - EXT4_SB(sb)->s_itb_per_group) || - in_range(ret_block + num - 1, le32_to_cpu(gdp->bg_inode_table), - EXT4_SB(sb)->s_itb_per_group)) + if (in_range(ext4_block_bitmap(gdp), ret_block, num) || + in_range(ext4_block_bitmap(gdp), ret_block, num) || + in_range(ret_block, ext4_inode_table(gdp), + EXT4_SB(sb)->s_itb_per_group) || + in_range(ret_block + num - 1, ext4_inode_table(gdp), + EXT4_SB(sb)->s_itb_per_group)) ext4_error(sb, "ext4_new_block", "Allocating block in system zone - " "blocks from "E3FSBLK", length %lu", @@ -1604,11 +1604,11 @@ allocated: jbd_unlock_bh_state(bitmap_bh); #endif - if (ret_block + num - 1 >= le32_to_cpu(es->s_blocks_count)) { + if (ret_block + num - 1 >= ext4_blocks_count(es)) { ext4_error(sb, "ext4_new_block", - "block("E3FSBLK") >= blocks count(%d) - " + "block("E3FSBLK") >= blocks count("E3FSBLK") - " "block_group = %lu, es == %p ", ret_block, - le32_to_cpu(es->s_blocks_count), group_no, es); + ext4_blocks_count(es), group_no, es); goto out; } @@ -1707,7 +1707,7 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) brelse(bitmap_bh); printk("ext4_count_free_blocks: stored = "E3FSBLK ", computed = "E3FSBLK", "E3FSBLK"\n", - le32_to_cpu(es->s_free_blocks_count), + EXT4_FREE_BLOCKS_COUNT(es), desc_count, bitmap_count); return bitmap_count; #else diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 94e1bb4..959b7fa 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -60,12 +60,12 @@ read_inode_bitmap(struct super_block * sb, unsigned long block_group) if (!desc) goto error_out; - bh = sb_bread(sb, le32_to_cpu(desc->bg_inode_bitmap)); + bh = sb_bread(sb, ext4_inode_bitmap(desc)); if (!bh) ext4_error(sb, "read_inode_bitmap", "Cannot read inode bitmap - " - "block_group = %lu, inode_bitmap = %u", - block_group, le32_to_cpu(desc->bg_inode_bitmap)); + "block_group = %lu, inode_bitmap = %llu", + block_group, ext4_inode_bitmap(desc)); error_out: return bh; } @@ -304,7 +304,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) goto fallback; } - blocks_per_dir = le32_to_cpu(es->s_blocks_count) - freeb; + blocks_per_dir = ext4_blocks_count(es) - freeb; sector_div(blocks_per_dir, ndirs); max_dirs = ndirs / ngroups + inodes_per_group / 16; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 9db8cff..effc38a 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2438,8 +2438,8 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, */ offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * EXT4_INODE_SIZE(sb); - block = le32_to_cpu(gdp[desc].bg_inode_table) + - (offset >> EXT4_BLOCK_SIZE_BITS(sb)); + block = ext4_inode_table(gdp + desc) + + (offset >> EXT4_BLOCK_SIZE_BITS(sb)); iloc->block_group = block_group; iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); @@ -2506,7 +2506,7 @@ static int __ext4_get_inode_loc(struct inode *inode, goto make_io; bitmap_bh = sb_getblk(inode->i_sb, - le32_to_cpu(desc->bg_inode_bitmap)); + ext4_inode_bitmap(desc)); if (!bitmap_bh) goto make_io; diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index c60bfed..3dbf91b 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -27,7 +27,7 @@ static int verify_group_input(struct super_block *sb, { struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; - ext4_fsblk_t start = le32_to_cpu(es->s_blocks_count); + ext4_fsblk_t start = ext4_blocks_count(es); ext4_fsblk_t end = start + input->blocks_count; unsigned group = input->group; ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; @@ -68,43 +68,43 @@ static int verify_group_input(struct super_block *sb, end - 1); else if (outside(input->block_bitmap, start, end)) ext4_warning(sb, __FUNCTION__, - "Block bitmap not in group (block %u)", + "Block bitmap not in group (block %llu)", input->block_bitmap); else if (outside(input->inode_bitmap, start, end)) ext4_warning(sb, __FUNCTION__, - "Inode bitmap not in group (block %u)", + "Inode bitmap not in group (block %llu)", input->inode_bitmap); else if (outside(input->inode_table, start, end) || outside(itend - 1, start, end)) ext4_warning(sb, __FUNCTION__, - "Inode table not in group (blocks %u-"E3FSBLK")", + "Inode table not in group (blocks %llu-%llu)", input->inode_table, itend - 1); else if (input->inode_bitmap == input->block_bitmap) ext4_warning(sb, __FUNCTION__, - "Block bitmap same as inode bitmap (%u)", + "Block bitmap same as inode bitmap (%llu)", input->block_bitmap); else if (inside(input->block_bitmap, input->inode_table, itend)) ext4_warning(sb, __FUNCTION__, - "Block bitmap (%u) in inode table (%u-"E3FSBLK")", + "Block bitmap (%llu) in inode table (%llu-%llu)", input->block_bitmap, input->inode_table, itend-1); else if (inside(input->inode_bitmap, input->inode_table, itend)) ext4_warning(sb, __FUNCTION__, - "Inode bitmap (%u) in inode table (%u-"E3FSBLK")", + "Inode bitmap (%llu) in inode table (%llu-%llu)", input->inode_bitmap, input->inode_table, itend-1); else if (inside(input->block_bitmap, start, metaend)) ext4_warning(sb, __FUNCTION__, - "Block bitmap (%u) in GDT table" + "Block bitmap (%llu) in GDT table" " ("E3FSBLK"-"E3FSBLK")", input->block_bitmap, start, metaend - 1); else if (inside(input->inode_bitmap, start, metaend)) ext4_warning(sb, __FUNCTION__, - "Inode bitmap (%u) in GDT table" + "Inode bitmap (%llu) in GDT table" " ("E3FSBLK"-"E3FSBLK")", input->inode_bitmap, start, metaend - 1); else if (inside(input->inode_table, start, metaend) || inside(itend - 1, start, metaend)) ext4_warning(sb, __FUNCTION__, - "Inode table (%u-"E3FSBLK") overlaps" + "Inode table ("E3FSBLK"-"E3FSBLK") overlaps" "GDT table ("E3FSBLK"-"E3FSBLK")", input->inode_table, itend - 1, start, metaend - 1); else @@ -286,6 +286,7 @@ exit_journal: return err; } + /* * Iterate through the groups which hold BACKUP superblock/GDT copies in an * ext4 filesystem. The counters should be initialized to 1, 5, and 7 before @@ -340,12 +341,15 @@ static int verify_reserved_gdb(struct super_block *sb, int gdbackups = 0; while ((grp = ext4_list_backups(sb, &three, &five, &seven)) < end) { - if (le32_to_cpu(*p++) != grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){ + if (le32_to_cpu(*p++) != + grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){ ext4_warning(sb, __FUNCTION__, "reserved GDT "E3FSBLK " missing grp %d ("E3FSBLK")", blk, grp, - grp * EXT4_BLOCKS_PER_GROUP(sb) + blk); + grp * + (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) + + blk); return -EINVAL; } if (++gdbackups > EXT4_ADDR_PER_BLOCK(sb)) @@ -731,8 +735,8 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) return -EPERM; } - if (le32_to_cpu(es->s_blocks_count) + input->blocks_count < - le32_to_cpu(es->s_blocks_count)) { + if (ext4_blocks_count(es) + input->blocks_count < + ext4_blocks_count(es)) { ext4_warning(sb, __FUNCTION__, "blocks_count overflow\n"); return -EINVAL; } @@ -830,9 +834,9 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) /* Update group descriptor block for new group */ gdp = (struct ext4_group_desc *)primary->b_data + gdb_off; - gdp->bg_block_bitmap = cpu_to_le32(input->block_bitmap); - gdp->bg_inode_bitmap = cpu_to_le32(input->inode_bitmap); - gdp->bg_inode_table = cpu_to_le32(input->inode_table); + ext4_block_bitmap_set(gdp, input->block_bitmap); /* LV FIXME */ + ext4_inode_bitmap_set(gdp, input->inode_bitmap); /* LV FIXME */ + ext4_inode_table_set(gdp, input->inode_table); /* LV FIXME */ gdp->bg_free_blocks_count = cpu_to_le16(input->free_blocks_count); gdp->bg_free_inodes_count = cpu_to_le16(EXT4_INODES_PER_GROUP(sb)); @@ -846,7 +850,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) * blocks/inodes before the group is live won't actually let us * allocate the new space yet. */ - es->s_blocks_count = cpu_to_le32(le32_to_cpu(es->s_blocks_count) + + ext4_blocks_count_set(es, ext4_blocks_count(es) + input->blocks_count); es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) + EXT4_INODES_PER_GROUP(sb)); @@ -882,7 +886,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) /* Update the reserved block counts only once the new group is * active. */ - es->s_r_blocks_count = cpu_to_le32(le32_to_cpu(es->s_r_blocks_count) + + ext4_r_blocks_count_set(es, ext4_r_blocks_count(es) + input->reserved_blocks); /* Update the free space counts */ @@ -933,7 +937,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, /* We don't need to worry about locking wrt other resizers just * yet: we're going to revalidate es->s_blocks_count after * taking lock_super() below. */ - o_blocks_count = le32_to_cpu(es->s_blocks_count); + o_blocks_count = ext4_blocks_count(es); o_groups_count = EXT4_SB(sb)->s_groups_count; if (test_opt(sb, DEBUG)) @@ -1004,7 +1008,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, } lock_super(sb); - if (o_blocks_count != le32_to_cpu(es->s_blocks_count)) { + if (o_blocks_count != ext4_blocks_count(es)) { ext4_warning(sb, __FUNCTION__, "multiple resizers run on filesystem!"); unlock_super(sb); @@ -1020,7 +1024,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_journal_stop(handle); goto exit_put; } - es->s_blocks_count = cpu_to_le32(o_blocks_count + add); + ext4_blocks_count_set(es, o_blocks_count + add); ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); sb->s_dirt = 1; unlock_super(sb); @@ -1032,8 +1036,8 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, if ((err = ext4_journal_stop(handle))) goto exit_put; if (test_opt(sb, DEBUG)) - printk(KERN_DEBUG "EXT4-fs: extended group to %u blocks\n", - le32_to_cpu(es->s_blocks_count)); + printk(KERN_DEBUG "EXT4-fs: extended group to %llu blocks\n", + ext4_blocks_count(es)); update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr, (char *)es, sizeof(struct ext4_super_block)); exit_put: diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 1d12e4f..b91dffd 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -62,6 +62,43 @@ static void ext4_unlockfs(struct super_block *sb); static void ext4_write_super (struct super_block * sb); static void ext4_write_super_lockfs(struct super_block *sb); + +ext4_fsblk_t ext4_block_bitmap(struct ext4_group_desc *bg) +{ + return le32_to_cpu(bg->bg_block_bitmap) | + ((ext4_fsblk_t)le16_to_cpu(bg->bg_block_bitmap_hi) << 32); +} + +ext4_fsblk_t ext4_inode_bitmap(struct ext4_group_desc *bg) +{ + return le32_to_cpu(bg->bg_inode_bitmap) | + ((ext4_fsblk_t)le16_to_cpu(bg->bg_inode_bitmap_hi) << 32); +} + +ext4_fsblk_t ext4_inode_table(struct ext4_group_desc *bg) +{ + return le32_to_cpu(bg->bg_inode_table) | + ((ext4_fsblk_t)le16_to_cpu(bg->bg_inode_table_hi) << 32); +} + +void ext4_block_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk) +{ + bg->bg_block_bitmap = cpu_to_le32((u32)blk); + bg->bg_block_bitmap_hi = cpu_to_le16(blk >> 32); +} + +void ext4_inode_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk) +{ + bg->bg_inode_bitmap = cpu_to_le32((u32)blk); + bg->bg_inode_bitmap_hi = cpu_to_le16(blk >> 32); +} + +void ext4_inode_table_set(struct ext4_group_desc *bg, ext4_fsblk_t blk) +{ + bg->bg_inode_table = cpu_to_le32((u32)blk); + bg->bg_inode_table_hi = cpu_to_le16(blk >> 32); +} + /* * Wrappers for jbd2_journal_start/end. * @@ -1182,6 +1219,9 @@ static int ext4_check_descriptors (struct super_block * sb) struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); ext4_fsblk_t last_block; + ext4_fsblk_t block_bitmap; + ext4_fsblk_t inode_bitmap; + ext4_fsblk_t inode_table; struct ext4_group_desc * gdp = NULL; int desc_block = 0; int i; @@ -1191,7 +1231,7 @@ static int ext4_check_descriptors (struct super_block * sb) for (i = 0; i < sbi->s_groups_count; i++) { if (i == sbi->s_groups_count - 1) - last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; + last_block = ext4_blocks_count(sbi->s_es) - 1; else last_block = first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); @@ -1199,42 +1239,39 @@ static int ext4_check_descriptors (struct super_block * sb) if ((i % EXT4_DESC_PER_BLOCK(sb)) == 0) gdp = (struct ext4_group_desc *) sbi->s_group_desc[desc_block++]->b_data; - if (le32_to_cpu(gdp->bg_block_bitmap) < first_block || - le32_to_cpu(gdp->bg_block_bitmap) > last_block) + block_bitmap = ext4_block_bitmap(gdp); + if (block_bitmap < first_block || block_bitmap > last_block) { ext4_error (sb, "ext4_check_descriptors", "Block bitmap for group %d" - " not in group (block %lu)!", - i, (unsigned long) - le32_to_cpu(gdp->bg_block_bitmap)); + " not in group (block "E3FSBLK")!", + i, block_bitmap); return 0; } - if (le32_to_cpu(gdp->bg_inode_bitmap) < first_block || - le32_to_cpu(gdp->bg_inode_bitmap) > last_block) + inode_bitmap = ext4_inode_bitmap(gdp); + if (inode_bitmap < first_block || inode_bitmap > last_block) { ext4_error (sb, "ext4_check_descriptors", "Inode bitmap for group %d" - " not in group (block %lu)!", - i, (unsigned long) - le32_to_cpu(gdp->bg_inode_bitmap)); + " not in group (block "E3FSBLK")!", + i, inode_bitmap); return 0; } - if (le32_to_cpu(gdp->bg_inode_table) < first_block || - le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group > - last_block) + inode_table = ext4_inode_table(gdp); + if (inode_table < first_block || + inode_table + sbi->s_itb_per_group > last_block) { ext4_error (sb, "ext4_check_descriptors", "Inode table for group %d" - " not in group (block %lu)!", - i, (unsigned long) - le32_to_cpu(gdp->bg_inode_table)); + " not in group (block "E3FSBLK")!", + i, inode_table); return 0; } first_block += EXT4_BLOCKS_PER_GROUP(sb); gdp++; } - sbi->s_es->s_free_blocks_count=cpu_to_le32(ext4_count_free_blocks(sb)); + ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb)); sbi->s_es->s_free_inodes_count=cpu_to_le32(ext4_count_free_inodes(sb)); return 1; } @@ -1411,6 +1448,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) int i; int needs_recovery; __le32 features; + __u64 blocks_count; sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) @@ -1620,7 +1658,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount; } - if (le32_to_cpu(es->s_blocks_count) > + if (ext4_blocks_count(es) > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { printk(KERN_ERR "EXT4-fs: filesystem on %s:" " too large to mount safely\n", sb->s_id); @@ -1632,9 +1670,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) if (EXT4_BLOCKS_PER_GROUP(sb) == 0) goto cantfind_ext4; - sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) - - le32_to_cpu(es->s_first_data_block) - 1) - / EXT4_BLOCKS_PER_GROUP(sb)) + 1; + blocks_count = (ext4_blocks_count(es) - + le32_to_cpu(es->s_first_data_block) + + EXT4_BLOCKS_PER_GROUP(sb) - 1); + do_div(blocks_count, EXT4_BLOCKS_PER_GROUP(sb)); + sbi->s_groups_count = blocks_count; db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / EXT4_DESC_PER_BLOCK(sb); sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *), @@ -1949,7 +1989,7 @@ static journal_t *ext4_get_dev_journal(struct super_block *sb, goto out_bdev; } - len = le32_to_cpu(es->s_blocks_count); + len = ext4_blocks_count(es); start = sb_block + 1; brelse(bh); /* we're done with the superblock */ @@ -2119,7 +2159,7 @@ static void ext4_commit_super (struct super_block * sb, if (!sbh) return; es->s_wtime = cpu_to_le32(get_seconds()); - es->s_free_blocks_count = cpu_to_le32(ext4_count_free_blocks(sb)); + ext4_free_blocks_count_set(es, ext4_count_free_blocks(sb)); es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb)); BUFFER_TRACE(sbh, "marking dirty"); mark_buffer_dirty(sbh); @@ -2312,7 +2352,7 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data) ext4_init_journal_params(sb, sbi->s_journal); if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY) || - n_blocks_count > le32_to_cpu(es->s_blocks_count)) { + n_blocks_count > ext4_blocks_count(es)) { if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) { err = -EROFS; goto restore_opts; @@ -2431,10 +2471,10 @@ static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) buf->f_type = EXT4_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; - buf->f_blocks = le32_to_cpu(es->s_blocks_count) - overhead; + buf->f_blocks = ext4_blocks_count(es) - overhead; buf->f_bfree = percpu_counter_sum(&sbi->s_freeblocks_counter); - buf->f_bavail = buf->f_bfree - le32_to_cpu(es->s_r_blocks_count); - if (buf->f_bfree < le32_to_cpu(es->s_r_blocks_count)) + buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es); + if (buf->f_bfree < ext4_r_blocks_count(es)) buf->f_bavail = 0; buf->f_files = le32_to_cpu(es->s_inodes_count); buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter); diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 63ed89f..8e5009e 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -128,10 +128,17 @@ struct ext4_group_desc __le16 bg_free_blocks_count; /* Free blocks count */ __le16 bg_free_inodes_count; /* Free inodes count */ __le16 bg_used_dirs_count; /* Directories count */ - __u16 bg_pad; - __le32 bg_reserved[3]; + __u16 bg_flags; + __le16 bg_block_bitmap_hi; /* Blocks bitmap block MSB */ + __le16 bg_inode_bitmap_hi; /* Inodes bitmap block MSB */ + __le16 bg_inode_table_hi; /* Inodes table block MSB */ + __u16 bg_reserved[3]; }; +#ifdef __KERNEL__ +#include +#include +#endif /* * Macro-instructions used to manage group descriptors */ @@ -194,9 +201,9 @@ struct ext4_group_desc /* Used to pass group descriptor data when online resize is done */ struct ext4_new_group_input { __u32 group; /* Group number for this data */ - __u32 block_bitmap; /* Absolute block number of block bitmap */ - __u32 inode_bitmap; /* Absolute block number of inode bitmap */ - __u32 inode_table; /* Absolute block number of inode table start */ + __u64 block_bitmap; /* Absolute block number of block bitmap */ + __u64 inode_bitmap; /* Absolute block number of inode bitmap */ + __u64 inode_table; /* Absolute block number of inode table start */ __u32 blocks_count; /* Total number of blocks in this group */ __u16 reserved_blocks; /* Number of reserved blocks in this group */ __u16 unused; @@ -205,9 +212,9 @@ struct ext4_new_group_input { /* The struct ext4_new_group_input in kernel space, with free_blocks_count */ struct ext4_new_group_data { __u32 group; - __u32 block_bitmap; - __u32 inode_bitmap; - __u32 inode_table; + __u64 block_bitmap; + __u64 inode_bitmap; + __u64 inode_table; __u32 blocks_count; __u16 reserved_blocks; __u16 unused; @@ -494,14 +501,18 @@ struct ext4_super_block { __u8 s_def_hash_version; /* Default hash version to use */ __u8 s_reserved_char_pad; __u16 s_reserved_word_pad; - __le32 s_default_mount_opts; +/*100*/ __le32 s_default_mount_opts; __le32 s_first_meta_bg; /* First metablock block group */ - __u32 s_reserved[190]; /* Padding to the end of the block */ + __le32 s_mkfs_time; /* When the filesystem was created */ + __le32 s_jnl_blocks[17]; /* Backup of the journal inode */ + /* 64bit support valid if EXT4_FEATURE_COMPAT_64BIT */ +/*150*/ __le32 s_blocks_count_hi; /* Blocks count */ + __le32 s_r_blocks_count_hi; /* Reserved blocks count */ + __le32 s_free_blocks_count_hi; /* Free blocks count */ + __u32 s_reserved[169]; /* Padding to the end of the block */ }; #ifdef __KERNEL__ -#include -#include static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb) { return sb->s_fs_info; @@ -588,12 +599,14 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) #define EXT4_FEATURE_INCOMPAT_JOURNAL_DEV 0x0008 /* Journal device */ #define EXT4_FEATURE_INCOMPAT_META_BG 0x0010 #define EXT4_FEATURE_INCOMPAT_EXTENTS 0x0040 /* extents support */ +#define EXT4_FEATURE_INCOMPAT_64BIT 0x0080 #define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR #define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ EXT4_FEATURE_INCOMPAT_RECOVER| \ EXT4_FEATURE_INCOMPAT_META_BG| \ - EXT4_FEATURE_INCOMPAT_EXTENTS) + EXT4_FEATURE_INCOMPAT_EXTENTS| \ + EXT4_FEATURE_INCOMPAT_64BIT) #define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ EXT4_FEATURE_RO_COMPAT_BTREE_DIR) @@ -888,6 +901,53 @@ extern void ext4_abort (struct super_block *, const char *, const char *, ...) extern void ext4_warning (struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); extern void ext4_update_dynamic_rev (struct super_block *sb); +extern ext4_fsblk_t ext4_block_bitmap(struct ext4_group_desc *bg); +extern ext4_fsblk_t ext4_inode_bitmap(struct ext4_group_desc *bg); +extern ext4_fsblk_t ext4_inode_table(struct ext4_group_desc *bg); +extern void ext4_block_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern void ext4_inode_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern void ext4_inode_table_set(struct ext4_group_desc *bg, ext4_fsblk_t blk); + +static inline ext4_fsblk_t ext4_blocks_count(struct ext4_super_block *es) +{ + return ((ext4_fsblk_t)le32_to_cpu(es->s_blocks_count_hi) << 32) | + le32_to_cpu(es->s_blocks_count); +} + +static inline ext4_fsblk_t ext4_r_blocks_count(struct ext4_super_block *es) +{ + return ((ext4_fsblk_t)le32_to_cpu(es->s_r_blocks_count_hi) << 32) | + le32_to_cpu(es->s_r_blocks_count); +} + +static inline ext4_fsblk_t ext4_free_blocks_count(struct ext4_super_block *es) +{ + return ((ext4_fsblk_t)le32_to_cpu(es->s_free_blocks_count_hi) << 32) | + le32_to_cpu(es->s_free_blocks_count); +} + +static inline void ext4_blocks_count_set(struct ext4_super_block *es, + ext4_fsblk_t blk) +{ + es->s_blocks_count = cpu_to_le32((u32)blk); + es->s_blocks_count_hi = cpu_to_le32(blk >> 32); +} + +static inline void ext4_free_blocks_count_set(struct ext4_super_block *es, + ext4_fsblk_t blk) +{ + es->s_free_blocks_count = cpu_to_le32((u32)blk); + es->s_free_blocks_count_hi = cpu_to_le32(blk >> 32); +} + +static inline void ext4_r_blocks_count_set(struct ext4_super_block *es, + ext4_fsblk_t blk) +{ + es->s_r_blocks_count = cpu_to_le32((u32)blk); + es->s_r_blocks_count_hi = cpu_to_le32(blk >> 32); +} + + #define ext4_std_error(sb, errno) \ do { \ -- cgit v0.10.2 From 2ae0210760aed9d626eaede5b63db95e198f7c8e Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:21:11 -0700 Subject: [PATCH] ext4: blk_type from sector_t to unsigned long long Change ext4 in-kernel block type (ext4_fsblk_t) from sector_t to unsigned long long. Remove ext4 block type string micro E3FSBLK, replaced with "%llu" [akpm@osdl.org: build fix] Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 6887151..df77ea8 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -103,7 +103,7 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) if (!bh) ext4_error (sb, "read_block_bitmap", "Cannot read block bitmap - " - "block_group = %d, block_bitmap = "E3FSBLK, + "block_group = %d, block_bitmap = %llu", block_group, ext4_block_bitmap(desc)); error_out: @@ -148,7 +148,7 @@ restart: rsv = list_entry(n, struct ext4_reserve_window_node, rsv_node); if (verbose) printk("reservation window 0x%p " - "start: "E3FSBLK", end: "E3FSBLK"\n", + "start: %llu, end: %llu\n", rsv, rsv->rsv_start, rsv->rsv_end); if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) { printk("Bad reservation %p (start >= end)\n", @@ -436,7 +436,7 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, block + count > ext4_blocks_count(es)) { ext4_error (sb, "ext4_free_blocks", "Freeing blocks not in datazone - " - "block = "E3FSBLK", count = %lu", block, count); + "block = %llu, count = %lu", block, count); goto error_return; } @@ -468,7 +468,7 @@ do_more: sbi->s_itb_per_group)) ext4_error (sb, "ext4_free_blocks", "Freeing blocks in system zones - " - "Block = "E3FSBLK", count = %lu", + "Block = %llu, count = %lu", block, count); /* @@ -552,7 +552,7 @@ do_more: bit + i, bitmap_bh->b_data)) { jbd_unlock_bh_state(bitmap_bh); ext4_error(sb, __FUNCTION__, - "bit already cleared for block "E3FSBLK, + "bit already cleared for block %llu", (ext4_fsblk_t)(block + i)); jbd_lock_bh_state(bitmap_bh); BUFFER_TRACE(bitmap_bh, "bit already cleared"); @@ -1569,7 +1569,7 @@ allocated: EXT4_SB(sb)->s_itb_per_group)) ext4_error(sb, "ext4_new_block", "Allocating block in system zone - " - "blocks from "E3FSBLK", length %lu", + "blocks from %llu, length %lu", ret_block, num); performed_allocation = 1; @@ -1606,7 +1606,7 @@ allocated: if (ret_block + num - 1 >= ext4_blocks_count(es)) { ext4_error(sb, "ext4_new_block", - "block("E3FSBLK") >= blocks count("E3FSBLK") - " + "block(%llu) >= blocks count(%llu) - " "block_group = %lu, es == %p ", ret_block, ext4_blocks_count(es), group_no, es); goto out; @@ -1705,8 +1705,8 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) bitmap_count += x; } brelse(bitmap_bh); - printk("ext4_count_free_blocks: stored = "E3FSBLK - ", computed = "E3FSBLK", "E3FSBLK"\n", + printk("ext4_count_free_blocks: stored = %llu" + ", computed = %llu, %llu\n", EXT4_FREE_BLOCKS_COUNT(es), desc_count, bitmap_count); return bitmap_count; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index e06e937..f72b756 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -281,10 +281,10 @@ static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) ext_debug("path:"); for (k = 0; k <= l; k++, path++) { if (path->p_idx) { - ext_debug(" %d->"E3FSBLK, le32_to_cpu(path->p_idx->ei_block), + ext_debug(" %d->%llu", le32_to_cpu(path->p_idx->ei_block), idx_pblock(path->p_idx)); } else if (path->p_ext) { - ext_debug(" %d:%d:"E3FSBLK" ", + ext_debug(" %d:%d:%llu ", le32_to_cpu(path->p_ext->ee_block), le16_to_cpu(path->p_ext->ee_len), ext_pblock(path->p_ext)); @@ -308,7 +308,7 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) ex = EXT_FIRST_EXTENT(eh); for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) { - ext_debug("%d:%d:"E3FSBLK" ", le32_to_cpu(ex->ee_block), + ext_debug("%d:%d:%llu ", le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len), ext_pblock(ex)); } ext_debug("\n"); @@ -426,7 +426,7 @@ ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) } path->p_ext = l - 1; - ext_debug(" -> %d:"E3FSBLK":%d ", + ext_debug(" -> %d:%llu:%d ", le32_to_cpu(path->p_ext->ee_block), ext_pblock(path->p_ext), le16_to_cpu(path->p_ext->ee_len)); @@ -687,7 +687,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, path[depth].p_ext++; while (path[depth].p_ext <= EXT_MAX_EXTENT(path[depth].p_hdr)) { - ext_debug("move %d:"E3FSBLK":%d in new leaf "E3FSBLK"\n", + ext_debug("move %d:%llu:%d in new leaf %llu\n", le32_to_cpu(path[depth].p_ext->ee_block), ext_pblock(path[depth].p_ext), le16_to_cpu(path[depth].p_ext->ee_len), @@ -752,7 +752,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, fidx->ei_block = border; ext4_idx_store_pblock(fidx, oldblock); - ext_debug("int.index at %d (block "E3FSBLK"): %lu -> "E3FSBLK"\n", i, + ext_debug("int.index at %d (block %llu): %lu -> %llu\n", i, newblock, (unsigned long) le32_to_cpu(border), oldblock); /* copy indexes */ @@ -764,7 +764,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, BUG_ON(EXT_MAX_INDEX(path[i].p_hdr) != EXT_LAST_INDEX(path[i].p_hdr)); while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) { - ext_debug("%d: move %d:%d in new index "E3FSBLK"\n", i, + ext_debug("%d: move %d:%d in new index %llu\n", i, le32_to_cpu(path[i].p_idx->ei_block), idx_pblock(path[i].p_idx), newblock); @@ -898,7 +898,7 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, neh = ext_inode_hdr(inode); fidx = EXT_FIRST_INDEX(neh); - ext_debug("new root: num %d(%d), lblock %d, ptr "E3FSBLK"\n", + ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n", le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max), le32_to_cpu(fidx->ei_block), idx_pblock(fidx)); @@ -1145,7 +1145,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, /* try to insert block into found extent and return */ if (ex && ext4_can_extents_be_merged(inode, ex, newext)) { - ext_debug("append %d block to %d:%d (from "E3FSBLK")\n", + ext_debug("append %d block to %d:%d (from %llu)\n", le16_to_cpu(newext->ee_len), le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len), ext_pblock(ex)); @@ -1204,7 +1204,7 @@ has_space: if (!nearex) { /* there is no extent in this leaf, create first one */ - ext_debug("first extent in the leaf: %d:"E3FSBLK":%d\n", + ext_debug("first extent in the leaf: %d:%llu:%d\n", le32_to_cpu(newext->ee_block), ext_pblock(newext), le16_to_cpu(newext->ee_len)); @@ -1216,7 +1216,7 @@ has_space: len = EXT_MAX_EXTENT(eh) - nearex; len = (len - 1) * sizeof(struct ext4_extent); len = len < 0 ? 0 : len; - ext_debug("insert %d:"E3FSBLK":%d after: nearest 0x%p, " + ext_debug("insert %d:%llu:%d after: nearest 0x%p, " "move %d from 0x%p to 0x%p\n", le32_to_cpu(newext->ee_block), ext_pblock(newext), @@ -1229,7 +1229,7 @@ has_space: BUG_ON(newext->ee_block == nearex->ee_block); len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext4_extent); len = len < 0 ? 0 : len; - ext_debug("insert %d:"E3FSBLK":%d before: nearest 0x%p, " + ext_debug("insert %d:%llu:%d before: nearest 0x%p, " "move %d from 0x%p to 0x%p\n", le32_to_cpu(newext->ee_block), ext_pblock(newext), @@ -1464,7 +1464,7 @@ ext4_ext_in_cache(struct inode *inode, unsigned long block, ex->ee_block = cpu_to_le32(cex->ec_block); ext4_ext_store_pblock(ex, cex->ec_start); ex->ee_len = cpu_to_le16(cex->ec_len); - ext_debug("%lu cached by %lu:%lu:"E3FSBLK"\n", + ext_debug("%lu cached by %lu:%lu:%llu\n", (unsigned long) block, (unsigned long) cex->ec_block, (unsigned long) cex->ec_len, @@ -1498,7 +1498,7 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1); if ((err = ext4_ext_dirty(handle, inode, path))) return err; - ext_debug("index is empty, remove it, free block "E3FSBLK"\n", leaf); + ext_debug("index is empty, remove it, free block %llu\n", leaf); bh = sb_find_get_block(inode->i_sb, leaf); ext4_forget(handle, 1, inode, bh, leaf); ext4_free_blocks(handle, inode, leaf, 1); @@ -1585,7 +1585,7 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t start; num = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - from; start = ext_pblock(ex) + le16_to_cpu(ex->ee_len) - num; - ext_debug("free last %lu blocks starting "E3FSBLK"\n", num, start); + ext_debug("free last %lu blocks starting %llu\n", num, start); for (i = 0; i < num; i++) { bh = sb_find_get_block(inode->i_sb, start + i); ext4_forget(handle, 0, inode, bh, start + i); @@ -1699,7 +1699,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, if (err) goto out; - ext_debug("new extent: %u:%u:"E3FSBLK"\n", block, num, + ext_debug("new extent: %u:%u:%llu\n", block, num, ext_pblock(ex)); ex--; ex_ee_block = le32_to_cpu(ex->ee_block); @@ -1816,7 +1816,7 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) path[i].p_idx); if (ext4_ext_more_to_rm(path + i)) { /* go to the next level */ - ext_debug("move to level %d (block "E3FSBLK")\n", + ext_debug("move to level %d (block %llu)\n", i + 1, idx_pblock(path[i].p_idx)); memset(path + i + 1, 0, sizeof(*path)); path[i+1].p_bh = @@ -1993,7 +1993,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, newblock = iblock - ee_block + ee_start; /* number of remaining blocks in the extent */ allocated = ee_len - (iblock - ee_block); - ext_debug("%d fit into %lu:%d -> "E3FSBLK"\n", (int) iblock, + ext_debug("%d fit into %lu:%d -> %llu\n", (int) iblock, ee_block, ee_len, newblock); ext4_ext_put_in_cache(inode, ee_block, ee_len, ee_start, EXT4_EXT_CACHE_EXTENT); @@ -2024,7 +2024,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err); if (!newblock) goto out2; - ext_debug("allocate new block: goal "E3FSBLK", found "E3FSBLK"/%lu\n", + ext_debug("allocate new block: goal %llu, found %llu/%lu\n", goal, newblock, allocated); /* try to insert new extent into found leaf and return */ diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index effc38a..99b82b5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2115,7 +2115,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, */ if (!bh) { ext4_error(inode->i_sb, "ext4_free_branches", - "Read failure, inode=%lu, block="E3FSBLK, + "Read failure, inode=%lu, block=%llu", inode->i_ino, nr); continue; } @@ -2466,7 +2466,7 @@ static int __ext4_get_inode_loc(struct inode *inode, if (!bh) { ext4_error (inode->i_sb, "ext4_get_inode_loc", "unable to read inode block - " - "inode=%lu, block="E3FSBLK, + "inode=%lu, block=%llu", inode->i_ino, block); return -EIO; } @@ -2548,7 +2548,7 @@ make_io: if (!buffer_uptodate(bh)) { ext4_error(inode->i_sb, "ext4_get_inode_loc", "unable to read inode block - " - "inode=%lu, block="E3FSBLK, + "inode=%lu, block=%llu", inode->i_ino, block); brelse(bh); return -EIO; diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 3dbf91b..3e96067 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -64,7 +64,7 @@ static int verify_group_input(struct super_block *sb, input->blocks_count); else if (!(bh = sb_bread(sb, end - 1))) ext4_warning(sb, __FUNCTION__, - "Cannot read last block ("E3FSBLK")", + "Cannot read last block (%llu)", end - 1); else if (outside(input->block_bitmap, start, end)) ext4_warning(sb, __FUNCTION__, @@ -94,18 +94,18 @@ static int verify_group_input(struct super_block *sb, else if (inside(input->block_bitmap, start, metaend)) ext4_warning(sb, __FUNCTION__, "Block bitmap (%llu) in GDT table" - " ("E3FSBLK"-"E3FSBLK")", + " (%llu-%llu)", input->block_bitmap, start, metaend - 1); else if (inside(input->inode_bitmap, start, metaend)) ext4_warning(sb, __FUNCTION__, "Inode bitmap (%llu) in GDT table" - " ("E3FSBLK"-"E3FSBLK")", + " (%llu-%llu)", input->inode_bitmap, start, metaend - 1); else if (inside(input->inode_table, start, metaend) || inside(itend - 1, start, metaend)) ext4_warning(sb, __FUNCTION__, - "Inode table ("E3FSBLK"-"E3FSBLK") overlaps" - "GDT table ("E3FSBLK"-"E3FSBLK")", + "Inode table (%llu-%llu) overlaps" + "GDT table (%llu-%llu)", input->inode_table, itend - 1, start, metaend - 1); else err = 0; @@ -344,8 +344,8 @@ static int verify_reserved_gdb(struct super_block *sb, if (le32_to_cpu(*p++) != grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){ ext4_warning(sb, __FUNCTION__, - "reserved GDT "E3FSBLK - " missing grp %d ("E3FSBLK")", + "reserved GDT %llu" + " missing grp %d (%llu)", blk, grp, grp * (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) + @@ -424,7 +424,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, data = (__le32 *)dind->b_data; if (le32_to_cpu(data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)]) != gdblock) { ext4_warning(sb, __FUNCTION__, - "new group %u GDT block "E3FSBLK" not reserved", + "new group %u GDT block %llu not reserved", input->group, gdblock); err = -EINVAL; goto exit_dind; @@ -547,7 +547,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, for (res = 0; res < reserved_gdb; res++, blk++) { if (le32_to_cpu(*data) != blk) { ext4_warning(sb, __FUNCTION__, - "reserved block "E3FSBLK + "reserved block %llu" " not at offset %ld", blk, (long)(data - (__le32 *)dind->b_data)); @@ -941,7 +941,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, o_groups_count = EXT4_SB(sb)->s_groups_count; if (test_opt(sb, DEBUG)) - printk(KERN_DEBUG "EXT4-fs: extending last group from "E3FSBLK" uto "E3FSBLK" blocks\n", + printk(KERN_DEBUG "EXT4-fs: extending last group from %llu uto %llu blocks\n", o_blocks_count, n_blocks_count); if (n_blocks_count == 0 || n_blocks_count == o_blocks_count) @@ -949,7 +949,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, if (n_blocks_count > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { printk(KERN_ERR "EXT4-fs: filesystem on %s:" - " too large to resize to "E3FSBLK" blocks safely\n", + " too large to resize to %llu blocks safely\n", sb->s_id, n_blocks_count); if (sizeof(sector_t) < 8) ext4_warning(sb, __FUNCTION__, @@ -984,7 +984,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, if (o_blocks_count + add < n_blocks_count) ext4_warning(sb, __FUNCTION__, - "will only finish group ("E3FSBLK + "will only finish group (%llu" " blocks, %u new)", o_blocks_count + add, add); @@ -1028,10 +1028,10 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); sb->s_dirt = 1; unlock_super(sb); - ext4_debug("freeing blocks %lu through "E3FSBLK"\n", o_blocks_count, + ext4_debug("freeing blocks %lu through %llu\n", o_blocks_count, o_blocks_count + add); ext4_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); - ext4_debug("freed blocks "E3FSBLK" through "E3FSBLK"\n", o_blocks_count, + ext4_debug("freed blocks %llu through %llu\n", o_blocks_count, o_blocks_count + add); if ((err = ext4_journal_stop(handle))) goto exit_put; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index b91dffd..d844175 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1244,7 +1244,7 @@ static int ext4_check_descriptors (struct super_block * sb) { ext4_error (sb, "ext4_check_descriptors", "Block bitmap for group %d" - " not in group (block "E3FSBLK")!", + " not in group (block %llu)!", i, block_bitmap); return 0; } @@ -1253,7 +1253,7 @@ static int ext4_check_descriptors (struct super_block * sb) { ext4_error (sb, "ext4_check_descriptors", "Inode bitmap for group %d" - " not in group (block "E3FSBLK")!", + " not in group (block %llu)!", i, inode_bitmap); return 0; } @@ -1263,7 +1263,7 @@ static int ext4_check_descriptors (struct super_block * sb) { ext4_error (sb, "ext4_check_descriptors", "Inode table for group %d" - " not in group (block "E3FSBLK")!", + " not in group (block %llu)!", i, inode_table); return 0; } diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 90f7d5c..63233cd 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -233,7 +233,7 @@ ext4_xattr_block_get(struct inode *inode, int name_index, const char *name, atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); if (ext4_xattr_check_block(bh)) { bad_block: ext4_error(inode->i_sb, __FUNCTION__, - "inode %lu: bad block "E3FSBLK, inode->i_ino, + "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; goto cleanup; @@ -375,7 +375,7 @@ ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); if (ext4_xattr_check_block(bh)) { ext4_error(inode->i_sb, __FUNCTION__, - "inode %lu: bad block "E3FSBLK, inode->i_ino, + "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; goto cleanup; @@ -647,7 +647,7 @@ ext4_xattr_block_find(struct inode *inode, struct ext4_xattr_info *i, le32_to_cpu(BHDR(bs->bh)->h_refcount)); if (ext4_xattr_check_block(bs->bh)) { ext4_error(sb, __FUNCTION__, - "inode %lu: bad block "E3FSBLK, inode->i_ino, + "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); error = -EIO; goto cleanup; @@ -848,7 +848,7 @@ cleanup_dquot: bad_block: ext4_error(inode->i_sb, __FUNCTION__, - "inode %lu: bad block "E3FSBLK, inode->i_ino, + "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); goto cleanup; @@ -1077,14 +1077,14 @@ ext4_xattr_delete_inode(handle_t *handle, struct inode *inode) bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); if (!bh) { ext4_error(inode->i_sb, __FUNCTION__, - "inode %lu: block "E3FSBLK" read error", inode->i_ino, + "inode %lu: block %llu read error", inode->i_ino, EXT4_I(inode)->i_file_acl); goto cleanup; } if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || BHDR(bh)->h_blocks != cpu_to_le32(1)) { ext4_error(inode->i_sb, __FUNCTION__, - "inode %lu: bad block "E3FSBLK, inode->i_ino, + "inode %lu: bad block %llu", inode->i_ino, EXT4_I(inode)->i_file_acl); goto cleanup; } diff --git a/include/linux/ext4_fs_i.h b/include/linux/ext4_fs_i.h index 2bed0ef..bb42379 100644 --- a/include/linux/ext4_fs_i.h +++ b/include/linux/ext4_fs_i.h @@ -25,13 +25,7 @@ typedef int ext4_grpblk_t; /* data type for filesystem-wide blocks number */ -typedef sector_t ext4_fsblk_t; - -#if BITS_PER_LONG == 64 -#define E3FSBLK "%lu" -#else -#define E3FSBLK "%llu" -#endif +typedef unsigned long long ext4_fsblk_t; struct ext4_reserve_window { ext4_fsblk_t _rsv_start; /* First byte reserved */ -- cgit v0.10.2 From 9b8f1f0106ab39ad58765d4e7c57189835f51127 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:21:13 -0700 Subject: [PATCH] ext4: removesector_t bits check Previously when in-kernel ext4 block type is sector_t, it's only 4 bits long under some 32bit arch (when CONFIG_LBD is not on). So we need to check the size of sector_t before we read 48bit long on-disk blocks to in-kernel blocks. These checks are unnecessary now as we changed the in-kernel blocks to unsigned longlong. Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index f72b756..926186a 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -53,8 +53,7 @@ static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) ext4_fsblk_t block; block = le32_to_cpu(ex->ee_start); - if (sizeof(ext4_fsblk_t) > 4) - block |= ((ext4_fsblk_t) le16_to_cpu(ex->ee_start_hi) << 31) << 1; + block |= ((ext4_fsblk_t) le16_to_cpu(ex->ee_start_hi) << 31) << 1; return block; } @@ -67,8 +66,7 @@ static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) ext4_fsblk_t block; block = le32_to_cpu(ix->ei_leaf); - if (sizeof(ext4_fsblk_t) > 4) - block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1; + block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1; return block; } @@ -80,8 +78,7 @@ static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) { ex->ee_start = cpu_to_le32((unsigned long) (pb & 0xffffffff)); - if (sizeof(ext4_fsblk_t) > 4) - ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); + ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); } /* @@ -92,8 +89,7 @@ static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) { ix->ei_leaf = cpu_to_le32((unsigned long) (pb & 0xffffffff)); - if (sizeof(ext4_fsblk_t) > 4) - ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); + ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); } static int ext4_ext_check_header(const char *function, struct inode *inode, diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 99b82b5..c05dc57 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2643,9 +2643,8 @@ void ext4_read_inode(struct inode * inode) ei->i_frag_size = raw_inode->i_fsize; #endif ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl); - if ((sizeof(sector_t) > 4) && - (EXT4_SB(inode->i_sb)->s_es->s_creator_os != - cpu_to_le32(EXT4_OS_HURD))) + if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != + cpu_to_le32(EXT4_OS_HURD)) ei->i_file_acl |= ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; if (!S_ISREG(inode->i_mode)) { @@ -2781,9 +2780,8 @@ static int ext4_do_update_inode(handle_t *handle, raw_inode->i_frag = ei->i_frag_no; raw_inode->i_fsize = ei->i_frag_size; #endif - if ((sizeof(sector_t) > 4) && - (EXT4_SB(inode->i_sb)->s_es->s_creator_os != - cpu_to_le32(EXT4_OS_HURD))) + if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != + cpu_to_le32(EXT4_OS_HURD)) raw_inode->i_file_acl_high = cpu_to_le16(ei->i_file_acl >> 32); raw_inode->i_file_acl = cpu_to_le32(ei->i_file_acl); -- cgit v0.10.2 From 18eba7aae080d4a5c0d850ea810e83d11f0a8d77 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Wed, 11 Oct 2006 01:21:13 -0700 Subject: [PATCH] jbd2: switch blks_type from sector_t to ull Similar to ext4, change blocks in JBD2 from sector_t to unsigned long long. Signed-off-by: Mingming Cao Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 1a9ce88..70b2ae1 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -272,7 +272,7 @@ write_out_data: } static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, - sector_t block) + unsigned long long block) { tag->t_blocknr = cpu_to_be32(block & (u32)~0); if (tag_bytes > JBD_TAG_SIZE32) @@ -293,7 +293,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) int bufs; int flags; int err; - sector_t blocknr; + unsigned long long blocknr; char *tagp = NULL; journal_header_t *header; journal_block_tag_t *tag = NULL; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 259e836..10db92c 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -271,7 +271,7 @@ static void journal_kill_thread(journal_t *journal) int jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, struct journal_head **jh_out, - sector_t blocknr) + unsigned long long blocknr) { int need_copy_out = 0; int done_copy_out = 0; @@ -555,7 +555,7 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid) * Log buffer allocation routines: */ -int jbd2_journal_next_log_block(journal_t *journal, sector_t *retp) +int jbd2_journal_next_log_block(journal_t *journal, unsigned long long *retp) { unsigned long blocknr; @@ -579,10 +579,10 @@ int jbd2_journal_next_log_block(journal_t *journal, sector_t *retp) * ready. */ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, - sector_t *retp) + unsigned long long *retp) { int err = 0; - sector_t ret; + unsigned long long ret; if (journal->j_inode) { ret = bmap(journal->j_inode, blocknr); @@ -618,7 +618,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) { struct buffer_head *bh; - sector_t blocknr; + unsigned long long blocknr; int err; err = jbd2_journal_next_log_block(journal, &blocknr); @@ -706,7 +706,7 @@ fail: */ journal_t * jbd2_journal_init_dev(struct block_device *bdev, struct block_device *fs_dev, - sector_t start, int len, int blocksize) + unsigned long long start, int len, int blocksize) { journal_t *journal = journal_init_common(); struct buffer_head *bh; @@ -753,7 +753,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal_t *journal = journal_init_common(); int err; int n; - sector_t blocknr; + unsigned long long blocknr; if (!journal) return NULL; @@ -819,7 +819,7 @@ static void journal_fail_superblock (journal_t *journal) static int journal_reset(journal_t *journal) { journal_superblock_t *sb = journal->j_superblock; - sector_t first, last; + unsigned long long first, last; first = be32_to_cpu(sb->s_first); last = be32_to_cpu(sb->s_maxlen); @@ -853,7 +853,7 @@ static int journal_reset(journal_t *journal) **/ int jbd2_journal_create(journal_t *journal) { - sector_t blocknr; + unsigned long long blocknr; struct buffer_head *bh; journal_superblock_t *sb; int i, err; diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 52054a83..9f10aca 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -70,7 +70,7 @@ static int do_readahead(journal_t *journal, unsigned int start) { int err; unsigned int max, nbufs, next; - sector_t blocknr; + unsigned long long blocknr; struct buffer_head *bh; struct buffer_head * bufs[MAXBUF]; @@ -132,7 +132,7 @@ static int jread(struct buffer_head **bhp, journal_t *journal, unsigned int offset) { int err; - sector_t blocknr; + unsigned long long blocknr; struct buffer_head *bh; *bhp = NULL; @@ -308,9 +308,9 @@ int jbd2_journal_skip_recovery(journal_t *journal) return err; } -static inline sector_t read_tag_block(int tag_bytes, journal_block_tag_t *tag) +static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag_t *tag) { - sector_t block = be32_to_cpu(tag->t_blocknr); + unsigned long long block = be32_to_cpu(tag->t_blocknr); if (tag_bytes > JBD_TAG_SIZE32) block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32; return block; @@ -452,7 +452,7 @@ static int do_one_pass(journal_t *journal, "block %ld in log\n", err, io_block); } else { - sector_t blocknr; + unsigned long long blocknr; J_ASSERT(obh != NULL); blocknr = read_tag_block(tag_bytes, @@ -592,7 +592,7 @@ static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, record_len = 8; while (offset + record_len <= max) { - sector_t blocknr; + unsigned long long blocknr; int err; if (record_len == 4) diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 3310a1d..380d199 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -81,7 +81,7 @@ struct jbd2_revoke_record_s { struct list_head hash; tid_t sequence; /* Used for recovery only */ - sector_t blocknr; + unsigned long long blocknr; }; @@ -106,7 +106,7 @@ static void flush_descriptor(journal_t *, struct journal_head *, int); /* Utility functions to maintain the revoke table */ /* Borrowed from buffer.c: this is a tried and tested block hash function */ -static inline int hash(journal_t *journal, sector_t block) +static inline int hash(journal_t *journal, unsigned long long block) { struct jbd2_revoke_table_s *table = journal->j_revoke; int hash_shift = table->hash_shift; @@ -117,7 +117,7 @@ static inline int hash(journal_t *journal, sector_t block) (hash << (hash_shift - 12))) & (table->hash_size - 1); } -static int insert_revoke_hash(journal_t *journal, sector_t blocknr, +static int insert_revoke_hash(journal_t *journal, unsigned long long blocknr, tid_t seq) { struct list_head *hash_list; @@ -147,7 +147,7 @@ oom: /* Find a revoke record in the journal's hash table. */ static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, - sector_t blocknr) + unsigned long long blocknr) { struct list_head *hash_list; struct jbd2_revoke_record_s *record; @@ -326,7 +326,7 @@ void jbd2_journal_destroy_revoke(journal_t *journal) * by one. */ -int jbd2_journal_revoke(handle_t *handle, sector_t blocknr, +int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr, struct buffer_head *bh_in) { struct buffer_head *bh = NULL; @@ -650,7 +650,7 @@ static void flush_descriptor(journal_t *journal, */ int jbd2_journal_set_revoke(journal_t *journal, - sector_t blocknr, + unsigned long long blocknr, tid_t sequence) { struct jbd2_revoke_record_s *record; @@ -674,7 +674,7 @@ int jbd2_journal_set_revoke(journal_t *journal, */ int jbd2_journal_test_revoke(journal_t *journal, - sector_t blocknr, + unsigned long long blocknr, tid_t sequence) { struct jbd2_revoke_record_s *record; diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 3c939c8..ddb1287 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -741,7 +741,7 @@ struct journal_s */ struct block_device *j_dev; int j_blocksize; - sector_t j_blk_offset; + unsigned long long j_blk_offset; /* * Device which holds the client fs. For internal journal this will be @@ -860,7 +860,7 @@ extern void __journal_clean_data_list(transaction_t *transaction); /* Log buffer allocation */ extern struct journal_head * jbd2_journal_get_descriptor_buffer(journal_t *); -int jbd2_journal_next_log_block(journal_t *, sector_t *); +int jbd2_journal_next_log_block(journal_t *, unsigned long long *); /* Commit management */ extern void jbd2_journal_commit_transaction(journal_t *); @@ -875,7 +875,7 @@ extern int jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, struct journal_head **jh_out, - sector_t blocknr); + unsigned long long blocknr); /* Transaction locking */ extern void __wait_on_journal (journal_t *); @@ -923,7 +923,7 @@ extern void jbd2_journal_unlock_updates (journal_t *); extern journal_t * jbd2_journal_init_dev(struct block_device *bdev, struct block_device *fs_dev, - sector_t start, int len, int bsize); + unsigned long long start, int len, int bsize); extern journal_t * jbd2_journal_init_inode (struct inode *); extern int jbd2_journal_update_format (journal_t *); extern int jbd2_journal_check_used_features @@ -944,7 +944,7 @@ extern void jbd2_journal_abort (journal_t *, int); extern int jbd2_journal_errno (journal_t *); extern void jbd2_journal_ack_err (journal_t *); extern int jbd2_journal_clear_err (journal_t *); -extern int jbd2_journal_bmap(journal_t *, unsigned long, sector_t *); +extern int jbd2_journal_bmap(journal_t *, unsigned long, unsigned long long *); extern int jbd2_journal_force_commit(journal_t *); /* @@ -977,13 +977,13 @@ extern void jbd2_journal_destroy_revoke_caches(void); extern int jbd2_journal_init_revoke_caches(void); extern void jbd2_journal_destroy_revoke(journal_t *); -extern int jbd2_journal_revoke (handle_t *, sector_t, struct buffer_head *); +extern int jbd2_journal_revoke (handle_t *, unsigned long long, struct buffer_head *); extern int jbd2_journal_cancel_revoke(handle_t *, struct journal_head *); extern void jbd2_journal_write_revoke_records(journal_t *, transaction_t *); /* Recovery revoke support */ -extern int jbd2_journal_set_revoke(journal_t *, sector_t, tid_t); -extern int jbd2_journal_test_revoke(journal_t *, sector_t, tid_t); +extern int jbd2_journal_set_revoke(journal_t *, unsigned long long, tid_t); +extern int jbd2_journal_test_revoke(journal_t *, unsigned long long, tid_t); extern void jbd2_journal_clear_revoke(journal_t *); extern void jbd2_journal_switch_revoke_table(journal_t *journal); -- cgit v0.10.2 From 0d1ee42f27d30eed1659f3e85bcbbc7b3711f61f Mon Sep 17 00:00:00 2001 From: Alexandre Ratchov Date: Wed, 11 Oct 2006 01:21:14 -0700 Subject: [PATCH] ext4: allow larger descriptor size make block group descriptor larger. Signed-off-by: Alexandre Ratchov Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index df77ea8..3dacb12 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -74,10 +74,12 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, return NULL; } - desc = (struct ext4_group_desc *) sbi->s_group_desc[group_desc]->b_data; + desc = (struct ext4_group_desc *)( + (__u8 *)sbi->s_group_desc[group_desc]->b_data + + offset * EXT4_DESC_SIZE(sb)); if (bh) *bh = sbi->s_group_desc[group_desc]; - return desc + offset; + return desc; } /** diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c05dc57..d03e7d8 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2432,14 +2432,16 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, return 0; } - gdp = (struct ext4_group_desc *)bh->b_data; + gdp = (struct ext4_group_desc *)((__u8 *)bh->b_data + + desc * EXT4_DESC_SIZE(sb)); /* * Figure out the offset within the block group inode table */ offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * EXT4_INODE_SIZE(sb); - block = ext4_inode_table(gdp + desc) + - (offset >> EXT4_BLOCK_SIZE_BITS(sb)); + block = ext4_inode_table(gdp) + (offset >> EXT4_BLOCK_SIZE_BITS(sb)); + + iloc->block_group = block_group; iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index d844175..bc8848b 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1268,7 +1268,8 @@ static int ext4_check_descriptors (struct super_block * sb) return 0; } first_block += EXT4_BLOCKS_PER_GROUP(sb); - gdp++; + gdp = (struct ext4_group_desc *) + ((__u8 *)gdp + EXT4_DESC_SIZE(sb)); } ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb)); @@ -1619,7 +1620,18 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sbi->s_frag_size, blocksize); goto failed_mount; } - sbi->s_frags_per_block = 1; + sbi->s_desc_size = le16_to_cpu(es->s_desc_size); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) { + if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE || + sbi->s_desc_size > EXT4_MAX_DESC_SIZE || + sbi->s_desc_size & (sbi->s_desc_size - 1)) { + printk(KERN_ERR + "EXT4-fs: unsupported descriptor size %ld\n", + sbi->s_desc_size); + goto failed_mount; + } + } else + sbi->s_desc_size = EXT4_MIN_DESC_SIZE; sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group); sbi->s_frags_per_group = le32_to_cpu(es->s_frags_per_group); sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group); @@ -1630,7 +1642,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto cantfind_ext4; sbi->s_itb_per_group = sbi->s_inodes_per_group / sbi->s_inodes_per_block; - sbi->s_desc_per_block = blocksize / sizeof(struct ext4_group_desc); + sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb); sbi->s_sbh = bh; sbi->s_mount_state = le16_to_cpu(es->s_state); sbi->s_addr_per_block_bits = log2(EXT4_ADDR_PER_BLOCK(sb)); diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 8e5009e..a3df2af 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -142,6 +142,9 @@ struct ext4_group_desc /* * Macro-instructions used to manage group descriptors */ +#define EXT4_MIN_DESC_SIZE 32 +#define EXT4_MAX_DESC_SIZE EXT4_MIN_BLOCK_SIZE +#define EXT4_DESC_SIZE(s) (EXT4_SB(s)->s_desc_size) #ifdef __KERNEL__ # define EXT4_BLOCKS_PER_GROUP(s) (EXT4_SB(s)->s_blocks_per_group) # define EXT4_DESC_PER_BLOCK(s) (EXT4_SB(s)->s_desc_per_block) @@ -149,7 +152,7 @@ struct ext4_group_desc # define EXT4_DESC_PER_BLOCK_BITS(s) (EXT4_SB(s)->s_desc_per_block_bits) #else # define EXT4_BLOCKS_PER_GROUP(s) ((s)->s_blocks_per_group) -# define EXT4_DESC_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (struct ext4_group_desc)) +# define EXT4_DESC_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / EXT4_DESC_SIZE(s)) # define EXT4_INODES_PER_GROUP(s) ((s)->s_inodes_per_group) #endif @@ -474,7 +477,7 @@ struct ext4_super_block { * things it doesn't understand... */ __le32 s_first_ino; /* First non-reserved inode */ - __le16 s_inode_size; /* size of inode structure */ + __le16 s_inode_size; /* size of inode structure */ __le16 s_block_group_nr; /* block group # of this superblock */ __le32 s_feature_compat; /* compatible feature set */ /*60*/ __le32 s_feature_incompat; /* incompatible feature set */ @@ -500,7 +503,7 @@ struct ext4_super_block { __le32 s_hash_seed[4]; /* HTREE hash seed */ __u8 s_def_hash_version; /* Default hash version to use */ __u8 s_reserved_char_pad; - __u16 s_reserved_word_pad; + __le16 s_desc_size; /* size of group descriptor */ /*100*/ __le32 s_default_mount_opts; __le32 s_first_meta_bg; /* First metablock block group */ __le32 s_mkfs_time; /* When the filesystem was created */ diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h index ce7a844..691a713 100644 --- a/include/linux/ext4_fs_sb.h +++ b/include/linux/ext4_fs_sb.h @@ -29,6 +29,7 @@ */ struct ext4_sb_info { unsigned long s_frag_size; /* Size of a fragment in bytes */ + unsigned long s_desc_size; /* Size of a group descriptor in bytes */ unsigned long s_frags_per_block;/* Number of fragments per block */ unsigned long s_inodes_per_block;/* Number of inodes per block */ unsigned long s_frags_per_group;/* Number of fragments in a group */ -- cgit v0.10.2 From 8fadc14323684c547f74cf2f4d13517c6c264731 Mon Sep 17 00:00:00 2001 From: Alexandre Ratchov Date: Wed, 11 Oct 2006 01:21:15 -0700 Subject: [PATCH] ext4: move block number hi bits move '_hi' bits of block numbers in the larger part of the block group descriptor structure Signed-off-by: Alexandre Ratchov Signed-off-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 3dacb12..3e85886 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -101,13 +101,13 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) desc = ext4_get_group_desc (sb, block_group, NULL); if (!desc) goto error_out; - bh = sb_bread(sb, ext4_block_bitmap(desc)); + bh = sb_bread(sb, ext4_block_bitmap(sb, desc)); if (!bh) ext4_error (sb, "read_block_bitmap", "Cannot read block bitmap - " "block_group = %d, block_bitmap = %llu", block_group, - ext4_block_bitmap(desc)); + ext4_block_bitmap(sb, desc)); error_out: return bh; } @@ -463,10 +463,10 @@ do_more: if (!desc) goto error_return; - if (in_range(ext4_block_bitmap(desc), block, count) || - in_range(ext4_inode_bitmap(desc), block, count) || - in_range(block, ext4_inode_table(desc), sbi->s_itb_per_group) || - in_range(block + count - 1, ext4_inode_table(desc), + if (in_range(ext4_block_bitmap(sb, desc), block, count) || + in_range(ext4_inode_bitmap(sb, desc), block, count) || + in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) || + in_range(block + count - 1, ext4_inode_table(sb, desc), sbi->s_itb_per_group)) ext4_error (sb, "ext4_free_blocks", "Freeing blocks in system zones - " @@ -1563,11 +1563,11 @@ allocated: ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no); - if (in_range(ext4_block_bitmap(gdp), ret_block, num) || - in_range(ext4_block_bitmap(gdp), ret_block, num) || - in_range(ret_block, ext4_inode_table(gdp), + if (in_range(ext4_block_bitmap(sb, gdp), ret_block, num) || + in_range(ext4_block_bitmap(sb, gdp), ret_block, num) || + in_range(ret_block, ext4_inode_table(sb, gdp), EXT4_SB(sb)->s_itb_per_group) || - in_range(ret_block + num - 1, ext4_inode_table(gdp), + in_range(ret_block + num - 1, ext4_inode_table(sb, gdp), EXT4_SB(sb)->s_itb_per_group)) ext4_error(sb, "ext4_new_block", "Allocating block in system zone - " diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 959b7fa..75608e1 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -60,12 +60,12 @@ read_inode_bitmap(struct super_block * sb, unsigned long block_group) if (!desc) goto error_out; - bh = sb_bread(sb, ext4_inode_bitmap(desc)); + bh = sb_bread(sb, ext4_inode_bitmap(sb, desc)); if (!bh) ext4_error(sb, "read_inode_bitmap", "Cannot read inode bitmap - " "block_group = %lu, inode_bitmap = %llu", - block_group, ext4_inode_bitmap(desc)); + block_group, ext4_inode_bitmap(sb, desc)); error_out: return bh; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index d03e7d8..0a60ec5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2439,9 +2439,8 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, */ offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * EXT4_INODE_SIZE(sb); - block = ext4_inode_table(gdp) + (offset >> EXT4_BLOCK_SIZE_BITS(sb)); - - + block = ext4_inode_table(sb, gdp) + + (offset >> EXT4_BLOCK_SIZE_BITS(sb)); iloc->block_group = block_group; iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); @@ -2508,7 +2507,7 @@ static int __ext4_get_inode_loc(struct inode *inode, goto make_io; bitmap_bh = sb_getblk(inode->i_sb, - ext4_inode_bitmap(desc)); + ext4_inode_bitmap(inode->i_sb, desc)); if (!bitmap_bh) goto make_io; diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 3e96067..1e95780 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -834,9 +834,9 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) /* Update group descriptor block for new group */ gdp = (struct ext4_group_desc *)primary->b_data + gdb_off; - ext4_block_bitmap_set(gdp, input->block_bitmap); /* LV FIXME */ - ext4_inode_bitmap_set(gdp, input->inode_bitmap); /* LV FIXME */ - ext4_inode_table_set(gdp, input->inode_table); /* LV FIXME */ + ext4_block_bitmap_set(sb, gdp, input->block_bitmap); /* LV FIXME */ + ext4_inode_bitmap_set(sb, gdp, input->inode_bitmap); /* LV FIXME */ + ext4_inode_table_set(sb, gdp, input->inode_table); /* LV FIXME */ gdp->bg_free_blocks_count = cpu_to_le16(input->free_blocks_count); gdp->bg_free_inodes_count = cpu_to_le16(EXT4_INODES_PER_GROUP(sb)); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index bc8848b..811011f 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -63,40 +63,52 @@ static void ext4_write_super (struct super_block * sb); static void ext4_write_super_lockfs(struct super_block *sb); -ext4_fsblk_t ext4_block_bitmap(struct ext4_group_desc *bg) +ext4_fsblk_t ext4_block_bitmap(struct super_block *sb, + struct ext4_group_desc *bg) { return le32_to_cpu(bg->bg_block_bitmap) | - ((ext4_fsblk_t)le16_to_cpu(bg->bg_block_bitmap_hi) << 32); + (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ? + (ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0); } -ext4_fsblk_t ext4_inode_bitmap(struct ext4_group_desc *bg) +ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb, + struct ext4_group_desc *bg) { return le32_to_cpu(bg->bg_inode_bitmap) | - ((ext4_fsblk_t)le16_to_cpu(bg->bg_inode_bitmap_hi) << 32); + (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ? + (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0); } -ext4_fsblk_t ext4_inode_table(struct ext4_group_desc *bg) +ext4_fsblk_t ext4_inode_table(struct super_block *sb, + struct ext4_group_desc *bg) { return le32_to_cpu(bg->bg_inode_table) | - ((ext4_fsblk_t)le16_to_cpu(bg->bg_inode_table_hi) << 32); + (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ? + (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0); } -void ext4_block_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk) +void ext4_block_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk) { bg->bg_block_bitmap = cpu_to_le32((u32)blk); - bg->bg_block_bitmap_hi = cpu_to_le16(blk >> 32); + if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT) + bg->bg_block_bitmap_hi = cpu_to_le32(blk >> 32); } -void ext4_inode_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk) +void ext4_inode_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk) { bg->bg_inode_bitmap = cpu_to_le32((u32)blk); - bg->bg_inode_bitmap_hi = cpu_to_le16(blk >> 32); + if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT) + bg->bg_inode_bitmap_hi = cpu_to_le32(blk >> 32); } -void ext4_inode_table_set(struct ext4_group_desc *bg, ext4_fsblk_t blk) +void ext4_inode_table_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk) { bg->bg_inode_table = cpu_to_le32((u32)blk); - bg->bg_inode_table_hi = cpu_to_le16(blk >> 32); + if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT) + bg->bg_inode_table_hi = cpu_to_le32(blk >> 32); } /* @@ -1239,7 +1251,7 @@ static int ext4_check_descriptors (struct super_block * sb) if ((i % EXT4_DESC_PER_BLOCK(sb)) == 0) gdp = (struct ext4_group_desc *) sbi->s_group_desc[desc_block++]->b_data; - block_bitmap = ext4_block_bitmap(gdp); + block_bitmap = ext4_block_bitmap(sb, gdp); if (block_bitmap < first_block || block_bitmap > last_block) { ext4_error (sb, "ext4_check_descriptors", @@ -1248,7 +1260,7 @@ static int ext4_check_descriptors (struct super_block * sb) i, block_bitmap); return 0; } - inode_bitmap = ext4_inode_bitmap(gdp); + inode_bitmap = ext4_inode_bitmap(sb, gdp); if (inode_bitmap < first_block || inode_bitmap > last_block) { ext4_error (sb, "ext4_check_descriptors", @@ -1257,7 +1269,7 @@ static int ext4_check_descriptors (struct super_block * sb) i, inode_bitmap); return 0; } - inode_table = ext4_inode_table(gdp); + inode_table = ext4_inode_table(sb, gdp); if (inode_table < first_block || inode_table + sbi->s_itb_per_group > last_block) { @@ -1622,11 +1634,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) } sbi->s_desc_size = le16_to_cpu(es->s_desc_size); if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) { - if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE || + if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT || sbi->s_desc_size > EXT4_MAX_DESC_SIZE || sbi->s_desc_size & (sbi->s_desc_size - 1)) { printk(KERN_ERR - "EXT4-fs: unsupported descriptor size %ld\n", + "EXT4-fs: unsupported descriptor size %lu\n", sbi->s_desc_size); goto failed_mount; } diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index a3df2af..296609b 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -129,10 +129,10 @@ struct ext4_group_desc __le16 bg_free_inodes_count; /* Free inodes count */ __le16 bg_used_dirs_count; /* Directories count */ __u16 bg_flags; - __le16 bg_block_bitmap_hi; /* Blocks bitmap block MSB */ - __le16 bg_inode_bitmap_hi; /* Inodes bitmap block MSB */ - __le16 bg_inode_table_hi; /* Inodes table block MSB */ - __u16 bg_reserved[3]; + __u32 bg_reserved[3]; + __le32 bg_block_bitmap_hi; /* Blocks bitmap block MSB */ + __le32 bg_inode_bitmap_hi; /* Inodes bitmap block MSB */ + __le32 bg_inode_table_hi; /* Inodes table block MSB */ }; #ifdef __KERNEL__ @@ -143,6 +143,7 @@ struct ext4_group_desc * Macro-instructions used to manage group descriptors */ #define EXT4_MIN_DESC_SIZE 32 +#define EXT4_MIN_DESC_SIZE_64BIT 64 #define EXT4_MAX_DESC_SIZE EXT4_MIN_BLOCK_SIZE #define EXT4_DESC_SIZE(s) (EXT4_SB(s)->s_desc_size) #ifdef __KERNEL__ @@ -904,12 +905,18 @@ extern void ext4_abort (struct super_block *, const char *, const char *, ...) extern void ext4_warning (struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); extern void ext4_update_dynamic_rev (struct super_block *sb); -extern ext4_fsblk_t ext4_block_bitmap(struct ext4_group_desc *bg); -extern ext4_fsblk_t ext4_inode_bitmap(struct ext4_group_desc *bg); -extern ext4_fsblk_t ext4_inode_table(struct ext4_group_desc *bg); -extern void ext4_block_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk); -extern void ext4_inode_bitmap_set(struct ext4_group_desc *bg, ext4_fsblk_t blk); -extern void ext4_inode_table_set(struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern ext4_fsblk_t ext4_block_bitmap(struct super_block *sb, + struct ext4_group_desc *bg); +extern ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb, + struct ext4_group_desc *bg); +extern ext4_fsblk_t ext4_inode_table(struct super_block *sb, + struct ext4_group_desc *bg); +extern void ext4_block_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern void ext4_inode_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk); +extern void ext4_inode_table_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk); static inline ext4_fsblk_t ext4_blocks_count(struct ext4_super_block *es) { -- cgit v0.10.2 From 72b64b594081ef0a0717f6aad77e891c72ed4afa Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:21:18 -0700 Subject: [PATCH] ext4 uninline ext4_get_group_no_and_offset() Way too big to inline. Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 3e85886..402475a 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -25,6 +25,24 @@ */ /* + * Calculate the block group number and offset, given a block number + */ +void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, + unsigned long *blockgrpp, ext4_grpblk_t *offsetp) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + ext4_grpblk_t offset; + + blocknr = blocknr - le32_to_cpu(es->s_first_data_block); + offset = sector_div(blocknr, EXT4_BLOCKS_PER_GROUP(sb)); + if (offsetp) + *offsetp = offset; + if (blockgrpp) + *blockgrpp = blocknr; + +} + +/* * The free blocks are managed by bitmaps. A file system contains several * blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap * block for inodes, N blocks for the inode table and data blocks. diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h index 296609b..498503e 100644 --- a/include/linux/ext4_fs.h +++ b/include/linux/ext4_fs.h @@ -769,26 +769,8 @@ ext4_group_first_block_no(struct super_block *sb, unsigned long group_no) */ #define ERR_BAD_DX_DIR -75000 -/* - * This function calculate the block group number and offset, - * given a block number - */ - -static inline void ext4_get_group_no_and_offset(struct super_block * sb, - ext4_fsblk_t blocknr, unsigned long* blockgrpp, - ext4_grpblk_t *offsetp) -{ - struct ext4_super_block *es = EXT4_SB(sb)->s_es; - ext4_grpblk_t offset; - - blocknr = blocknr - le32_to_cpu(es->s_first_data_block); - offset = sector_div(blocknr, EXT4_BLOCKS_PER_GROUP(sb)); - if (offsetp) - *offsetp = offset; - if (blockgrpp) - *blockgrpp = blocknr; - -} +void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, + unsigned long *blockgrpp, ext4_grpblk_t *offsetp); /* * Function prototypes -- cgit v0.10.2 From f4e5bc244f23ee024a4dfa034b591b219b2bfb8f Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:21:19 -0700 Subject: [PATCH] ext4 64 bit divide fix With CONFIG_LBD=n, sector_div() expands to a plain old divide. But ext4 is _not_ passing in a sector_t as the first argument, so... fs/built-in.o: In function `ext4_get_group_no_and_offset': fs/ext4/balloc.c:39: undefined reference to `__umoddi3' fs/ext4/balloc.c:41: undefined reference to `__udivdi3' fs/built-in.o: In function `find_group_orlov': fs/ext4/ialloc.c:278: undefined reference to `__udivdi3' fs/built-in.o: In function `ext4_fill_super': fs/ext4/super.c:1488: undefined reference to `__udivdi3' fs/ext4/super.c:1488: undefined reference to `__umoddi3' fs/ext4/super.c:1594: undefined reference to `__udivdi3' fs/ext4/super.c:1601: undefined reference to `__umoddi3' Fix that up by calling do_div() directly. Also cast the arg to u64. do_div() is only defined on u64, and ext4_fsblk_t is supposed to be opaque. Note especially the changes to find_group_orlov(). It was attempting to do do_div(int, unsigned long long); which is royally screwed up. Switched it to plain old divide. Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 402475a..5d45582 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -34,7 +34,7 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, ext4_grpblk_t offset; blocknr = blocknr - le32_to_cpu(es->s_first_data_block); - offset = sector_div(blocknr, EXT4_BLOCKS_PER_GROUP(sb)); + offset = do_div(blocknr, EXT4_BLOCKS_PER_GROUP(sb)); if (offsetp) *offsetp = offset; if (blockgrpp) diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 75608e1..c88b439 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -275,7 +275,7 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) avefreei = freei / ngroups; freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter); avefreeb = freeb; - sector_div(avefreeb, ngroups); + do_div(avefreeb, ngroups); ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter); if ((parent == sb->s_root->d_inode) || @@ -305,14 +305,14 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent) } blocks_per_dir = ext4_blocks_count(es) - freeb; - sector_div(blocks_per_dir, ndirs); + do_div(blocks_per_dir, ndirs); max_dirs = ndirs / ngroups + inodes_per_group / 16; min_inodes = avefreei - inodes_per_group / 4; min_blocks = avefreeb - EXT4_BLOCKS_PER_GROUP(sb) / 4; max_debt = EXT4_BLOCKS_PER_GROUP(sb); - sector_div(max_debt, max(blocks_per_dir, (ext4_fsblk_t)BLOCK_COST)); + max_debt /= max_t(int, blocks_per_dir, BLOCK_COST); if (max_debt * INODE_COST > inodes_per_group) max_debt = inodes_per_group / INODE_COST; if (max_debt > 255) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 811011f..f7ea022 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1485,7 +1485,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) */ if (blocksize != EXT4_MIN_BLOCK_SIZE) { logic_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; - offset = sector_div(logic_sb_block, blocksize); + offset = do_div(logic_sb_block, blocksize); } else { logic_sb_block = sb_block; } @@ -1591,7 +1591,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) brelse (bh); sb_set_blocksize(sb, blocksize); logic_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; - offset = sector_div(logic_sb_block, blocksize); + offset = do_div(logic_sb_block, blocksize); bh = sb_bread(sb, logic_sb_block); if (!bh) { printk(KERN_ERR -- cgit v0.10.2 From 70bbb3e0a07c1ff2dd383761b12f865b6002a7a0 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:21:20 -0700 Subject: [PATCH] ext4: rename logic_sb_block I assume this means "logical sb block". So call it that. I still don't understand the name though. A block is a block. What's different about this one? Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f7ea022..6cf1f18 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1423,8 +1423,7 @@ static loff_t ext4_max_size(int bits) } static ext4_fsblk_t descriptor_loc(struct super_block *sb, - ext4_fsblk_t logic_sb_block, - int nr) + ext4_fsblk_t logical_sb_block, int nr) { struct ext4_sb_info *sbi = EXT4_SB(sb); unsigned long bg, first_meta_bg; @@ -1434,7 +1433,7 @@ static ext4_fsblk_t descriptor_loc(struct super_block *sb, if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG) || nr < first_meta_bg) - return (logic_sb_block + nr + 1); + return logical_sb_block + nr + 1; bg = sbi->s_desc_per_block * nr; if (ext4_bg_has_super(sb, bg)) has_super = 1; @@ -1449,7 +1448,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) struct ext4_sb_info *sbi; ext4_fsblk_t block; ext4_fsblk_t sb_block = get_sb_block(&data); - ext4_fsblk_t logic_sb_block; + ext4_fsblk_t logical_sb_block; unsigned long offset = 0; unsigned int journal_inum = 0; unsigned long journal_devnum = 0; @@ -1484,13 +1483,13 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) * block sizes. We need to calculate the offset from buffer start. */ if (blocksize != EXT4_MIN_BLOCK_SIZE) { - logic_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; - offset = do_div(logic_sb_block, blocksize); + logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; + offset = do_div(logical_sb_block, blocksize); } else { - logic_sb_block = sb_block; + logical_sb_block = sb_block; } - if (!(bh = sb_bread(sb, logic_sb_block))) { + if (!(bh = sb_bread(sb, logical_sb_block))) { printk (KERN_ERR "EXT4-fs: unable to read superblock\n"); goto out_fail; } @@ -1590,9 +1589,9 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) brelse (bh); sb_set_blocksize(sb, blocksize); - logic_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; - offset = do_div(logic_sb_block, blocksize); - bh = sb_bread(sb, logic_sb_block); + logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; + offset = do_div(logical_sb_block, blocksize); + bh = sb_bread(sb, logical_sb_block); if (!bh) { printk(KERN_ERR "EXT4-fs: Can't read superblock on 2nd try.\n"); @@ -1711,7 +1710,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) bgl_lock_init(&sbi->s_blockgroup_lock); for (i = 0; i < db_count; i++) { - block = descriptor_loc(sb, logic_sb_block, i); + block = descriptor_loc(sb, logical_sb_block, i); sbi->s_group_desc[i] = sb_bread(sb, block); if (!sbi->s_group_desc[i]) { printk (KERN_ERR "EXT4-fs: " -- cgit v0.10.2 From ceea16bf85fb46035dda1db263ca29b0e07e22ba Mon Sep 17 00:00:00 2001 From: Dmitry Mishin Date: Wed, 11 Oct 2006 01:21:21 -0700 Subject: [PATCH] ext4: errors behaviour fix Current error behaviour for ext2 and ext3 filesystems does not fully correspond to the documentation and should be fixed. According to man 8 mount, ext2 and ext3 file systems allow to set one of 3 different on-errors behaviours: ---- start of quote man 8 mount ---- errors=continue / errors=remount-ro / errors=panic Define the behaviour when an error is encountered. (Either ignore errors and just mark the file system erroneous and continue, or remount the file system read-only, or panic and halt the system.) The default is set in the filesystem superblock, and can be changed using tune2fs(8). ---- end of quote ---- However EXT3_ERRORS_CONTINUE is not read from the superblock, and thus ERRORS_CONT is not saved on the sbi->s_mount_opt. It leads to the incorrect handle of errors on ext3. Then we've checked corresponding code in ext2 and discovered that it is buggy as well: - EXT2_ERRORS_CONTINUE is not read from the superblock (the same); - parse_option() does not clean the alternative values and thus something like (ERRORS_CONT|ERRORS_RO) can be set; - if options are omitted, parse_option() does not set any of these options. Therefore it is possible to set any combination of these options on the ext2: - none of them may be set: EXT2_ERRORS_CONTINUE on superblock / empty mount options; - any of them may be set using mount options; - 2 any options may be set: by using EXT2_ERRORS_RO/EXT2_ERRORS_PANIC on the superblock and other value in mount options; - and finally all three options may be set by adding third option in remount. Currently ext2 uses these values only in ext2_error() and it is not leading to any noticeable troubles. However somebody may be discouraged when he will try to workaround EXT2_ERRORS_PANIC on the superblock by using errors=continue in mount options. This patch: EXT4_ERRORS_CONTINUE should be taken from the superblock as default value for error behaviour. Signed-off-by: Dmitry Mishin Acked-by: Vasily Averin Acked-by: Kirill Korotaev Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 6cf1f18..1fa3bdc 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1526,6 +1526,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, ERRORS_PANIC); else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_RO) set_opt(sbi->s_mount_opt, ERRORS_RO); + else + set_opt(sbi->s_mount_opt, ERRORS_CONT); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); -- cgit v0.10.2 From 63f5793351d821749979e36889f9c089c6028c83 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:21:24 -0700 Subject: [PATCH] ext4 whitespace cleanups Someone's tab key is emitting spaces. Attempt to repair some of the damage. Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index 0a965dd..9e88254 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -141,7 +141,7 @@ ext4_iget_acl(struct inode *inode, struct posix_acl **i_acl) static inline void ext4_iset_acl(struct inode *inode, struct posix_acl **i_acl, - struct posix_acl *acl) + struct posix_acl *acl) { spin_lock(&inode->i_lock); if (*i_acl != EXT4_ACL_NOT_CACHED) @@ -375,7 +375,7 @@ int ext4_acl_chmod(struct inode *inode) { struct posix_acl *acl, *clone; - int error; + int error; if (S_ISLNK(inode->i_mode)) return -EOPNOTSUPP; diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 18ac173a..f859578 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -188,8 +188,9 @@ revalidate: de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); if (!ext4_check_dir_entry ("ext4_readdir", inode, de, bh, offset)) { - /* On error, skip the f_pos to the - next block. */ + /* + * On error, skip the f_pos to the next block + */ filp->f_pos = (filp->f_pos | (sb->s_blocksize - 1)) + 1; brelse (bh); @@ -508,7 +509,7 @@ finished: static int ext4_release_dir (struct inode * inode, struct file * filp) { - if (filp->private_data) + if (filp->private_data) ext4_htree_free_dir_info(filp->private_data); return 0; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 926186a..2608dce 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1014,7 +1014,7 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) * returns first allocated block from next leaf or EXT_MAX_BLOCK */ static unsigned ext4_ext_next_leaf_block(struct inode *inode, - struct ext4_ext_path *path) + struct ext4_ext_path *path) { int depth; @@ -1097,8 +1097,8 @@ static int inline ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, struct ext4_extent *ex2) { - if (le32_to_cpu(ex1->ee_block) + le16_to_cpu(ex1->ee_len) - != le32_to_cpu(ex2->ee_block)) + if (le32_to_cpu(ex1->ee_block) + le16_to_cpu(ex1->ee_len) != + le32_to_cpu(ex2->ee_block)) return 0; /* @@ -1113,7 +1113,7 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, return 0; #endif - if (ext_pblock(ex1) + le16_to_cpu(ex1->ee_len) == ext_pblock(ex2)) + if (ext_pblock(ex1) + le16_to_cpu(ex1->ee_len) == ext_pblock(ex2)) return 1; return 0; } @@ -2008,9 +2008,9 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, goto out2; } /* - * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary. - */ + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary. + */ if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) ext4_init_block_alloc_info(inode); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index f98b999..8b1bd03 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -241,13 +241,13 @@ static inline unsigned dx_node_limit (struct inode *dir) #ifdef DX_DEBUG static void dx_show_index (char * label, struct dx_entry *entries) { - int i, n = dx_get_count (entries); + int i, n = dx_get_count (entries); printk("%s index ", label); - for (i = 0; i < n; i++) - { - printk("%x->%u ", i? dx_get_hash(entries + i): 0, dx_get_block(entries + i)); - } - printk("\n"); + for (i = 0; i < n; i++) { + printk("%x->%u ", i? dx_get_hash(entries + i) : + 0, dx_get_block(entries + i)); + } + printk("\n"); } struct stats @@ -688,28 +688,26 @@ static int dx_make_map (struct ext4_dir_entry_2 *de, int size, static void dx_sort_map (struct dx_map_entry *map, unsigned count) { - struct dx_map_entry *p, *q, *top = map + count - 1; - int more; - /* Combsort until bubble sort doesn't suck */ - while (count > 2) - { - count = count*10/13; - if (count - 9 < 2) /* 9, 10 -> 11 */ - count = 11; - for (p = top, q = p - count; q >= map; p--, q--) - if (p->hash < q->hash) - swap(*p, *q); - } - /* Garden variety bubble sort */ - do { - more = 0; - q = top; - while (q-- > map) - { - if (q[1].hash >= q[0].hash) + struct dx_map_entry *p, *q, *top = map + count - 1; + int more; + /* Combsort until bubble sort doesn't suck */ + while (count > 2) { + count = count*10/13; + if (count - 9 < 2) /* 9, 10 -> 11 */ + count = 11; + for (p = top, q = p - count; q >= map; p--, q--) + if (p->hash < q->hash) + swap(*p, *q); + } + /* Garden variety bubble sort */ + do { + more = 0; + q = top; + while (q-- > map) { + if (q[1].hash >= q[0].hash) continue; - swap(*(q+1), *q); - more = 1; + swap(*(q+1), *q); + more = 1; } } while(more); } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 1fa3bdc..b4b022a 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1191,9 +1191,10 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, "running e2fsck is recommended\n"); #if 0 /* @@@ We _will_ want to clear the valid bit if we find - inconsistencies, to force a fsck at reboot. But for - a plain journaled filesystem we can keep it set as - valid forever! :) */ + * inconsistencies, to force a fsck at reboot. But for + * a plain journaled filesystem we can keep it set as + * valid forever! :) + */ es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT4_VALID_FS); #endif if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) @@ -1791,8 +1792,9 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) switch (test_opt(sb, DATA_FLAGS)) { case 0: /* No mode set, assume a default based on the journal - capabilities: ORDERED_DATA if the journal can - cope, else JOURNAL_DATA */ + * capabilities: ORDERED_DATA if the journal can + * cope, else JOURNAL_DATA + */ if (jbd2_journal_check_available_features (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) set_opt(sbi->s_mount_opt, ORDERED_DATA); @@ -2802,7 +2804,7 @@ static int __init init_ext4_fs(void) err = init_inodecache(); if (err) goto out1; - err = register_filesystem(&ext4dev_fs_type); + err = register_filesystem(&ext4dev_fs_type); if (err) goto out; return 0; -- cgit v0.10.2 From fc513a333b45f8913d40c8241a0cb61be79e1c60 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Wed, 11 Oct 2006 01:21:25 -0700 Subject: [PATCH] Documentation/filesystems/ext4.txt This file, ext4.txt, was put together with information from Andrew Morton, Andreas Dilger, Suparna Bhattacharya, and Ted Ts'o. I copied the mount options, with the exception of "extents", from ext3.txt, so if anyone is aware of anything out-of-date, please let me know. Signed-off-by: Dave Kleikamp Cc: Theodore Ts'o Cc: Suparna Bhattacharya Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/filesystems/00-INDEX b/Documentation/filesystems/00-INDEX index 3c384c0..4dc28cc 100644 --- a/Documentation/filesystems/00-INDEX +++ b/Documentation/filesystems/00-INDEX @@ -34,6 +34,8 @@ ext2.txt - info, mount options and specifications for the Ext2 filesystem. ext3.txt - info, mount options and specifications for the Ext3 filesystem. +ext4.txt + - info, mount options and specifications for the Ext4 filesystem. files.txt - info on file management in the Linux kernel. fuse.txt diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt new file mode 100644 index 0000000..6a4adca --- /dev/null +++ b/Documentation/filesystems/ext4.txt @@ -0,0 +1,236 @@ + +Ext4 Filesystem +=============== + +This is a development version of the ext4 filesystem, an advanced level +of the ext3 filesystem which incorporates scalability and reliability +enhancements for supporting large filesystems (64 bit) in keeping with +increasing disk capacities and state-of-the-art feature requirements. + +Mailing list: linux-ext4@vger.kernel.org + + +1. Quick usage instructions: +=========================== + + - Grab updated e2fsprogs from + ftp://ftp.kernel.org/pub/linux/kernel/people/tytso/e2fsprogs-interim/ + This is a patchset on top of e2fsprogs-1.39, which can be found at + ftp://ftp.kernel.org/pub/linux/kernel/people/tytso/e2fsprogs/ + + - It's still mke2fs -j /dev/hda1 + + - mount /dev/hda1 /wherever -t ext4dev + + - To enable extents, + + mount /dev/hda1 /wherever -t ext4dev -o extents + + - The filesystem is compatible with the ext3 driver until you add a file + which has extents (ie: `mount -o extents', then create a file). + + NOTE: The "extents" mount flag is temporary. It will soon go away and + extents will be enabled by the "-o extents" flag to mke2fs or tune2fs + + - When comparing performance with other filesystems, remember that + ext3/4 by default offers higher data integrity guarantees than most. So + when comparing with a metadata-only journalling filesystem, use `mount -o + data=writeback'. And you might as well use `mount -o nobh' too along + with it. Making the journal larger than the mke2fs default often helps + performance with metadata-intensive workloads. + +2. Features +=========== + +2.1 Currently available + +* ability to use filesystems > 16TB +* extent format reduces metadata overhead (RAM, IO for access, transactions) +* extent format more robust in face of on-disk corruption due to magics, +* internal redunancy in tree + +2.1 Previously available, soon to be enabled by default by "mkefs.ext4": + +* dir_index and resize inode will be on by default +* large inodes will be used by default for fast EAs, nsec timestamps, etc + +2.2 Candidate features for future inclusion + +There are several under discussion, whether they all make it in is +partly a function of how much time everyone has to work on them: + +* improved file allocation (multi-block alloc, delayed alloc; basically done) +* fix 32000 subdirectory limit (patch exists, needs some e2fsck work) +* nsec timestamps for mtime, atime, ctime, create time (patch exists, + needs some e2fsck work) +* inode version field on disk (NFSv4, Lustre; prototype exists) +* reduced mke2fs/e2fsck time via uninitialized groups (prototype exists) +* journal checksumming for robustness, performance (prototype exists) +* persistent file preallocation (e.g for streaming media, databases) + +Features like metadata checksumming have been discussed and planned for +a bit but no patches exist yet so I'm not sure they're in the near-term +roadmap. + +The big performance win will come with mballoc and delalloc. CFS has +been using mballoc for a few years already with Lustre, and IBM + Bull +did a lot of benchmarking on it. The reason it isn't in the first set of +patches is partly a manageability issue, and partly because it doesn't +directly affect the on-disk format (outside of much better allocation) +so it isn't critical to get into the first round of changes. I believe +Alex is working on a new set of patches right now. + +3. Options +========== + +When mounting an ext4 filesystem, the following option are accepted: +(*) == default + +extents ext4 will use extents to address file data. The + file system will no longer be mountable by ext3. + +journal=update Update the ext4 file system's journal to the current + format. + +journal=inum When a journal already exists, this option is ignored. + Otherwise, it specifies the number of the inode which + will represent the ext4 file system's journal file. + +journal_dev=devnum When the external journal device's major/minor numbers + have changed, this option allows the user to specify + the new journal location. The journal device is + identified through its new major/minor numbers encoded + in devnum. + +noload Don't load the journal on mounting. + +data=journal All data are committed into the journal prior to being + written into the main file system. + +data=ordered (*) All data are forced directly out to the main file + system prior to its metadata being committed to the + journal. + +data=writeback Data ordering is not preserved, data may be written + into the main file system after its metadata has been + committed to the journal. + +commit=nrsec (*) Ext4 can be told to sync all its data and metadata + every 'nrsec' seconds. The default value is 5 seconds. + This means that if you lose your power, you will lose + as much as the latest 5 seconds of work (your + filesystem will not be damaged though, thanks to the + journaling). This default value (or any low value) + will hurt performance, but it's good for data-safety. + Setting it to 0 will have the same effect as leaving + it at the default (5 seconds). + Setting it to very large values will improve + performance. + +barrier=1 This enables/disables barriers. barrier=0 disables + it, barrier=1 enables it. + +orlov (*) This enables the new Orlov block allocator. It is + enabled by default. + +oldalloc This disables the Orlov block allocator and enables + the old block allocator. Orlov should have better + performance - we'd like to get some feedback if it's + the contrary for you. + +user_xattr Enables Extended User Attributes. Additionally, you + need to have extended attribute support enabled in the + kernel configuration (CONFIG_EXT4_FS_XATTR). See the + attr(5) manual page and http://acl.bestbits.at/ to + learn more about extended attributes. + +nouser_xattr Disables Extended User Attributes. + +acl Enables POSIX Access Control Lists support. + Additionally, you need to have ACL support enabled in + the kernel configuration (CONFIG_EXT4_FS_POSIX_ACL). + See the acl(5) manual page and http://acl.bestbits.at/ + for more information. + +noacl This option disables POSIX Access Control List + support. + +reservation + +noreservation + +bsddf (*) Make 'df' act like BSD. +minixdf Make 'df' act like Minix. + +check=none Don't do extra checking of bitmaps on mount. +nocheck + +debug Extra debugging information is sent to syslog. + +errors=remount-ro(*) Remount the filesystem read-only on an error. +errors=continue Keep going on a filesystem error. +errors=panic Panic and halt the machine if an error occurs. + +grpid Give objects the same group ID as their creator. +bsdgroups + +nogrpid (*) New objects have the group ID of their creator. +sysvgroups + +resgid=n The group ID which may use the reserved blocks. + +resuid=n The user ID which may use the reserved blocks. + +sb=n Use alternate superblock at this location. + +quota +noquota +grpquota +usrquota + +bh (*) ext4 associates buffer heads to data pages to +nobh (a) cache disk block mapping information + (b) link pages into transaction to provide + ordering guarantees. + "bh" option forces use of buffer heads. + "nobh" option tries to avoid associating buffer + heads (supported only for "writeback" mode). + + +Data Mode +--------- +There are 3 different data modes: + +* writeback mode +In data=writeback mode, ext4 does not journal data at all. This mode provides +a similar level of journaling as that of XFS, JFS, and ReiserFS in its default +mode - metadata journaling. A crash+recovery can cause incorrect data to +appear in files which were written shortly before the crash. This mode will +typically provide the best ext4 performance. + +* ordered mode +In data=ordered mode, ext4 only officially journals metadata, but it logically +groups metadata and data blocks into a single unit called a transaction. When +it's time to write the new metadata out to disk, the associated data blocks +are written first. In general, this mode performs slightly slower than +writeback but significantly faster than journal mode. + +* journal mode +data=journal mode provides full data and metadata journaling. All new data is +written to the journal first, and then to its final location. +In the event of a crash, the journal can be replayed, bringing both data and +metadata into a consistent state. This mode is the slowest except when data +needs to be read from and written to disk at the same time where it +outperforms all others modes. + +References +========== + +kernel source: + + +programs: http://e2fsprogs.sourceforge.net/ + http://ext2resize.sourceforge.net + +useful links: http://fedoraproject.org/wiki/ext3-devel + http://www.bullopensource.org/ext4/ -- cgit v0.10.2 From 469340236a7c9673df3e6a2425f559517f01464e Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 11 Oct 2006 01:21:26 -0700 Subject: [PATCH] mm: kevent threads: use MPOL_DEFAULT Switch the memory policy of the kevent threads to MPOL_DEFAULT while leaving the kzalloc of the workqueue structure on interleave. This means that all code executed in the context of the kevent thread is allocating node local. Signed-off-by: Christoph Lameter Cc: Christoph Lameter Cc: Alok Kataria Cc: Andi Kleen Cc: Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/workqueue.c b/kernel/workqueue.c index cfc737b..3df9bfc 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -28,6 +28,7 @@ #include #include #include +#include /* * The per-CPU workqueue (if single thread, we always use the first @@ -245,6 +246,12 @@ static int worker_thread(void *__cwq) sigprocmask(SIG_BLOCK, &blocked, NULL); flush_signals(current); + /* + * We inherited MPOL_INTERLEAVE from the booting kernel. + * Set MPOL_DEFAULT to insure node local allocations. + */ + numa_default_policy(); + /* SIG_IGN makes children autoreap: see do_notify_parent(). */ sa.sa.sa_handler = SIG_IGN; sa.sa.sa_flags = 0; -- cgit v0.10.2 From b16bc64d1aed40fb9cff9187061005b2a89b5d5d Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Wed, 11 Oct 2006 01:21:27 -0700 Subject: [PATCH] move rmap BUG_ON outside DEBUG_VM We have a persistent dribble of reports of this BUG triggering. Its extended diagnostics were recently made conditional on CONFIG_DEBUG_VM, which was a bad idea - we want to know about it. Signed-off-by: Dave Jones Cc: Nick Piggin Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/rmap.c b/mm/rmap.c index e2155d7..a9136d8 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -576,15 +576,14 @@ void page_add_file_rmap(struct page *page) void page_remove_rmap(struct page *page) { if (atomic_add_negative(-1, &page->_mapcount)) { -#ifdef CONFIG_DEBUG_VM if (unlikely(page_mapcount(page) < 0)) { printk (KERN_EMERG "Eeek! page_mapcount(page) went negative! (%d)\n", page_mapcount(page)); printk (KERN_EMERG " page->flags = %lx\n", page->flags); printk (KERN_EMERG " page->count = %x\n", page_count(page)); printk (KERN_EMERG " page->mapping = %p\n", page->mapping); + BUG(); } -#endif - BUG_ON(page_mapcount(page) < 0); + /* * It would be tidy to reset the PageAnon mapping here, * but that might overwrite a racing page_add_anon_rmap -- cgit v0.10.2 From 699397499742d1245ea5d677a08fa265df666d2d Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Wed, 11 Oct 2006 01:21:28 -0700 Subject: [PATCH] Fix do_mbind warning with CONFIG_MIGRATION=n With CONFIG_MIGRATION=n mm/mempolicy.c: In function 'do_mbind': mm/mempolicy.c:796: warning: passing argument 2 of 'migrate_pages' from incompatible pointer type Signed-off-by: Keith Owens Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 25788b1..617fb31 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -727,7 +727,7 @@ int do_migrate_pages(struct mm_struct *mm, return -ENOSYS; } -static struct page *new_vma_page(struct page *page, unsigned long private) +static struct page *new_vma_page(struct page *page, unsigned long private, int **x) { return NULL; } -- cgit v0.10.2 From dafb13673c463bc2aade4a4819704dde0f5fa37f Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 11 Oct 2006 01:21:30 -0700 Subject: [PATCH] mm: arch_free_page fix After the PG_reserved check was added, arch_free_page was being called in the wrong place (it could be called for a page we don't actually want to free). Fix that. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8246e83..c5caac2 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -495,7 +495,6 @@ static void __free_pages_ok(struct page *page, unsigned int order) int i; int reserved = 0; - arch_free_page(page, order); if (!PageHighMem(page)) debug_check_no_locks_freed(page_address(page), PAGE_SIZE<mapping = NULL; if (free_pages_check(page)) return; + arch_free_page(page, 0); kernel_map_pages(page, 1, 0); pcp = &zone_pcp(zone, get_cpu())->pcp[cold]; -- cgit v0.10.2 From 9858db504caedb2424b9a32744c23f9a81ec1731 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 11 Oct 2006 01:21:30 -0700 Subject: [PATCH] mm: locks_freed fix Move the lock debug checks below the page reserved checks. Also, having debug_check_no_locks_freed in kernel_map_pages is wrong. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/mm.h b/include/linux/mm.h index 2614662..5a6068f 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1103,12 +1103,7 @@ static inline void vm_stat_account(struct mm_struct *mm, #ifndef CONFIG_DEBUG_PAGEALLOC static inline void -kernel_map_pages(struct page *page, int numpages, int enable) -{ - if (!PageHighMem(page) && !enable) - debug_check_no_locks_freed(page_address(page), - numpages * PAGE_SIZE); -} +kernel_map_pages(struct page *page, int numpages, int enable) {} #endif extern struct vm_area_struct *get_gate_vma(struct task_struct *tsk); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c5caac2..40db96a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -495,15 +495,13 @@ static void __free_pages_ok(struct page *page, unsigned int order) int i; int reserved = 0; - if (!PageHighMem(page)) - debug_check_no_locks_freed(page_address(page), - PAGE_SIZE< Date: Wed, 11 Oct 2006 01:21:31 -0700 Subject: [PATCH] uml: revert wrong patch Andi Kleen pointed out that -mcmodel=kernel does not make sense for userspace code and would stop everything from working, and pointed out the correct fix for the original bug (not easy to do for me). Reverts part of commit 06837504de7b4883e92af207dbbab4310d0db0ed. Cc: Andi Kleen Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64 index 11154b6..87d6373 100644 --- a/arch/um/Makefile-x86_64 +++ b/arch/um/Makefile-x86_64 @@ -4,7 +4,7 @@ core-y += arch/um/sys-x86_64/ START := 0x60000000 -_extra_flags_ = -fno-builtin -m64 -mcmodel=kernel +_extra_flags_ = -fno-builtin -m64 #We #undef __x86_64__ for kernelspace, not for userspace where #it's needed for headers to work! -- cgit v0.10.2 From 21c935e5298a3bc4f16cdff7ce3b1efab92b995e Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:32 -0700 Subject: [PATCH] uml: correct removal of pte_mkexec Correct commit 5906e4171ad61ce68de95e51b773146707671f80 - this makes more sense: we turn pte_mkexec + pte_wrprotect to pte_mkread. However, due to a bug in pte_mkread, it does the exact same thing as pte_mkwrite, so this patch improves the code but does not change anything in practice. The pte_mkread bug is fixed separately, as it may have big impact. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c index c17eddc..2c6d090 100644 --- a/arch/um/kernel/skas/mmu.c +++ b/arch/um/kernel/skas/mmu.c @@ -60,10 +60,7 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc, #endif *pte = mk_pte(virt_to_page(kernel), __pgprot(_PAGE_PRESENT)); - /* This is wrong for the code page, but it doesn't matter since the - * stub is mapped by hand with the correct permissions. - */ - *pte = pte_mkwrite(*pte); + *pte = pte_mkread(*pte); return(0); out_pmd: -- cgit v0.10.2 From e27e80b3da7ad6b90185bd689879888907104a40 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:33 -0700 Subject: [PATCH] uml: readd forgot prototype This was forgot in a previous patch so UML does not compile with TT mode enabled. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/include/os.h b/arch/um/include/os.h index 120ca21..6516f6d 100644 --- a/arch/um/include/os.h +++ b/arch/um/include/os.h @@ -201,6 +201,7 @@ extern int os_getpgrp(void); #ifdef UML_CONFIG_MODE_TT extern void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int)); +extern void stop(void); #endif extern void init_new_thread_signals(void); extern int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr); -- cgit v0.10.2 From d875f9fd3f2369bf6f4d0e9989f00fe610eac470 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:34 -0700 Subject: [PATCH] uml: make TT mode compile after setjmp-related changes Make TT mode compile after the introduction of klibc's implementation of setjmp. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/tt/uaccess_user.c b/arch/um/kernel/tt/uaccess_user.c index 6c92bbc..ed1abcf 100644 --- a/arch/um/kernel/tt/uaccess_user.c +++ b/arch/um/kernel/tt/uaccess_user.c @@ -4,13 +4,13 @@ * Licensed under the GPL */ -#include #include #include "user_util.h" #include "uml_uaccess.h" #include "task.h" #include "kern_util.h" #include "os.h" +#include "longjmp.h" int __do_copy_from_user(void *to, const void *from, int n, void **fault_addr, void **fault_catcher) @@ -80,10 +80,10 @@ int __do_strnlen_user(const char *str, unsigned long n, struct tt_regs save = TASK_REGS(get_current())->tt; int ret; unsigned long *faddrp = (unsigned long *)fault_addr; - sigjmp_buf jbuf; + jmp_buf jbuf; *fault_catcher = &jbuf; - if(sigsetjmp(jbuf, 1) == 0) + if(UML_SETJMP(&jbuf) == 0) ret = strlen(str) + 1; else ret = *faddrp - (unsigned long) str; diff --git a/arch/um/os-Linux/tt.c b/arch/um/os-Linux/tt.c index 5461a06..3dc3a02 100644 --- a/arch/um/os-Linux/tt.c +++ b/arch/um/os-Linux/tt.c @@ -10,7 +10,6 @@ #include #include #include -#include #include #include #include -- cgit v0.10.2 From b5cdb5797d364a112879e49cc708083853ffc592 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:34 -0700 Subject: [PATCH] uml: make UML_SETJMP always safe If enable is moved by GCC in a register its value may not be preserved after coming back there with longjmp(). So, mark it as volatile to prevent this; this is suggested (it seems) in info gcc, when it talks about -Wuninitialized. I re-read this and it seems to say something different, but I still believe this may be needed. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/include/longjmp.h b/arch/um/include/longjmp.h index e93c6d3..e860bc584 100644 --- a/arch/um/include/longjmp.h +++ b/arch/um/include/longjmp.h @@ -12,7 +12,8 @@ extern void longjmp(jmp_buf, int); } while(0) #define UML_SETJMP(buf) ({ \ - int n, enable; \ + int n; \ + volatile int enable; \ enable = get_signals(); \ n = setjmp(*buf); \ if(n != 0) \ -- cgit v0.10.2 From 1b4ad242fcfcf1e261604cb91ee5fb8032900997 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:35 -0700 Subject: [PATCH] uml: fix processor selection to exclude unsupported processors and features Makes UML compile on any possible processor choice. The two problems were: *) x86 code, when 386 is selected, checks at runtime boot_cpuflags, which we do not have. *) 3Dnow support for memcpy() et al. does not compile currently and fixing this is not trivial, so simply disable it; with this change, if one selects MK7 UML compiles (while it did not). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/Kconfig.cpu b/arch/i386/Kconfig.cpu index 21c9a4e..fc4f2ab 100644 --- a/arch/i386/Kconfig.cpu +++ b/arch/i386/Kconfig.cpu @@ -7,6 +7,7 @@ choice config M386 bool "386" + depends on !UML ---help--- This is the processor type of your CPU. This information is used for optimizing purposes. In order to compile a kernel that can run on @@ -301,7 +302,7 @@ config X86_USE_PPRO_CHECKSUM config X86_USE_3DNOW bool - depends on MCYRIXIII || MK7 || MGEODE_LX + depends on (MCYRIXIII || MK7 || MGEODE_LX) && !UML default y config X86_OOSTORE -- cgit v0.10.2 From 69fada32d868d7f4be128ea4df8fbe4fd897fc34 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:36 -0700 Subject: [PATCH] uml: fix uname under setarch i386 On a 64bit Uml, if run under "setarch i386" (which a user did), uname() currently returns the obtained i686 as machine - fix that. Btw, I'm quite surprised that under setarch i386 a 64-bit binary can run. Cc: Andi Kleen Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/os-Linux/util.c b/arch/um/os-Linux/util.c index 3f5b151..56b8a50 100644 --- a/arch/um/os-Linux/util.c +++ b/arch/um/os-Linux/util.c @@ -80,11 +80,18 @@ void setup_machinename(char *machine_out) struct utsname host; uname(&host); -#if defined(UML_CONFIG_UML_X86) && !defined(UML_CONFIG_64BIT) +#ifdef UML_CONFIG_UML_X86 +# ifndef UML_CONFIG_64BIT if (!strcmp(host.machine, "x86_64")) { strcpy(machine_out, "i686"); return; } +# else + if (!strcmp(host.machine, "i686")) { + strcpy(machine_out, "x86_64"); + return; + } +# endif #endif strcpy(machine_out, host.machine); } -- cgit v0.10.2 From 711553efa5b8581365084cacf3585a36ca253bab Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:37 -0700 Subject: [PATCH] uml: declare in Kconfig our partial LOCKDEP support Declare UML partial support for LOCKDEP - however IRQFLAGS tracing requires some coding which nobody did yet, so we cannot run full lockdep on UML. Grep for CONFIG_TRACE_IRQFLAGS on i386 code to find their implementation. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig b/arch/um/Kconfig index d753075..450547a 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -25,6 +25,19 @@ config PCI config PCMCIA bool +# Yet to do! +config TRACE_IRQFLAGS_SUPPORT + bool + default n + +config LOCKDEP_SUPPORT + bool + default y + +config STACKTRACE_SUPPORT + bool + default y + config GENERIC_CALIBRATE_DELAY bool default y -- cgit v0.10.2 From 2465b8580a5649ee789f213d39c1b49d84fefe8f Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:38 -0700 Subject: [PATCH] uml: allow using again x86/x86_64 crypto code Enable compilation of x86_64 crypto code;, and add the needed constant to make the code compile again (that macro was added to i386 asm-offsets between 2.6.17 and 2.6.18, in 6c2bb98bc33ae33c7a33a133a4cd5a06395fece5). Cc: Herbert Xu Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64 index 87d6373..d278682 100644 --- a/arch/um/Makefile-x86_64 +++ b/arch/um/Makefile-x86_64 @@ -1,7 +1,7 @@ # Copyright 2003 - 2004 Pathscale, Inc # Released under the GPL -core-y += arch/um/sys-x86_64/ +core-y += arch/um/sys-x86_64/ arch/x86_64/crypto/ START := 0x60000000 _extra_flags_ = -fno-builtin -m64 diff --git a/arch/um/include/common-offsets.h b/arch/um/include/common-offsets.h index 356390d..39bb210 100644 --- a/arch/um/include/common-offsets.h +++ b/arch/um/include/common-offsets.h @@ -15,3 +15,4 @@ DEFINE_STR(UM_KERN_DEBUG, KERN_DEBUG); DEFINE(UM_ELF_CLASS, ELF_CLASS); DEFINE(UM_ELFCLASS32, ELFCLASS32); DEFINE(UM_ELFCLASS64, ELFCLASS64); +DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx)); diff --git a/arch/um/include/sysdep-i386/kernel-offsets.h b/arch/um/include/sysdep-i386/kernel-offsets.h index 2c13de3..2e58c4c 100644 --- a/arch/um/include/sysdep-i386/kernel-offsets.h +++ b/arch/um/include/sysdep-i386/kernel-offsets.h @@ -1,6 +1,7 @@ #include #include #include +#include #include #define DEFINE(sym, val) \ diff --git a/arch/um/include/sysdep-x86_64/kernel-offsets.h b/arch/um/include/sysdep-x86_64/kernel-offsets.h index 91d129f..4cbfbb9 100644 --- a/arch/um/include/sysdep-x86_64/kernel-offsets.h +++ b/arch/um/include/sysdep-x86_64/kernel-offsets.h @@ -2,6 +2,7 @@ #include #include #include +#include #include #include -- cgit v0.10.2 From d27ecef7c63064c1f2eadf413e694e65a34f1f79 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:40 -0700 Subject: [PATCH] uml: asm offsets duplication removal Unify macros common to x86 and x86_64 kernel-offsets.h files. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/include/common-offsets.h b/arch/um/include/common-offsets.h index 39bb210..461175f 100644 --- a/arch/um/include/common-offsets.h +++ b/arch/um/include/common-offsets.h @@ -1,9 +1,16 @@ /* for use by sys-$SUBARCH/kernel-offsets.c */ +DEFINE(KERNEL_MADV_REMOVE, MADV_REMOVE); +#ifdef CONFIG_MODE_TT +OFFSET(HOST_TASK_EXTERN_PID, task_struct, thread.mode.tt.extern_pid); +#endif + OFFSET(HOST_TASK_REGS, task_struct, thread.regs); OFFSET(HOST_TASK_PID, task_struct, pid); + DEFINE(UM_KERN_PAGE_SIZE, PAGE_SIZE); DEFINE(UM_NSEC_PER_SEC, NSEC_PER_SEC); + DEFINE_STR(UM_KERN_EMERG, KERN_EMERG); DEFINE_STR(UM_KERN_ALERT, KERN_ALERT); DEFINE_STR(UM_KERN_CRIT, KERN_CRIT); @@ -12,7 +19,10 @@ DEFINE_STR(UM_KERN_WARNING, KERN_WARNING); DEFINE_STR(UM_KERN_NOTICE, KERN_NOTICE); DEFINE_STR(UM_KERN_INFO, KERN_INFO); DEFINE_STR(UM_KERN_DEBUG, KERN_DEBUG); + DEFINE(UM_ELF_CLASS, ELF_CLASS); DEFINE(UM_ELFCLASS32, ELFCLASS32); DEFINE(UM_ELFCLASS64, ELFCLASS64); + +/* For crypto assembler code. */ DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx)); diff --git a/arch/um/include/sysdep-i386/kernel-offsets.h b/arch/um/include/sysdep-i386/kernel-offsets.h index 2e58c4c..97ec9d8 100644 --- a/arch/um/include/sysdep-i386/kernel-offsets.h +++ b/arch/um/include/sysdep-i386/kernel-offsets.h @@ -18,9 +18,5 @@ void foo(void) { OFFSET(HOST_TASK_DEBUGREGS, task_struct, thread.arch.debugregs); - DEFINE(KERNEL_MADV_REMOVE, MADV_REMOVE); -#ifdef CONFIG_MODE_TT - OFFSET(HOST_TASK_EXTERN_PID, task_struct, thread.mode.tt.extern_pid); -#endif #include } diff --git a/arch/um/include/sysdep-x86_64/kernel-offsets.h b/arch/um/include/sysdep-x86_64/kernel-offsets.h index 4cbfbb9..a307237 100644 --- a/arch/um/include/sysdep-x86_64/kernel-offsets.h +++ b/arch/um/include/sysdep-x86_64/kernel-offsets.h @@ -19,9 +19,5 @@ void foo(void) { - DEFINE(KERNEL_MADV_REMOVE, MADV_REMOVE); -#ifdef CONFIG_MODE_TT - OFFSET(HOST_TASK_EXTERN_PID, task_struct, thread.mode.tt.extern_pid); -#endif #include } -- cgit v0.10.2 From f2d57151bd3110ae3f0db3e5770ab1a46d1b647a Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:40 -0700 Subject: [PATCH] uml: remove duplicate export The export is together with the definition, in arch/x86_64/lib/csum-partial.c, which is compiled in by arch/um/sys-x86_64/Makefile. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/sys-x86_64/ksyms.c b/arch/um/sys-x86_64/ksyms.c index 8592738..12c5936 100644 --- a/arch/um/sys-x86_64/ksyms.c +++ b/arch/um/sys-x86_64/ksyms.c @@ -14,6 +14,3 @@ EXPORT_SYMBOL(__up_wakeup); /*XXX: we need them because they would be exported by x86_64 */ EXPORT_SYMBOL(__memcpy); - -/* Networking helper routines. */ -EXPORT_SYMBOL(ip_compute_csum); -- cgit v0.10.2 From ca316fcf641532ee70dfc59daf632d1685b7d902 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:41 -0700 Subject: [PATCH] uml: deprecate CONFIG_MODE_TT Deprecate TT mode in Kconfig so that users won't select it, update the MODE_SKAS description (it was largely obsolete and misleadin) and btw describe advantages for high memory usage with CONFIG_STATIC_LINK. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 450547a..78fb619 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -50,13 +50,15 @@ config IRQ_RELEASE_METHOD menu "UML-specific options" config MODE_TT - bool "Tracing thread support" + bool "Tracing thread support (DEPRECATED)" default n help This option controls whether tracing thread support is compiled - into UML. This option is largely obsolete, given that skas0 provides + into UML. This option is largely obsolete, given that skas0 provides skas security and performance without needing to patch the host. - It is safe to say 'N' here. + It is safe to say 'N' here; saying 'Y' may cause additional problems + with the resulting binary even if you run UML in SKAS mode, and running + in TT mode is strongly *NOT RECOMMENDED*. config STATIC_LINK bool "Force a static link" @@ -69,6 +71,9 @@ config STATIC_LINK for use in a chroot jail. So, if you intend to run UML inside a chroot, and you disable CONFIG_MODE_TT, you probably want to say Y here. + Additionally, this option enables using higher memory spaces (up to + 2.75G) for UML - disabling CONFIG_MODE_TT and enabling this option leads + to best results for this. config KERNEL_HALF_GIGS int "Kernel address space size (in .5G units)" @@ -85,10 +90,13 @@ config MODE_SKAS default y help This option controls whether skas (separate kernel address space) - support is compiled in. If you have applied the skas patch to the - host, then you certainly want to say Y here (and consider saying N - to CONFIG_MODE_TT). Otherwise, it is safe to say Y. Disabling this - option will shrink the UML binary slightly. + support is compiled in. + Unless you have specific needs to use TT mode (which applies almost only + to developers), you should say Y here. + SKAS mode will make use of the SKAS3 patch if it is applied on the host + (and your UML will run in SKAS3 mode), but if no SKAS patch is applied + on the host it will run in SKAS0 mode, which is anyway faster than TT + mode. source "arch/um/Kconfig.arch" source "mm/Kconfig" -- cgit v0.10.2 From c21be1c9c1ab42fe285a74f184e1acbc37ee084b Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Wed, 11 Oct 2006 01:21:42 -0700 Subject: [PATCH] uml: allow finer tuning for host VMSPLIT setting Now that various memory splits are enabled, add a config option allowing the user to compile UML for its need - HOST_2G_2G allowed to choose either 3G/1G or 2G/2G, and enabling it reduced the usable virtual memory. Detecting this at run time should be implemented in the future, but we must make the stop-gap measure work well enough (this is valid in _many_ cases). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386 index f6eb72d..f191a55 100644 --- a/arch/um/Kconfig.i386 +++ b/arch/um/Kconfig.i386 @@ -16,23 +16,42 @@ config SEMAPHORE_SLEEPERS bool default y -config HOST_2G_2G - bool "2G/2G host address space split" - default n - help - This is needed when the host on which you run has a 2G/2G memory - split, instead of the customary 3G/1G. - - Note that to enable such a host - configuration, which makes sense only in some cases, you need special - host patches. - - So, if you do not know what to do here, say 'N'. +choice + prompt "Host memory split" + default HOST_VMSPLIT_3G + ---help--- + This is needed when the host kernel on which you run has a non-default + (like 2G/2G) memory split, instead of the customary 3G/1G. If you did + not recompile your own kernel but use the default distro's one, you can + safely accept the "Default split" option. + + It can be enabled on recent (>=2.6.16-rc2) vanilla kernels via + CONFIG_VM_SPLIT_*, or on previous kernels with special patches (-ck + patchset by Con Kolivas, or other ones) - option names match closely the + host CONFIG_VM_SPLIT_* ones. + + A lower setting (where 1G/3G is lowest and 3G/1G is higher) will + tolerate even more "normal" host kernels, but an higher setting will be + stricter. + + So, if you do not know what to do here, say 'Default split'. + + config HOST_VMSPLIT_3G + bool "Default split (3G/1G user/kernel host split)" + config HOST_VMSPLIT_3G_OPT + bool "3G/1G user/kernel host split (for full 1G low memory)" + config HOST_VMSPLIT_2G + bool "2G/2G user/kernel host split" + config HOST_VMSPLIT_1G + bool "1G/3G user/kernel host split" +endchoice config TOP_ADDR - hex - default 0xc0000000 if !HOST_2G_2G - default 0x80000000 if HOST_2G_2G + hex + default 0xB0000000 if HOST_VMSPLIT_3G_OPT + default 0x78000000 if HOST_VMSPLIT_2G + default 0x40000000 if HOST_VMSPLIT_1G + default 0xC0000000 config 3_LEVEL_PGTABLES bool "Three-level pagetables (EXPERIMENTAL)" -- cgit v0.10.2 From 0f836e5fecf59d0d0353e9af11fd14a32a3001ae Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 11 Oct 2006 01:21:43 -0700 Subject: [PATCH] Add CONFIG_HEADERS_CHECK option to automatically run 'make headers_check' In order to encourage people to notice when they break the exported headers, add a config option which automatically runs the sanity checks when building vmlinux. That way, those who use allyesconfig will notice failures. Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Makefile b/Makefile index 274b780..f242829 100644 --- a/Makefile +++ b/Makefile @@ -741,6 +741,9 @@ endif # ifdef CONFIG_KALLSYMS # vmlinux image - including updated kernel symbols vmlinux: $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) $(kallsyms.o) FORCE +ifdef CONFIG_HEADERS_CHECK + $(Q)$(MAKE) headers_check +endif $(call if_changed_rule,vmlinux__) $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost $@ $(Q)rm -f .old_version diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 2d7cb0d..8fd2dbf 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -371,6 +371,19 @@ config FORCED_INLINING become the default in the future, until then this option is there to test gcc for this. +config HEADERS_CHECK + bool "Run 'make headers_check' when building vmlinux" + help + This option will extract the user-visible kernel headers whenever + building the kernel, and will run basic sanity checks on them to + ensure that exported files do not attempt to include files which + were not exported, etc. + + If you're making modifications to header files which are + relevant for userspace, say 'Y', and check the headers + exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in + your build tree), to make sure they're suitable. + config RCU_TORTURE_TEST tristate "torture tests for RCU" depends on DEBUG_KERNEL -- cgit v0.10.2 From b611967de4dc5c52049676c4369dcac622a7cdfe Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Wed, 11 Oct 2006 01:21:44 -0700 Subject: [PATCH] epoll_pwait() Implement the epoll_pwait system call, that extend the event wait mechanism with the same logic ppoll and pselect do. The definition of epoll_pwait is: int epoll_pwait(int epfd, struct epoll_event *events, int maxevents, int timeout, const sigset_t *sigmask, size_t sigsetsize); The difference between the vanilla epoll_wait and epoll_pwait is that the latter allows the caller to specify a signal mask to be set while waiting for events. Hence epoll_pwait will wait until either one monitored event, or an unmasked signal happen. If sigmask is NULL, the epoll_pwait system call will act exactly like epoll_wait. For the POSIX definition of pselect, information is available here: http://www.opengroup.org/onlinepubs/009695399/functions/select.html Signed-off-by: Davide Libenzi Cc: David Woodhouse Cc: Andi Kleen Cc: Michael Kerrisk Cc: Ulrich Drepper Cc: Roland McGrath Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index 7e639f7..2697e92 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -318,3 +318,4 @@ ENTRY(sys_call_table) .long sys_vmsplice .long sys_move_pages .long sys_getcpu + .long sys_epoll_pwait diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 557d5b6..ae228ec 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -105,6 +105,8 @@ /* Maximum msec timeout value storeable in a long int */ #define EP_MAX_MSTIMEO min(1000ULL * MAX_SCHEDULE_TIMEOUT / HZ, (LONG_MAX - 999ULL) / HZ) +#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event)) + struct epoll_filefd { struct file *file; @@ -497,7 +499,7 @@ void eventpoll_release_file(struct file *file) */ asmlinkage long sys_epoll_create(int size) { - int error, fd; + int error, fd = -1; struct eventpoll *ep; struct inode *inode; struct file *file; @@ -640,7 +642,6 @@ eexit_1: return error; } -#define MAX_EVENTS (INT_MAX / sizeof(struct epoll_event)) /* * Implement the event wait interface for the eventpoll file. It is the kernel @@ -657,7 +658,7 @@ asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events, current, epfd, events, maxevents, timeout)); /* The maximum number of event must be greater than zero */ - if (maxevents <= 0 || maxevents > MAX_EVENTS) + if (maxevents <= 0 || maxevents > EP_MAX_EVENTS) return -EINVAL; /* Verify that the area passed by the user is writeable */ @@ -699,6 +700,55 @@ eexit_1: } +#ifdef TIF_RESTORE_SIGMASK + +/* + * Implement the event wait interface for the eventpoll file. It is the kernel + * part of the user space epoll_pwait(2). + */ +asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events, + int maxevents, int timeout, const sigset_t __user *sigmask, + size_t sigsetsize) +{ + int error; + sigset_t ksigmask, sigsaved; + + /* + * If the caller wants a certain signal mask to be set during the wait, + * we apply it here. + */ + if (sigmask) { + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) + return -EFAULT; + sigdelsetmask(&ksigmask, sigmask(SIGKILL) | sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + error = sys_epoll_wait(epfd, events, maxevents, timeout); + + /* + * If we changed the signal mask, we need to restore the original one. + * In case we've got a signal while waiting, we do not restore the + * signal mask yet, and we allow do_signal() to deliver the signal on + * the way back to userspace, before the signal mask is restored. + */ + if (sigmask) { + if (error == -EINTR) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_thread_flag(TIF_RESTORE_SIGMASK); + } else + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + } + + return error; +} + +#endif /* #ifdef TIF_RESTORE_SIGMASK */ + + /* * Creates the file descriptor to be used by the epoll interface. */ diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index 3ca7ab9..beeeaf6 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -324,10 +324,11 @@ #define __NR_vmsplice 316 #define __NR_move_pages 317 #define __NR_getcpu 318 +#define __NR_epoll_pwait 319 #ifdef __KERNEL__ -#define NR_syscalls 319 +#define NR_syscalls 320 #include /* diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index b0ace3f..1912c6c 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -431,6 +431,10 @@ asmlinkage long sys_epoll_ctl(int epfd, int op, int fd, struct epoll_event __user *event); asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events, int maxevents, int timeout); +asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events, + int maxevents, int timeout, + const sigset_t __user *sigmask, + size_t sigsetsize); asmlinkage long sys_gethostname(char __user *name, int len); asmlinkage long sys_sethostname(char __user *name, int len); asmlinkage long sys_setdomainname(char __user *name, int len); -- cgit v0.10.2 From e0ab2928cc2202f13f0574d4c6f567f166d307eb Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Wed, 11 Oct 2006 01:21:45 -0700 Subject: [PATCH] Add carta_random32() library routine This is a follow-up patch based on the review for perfmon2. This patch adds the carta_random32() library routine + carta_random32.h header file. This is fast, simple, and efficient pseudo number generator algorithm. We use it in perfmon2 to randomize the sampling periods. In this context, we do not need any fancy randomizer. Signed-off-by: stephane eranian Cc: David Mosberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/carta_random32.h b/include/linux/carta_random32.h new file mode 100644 index 0000000..f6f3bd9 --- /dev/null +++ b/include/linux/carta_random32.h @@ -0,0 +1,29 @@ +/* + * Fast, simple, yet decent quality random number generator based on + * a paper by David G. Carta ("Two Fast Implementations of the + * `Minimal Standard' Random Number Generator," Communications of the + * ACM, January, 1990). + * + * Copyright (c) 2002-2006 Hewlett-Packard Development Company, L.P. + * Contributed by Stephane Eranian + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + * 02111-1307 USA + */ +#ifndef _LINUX_CARTA_RANDOM32_H_ +#define _LINUX_CARTA_RANDOM32_H_ + +u64 carta_random32(u64 seed); + +#endif /* _LINUX_CARTA_RANDOM32_H_ */ diff --git a/lib/Makefile b/lib/Makefile index 8e6662b..59070db 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -5,7 +5,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \ bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \ idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \ - sha1.o irq_regs.o + sha1.o irq_regs.o carta_random32.o lib-$(CONFIG_MMU) += ioremap.o lib-$(CONFIG_SMP) += cpumask.o diff --git a/lib/carta_random32.c b/lib/carta_random32.c new file mode 100644 index 0000000..ca82df7 --- /dev/null +++ b/lib/carta_random32.c @@ -0,0 +1,41 @@ +/* + * Copyright (c) 2002-2006 Hewlett-Packard Development Company, L.P. + * Contributed by David Mosberger-Tang + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + * 02111-1307 USA + */ +#include +#include + +/* + * Fast, simple, yet decent quality random number generator based on + * a paper by David G. Carta ("Two Fast Implementations of the + * `Minimal Standard' Random Number Generator," Communications of the + * ACM, January, 1990). + */ +u64 carta_random32 (u64 seed) +{ +# define A 16807 +# define M ((u32) 1 << 31) + u64 s, prod = A * seed, p, q; + + p = (prod >> 31) & (M - 1); + q = (prod >> 0) & (M - 1); + s = p + q; + if (s >= M) + s -= M - 1; + return s; +} +EXPORT_SYMBOL_GPL(carta_random32); -- cgit v0.10.2 From e5657933863f43cc6bb76a54d659303dafaa9e58 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:21:46 -0700 Subject: [PATCH] grow_buffers() infinite loop fix If grow_buffers() is for some reason passed a block number which wants to lie outside the maximum-addressable pagecache range (PAGE_SIZE * 4G bytes) then it will accidentally truncate `index' and will then instnatiate a page at the wrong pagecache offset. This causes __getblk_slow() to go into an infinite loop. This can happen with corrupted disks, or with software errors elsewhere. Detect that, and handle it. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/buffer.c b/fs/buffer.c index eeb8ac1..2a7828c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1042,8 +1042,21 @@ grow_buffers(struct block_device *bdev, sector_t block, int size) } while ((size << sizebits) < PAGE_SIZE); index = block >> sizebits; - block = index << sizebits; + /* + * Check for a block which wants to lie outside our maximum possible + * pagecache index. (this comparison is done using sector_t types). + */ + if (unlikely(index != block >> sizebits)) { + char b[BDEVNAME_SIZE]; + + printk(KERN_ERR "%s: requested out-of-range block %llu for " + "device %s\n", + __FUNCTION__, (unsigned long long)block, + bdevname(bdev, b)); + return -EIO; + } + block = index << sizebits; /* Create a page with the proper size buffers.. */ page = grow_dev_page(bdev, block, index, size); if (!page) @@ -1070,12 +1083,16 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size) for (;;) { struct buffer_head * bh; + int ret; bh = __find_get_block(bdev, block, size); if (bh) return bh; - if (!grow_buffers(bdev, block, size)) + ret = grow_buffers(bdev, block, size); + if (ret < 0) + return NULL; + if (ret == 0) free_more_memory(); } } -- cgit v0.10.2 From 3719bc5c22c9025bf1c909fe8b527ebf1de9a153 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 11 Oct 2006 01:21:47 -0700 Subject: [PATCH] ide-generic: jmicron fix Some people find their Jmicron pata port reports its disabled even though it has devices on it and was boot probed. Fix this (Candidate for 2.6.18.*, less so for 2.6.19 as we've got a proper jmicron driver on the merge for that to replace ide-generic support) Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c index 965c436..5b77a5b 100644 --- a/drivers/ide/pci/generic.c +++ b/drivers/ide/pci/generic.c @@ -237,10 +237,12 @@ static int __devinit generic_init_one(struct pci_dev *dev, const struct pci_devi if (dev->vendor == PCI_VENDOR_ID_JMICRON && PCI_FUNC(dev->devfn) != 1) goto out; - pci_read_config_word(dev, PCI_COMMAND, &command); - if (!(command & PCI_COMMAND_IO)) { - printk(KERN_INFO "Skipping disabled %s IDE controller.\n", d->name); - goto out; + if (dev->vendor != PCI_VENDOR_ID_JMICRON) { + pci_read_config_word(dev, PCI_COMMAND, &command); + if (!(command & PCI_COMMAND_IO)) { + printk(KERN_INFO "Skipping disabled %s IDE controller.\n", d->name); + goto out; + } } ret = ide_setup_pci_device(dev, d); out: -- cgit v0.10.2 From fa3ba2e81ea23416272a22009bba95954c81969c Mon Sep 17 00:00:00 2001 From: Florin Malita Date: Wed, 11 Oct 2006 01:21:48 -0700 Subject: [PATCH] fix Module taint flags listing in Oops/panic Module taint flags listing in Oops/panic has a couple of issues: * taint_flags() doesn't null-terminate the buffer after printing the flags * per-module taints are only set if the kernel is not already tainted (with that particular flag) => only the first offending module gets its taint info correctly updated Some additional changes: * 'license_gplok' is no longer needed - equivalent to !(taints & TAINT_PROPRIETARY_MODULE) - so we can drop it from struct module * exporting module taint info via /proc/module: pwc 88576 0 - Live 0xf8c32000 evilmod 6784 1 pwc, Live 0xf8bbf000 (PF) Signed-off-by: Florin Malita Cc: "Randy.Dunlap" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/module.h b/include/linux/module.h index 4b2d809..d1d00ce 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -317,9 +317,6 @@ struct module /* Am I unsafe to unload? */ int unsafe; - /* Am I GPL-compatible */ - int license_gplok; - unsigned int taints; /* same bits as kernel:tainted */ #ifdef CONFIG_MODULE_UNLOAD diff --git a/kernel/module.c b/kernel/module.c index 7f60e78..67009bd 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -87,6 +87,12 @@ static inline int strong_try_module_get(struct module *mod) return try_module_get(mod); } +static inline void add_taint_module(struct module *mod, unsigned flag) +{ + add_taint(flag); + mod->taints |= flag; +} + /* A thread that wants to hold a reference to a module only while it * is running can call ths to safely exit. * nfsd and lockd use this. @@ -847,12 +853,10 @@ static int check_version(Elf_Shdr *sechdrs, return 0; } /* Not in module's version table. OK, but that taints the kernel. */ - if (!(tainted & TAINT_FORCED_MODULE)) { + if (!(tainted & TAINT_FORCED_MODULE)) printk("%s: no version for \"%s\" found: kernel tainted.\n", mod->name, symname); - add_taint(TAINT_FORCED_MODULE); - mod->taints |= TAINT_FORCED_MODULE; - } + add_taint_module(mod, TAINT_FORCED_MODULE); return 1; } @@ -910,7 +914,8 @@ static unsigned long resolve_symbol(Elf_Shdr *sechdrs, unsigned long ret; const unsigned long *crc; - ret = __find_symbol(name, &owner, &crc, mod->license_gplok); + ret = __find_symbol(name, &owner, &crc, + !(mod->taints & TAINT_PROPRIETARY_MODULE)); if (ret) { /* use_module can fail due to OOM, or module unloading */ if (!check_version(sechdrs, versindex, name, mod, crc) || @@ -1335,12 +1340,11 @@ static void set_license(struct module *mod, const char *license) if (!license) license = "unspecified"; - mod->license_gplok = license_is_gpl_compatible(license); - if (!mod->license_gplok && !(tainted & TAINT_PROPRIETARY_MODULE)) { - printk(KERN_WARNING "%s: module license '%s' taints kernel.\n", - mod->name, license); - add_taint(TAINT_PROPRIETARY_MODULE); - mod->taints |= TAINT_PROPRIETARY_MODULE; + if (!license_is_gpl_compatible(license)) { + if (!(tainted & TAINT_PROPRIETARY_MODULE)) + printk(KERN_WARNING "%s: module license '%s' taints" + "kernel.\n", mod->name, license); + add_taint_module(mod, TAINT_PROPRIETARY_MODULE); } } @@ -1619,8 +1623,7 @@ static struct module *load_module(void __user *umod, modmagic = get_modinfo(sechdrs, infoindex, "vermagic"); /* This is allowed: modprobe --force will invalidate it. */ if (!modmagic) { - add_taint(TAINT_FORCED_MODULE); - mod->taints |= TAINT_FORCED_MODULE; + add_taint_module(mod, TAINT_FORCED_MODULE); printk(KERN_WARNING "%s: no version magic, tainting kernel.\n", mod->name); } else if (!same_magic(modmagic, vermagic)) { @@ -1714,14 +1717,10 @@ static struct module *load_module(void __user *umod, /* Set up license info based on the info section */ set_license(mod, get_modinfo(sechdrs, infoindex, "license")); - if (strcmp(mod->name, "ndiswrapper") == 0) { - add_taint(TAINT_PROPRIETARY_MODULE); - mod->taints |= TAINT_PROPRIETARY_MODULE; - } - if (strcmp(mod->name, "driverloader") == 0) { - add_taint(TAINT_PROPRIETARY_MODULE); - mod->taints |= TAINT_PROPRIETARY_MODULE; - } + if (strcmp(mod->name, "ndiswrapper") == 0) + add_taint_module(mod, TAINT_PROPRIETARY_MODULE); + if (strcmp(mod->name, "driverloader") == 0) + add_taint_module(mod, TAINT_PROPRIETARY_MODULE); /* Set up MODINFO_ATTR fields */ setup_modinfo(mod, sechdrs, infoindex); @@ -1766,8 +1765,7 @@ static struct module *load_module(void __user *umod, (mod->num_unused_gpl_syms && !unusedgplcrcindex)) { printk(KERN_WARNING "%s: No versions for exported symbols." " Tainting kernel.\n", mod->name); - add_taint(TAINT_FORCED_MODULE); - mod->taints |= TAINT_FORCED_MODULE; + add_taint_module(mod, TAINT_FORCED_MODULE); } #endif @@ -2132,9 +2130,33 @@ static void m_stop(struct seq_file *m, void *p) mutex_unlock(&module_mutex); } +static char *taint_flags(unsigned int taints, char *buf) +{ + int bx = 0; + + if (taints) { + buf[bx++] = '('; + if (taints & TAINT_PROPRIETARY_MODULE) + buf[bx++] = 'P'; + if (taints & TAINT_FORCED_MODULE) + buf[bx++] = 'F'; + /* + * TAINT_FORCED_RMMOD: could be added. + * TAINT_UNSAFE_SMP, TAINT_MACHINE_CHECK, TAINT_BAD_PAGE don't + * apply to modules. + */ + buf[bx++] = ')'; + } + buf[bx] = '\0'; + + return buf; +} + static int m_show(struct seq_file *m, void *p) { struct module *mod = list_entry(p, struct module, list); + char buf[8]; + seq_printf(m, "%s %lu", mod->name, mod->init_size + mod->core_size); print_unload_info(m, mod); @@ -2147,6 +2169,10 @@ static int m_show(struct seq_file *m, void *p) /* Used by oprofile and other similar tools. */ seq_printf(m, " 0x%p", mod->module_core); + /* Taints info */ + if (mod->taints) + seq_printf(m, " %s", taint_flags(mod->taints, buf)); + seq_printf(m, "\n"); return 0; } @@ -2235,28 +2261,6 @@ struct module *module_text_address(unsigned long addr) return mod; } -static char *taint_flags(unsigned int taints, char *buf) -{ - *buf = '\0'; - if (taints) { - int bx; - - buf[0] = '('; - bx = 1; - if (taints & TAINT_PROPRIETARY_MODULE) - buf[bx++] = 'P'; - if (taints & TAINT_FORCED_MODULE) - buf[bx++] = 'F'; - /* - * TAINT_FORCED_RMMOD: could be added. - * TAINT_UNSAFE_SMP, TAINT_MACHINE_CHECK, TAINT_BAD_PAGE don't - * apply to modules. - */ - buf[bx] = ')'; - } - return buf; -} - /* Don't grab lock, we're oopsing. */ void print_modules(void) { -- cgit v0.10.2 From 2245d7c21f5bb1f104ee1226ebcb3dd5b9acfff6 Mon Sep 17 00:00:00 2001 From: Dmitry Mishin Date: Wed, 11 Oct 2006 01:21:49 -0700 Subject: [PATCH] ext3: errors behaviour fix Current error behaviour for ext2 and ext3 filesystems does not fully correspond to the documentation and should be fixed. According to man 8 mount, ext2 and ext3 file systems allow to set one of 3 different on-errors behaviours: ---- start of quote man 8 mount ---- errors=continue / errors=remount-ro / errors=panic Define the behaviour when an error is encountered. (Either ignore errors and just mark the file system erroneous and continue, or remount the file system read-only, or panic and halt the system.) The default is set in the filesystem superblock, and can be changed using tune2fs(8). ---- end of quote ---- However EXT3_ERRORS_CONTINUE is not read from the superblock, and thus ERRORS_CONT is not saved on the sbi->s_mount_opt. It leads to the incorrect handle of errors on ext3. Then we've checked corresponding code in ext2 and discovered that it is buggy as well: - EXT2_ERRORS_CONTINUE is not read from the superblock (the same); - parse_option() does not clean the alternative values and thus something like (ERRORS_CONT|ERRORS_RO) can be set; - if options are omitted, parse_option() does not set any of these options. Therefore it is possible to set any combination of these options on the ext2: - none of them may be set: EXT2_ERRORS_CONTINUE on superblock / empty mount options; - any of them may be set using mount options; - 2 any options may be set: by using EXT2_ERRORS_RO/EXT2_ERRORS_PANIC on the superblock and other value in mount options; - and finally all three options may be set by adding third option in remount. Currently ext2 uses these values only in ext2_error() and it is not leading to any noticeable troubles. However somebody may be discouraged when he will try to workaround EXT2_ERRORS_PANIC on the superblock by using errors=continue in mount options. This patch: EXT3_ERRORS_CONTINUE should be taken from the superblock as default value for error behaviour. Signed-off-by: Dmitry Mishin Acked-by: Vasily Averin Acked-by: Kirill Korotaev Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 8bfd56e..afc2d4f 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1470,6 +1470,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, ERRORS_PANIC); else if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_RO) set_opt(sbi->s_mount_opt, ERRORS_RO); + else + set_opt(sbi->s_mount_opt, ERRORS_CONT); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); -- cgit v0.10.2 From 5a2b4062f5adf2218b9b021e1c33f374bf142cb2 Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Wed, 11 Oct 2006 01:21:50 -0700 Subject: [PATCH] ext2: errors behaviour fix Current error behaviour for ext2 and ext3 filesystems does not fully correspond to the documentation and should be fixed. According to man 8 mount, ext2 and ext3 file systems allow to set one of 3 different on-errors behaviours: ---- start of quote man 8 mount ---- errors=continue / errors=remount-ro / errors=panic Define the behaviour when an error is encountered. (Either ignore errors and just mark the file system erroneous and continue, or remount the file system read-only, or panic and halt the system.) The default is set in the filesystem superblock, and can be changed using tune2fs(8). ---- end of quote ---- However EXT3_ERRORS_CONTINUE is not read from the superblock, and thus ERRORS_CONT is not saved on the sbi->s_mount_opt. It leads to the incorrect handle of errors on ext3. Then we've checked corresponding code in ext2 and discovered that it is buggy as well: - EXT2_ERRORS_CONTINUE is not read from the superblock (the same); - parse_option() does not clean the alternative values and thus something like (ERRORS_CONT|ERRORS_RO) can be set; - if options are omitted, parse_option() does not set any of these options. Therefore it is possible to set any combination of these options on the ext2: - none of them may be set: EXT2_ERRORS_CONTINUE on superblock / empty mount options; - any of them may be set using mount options; - 2 any options may be set: by using EXT2_ERRORS_RO/EXT2_ERRORS_PANIC on the superblock and other value in mount options; - and finally all three options may be set by adding third option in remount. Currently ext2 uses these values only in ext2_error() and it is not leading to any noticeable troubles. However somebody may be discouraged when he will try to workaround EXT2_ERRORS_PANIC on the superblock by using errors=continue in mount options. This patch: EXT2_ERRORS_CONTINUE should be read from the superblock as default value for error behaviour. parse_option() should clean the alternative options and should not change default value taken from the superblock. Signed-off-by: Vasily Averin Acked-by: Kirill Korotaev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 513cd42..d8b9abd 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -364,7 +364,6 @@ static int parse_options (char * options, { char * p; substring_t args[MAX_OPT_ARGS]; - unsigned long kind = EXT2_MOUNT_ERRORS_CONT; int option; if (!options) @@ -404,13 +403,19 @@ static int parse_options (char * options, /* *sb_block = match_int(&args[0]); */ break; case Opt_err_panic: - kind = EXT2_MOUNT_ERRORS_PANIC; + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_RO); + set_opt (sbi->s_mount_opt, ERRORS_PANIC); break; case Opt_err_ro: - kind = EXT2_MOUNT_ERRORS_RO; + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_RO); break; case Opt_err_cont: - kind = EXT2_MOUNT_ERRORS_CONT; + clear_opt (sbi->s_mount_opt, ERRORS_RO); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_CONT); break; case Opt_nouid32: set_opt (sbi->s_mount_opt, NO_UID32); @@ -489,7 +494,6 @@ static int parse_options (char * options, return 0; } } - sbi->s_mount_opt |= kind; return 1; } @@ -715,6 +719,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, ERRORS_PANIC); else if (le16_to_cpu(sbi->s_es->s_errors) == EXT2_ERRORS_RO) set_opt(sbi->s_mount_opt, ERRORS_RO); + else + set_opt(sbi->s_mount_opt, ERRORS_CONT); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); -- cgit v0.10.2 From f33d9bd50478c9a969b65f58feb6b69a3ad478cb Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:21:51 -0700 Subject: [PATCH] tpm: fix error handling - handle sysfs error - handle driver model errors - de-obfuscate platform_device_register_simple() call, which included an assignment in between two function calls, in the same C statement. Signed-off-by: Jeff Garzik Acked-by: Kylene Hall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index a082a2e..6ad2d3b 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -1153,7 +1153,14 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vend spin_unlock(&driver_lock); - sysfs_create_group(&dev->kobj, chip->vendor.attr_group); + if (sysfs_create_group(&dev->kobj, chip->vendor.attr_group)) { + list_del(&chip->list); + put_device(dev); + clear_bit(chip->dev_num, dev_mask); + kfree(chip); + kfree(devname); + return NULL; + } chip->bios_dir = tpm_bios_log_setup(devname); diff --git a/drivers/char/tpm/tpm_atmel.c b/drivers/char/tpm/tpm_atmel.c index ad8ffe4..1ab0896 100644 --- a/drivers/char/tpm/tpm_atmel.c +++ b/drivers/char/tpm/tpm_atmel.c @@ -184,7 +184,9 @@ static int __init init_atmel(void) unsigned long base; struct tpm_chip *chip; - driver_register(&atml_drv); + rc = driver_register(&atml_drv); + if (rc) + return rc; if ((iobase = atmel_get_base_addr(&base, ®ion_size)) == NULL) { rc = -ENODEV; @@ -195,10 +197,8 @@ static int __init init_atmel(void) (atmel_request_region (tpm_atmel.base, region_size, "tpm_atmel0") == NULL) ? 0 : 1; - - if (IS_ERR - (pdev = - platform_device_register_simple("tpm_atmel", -1, NULL, 0))) { + pdev = platform_device_register_simple("tpm_atmel", -1, NULL, 0); + if (IS_ERR(pdev)) { rc = PTR_ERR(pdev); goto err_rel_reg; } diff --git a/drivers/char/tpm/tpm_nsc.c b/drivers/char/tpm/tpm_nsc.c index 26287aa..608f730 100644 --- a/drivers/char/tpm/tpm_nsc.c +++ b/drivers/char/tpm/tpm_nsc.c @@ -284,7 +284,7 @@ static struct device_driver nsc_drv = { static int __init init_nsc(void) { int rc = 0; - int lo, hi; + int lo, hi, err; int nscAddrBase = TPM_ADDR; struct tpm_chip *chip; unsigned long base; @@ -297,7 +297,9 @@ static int __init init_nsc(void) return -ENODEV; } - driver_register(&nsc_drv); + err = driver_register(&nsc_drv); + if (err) + return err; hi = tpm_read_index(nscAddrBase, TPM_NSC_BASE0_HI); lo = tpm_read_index(nscAddrBase, TPM_NSC_BASE0_LO); -- cgit v0.10.2 From beed33a816204cb402c69266475b6a60a2433ceb Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 11 Oct 2006 01:21:52 -0700 Subject: [PATCH] sched: likely profiling This likely profiling is pretty fun. I found a few possible problems in sched.c. This patch may be not measurable, but when I did measure long ago, nooping (un)likely cost a couple of % on scheduler heavy benchmarks, so it all adds up. Tweak some branch hints: - the 2nd 64 bits in the bitmask is likely to be populated, because it contains the first 28 bits (nearly 3/4) of the normal priorities. (ratio of 669669:691 ~= 1000:1). - it isn't unlikely that context switching switches to another process. it might be very rapidly switching to and from the idle process (ratio of 475815:419004 and 471330:423544). Let the branch predictor decide. - preempt_enable seems to be very often called in a nested preempt_disable or with interrupts disabled (ratio of 3567760:87965 ~= 40:1) Signed-off-by: Nick Piggin Acked-by: Ingo Molnar Cc: Daniel Walker Cc: Hua Zhong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-generic/bitops/sched.h b/include/asm-generic/bitops/sched.h index 5ef93a4..815bb01 100644 --- a/include/asm-generic/bitops/sched.h +++ b/include/asm-generic/bitops/sched.h @@ -15,7 +15,7 @@ static inline int sched_find_first_bit(const unsigned long *b) #if BITS_PER_LONG == 64 if (unlikely(b[0])) return __ffs(b[0]); - if (unlikely(b[1])) + if (likely(b[1])) return __ffs(b[1]) + 64; return __ffs(b[2]) + 128; #elif BITS_PER_LONG == 32 diff --git a/kernel/sched.c b/kernel/sched.c index 53608a5..094b568 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1822,14 +1822,14 @@ context_switch(struct rq *rq, struct task_struct *prev, struct mm_struct *mm = next->mm; struct mm_struct *oldmm = prev->active_mm; - if (unlikely(!mm)) { + if (!mm) { next->active_mm = oldmm; atomic_inc(&oldmm->mm_count); enter_lazy_tlb(oldmm, next); } else switch_mm(oldmm, mm, next); - if (unlikely(!prev->mm)) { + if (!prev->mm) { prev->active_mm = NULL; WARN_ON(rq->prev_mm); rq->prev_mm = oldmm; @@ -3491,7 +3491,7 @@ asmlinkage void __sched preempt_schedule(void) * If there is a non-zero preempt_count or interrupts are disabled, * we do not want to preempt the current task. Just return.. */ - if (unlikely(ti->preempt_count || irqs_disabled())) + if (likely(ti->preempt_count || irqs_disabled())) return; need_resched: -- cgit v0.10.2 From 8258d4a574d3a8c01f0ef68aa26b969398a0e140 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:21:53 -0700 Subject: [PATCH] invalidate_inode_pages2_range() debug A failure in invalidate_inode_pages2_range() can result in unpleasant things happening in NFS (at least). Stick a WARN_ON_ONCE() in there so we can find out if it happens, and maybe why. (akpm: might be a -mm-only patch, we'll see..) Cc: Chuck Lever Cc: Trond Myklebust Cc: Steve Dickson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/truncate.c b/mm/truncate.c index f4edbc1..fca2883 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -396,6 +396,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping, pagevec_release(&pvec); cond_resched(); } + WARN_ON_ONCE(ret); return ret; } EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range); -- cgit v0.10.2 From 2e3ad8af43c2f555c1be6c02581f6cc939dcb71c Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:21:53 -0700 Subject: [PATCH] x86/microcode: handle sysfs error Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index 9b94797..c4d0291 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -656,14 +656,18 @@ static struct attribute_group mc_attr_group = { static int mc_sysdev_add(struct sys_device *sys_dev) { - int cpu = sys_dev->id; + int err, cpu = sys_dev->id; struct ucode_cpu_info *uci = ucode_cpu_info + cpu; if (!cpu_online(cpu)) return 0; + pr_debug("Microcode:CPU %d added\n", cpu); memset(uci, 0, sizeof(*uci)); - sysfs_create_group(&sys_dev->kobj, &mc_attr_group); + + err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group); + if (err) + return err; microcode_init_cpu(cpu); return 0; -- cgit v0.10.2 From 39484e53bb00f55b6303a908070db133608ef2a5 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 11 Oct 2006 01:21:54 -0700 Subject: [PATCH] 32-bit compatibility HDIO IOCTLs A couple of HDIO IOCTLs are not yet handled and a few others are marked as using a pointer rather than an unsigned long. The formers include: HDIO_GET_WCACHE, HDIO_GET_ACOUSTIC, HDIO_GET_ADDRESS and HDIO_GET_BUSSTATE. The latters are: HDIO_SET_MULTCOUNT, HDIO_SET_UNMASKINTR, HDIO_SET_KEEPSETTINGS, HDIO_SET_32BIT, HDIO_SET_NOWERR, HDIO_SET_DMA, HDIO_SET_PIO_MODE and HDIO_SET_NICE. Additionally 0x330 used to be HDIO_GETGEO_BIG and may be issued by 32-bit `hdparm' run on a 64-bit kernel making Linux complain loudly. This is a fix for these issues. Signed-off-by: Maciej W. Rozycki Cc: Alan Cox Acked-by: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 27ca1aa..a91f262 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -2438,13 +2438,17 @@ HANDLE_IOCTL(0x1260, broken_blkgetsize) HANDLE_IOCTL(BLKFRAGET, w_long) HANDLE_IOCTL(BLKSECTGET, w_long) HANDLE_IOCTL(BLKPG, blkpg_ioctl_trans) -HANDLE_IOCTL(HDIO_GET_KEEPSETTINGS, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_UNMASKINTR, hdio_ioctl_trans) -HANDLE_IOCTL(HDIO_GET_DMA, hdio_ioctl_trans) -HANDLE_IOCTL(HDIO_GET_32BIT, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_MULTCOUNT, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_KEEPSETTINGS, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_32BIT, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_NOWERR, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_DMA, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_NICE, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_WCACHE, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_ACOUSTIC, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_ADDRESS, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_BUSSTATE, hdio_ioctl_trans) HANDLE_IOCTL(FDSETPRM32, fd_ioctl_trans) HANDLE_IOCTL(FDDEFPRM32, fd_ioctl_trans) HANDLE_IOCTL(FDGETPRM32, fd_ioctl_trans) diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h index 4e1663d..cfdb4f6a 100644 --- a/include/linux/compat_ioctl.h +++ b/include/linux/compat_ioctl.h @@ -61,17 +61,23 @@ COMPATIBLE_IOCTL(FIGETBSZ) * Some need translations, these do not. */ COMPATIBLE_IOCTL(HDIO_GET_IDENTITY) -COMPATIBLE_IOCTL(HDIO_SET_DMA) -COMPATIBLE_IOCTL(HDIO_SET_UNMASKINTR) -COMPATIBLE_IOCTL(HDIO_SET_NOWERR) -COMPATIBLE_IOCTL(HDIO_SET_32BIT) -COMPATIBLE_IOCTL(HDIO_SET_MULTCOUNT) -COMPATIBLE_IOCTL(HDIO_DRIVE_CMD) COMPATIBLE_IOCTL(HDIO_DRIVE_TASK) -COMPATIBLE_IOCTL(HDIO_SET_PIO_MODE) -COMPATIBLE_IOCTL(HDIO_SET_NICE) -COMPATIBLE_IOCTL(HDIO_SET_KEEPSETTINGS) +COMPATIBLE_IOCTL(HDIO_DRIVE_CMD) +ULONG_IOCTL(HDIO_SET_MULTCOUNT) +ULONG_IOCTL(HDIO_SET_UNMASKINTR) +ULONG_IOCTL(HDIO_SET_KEEPSETTINGS) +ULONG_IOCTL(HDIO_SET_32BIT) +ULONG_IOCTL(HDIO_SET_NOWERR) +ULONG_IOCTL(HDIO_SET_DMA) +ULONG_IOCTL(HDIO_SET_PIO_MODE) +ULONG_IOCTL(HDIO_SET_NICE) +ULONG_IOCTL(HDIO_SET_WCACHE) +ULONG_IOCTL(HDIO_SET_ACOUSTIC) +ULONG_IOCTL(HDIO_SET_BUSSTATE) +ULONG_IOCTL(HDIO_SET_ADDRESS) COMPATIBLE_IOCTL(HDIO_SCAN_HWIF) +/* 0x330 is reserved -- it used to be HDIO_GETGEO_BIG */ +COMPATIBLE_IOCTL(0x330) /* 0x02 -- Floppy ioctls */ COMPATIBLE_IOCTL(FDMSGON) COMPATIBLE_IOCTL(FDMSGOFF) -- cgit v0.10.2 From 01a3ee2b203e511e20f98b85a9172fd32c53e87c Mon Sep 17 00:00:00 2001 From: Reinette Chatre Date: Wed, 11 Oct 2006 01:21:55 -0700 Subject: [PATCH] bitmap: parse input from kernel and user buffers lib/bitmap.c:bitmap_parse() is a library function that received as input a user buffer. This seemed to have originated from the way the write_proc function of the /proc filesystem operates. This has been reworked to not use kmalloc and eliminates a lot of get_user() overhead by performing one access_ok before using __get_user(). We need to test if we are in kernel or user space (is_user) and access the buffer differently. We cannot use __get_user() to access kernel addresses in all cases, for example in architectures with separate address space for kernel and user. This function will be useful for other uses as well; for example, taking input for /sysfs instead of /proc, so it was changed to accept kernel buffers. We have this use for the Linux UWB project, as part as the upcoming bandwidth allocator code. Only a few routines used this function and they were changed too. Signed-off-by: Reinette Chatre Signed-off-by: Inaky Perez-Gonzalez Cc: Paul Jackson Cc: Joe Korty Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h index dcc5de7..64b4641 100644 --- a/include/linux/bitmap.h +++ b/include/linux/bitmap.h @@ -46,7 +46,8 @@ * bitmap_remap(dst, src, old, new, nbits) *dst = map(old, new)(src) * bitmap_bitremap(oldbit, old, new, nbits) newbit = map(old, new)(oldbit) * bitmap_scnprintf(buf, len, src, nbits) Print bitmap src to buf - * bitmap_parse(ubuf, ulen, dst, nbits) Parse bitmap dst from user buf + * bitmap_parse(buf, buflen, dst, nbits) Parse bitmap dst from kernel buf + * bitmap_parse_user(ubuf, ulen, dst, nbits) Parse bitmap dst from user buf * bitmap_scnlistprintf(buf, len, src, nbits) Print bitmap src as list to buf * bitmap_parselist(buf, dst, nbits) Parse bitmap dst from list * bitmap_find_free_region(bitmap, bits, order) Find and allocate bit region @@ -106,7 +107,9 @@ extern int __bitmap_weight(const unsigned long *bitmap, int bits); extern int bitmap_scnprintf(char *buf, unsigned int len, const unsigned long *src, int nbits); -extern int bitmap_parse(const char __user *ubuf, unsigned int ulen, +extern int __bitmap_parse(const char *buf, unsigned int buflen, int is_user, + unsigned long *dst, int nbits); +extern int bitmap_parse_user(const char __user *ubuf, unsigned int ulen, unsigned long *dst, int nbits); extern int bitmap_scnlistprintf(char *buf, unsigned int len, const unsigned long *src, int nbits); @@ -270,6 +273,12 @@ static inline void bitmap_shift_left(unsigned long *dst, __bitmap_shift_left(dst, src, n, nbits); } +static inline int bitmap_parse(const char *buf, unsigned int buflen, + unsigned long *maskp, int nmaskbits) +{ + return __bitmap_parse(buf, buflen, 0, maskp, nmaskbits); +} + #endif /* __ASSEMBLY__ */ #endif /* __LINUX_BITMAP_H */ diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index b268a3c..d0e8c8b 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -8,8 +8,8 @@ * See detailed comments in the file linux/bitmap.h describing the * data type on which these cpumasks are based. * - * For details of cpumask_scnprintf() and cpumask_parse(), - * see bitmap_scnprintf() and bitmap_parse() in lib/bitmap.c. + * For details of cpumask_scnprintf() and cpumask_parse_user(), + * see bitmap_scnprintf() and bitmap_parse_user() in lib/bitmap.c. * For details of cpulist_scnprintf() and cpulist_parse(), see * bitmap_scnlistprintf() and bitmap_parselist(), also in bitmap.c. * For details of cpu_remap(), see bitmap_bitremap in lib/bitmap.c @@ -49,7 +49,7 @@ * unsigned long *cpus_addr(mask) Array of unsigned long's in mask * * int cpumask_scnprintf(buf, len, mask) Format cpumask for printing - * int cpumask_parse(ubuf, ulen, mask) Parse ascii string as cpumask + * int cpumask_parse_user(ubuf, ulen, mask) Parse ascii string as cpumask * int cpulist_scnprintf(buf, len, mask) Format cpumask as list for printing * int cpulist_parse(buf, map) Parse ascii string as cpulist * int cpu_remap(oldbit, old, new) newbit = map(old, new)(oldbit) @@ -273,12 +273,12 @@ static inline int __cpumask_scnprintf(char *buf, int len, return bitmap_scnprintf(buf, len, srcp->bits, nbits); } -#define cpumask_parse(ubuf, ulen, dst) \ - __cpumask_parse((ubuf), (ulen), &(dst), NR_CPUS) -static inline int __cpumask_parse(const char __user *buf, int len, +#define cpumask_parse_user(ubuf, ulen, dst) \ + __cpumask_parse_user((ubuf), (ulen), &(dst), NR_CPUS) +static inline int __cpumask_parse_user(const char __user *buf, int len, cpumask_t *dstp, int nbits) { - return bitmap_parse(buf, len, dstp->bits, nbits); + return bitmap_parse_user(buf, len, dstp->bits, nbits); } #define cpulist_scnprintf(buf, len, src) \ diff --git a/include/linux/nodemask.h b/include/linux/nodemask.h index 5dce5c2..b1063e9 100644 --- a/include/linux/nodemask.h +++ b/include/linux/nodemask.h @@ -8,8 +8,8 @@ * See detailed comments in the file linux/bitmap.h describing the * data type on which these nodemasks are based. * - * For details of nodemask_scnprintf() and nodemask_parse(), - * see bitmap_scnprintf() and bitmap_parse() in lib/bitmap.c. + * For details of nodemask_scnprintf() and nodemask_parse_user(), + * see bitmap_scnprintf() and bitmap_parse_user() in lib/bitmap.c. * For details of nodelist_scnprintf() and nodelist_parse(), see * bitmap_scnlistprintf() and bitmap_parselist(), also in bitmap.c. * For details of node_remap(), see bitmap_bitremap in lib/bitmap.c. @@ -51,7 +51,7 @@ * unsigned long *nodes_addr(mask) Array of unsigned long's in mask * * int nodemask_scnprintf(buf, len, mask) Format nodemask for printing - * int nodemask_parse(ubuf, ulen, mask) Parse ascii string as nodemask + * int nodemask_parse_user(ubuf, ulen, mask) Parse ascii string as nodemask * int nodelist_scnprintf(buf, len, mask) Format nodemask as list for printing * int nodelist_parse(buf, map) Parse ascii string as nodelist * int node_remap(oldbit, old, new) newbit = map(old, new)(oldbit) @@ -288,12 +288,12 @@ static inline int __nodemask_scnprintf(char *buf, int len, return bitmap_scnprintf(buf, len, srcp->bits, nbits); } -#define nodemask_parse(ubuf, ulen, dst) \ - __nodemask_parse((ubuf), (ulen), &(dst), MAX_NUMNODES) -static inline int __nodemask_parse(const char __user *buf, int len, +#define nodemask_parse_user(ubuf, ulen, dst) \ + __nodemask_parse_user((ubuf), (ulen), &(dst), MAX_NUMNODES) +static inline int __nodemask_parse_user(const char __user *buf, int len, nodemask_t *dstp, int nbits) { - return bitmap_parse(buf, len, dstp->bits, nbits); + return bitmap_parse_user(buf, len, dstp->bits, nbits); } #define nodelist_scnprintf(buf, len, src) \ diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 607c780..9a35266 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -57,7 +57,7 @@ static int irq_affinity_write_proc(struct file *file, const char __user *buffer, if (!irq_desc[irq].chip->set_affinity || no_irq_affinity) return -EIO; - err = cpumask_parse(buffer, count, new_value); + err = cpumask_parse_user(buffer, count, new_value); if (err) return err; diff --git a/kernel/profile.c b/kernel/profile.c index 857300a..f940b46 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -399,7 +399,7 @@ static int prof_cpu_mask_write_proc (struct file *file, const char __user *buffe unsigned long full_count = count, err; cpumask_t new_value; - err = cpumask_parse(buffer, count, new_value); + err = cpumask_parse_user(buffer, count, new_value); if (err) return err; diff --git a/lib/bitmap.c b/lib/bitmap.c index d71e38c..037fa9a 100644 --- a/lib/bitmap.c +++ b/lib/bitmap.c @@ -316,10 +316,11 @@ int bitmap_scnprintf(char *buf, unsigned int buflen, EXPORT_SYMBOL(bitmap_scnprintf); /** - * bitmap_parse - convert an ASCII hex string into a bitmap. - * @ubuf: pointer to buffer in user space containing string. - * @ubuflen: buffer size in bytes. If string is smaller than this + * __bitmap_parse - convert an ASCII hex string into a bitmap. + * @buf: pointer to buffer containing string. + * @buflen: buffer size in bytes. If string is smaller than this * then it must be terminated with a \0. + * @is_user: location of buffer, 0 indicates kernel space * @maskp: pointer to bitmap array that will contain result. * @nmaskbits: size of bitmap, in bits. * @@ -330,11 +331,13 @@ EXPORT_SYMBOL(bitmap_scnprintf); * characters and for grouping errors such as "1,,5", ",44", "," and "". * Leading and trailing whitespace accepted, but not embedded whitespace. */ -int bitmap_parse(const char __user *ubuf, unsigned int ubuflen, - unsigned long *maskp, int nmaskbits) +int __bitmap_parse(const char *buf, unsigned int buflen, + int is_user, unsigned long *maskp, + int nmaskbits) { int c, old_c, totaldigits, ndigits, nchunks, nbits; u32 chunk; + const char __user *ubuf = buf; bitmap_zero(maskp, nmaskbits); @@ -343,11 +346,15 @@ int bitmap_parse(const char __user *ubuf, unsigned int ubuflen, chunk = ndigits = 0; /* Get the next chunk of the bitmap */ - while (ubuflen) { + while (buflen) { old_c = c; - if (get_user(c, ubuf++)) - return -EFAULT; - ubuflen--; + if (is_user) { + if (__get_user(c, ubuf++)) + return -EFAULT; + } + else + c = *buf++; + buflen--; if (isspace(c)) continue; @@ -388,11 +395,36 @@ int bitmap_parse(const char __user *ubuf, unsigned int ubuflen, nbits += (nchunks == 1) ? nbits_to_hold_value(chunk) : CHUNKSZ; if (nbits > nmaskbits) return -EOVERFLOW; - } while (ubuflen && c == ','); + } while (buflen && c == ','); return 0; } -EXPORT_SYMBOL(bitmap_parse); +EXPORT_SYMBOL(__bitmap_parse); + +/** + * bitmap_parse_user() + * + * @ubuf: pointer to user buffer containing string. + * @ulen: buffer size in bytes. If string is smaller than this + * then it must be terminated with a \0. + * @maskp: pointer to bitmap array that will contain result. + * @nmaskbits: size of bitmap, in bits. + * + * Wrapper for __bitmap_parse(), providing it with user buffer. + * + * We cannot have this as an inline function in bitmap.h because it needs + * linux/uaccess.h to get the access_ok() declaration and this causes + * cyclic dependencies. + */ +int bitmap_parse_user(const char __user *ubuf, + unsigned int ulen, unsigned long *maskp, + int nmaskbits) +{ + if (!access_ok(VERIFY_READ, ubuf, ulen)) + return -EFAULT; + return __bitmap_parse((const char *)ubuf, ulen, 1, maskp, nmaskbits); +} +EXPORT_SYMBOL(bitmap_parse_user); /* * bscnl_emit(buf, buflen, rbot, rtop, bp) -- cgit v0.10.2 From cd0810410beb86c570aeb3dcd3dc1fc5ab9e6077 Mon Sep 17 00:00:00 2001 From: Matthias Urlichs Date: Wed, 11 Oct 2006 01:21:57 -0700 Subject: [PATCH] document the core-dump-to-a-pipe patch The pipe-a-coredump-to-a-program feature was undocumented. *Grumble*. NB: a good enhancement to that patch would be: save all the stuff that a core file can get from the %x expansions in the environment. Signed-off-by: Matthias Urlichs Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt index 89bf8c2..0bc7f1e 100644 --- a/Documentation/sysctl/kernel.txt +++ b/Documentation/sysctl/kernel.txt @@ -86,7 +86,7 @@ valid for 30 seconds. core_pattern: core_pattern is used to specify a core dumpfile pattern name. -. max length 64 characters; default value is "core" +. max length 128 characters; default value is "core" . core_pattern is used as a pattern template for the output filename; certain string patterns (beginning with '%') are substituted with their actual values. @@ -105,6 +105,9 @@ core_pattern is used to specify a core dumpfile pattern name. %h hostname %e executable filename % both are dropped +. If the first character of the pattern is a '|', the kernel will treat + the rest of the pattern as a command to run. The core dump will be + written to the standard input of that program instead of to a file. ============================================================== -- cgit v0.10.2 From 887ed2f3aecde2ff24e06666932dc5f144745044 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 11 Oct 2006 01:21:58 -0700 Subject: [PATCH] VM: Fix the gfp_mask in invalidate_complete_page2 If try_to_release_page() is called with a zero gfp mask, then the filesystem is effectively denied the possibility of sleeping while attempting to release the page. There doesn't appear to be any valid reason why this should be banned, given that we're not calling this from a memory allocation context. For this reason, change the gfp_mask argument of the call to GFP_KERNEL. Signed-off-by: Trond Myklebust Cc: Steve Dickson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/truncate.c b/mm/truncate.c index fca2883..11ca480 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -302,7 +302,7 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) if (page->mapping != mapping) return 0; - if (PagePrivate(page) && !try_to_release_page(page, 0)) + if (PagePrivate(page) && !try_to_release_page(page, GFP_KERNEL)) return 0; write_lock_irq(&mapping->tree_lock); -- cgit v0.10.2 From 97e860d364aa9d08e895ecb619b9122ae2c70df8 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Wed, 11 Oct 2006 01:21:59 -0700 Subject: [PATCH] Remove unnecessary check in fs/fat/inode.c Aince all callers dereference sb, and this function does so earlier too, we dont need the check. Signed-off-by: Eric Sesterhenn Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 4613cb2..78945b5 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -1472,7 +1472,7 @@ int fat_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2) ret = writeback_inode(i1); if (!ret && i2) ret = writeback_inode(i2); - if (!ret && sb) { + if (!ret) { struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping; ret = filemap_flush(mapping); } -- cgit v0.10.2 From 8c58165108e26d18849a0138c719e680f281197a Mon Sep 17 00:00:00 2001 From: Monakhov Dmitriy Date: Wed, 11 Oct 2006 01:22:00 -0700 Subject: [PATCH] D-cache aliasing issue in __block_prepare_write A couple of flush_dcache_page()s are missing on the I/O-error paths. Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/buffer.c b/fs/buffer.c index 2a7828c..f65ef88 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1854,6 +1854,7 @@ static int __block_prepare_write(struct inode *inode, struct page *page, clear_buffer_new(bh); kaddr = kmap_atomic(page, KM_USER0); memset(kaddr+block_start, 0, bh->b_size); + flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); set_buffer_uptodate(bh); mark_buffer_dirty(bh); @@ -2360,6 +2361,7 @@ failed: */ kaddr = kmap_atomic(page, KM_USER0); memset(kaddr, 0, PAGE_CACHE_SIZE); + flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); set_page_dirty(page); -- cgit v0.10.2 From 53d5ed627df852ba8bab7f70df25290bd733792c Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 11 Oct 2006 01:22:01 -0700 Subject: [PATCH] Use linux/io.h instead of asm/io.h In preparation for moving check_signature, change these users from asm/io.h to linux/io.h Signed-off-by: Matthew Wilcox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/xd.c b/drivers/block/xd.c index 10cc387..0d97b7e 100644 --- a/drivers/block/xd.c +++ b/drivers/block/xd.c @@ -48,9 +48,9 @@ #include #include #include +#include #include -#include #include #include diff --git a/drivers/input/misc/wistron_btns.c b/drivers/input/misc/wistron_btns.c index 4639537..7b9d1c1 100644 --- a/drivers/input/misc/wistron_btns.c +++ b/drivers/input/misc/wistron_btns.c @@ -17,7 +17,7 @@ * with this program; if not, write to the Free Software Foundation, Inc., * 59 Temple Place Suite 330, Boston, MA 02111-1307, USA. */ -#include +#include #include #include #include diff --git a/drivers/net/eth16i.c b/drivers/net/eth16i.c index 8cc3c33..b7b8bc2 100644 --- a/drivers/net/eth16i.c +++ b/drivers/net/eth16i.c @@ -162,9 +162,9 @@ static char *version = #include #include #include +#include #include -#include #include diff --git a/drivers/scsi/aha152x.c b/drivers/scsi/aha152x.c index a0d1cee..306f46b 100644 --- a/drivers/scsi/aha152x.c +++ b/drivers/scsi/aha152x.c @@ -238,7 +238,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/scsi/dtc.c b/drivers/scsi/dtc.c index 0d5713d..5475672 100644 --- a/drivers/scsi/dtc.c +++ b/drivers/scsi/dtc.c @@ -82,7 +82,7 @@ #include #include #include -#include +#include #include "scsi.h" #include #include "dtc.h" diff --git a/drivers/scsi/fdomain.c b/drivers/scsi/fdomain.c index 41b05fc4..72794a7 100644 --- a/drivers/scsi/fdomain.c +++ b/drivers/scsi/fdomain.c @@ -278,9 +278,9 @@ #include #include #include +#include #include -#include #include #include diff --git a/drivers/scsi/seagate.c b/drivers/scsi/seagate.c index 8ff1f28..5ffec27 100644 --- a/drivers/scsi/seagate.c +++ b/drivers/scsi/seagate.c @@ -97,8 +97,8 @@ #include #include #include +#include -#include #include #include diff --git a/drivers/scsi/t128.c b/drivers/scsi/t128.c index 2df6747..0b7a70f 100644 --- a/drivers/scsi/t128.c +++ b/drivers/scsi/t128.c @@ -109,7 +109,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/scsi/wd7000.c b/drivers/scsi/wd7000.c index 331e1cf..30be765 100644 --- a/drivers/scsi/wd7000.c +++ b/drivers/scsi/wd7000.c @@ -178,10 +178,10 @@ #include #include #include +#include #include #include -#include #include #include -- cgit v0.10.2 From e50190a8341485b413f599033cb74649f849d939 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 11 Oct 2006 01:22:02 -0700 Subject: [PATCH] Consolidate check_signature There's nothing arch-specific about check_signature(), so move it to . Use a cross between the Alpha and i386 implementations as the generic one. Signed-off-by: Matthew Wilcox Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-alpha/io.h b/include/asm-alpha/io.h index f5ae98c..5d15af2 100644 --- a/include/asm-alpha/io.h +++ b/include/asm-alpha/io.h @@ -533,19 +533,6 @@ extern void outsl (unsigned long port, const void *src, unsigned long count); #define eth_io_copy_and_sum(skb,src,len,unused) \ memcpy_fromio((skb)->data,src,len) -static inline int -check_signature(const volatile void __iomem *io_addr, - const unsigned char *signature, int length) -{ - do { - if (readb(io_addr) != *signature) - return 0; - io_addr++; - signature++; - } while (--length); - return 1; -} - /* * The Alpha Jensen hardware for some rather strange reason puts * the RTC clock at 0x170 instead of 0x70. Probably due to some diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h index 34aaaac..ae999fd 100644 --- a/include/asm-arm/io.h +++ b/include/asm-arm/io.h @@ -193,23 +193,6 @@ extern void _memset_io(volatile void __iomem *, int, size_t); #define eth_io_copy_and_sum(s,c,l,b) \ eth_copy_and_sum((s),__mem_pci(c),(l),(b)) -static inline int -check_signature(void __iomem *io_addr, const unsigned char *signature, - int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - #elif !defined(readb) #define readb(c) (__readwrite_bug("readb"),0) diff --git a/include/asm-frv/io.h b/include/asm-frv/io.h index 7765f55..20e44fe0 100644 --- a/include/asm-frv/io.h +++ b/include/asm-frv/io.h @@ -385,27 +385,6 @@ static inline void pci_iounmap(struct pci_dev *dev, void __iomem *p) */ #define xlate_dev_kmem_ptr(p) p -/* - * Check BIOS signature - */ -static inline int check_signature(volatile void __iomem *io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - - retval = 1; -out: - return retval; -} - #endif /* __KERNEL__ */ #endif /* _ASM_IO_H */ diff --git a/include/asm-i386/io.h b/include/asm-i386/io.h index b3724fe..68df0dc3 100644 --- a/include/asm-i386/io.h +++ b/include/asm-i386/io.h @@ -224,33 +224,6 @@ static inline void memcpy_toio(volatile void __iomem *dst, const void *src, int #define eth_io_copy_and_sum(a,b,c,d) eth_copy_and_sum((a),(void __force *)(b),(c),(d)) -/** - * check_signature - find BIOS signatures - * @io_addr: mmio address to check - * @signature: signature block - * @length: length of signature - * - * Perform a signature comparison with the mmio address io_addr. This - * address should have been obtained by ioremap. - * Returns 1 on a match. - */ - -static inline int check_signature(volatile void __iomem * io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - /* * Cache management * diff --git a/include/asm-m32r/io.h b/include/asm-m32r/io.h index 70ad1c9..d06933b 100644 --- a/include/asm-m32r/io.h +++ b/include/asm-m32r/io.h @@ -166,38 +166,6 @@ static inline void _writel(unsigned long l, unsigned long addr) #define flush_write_buffers() do { } while (0) /* M32R_FIXME */ -/** - * check_signature - find BIOS signatures - * @io_addr: mmio address to check - * @signature: signature block - * @length: length of signature - * - * Perform a signature comparison with the ISA mmio address io_addr. - * Returns 1 on a match. - * - * This function is deprecated. New drivers should use ioremap and - * check_signature. - */ - -static inline int check_signature(void __iomem *io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; -#if 0 -printk("check_signature\n"); - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: -#endif - return retval; -} - static inline void memset_io(volatile void __iomem *addr, unsigned char val, int count) { diff --git a/include/asm-mips/io.h b/include/asm-mips/io.h index df624e1..c2d124b 100644 --- a/include/asm-mips/io.h +++ b/include/asm-mips/io.h @@ -562,32 +562,6 @@ extern void pci_iounmap(struct pci_dev *dev, void __iomem *); #define eth_io_copy_and_sum(skb,src,len,unused) memcpy_fromio((skb)->data,(src),(len)) /* - * check_signature - find BIOS signatures - * @io_addr: mmio address to check - * @signature: signature block - * @length: length of signature - * - * Perform a signature comparison with the mmio address io_addr. This - * address should have been obtained by ioremap. - * Returns 1 on a match. - */ -static inline int check_signature(char __iomem *io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - -/* * The caches on some architectures aren't dma-coherent and have need to * handle this in software. There are three types of operations that * can be applied to dma buffers. diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h index cbbd8c6..3baff8b 100644 --- a/include/asm-powerpc/io.h +++ b/include/asm-powerpc/io.h @@ -404,32 +404,6 @@ static inline void __out_be64(volatile unsigned long __iomem *addr, unsigned lon #include -/** - * check_signature - find BIOS signatures - * @io_addr: mmio address to check - * @signature: signature block - * @length: length of signature - * - * Perform a signature comparison with the mmio address io_addr. This - * address should have been obtained by ioremap. - * Returns 1 on a match. - */ -static inline int check_signature(const volatile void __iomem * io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - /* Nothing to do */ #define dma_cache_inv(_start,_size) do { } while (0) diff --git a/include/asm-ppc/io.h b/include/asm-ppc/io.h index 3d9a9e6..a4c411b 100644 --- a/include/asm-ppc/io.h +++ b/include/asm-ppc/io.h @@ -439,22 +439,6 @@ extern inline void * phys_to_virt(unsigned long address) #define iobarrier_r() eieio() #define iobarrier_w() eieio() -static inline int check_signature(volatile void __iomem * io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - /* * Here comes the ppc implementation of the IOMAP * interfaces. diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h index ed12d38..a0e55b0 100644 --- a/include/asm-sh/io.h +++ b/include/asm-sh/io.h @@ -304,22 +304,6 @@ __ioremap_mode(unsigned long offset, unsigned long size, unsigned long flags) #define iounmap(addr) \ __iounmap((addr)) -static inline int check_signature(char __iomem *io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - /* * The caches on some architectures aren't dma-coherent and have need to * handle this in software. There are three types of operations that diff --git a/include/asm-sh64/io.h b/include/asm-sh64/io.h index 252fedb..14d8e7b 100644 --- a/include/asm-sh64/io.h +++ b/include/asm-sh64/io.h @@ -178,22 +178,6 @@ extern void iounmap(void *addr); unsigned long onchip_remap(unsigned long addr, unsigned long size, const char* name); extern void onchip_unmap(unsigned long vaddr); -static __inline__ int check_signature(volatile void __iomem *io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - /* * The caches on some architectures aren't dma-coherent and have need to * handle this in software. There are three types of operations that diff --git a/include/asm-sparc64/io.h b/include/asm-sparc64/io.h index 0056770..30b912d 100644 --- a/include/asm-sparc64/io.h +++ b/include/asm-sparc64/io.h @@ -440,21 +440,6 @@ _memcpy_toio(volatile void __iomem *dst, const void *src, __kernel_size_t n) #define memcpy_toio(d,s,sz) _memcpy_toio(d,s,sz) -static inline int check_signature(void __iomem *io_addr, - const unsigned char *signature, - int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature++) - goto out; - io_addr++; - } while (--length); - retval = 1; -out: - return retval; -} - #define mmiowb() #ifdef __KERNEL__ diff --git a/include/asm-x86_64/io.h b/include/asm-x86_64/io.h index 70e91fe..6ee9fad 100644 --- a/include/asm-x86_64/io.h +++ b/include/asm-x86_64/io.h @@ -254,33 +254,6 @@ void memset_io(volatile void __iomem *a, int b, size_t c); #define eth_io_copy_and_sum(a,b,c,d) eth_copy_and_sum((a),(void *)(b),(c),(d)) -/** - * check_signature - find BIOS signatures - * @io_addr: mmio address to check - * @signature: signature block - * @length: length of signature - * - * Perform a signature comparison with the mmio address io_addr. This - * address should have been obtained by ioremap. - * Returns 1 on a match. - */ - -static inline int check_signature(void __iomem *io_addr, - const unsigned char *signature, int length) -{ - int retval = 0; - do { - if (readb(io_addr) != *signature) - goto out; - io_addr++; - signature++; - length--; - } while (length); - retval = 1; -out: - return retval; -} - /* Nothing to do */ #define dma_cache_inv(_start,_size) do { } while (0) diff --git a/include/linux/io.h b/include/linux/io.h index 2ad96c3..81877ea 100644 --- a/include/linux/io.h +++ b/include/linux/io.h @@ -28,4 +28,31 @@ void __iowrite64_copy(void __iomem *to, const void *from, size_t count); int ioremap_page_range(unsigned long addr, unsigned long end, unsigned long phys_addr, pgprot_t prot); +/** + * check_signature - find BIOS signatures + * @io_addr: mmio address to check + * @signature: signature block + * @length: length of signature + * + * Perform a signature comparison with the mmio address io_addr. This + * address should have been obtained by ioremap. + * Returns 1 on a match. + */ + +static inline int check_signature(const volatile void __iomem *io_addr, + const unsigned char *signature, int length) +{ + int retval = 0; + do { + if (readb(io_addr) != *signature) + goto out; + io_addr++; + signature++; + length--; + } while (length); + retval = 1; +out: + return retval; +} + #endif /* _LINUX_IO_H */ -- cgit v0.10.2 From 53bc5b2db16ceefdd972b9ffd1f7bde5c427939e Mon Sep 17 00:00:00 2001 From: Aneesh Kumar Date: Wed, 11 Oct 2006 01:22:03 -0700 Subject: [PATCH] Fix typos in mm/shmem_acl.c Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/shmem_acl.c b/mm/shmem_acl.c index c946bf4..f5664c5 100644 --- a/mm/shmem_acl.c +++ b/mm/shmem_acl.c @@ -35,7 +35,7 @@ shmem_get_acl(struct inode *inode, int type) } /** - * shmem_get_acl - generic_acl_operations->setacl() operation + * shmem_set_acl - generic_acl_operations->setacl() operation */ static void shmem_set_acl(struct inode *inode, int type, struct posix_acl *acl) -- cgit v0.10.2 From fbab41ccc479b6b0ba15c137af9e0b1c100bff24 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 11 Oct 2006 01:22:04 -0700 Subject: [PATCH] HT_IRQ must depend on PCI CONFIG_PCI=n, CONFIG_HT_IRQ=y results in the following compile error: ... LD vmlinux arch/i386/mach-generic/built-in.o: In function `apicid_to_node': summit.c:(.text+0x53): undefined reference to `apicid_2_node' arch/i386/kernel/built-in.o: In function `arch_setup_ht_irq': (.text+0xcf79): undefined reference to `write_ht_irq_low' arch/i386/kernel/built-in.o: In function `arch_setup_ht_irq': (.text+0xcf85): undefined reference to `write_ht_irq_high' arch/i386/kernel/built-in.o: In function `k7nops': alternative.c:(.data+0x1358): undefined reference to `mask_ht_irq' alternative.c:(.data+0x1360): undefined reference to `unmask_ht_irq' make[1]: *** [vmlinux] Error 1 Bug report by Jesper Juhl. Signed-off-by: Adrian Bunk Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index 3029412..ecc50db 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -55,7 +55,7 @@ config PCI_DEBUG config HT_IRQ bool "Interrupts on hypertransport devices" default y - depends on X86_LOCAL_APIC && X86_IO_APIC + depends on PCI && X86_LOCAL_APIC && X86_IO_APIC help This allows native hypertransport devices to use interrupts. -- cgit v0.10.2 From 2ecd05ae68a903761e736e9e0aca40d6ace4319e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 11 Oct 2006 01:22:05 -0700 Subject: [PATCH] fs/*: use BUILD_BUG_ON Signed-off-by: Alexey Dobriyan Cc: David Woodhouse Cc: David Howells Cc: Mark Fasheh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/afs/dir.c b/fs/afs/dir.c index cf8a2cb..a6ec75c 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -211,8 +211,8 @@ static int afs_dir_open(struct inode *inode, struct file *file) { _enter("{%lu}", inode->i_ino); - BUG_ON(sizeof(union afs_dir_block) != 2048); - BUG_ON(sizeof(union afs_dirent) != 32); + BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_dirent) != 32); if (AFS_FS_I(inode)->flags & AFS_VNODE_DELETED) return -ENOENT; @@ -446,8 +446,8 @@ static struct dentry *afs_dir_lookup(struct inode *dir, struct dentry *dentry, _enter("{%lu},%p{%s}", dir->i_ino, dentry, dentry->d_name.name); /* insanity checks first */ - BUG_ON(sizeof(union afs_dir_block) != 2048); - BUG_ON(sizeof(union afs_dirent) != 32); + BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_dirent) != 32); if (dentry->d_name.len > 255) { _leave(" = -ENAMETOOLONG"); diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 6de3745..bc4b810 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -334,10 +334,10 @@ static int __init init_jffs2_fs(void) which means just 'no padding', without the alignment thing. But GCC doesn't have that -- we have to just hope the structs are the right sizes, instead. */ - BUG_ON(sizeof(struct jffs2_unknown_node) != 12); - BUG_ON(sizeof(struct jffs2_raw_dirent) != 40); - BUG_ON(sizeof(struct jffs2_raw_inode) != 68); - BUG_ON(sizeof(struct jffs2_raw_summary) != 32); + BUILD_BUG_ON(sizeof(struct jffs2_unknown_node) != 12); + BUILD_BUG_ON(sizeof(struct jffs2_raw_dirent) != 40); + BUILD_BUG_ON(sizeof(struct jffs2_raw_inode) != 68); + BUILD_BUG_ON(sizeof(struct jffs2_raw_summary) != 32); printk(KERN_INFO "JFFS2 version 2.2." #ifdef CONFIG_JFFS2_FS_WRITEBUFFER diff --git a/fs/minix/inode.c b/fs/minix/inode.c index c11a4b9..1e36bae 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -149,12 +149,8 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) return -ENOMEM; s->s_fs_info = sbi; - /* N.B. These should be compile-time tests. - Unfortunately that is impossible. */ - if (32 != sizeof (struct minix_inode)) - panic("bad V1 i-node size"); - if (64 != sizeof(struct minix2_inode)) - panic("bad V2 i-node size"); + BUILD_BUG_ON(32 != sizeof (struct minix_inode)); + BUILD_BUG_ON(64 != sizeof(struct minix2_inode)); if (!sb_set_blocksize(s, BLOCK_SIZE)) goto out_bad_hblock; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 4c29cd7..76b46eb 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -339,7 +339,7 @@ static unsigned long long ocfs2_max_file_offset(unsigned int blockshift) #if BITS_PER_LONG == 32 # if defined(CONFIG_LBD) - BUG_ON(sizeof(sector_t) != 8); + BUILD_BUG_ON(sizeof(sector_t) != 8); pagefactor = PAGE_CACHE_SIZE; bitshift = BITS_PER_LONG; # else diff --git a/fs/sysv/super.c b/fs/sysv/super.c index 350cba5..dc9e7dc 100644 --- a/fs/sysv/super.c +++ b/fs/sysv/super.c @@ -358,16 +358,11 @@ static int sysv_fill_super(struct super_block *sb, void *data, int silent) unsigned long blocknr; int size = 0, i; - if (1024 != sizeof (struct xenix_super_block)) - panic("Xenix FS: bad superblock size"); - if (512 != sizeof (struct sysv4_super_block)) - panic("SystemV FS: bad superblock size"); - if (512 != sizeof (struct sysv2_super_block)) - panic("SystemV FS: bad superblock size"); - if (500 != sizeof (struct coh_super_block)) - panic("Coherent FS: bad superblock size"); - if (64 != sizeof (struct sysv_inode)) - panic("sysv fs: bad inode size"); + BUILD_BUG_ON(1024 != sizeof (struct xenix_super_block)); + BUILD_BUG_ON(512 != sizeof (struct sysv4_super_block)); + BUILD_BUG_ON(512 != sizeof (struct sysv2_super_block)); + BUILD_BUG_ON(500 != sizeof (struct coh_super_block)); + BUILD_BUG_ON(64 != sizeof (struct sysv_inode)); sbi = kzalloc(sizeof(struct sysv_sb_info), GFP_KERNEL); if (!sbi) -- cgit v0.10.2 From 39913b31d0cd11b5b18a303e220c95ffbd9e1a88 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 11 Oct 2006 01:22:06 -0700 Subject: [PATCH] DAC960: use memmove for overlapping areas Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/DAC960.h b/drivers/block/DAC960.h index cec539e..6148073 100644 --- a/drivers/block/DAC960.h +++ b/drivers/block/DAC960.h @@ -4379,8 +4379,8 @@ static inline void DAC960_P_To_PD_TranslateEnquiry(void *Enquiry) static inline void DAC960_P_To_PD_TranslateDeviceState(void *DeviceState) { memcpy(DeviceState + 2, DeviceState + 3, 1); - memcpy(DeviceState + 4, DeviceState + 5, 2); - memcpy(DeviceState + 6, DeviceState + 8, 4); + memmove(DeviceState + 4, DeviceState + 5, 2); + memmove(DeviceState + 6, DeviceState + 8, 4); } static inline -- cgit v0.10.2 From 3dc3099a9b2c346b16383597fadaa79a05a52388 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 11 Oct 2006 01:22:06 -0700 Subject: [PATCH] lockdep: use BUILD_BUG_ON Signed-off-by: Alexey Dobriyan Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 4c05534..805a322 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -1114,8 +1114,6 @@ static int count_matching_names(struct lock_class *new_class) return count + 1; } -extern void __error_too_big_MAX_LOCKDEP_SUBCLASSES(void); - /* * Register a lock's class in the hash-table, if the class is not present * yet. Otherwise we look it up. We cache the result in the lock object @@ -1153,8 +1151,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) * (or spin_lock_init()) call - which acts as the key. For static * locks we use the lock object itself as the key. */ - if (sizeof(struct lock_class_key) > sizeof(struct lock_class)) - __error_too_big_MAX_LOCKDEP_SUBCLASSES(); + BUILD_BUG_ON(sizeof(struct lock_class_key) > sizeof(struct lock_class)); key = lock->key->subkeys + subclass; -- cgit v0.10.2 From 5fcce7432a731e67f1cd3ef1e71dca9ea84dedb1 Mon Sep 17 00:00:00 2001 From: Aneesh Kumar Date: Wed, 11 Oct 2006 01:22:07 -0700 Subject: [PATCH] fix lockdep-design.txt I was looking at lockdep-desing.txt and i guess i am confused with the changes with respect to fd7bcea35e7efb108c34ee2b3840942a3749cadb. It says + '.' acquired while irqs enabled + '+' acquired in irq context + '-' acquired in process context with irqs disabled + '?' read-acquired both with irqs enabled and in irq context + But the get_usage_chars() function does this for '-' if (class->usage_mask & LOCKF_ENABLED_HARDIRQS) *c1 = '-'; So i guess what would be correct would be '.' acquired while irqs disabled '+' acquired in irq context '-' acquired with irqs enabled '?' read acquired in irq context with irqs enabled. Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/lockdep-design.txt b/Documentation/lockdep-design.txt index dab123d..4887730 100644 --- a/Documentation/lockdep-design.txt +++ b/Documentation/lockdep-design.txt @@ -50,10 +50,10 @@ The bit position indicates hardirq, softirq, hardirq-read, softirq-read respectively, and the character displayed in each indicates: - '.' acquired while irqs enabled + '.' acquired while irqs disabled '+' acquired in irq context - '-' acquired in process context with irqs disabled - '?' read-acquired both with irqs enabled and in irq context + '-' acquired with irqs enabled + '?' read acquired in irq context with irqs enabled. Unused mutexes cannot be part of the cause of an error. -- cgit v0.10.2 From 256a6b41365e17cebe5c2fc91ddff716c9aa055a Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 11 Oct 2006 01:22:08 -0700 Subject: [PATCH] lockdep: fix printk recursion logic Bug reported and fixed by Tilman Schmidt : if lockdep is enabled then log messages make it to /var/log/messages belatedly. The reason is a missed wakeup of klogd. Initially there was only a lockdep_internal() protection against lockdep recursion within vprintk() - it grew the 'outer' lockdep_off()/on() protection only later on. But that lockdep_off() made the release_console_sem() within vprintk() always happen under the lockdep_internal() condition, causing the bug. The right solution to remove the inner protection against recursion here - the outer one is enough. Signed-off-by: Ingo Molnar Cc: Tilman Schmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/printk.c b/kernel/printk.c index 771f5e8..f7d427e 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -820,15 +820,8 @@ void release_console_sem(void) console_locked = 0; up(&console_sem); spin_unlock_irqrestore(&logbuf_lock, flags); - if (wake_klogd && !oops_in_progress && waitqueue_active(&log_wait)) { - /* - * If we printk from within the lock dependency code, - * from within the scheduler code, then do not lock - * up due to self-recursion: - */ - if (!lockdep_internal()) - wake_up_interruptible(&log_wait); - } + if (wake_klogd && !oops_in_progress && waitqueue_active(&log_wait)) + wake_up_interruptible(&log_wait); } EXPORT_SYMBOL(release_console_sem); -- cgit v0.10.2 From 35e38a6e03bedba7c633ae03d286accd85ab6a42 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 11 Oct 2006 01:22:09 -0700 Subject: [PATCH] kernel-doc: fix function name in usercopy.c Fix kernel-doc function name in usercopy.c. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/lib/usercopy.c b/arch/i386/lib/usercopy.c index 08502fc..258df6b 100644 --- a/arch/i386/lib/usercopy.c +++ b/arch/i386/lib/usercopy.c @@ -179,7 +179,7 @@ __clear_user(void __user *to, unsigned long n) EXPORT_SYMBOL(__clear_user); /** - * strlen_user: - Get the size of a string in user space. + * strnlen_user: - Get the size of a string in user space. * @s: The string to measure. * @n: The maximum valid length * -- cgit v0.10.2 From 9c7fff6ef36526fb54694ee8201870f98b6a1747 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 11 Oct 2006 01:22:10 -0700 Subject: [PATCH] uaccess.h: match kernel-doc and function names Place kernel-doc function comment header immediately before the function that is being documented. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-i386/uaccess.h b/include/asm-i386/uaccess.h index 54d905e..eef5133 100644 --- a/include/asm-i386/uaccess.h +++ b/include/asm-i386/uaccess.h @@ -404,20 +404,6 @@ unsigned long __must_check __copy_from_user_ll_nocache_nozero(void *to, * anything, so this is accurate. */ -/** - * __copy_to_user: - Copy a block of data into user space, with less checking. - * @to: Destination address, in user space. - * @from: Source address, in kernel space. - * @n: Number of bytes to copy. - * - * Context: User context only. This function may sleep. - * - * Copy data from kernel space to user space. Caller must check - * the specified block with access_ok() before calling this function. - * - * Returns number of bytes that could not be copied. - * On success, this will be zero. - */ static __always_inline unsigned long __must_check __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) { @@ -439,35 +425,27 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) return __copy_to_user_ll(to, from, n); } -static __always_inline unsigned long __must_check -__copy_to_user(void __user *to, const void *from, unsigned long n) -{ - might_sleep(); - return __copy_to_user_inatomic(to, from, n); -} - /** - * __copy_from_user: - Copy a block of data from user space, with less checking. - * @to: Destination address, in kernel space. - * @from: Source address, in user space. + * __copy_to_user: - Copy a block of data into user space, with less checking. + * @to: Destination address, in user space. + * @from: Source address, in kernel space. * @n: Number of bytes to copy. * * Context: User context only. This function may sleep. * - * Copy data from user space to kernel space. Caller must check + * Copy data from kernel space to user space. Caller must check * the specified block with access_ok() before calling this function. * * Returns number of bytes that could not be copied. * On success, this will be zero. - * - * If some data could not be copied, this function will pad the copied - * data to the requested size using zero bytes. - * - * An alternate version - __copy_from_user_inatomic() - may be called from - * atomic context and will fail rather than sleep. In this case the - * uncopied bytes will *NOT* be padded with zeros. See fs/filemap.h - * for explanation of why this is needed. */ +static __always_inline unsigned long __must_check +__copy_to_user(void __user *to, const void *from, unsigned long n) +{ + might_sleep(); + return __copy_to_user_inatomic(to, from, n); +} + static __always_inline unsigned long __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) { @@ -493,6 +471,29 @@ __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) } return __copy_from_user_ll_nozero(to, from, n); } + +/** + * __copy_from_user: - Copy a block of data from user space, with less checking. + * @to: Destination address, in kernel space. + * @from: Source address, in user space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from user space to kernel space. Caller must check + * the specified block with access_ok() before calling this function. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + * + * If some data could not be copied, this function will pad the copied + * data to the requested size using zero bytes. + * + * An alternate version - __copy_from_user_inatomic() - may be called from + * atomic context and will fail rather than sleep. In this case the + * uncopied bytes will *NOT* be padded with zeros. See fs/filemap.h + * for explanation of why this is needed. + */ static __always_inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) { -- cgit v0.10.2 From 32e794015aa698fd6bf1ec21b713f900a707fb15 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 11 Oct 2006 01:22:10 -0700 Subject: [PATCH] kernel-doc: drop various "inline" qualifiers Drop __inline, __always_inline, and noinline in the produced kernel-doc output, similar to other pseudo directives. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/kernel-doc b/scripts/kernel-doc index 00d1ad1..7bc53a9 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -1518,6 +1518,9 @@ sub dump_function($$) { $prototype =~ s/^asmlinkage +//; $prototype =~ s/^inline +//; $prototype =~ s/^__inline__ +//; + $prototype =~ s/^__inline +//; + $prototype =~ s/^__always_inline +//; + $prototype =~ s/^noinline +//; $prototype =~ s/__devinit +//; $prototype =~ s/^#define +//; #ak added $prototype =~ s/__attribute__ \(\([a-z,]*\)\)//; -- cgit v0.10.2 From c751c1dbb1289d220a8a175ba0df47706ce95a7e Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Wed, 11 Oct 2006 01:22:11 -0700 Subject: [PATCH] include linux/types.h in linux/nbd.h The nbd header uses __be32 and such types but doesn't actually include the header that defines these things (linux/types.h); so let's include it. Signed-off-by: Mike Frysinger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/nbd.h b/include/linux/nbd.h index e712e7d..d6b6dc0 100644 --- a/include/linux/nbd.h +++ b/include/linux/nbd.h @@ -15,6 +15,8 @@ #ifndef LINUX_NBD_H #define LINUX_NBD_H +#include + #define NBD_SET_SOCK _IO( 0xab, 0 ) #define NBD_SET_BLKSIZE _IO( 0xab, 1 ) #define NBD_SET_SIZE _IO( 0xab, 2 ) -- cgit v0.10.2 From 272057447f646c51bc77c60044eb21c683fa366d Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 11 Oct 2006 01:22:12 -0700 Subject: [PATCH] kernel-doc: make parameter description indentation uniform - In parameter descriptions, strip all whitespace between the parameter name (e.g., @len) and its description so that the description is indented uniformly in text and man page modes. Previously, spaces or tabs (which are used for cleaner source code viewing) affected the produced output in a negative way. Before (man mode): to Destination address, in user space. from Source address, in kernel space. n Number of bytes to copy. After (man mode): to Destination address, in user space. from Source address, in kernel space. n Number of bytes to copy. - Fix/clarify a few function description comments. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/kernel-doc b/scripts/kernel-doc index 7bc53a9..187f5de 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -1262,7 +1262,9 @@ sub output_intro_text(%) { } ## -# generic output function for typedefs +# generic output function for all types (function, struct/union, typedef, enum); +# calls the generated, variable output_ function name based on +# functype and output_mode sub output_declaration { no strict 'refs'; my $name = shift; @@ -1278,8 +1280,7 @@ sub output_declaration { } ## -# generic output function - calls the right one based -# on current output mode. +# generic output function - calls the right one based on current output mode. sub output_intro { no strict 'refs'; my $func = "output_intro_".$output_mode; @@ -1781,8 +1782,9 @@ sub process_file($) { $in_doc_sect = 1; $contents = $newcontents; if ($contents ne "") { - if (substr($contents, 0, 1) eq " ") { - $contents = substr($contents, 1); + while ((substr($contents, 0, 1) eq " ") || + substr($contents, 0, 1) eq "\t") { + $contents = substr($contents, 1); } $contents .= "\n"; } -- cgit v0.10.2 From 9e42ef777f62277ea9bb70976be65bb374e00b9c Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 01:22:13 -0700 Subject: [PATCH] dell_rbu: printk() warning fix drivers/firmware/dell_rbu.c: In function 'packetize_data': drivers/firmware/dell_rbu.c:252: warning: format '%lu' expects type 'long unsigned int', but argument 3 has type 'int' Cc: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/firmware/dell_rbu.c b/drivers/firmware/dell_rbu.c index fc17599..8136d77 100644 --- a/drivers/firmware/dell_rbu.c +++ b/drivers/firmware/dell_rbu.c @@ -249,7 +249,7 @@ static int packetize_data(void *data, size_t length) if ((rc = create_packet(temp, packet_length))) return rc; - pr_debug("%p:%lu\n", temp, (end - temp)); + pr_debug("%p:%td\n", temp, (end - temp)); temp += packet_length; } -- cgit v0.10.2 From edc666e2ff9ec2e4e9510f1127c68c22cffc93f6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 11 Oct 2006 01:22:14 -0700 Subject: [PATCH] ReiserFS: Make sure all dentries refs are released before calling kill_block_super() Make sure all dentries refs are released before calling kill_block_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true. What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_block_super(). Changes made in [try #2]: (*) reiserfs_kill_sb() now checks that the superblock FS info pointer is set before trying to dereference it. Signed-off-by: David Howells Cc: "Rafael J. Wysocki" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index c89aa23..9041802 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -430,20 +430,29 @@ int remove_save_link(struct inode *inode, int truncate) return journal_end(&th, inode->i_sb, JOURNAL_PER_BALANCE_CNT); } -static void reiserfs_put_super(struct super_block *s) +static void reiserfs_kill_sb(struct super_block *s) { - struct reiserfs_transaction_handle th; - th.t_trans_id = 0; + if (REISERFS_SB(s)) { + if (REISERFS_SB(s)->xattr_root) { + d_invalidate(REISERFS_SB(s)->xattr_root); + dput(REISERFS_SB(s)->xattr_root); + REISERFS_SB(s)->xattr_root = NULL; + } - if (REISERFS_SB(s)->xattr_root) { - d_invalidate(REISERFS_SB(s)->xattr_root); - dput(REISERFS_SB(s)->xattr_root); + if (REISERFS_SB(s)->priv_root) { + d_invalidate(REISERFS_SB(s)->priv_root); + dput(REISERFS_SB(s)->priv_root); + REISERFS_SB(s)->priv_root = NULL; + } } - if (REISERFS_SB(s)->priv_root) { - d_invalidate(REISERFS_SB(s)->priv_root); - dput(REISERFS_SB(s)->priv_root); - } + kill_block_super(s); +} + +static void reiserfs_put_super(struct super_block *s) +{ + struct reiserfs_transaction_handle th; + th.t_trans_id = 0; /* change file system state to current state if it was mounted with read-write permissions */ if (!(s->s_flags & MS_RDONLY)) { @@ -2156,7 +2165,7 @@ struct file_system_type reiserfs_fs_type = { .owner = THIS_MODULE, .name = "reiserfs", .get_sb = get_super_block, - .kill_sb = kill_block_super, + .kill_sb = reiserfs_kill_sb, .fs_flags = FS_REQUIRES_DEV, }; -- cgit v0.10.2 From 6ce315234aefcbc599dea390c15672156ebf9e7b Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 11 Oct 2006 01:22:15 -0700 Subject: [PATCH] AUTOFS: Make sure all dentries refs are released before calling kill_anon_super() Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true. What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super(). This makes the struct autofs_sb_info::root member variable redundant (since sb->s_root is still available), and so that is removed. The calls to shrink_dcache_sb() are also removed since they're also redundant as shrink_dcache_for_umount() will now be called after the cleanup routine. Signed-off-by: David Howells Acked-by: Ian Kent Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 480ab17..b13f32c 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -94,7 +94,6 @@ struct autofs_wait_queue { struct autofs_sb_info { u32 magic; - struct dentry *root; int pipefd; struct file *pipe; pid_t oz_pgrp; @@ -229,4 +228,4 @@ out: } void autofs4_dentry_release(struct dentry *); - +extern void autofs4_kill_sb(struct super_block *); diff --git a/fs/autofs4/init.c b/fs/autofs4/init.c index 5d91933..723a1c5e 100644 --- a/fs/autofs4/init.c +++ b/fs/autofs4/init.c @@ -24,7 +24,7 @@ static struct file_system_type autofs_fs_type = { .owner = THIS_MODULE, .name = "autofs", .get_sb = autofs_get_sb, - .kill_sb = kill_anon_super, + .kill_sb = autofs4_kill_sb, }; static int __init init_autofs4_fs(void) diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 800ce87..51fd859 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -96,7 +96,7 @@ void autofs4_free_ino(struct autofs_info *ino) */ static void autofs4_force_release(struct autofs_sb_info *sbi) { - struct dentry *this_parent = sbi->root; + struct dentry *this_parent = sbi->sb->s_root; struct list_head *next; spin_lock(&dcache_lock); @@ -127,7 +127,7 @@ resume: spin_lock(&dcache_lock); } - if (this_parent != sbi->root) { + if (this_parent != sbi->sb->s_root) { struct dentry *dentry = this_parent; next = this_parent->d_u.d_child.next; @@ -140,15 +140,9 @@ resume: goto resume; } spin_unlock(&dcache_lock); - - dput(sbi->root); - sbi->root = NULL; - shrink_dcache_sb(sbi->sb); - - return; } -static void autofs4_put_super(struct super_block *sb) +void autofs4_kill_sb(struct super_block *sb) { struct autofs_sb_info *sbi = autofs4_sbi(sb); @@ -163,6 +157,7 @@ static void autofs4_put_super(struct super_block *sb) kfree(sbi); DPRINTK("shutting down"); + kill_anon_super(sb); } static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) @@ -189,7 +184,6 @@ static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) } static struct super_operations autofs4_sops = { - .put_super = autofs4_put_super, .statfs = simple_statfs, .show_options = autofs4_show_options, }; @@ -315,7 +309,6 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) s->s_fs_info = sbi; sbi->magic = AUTOFS_SBI_MAGIC; - sbi->root = NULL; sbi->pipefd = -1; sbi->catatonic = 0; sbi->exp_timeout = 0; @@ -397,13 +390,6 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) sbi->pipefd = pipefd; /* - * Take a reference to the root dentry so we get a chance to - * clean up the dentry tree on umount. - * See autofs4_force_release. - */ - sbi->root = dget(root); - - /* * Success! Install the root dentry now to indicate completion. */ s->s_root = root; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index ce103e7..c0a6c8d 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -45,7 +45,6 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) fput(sbi->pipe); /* Close the pipe */ sbi->pipe = NULL; } - shrink_dcache_sb(sbi->sb); } static int autofs4_write(struct file *file, const void *addr, int bytes) -- cgit v0.10.2 From c636ebdb186bf37f98d3839f69293597723edb36 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 11 Oct 2006 01:22:19 -0700 Subject: [PATCH] VFS: Destroy the dentries contributed by a superblock on unmounting The attached patch destroys all the dentries attached to a superblock in one go by: (1) Destroying the tree rooted at s_root. (2) Destroying every entry in the anon list, one at a time. (3) Each entry in the anon list has its subtree consumed from the leaves inwards. This reduces the amount of work generic_shutdown_super() does, and avoids iterating through the dentry_unused list. Note that locking is almost entirely absent in the shrink_dcache_for_umount*() functions added by this patch. This is because: (1) at the point the filesystem calls generic_shutdown_super(), it is not permitted to further touch the superblock's set of dentries, and nor may it remove aliases from inodes; (2) the dcache memory shrinker now skips dentries that are being unmounted; and (3) the superblock no longer has any external references through which the VFS can reach it. Given these points, the only locking we need to do is when we remove dentries from the unused list and the name hashes, which we do a directory's worth at a time. We also don't need to guard against reference counts going to zero unexpectedly and removing bits of the tree we're working on as nothing else can call dput(). A cut down version of dentry_iput() has been folded into shrink_dcache_for_umount_subtree() function. Apart from not needing to unlock things, it also doesn't need to check for inotify watches. In this version of the patch, the complaint about a dentry still being in use has been expanded from a single BUG_ON() and now gives much more information. Signed-off-by: David Howells Acked-by: NeilBrown Acked-by: Ian Kent Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/dcache.c b/fs/dcache.c index 2355bdd..2bac4ba 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -549,6 +549,136 @@ repeat: } /* + * destroy a single subtree of dentries for unmount + * - see the comments on shrink_dcache_for_umount() for a description of the + * locking + */ +static void shrink_dcache_for_umount_subtree(struct dentry *dentry) +{ + struct dentry *parent; + + BUG_ON(!IS_ROOT(dentry)); + + /* detach this root from the system */ + spin_lock(&dcache_lock); + if (!list_empty(&dentry->d_lru)) { + dentry_stat.nr_unused--; + list_del_init(&dentry->d_lru); + } + __d_drop(dentry); + spin_unlock(&dcache_lock); + + for (;;) { + /* descend to the first leaf in the current subtree */ + while (!list_empty(&dentry->d_subdirs)) { + struct dentry *loop; + + /* this is a branch with children - detach all of them + * from the system in one go */ + spin_lock(&dcache_lock); + list_for_each_entry(loop, &dentry->d_subdirs, + d_u.d_child) { + if (!list_empty(&loop->d_lru)) { + dentry_stat.nr_unused--; + list_del_init(&loop->d_lru); + } + + __d_drop(loop); + cond_resched_lock(&dcache_lock); + } + spin_unlock(&dcache_lock); + + /* move to the first child */ + dentry = list_entry(dentry->d_subdirs.next, + struct dentry, d_u.d_child); + } + + /* consume the dentries from this leaf up through its parents + * until we find one with children or run out altogether */ + do { + struct inode *inode; + + if (atomic_read(&dentry->d_count) != 0) { + printk(KERN_ERR + "BUG: Dentry %p{i=%lx,n=%s}" + " still in use (%d)" + " [unmount of %s %s]\n", + dentry, + dentry->d_inode ? + dentry->d_inode->i_ino : 0UL, + dentry->d_name.name, + atomic_read(&dentry->d_count), + dentry->d_sb->s_type->name, + dentry->d_sb->s_id); + BUG(); + } + + parent = dentry->d_parent; + if (parent == dentry) + parent = NULL; + else + atomic_dec(&parent->d_count); + + list_del(&dentry->d_u.d_child); + dentry_stat.nr_dentry--; /* For d_free, below */ + + inode = dentry->d_inode; + if (inode) { + dentry->d_inode = NULL; + list_del_init(&dentry->d_alias); + if (dentry->d_op && dentry->d_op->d_iput) + dentry->d_op->d_iput(dentry, inode); + else + iput(inode); + } + + d_free(dentry); + + /* finished when we fall off the top of the tree, + * otherwise we ascend to the parent and move to the + * next sibling if there is one */ + if (!parent) + return; + + dentry = parent; + + } while (list_empty(&dentry->d_subdirs)); + + dentry = list_entry(dentry->d_subdirs.next, + struct dentry, d_u.d_child); + } +} + +/* + * destroy the dentries attached to a superblock on unmounting + * - we don't need to use dentry->d_lock, and only need dcache_lock when + * removing the dentry from the system lists and hashes because: + * - the superblock is detached from all mountings and open files, so the + * dentry trees will not be rearranged by the VFS + * - s_umount is write-locked, so the memory pressure shrinker will ignore + * any dentries belonging to this superblock that it comes across + * - the filesystem itself is no longer permitted to rearrange the dentries + * in this superblock + */ +void shrink_dcache_for_umount(struct super_block *sb) +{ + struct dentry *dentry; + + if (down_read_trylock(&sb->s_umount)) + BUG(); + + dentry = sb->s_root; + sb->s_root = NULL; + atomic_dec(&dentry->d_count); + shrink_dcache_for_umount_subtree(dentry); + + while (!hlist_empty(&sb->s_anon)) { + dentry = hlist_entry(sb->s_anon.first, struct dentry, d_hash); + shrink_dcache_for_umount_subtree(dentry); + } +} + +/* * Search for at least 1 mount point in the dentry's subdirs. * We descend to the next level whenever the d_subdirs * list is non-empty and continue searching. diff --git a/fs/super.c b/fs/super.c index aec99dd..47e554c 100644 --- a/fs/super.c +++ b/fs/super.c @@ -260,17 +260,17 @@ int fsync_super(struct super_block *sb) * that need destruction out of superblock, call generic_shutdown_super() * and release aforementioned objects. Note: dentries and inodes _are_ * taken care of and do not need specific handling. + * + * Upon calling this function, the filesystem may no longer alter or + * rearrange the set of dentries belonging to this super_block, nor may it + * change the attachments of dentries to inodes. */ void generic_shutdown_super(struct super_block *sb) { - struct dentry *root = sb->s_root; struct super_operations *sop = sb->s_op; - if (root) { - sb->s_root = NULL; - shrink_dcache_parent(root); - shrink_dcache_sb(sb); - dput(root); + if (sb->s_root) { + shrink_dcache_for_umount(sb); fsync_super(sb); lock_super(sb); sb->s_flags &= ~MS_ACTIVE; diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 44605be..63f64a9 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -230,6 +230,7 @@ extern struct dentry * d_alloc_anon(struct inode *); extern struct dentry * d_splice_alias(struct inode *, struct dentry *); extern void shrink_dcache_sb(struct super_block *); extern void shrink_dcache_parent(struct dentry *); +extern void shrink_dcache_for_umount(struct super_block *); extern int d_invalidate(struct dentry *); /* only used at mount-time */ -- cgit v0.10.2 From 41bfcfd9ac0fbb59aaaa18e3ed5774e85b716de4 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:22:20 -0700 Subject: [PATCH] firmware/dell_rbu: handle sysfs errors Signed-off-by: Jeff Garzik Cc: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/firmware/dell_rbu.c b/drivers/firmware/dell_rbu.c index 8136d77..08b1617 100644 --- a/drivers/firmware/dell_rbu.c +++ b/drivers/firmware/dell_rbu.c @@ -718,14 +718,27 @@ static int __init dcdrbu_init(void) return -EIO; } - sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_data_attr); - sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_image_type_attr); - sysfs_create_bin_file(&rbu_device->dev.kobj, + rc = sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_data_attr); + if (rc) + goto out_devreg; + rc = sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_image_type_attr); + if (rc) + goto out_data; + rc = sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_packet_size_attr); + if (rc) + goto out_imtype; rbu_data.entry_created = 0; - return rc; + return 0; +out_imtype: + sysfs_remove_bin_file(&rbu_device->dev.kobj, &rbu_image_type_attr); +out_data: + sysfs_remove_bin_file(&rbu_device->dev.kobj, &rbu_data_attr); +out_devreg: + platform_device_unregister(rbu_device); + return rc; } static __exit void dcdrbu_exit(void) -- cgit v0.10.2 From 5e59393ec242d7b772356c95e2be48384cd0c5d7 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:22:21 -0700 Subject: [PATCH] ipmi: handle sysfs errors Signed-off-by: Jeff Garzik Acked-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 2455e8d..34a4fd1 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -1928,13 +1928,8 @@ static ssize_t guid_show(struct device *dev, struct device_attribute *attr, (long long) bmc->guid[8]); } -static void -cleanup_bmc_device(struct kref *ref) +static void remove_files(struct bmc_device *bmc) { - struct bmc_device *bmc; - - bmc = container_of(ref, struct bmc_device, refcount); - device_remove_file(&bmc->dev->dev, &bmc->device_id_attr); device_remove_file(&bmc->dev->dev, @@ -1951,12 +1946,23 @@ cleanup_bmc_device(struct kref *ref) &bmc->manufacturer_id_attr); device_remove_file(&bmc->dev->dev, &bmc->product_id_attr); + if (bmc->id.aux_firmware_revision_set) device_remove_file(&bmc->dev->dev, &bmc->aux_firmware_rev_attr); if (bmc->guid_set) device_remove_file(&bmc->dev->dev, &bmc->guid_attr); +} + +static void +cleanup_bmc_device(struct kref *ref) +{ + struct bmc_device *bmc; + + bmc = container_of(ref, struct bmc_device, refcount); + + remove_files(bmc); platform_device_unregister(bmc->dev); kfree(bmc); } @@ -1977,6 +1983,79 @@ static void ipmi_bmc_unregister(ipmi_smi_t intf) mutex_unlock(&ipmidriver_mutex); } +static int create_files(struct bmc_device *bmc) +{ + int err; + + err = device_create_file(&bmc->dev->dev, + &bmc->device_id_attr); + if (err) goto out; + err = device_create_file(&bmc->dev->dev, + &bmc->provides_dev_sdrs_attr); + if (err) goto out_devid; + err = device_create_file(&bmc->dev->dev, + &bmc->revision_attr); + if (err) goto out_sdrs; + err = device_create_file(&bmc->dev->dev, + &bmc->firmware_rev_attr); + if (err) goto out_rev; + err = device_create_file(&bmc->dev->dev, + &bmc->version_attr); + if (err) goto out_firm; + err = device_create_file(&bmc->dev->dev, + &bmc->add_dev_support_attr); + if (err) goto out_version; + err = device_create_file(&bmc->dev->dev, + &bmc->manufacturer_id_attr); + if (err) goto out_add_dev; + err = device_create_file(&bmc->dev->dev, + &bmc->product_id_attr); + if (err) goto out_manu; + if (bmc->id.aux_firmware_revision_set) { + err = device_create_file(&bmc->dev->dev, + &bmc->aux_firmware_rev_attr); + if (err) goto out_prod_id; + } + if (bmc->guid_set) { + err = device_create_file(&bmc->dev->dev, + &bmc->guid_attr); + if (err) goto out_aux_firm; + } + + return 0; + +out_aux_firm: + if (bmc->id.aux_firmware_revision_set) + device_remove_file(&bmc->dev->dev, + &bmc->aux_firmware_rev_attr); +out_prod_id: + device_remove_file(&bmc->dev->dev, + &bmc->product_id_attr); +out_manu: + device_remove_file(&bmc->dev->dev, + &bmc->manufacturer_id_attr); +out_add_dev: + device_remove_file(&bmc->dev->dev, + &bmc->add_dev_support_attr); +out_version: + device_remove_file(&bmc->dev->dev, + &bmc->version_attr); +out_firm: + device_remove_file(&bmc->dev->dev, + &bmc->firmware_rev_attr); +out_rev: + device_remove_file(&bmc->dev->dev, + &bmc->revision_attr); +out_sdrs: + device_remove_file(&bmc->dev->dev, + &bmc->provides_dev_sdrs_attr); +out_devid: + device_remove_file(&bmc->dev->dev, + &bmc->device_id_attr); +out: + return err; +} + static int ipmi_bmc_register(ipmi_smi_t intf) { int rv; @@ -2051,7 +2130,6 @@ static int ipmi_bmc_register(ipmi_smi_t intf) bmc->provides_dev_sdrs_attr.attr.mode = S_IRUGO; bmc->provides_dev_sdrs_attr.show = provides_dev_sdrs_show; - bmc->revision_attr.attr.name = "revision"; bmc->revision_attr.attr.owner = THIS_MODULE; bmc->revision_attr.attr.mode = S_IRUGO; @@ -2093,28 +2171,14 @@ static int ipmi_bmc_register(ipmi_smi_t intf) bmc->aux_firmware_rev_attr.attr.mode = S_IRUGO; bmc->aux_firmware_rev_attr.show = aux_firmware_rev_show; - device_create_file(&bmc->dev->dev, - &bmc->device_id_attr); - device_create_file(&bmc->dev->dev, - &bmc->provides_dev_sdrs_attr); - device_create_file(&bmc->dev->dev, - &bmc->revision_attr); - device_create_file(&bmc->dev->dev, - &bmc->firmware_rev_attr); - device_create_file(&bmc->dev->dev, - &bmc->version_attr); - device_create_file(&bmc->dev->dev, - &bmc->add_dev_support_attr); - device_create_file(&bmc->dev->dev, - &bmc->manufacturer_id_attr); - device_create_file(&bmc->dev->dev, - &bmc->product_id_attr); - if (bmc->id.aux_firmware_revision_set) - device_create_file(&bmc->dev->dev, - &bmc->aux_firmware_rev_attr); - if (bmc->guid_set) - device_create_file(&bmc->dev->dev, - &bmc->guid_attr); + rv = create_files(bmc); + if (rv) { + mutex_lock(&ipmidriver_mutex); + platform_device_unregister(bmc->dev); + mutex_unlock(&ipmidriver_mutex); + + return rv; + } printk(KERN_INFO "ipmi: Found new BMC (man_id: 0x%6.6x, " -- cgit v0.10.2 From 42ddfd6859b9d57490c94d26e29a43ffd78366e5 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:22:22 -0700 Subject: [PATCH] EISA: handle sysfs errors Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c index 3a365e1..d944647 100644 --- a/drivers/eisa/eisa-bus.c +++ b/drivers/eisa/eisa-bus.c @@ -226,14 +226,26 @@ static int __init eisa_init_device (struct eisa_root_device *root, static int __init eisa_register_device (struct eisa_device *edev) { - if (device_register (&edev->dev)) - return -1; + int rc = device_register (&edev->dev); + if (rc) + return rc; - device_create_file (&edev->dev, &dev_attr_signature); - device_create_file (&edev->dev, &dev_attr_enabled); - device_create_file (&edev->dev, &dev_attr_modalias); + rc = device_create_file (&edev->dev, &dev_attr_signature); + if (rc) goto err_devreg; + rc = device_create_file (&edev->dev, &dev_attr_enabled); + if (rc) goto err_sig; + rc = device_create_file (&edev->dev, &dev_attr_modalias); + if (rc) goto err_enab; return 0; + +err_enab: + device_remove_file (&edev->dev, &dev_attr_enabled); +err_sig: + device_remove_file (&edev->dev, &dev_attr_signature); +err_devreg: + device_unregister(&edev->dev); + return rc; } static int __init eisa_request_resources (struct eisa_root_device *root, -- cgit v0.10.2 From 69b2186c5fcb335e29c558e3b4e410e1939b5cc8 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:22:23 -0700 Subject: [PATCH] firmware/efivars: handle error Signed-off-by: Jeff Garzik Acked-by: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c index 8ebce1c03..5ab5e39 100644 --- a/drivers/firmware/efivars.c +++ b/drivers/firmware/efivars.c @@ -639,7 +639,12 @@ efivar_create_sysfs_entry(unsigned long variable_name_size, kobject_set_name(&new_efivar->kobj, "%s", short_name); kobj_set_kset_s(new_efivar, vars_subsys); - kobject_register(&new_efivar->kobj); + i = kobject_register(&new_efivar->kobj); + if (i) { + kfree(short_name); + kfree(new_efivar); + return 1; + } kfree(short_name); short_name = NULL; -- cgit v0.10.2 From 49a6cbe1cd8a72451d9d6ab5b1e163f17c1bbee3 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:22:23 -0700 Subject: [PATCH] drivers/mca: handle sysfs errors Also includes a kmalloc->kzalloc cleanup. Signed-off-by: Jeff Garzik Cc: James Bottomley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mca/mca-bus.c b/drivers/mca/mca-bus.c index 09baa43..da862e4 100644 --- a/drivers/mca/mca-bus.c +++ b/drivers/mca/mca-bus.c @@ -100,6 +100,7 @@ static DEVICE_ATTR(pos, S_IRUGO, mca_show_pos, NULL); int __init mca_register_device(int bus, struct mca_device *mca_dev) { struct mca_bus *mca_bus = mca_root_busses[bus]; + int rc; mca_dev->dev.parent = &mca_bus->dev; mca_dev->dev.bus = &mca_bus_type; @@ -108,13 +109,23 @@ int __init mca_register_device(int bus, struct mca_device *mca_dev) mca_dev->dev.dma_mask = &mca_dev->dma_mask; mca_dev->dev.coherent_dma_mask = mca_dev->dma_mask; - if (device_register(&mca_dev->dev)) - return 0; + rc = device_register(&mca_dev->dev); + if (rc) + goto err_out; - device_create_file(&mca_dev->dev, &dev_attr_id); - device_create_file(&mca_dev->dev, &dev_attr_pos); + rc = device_create_file(&mca_dev->dev, &dev_attr_id); + if (rc) goto err_out_devreg; + rc = device_create_file(&mca_dev->dev, &dev_attr_pos); + if (rc) goto err_out_id; return 1; + +err_out_id: + device_remove_file(&mca_dev->dev, &dev_attr_id); +err_out_devreg: + device_unregister(&mca_dev->dev); +err_out: + return 0; } /* */ @@ -130,13 +141,16 @@ struct mca_bus * __devinit mca_attach_bus(int bus) return NULL; } - mca_bus = kmalloc(sizeof(struct mca_bus), GFP_KERNEL); + mca_bus = kzalloc(sizeof(struct mca_bus), GFP_KERNEL); if (!mca_bus) return NULL; - memset(mca_bus, 0, sizeof(struct mca_bus)); + sprintf(mca_bus->dev.bus_id,"mca%d",bus); sprintf(mca_bus->name,"Host %s MCA Bridge", bus ? "Secondary" : "Primary"); - device_register(&mca_bus->dev); + if (device_register(&mca_bus->dev)) { + kfree(mca_bus); + return NULL; + } mca_root_busses[bus] = mca_bus; -- cgit v0.10.2 From bf02c082bf7a464518d45b9c178b8aa83f74dd5d Mon Sep 17 00:00:00 2001 From: Andreas Mohr Date: Wed, 11 Oct 2006 01:22:24 -0700 Subject: [PATCH] fs/bio.c: tweaks - Calculate a variable in bvec_alloc_bs() only once needed, not earlier (bio.o down from 18408 to 18376 Bytes, 32 Bytes saved, probably due to data locality improvements). - Init variable idx to silence a gcc warning which already existed in the unmodified original base file (bvec_alloc_bs() handles idx correctly, so there's no need for the warning): fs/bio.c: In function `bio_alloc_bioset': fs/bio.c:169: warning: `idx' may be used uninitialized in this function Signed-off-by: Andreas Mohr Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/bio.c b/fs/bio.c index 8f93e93..f95c874 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -79,7 +79,6 @@ static struct bio_set *fs_bio_set; static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs) { struct bio_vec *bvl; - struct biovec_slab *bp; /* * see comment near bvec_array define! @@ -98,10 +97,12 @@ static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned lon * idx now points to the pool we want to allocate from */ - bp = bvec_slabs + *idx; bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); - if (bvl) + if (bvl) { + struct biovec_slab *bp = bvec_slabs + *idx; + memset(bvl, 0, bp->nr_vecs * sizeof(struct bio_vec)); + } return bvl; } @@ -166,7 +167,7 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) bio_init(bio); if (likely(nr_iovecs)) { - unsigned long idx; + unsigned long idx = 0; /* shut up gcc */ bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); if (unlikely(!bvl)) { -- cgit v0.10.2 From 76fd020937f2d09f76a4cd8dbae1f3bec640ff0b Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 01:22:25 -0700 Subject: [PATCH] ISDN: several minor fixes pcbit: kill 'may be used uninitialized' warning. although the code does eventually fill the 32 bits it cares about, the variable truly is accessed uninitialized in each macro. Easier to just clean it up now. sc: fix a ton of obviously incorrect printk's (some with missing arguments even) Signed-off-by: Jeff Garzik Acked-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/pcbit/layer2.c b/drivers/isdn/pcbit/layer2.c index 13e7d21..937fd21 100644 --- a/drivers/isdn/pcbit/layer2.c +++ b/drivers/isdn/pcbit/layer2.c @@ -311,6 +311,7 @@ pcbit_deliver(void *data) dev->read_queue = frame->next; spin_unlock_irqrestore(&dev->lock, flags); + msg = 0; SET_MSG_CPU(msg, 0); SET_MSG_PROC(msg, 0); SET_MSG_CMD(msg, frame->skb->data[2]); diff --git a/drivers/isdn/sc/init.c b/drivers/isdn/sc/init.c index 222ca7c..06c9872 100644 --- a/drivers/isdn/sc/init.c +++ b/drivers/isdn/sc/init.c @@ -98,13 +98,14 @@ static int __init sc_init(void) * Confirm the I/O Address with a test */ if(io[b] == 0) { - pr_debug("I/O Address 0x%x is in use.\n"); + pr_debug("I/O Address invalid.\n"); continue; } outb(0x18, io[b] + 0x400 * EXP_PAGE0); if(inb(io[b] + 0x400 * EXP_PAGE0) != 0x18) { - pr_debug("I/O Base 0x%x fails test\n"); + pr_debug("I/O Base 0x%x fails test\n", + io[b] + 0x400 * EXP_PAGE0); continue; } } @@ -158,8 +159,8 @@ static int __init sc_init(void) outb(0xFF, io[b] + RESET_OFFSET); msleep_interruptible(10000); } - pr_debug("RAM Base for board %d is 0x%x, %s probe\n", b, ram[b], - ram[b] == 0 ? "will" : "won't"); + pr_debug("RAM Base for board %d is 0x%lx, %s probe\n", b, + ram[b], ram[b] == 0 ? "will" : "won't"); if(ram[b]) { /* @@ -168,7 +169,7 @@ static int __init sc_init(void) * board model */ if(request_region(ram[b], SRAM_PAGESIZE, "sc test")) { - pr_debug("request_region for RAM base 0x%x succeeded\n", ram[b]); + pr_debug("request_region for RAM base 0x%lx succeeded\n", ram[b]); model = identify_board(ram[b], io[b]); release_region(ram[b], SRAM_PAGESIZE); } @@ -204,7 +205,7 @@ static int __init sc_init(void) * Nope, there was no place in RAM for the * board, or it couldn't be identified */ - pr_debug("Failed to find an adapter at 0x%x\n", ram[b]); + pr_debug("Failed to find an adapter at 0x%lx\n", ram[b]); continue; } @@ -451,7 +452,7 @@ static int identify_board(unsigned long rambase, unsigned int iobase) HWConfig_pl hwci; int x; - pr_debug("Attempting to identify adapter @ 0x%x io 0x%x\n", + pr_debug("Attempting to identify adapter @ 0x%lx io 0x%x\n", rambase, iobase); /* @@ -490,7 +491,7 @@ static int identify_board(unsigned long rambase, unsigned int iobase) outb(PRI_BASEPG_VAL, pgport); msleep_interruptible(1000); sig = readl(rambase + SIG_OFFSET); - pr_debug("Looking for a signature, got 0x%x\n", sig); + pr_debug("Looking for a signature, got 0x%lx\n", sig); if(sig == SIGNATURE) return PRI_BOARD; @@ -500,7 +501,7 @@ static int identify_board(unsigned long rambase, unsigned int iobase) outb(BRI_BASEPG_VAL, pgport); msleep_interruptible(1000); sig = readl(rambase + SIG_OFFSET); - pr_debug("Looking for a signature, got 0x%x\n", sig); + pr_debug("Looking for a signature, got 0x%lx\n", sig); if(sig == SIGNATURE) return BRI_BOARD; @@ -510,7 +511,7 @@ static int identify_board(unsigned long rambase, unsigned int iobase) * Try to spot a card */ sig = readl(rambase + SIG_OFFSET); - pr_debug("Looking for a signature, got 0x%x\n", sig); + pr_debug("Looking for a signature, got 0x%lx\n", sig); if(sig != SIGNATURE) return -1; @@ -540,7 +541,7 @@ static int identify_board(unsigned long rambase, unsigned int iobase) memcpy_fromio(&rcvmsg, &(dpm->rsp_queue[dpm->rsp_tail]), MSG_LEN); pr_debug("Got HWConfig response, status = 0x%x\n", rcvmsg.rsp_status); memcpy(&hwci, &(rcvmsg.msg_data.HWCresponse), sizeof(HWConfig_pl)); - pr_debug("Hardware Config: Interface: %s, RAM Size: %d, Serial: %s\n" + pr_debug("Hardware Config: Interface: %s, RAM Size: %ld, Serial: %s\n" " Part: %s, Rev: %s\n", hwci.st_u_sense ? "S/T" : "U", hwci.ram_size, hwci.serial_no, hwci.part_no, hwci.rev_no); diff --git a/drivers/isdn/sc/packet.c b/drivers/isdn/sc/packet.c index f50defc..1e04676 100644 --- a/drivers/isdn/sc/packet.c +++ b/drivers/isdn/sc/packet.c @@ -44,7 +44,7 @@ int sndpkt(int devId, int channel, struct sk_buff *data) return -ENODEV; } - pr_debug("%s: sndpkt: frst = 0x%x nxt = %d f = %d n = %d\n", + pr_debug("%s: sndpkt: frst = 0x%lx nxt = %d f = %d n = %d\n", sc_adapter[card]->devicename, sc_adapter[card]->channel[channel].first_sendbuf, sc_adapter[card]->channel[channel].next_sendbuf, @@ -66,7 +66,7 @@ int sndpkt(int devId, int channel, struct sk_buff *data) ReqLnkWrite.buff_offset = sc_adapter[card]->channel[channel].next_sendbuf * BUFFER_SIZE + sc_adapter[card]->channel[channel].first_sendbuf; ReqLnkWrite.msg_len = data->len; /* sk_buff size */ - pr_debug("%s: writing %d bytes to buffer offset 0x%x\n", + pr_debug("%s: writing %d bytes to buffer offset 0x%lx\n", sc_adapter[card]->devicename, ReqLnkWrite.msg_len, ReqLnkWrite.buff_offset); memcpy_toshmem(card, (char *)ReqLnkWrite.buff_offset, data->data, ReqLnkWrite.msg_len); @@ -74,7 +74,7 @@ int sndpkt(int devId, int channel, struct sk_buff *data) /* * sendmessage */ - pr_debug("%s: sndpkt size=%d, buf_offset=0x%x buf_indx=%d\n", + pr_debug("%s: sndpkt size=%d, buf_offset=0x%lx buf_indx=%d\n", sc_adapter[card]->devicename, ReqLnkWrite.msg_len, ReqLnkWrite.buff_offset, sc_adapter[card]->channel[channel].next_sendbuf); @@ -124,7 +124,7 @@ void rcvpkt(int card, RspMessage *rcvmsg) return; } skb_put(skb, rcvmsg->msg_data.response.msg_len); - pr_debug("%s: getting data from offset: 0x%x\n", + pr_debug("%s: getting data from offset: 0x%lx\n", sc_adapter[card]->devicename, rcvmsg->msg_data.response.buff_offset); memcpy_fromshmem(card, @@ -143,7 +143,7 @@ void rcvpkt(int card, RspMessage *rcvmsg) /* memset_shmem(card, rcvmsg->msg_data.response.buff_offset, 0, BUFFER_SIZE); */ newll.buff_offset = rcvmsg->msg_data.response.buff_offset; newll.msg_len = BUFFER_SIZE; - pr_debug("%s: recycled buffer at offset 0x%x size %d\n", + pr_debug("%s: recycled buffer at offset 0x%lx size %d\n", sc_adapter[card]->devicename, newll.buff_offset, newll.msg_len); sendmessage(card, CEPID, ceReqTypeLnk, ceReqClass1, ceReqLnkRead, @@ -186,7 +186,7 @@ int setup_buffers(int card, int c) sc_adapter[card]->channel[c-1].num_sendbufs = nBuffers / 2; sc_adapter[card]->channel[c-1].free_sendbufs = nBuffers / 2; sc_adapter[card]->channel[c-1].next_sendbuf = 0; - pr_debug("%s: send buffer setup complete: first=0x%x n=%d f=%d, nxt=%d\n", + pr_debug("%s: send buffer setup complete: first=0x%lx n=%d f=%d, nxt=%d\n", sc_adapter[card]->devicename, sc_adapter[card]->channel[c-1].first_sendbuf, sc_adapter[card]->channel[c-1].num_sendbufs, @@ -203,7 +203,7 @@ int setup_buffers(int card, int c) ((sc_adapter[card]->channel[c-1].first_sendbuf + (nBuffers / 2) * buffer_size) + (buffer_size * i)); RcvBuffOffset.msg_len = buffer_size; - pr_debug("%s: adding RcvBuffer #%d offset=0x%x sz=%d bufsz:%d\n", + pr_debug("%s: adding RcvBuffer #%d offset=0x%lx sz=%d bufsz:%d\n", sc_adapter[card]->devicename, i + 1, RcvBuffOffset.buff_offset, RcvBuffOffset.msg_len,buffer_size); diff --git a/drivers/isdn/sc/shmem.c b/drivers/isdn/sc/shmem.c index 24854826..6f58862 100644 --- a/drivers/isdn/sc/shmem.c +++ b/drivers/isdn/sc/shmem.c @@ -61,7 +61,7 @@ void memcpy_toshmem(int card, void *dest, const void *src, size_t n) spin_unlock_irqrestore(&sc_adapter[card]->lock, flags); pr_debug("%s: set page to %#x\n",sc_adapter[card]->devicename, ((sc_adapter[card]->shmem_magic + ch * SRAM_PAGESIZE)>>14)|0x80); - pr_debug("%s: copying %d bytes from %#x to %#x\n", + pr_debug("%s: copying %d bytes from %#lx to %#lx\n", sc_adapter[card]->devicename, n, (unsigned long) src, sc_adapter[card]->rambase + ((unsigned long) dest %0x4000)); -- cgit v0.10.2 From 5f6e3c836508926e50cebe17ad87f59666a7fb47 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 11 Oct 2006 01:22:26 -0700 Subject: [PATCH] md: use BUILD_BUG_ON Signed-off-by: Alexey Dobriyan Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 8e67634..d47d38a 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1413,7 +1413,7 @@ int bitmap_create(mddev_t *mddev) int err; sector_t start; - BUG_ON(sizeof(bitmap_super_t) != 256); + BUILD_BUG_ON(sizeof(bitmap_super_t) != 256); if (!file && !mddev->bitmap_offset) /* bitmap disabled, nothing to do */ return 0; -- cgit v0.10.2 From 35d59fc5d6f318a28a99c5936171afd4edef28c8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:22:44 +0100 Subject: [PATCH] arm __user annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c index dedbb44..a657a28 100644 --- a/arch/arm/vfp/vfpmodule.c +++ b/arch/arm/vfp/vfpmodule.c @@ -90,7 +90,7 @@ void vfp_raise_sigfpe(unsigned int sicode, struct pt_regs *regs) info.si_signo = SIGFPE; info.si_code = sicode; - info.si_addr = (void *)(instruction_pointer(regs) - 4); + info.si_addr = (void __user *)(instruction_pointer(regs) - 4); /* * This is the same as NWFPE, because it's not clear what -- cgit v0.10.2 From fc048b5b0f2554bc953a8ada5b2e3b82bde2fcb0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:22:54 +0100 Subject: [PATCH] arm: use unsigned long instead of unsigned int in get_user() Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/include/asm-arm/uaccess.h b/include/asm-arm/uaccess.h index 87aba57..09ad0ca 100644 --- a/include/asm-arm/uaccess.h +++ b/include/asm-arm/uaccess.h @@ -110,7 +110,7 @@ extern int __get_user_4(void *); #define get_user(x,p) \ ({ \ const register typeof(*(p)) __user *__p asm("r0") = (p);\ - register unsigned int __r2 asm("r2"); \ + register unsigned long __r2 asm("r2"); \ register int __e asm("r0"); \ switch (sizeof(*(__p))) { \ case 1: \ -- cgit v0.10.2 From 399ad77b9098ed2eb27cbfbeb6449c1caab3c18e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:22:34 +0100 Subject: [PATCH] arm-versatile iomem annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c index 2aa150b..3b85761 100644 --- a/arch/arm/mach-versatile/core.c +++ b/arch/arm/mach-versatile/core.c @@ -188,12 +188,12 @@ static struct map_desc versatile_io_desc[] __initdata = { .length = SZ_4K, .type = MT_DEVICE }, { - .virtual = VERSATILE_PCI_VIRT_BASE, + .virtual = (unsigned long)VERSATILE_PCI_VIRT_BASE, .pfn = __phys_to_pfn(VERSATILE_PCI_BASE), .length = VERSATILE_PCI_BASE_SIZE, .type = MT_DEVICE }, { - .virtual = VERSATILE_PCI_CFG_VIRT_BASE, + .virtual = (unsigned long)VERSATILE_PCI_CFG_VIRT_BASE, .pfn = __phys_to_pfn(VERSATILE_PCI_CFG_BASE), .length = VERSATILE_PCI_CFG_BASE_SIZE, .type = MT_DEVICE diff --git a/arch/arm/mach-versatile/pci.c b/arch/arm/mach-versatile/pci.c index 13bbd08..5cd0b5d 100644 --- a/arch/arm/mach-versatile/pci.c +++ b/arch/arm/mach-versatile/pci.c @@ -40,14 +40,15 @@ * Cfg 42000000 - 42FFFFFF PCI config * */ -#define SYS_PCICTL IO_ADDRESS(VERSATILE_SYS_PCICTL) -#define PCI_IMAP0 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x0) -#define PCI_IMAP1 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x4) -#define PCI_IMAP2 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x8) -#define PCI_SMAP0 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x10) -#define PCI_SMAP1 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x14) -#define PCI_SMAP2 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x18) -#define PCI_SELFID IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0xc) +#define __IO_ADDRESS(n) ((void __iomem *)(unsigned long)IO_ADDRESS(n)) +#define SYS_PCICTL __IO_ADDRESS(VERSATILE_SYS_PCICTL) +#define PCI_IMAP0 __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x0) +#define PCI_IMAP1 __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x4) +#define PCI_IMAP2 __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x8) +#define PCI_SMAP0 __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x10) +#define PCI_SMAP1 __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x14) +#define PCI_SMAP2 __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x18) +#define PCI_SELFID __IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0xc) #define DEVICE_ID_OFFSET 0x00 #define CSR_OFFSET 0x04 @@ -76,7 +77,7 @@ static int __init versatile_pci_slot_ignore(char *str) __setup("pci_slot_ignore=", versatile_pci_slot_ignore); -static unsigned long __pci_addr(struct pci_bus *bus, +static void __iomem *__pci_addr(struct pci_bus *bus, unsigned int devfn, int offset) { unsigned int busnr = bus->number; @@ -91,14 +92,14 @@ static unsigned long __pci_addr(struct pci_bus *bus, if (devfn > 255) BUG(); - return (VERSATILE_PCI_CFG_VIRT_BASE | (busnr << 16) | + return VERSATILE_PCI_CFG_VIRT_BASE + ((busnr << 16) | (PCI_SLOT(devfn) << 11) | (PCI_FUNC(devfn) << 8) | offset); } static int versatile_read_config(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 *val) { - unsigned long addr = __pci_addr(bus, devfn, where); + void __iomem *addr = __pci_addr(bus, devfn, where & ~3); u32 v; int slot = PCI_SLOT(devfn); @@ -121,13 +122,12 @@ static int versatile_read_config(struct pci_bus *bus, unsigned int devfn, int wh break; case 2: - v = __raw_readl(addr & ~3); - if (addr & 2) v >>= 16; + v = __raw_readl(addr); + if (where & 2) v >>= 16; v &= 0xffff; break; default: - addr &= ~3; v = __raw_readl(addr); break; } @@ -140,7 +140,7 @@ static int versatile_read_config(struct pci_bus *bus, unsigned int devfn, int wh static int versatile_write_config(struct pci_bus *bus, unsigned int devfn, int where, int size, u32 val) { - unsigned long addr = __pci_addr(bus, devfn, where); + void __iomem *addr = __pci_addr(bus, devfn, where); int slot = PCI_SLOT(devfn); if (pci_slot_ignore & (1 << slot)) { @@ -279,7 +279,7 @@ int __init pci_versatile_setup(int nr, struct pci_sys_data *sys) printk("PCI core found (slot %d)\n",myslot); __raw_writel(myslot, PCI_SELFID); - local_pci_cfg_base = (void *) VERSATILE_PCI_CFG_VIRT_BASE + (myslot << 11); + local_pci_cfg_base = VERSATILE_PCI_CFG_VIRT_BASE + (myslot << 11); val = __raw_readl(local_pci_cfg_base + CSR_OFFSET); val |= PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER | PCI_COMMAND_INVALIDATE; diff --git a/include/asm-arm/arch-versatile/hardware.h b/include/asm-arm/arch-versatile/hardware.h index 41c1bee..edc0659 100644 --- a/include/asm-arm/arch-versatile/hardware.h +++ b/include/asm-arm/arch-versatile/hardware.h @@ -28,8 +28,8 @@ /* * PCI space virtual addresses */ -#define VERSATILE_PCI_VIRT_BASE 0xe8000000 -#define VERSATILE_PCI_CFG_VIRT_BASE 0xe9000000 +#define VERSATILE_PCI_VIRT_BASE (void __iomem *)0xe8000000ul +#define VERSATILE_PCI_CFG_VIRT_BASE (void __iomem *)0xe9000000ul #if 0 #define VERSATILE_PCI_VIRT_MEM_BASE0 0xf4000000 -- cgit v0.10.2 From 7857a7f59d5d0b4e8c15606a79c0bcdbdef18ab6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:24:25 +0100 Subject: [PATCH] m32r: C99 initializers in setup.c Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m32r/kernel/setup.c b/arch/m32r/kernel/setup.c index 3f35ab3..0e7778b 100644 --- a/arch/m32r/kernel/setup.c +++ b/arch/m32r/kernel/setup.c @@ -369,10 +369,10 @@ static void c_stop(struct seq_file *m, void *v) } struct seq_operations cpuinfo_op = { - start: c_start, - next: c_next, - stop: c_stop, - show: show_cpuinfo, + .start = c_start, + .next = c_next, + .stop = c_stop, + .show = show_cpuinfo, }; #endif /* CONFIG_PROC_FS */ -- cgit v0.10.2 From 870e75a2930a1db02c7a5c09a13edcb4e3b07838 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:24:45 +0100 Subject: [PATCH] m32r: signal __user annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c index a9174ef..b60cea4 100644 --- a/arch/m32r/kernel/signal.c +++ b/arch/m32r/kernel/signal.c @@ -33,7 +33,7 @@ int do_signal(struct pt_regs *, sigset_t *); asmlinkage int -sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize, +sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize, unsigned long r2, unsigned long r3, unsigned long r4, unsigned long r5, unsigned long r6, struct pt_regs *regs) { @@ -78,8 +78,8 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss, struct rt_sigframe { int sig; - struct siginfo *pinfo; - void *puc; + struct siginfo __user *pinfo; + void __user *puc; struct siginfo info; struct ucontext uc; // struct _fpstate fpstate; -- cgit v0.10.2 From 12ea59e8109d7192ecb2ac994588e24b11ab0428 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:24:35 +0100 Subject: [PATCH] m32r: NULL noise removal Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m32r/kernel/setup_mappi.c b/arch/m32r/kernel/setup_mappi.c index 67dbbdc..6b2d77d 100644 --- a/arch/m32r/kernel/setup_mappi.c +++ b/arch/m32r/kernel/setup_mappi.c @@ -86,7 +86,7 @@ void __init init_IRQ(void) /* INT0 : LAN controller (RTL8019AS) */ irq_desc[M32R_IRQ_INT0].status = IRQ_DISABLED; irq_desc[M32R_IRQ_INT0].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_INT0].action = 0; + irq_desc[M32R_IRQ_INT0].action = NULL; irq_desc[M32R_IRQ_INT0].depth = 1; icu_data[M32R_IRQ_INT0].icucr = M32R_ICUCR_IEN|M32R_ICUCR_ISMOD10; disable_mappi_irq(M32R_IRQ_INT0); @@ -95,7 +95,7 @@ void __init init_IRQ(void) /* MFT2 : system timer */ irq_desc[M32R_IRQ_MFT2].status = IRQ_DISABLED; irq_desc[M32R_IRQ_MFT2].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_MFT2].action = 0; + irq_desc[M32R_IRQ_MFT2].action = NULL; irq_desc[M32R_IRQ_MFT2].depth = 1; icu_data[M32R_IRQ_MFT2].icucr = M32R_ICUCR_IEN; disable_mappi_irq(M32R_IRQ_MFT2); @@ -104,7 +104,7 @@ void __init init_IRQ(void) /* SIO0_R : uart receive data */ irq_desc[M32R_IRQ_SIO0_R].status = IRQ_DISABLED; irq_desc[M32R_IRQ_SIO0_R].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_SIO0_R].action = 0; + irq_desc[M32R_IRQ_SIO0_R].action = NULL; irq_desc[M32R_IRQ_SIO0_R].depth = 1; icu_data[M32R_IRQ_SIO0_R].icucr = 0; disable_mappi_irq(M32R_IRQ_SIO0_R); @@ -112,7 +112,7 @@ void __init init_IRQ(void) /* SIO0_S : uart send data */ irq_desc[M32R_IRQ_SIO0_S].status = IRQ_DISABLED; irq_desc[M32R_IRQ_SIO0_S].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_SIO0_S].action = 0; + irq_desc[M32R_IRQ_SIO0_S].action = NULL; irq_desc[M32R_IRQ_SIO0_S].depth = 1; icu_data[M32R_IRQ_SIO0_S].icucr = 0; disable_mappi_irq(M32R_IRQ_SIO0_S); @@ -120,7 +120,7 @@ void __init init_IRQ(void) /* SIO1_R : uart receive data */ irq_desc[M32R_IRQ_SIO1_R].status = IRQ_DISABLED; irq_desc[M32R_IRQ_SIO1_R].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_SIO1_R].action = 0; + irq_desc[M32R_IRQ_SIO1_R].action = NULL; irq_desc[M32R_IRQ_SIO1_R].depth = 1; icu_data[M32R_IRQ_SIO1_R].icucr = 0; disable_mappi_irq(M32R_IRQ_SIO1_R); @@ -128,7 +128,7 @@ void __init init_IRQ(void) /* SIO1_S : uart send data */ irq_desc[M32R_IRQ_SIO1_S].status = IRQ_DISABLED; irq_desc[M32R_IRQ_SIO1_S].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_SIO1_S].action = 0; + irq_desc[M32R_IRQ_SIO1_S].action = NULL; irq_desc[M32R_IRQ_SIO1_S].depth = 1; icu_data[M32R_IRQ_SIO1_S].icucr = 0; disable_mappi_irq(M32R_IRQ_SIO1_S); @@ -138,7 +138,7 @@ void __init init_IRQ(void) /* INT1 : pccard0 interrupt */ irq_desc[M32R_IRQ_INT1].status = IRQ_DISABLED; irq_desc[M32R_IRQ_INT1].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_INT1].action = 0; + irq_desc[M32R_IRQ_INT1].action = NULL; irq_desc[M32R_IRQ_INT1].depth = 1; icu_data[M32R_IRQ_INT1].icucr = M32R_ICUCR_IEN | M32R_ICUCR_ISMOD00; disable_mappi_irq(M32R_IRQ_INT1); @@ -146,7 +146,7 @@ void __init init_IRQ(void) /* INT2 : pccard1 interrupt */ irq_desc[M32R_IRQ_INT2].status = IRQ_DISABLED; irq_desc[M32R_IRQ_INT2].chip = &mappi_irq_type; - irq_desc[M32R_IRQ_INT2].action = 0; + irq_desc[M32R_IRQ_INT2].action = NULL; irq_desc[M32R_IRQ_INT2].depth = 1; icu_data[M32R_IRQ_INT2].icucr = M32R_ICUCR_IEN | M32R_ICUCR_ISMOD00; disable_mappi_irq(M32R_IRQ_INT2); diff --git a/arch/m32r/kernel/smp.c b/arch/m32r/kernel/smp.c index 722e21f..3601291 100644 --- a/arch/m32r/kernel/smp.c +++ b/arch/m32r/kernel/smp.c @@ -231,7 +231,7 @@ void smp_flush_tlb_all(void) local_irq_save(flags); __flush_tlb_all(); local_irq_restore(flags); - smp_call_function(flush_tlb_all_ipi, 0, 1, 1); + smp_call_function(flush_tlb_all_ipi, NULL, 1, 1); preempt_enable(); } diff --git a/arch/m32r/kernel/traps.c b/arch/m32r/kernel/traps.c index c1daf2c..97e0b1c 100644 --- a/arch/m32r/kernel/traps.c +++ b/arch/m32r/kernel/traps.c @@ -268,7 +268,7 @@ static __inline__ void do_trap(int trapnr, int signr, const char * str, #define DO_ERROR(trapnr, signr, str, name) \ asmlinkage void do_##name(struct pt_regs * regs, long error_code) \ { \ - do_trap(trapnr, signr, 0, regs, error_code, NULL); \ + do_trap(trapnr, signr, NULL, regs, error_code, NULL); \ } #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \ -- cgit v0.10.2 From fd2c903b89a6c3cec9388eb24a134ebc1be20747 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:24:55 +0100 Subject: [PATCH] m32r: more __user annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m32r/kernel/sys_m32r.c b/arch/m32r/kernel/sys_m32r.c index b567351..b4e7bcb 100644 --- a/arch/m32r/kernel/sys_m32r.c +++ b/arch/m32r/kernel/sys_m32r.c @@ -31,7 +31,7 @@ /* * sys_tas() - test-and-set */ -asmlinkage int sys_tas(int *addr) +asmlinkage int sys_tas(int __user *addr) { int oldval; @@ -90,7 +90,7 @@ sys_pipe(unsigned long r0, unsigned long r1, unsigned long r2, error = do_pipe(fd); if (!error) { - if (copy_to_user((void *)r0, (void *)fd, 2*sizeof(int))) + if (copy_to_user((void __user *)r0, fd, 2*sizeof(int))) error = -EFAULT; } return error; @@ -201,7 +201,7 @@ asmlinkage int sys_ipc(uint call, int first, int second, } } -asmlinkage int sys_uname(struct old_utsname * name) +asmlinkage int sys_uname(struct old_utsname __user * name) { int err; if (!name) -- cgit v0.10.2 From 4b4fcaa1a9eec90b44b66a67af6e130349ba008e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:25:45 +0100 Subject: [PATCH] misuse of strstr Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c index 1f94dd3..cdd1694 100644 --- a/fs/gfs2/locking/dlm/mount.c +++ b/fs/gfs2/locking/dlm/mount.c @@ -45,7 +45,7 @@ static struct gdlm_ls *init_gdlm(lm_callback_t cb, struct gfs2_sbd *sdp, strncpy(buf, table_name, 256); buf[255] = '\0'; - p = strstr(buf, ":"); + p = strchr(buf, ':'); if (!p) { log_info("invalid table_name \"%s\"", table_name); kfree(ls); -- cgit v0.10.2 From b971018bae94bb43ae2402f884684ad69e85f931 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:27:57 +0100 Subject: [PATCH] m68k uaccess __user annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/include/asm-m68k/uaccess.h b/include/asm-m68k/uaccess.h index 88b1f47..e4c9f08 100644 --- a/include/asm-m68k/uaccess.h +++ b/include/asm-m68k/uaccess.h @@ -76,7 +76,7 @@ asm volatile ("\n" \ break; \ case 8: \ { \ - const void *__pu_ptr = (ptr); \ + const void __user *__pu_ptr = (ptr); \ asm volatile ("\n" \ "1: moves.l %2,(%1)+\n" \ "2: moves.l %R2,(%1)\n" \ @@ -125,7 +125,7 @@ asm volatile ("\n" \ " .previous" \ : "+d" (res), "=&" #reg (__gu_val) \ : "m" (*(ptr)), "i" (err)); \ - (x) = (typeof(*(ptr)))(long)__gu_val; \ + (x) = (typeof(*(ptr)))(unsigned long)__gu_val; \ }) #define __get_user(x, ptr) \ @@ -221,16 +221,16 @@ __constant_copy_from_user(void *to, const void __user *from, unsigned long n) switch (n) { case 1: - __get_user_asm(res, *(u8 *)to, (u8 *)from, u8, b, d, 1); + __get_user_asm(res, *(u8 *)to, (u8 __user *)from, u8, b, d, 1); break; case 2: - __get_user_asm(res, *(u16 *)to, (u16 *)from, u16, w, d, 2); + __get_user_asm(res, *(u16 *)to, (u16 __user *)from, u16, w, d, 2); break; case 3: __constant_copy_from_user_asm(res, to, from, tmp, 3, w, b,); break; case 4: - __get_user_asm(res, *(u32 *)to, (u32 *)from, u32, l, r, 4); + __get_user_asm(res, *(u32 *)to, (u32 __user *)from, u32, l, r, 4); break; case 5: __constant_copy_from_user_asm(res, to, from, tmp, 5, l, b,); @@ -302,16 +302,16 @@ __constant_copy_to_user(void __user *to, const void *from, unsigned long n) switch (n) { case 1: - __put_user_asm(res, *(u8 *)from, (u8 *)to, b, d, 1); + __put_user_asm(res, *(u8 *)from, (u8 __user *)to, b, d, 1); break; case 2: - __put_user_asm(res, *(u16 *)from, (u16 *)to, w, d, 2); + __put_user_asm(res, *(u16 *)from, (u16 __user *)to, w, d, 2); break; case 3: __constant_copy_to_user_asm(res, to, from, tmp, 3, w, b,); break; case 4: - __put_user_asm(res, *(u32 *)from, (u32 *)to, l, r, 4); + __put_user_asm(res, *(u32 *)from, (u32 __user *)to, l, r, 4); break; case 5: __constant_copy_to_user_asm(res, to, from, tmp, 5, l, b,); -- cgit v0.10.2 From 1e5c374d3833f816b4840227c6949f689af0cb44 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:28:07 +0100 Subject: [PATCH] misc m68k __user annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c index 45a4664..24e83d5 100644 --- a/arch/m68k/kernel/process.c +++ b/arch/m68k/kernel/process.c @@ -221,13 +221,13 @@ asmlinkage int m68k_clone(struct pt_regs *regs) { unsigned long clone_flags; unsigned long newsp; - int *parent_tidptr, *child_tidptr; + int __user *parent_tidptr, *child_tidptr; /* syscall2 puts clone_flags in d1 and usp in d2 */ clone_flags = regs->d1; newsp = regs->d2; - parent_tidptr = (int *)regs->d3; - child_tidptr = (int *)regs->d4; + parent_tidptr = (int __user *)regs->d3; + child_tidptr = (int __user *)regs->d4; if (!newsp) newsp = rdusp(); return do_fork(clone_flags, newsp, regs, 0, @@ -361,7 +361,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump) /* * sys_execve() executes a new program. */ -asmlinkage int sys_execve(char *name, char **argv, char **envp) +asmlinkage int sys_execve(char __user *name, char __user * __user *argv, char __user * __user *envp) { int error; char * filename; diff --git a/arch/m68k/kernel/traps.c b/arch/m68k/kernel/traps.c index 4569406..759fa244 100644 --- a/arch/m68k/kernel/traps.c +++ b/arch/m68k/kernel/traps.c @@ -326,13 +326,13 @@ static inline int do_040writeback1(unsigned short wbs, unsigned long wba, switch (wbs & WBSIZ_040) { case BA_SIZE_BYTE: - res = put_user(wbd & 0xff, (char *)wba); + res = put_user(wbd & 0xff, (char __user *)wba); break; case BA_SIZE_WORD: - res = put_user(wbd & 0xffff, (short *)wba); + res = put_user(wbd & 0xffff, (short __user *)wba); break; case BA_SIZE_LONG: - res = put_user(wbd, (int *)wba); + res = put_user(wbd, (int __user *)wba); break; } -- cgit v0.10.2 From 437111ca381263520d23c877e55e0a83558e79da Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:28:17 +0100 Subject: [PATCH] sun3 __iomem annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m68k/mm/sun3kmap.c b/arch/m68k/mm/sun3kmap.c index 7f0d86f..8caa459 100644 --- a/arch/m68k/mm/sun3kmap.c +++ b/arch/m68k/mm/sun3kmap.c @@ -59,7 +59,7 @@ static inline void do_pmeg_mapin(unsigned long phys, unsigned long virt, } } -void *sun3_ioremap(unsigned long phys, unsigned long size, +void __iomem *sun3_ioremap(unsigned long phys, unsigned long size, unsigned long type) { struct vm_struct *area; @@ -101,19 +101,19 @@ void *sun3_ioremap(unsigned long phys, unsigned long size, virt += seg_pages * PAGE_SIZE; } - return (void *)ret; + return (void __iomem *)ret; } -void *__ioremap(unsigned long phys, unsigned long size, int cache) +void __iomem *__ioremap(unsigned long phys, unsigned long size, int cache) { return sun3_ioremap(phys, size, SUN3_PAGE_TYPE_IO); } -void iounmap(void *addr) +void iounmap(void __iomem *addr) { vfree((void *)(PAGE_MASK & (unsigned long)addr)); } diff --git a/drivers/net/sun3_82586.c b/drivers/net/sun3_82586.c index d1d1885..a3220a9 100644 --- a/drivers/net/sun3_82586.c +++ b/drivers/net/sun3_82586.c @@ -330,7 +330,7 @@ out2: out1: free_netdev(dev); out: - iounmap((void *)ioaddr); + iounmap((void __iomem *)ioaddr); return ERR_PTR(err); } diff --git a/drivers/net/sun3lance.c b/drivers/net/sun3lance.c index 91c7654..b865db3 100644 --- a/drivers/net/sun3lance.c +++ b/drivers/net/sun3lance.c @@ -286,7 +286,7 @@ struct net_device * __init sun3lance_probe(int unit) out1: #ifdef CONFIG_SUN3 - iounmap((void *)dev->base_addr); + iounmap((void __iomem *)dev->base_addr); #endif out: free_netdev(dev); @@ -326,7 +326,7 @@ static int __init lance_probe( struct net_device *dev) ioaddr_probe[1] = tmp2; #ifdef CONFIG_SUN3 - iounmap((void *)ioaddr); + iounmap((void __iomem *)ioaddr); #endif return 0; } @@ -956,7 +956,7 @@ void cleanup_module(void) { unregister_netdev(sun3lance_dev); #ifdef CONFIG_SUN3 - iounmap((void *)sun3lance_dev->base_addr); + iounmap((void __iomem *)sun3lance_dev->base_addr); #endif free_netdev(sun3lance_dev); } -- cgit v0.10.2 From 2e811488cedddefb9d1df97c260b6048ea8ef835 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:28:27 +0100 Subject: [PATCH] clean m68k ksyms sun3_ksyms gone, m68k_ksyms trimmed down to exports of the assembler ones, for sun3 added the missing exports of __ioremap() and iounmap(). Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/m68k/kernel/m68k_ksyms.c b/arch/m68k/kernel/m68k_ksyms.c index f9636e8..6fc69c7 100644 --- a/arch/m68k/kernel/m68k_ksyms.c +++ b/arch/m68k/kernel/m68k_ksyms.c @@ -1,61 +1,10 @@ #include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include #include -#include asmlinkage long long __ashldi3 (long long, int); asmlinkage long long __ashrdi3 (long long, int); asmlinkage long long __lshrdi3 (long long, int); asmlinkage long long __muldi3 (long long, long long); -extern char m68k_debug_device[]; - -/* platform dependent support */ - -EXPORT_SYMBOL(m68k_machtype); -EXPORT_SYMBOL(m68k_cputype); -EXPORT_SYMBOL(m68k_is040or060); -EXPORT_SYMBOL(m68k_realnum_memory); -EXPORT_SYMBOL(m68k_memory); -#ifndef CONFIG_SUN3 -EXPORT_SYMBOL(cache_push); -EXPORT_SYMBOL(cache_clear); -#ifndef CONFIG_SINGLE_MEMORY_CHUNK -EXPORT_SYMBOL(mm_vtop); -EXPORT_SYMBOL(mm_ptov); -EXPORT_SYMBOL(mm_end_of_chunk); -#else -EXPORT_SYMBOL(m68k_memoffset); -#endif /* !CONFIG_SINGLE_MEMORY_CHUNK */ -EXPORT_SYMBOL(__ioremap); -EXPORT_SYMBOL(iounmap); -EXPORT_SYMBOL(kernel_set_cachemode); -#endif /* !CONFIG_SUN3 */ -EXPORT_SYMBOL(m68k_debug_device); -EXPORT_SYMBOL(mach_hwclk); -EXPORT_SYMBOL(mach_get_ss); -EXPORT_SYMBOL(mach_get_rtc_pll); -EXPORT_SYMBOL(mach_set_rtc_pll); -#ifdef CONFIG_INPUT_M68K_BEEP_MODULE -EXPORT_SYMBOL(mach_beep); -#endif -EXPORT_SYMBOL(dump_fpu); -EXPORT_SYMBOL(dump_thread); -EXPORT_SYMBOL(kernel_thread); -#ifdef CONFIG_VME -EXPORT_SYMBOL(vme_brdtype); -#endif /* The following are special because they're not called explicitly (the C compiler generates them). Fortunately, diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c index 24e83d5..99fc122 100644 --- a/arch/m68k/kernel/process.c +++ b/arch/m68k/kernel/process.c @@ -187,6 +187,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) set_fs (fs); return pid; } +EXPORT_SYMBOL(kernel_thread); void flush_thread(void) { @@ -311,6 +312,7 @@ int dump_fpu (struct pt_regs *regs, struct user_m68kfp_struct *fpu) : "memory"); return 1; } +EXPORT_SYMBOL(dump_fpu); /* * fill in the user structure for a core dump.. @@ -357,6 +359,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump) /* dump floating point stuff */ dump->u_fpvalid = dump_fpu (regs, &dump->m68kfp); } +EXPORT_SYMBOL(dump_thread); /* * sys_execve() executes a new program. diff --git a/arch/m68k/kernel/setup.c b/arch/m68k/kernel/setup.c index 42d5b85..9af3ee0 100644 --- a/arch/m68k/kernel/setup.c +++ b/arch/m68k/kernel/setup.c @@ -42,27 +42,37 @@ unsigned long m68k_machtype; unsigned long m68k_cputype; +EXPORT_SYMBOL(m68k_machtype); +EXPORT_SYMBOL(m68k_cputype); unsigned long m68k_fputype; unsigned long m68k_mmutype; #ifdef CONFIG_VME unsigned long vme_brdtype; +EXPORT_SYMBOL(vme_brdtype); #endif int m68k_is040or060; +EXPORT_SYMBOL(m68k_is040or060); extern int end; extern unsigned long availmem; int m68k_num_memory; int m68k_realnum_memory; +EXPORT_SYMBOL(m68k_realnum_memory); +#ifdef CONFIG_SINGLE_MEMORY_CHUNK unsigned long m68k_memoffset; +EXPORT_SYMBOL(m68k_memoffset); +#endif struct mem_info m68k_memory[NUM_MEMINFO]; +EXPORT_SYMBOL(m68k_memory); static struct mem_info m68k_ramdisk; static char m68k_command_line[CL_SIZE]; char m68k_debug_device[6] = ""; +EXPORT_SYMBOL(m68k_debug_device); void (*mach_sched_init) (irq_handler_t handler) __initdata = NULL; /* machine dependent irq functions */ @@ -72,10 +82,14 @@ int (*mach_get_hardware_list) (char *buffer); /* machine dependent timer functions */ unsigned long (*mach_gettimeoffset) (void); int (*mach_hwclk) (int, struct rtc_time*); +EXPORT_SYMBOL(mach_hwclk); int (*mach_set_clock_mmss) (unsigned long); unsigned int (*mach_get_ss)(void); int (*mach_get_rtc_pll)(struct rtc_pll_info *); int (*mach_set_rtc_pll)(struct rtc_pll_info *); +EXPORT_SYMBOL(mach_get_ss); +EXPORT_SYMBOL(mach_get_rtc_pll); +EXPORT_SYMBOL(mach_set_rtc_pll); void (*mach_reset)( void ); void (*mach_halt)( void ); void (*mach_power_off)( void ); @@ -89,6 +103,7 @@ void (*mach_l2_flush) (int); #endif #if defined(CONFIG_INPUT_M68K_BEEP) || defined(CONFIG_INPUT_M68K_BEEP_MODULE) void (*mach_beep)(unsigned int, unsigned int); +EXPORT_SYMBOL(mach_beep); #endif #if defined(CONFIG_ISA) && defined(MULTI_ISA) int isa_type; diff --git a/arch/m68k/mm/kmap.c b/arch/m68k/mm/kmap.c index f46f049..b54ef17 100644 --- a/arch/m68k/mm/kmap.c +++ b/arch/m68k/mm/kmap.c @@ -7,6 +7,7 @@ * used by other architectures /Roman Zippel */ +#include #include #include #include @@ -219,6 +220,7 @@ void __iomem *__ioremap(unsigned long physaddr, unsigned long size, int cachefla return (void __iomem *)retaddr; } +EXPORT_SYMBOL(__ioremap); /* * Unmap a ioremap()ed region again @@ -234,6 +236,7 @@ void iounmap(void __iomem *addr) free_io_area((__force void *)addr); #endif } +EXPORT_SYMBOL(iounmap); /* * __iounmap unmaps nearly everything, so be careful @@ -360,3 +363,4 @@ void kernel_set_cachemode(void *addr, unsigned long size, int cmode) flush_tlb_all(); } +EXPORT_SYMBOL(kernel_set_cachemode); diff --git a/arch/m68k/mm/memory.c b/arch/m68k/mm/memory.c index a0c095e..0f88812 100644 --- a/arch/m68k/mm/memory.c +++ b/arch/m68k/mm/memory.c @@ -4,6 +4,7 @@ * Copyright (C) 1995 Hamish Macdonald */ +#include #include #include #include @@ -157,9 +158,8 @@ unsigned long mm_vtop(unsigned long vaddr) return -1; } -#endif +EXPORT_SYMBOL(mm_vtop); -#ifndef CONFIG_SINGLE_MEMORY_CHUNK unsigned long mm_ptov (unsigned long paddr) { int i = 0; @@ -185,6 +185,7 @@ unsigned long mm_ptov (unsigned long paddr) #endif return -1; } +EXPORT_SYMBOL(mm_ptov); #endif /* invalidate page in both caches */ @@ -298,6 +299,7 @@ void cache_clear (unsigned long paddr, int len) mach_l2_flush(0); #endif } +EXPORT_SYMBOL(cache_clear); /* probably can be unexported */ /* @@ -350,6 +352,7 @@ void cache_push (unsigned long paddr, int len) mach_l2_flush(1); #endif } +EXPORT_SYMBOL(cache_push); /* probably can be unexported */ #ifndef CONFIG_SINGLE_MEMORY_CHUNK int mm_end_of_chunk (unsigned long addr, int len) @@ -361,4 +364,5 @@ int mm_end_of_chunk (unsigned long addr, int len) return 1; return 0; } +EXPORT_SYMBOL(mm_end_of_chunk); #endif diff --git a/arch/m68k/mm/sun3kmap.c b/arch/m68k/mm/sun3kmap.c index 8caa459..1af24cb 100644 --- a/arch/m68k/mm/sun3kmap.c +++ b/arch/m68k/mm/sun3kmap.c @@ -8,6 +8,7 @@ * for more details. */ +#include #include #include #include @@ -112,11 +113,13 @@ void __iomem *__ioremap(unsigned long phys, unsigned long size, int cache) return sun3_ioremap(phys, size, SUN3_PAGE_TYPE_IO); } +EXPORT_SYMBOL(__ioremap); void iounmap(void __iomem *addr) { vfree((void *)(PAGE_MASK & (unsigned long)addr)); } +EXPORT_SYMBOL(iounmap); /* sun3_map_test(addr, val) -- Reads a byte from addr, storing to val, * trapping the potential read fault. Returns 0 if the access faulted, diff --git a/arch/m68k/sun3/Makefile b/arch/m68k/sun3/Makefile index 4d4f069..be1a847 100644 --- a/arch/m68k/sun3/Makefile +++ b/arch/m68k/sun3/Makefile @@ -2,6 +2,6 @@ # Makefile for Linux arch/m68k/sun3 source directory # -obj-y := sun3_ksyms.o sun3ints.o sun3dvma.o sbus.o idprom.o +obj-y := sun3ints.o sun3dvma.o sbus.o idprom.o obj-$(CONFIG_SUN3) += config.o mmu_emu.o leds.o dvma.o intersil.o diff --git a/arch/m68k/sun3/idprom.c b/arch/m68k/sun3/idprom.c index 02c1fee6..dca6ab6 100644 --- a/arch/m68k/sun3/idprom.c +++ b/arch/m68k/sun3/idprom.c @@ -6,6 +6,7 @@ * Sun3/3x models added by David Monro (davidm@psrg.cs.usyd.edu.au) */ +#include #include #include #include @@ -16,6 +17,8 @@ #include /* Fun with Sun released architectures. */ struct idprom *idprom; +EXPORT_SYMBOL(idprom); + static struct idprom idprom_buffer; /* Here is the master table of Sun machines which use some implementation diff --git a/arch/m68k/sun3/sun3_ksyms.c b/arch/m68k/sun3/sun3_ksyms.c deleted file mode 100644 index 43e5a9a..0000000 --- a/arch/m68k/sun3/sun3_ksyms.c +++ /dev/null @@ -1,13 +0,0 @@ -#include -#include -#include -#include - -/* - * Add things here when you find the need for it. - */ -EXPORT_SYMBOL(dvma_map_align); -EXPORT_SYMBOL(dvma_unmap); -EXPORT_SYMBOL(dvma_malloc_align); -EXPORT_SYMBOL(dvma_free); -EXPORT_SYMBOL(idprom); diff --git a/arch/m68k/sun3/sun3dvma.c b/arch/m68k/sun3/sun3dvma.c index a2bc2da..8709677 100644 --- a/arch/m68k/sun3/sun3dvma.c +++ b/arch/m68k/sun3/sun3dvma.c @@ -6,6 +6,7 @@ * Contains common routines for sun3/sun3x DVMA management. */ +#include #include #include #include @@ -312,6 +313,7 @@ inline unsigned long dvma_map_align(unsigned long kaddr, int len, int align) BUG(); return 0; } +EXPORT_SYMBOL(dvma_map_align); void dvma_unmap(void *baddr) { @@ -327,7 +329,7 @@ void dvma_unmap(void *baddr) return; } - +EXPORT_SYMBOL(dvma_unmap); void *dvma_malloc_align(unsigned long len, unsigned long align) { @@ -367,6 +369,7 @@ void *dvma_malloc_align(unsigned long len, unsigned long align) return (void *)vaddr; } +EXPORT_SYMBOL(dvma_malloc_align); void dvma_free(void *vaddr) { @@ -374,3 +377,4 @@ void dvma_free(void *vaddr) return; } +EXPORT_SYMBOL(dvma_free); -- cgit v0.10.2 From 2db5f59ca74d911f93c39494db1581c3c93d5a29 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:28:37 +0100 Subject: [PATCH] amiga_floppy_init() in non-modular case It used to be called directly, but that got lost in 2.1.87-pre1. Similar breakage in ataflop got fixed 3 years ago, this one had gone unnoticed. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/block/amiflop.c b/drivers/block/amiflop.c index 5d254b7..5d65621 100644 --- a/drivers/block/amiflop.c +++ b/drivers/block/amiflop.c @@ -1709,10 +1709,13 @@ static struct kobject *floppy_find(dev_t dev, int *part, void *data) return get_disk(unit[drive].gendisk); } -int __init amiga_floppy_init(void) +static int __init amiga_floppy_init(void) { int i, ret; + if (!MACH_IS_AMIGA) + return -ENXIO; + if (!AMIGAHW_PRESENT(AMI_FLOPPY)) return -ENXIO; @@ -1809,15 +1812,9 @@ out_blkdev: return ret; } +module_init(amiga_floppy_init); #ifdef MODULE -int init_module(void) -{ - if (!MACH_IS_AMIGA) - return -ENXIO; - return amiga_floppy_init(); -} - #if 0 /* not safe to unload */ void cleanup_module(void) { -- cgit v0.10.2 From f39d88adc5daf544cf9ae666a097b595b212871e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:28:47 +0100 Subject: [PATCH] z2_init() in non-modular case ... another victim - this time of 2.5.1-pre2 Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/block/z2ram.c b/drivers/block/z2ram.c index 82ddbdd..7cc2685 100644 --- a/drivers/block/z2ram.c +++ b/drivers/block/z2ram.c @@ -329,7 +329,7 @@ static struct kobject *z2_find(dev_t dev, int *part, void *data) static struct request_queue *z2_queue; -int __init +static int __init z2_init(void) { int ret; @@ -370,26 +370,7 @@ err: return ret; } -#if defined(MODULE) - -MODULE_LICENSE("GPL"); - -int -init_module( void ) -{ - int error; - - error = z2_init(); - if ( error == 0 ) - { - printk( KERN_INFO DEVICE_NAME ": loaded as module\n" ); - } - - return error; -} - -void -cleanup_module( void ) +static void __exit z2_exit(void) { int i, j; blk_unregister_region(MKDEV(Z2RAM_MAJOR, 0), 256); @@ -425,4 +406,7 @@ cleanup_module( void ) return; } -#endif + +module_init(z2_init); +module_exit(z2_exit); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 9ab6a45394715918b025f2d3b799477295af4cc0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:40:22 +0100 Subject: [PATCH] remove bogus arch-specific syscall exports Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c index da69e66..4779f47 100644 --- a/arch/arm/kernel/armksyms.c +++ b/arch/arm/kernel/armksyms.c @@ -178,9 +178,3 @@ EXPORT_SYMBOL(_find_next_zero_bit_be); EXPORT_SYMBOL(_find_first_bit_be); EXPORT_SYMBOL(_find_next_bit_be); #endif - - /* syscalls */ -EXPORT_SYMBOL(sys_write); -EXPORT_SYMBOL(sys_lseek); -EXPORT_SYMBOL(sys_exit); -EXPORT_SYMBOL(sys_wait4); diff --git a/arch/arm26/kernel/armksyms.c b/arch/arm26/kernel/armksyms.c index 07907b6..93293d0 100644 --- a/arch/arm26/kernel/armksyms.c +++ b/arch/arm26/kernel/armksyms.c @@ -202,14 +202,6 @@ EXPORT_SYMBOL(_find_next_zero_bit_le); EXPORT_SYMBOL(elf_platform); EXPORT_SYMBOL(elf_hwcap); - /* syscalls */ -EXPORT_SYMBOL(sys_write); -EXPORT_SYMBOL(sys_read); -EXPORT_SYMBOL(sys_lseek); -EXPORT_SYMBOL(sys_open); -EXPORT_SYMBOL(sys_exit); -EXPORT_SYMBOL(sys_wait4); - #ifdef CONFIG_PREEMPT EXPORT_SYMBOL(kernel_flag); #endif diff --git a/arch/parisc/kernel/parisc_ksyms.c b/arch/parisc/kernel/parisc_ksyms.c index 6d57553..8f6a0b3 100644 --- a/arch/parisc/kernel/parisc_ksyms.c +++ b/arch/parisc/kernel/parisc_ksyms.c @@ -69,10 +69,6 @@ EXPORT_SYMBOL(memcpy_toio); EXPORT_SYMBOL(memcpy_fromio); EXPORT_SYMBOL(memset_io); -#include -EXPORT_SYMBOL(sys_lseek); -EXPORT_SYMBOL(sys_write); - #include EXPORT_SYMBOL(__up); EXPORT_SYMBOL(__down_interruptible); diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c index 9f19e83..90b5ef5 100644 --- a/arch/s390/kernel/s390_ksyms.c +++ b/arch/s390/kernel/s390_ksyms.c @@ -51,4 +51,3 @@ EXPORT_SYMBOL(csum_fold); EXPORT_SYMBOL(console_mode); EXPORT_SYMBOL(console_devno); EXPORT_SYMBOL(console_irq); -EXPORT_SYMBOL(sys_wait4); -- cgit v0.10.2 From cff52daffa080eff6353f44df418b080dacefb96 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:40:22 +0100 Subject: [PATCH] alpha_ksyms.c cleanup taken exports to actual definitions of symbols being exported. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/alpha/kernel/alpha_ksyms.c b/arch/alpha/kernel/alpha_ksyms.c index 8b02420..692809e 100644 --- a/arch/alpha/kernel/alpha_ksyms.c +++ b/arch/alpha/kernel/alpha_ksyms.c @@ -6,41 +6,14 @@ */ #include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include #include -#include #include -#include #include -#include #include -#include #include -#include -#include -#include -#include -#include #include -extern struct hwrpb_struct *hwrpb; -extern spinlock_t rtc_lock; - /* these are C runtime functions with special calling conventions: */ extern void __divl (void); extern void __reml (void); @@ -52,14 +25,9 @@ extern void __divqu (void); extern void __remqu (void); EXPORT_SYMBOL(alpha_mv); -EXPORT_SYMBOL(screen_info); -EXPORT_SYMBOL(perf_irq); EXPORT_SYMBOL(callback_getenv); EXPORT_SYMBOL(callback_setenv); EXPORT_SYMBOL(callback_save_env); -#ifdef CONFIG_ALPHA_GENERIC -EXPORT_SYMBOL(alpha_using_srm); -#endif /* CONFIG_ALPHA_GENERIC */ /* platform dependent support */ EXPORT_SYMBOL(strcat); @@ -77,47 +45,14 @@ EXPORT_SYMBOL(__constant_c_memset); EXPORT_SYMBOL(copy_page); EXPORT_SYMBOL(clear_page); -EXPORT_SYMBOL(__direct_map_base); -EXPORT_SYMBOL(__direct_map_size); - -#ifdef CONFIG_PCI -EXPORT_SYMBOL(pci_alloc_consistent); -EXPORT_SYMBOL(pci_free_consistent); -EXPORT_SYMBOL(pci_map_single); -EXPORT_SYMBOL(pci_map_page); -EXPORT_SYMBOL(pci_unmap_single); -EXPORT_SYMBOL(pci_unmap_page); -EXPORT_SYMBOL(pci_map_sg); -EXPORT_SYMBOL(pci_unmap_sg); -EXPORT_SYMBOL(pci_dma_supported); -EXPORT_SYMBOL(pci_dac_dma_supported); -EXPORT_SYMBOL(pci_dac_page_to_dma); -EXPORT_SYMBOL(pci_dac_dma_to_page); -EXPORT_SYMBOL(pci_dac_dma_to_offset); -EXPORT_SYMBOL(alpha_gendev_to_pci); -#endif -EXPORT_SYMBOL(dma_set_mask); - -EXPORT_SYMBOL(dump_thread); -EXPORT_SYMBOL(dump_elf_thread); -EXPORT_SYMBOL(dump_elf_task); -EXPORT_SYMBOL(dump_elf_task_fp); -EXPORT_SYMBOL(hwrpb); -EXPORT_SYMBOL(start_thread); EXPORT_SYMBOL(alpha_read_fp_reg); EXPORT_SYMBOL(alpha_read_fp_reg_s); EXPORT_SYMBOL(alpha_write_fp_reg); EXPORT_SYMBOL(alpha_write_fp_reg_s); -/* In-kernel system calls. */ +/* entry.S */ EXPORT_SYMBOL(kernel_thread); -EXPORT_SYMBOL(sys_dup); -EXPORT_SYMBOL(sys_exit); -EXPORT_SYMBOL(sys_write); -EXPORT_SYMBOL(sys_lseek); EXPORT_SYMBOL(kernel_execve); -EXPORT_SYMBOL(sys_setsid); -EXPORT_SYMBOL(sys_wait4); /* Networking helper routines. */ EXPORT_SYMBOL(csum_tcpudp_magic); @@ -134,10 +69,6 @@ EXPORT_SYMBOL(alpha_fp_emul_imprecise); EXPORT_SYMBOL(alpha_fp_emul); #endif -#ifdef CONFIG_ALPHA_BROKEN_IRQ_MASK -EXPORT_SYMBOL(__min_ipl); -#endif - /* * The following are specially called from the uaccess assembly stubs. */ @@ -160,27 +91,10 @@ EXPORT_SYMBOL(up); */ #ifdef CONFIG_SMP -EXPORT_SYMBOL(flush_tlb_mm); -EXPORT_SYMBOL(flush_tlb_range); -EXPORT_SYMBOL(flush_tlb_page); -EXPORT_SYMBOL(smp_imb); -EXPORT_SYMBOL(cpu_data); -EXPORT_SYMBOL(smp_num_cpus); -EXPORT_SYMBOL(smp_call_function); -EXPORT_SYMBOL(smp_call_function_on_cpu); EXPORT_SYMBOL(_atomic_dec_and_lock); #endif /* CONFIG_SMP */ /* - * NUMA specific symbols - */ -#ifdef CONFIG_DISCONTIGMEM -EXPORT_SYMBOL(node_data); -#endif /* CONFIG_DISCONTIGMEM */ - -EXPORT_SYMBOL(rtc_lock); - -/* * The following are special because they're not called * explicitly (the C compiler or assembler generates them in * response to division operations). Fortunately, their @@ -200,8 +114,3 @@ EXPORT_SYMBOL(__remqu); EXPORT_SYMBOL(memcpy); EXPORT_SYMBOL(memset); EXPORT_SYMBOL(memchr); - -#ifdef CONFIG_ALPHA_IRONGATE -EXPORT_SYMBOL(irongate_ioremap); -EXPORT_SYMBOL(irongate_iounmap); -#endif diff --git a/arch/alpha/kernel/core_irongate.c b/arch/alpha/kernel/core_irongate.c index 138d497..e4a0bcf 100644 --- a/arch/alpha/kernel/core_irongate.c +++ b/arch/alpha/kernel/core_irongate.c @@ -404,6 +404,7 @@ irongate_ioremap(unsigned long addr, unsigned long size) #endif return (void __iomem *)vaddr; } +EXPORT_SYMBOL(irongate_ioremap); void irongate_iounmap(volatile void __iomem *xaddr) @@ -414,3 +415,4 @@ irongate_iounmap(volatile void __iomem *xaddr) if (addr) return vfree((void *)(PAGE_MASK & addr)); } +EXPORT_SYMBOL(irongate_iounmap); diff --git a/arch/alpha/kernel/irq_alpha.c b/arch/alpha/kernel/irq_alpha.c index 6dd126b..e16aeb6 100644 --- a/arch/alpha/kernel/irq_alpha.c +++ b/arch/alpha/kernel/irq_alpha.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include @@ -16,6 +17,7 @@ /* Hack minimum IPL during interrupt processing for broken hardware. */ #ifdef CONFIG_ALPHA_BROKEN_IRQ_MASK int __min_ipl; +EXPORT_SYMBOL(__min_ipl); #endif /* @@ -30,6 +32,7 @@ dummy_perf(unsigned long vector, struct pt_regs *regs) } void (*perf_irq)(unsigned long, struct pt_regs *) = dummy_perf; +EXPORT_SYMBOL(perf_irq); /* * The main interrupt entry point. diff --git a/arch/alpha/kernel/pci-noop.c b/arch/alpha/kernel/pci-noop.c index fff5cf9..174b729 100644 --- a/arch/alpha/kernel/pci-noop.c +++ b/arch/alpha/kernel/pci-noop.c @@ -201,6 +201,7 @@ dma_set_mask(struct device *dev, u64 mask) return 0; } +EXPORT_SYMBOL(dma_set_mask); void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) { diff --git a/arch/alpha/kernel/pci_iommu.c b/arch/alpha/kernel/pci_iommu.c index c468e312..6e7d1fe 100644 --- a/arch/alpha/kernel/pci_iommu.c +++ b/arch/alpha/kernel/pci_iommu.c @@ -300,6 +300,7 @@ pci_map_single(struct pci_dev *pdev, void *cpu_addr, size_t size, int dir) dac_allowed = pdev ? pci_dac_dma_supported(pdev, pdev->dma_mask) : 0; return pci_map_single_1(pdev, cpu_addr, size, dac_allowed); } +EXPORT_SYMBOL(pci_map_single); dma_addr_t pci_map_page(struct pci_dev *pdev, struct page *page, unsigned long offset, @@ -314,6 +315,7 @@ pci_map_page(struct pci_dev *pdev, struct page *page, unsigned long offset, return pci_map_single_1(pdev, (char *)page_address(page) + offset, size, dac_allowed); } +EXPORT_SYMBOL(pci_map_page); /* Unmap a single streaming mode DMA translation. The DMA_ADDR and SIZE must match what was provided for in a previous pci_map_single @@ -379,6 +381,7 @@ pci_unmap_single(struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, DBGA2("pci_unmap_single: sg [%lx,%lx] np %ld from %p\n", dma_addr, size, npages, __builtin_return_address(0)); } +EXPORT_SYMBOL(pci_unmap_single); void pci_unmap_page(struct pci_dev *pdev, dma_addr_t dma_addr, @@ -386,6 +389,7 @@ pci_unmap_page(struct pci_dev *pdev, dma_addr_t dma_addr, { pci_unmap_single(pdev, dma_addr, size, direction); } +EXPORT_SYMBOL(pci_unmap_page); /* Allocate and map kernel buffer using consistent mode DMA for PCI device. Returns non-NULL cpu-view pointer to the buffer if @@ -427,6 +431,7 @@ try_again: return cpu_addr; } +EXPORT_SYMBOL(pci_alloc_consistent); /* Free and unmap a consistent DMA buffer. CPU_ADDR and DMA_ADDR must be values that were returned from pci_alloc_consistent. SIZE must @@ -444,7 +449,7 @@ pci_free_consistent(struct pci_dev *pdev, size_t size, void *cpu_addr, DBGA2("pci_free_consistent: [%x,%lx] from %p\n", dma_addr, size, __builtin_return_address(0)); } - +EXPORT_SYMBOL(pci_free_consistent); /* Classify the elements of the scatterlist. Write dma_address of each element with: @@ -672,6 +677,7 @@ pci_map_sg(struct pci_dev *pdev, struct scatterlist *sg, int nents, pci_unmap_sg(pdev, start, out - start, direction); return 0; } +EXPORT_SYMBOL(pci_map_sg); /* Unmap a set of streaming mode DMA translations. Again, cpu read rules concerning calls here are the same as for pci_unmap_single() @@ -752,6 +758,7 @@ pci_unmap_sg(struct pci_dev *pdev, struct scatterlist *sg, int nents, DBGA("pci_unmap_sg: %ld entries\n", nents - (end - sg)); } +EXPORT_SYMBOL(pci_unmap_sg); /* Return whether the given PCI device DMA address mask can be @@ -786,6 +793,7 @@ pci_dma_supported(struct pci_dev *pdev, u64 mask) return 0; } +EXPORT_SYMBOL(pci_dma_supported); /* @@ -908,6 +916,7 @@ pci_dac_dma_supported(struct pci_dev *dev, u64 mask) return ok; } +EXPORT_SYMBOL(pci_dac_dma_supported); dma64_addr_t pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, @@ -917,6 +926,7 @@ pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, + __pa(page_address(page)) + (dma64_addr_t) offset); } +EXPORT_SYMBOL(pci_dac_page_to_dma); struct page * pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr) @@ -924,13 +934,14 @@ pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr) unsigned long paddr = (dma_addr & PAGE_MASK) - alpha_mv.pci_dac_offset; return virt_to_page(__va(paddr)); } +EXPORT_SYMBOL(pci_dac_dma_to_page); unsigned long pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr) { return (dma_addr & ~PAGE_MASK); } - +EXPORT_SYMBOL(pci_dac_dma_to_offset); /* Helper for generic DMA-mapping functions. */ @@ -957,6 +968,7 @@ alpha_gendev_to_pci(struct device *dev) /* This assumes ISA bus master with dma_mask 0xffffff. */ return NULL; } +EXPORT_SYMBOL(alpha_gendev_to_pci); int dma_set_mask(struct device *dev, u64 mask) @@ -969,3 +981,4 @@ dma_set_mask(struct device *dev, u64 mask) return 0; } +EXPORT_SYMBOL(dma_set_mask); diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c index b3a8a29..3370e6f 100644 --- a/arch/alpha/kernel/process.c +++ b/arch/alpha/kernel/process.c @@ -205,6 +205,7 @@ start_thread(struct pt_regs * regs, unsigned long pc, unsigned long sp) regs->ps = 8; wrusp(sp); } +EXPORT_SYMBOL(start_thread); /* * Free current thread data structures etc.. @@ -376,6 +377,7 @@ dump_thread(struct pt_regs * pt, struct user * dump) dump->regs[EF_A2] = pt->r18; memcpy((char *)dump->regs + EF_SIZE, sw->fp, 32 * 8); } +EXPORT_SYMBOL(dump_thread); /* * Fill in the user structure for a ELF core dump. @@ -424,6 +426,7 @@ dump_elf_thread(elf_greg_t *dest, struct pt_regs *pt, struct thread_info *ti) useful value of the thread's UNIQUE field. */ dest[32] = ti->pcb.unique; } +EXPORT_SYMBOL(dump_elf_thread); int dump_elf_task(elf_greg_t *dest, struct task_struct *task) @@ -431,6 +434,7 @@ dump_elf_task(elf_greg_t *dest, struct task_struct *task) dump_elf_thread(dest, task_pt_regs(task), task_thread_info(task)); return 1; } +EXPORT_SYMBOL(dump_elf_task); int dump_elf_task_fp(elf_fpreg_t *dest, struct task_struct *task) @@ -439,6 +443,7 @@ dump_elf_task_fp(elf_fpreg_t *dest, struct task_struct *task) memcpy(dest, sw->fp, 32 * 8); return 1; } +EXPORT_SYMBOL(dump_elf_task_fp); /* * sys_execve() executes a new program. diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c index a94e6d9..1aea7c7 100644 --- a/arch/alpha/kernel/setup.c +++ b/arch/alpha/kernel/setup.c @@ -66,6 +66,7 @@ static struct notifier_block alpha_panic_block = { struct hwrpb_struct *hwrpb; +EXPORT_SYMBOL(hwrpb); unsigned long srm_hae; int alpha_l1i_cacheshape; @@ -111,6 +112,7 @@ unsigned long alpha_agpgart_size = DEFAULT_AGP_APER_SIZE; #ifdef CONFIG_ALPHA_GENERIC struct alpha_machine_vector alpha_mv; int alpha_using_srm; +EXPORT_SYMBOL(alpha_using_srm); #endif static struct alpha_machine_vector *get_sysvec(unsigned long, unsigned long, @@ -137,6 +139,8 @@ struct screen_info screen_info = { .orig_video_points = 16 }; +EXPORT_SYMBOL(screen_info); + /* * The direct map I/O window, if any. This should be the same * for all busses, since it's used by virt_to_bus. @@ -144,6 +148,8 @@ struct screen_info screen_info = { unsigned long __direct_map_base; unsigned long __direct_map_size; +EXPORT_SYMBOL(__direct_map_base); +EXPORT_SYMBOL(__direct_map_size); /* * Declare all of the machine vectors. diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c index 596780e..d1ec4f5 100644 --- a/arch/alpha/kernel/smp.c +++ b/arch/alpha/kernel/smp.c @@ -52,6 +52,7 @@ /* A collection of per-processor data. */ struct cpuinfo_alpha cpu_data[NR_CPUS]; +EXPORT_SYMBOL(cpu_data); /* A collection of single bit ipi messages. */ static struct { @@ -74,6 +75,7 @@ EXPORT_SYMBOL(cpu_online_map); int smp_num_probed; /* Internal processor count */ int smp_num_cpus = 1; /* Number that came online. */ +EXPORT_SYMBOL(smp_num_cpus); extern void calibrate_delay(void); @@ -790,6 +792,7 @@ smp_call_function_on_cpu (void (*func) (void *info), void *info, int retry, return 0; } +EXPORT_SYMBOL(smp_call_function_on_cpu); int smp_call_function (void (*func) (void *info), void *info, int retry, int wait) @@ -797,6 +800,7 @@ smp_call_function (void (*func) (void *info), void *info, int retry, int wait) return smp_call_function_on_cpu (func, info, retry, wait, cpu_online_map); } +EXPORT_SYMBOL(smp_call_function); static void ipi_imb(void *ignored) @@ -811,6 +815,7 @@ smp_imb(void) if (on_each_cpu(ipi_imb, NULL, 1, 1)) printk(KERN_CRIT "smp_imb: timed out\n"); } +EXPORT_SYMBOL(smp_imb); static void ipi_flush_tlb_all(void *ignored) @@ -866,6 +871,7 @@ flush_tlb_mm(struct mm_struct *mm) preempt_enable(); } +EXPORT_SYMBOL(flush_tlb_mm); struct flush_tlb_page_struct { struct vm_area_struct *vma; @@ -918,6 +924,7 @@ flush_tlb_page(struct vm_area_struct *vma, unsigned long addr) preempt_enable(); } +EXPORT_SYMBOL(flush_tlb_page); void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long end) @@ -925,6 +932,7 @@ flush_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long e /* On the Alpha we always flush the whole user tlb. */ flush_tlb_mm(vma->vm_mm); } +EXPORT_SYMBOL(flush_tlb_range); static void ipi_flush_icache_page(void *x) diff --git a/arch/alpha/kernel/time.c b/arch/alpha/kernel/time.c index cf06665..d7053eb 100644 --- a/arch/alpha/kernel/time.c +++ b/arch/alpha/kernel/time.c @@ -57,6 +57,7 @@ static int set_rtc_mmss(unsigned long); DEFINE_SPINLOCK(rtc_lock); +EXPORT_SYMBOL(rtc_lock); #define TICK_SIZE (tick_nsec / 1000) diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c index b826f58..e3e3806 100644 --- a/arch/alpha/mm/numa.c +++ b/arch/alpha/mm/numa.c @@ -13,12 +13,14 @@ #include #include #include +#include #include #include pg_data_t node_data[MAX_NUMNODES]; bootmem_data_t node_bdata[MAX_NUMNODES]; +EXPORT_SYMBOL(node_data); #undef DEBUG_DISCONTIG #ifdef DEBUG_DISCONTIG -- cgit v0.10.2 From f061c5847bcc72eebf6a783f458d42092eac1b6a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Oct 2006 17:45:47 +0100 Subject: [PATCH] i2Output always takes kernel data now Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/char/ip2/i2lib.c b/drivers/char/ip2/i2lib.c index fc944d3..54d93f0 100644 --- a/drivers/char/ip2/i2lib.c +++ b/drivers/char/ip2/i2lib.c @@ -1007,7 +1007,7 @@ i2InputAvailable(i2ChanStrPtr pCh) // applications that one cannot break out of. //****************************************************************************** static int -i2Output(i2ChanStrPtr pCh, const char *pSource, int count, int user ) +i2Output(i2ChanStrPtr pCh, const char *pSource, int count) { i2eBordStrPtr pB; unsigned char *pInsert; @@ -1020,7 +1020,7 @@ i2Output(i2ChanStrPtr pCh, const char *pSource, int count, int user ) int bailout = 10; - ip2trace (CHANN, ITRC_OUTPUT, ITRC_ENTER, 2, count, user ); + ip2trace (CHANN, ITRC_OUTPUT, ITRC_ENTER, 2, count, 0 ); // Ensure channel structure seems real if ( !i2Validate ( pCh ) ) @@ -1087,12 +1087,7 @@ i2Output(i2ChanStrPtr pCh, const char *pSource, int count, int user ) DATA_COUNT_OF(pInsert) = amountToMove; // Move the data - if ( user ) { - rc = copy_from_user((char*)(DATA_OF(pInsert)), pSource, - amountToMove ); - } else { - memcpy( (char*)(DATA_OF(pInsert)), pSource, amountToMove ); - } + memcpy( (char*)(DATA_OF(pInsert)), pSource, amountToMove ); // Adjust pointers and indices pSource += amountToMove; pCh->Obuf_char_count += amountToMove; diff --git a/drivers/char/ip2/i2lib.h b/drivers/char/ip2/i2lib.h index 952e113..e559e9b 100644 --- a/drivers/char/ip2/i2lib.h +++ b/drivers/char/ip2/i2lib.h @@ -332,7 +332,7 @@ static int i2QueueCommands(int, i2ChanStrPtr, int, int, cmdSyntaxPtr,...); static int i2GetStatus(i2ChanStrPtr, int); static int i2Input(i2ChanStrPtr); static int i2InputFlush(i2ChanStrPtr); -static int i2Output(i2ChanStrPtr, const char *, int, int); +static int i2Output(i2ChanStrPtr, const char *, int); static int i2OutputFree(i2ChanStrPtr); static int i2ServiceBoard(i2eBordStrPtr); static void i2DrainOutput(i2ChanStrPtr, int); diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index 858ba54..a3f32d4 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -1704,7 +1704,7 @@ ip2_write( PTTY tty, const unsigned char *pData, int count) /* This is the actual move bit. Make sure it does what we need!!!!! */ WRITE_LOCK_IRQSAVE(&pCh->Pbuf_spinlock,flags); - bytesSent = i2Output( pCh, pData, count, 0 ); + bytesSent = i2Output( pCh, pData, count); WRITE_UNLOCK_IRQRESTORE(&pCh->Pbuf_spinlock,flags); ip2trace (CHANN, ITRC_WRITE, ITRC_RETURN, 1, bytesSent ); @@ -1764,7 +1764,7 @@ ip2_flush_chars( PTTY tty ) // // We may need to restart i2Output if it does not fullfill this request // - strip = i2Output( pCh, pCh->Pbuf, pCh->Pbuf_stuff, 0 ); + strip = i2Output( pCh, pCh->Pbuf, pCh->Pbuf_stuff); if ( strip != pCh->Pbuf_stuff ) { memmove( pCh->Pbuf, &pCh->Pbuf[strip], pCh->Pbuf_stuff - strip ); } -- cgit v0.10.2 From ea5742830ada301de22a68e604b48fb6107c270f Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Tue, 10 Oct 2006 16:56:11 +0900 Subject: [MIPS] Fix timer setup for Jazz Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/jazz/setup.c b/arch/mips/jazz/setup.c index 6dc4135..d848f1a 100644 --- a/arch/mips/jazz/setup.c +++ b/arch/mips/jazz/setup.c @@ -37,7 +37,7 @@ extern void jazz_machine_restart(char *command); extern void jazz_machine_halt(void); extern void jazz_machine_power_off(void); -void __init plat_time_init(struct irqaction *irq) +void __init plat_timer_setup(struct irqaction *irq) { /* set the clock to 100 Hz */ r4030_write_reg32(JAZZ_TIMER_INTERVAL, 9); -- cgit v0.10.2 From f94054121efcba573f12101a75a4be93e871b309 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 10 Oct 2006 15:44:10 +0100 Subject: [MIPS] Workaround for bug in gcc -EB / -EL options. Certain gcc versions upto gcc 4.1.1 (probably 4.2-subversion as of 2006-10-10 don't properly change the the predefined symbols if -EB / -EL are used, so we kludge that here. A bug has been filed at http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29413. Signed-off-by: Ralf Baechle diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 2124350..641aa30 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -91,8 +91,17 @@ cflags-y += -ffreestanding # carefully avoid to add it redundantly because gcc 3.3/3.4 complains # when fed the toolchain default! # -cflags-$(CONFIG_CPU_BIG_ENDIAN) += $(shell $(CC) -dumpmachine |grep -q 'mips.*el-.*' && echo -EB -D__MIPSEB__) -cflags-$(CONFIG_CPU_LITTLE_ENDIAN) += $(shell $(CC) -dumpmachine |grep -q 'mips.*el-.*' || echo -EL -D__MIPSEL__) +# Certain gcc versions upto gcc 4.1.1 (probably 4.2-subversion as of +# 2006-10-10 don't properly change the the predefined symbols if -EB / -EL +# are used, so we kludge that here. A bug has been filed at +# http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29413. +# +undef-all += -UMIPSEB -U_MIPSEB -U__MIPSEB -U__MIPSEB__ +undef-all += -UMIPSEL -U_MIPSEL -U__MIPSEL -U__MIPSEL__ +predef-be += -DMIPSEB -D_MIPSEB -D__MIPSEB -D__MIPSEB__ +predef-le += -DMIPSEL -D_MIPSEL -D__MIPSEL -D__MIPSEL__ +cflags-$(CONFIG_CPU_BIG_ENDIAN) += $(shell $(CC) -dumpmachine |grep -q 'mips.*el-.*' && echo -EB $(undef-all) $(predef-be)) +cflags-$(CONFIG_CPU_LITTLE_ENDIAN) += $(shell $(CC) -dumpmachine |grep -q 'mips.*el-.*' || echo -EL $(undef-all) $(predef-le)) cflags-$(CONFIG_SB1XXX_CORELIS) += $(call cc-option,-mno-sched-prolog) \ -fno-omit-frame-pointer -- cgit v0.10.2 From dff9262ed1491a1e531dc56e687605b5e4cd488d Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Wed, 11 Oct 2006 01:07:01 +0900 Subject: [MIPS] does not need pt_regs anymore. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle diff --git a/include/asm-mips/irq.h b/include/asm-mips/irq.h index 1a9804c..0ce2a80 100644 --- a/include/asm-mips/irq.h +++ b/include/asm-mips/irq.h @@ -24,8 +24,6 @@ static inline int irq_canonicalize(int irq) #define irq_canonicalize(irq) (irq) /* Sane hardware, sane code ... */ #endif -struct pt_regs; - extern asmlinkage unsigned int do_IRQ(unsigned int irq); #ifdef CONFIG_MIPS_MT_SMTC -- cgit v0.10.2 From 9b95e629eab59ee140fe2b17bbd7fea6821c6085 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Tue, 10 Oct 2006 22:46:52 +0900 Subject: [MIPS] Optimize and cleanup get_saved_sp, set_saved_sp If CONFIG_BUILD_ELF64 was not selected and gcc had -msym32 option (i.e. 4.0 or newer), there is no point to use %highest, %higher for kernel symbols. This patch also fixes 64-bit SMTC version of get_saved_sp() which is broken but harmless since there is no such CPUs for now. A bonus is set_saved_sp() and SMP version of get_saved_sp() are more readable now. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle diff --git a/include/asm-mips/stackframe.h b/include/asm-mips/stackframe.h index 158a4cd..1fae5dc 100644 --- a/include/asm-mips/stackframe.h +++ b/include/asm-mips/stackframe.h @@ -59,69 +59,43 @@ .endm #ifdef CONFIG_SMP - .macro get_saved_sp /* SMP variation */ -#ifdef CONFIG_32BIT #ifdef CONFIG_MIPS_MT_SMTC - .set mips32 - mfc0 k0, CP0_TCBIND; - .set mips0 - lui k1, %hi(kernelsp) - srl k0, k0, 19 - /* No need to shift down and up to clear bits 0-1 */ +#define PTEBASE_SHIFT 19 /* TCBIND */ #else - mfc0 k0, CP0_CONTEXT - lui k1, %hi(kernelsp) - srl k0, k0, 23 -#endif - addu k1, k0 - LONG_L k1, %lo(kernelsp)(k1) +#define PTEBASE_SHIFT 23 /* CONTEXT */ #endif -#ifdef CONFIG_64BIT + .macro get_saved_sp /* SMP variation */ #ifdef CONFIG_MIPS_MT_SMTC - .set mips64 - mfc0 k0, CP0_TCBIND; - .set mips0 - lui k0, %highest(kernelsp) - dsrl k1, 19 - /* No need to shift down and up to clear bits 0-2 */ + mfc0 k0, CP0_TCBIND #else - MFC0 k1, CP0_CONTEXT - lui k0, %highest(kernelsp) - dsrl k1, 23 - daddiu k0, %higher(kernelsp) - dsll k0, k0, 16 - daddiu k0, %hi(kernelsp) - dsll k0, k0, 16 -#endif /* CONFIG_MIPS_MT_SMTC */ - daddu k1, k1, k0 + MFC0 k0, CP0_CONTEXT +#endif +#if defined(CONFIG_BUILD_ELF64) || (defined(CONFIG_64BIT) && __GNUC__ < 4) + lui k1, %highest(kernelsp) + daddiu k1, %higher(kernelsp) + dsll k1, 16 + daddiu k1, %hi(kernelsp) + dsll k1, 16 +#else + lui k1, %hi(kernelsp) +#endif + LONG_SRL k0, PTEBASE_SHIFT + LONG_ADDU k1, k0 LONG_L k1, %lo(kernelsp)(k1) -#endif /* CONFIG_64BIT */ .endm .macro set_saved_sp stackp temp temp2 -#ifdef CONFIG_32BIT -#ifdef CONFIG_MIPS_MT_SMTC - mfc0 \temp, CP0_TCBIND - srl \temp, 19 -#else - mfc0 \temp, CP0_CONTEXT - srl \temp, 23 -#endif -#endif -#ifdef CONFIG_64BIT #ifdef CONFIG_MIPS_MT_SMTC mfc0 \temp, CP0_TCBIND - dsrl \temp, 19 #else MFC0 \temp, CP0_CONTEXT - dsrl \temp, 23 -#endif #endif + LONG_SRL \temp, PTEBASE_SHIFT LONG_S \stackp, kernelsp(\temp) .endm #else .macro get_saved_sp /* Uniprocessor variation */ -#ifdef CONFIG_64BIT +#if defined(CONFIG_BUILD_ELF64) || (defined(CONFIG_64BIT) && __GNUC__ < 4) lui k1, %highest(kernelsp) daddiu k1, %higher(kernelsp) dsll k1, k1, 16 -- cgit v0.10.2 From 472f291df59fc1c5eb3ade74a50bfa222030e736 Mon Sep 17 00:00:00 2001 From: Mark Mason Date: Tue, 10 Oct 2006 17:03:10 -0700 Subject: [MIPS] Fix compilation warnings in arch/mips/sibyte/bcm1480/smp.c Signed-off-by: Mark Mason Signed-off-by: Ralf Baechle diff --git a/arch/mips/sibyte/bcm1480/smp.c b/arch/mips/sibyte/bcm1480/smp.c index 6eac36d..bf32827 100644 --- a/arch/mips/sibyte/bcm1480/smp.c +++ b/arch/mips/sibyte/bcm1480/smp.c @@ -34,21 +34,21 @@ extern void smp_call_function_interrupt(void); * independent of board/firmware */ -static void *mailbox_0_set_regs[] = { +static volatile void *mailbox_0_set_regs[] = { IOADDR(A_BCM1480_IMR_CPU0_BASE + R_BCM1480_IMR_MAILBOX_0_SET_CPU), IOADDR(A_BCM1480_IMR_CPU1_BASE + R_BCM1480_IMR_MAILBOX_0_SET_CPU), IOADDR(A_BCM1480_IMR_CPU2_BASE + R_BCM1480_IMR_MAILBOX_0_SET_CPU), IOADDR(A_BCM1480_IMR_CPU3_BASE + R_BCM1480_IMR_MAILBOX_0_SET_CPU), }; -static void *mailbox_0_clear_regs[] = { +static volatile void *mailbox_0_clear_regs[] = { IOADDR(A_BCM1480_IMR_CPU0_BASE + R_BCM1480_IMR_MAILBOX_0_CLR_CPU), IOADDR(A_BCM1480_IMR_CPU1_BASE + R_BCM1480_IMR_MAILBOX_0_CLR_CPU), IOADDR(A_BCM1480_IMR_CPU2_BASE + R_BCM1480_IMR_MAILBOX_0_CLR_CPU), IOADDR(A_BCM1480_IMR_CPU3_BASE + R_BCM1480_IMR_MAILBOX_0_CLR_CPU), }; -static void *mailbox_0_regs[] = { +static volatile void *mailbox_0_regs[] = { IOADDR(A_BCM1480_IMR_CPU0_BASE + R_BCM1480_IMR_MAILBOX_0_CPU), IOADDR(A_BCM1480_IMR_CPU1_BASE + R_BCM1480_IMR_MAILBOX_0_CPU), IOADDR(A_BCM1480_IMR_CPU2_BASE + R_BCM1480_IMR_MAILBOX_0_CPU), -- cgit v0.10.2 From 04d4d7d5e36957b2d941310fc8243aa7bf036f3b Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 11 Oct 2006 01:22:12 +0100 Subject: [MIPS] Cleanup definitions of speed_t and tcflag_t. Signed-off-by: Ralf Baechle diff --git a/include/asm-mips/termbits.h b/include/asm-mips/termbits.h index fa6d04d..b62ec7c 100644 --- a/include/asm-mips/termbits.h +++ b/include/asm-mips/termbits.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1995, 1996, 1999, 2001 Ralf Baechle + * Copyright (C) 1995, 96, 99, 2001, 06 Ralf Baechle * Copyright (C) 1999 Silicon Graphics, Inc. * Copyright (C) 2001 MIPS Technologies, Inc. */ @@ -13,14 +13,8 @@ #include typedef unsigned char cc_t; -#if (_MIPS_SZLONG == 32) -typedef unsigned long speed_t; -typedef unsigned long tcflag_t; -#endif -#if (_MIPS_SZLONG == 64) -typedef __u32 speed_t; -typedef __u32 tcflag_t; -#endif +typedef unsigned int speed_t; +typedef unsigned int tcflag_t; /* * The ABI says nothing about NCC but seems to use NCCS as -- cgit v0.10.2 From 67672f5b1f90ff19b970727b3fe2fa2b3cee3bef Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 11 Oct 2006 01:42:30 +0100 Subject: [MIPS] BigSur: More useful defconfig. Signed-off-by: Ralf Baechle diff --git a/arch/mips/configs/bigsur_defconfig b/arch/mips/configs/bigsur_defconfig index c6a0159..ba3bf73 100644 --- a/arch/mips/configs/bigsur_defconfig +++ b/arch/mips/configs/bigsur_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc1 -# Thu Jul 6 10:02:58 2006 +# Linux kernel version: 2.6.19-rc1 +# Wed Oct 11 01:41:41 2006 # CONFIG_MIPS=y @@ -25,8 +25,6 @@ CONFIG_MIPS=y # CONFIG_MIPS_COBALT is not set # CONFIG_MACH_DECSTATION is not set # CONFIG_MIPS_EV64120 is not set -# CONFIG_MIPS_IVR is not set -# CONFIG_MIPS_ITE8172 is not set # CONFIG_MACH_JAZZ is not set # CONFIG_LASAT is not set # CONFIG_MIPS_ATLAS is not set @@ -83,6 +81,7 @@ CONFIG_RWSEM_GENERIC_SPINLOCK=y CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y CONFIG_DMA_COHERENT=y CONFIG_CPU_BIG_ENDIAN=y @@ -132,8 +131,8 @@ CONFIG_PAGE_SIZE_4KB=y # CONFIG_PAGE_SIZE_64KB is not set # CONFIG_SIBYTE_DMA_PAGEOPS is not set CONFIG_MIPS_MT_DISABLED=y -# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_MT_SMP is not set +# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_VPE_LOADER is not set CONFIG_CPU_HAS_LLSC=y CONFIG_CPU_HAS_SYNC=y @@ -185,9 +184,11 @@ CONFIG_LOCALVERSION="" CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set # CONFIG_POSIX_MQUEUE is not set # CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y +# CONFIG_TASKSTATS is not set +# CONFIG_UTS_NS is not set # CONFIG_AUDIT is not set CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y @@ -195,7 +196,9 @@ CONFIG_IKCONFIG_PROC=y # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y CONFIG_EMBEDDED=y +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -204,12 +207,12 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y CONFIG_BASE_FULL=y -CONFIG_RT_MUTEXES=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -228,6 +231,7 @@ CONFIG_STOP_MACHINE=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_BLK_DEV_IO_TRACE is not set # @@ -249,18 +253,17 @@ CONFIG_DEFAULT_IOSCHED="anticipatory" CONFIG_HW_HAS_PCI=y CONFIG_PCI=y CONFIG_PCI_DOMAINS=y +# CONFIG_PCI_MULTITHREAD_PROBE is not set CONFIG_PCI_DEBUG=y CONFIG_MMU=y # # PCCARD (PCMCIA/CardBus) support # -# CONFIG_PCCARD is not set # # PCI Hotplug Support # -# CONFIG_HOTPLUG_PCI is not set # # Executable file formats @@ -271,7 +274,7 @@ CONFIG_BINFMT_ELF=y CONFIG_MIPS32_COMPAT=y CONFIG_COMPAT=y CONFIG_MIPS32_O32=y -# CONFIG_MIPS32_N32 is not set +CONFIG_MIPS32_N32=y CONFIG_BINFMT_ELF32=y # @@ -288,6 +291,7 @@ CONFIG_PACKET_MMAP=y CONFIG_UNIX=y CONFIG_XFRM=y CONFIG_XFRM_USER=m +# CONFIG_XFRM_SUB_POLICY is not set CONFIG_NET_KEY=y CONFIG_INET=y # CONFIG_IP_MULTICAST is not set @@ -308,10 +312,12 @@ CONFIG_IP_PNP_BOOTP=y # CONFIG_INET_TUNNEL is not set CONFIG_INET_XFRM_MODE_TRANSPORT=m CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set @@ -341,7 +347,6 @@ CONFIG_NETWORK_SECMARK=y # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -368,7 +373,6 @@ CONFIG_NETWORK_SECMARK=y # CONFIG_STANDALONE=y CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set # CONFIG_DEBUG_DRIVER is not set # CONFIG_SYS_HYPERVISOR is not set @@ -404,7 +408,7 @@ CONFIG_BLK_DEV_LOOP=m CONFIG_BLK_DEV_NBD=m # CONFIG_BLK_DEV_SX8 is not set # CONFIG_BLK_DEV_RAM is not set -# CONFIG_BLK_DEV_INITRD is not set +CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set @@ -412,6 +416,7 @@ CONFIG_BLK_DEV_NBD=m # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y +CONFIG_IDE_MAX_HWIFS=4 CONFIG_BLK_DEV_IDE=y # @@ -429,10 +434,40 @@ CONFIG_BLK_DEV_IDEFLOPPY=y # IDE chipset support/bugfixes # CONFIG_IDE_GENERIC=y -# CONFIG_BLK_DEV_IDEPCI is not set +CONFIG_BLK_DEV_IDEPCI=y +# CONFIG_IDEPCI_SHARE_IRQ is not set +# CONFIG_BLK_DEV_OFFBOARD is not set +CONFIG_BLK_DEV_GENERIC=y +# CONFIG_BLK_DEV_OPTI621 is not set +CONFIG_BLK_DEV_IDEDMA_PCI=y +# CONFIG_BLK_DEV_IDEDMA_FORCED is not set +# CONFIG_IDEDMA_PCI_AUTO is not set +# CONFIG_BLK_DEV_AEC62XX is not set +# CONFIG_BLK_DEV_ALI15X3 is not set +# CONFIG_BLK_DEV_AMD74XX is not set +CONFIG_BLK_DEV_CMD64X=y +# CONFIG_BLK_DEV_TRIFLEX is not set +# CONFIG_BLK_DEV_CY82C693 is not set +# CONFIG_BLK_DEV_CS5520 is not set +# CONFIG_BLK_DEV_CS5530 is not set +# CONFIG_BLK_DEV_HPT34X is not set +# CONFIG_BLK_DEV_HPT366 is not set +# CONFIG_BLK_DEV_JMICRON is not set +# CONFIG_BLK_DEV_SC1200 is not set +# CONFIG_BLK_DEV_PIIX is not set +# CONFIG_BLK_DEV_IT821X is not set +# CONFIG_BLK_DEV_NS87415 is not set +# CONFIG_BLK_DEV_PDC202XX_OLD is not set +# CONFIG_BLK_DEV_PDC202XX_NEW is not set +# CONFIG_BLK_DEV_SVWKS is not set +# CONFIG_BLK_DEV_SIIMAGE is not set +# CONFIG_BLK_DEV_SLC90E66 is not set +# CONFIG_BLK_DEV_TRM290 is not set +# CONFIG_BLK_DEV_VIA82CXXX is not set # CONFIG_BLK_DEV_IDE_SWARM is not set # CONFIG_IDE_ARM is not set -# CONFIG_BLK_DEV_IDEDMA is not set +CONFIG_BLK_DEV_IDEDMA=y +# CONFIG_IDEDMA_IVB is not set # CONFIG_IDEDMA_AUTO is not set # CONFIG_BLK_DEV_HD is not set @@ -441,6 +476,12 @@ CONFIG_IDE_GENERIC=y # # CONFIG_RAID_ATTRS is not set # CONFIG_SCSI is not set +# CONFIG_SCSI_NETLINK is not set + +# +# Serial ATA (prod) and Parallel ATA (experimental) drivers +# +# CONFIG_ATA is not set # # Multi-device support (RAID and LVM) @@ -516,6 +557,7 @@ CONFIG_NET_SB1250_MAC=y # CONFIG_SK98LIN is not set # CONFIG_TIGON3 is not set # CONFIG_BNX2 is not set +# CONFIG_QLA3XXX is not set # # Ethernet (10000 Mbit) @@ -650,7 +692,6 @@ CONFIG_I2C_CHARDEV=y # CONFIG_I2C_ALGOBIT is not set # CONFIG_I2C_ALGOPCF is not set # CONFIG_I2C_ALGOPCA is not set -CONFIG_I2C_ALGO_SIBYTE=y # # I2C Hardware Bus support @@ -712,12 +753,12 @@ CONFIG_I2C_DEBUG_CHIP=y # # Misc devices # +# CONFIG_TIFM_CORE is not set # # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -729,6 +770,7 @@ CONFIG_VIDEO_V4L2=y # # CONFIG_FIRMWARE_EDID is not set # CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set # # Sound @@ -811,6 +853,7 @@ CONFIG_FS_MBCACHE=y # CONFIG_JFS_FS is not set CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set @@ -840,8 +883,10 @@ CONFIG_DNOTIFY=y # CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y -# CONFIG_TMPFS is not set +CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y # CONFIG_CONFIGFS_FS is not set @@ -851,6 +896,7 @@ CONFIG_RAMFS=y # # CONFIG_ADFS_FS is not set # CONFIG_AFFS_FS is not set +# CONFIG_ECRYPT_FS is not set # CONFIG_HFS_FS is not set # CONFIG_HFSPLUS_FS is not set # CONFIG_BEFS_FS is not set @@ -881,7 +927,6 @@ CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set # CONFIG_CIFS is not set -# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -899,6 +944,10 @@ CONFIG_MSDOS_PARTITION=y # CONFIG_NLS is not set # +# Distributed Lock Manager +# + +# # Profiling support # # CONFIG_PROFILING is not set @@ -907,7 +956,8 @@ CONFIG_MSDOS_PARTITION=y # Kernel hacking # CONFIG_TRACE_IRQFLAGS_SUPPORT=y -CONFIG_PRINTK_TIME=y +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y @@ -920,12 +970,15 @@ CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_DEBUG_SPINLOCK is not set CONFIG_DEBUG_MUTEXES=y # CONFIG_DEBUG_RWSEMS is not set +# CONFIG_DEBUG_LOCK_ALLOC is not set +# CONFIG_PROVE_LOCKING is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set # CONFIG_DEBUG_KOBJECT is not set # CONFIG_DEBUG_INFO is not set # CONFIG_DEBUG_FS is not set # CONFIG_DEBUG_VM is not set +# CONFIG_DEBUG_LIST is not set CONFIG_FORCED_INLINING=y # CONFIG_RCU_TORTURE_TEST is not set CONFIG_CROSSCOMPILE=y @@ -946,6 +999,10 @@ CONFIG_KEYS_DEBUG_PROC_KEYS=y # Cryptographic options # CONFIG_CRYPTO=y +CONFIG_CRYPTO_ALGAPI=y +CONFIG_CRYPTO_BLKCIPHER=m +CONFIG_CRYPTO_HASH=y +CONFIG_CRYPTO_MANAGER=m CONFIG_CRYPTO_HMAC=y CONFIG_CRYPTO_NULL=y CONFIG_CRYPTO_MD4=y @@ -955,9 +1012,12 @@ CONFIG_CRYPTO_SHA256=y CONFIG_CRYPTO_SHA512=y CONFIG_CRYPTO_WP512=m CONFIG_CRYPTO_TGR192=m +CONFIG_CRYPTO_ECB=m +CONFIG_CRYPTO_CBC=m CONFIG_CRYPTO_DES=y CONFIG_CRYPTO_BLOWFISH=y CONFIG_CRYPTO_TWOFISH=y +CONFIG_CRYPTO_TWOFISH_COMMON=y CONFIG_CRYPTO_SERPENT=y CONFIG_CRYPTO_AES=m # CONFIG_CRYPTO_CAST5 is not set -- cgit v0.10.2 From c11b3c1bc0250027ff05665898f8d8eec40b7e49 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 11 Oct 2006 18:35:33 +0100 Subject: [MIPS] IP27: Make declaration of setup_replication_mask a proper prototype. Signed-off-by: Ralf Baechle diff --git a/arch/mips/sgi-ip27/ip27-klnuma.c b/arch/mips/sgi-ip27/ip27-klnuma.c index d777b7d..f9f404a 100644 --- a/arch/mips/sgi-ip27/ip27-klnuma.c +++ b/arch/mips/sgi-ip27/ip27-klnuma.c @@ -26,7 +26,7 @@ static cpumask_t ktext_repmask; * kernel. For example, we should never put a copy on a headless node, * and we should respect the topology of the machine. */ -void __init setup_replication_mask() +void __init setup_replication_mask(void) { cnodeid_t cnode; -- cgit v0.10.2 From 9a244b95ddb62a17b62f4b061b6e13ca4d177942 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 11 Oct 2006 19:30:03 +0100 Subject: [MIPS] Pass NULL not 0 for pointer value. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c index 1af3612..db80957 100644 --- a/arch/mips/kernel/smp.c +++ b/arch/mips/kernel/smp.c @@ -310,7 +310,7 @@ static void flush_tlb_all_ipi(void *info) void flush_tlb_all(void) { - on_each_cpu(flush_tlb_all_ipi, 0, 1, 1); + on_each_cpu(flush_tlb_all_ipi, NULL, 1, 1); } static void flush_tlb_mm_ipi(void *mm) -- cgit v0.10.2 From 70903ca004fef17b0f6483714baefdb2f6ecceb0 Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 11 Oct 2006 18:49:24 +0000 Subject: [CIFS] Do not need to adjust for Jan/Feb for leap day calculation in 2100 (year divisible by 100) Signed-off-by: Yehuda Sadeh Weinraub Signed-off-by: Steve French diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 32562d1..3adbd12 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -956,7 +956,8 @@ struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) days = days - 1; /* do not count leap year for the year 2100 */ /* adjust for leap year where we are still before leap day */ - days -= ((year & 0x03) == 0) && (month < 2 ? 1 : 0); + if(year != 120) + days -= ((year & 0x03) == 0) && (month < 2 ? 1 : 0); sec += 24 * 60 * 60 * days; ts.tv_sec = sec; -- cgit v0.10.2 From 80060362aaefec507ac2d7a7bd156716d7a7ca91 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 10 Oct 2006 03:40:44 -0400 Subject: [WATCHDOG] watchdog/iTCO_wdt: fix bug related to gcc uninit warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc emits the following warning: drivers/char/watchdog/iTCO_wdt.c: In function ‘iTCO_wdt_ioctl’: drivers/char/watchdog/iTCO_wdt.c:429: warning: ‘time_left’ may be used uninitialized in this function This indicates a condition near enough to a bug, to want to fix. iTCO_wdt_get_timeleft() stores a value in 'time_left' iff iTCO_version==(1 or 2). This driver only supports versions 1 or 2, so this is ok. However, since (a) the return value of iTCO_wdt_get_timeleft() is handled anyway, (b) it fixes the warning, and (c) it future-proofs the driver, we go ahead and add the obvious return value. Signed-off-by: Jeff Garzik Signed-off-by: Wim Van Sebroeck Signed-off-by: Andrew Morton diff --git a/drivers/char/watchdog/iTCO_wdt.c b/drivers/char/watchdog/iTCO_wdt.c index 505aae9..b6f29cb 100644 --- a/drivers/char/watchdog/iTCO_wdt.c +++ b/drivers/char/watchdog/iTCO_wdt.c @@ -368,7 +368,8 @@ static int iTCO_wdt_get_timeleft (int *time_left) spin_unlock(&iTCO_wdt_private.io_lock); *time_left = (val8 * 6) / 10; - } + } else + return -EINVAL; return 0; } @@ -439,7 +440,6 @@ static int iTCO_wdt_ioctl (struct inode *inode, struct file *file, { int new_options, retval = -EINVAL; int new_heartbeat; - int time_left; void __user *argp = (void __user *)arg; int __user *p = argp; static struct watchdog_info ident = { @@ -499,6 +499,8 @@ static int iTCO_wdt_ioctl (struct inode *inode, struct file *file, case WDIOC_GETTIMELEFT: { + int time_left; + if (iTCO_wdt_get_timeleft(&time_left)) return -EINVAL; -- cgit v0.10.2 From 533f90af6d90b9e4859a158565385d1d84a79f75 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Oct 2006 00:02:32 +0000 Subject: [CIFS] Fix old DOS time conversion to handle timezone Signed-off-by: Steve French diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 3adbd12..992e80e 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -942,7 +942,7 @@ struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) cERROR(1,("illegal date, month %d day: %d", month, days)); month -= 1; days += total_days_of_prev_months[month]; - days += 3653; /* account for difference in days between 1980 and 1970 */ + days += 3652; /* account for difference in days between 1980 and 1970 */ year = sd->Year; days += year * 365; days += (year/4); /* leap year */ diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 81e7b2e..80e6ebd 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -106,6 +106,17 @@ static int construct_dentry(struct qstr *qstring, struct file *file, return rc; } +static void AdjustForTZ(struct cifsTconInfo * tcon, struct inode * inode) +{ + if((tcon) && (tcon->ses) && (tcon->ses->server)) { + inode->i_ctime.tv_sec += tcon->ses->server.timeAdj; + inode->i_mtime.tv_sec += tcon->ses->server.timeAdj; + inode->i_atime.tv_sec += tcon->ses->server.timeAdj; + } + return; +} + + static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, char * buf, int *pobject_type, int isNewInode) { @@ -148,7 +159,7 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, tmp_inode->i_ctime = cnvrtDosUnixTm( le16_to_cpu(pfindData->LastWriteDate), le16_to_cpu(pfindData->LastWriteTime)); - + AdjustForTZ(cifs_sb->tcon, tmp_inode); attr = le16_to_cpu(pfindData->Attributes); allocation_size = le32_to_cpu(pfindData->AllocationSize); end_of_file = le32_to_cpu(pfindData->DataSize); -- cgit v0.10.2 From ddae957da48cc381c1472a8909905e1818e4afdd Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Oct 2006 01:23:29 +0000 Subject: [CIFS] fix typo in previous patch Signed-off-by: Steve French diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 80e6ebd..acbabc0 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -109,9 +109,9 @@ static int construct_dentry(struct qstr *qstring, struct file *file, static void AdjustForTZ(struct cifsTconInfo * tcon, struct inode * inode) { if((tcon) && (tcon->ses) && (tcon->ses->server)) { - inode->i_ctime.tv_sec += tcon->ses->server.timeAdj; - inode->i_mtime.tv_sec += tcon->ses->server.timeAdj; - inode->i_atime.tv_sec += tcon->ses->server.timeAdj; + inode->i_ctime.tv_sec += tcon->ses->server->timeAdj; + inode->i_mtime.tv_sec += tcon->ses->server->timeAdj; + inode->i_atime.tv_sec += tcon->ses->server->timeAdj; } return; } -- cgit v0.10.2 From 8884c4cb8b621963b5eb4a9ae45070bd0cb7085f Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 12 Oct 2006 11:56:31 +0900 Subject: sh: Default enable R7780RP IRQs. Now that we've started accounting for spurious IRQs, change the logic somewhat so that we have a better chance of catching them. Signed-off-by: Paul Mundt diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c index 5519d3d..b544772 100644 --- a/arch/sh/boards/renesas/r7780rp/irq.c +++ b/arch/sh/boards/renesas/r7780rp/irq.c @@ -49,6 +49,6 @@ void __init init_r7780rp_IRQ(void) disable_irq_nosync(i); set_irq_chip_and_handler(i, &r7780rp_irq_chip, handle_level_irq); - disable_r7780rp_irq(i); + enable_r7780rp_irq(i); } } -- cgit v0.10.2 From baf4326e49801526e4516e4de7f37b5e51468c49 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 12 Oct 2006 12:03:04 +0900 Subject: sh: interrupt exception handling rework Kill off interrupt_table for all of the CPU subtypes, we now default in to stepping in to do_IRQ() for _all_ IRQ exceptions and counting the spurious ones, rather than simply flipping on the ones we cared about. This and enabling the IRQ by default automatically has already uncovered a couple of bugs and IRQs that weren't being caught, as well as some that are being generated far too often (SCI Tx Data Empty, for example). The general rationale is to use a marker for interrupt exceptions, test for it in the handle_exception() path, and skip out to do_IRQ() if it's found. Everything else follows the same behaviour of finding the cached EXPEVT value in r2/r2_bank, we just rip out the INTEVT read from entry.S entirely (except for in the kGDB NMI case, which is another matter). Note that while this changes the do_IRQ() semantics regarding r4 handling, they were fundamentally broken anyways (relying entirely on r2_bank for the cached code). With this, we do the INTEVT read from do_IRQ() itself (in the CONFIG_CPU_HAS_INTEVT case), or fall back on r4 for the muxed IRQ number, which should also be closer to what SH-2 and SH-2A want anyways. Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/cpu/sh3/ex.S b/arch/sh/kernel/cpu/sh3/ex.S index 44daf44..6be46f0 100644 --- a/arch/sh/kernel/cpu/sh3/ex.S +++ b/arch/sh/kernel/cpu/sh3/ex.S @@ -49,198 +49,3 @@ ENTRY(nmi_slot) #endif ENTRY(user_break_point_trap) .long break_point_trap /* 1E0 */ -ENTRY(interrupt_table) - ! external hardware - .long do_IRQ ! 0000 /* 200 */ - .long do_IRQ ! 0001 - .long do_IRQ ! 0010 - .long do_IRQ ! 0011 - .long do_IRQ ! 0100 - .long do_IRQ ! 0101 - .long do_IRQ ! 0110 - .long do_IRQ ! 0111 - .long do_IRQ ! 1000 /* 300 */ - .long do_IRQ ! 1001 - .long do_IRQ ! 1010 - .long do_IRQ ! 1011 - .long do_IRQ ! 1100 - .long do_IRQ ! 1101 - .long do_IRQ ! 1110 - .long exception_error - ! Internal hardware - .long do_IRQ ! TMU0 tuni0 /* 400 */ - .long do_IRQ ! TMU1 tuni1 - .long do_IRQ ! TMU2 tuni2 - .long do_IRQ ! ticpi2 - .long do_IRQ ! RTC ati - .long do_IRQ ! pri - .long do_IRQ ! cui - .long do_IRQ ! SCI eri - .long do_IRQ ! rxi /* 500 */ - .long do_IRQ ! txi - .long do_IRQ ! tei - .long do_IRQ ! WDT iti /* 560 */ - .long do_IRQ ! REF rcmi - .long do_IRQ ! rovi - .long do_IRQ - .long do_IRQ /* 5E0 */ -#if defined(CONFIG_CPU_SUBTYPE_SH7707) || \ - defined(CONFIG_CPU_SUBTYPE_SH7709) || \ - defined(CONFIG_CPU_SUBTYPE_SH7706) || \ - defined(CONFIG_CPU_SUBTYPE_SH7300) || \ - defined(CONFIG_CPU_SUBTYPE_SH7705) || \ - defined(CONFIG_CPU_SUBTYPE_SH7710) - .long do_IRQ ! 32 IRQ irq0 /* 600 */ - .long do_IRQ ! 33 irq1 - .long do_IRQ ! 34 irq2 - .long do_IRQ ! 35 irq3 - .long do_IRQ ! 36 irq4 - .long do_IRQ ! 37 irq5 - .long do_IRQ ! 38 - .long do_IRQ ! 39 - .long do_IRQ ! 40 PINT pint0-7 /* 700 */ - .long do_IRQ ! 41 pint8-15 - .long do_IRQ ! 42 - .long do_IRQ ! 43 - .long do_IRQ ! 44 - .long do_IRQ ! 45 - .long do_IRQ ! 46 - .long do_IRQ ! 47 - .long do_IRQ ! 48 DMAC dei0 /* 800 */ - .long do_IRQ ! 49 dei1 - .long do_IRQ ! 50 dei2 - .long do_IRQ ! 51 dei3 - .long do_IRQ ! 52 IrDA eri1 - .long do_IRQ ! 53 rxi1 - .long do_IRQ ! 54 bri1 - .long do_IRQ ! 55 txi1 - .long do_IRQ ! 56 SCIF eri2 - .long do_IRQ ! 57 rxi2 - .long do_IRQ ! 58 bri2 - .long do_IRQ ! 59 txi2 - .long do_IRQ ! 60 ADC adi /* 980 */ -#if defined(CONFIG_CPU_SUBTYPE_SH7705) - .long exception_none ! 61 /* 9A0 */ - .long exception_none ! 62 - .long exception_none ! 63 - .long exception_none ! 64 /* A00 */ - .long do_IRQ ! 65 USB usi0 - .long do_IRQ ! 66 usi1 - .long exception_none ! 67 - .long exception_none ! 68 - .long exception_none ! 69 - .long exception_none ! 70 - .long exception_none ! 71 - .long exception_none ! 72 /* B00 */ - .long exception_none ! 73 - .long exception_none ! 74 - .long exception_none ! 75 - .long exception_none ! 76 - .long exception_none ! 77 - .long exception_none ! 78 - .long exception_none ! 79 - .long do_IRQ ! 80 TPU0 tpi0 /* C00 */ - .long do_IRQ ! 81 TPU1 tpi1 - .long exception_none ! 82 - .long exception_none ! 83 - .long do_IRQ ! 84 TPU2 tpi2 - .long do_IRQ ! 85 TPU3 tpi3 /* CA0 */ -#endif -#if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7300) - .long do_IRQ ! 61 LCDC lcdi /* 9A0 */ - .long do_IRQ ! 62 PCC pcc0i - .long do_IRQ ! 63 pcc1i /* 9E0 */ -#endif -#if defined(CONFIG_CPU_SUBTYPE_SH7710) - .long exception_none ! 61 /* 9A0 */ - .long exception_none ! 62 - .long exception_none ! 63 - .long exception_none ! 64 /* A00 */ - .long exception_none ! 65 - .long exception_none ! 66 - .long exception_none ! 67 - .long exception_none ! 68 - .long exception_none ! 69 - .long exception_none ! 70 - .long exception_none ! 71 - .long exception_none ! 72 /* B00 */ - .long exception_none ! 73 - .long exception_none ! 74 - .long exception_none ! 75 - .long do_IRQ ! 76 DMAC2 dei4 /* B80 */ - .long do_IRQ ! 77 DMAC2 dei5 - .long exception_none ! 78 - .long do_IRQ ! 79 IPSEC ipseci /* BE0 */ - .long do_IRQ ! 80 EDMAC eint0 /* C00 */ - .long do_IRQ ! 81 EDMAC eint1 - .long do_IRQ ! 82 EDMAC eint2 - .long exception_none ! 83 /* C60 */ - .long exception_none ! 84 - .long exception_none ! 85 - .long exception_none ! 86 - .long exception_none ! 87 - .long exception_none ! 88 /* D00 */ - .long exception_none ! 89 - .long exception_none ! 90 - .long exception_none ! 91 - .long exception_none ! 92 - .long exception_none ! 93 - .long exception_none ! 94 - .long exception_none ! 95 - .long do_IRQ ! 96 SIOF eri0 /* E00 */ - .long do_IRQ ! 97 txi0 - .long do_IRQ ! 98 rxi0 - .long do_IRQ ! 99 cci0 - .long do_IRQ ! 100 eri1 /* E80 */ - .long do_IRQ ! 101 txi1 - .long do_IRQ ! 102 rxi2 - .long do_IRQ ! 103 cci3 -#endif -#if defined(CONFIG_CPU_SUBTYPE_SH7300) - .long do_IRQ ! 64 - .long do_IRQ ! 65 - .long do_IRQ ! 66 - .long do_IRQ ! 67 - .long do_IRQ ! 68 - .long do_IRQ ! 69 - .long do_IRQ ! 70 - .long do_IRQ ! 71 - .long do_IRQ ! 72 - .long do_IRQ ! 73 - .long do_IRQ ! 74 - .long do_IRQ ! 75 - .long do_IRQ ! 76 - .long do_IRQ ! 77 - .long do_IRQ ! 78 - .long do_IRQ ! 79 - .long do_IRQ ! 80 SCIF0(SH7300) - .long do_IRQ ! 81 - .long do_IRQ ! 82 - .long do_IRQ ! 83 - .long do_IRQ ! 84 - .long do_IRQ ! 85 - .long do_IRQ ! 86 - .long do_IRQ ! 87 - .long do_IRQ ! 88 - .long do_IRQ ! 89 - .long do_IRQ ! 90 - .long do_IRQ ! 91 - .long do_IRQ ! 92 - .long do_IRQ ! 93 - .long do_IRQ ! 94 - .long do_IRQ ! 95 - .long do_IRQ ! 96 - .long do_IRQ ! 97 - .long do_IRQ ! 98 - .long do_IRQ ! 99 - .long do_IRQ ! 100 - .long do_IRQ ! 101 - .long do_IRQ ! 102 - .long do_IRQ ! 103 - .long do_IRQ ! 104 - .long do_IRQ ! 105 - .long do_IRQ ! 106 - .long do_IRQ ! 107 - .long do_IRQ ! 108 -#endif -#endif diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S index 7146893..3f4cd04 100644 --- a/arch/sh/kernel/cpu/sh4/ex.S +++ b/arch/sh/kernel/cpu/sh4/ex.S @@ -53,503 +53,3 @@ ENTRY(nmi_slot) #endif ENTRY(user_break_point_trap) .long break_point_trap /* 1E0 */ -ENTRY(interrupt_table) - ! external hardware - .long do_IRQ ! 0000 /* 200 */ - .long do_IRQ ! 0001 - .long do_IRQ ! 0010 - .long do_IRQ ! 0011 - .long do_IRQ ! 0100 - .long do_IRQ ! 0101 - .long do_IRQ ! 0110 - .long do_IRQ ! 0111 - .long do_IRQ ! 1000 /* 300 */ - .long do_IRQ ! 1001 - .long do_IRQ ! 1010 - .long do_IRQ ! 1011 - .long do_IRQ ! 1100 - .long do_IRQ ! 1101 - .long do_IRQ ! 1110 - .long exception_error - ! Internal hardware -#ifndef CONFIG_CPU_SUBTYPE_SH7780 - .long do_IRQ ! TMU0 tuni0 /* 400 */ - .long do_IRQ ! TMU1 tuni1 - .long do_IRQ ! TMU2 tuni2 - .long do_IRQ ! ticpi2 -#if defined(CONFIG_CPU_SUBTYPE_SH7760) - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error /* 500 */ - .long exception_error - .long exception_error -#else - .long do_IRQ ! RTC ati - .long do_IRQ ! pri - .long do_IRQ ! cui - .long do_IRQ ! SCI eri - .long do_IRQ ! rxi /* 500 */ - .long do_IRQ ! txi - .long do_IRQ ! tei -#endif - .long do_IRQ ! WDT iti /* 560 */ - .long do_IRQ ! REF rcmi - .long do_IRQ ! rovi - .long do_IRQ - .long do_IRQ /* 5E0 */ - .long do_IRQ ! 32 Hitachi UDI /* 600 */ - .long do_IRQ ! 33 GPIO - .long do_IRQ ! 34 DMAC dmte0 - .long do_IRQ ! 35 dmte1 - .long do_IRQ ! 36 dmte2 - .long do_IRQ ! 37 dmte3 - .long do_IRQ ! 38 dmae - .long exception_error ! 39 /* 6E0 */ -#if defined(CONFIG_CPU_SUBTYPE_SH7760) - .long exception_error /* 700 */ - .long exception_error - .long exception_error - .long exception_error /* 760 */ -#else - .long do_IRQ ! 40 SCIF eri /* 700 */ - .long do_IRQ ! 41 rxi - .long do_IRQ ! 42 bri - .long do_IRQ ! 43 txi -#endif -#if CONFIG_NR_ONCHIP_DMA_CHANNELS == 8 - .long do_IRQ ! 44 DMAC dmte4 /* 780 */ - .long do_IRQ ! 45 dmte5 - .long do_IRQ ! 46 dmte6 - .long do_IRQ ! 47 dmte7 /* 7E0 */ -#elif defined(CONFIG_CPU_SUBTYPE_SH7343) - .long do_IRQ ! 44 IIC1 ali /* 780 */ - .long do_IRQ ! 45 tacki - .long do_IRQ ! 46 waiti - .long do_IRQ ! 47 dtei /* 7E0 */ - .long do_IRQ ! 48 DMAC dei0 /* 800 */ - .long do_IRQ ! 49 dei1 /* 820 */ -#else - .long exception_error ! 44 /* 780 */ - .long exception_error ! 45 - .long exception_error ! 46 - .long exception_error ! 47 -#endif -#if defined(CONFIG_SH_FPU) - .long do_fpu_state_restore ! 48 /* 800 */ - .long do_fpu_state_restore ! 49 /* 820 */ -#elif !defined(CONFIG_CPU_SUBTYPE_SH7343) && \ - !defined(CONFIG_CPU_SUBTYPE_SH73180) - .long exception_error - .long exception_error -#endif -#if defined(CONFIG_CPU_SUBTYPE_SH7751) - .long exception_error /* 840 */ - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error /* 900 */ - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! PCI serr /* A00 */ - .long do_IRQ ! dma3 - .long do_IRQ ! dma2 - .long do_IRQ ! dma1 - .long do_IRQ ! dma0 - .long do_IRQ ! pwon - .long do_IRQ ! pwdwn - .long do_IRQ ! err - .long do_IRQ ! TMU3 tuni3 /* B00 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! TMU4 tuni4 /* B80 */ -#elif defined(CONFIG_CPU_SUBTYPE_SH7760) - .long do_IRQ ! IRQ irq6 /* 840 */ - .long do_IRQ ! irq7 - .long do_IRQ ! SCIF eri0 - .long do_IRQ ! rxi0 - .long do_IRQ ! bri0 - .long do_IRQ ! txi0 - .long do_IRQ ! HCAN2 cani0 /* 900 */ - .long do_IRQ ! cani1 - .long do_IRQ ! SSI ssii0 - .long do_IRQ ! ssii1 - .long do_IRQ ! HAC haci0 - .long do_IRQ ! haci1 - .long do_IRQ ! IIC iici0 - .long do_IRQ ! iici1 - .long do_IRQ ! USB usbi /* A00 */ - .long do_IRQ ! LCDC vint - .long exception_error - .long exception_error - .long do_IRQ ! DMABRG dmabrgi0 - .long do_IRQ ! dmabrgi1 - .long do_IRQ ! dmabrgi2 - .long exception_error - .long do_IRQ ! SCIF eri1 /* B00 */ - .long do_IRQ ! rxi1 - .long do_IRQ ! bri1 - .long do_IRQ ! txi1 - .long do_IRQ ! eri2 - .long do_IRQ ! rxi2 - .long do_IRQ ! bri2 - .long do_IRQ ! txi2 - .long do_IRQ ! SIM simeri /* C00 */ - .long do_IRQ ! simrxi - .long do_IRQ ! simtxi - .long do_IRQ ! simtei - .long do_IRQ ! HSPI spii - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! MMCIF mmci0 /* D00 */ - .long do_IRQ ! mmci1 - .long do_IRQ ! mmci2 - .long do_IRQ ! mmci3 - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error /* E00 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! MFI mfii - .long exception_error - .long exception_error - .long exception_error - .long exception_error /* F00 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! ADC adi - .long do_IRQ ! CMT cmti /* FA0 */ -#elif defined(CONFIG_CPU_SUBTYPE_SH73180) || defined(CONFIG_CPU_SUBTYPE_SH7343) - .long do_IRQ ! 50 0x840 - .long do_IRQ ! 51 0x860 - .long do_IRQ ! 52 0x880 - .long do_IRQ ! 53 0x8a0 - .long do_IRQ ! 54 0x8c0 - .long do_IRQ ! 55 0x8e0 - .long do_IRQ ! 56 0x900 - .long do_IRQ ! 57 0x920 - .long do_IRQ ! 58 0x940 - .long do_IRQ ! 59 0x960 - .long do_IRQ ! 60 0x980 - .long do_IRQ ! 61 0x9a0 - .long do_IRQ ! 62 0x9c0 - .long do_IRQ ! 63 0x9e0 - .long do_IRQ ! 64 0xa00 - .long do_IRQ ! 65 0xa20 - .long do_IRQ ! 66 0xa40 - .long do_IRQ ! 67 0xa60 - .long do_IRQ ! 68 0xa80 - .long do_IRQ ! 69 0xaa0 - .long do_IRQ ! 70 0xac0 - .long do_IRQ ! 71 0xae0 - .long do_IRQ ! 72 0xb00 - .long do_IRQ ! 73 0xb20 - .long do_IRQ ! 74 0xb40 - .long do_IRQ ! 75 0xb60 - .long do_IRQ ! 76 0xb80 - .long do_IRQ ! 77 0xba0 - .long do_IRQ ! 78 0xbc0 - .long do_IRQ ! 79 0xbe0 - .long do_IRQ ! 80 0xc00 - .long do_IRQ ! 81 0xc20 - .long do_IRQ ! 82 0xc40 - .long do_IRQ ! 83 0xc60 - .long do_IRQ ! 84 0xc80 - .long do_IRQ ! 85 0xca0 - .long do_IRQ ! 86 0xcc0 - .long do_IRQ ! 87 0xce0 - .long do_IRQ ! 88 0xd00 - .long do_IRQ ! 89 0xd20 - .long do_IRQ ! 90 0xd40 - .long do_IRQ ! 91 0xd60 - .long do_IRQ ! 92 0xd80 - .long do_IRQ ! 93 0xda0 - .long do_IRQ ! 94 0xdc0 - .long do_IRQ ! 95 0xde0 - .long do_IRQ ! 96 0xe00 - .long do_IRQ ! 97 0xe20 - .long do_IRQ ! 98 0xe40 - .long do_IRQ ! 99 0xe60 - .long do_IRQ ! 100 0xe80 - .long do_IRQ ! 101 0xea0 - .long do_IRQ ! 102 0xec0 - .long do_IRQ ! 103 0xee0 - .long do_IRQ ! 104 0xf00 - .long do_IRQ ! 105 0xf20 - .long do_IRQ ! 106 0xf40 - .long do_IRQ ! 107 0xf60 - .long do_IRQ ! 108 0xf80 -#elif defined(CONFIG_CPU_SUBTYPE_ST40STB1) - .long exception_error ! 50 0x840 - .long exception_error ! 51 0x860 - .long exception_error ! 52 0x880 - .long exception_error ! 53 0x8a0 - .long exception_error ! 54 0x8c0 - .long exception_error ! 55 0x8e0 - .long exception_error ! 56 0x900 - .long exception_error ! 57 0x920 - .long exception_error ! 58 0x940 - .long exception_error ! 59 0x960 - .long exception_error ! 60 0x980 - .long exception_error ! 61 0x9a0 - .long exception_error ! 62 0x9c0 - .long exception_error ! 63 0x9e0 - .long do_IRQ ! 64 0xa00 PCI serr - .long do_IRQ ! 65 0xa20 err - .long do_IRQ ! 66 0xa40 ad - .long do_IRQ ! 67 0xa60 pwr_dwn - .long exception_error ! 68 0xa80 - .long exception_error ! 69 0xaa0 - .long exception_error ! 70 0xac0 - .long exception_error ! 71 0xae0 - .long do_IRQ ! 72 0xb00 DMA INT0 - .long do_IRQ ! 73 0xb20 INT1 - .long do_IRQ ! 74 0xb40 INT2 - .long do_IRQ ! 75 0xb60 INT3 - .long do_IRQ ! 76 0xb80 INT4 - .long exception_error ! 77 0xba0 - .long do_IRQ ! 78 0xbc0 DMA ERR - .long exception_error ! 79 0xbe0 - .long do_IRQ ! 80 0xc00 PIO0 - .long do_IRQ ! 81 0xc20 PIO1 - .long do_IRQ ! 82 0xc40 PIO2 - .long exception_error ! 83 0xc60 - .long exception_error ! 84 0xc80 - .long exception_error ! 85 0xca0 - .long exception_error ! 86 0xcc0 - .long exception_error ! 87 0xce0 - .long exception_error ! 88 0xd00 - .long exception_error ! 89 0xd20 - .long exception_error ! 90 0xd40 - .long exception_error ! 91 0xd60 - .long exception_error ! 92 0xd80 - .long exception_error ! 93 0xda0 - .long exception_error ! 94 0xdc0 - .long exception_error ! 95 0xde0 - .long exception_error ! 96 0xe00 - .long exception_error ! 97 0xe20 - .long exception_error ! 98 0xe40 - .long exception_error ! 99 0xe60 - .long exception_error ! 100 0xe80 - .long exception_error ! 101 0xea0 - .long exception_error ! 102 0xec0 - .long exception_error ! 103 0xee0 - .long exception_error ! 104 0xf00 - .long exception_error ! 105 0xf20 - .long exception_error ! 106 0xf40 - .long exception_error ! 107 0xf60 - .long exception_error ! 108 0xf80 - .long exception_error ! 109 0xfa0 - .long exception_error ! 110 0xfc0 - .long exception_error ! 111 0xfe0 - .long do_IRQ ! 112 0x1000 Mailbox - .long exception_error ! 113 0x1020 - .long exception_error ! 114 0x1040 - .long exception_error ! 115 0x1060 - .long exception_error ! 116 0x1080 - .long exception_error ! 117 0x10a0 - .long exception_error ! 118 0x10c0 - .long exception_error ! 119 0x10e0 - .long exception_error ! 120 0x1100 - .long exception_error ! 121 0x1120 - .long exception_error ! 122 0x1140 - .long exception_error ! 123 0x1160 - .long exception_error ! 124 0x1180 - .long exception_error ! 125 0x11a0 - .long exception_error ! 126 0x11c0 - .long exception_error ! 127 0x11e0 - .long exception_error ! 128 0x1200 - .long exception_error ! 129 0x1220 - .long exception_error ! 130 0x1240 - .long exception_error ! 131 0x1260 - .long exception_error ! 132 0x1280 - .long exception_error ! 133 0x12a0 - .long exception_error ! 134 0x12c0 - .long exception_error ! 135 0x12e0 - .long exception_error ! 136 0x1300 - .long exception_error ! 137 0x1320 - .long exception_error ! 138 0x1340 - .long exception_error ! 139 0x1360 - .long do_IRQ ! 140 0x1380 EMPI INV_ADDR - .long exception_error ! 141 0x13a0 - .long exception_error ! 142 0x13c0 - .long exception_error ! 143 0x13e0 -#elif defined(CONFIG_CPU_SUBTYPE_SH7770) - .long do_IRQ ! 50 0x840 - .long do_IRQ ! 51 0x860 - .long do_IRQ ! 52 0x880 - .long do_IRQ ! 53 0x8a0 - .long do_IRQ ! 54 0x8c0 - .long do_IRQ ! 55 0x8e0 - .long do_IRQ ! 56 0x900 - .long do_IRQ ! 57 0x920 - .long do_IRQ ! 58 0x940 - .long do_IRQ ! 59 0x960 - .long do_IRQ ! 60 0x980 - .long do_IRQ ! 61 0x9a0 - .long do_IRQ ! 62 0x9c0 - .long do_IRQ ! 63 0x9e0 - .long do_IRQ ! 64 0xa00 - .long do_IRQ ! 65 0xa20 - .long do_IRQ ! 66 0xa4d - .long do_IRQ ! 67 0xa60 - .long do_IRQ ! 68 0xa80 - .long do_IRQ ! 69 0xaa0 - .long do_IRQ ! 70 0xac0 - .long do_IRQ ! 71 0xae0 - .long do_IRQ ! 72 0xb00 - .long do_IRQ ! 73 0xb20 - .long do_IRQ ! 74 0xb40 - .long do_IRQ ! 75 0xb60 - .long do_IRQ ! 76 0xb80 - .long do_IRQ ! 77 0xba0 - .long do_IRQ ! 78 0xbc0 - .long do_IRQ ! 79 0xbe0 - .long do_IRQ ! 80 0xc00 - .long do_IRQ ! 81 0xc20 - .long do_IRQ ! 82 0xc40 - .long do_IRQ ! 83 0xc60 - .long do_IRQ ! 84 0xc80 - .long do_IRQ ! 85 0xca0 - .long do_IRQ ! 86 0xcc0 - .long do_IRQ ! 87 0xce0 - .long do_IRQ ! 88 0xd00 - .long do_IRQ ! 89 0xd20 - .long do_IRQ ! 90 0xd40 - .long do_IRQ ! 91 0xd60 - .long do_IRQ ! 92 0xd80 - .long do_IRQ ! 93 0xda0 - .long do_IRQ ! 94 0xdc0 - .long do_IRQ ! 95 0xde0 - .long do_IRQ ! 96 0xe00 - .long do_IRQ ! 97 0xe20 - .long do_IRQ ! 98 0xe40 - .long do_IRQ ! 99 0xe60 - .long do_IRQ ! 100 0xe80 - .long do_IRQ ! 101 0xea0 - .long do_IRQ ! 102 0xec0 - .long do_IRQ ! 103 0xee0 - .long do_IRQ ! 104 0xf00 - .long do_IRQ ! 105 0xf20 - .long do_IRQ ! 106 0xf40 - .long do_IRQ ! 107 0xf60 - .long do_IRQ ! 108 0xf80 -#endif -#else - .long exception_error /* 400 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! RTC ati - .long do_IRQ ! pri - .long do_IRQ ! cui - .long exception_error - .long exception_error /* 500 */ - .long exception_error - .long exception_error - .long do_IRQ ! WDT iti /* 560 */ - .long do_IRQ ! TMU-ch0 - .long do_IRQ ! TMU-ch1 - .long do_IRQ ! TMU-ch2 - .long do_IRQ ! ticpi2 /* 5E0 */ - .long do_IRQ ! 32 Hitachi UDI /* 600 */ - .long exception_error - .long do_IRQ ! 34 DMAC dmte0 - .long do_IRQ ! 35 dmte1 - .long do_IRQ ! 36 dmte2 - .long do_IRQ ! 37 dmte3 - .long do_IRQ ! 38 dmae - .long exception_error ! 39 /* 6E0 */ - .long do_IRQ ! 40 SCIF-ch0 eri /* 700 */ - .long do_IRQ ! 41 rxi - .long do_IRQ ! 42 bri - .long do_IRQ ! 43 txi - .long do_IRQ ! 44 DMAC dmte4 /* 780 */ - .long do_IRQ ! 45 dmte5 - .long do_IRQ ! 46 dmte6 - .long do_IRQ ! 47 dmte7 /* 7E0 */ -#if defined(CONFIG_SH_FPU) - .long do_fpu_state_restore ! 48 /* 800 */ - .long do_fpu_state_restore ! 49 /* 820 */ -#else - .long exception_error - .long exception_error -#endif - .long exception_error /* 840 */ - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! 56 CMT /* 900 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! 60 HAC - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! PCI serr /* A00 */ - .long do_IRQ ! INTA - .long do_IRQ ! INTB - .long do_IRQ ! INTC - .long do_IRQ ! INTD - .long do_IRQ ! err - .long do_IRQ ! pwd3 - .long do_IRQ ! pwd2 - .long do_IRQ ! pwd1 /* B00 */ - .long do_IRQ ! pwd0 - .long exception_error - .long exception_error - .long do_IRQ ! SCIF-ch1 eri /* B80 */ - .long do_IRQ ! rxi - .long do_IRQ ! bri - .long do_IRQ ! txi - .long do_IRQ ! SIOF /* C00 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! HSPI /* C80 */ - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! MMCIF fatat /* D00 */ - .long do_IRQ ! tran - .long do_IRQ ! err - .long do_IRQ ! frdy - .long do_IRQ ! DMAC dmint8 /* D80 */ - .long do_IRQ ! dmint9 - .long do_IRQ ! dmint10 - .long do_IRQ ! dmint11 - .long do_IRQ ! TMU-ch3 /* E00 */ - .long do_IRQ ! TMU-ch4 - .long do_IRQ ! TMU-ch5 - .long exception_error - .long do_IRQ ! SSI - .long exception_error - .long exception_error - .long exception_error - .long do_IRQ ! FLCTL flste /* F00 */ - .long do_IRQ ! fltend - .long do_IRQ ! fltrq0 - .long do_IRQ ! fltrq1 - .long do_IRQ ! GPIO gpioi0 /* F80 */ - .long do_IRQ ! gpioi1 - .long do_IRQ ! gpioi2 - .long do_IRQ ! gpioi3 -#endif - diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S index 97c571f..39aaefb 100644 --- a/arch/sh/kernel/entry.S +++ b/arch/sh/kernel/entry.S @@ -1,9 +1,8 @@ -/* $Id: entry.S,v 1.37 2004/06/11 13:02:46 doyu Exp $ - * +/* * linux/arch/sh/entry.S * * Copyright (C) 1999, 2000, 2002 Niibe Yutaka - * Copyright (C) 2003 Paul Mundt + * Copyright (C) 2003 - 2006 Paul Mundt * * This file is subject to the terms and conditions of the GNU General Public * License. See the file "COPYING" in the main directory of this archive @@ -78,7 +77,6 @@ OFF_TRA = (16*4+6*4) #define k3 r3 #define k4 r4 -#define k_ex_code r2_bank /* r2_bank1 */ #define g_imask r6 /* r6_bank1 */ #define k_g_imask r6_bank /* r6_bank1 */ #define current r7 /* r7_bank1 */ @@ -691,7 +689,7 @@ interrupt: 0: #endif /* defined(CONFIG_KGDB_NMI) */ bra handle_exception - mov.l @k2, k2 + mov #-1, k2 ! interrupt exception marker .align 2 1: .long EXPEVT @@ -717,8 +715,7 @@ ENTRY(handle_exception) add current, k1 mov k1, r15 ! change to kernel stack ! -1: mov #-1, k4 - mov.l 2f, k1 +1: mov.l 2f, k1 ! #ifdef CONFIG_SH_DSP mov.l r2, @-r15 ! Save r2, we need another reg @@ -763,6 +760,8 @@ skip_save: #endif ! Save the user registers on the stack. mov.l k2, @-r15 ! EXPEVT + + mov #-1, k4 mov.l k4, @-r15 ! set TRA (default: -1) ! sts.l macl, @-r15 @@ -797,8 +796,21 @@ skip_save: mov.l r2, @-r15 mov.l r1, @-r15 mov.l r0, @-r15 - ! Then, dispatch to the handler, according to the exception code. - stc k_ex_code, r8 + + /* + * This gets a bit tricky.. in the INTEVT case we don't want to use + * the VBR offset as a destination in the jump call table, since all + * of the destinations are the same. In this case, (interrupt) sets + * a marker in r2 (now r2_bank since SR.RB changed), which we check + * to determine the exception type. For all other exceptions, we + * forcibly read EXPEVT from memory and fix up the jump address, in + * the interrupt exception case we jump to do_IRQ() and defer the + * INTEVT read until there. As a bonus, we can also clean up the SR.RB + * checks that do_IRQ() was doing.. + */ + stc r2_bank, r8 + cmp/pz r8 + bf interrupt_exception shlr2 r8 shlr r8 mov.l 4f, r9 @@ -806,6 +818,8 @@ skip_save: mov.l @r9, r9 jmp @r9 nop + rts + nop .align 2 1: .long 0x00001000 ! DSP=1 @@ -813,8 +827,17 @@ skip_save: 3: .long 0xcfffffff ! RB=0, BL=0 4: .long exception_handling_table +interrupt_exception: + mov.l 1f, r9 + jmp @r9 + nop + rts + nop + + .align 2 +1: .long do_IRQ + .align 2 ENTRY(exception_none) rts nop - diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c index 3b93682..acf2602 100644 --- a/arch/sh/kernel/irq.c +++ b/arch/sh/kernel/irq.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -26,6 +27,7 @@ atomic_t irq_err_count; */ void ack_bad_irq(unsigned int irq) { + atomic_inc(&irq_err_count); printk("unexpected IRQ trap at vector %02x\n", irq); } @@ -85,7 +87,7 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5, struct pt_regs regs) { struct pt_regs *old_regs = set_irq_regs(®s); - int irq = r4; + int irq; #ifdef CONFIG_4KSTACKS union irq_ctx *curctx, *irqctx; #endif @@ -109,20 +111,9 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5, #endif #ifdef CONFIG_CPU_HAS_INTEVT - __asm__ __volatile__ ( -#ifdef CONFIG_CPU_HAS_SR_RB - "stc r2_bank, %0\n\t" + irq = (ctrl_inl(INTEVT) >> 5) - 16; #else - "mov.l @%1, %0\n\t" -#endif - "shlr2 %0\n\t" - "shlr2 %0\n\t" - "shlr %0\n\t" - "add #-16, %0\n\t" - : "=z" (irq), "=r" (r4) - : "1" (INTEVT) - : "memory" - ); + irq = r4; #endif irq = irq_demux(irq); @@ -147,9 +138,9 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5, __asm__ __volatile__ ( "mov %0, r4 \n" "mov r15, r9 \n" - "jsr @%2 \n" + "jsr @%1 \n" /* swith to the irq stack */ - " mov %3, r15 \n" + " mov %2, r15 \n" /* restore the stack (ring zero) */ "mov r9, r15 \n" : /* no outputs */ -- cgit v0.10.2 From 8ae91b9ad88a130cd50fc0b78b16e7b9510b8067 Mon Sep 17 00:00:00 2001 From: Ryusuke Sakato Date: Thu, 12 Oct 2006 12:16:13 +0900 Subject: sh: SH-4A UBC support A simple patch to enable the UBC on SH-4A. Signed-off-by: Ryusuke Sakato Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c index 0b1d5dd..91516dc 100644 --- a/arch/sh/kernel/process.c +++ b/arch/sh/kernel/process.c @@ -5,6 +5,7 @@ * Copyright (C) 1995 Linus Torvalds * * SuperH version: Copyright (C) 1999, 2000 Niibe Yutaka & Kaz Kojima + * Copyright (C) 2006 Lineo Solutions Inc. support SH4A UBC */ /* @@ -290,6 +291,24 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp, static void ubc_set_tracing(int asid, unsigned long pc) { +#if defined(CONFIG_CPU_SH4A) + unsigned long val; + + val = (UBC_CBR_ID_INST | UBC_CBR_RW_READ | UBC_CBR_CE); + val |= (UBC_CBR_AIE | UBC_CBR_AIV_SET(asid)); + + ctrl_outl(val, UBC_CBR0); + ctrl_outl(pc, UBC_CAR0); + ctrl_outl(0x0, UBC_CAMR0); + ctrl_outl(0x0, UBC_CBCR); + + val = (UBC_CRR_RES | UBC_CRR_PCB | UBC_CRR_BIE); + ctrl_outl(val, UBC_CRR0); + + /* Read UBC register that we writed last. For chekking UBC Register changed */ + val = ctrl_inl(UBC_CRR0); + +#else /* CONFIG_CPU_SH4A */ ctrl_outl(pc, UBC_BARA); #ifdef CONFIG_MMU @@ -307,6 +326,7 @@ ubc_set_tracing(int asid, unsigned long pc) ctrl_outw(BBR_INST | BBR_READ, UBC_BBRA); ctrl_outw(BRCR_PCBA, UBC_BRCR); } +#endif /* CONFIG_CPU_SH4A */ } /* @@ -359,8 +379,13 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne #endif ubc_set_tracing(asid, next->thread.ubc_pc); } else { +#if defined(CONFIG_CPU_SH4A) + ctrl_outl(UBC_CBR_INIT, UBC_CBR0); + ctrl_outl(UBC_CRR_INIT, UBC_CRR0); +#else ctrl_outw(0, UBC_BBRA); ctrl_outw(0, UBC_BBRB); +#endif } return prev; @@ -460,8 +485,13 @@ asmlinkage void break_point_trap(unsigned long r4, unsigned long r5, struct pt_regs regs) { /* Clear tracing. */ +#if defined(CONFIG_CPU_SH4A) + ctrl_outl(UBC_CBR_INIT, UBC_CBR0); + ctrl_outl(UBC_CRR_INIT, UBC_CRR0); +#else ctrl_outw(0, UBC_BBRA); ctrl_outw(0, UBC_BBRB); +#endif current->thread.ubc_pc = 0; ubc_usercnt -= 1; diff --git a/include/asm-sh/cpu-sh4/ubc.h b/include/asm-sh/cpu-sh4/ubc.h index 3d09431..c86e170 100644 --- a/include/asm-sh/cpu-sh4/ubc.h +++ b/include/asm-sh/cpu-sh4/ubc.h @@ -3,6 +3,7 @@ * * Copyright (C) 1999 Niibe Yutaka * Copyright (C) 2003 Paul Mundt + * Copyright (C) 2006 Lineo Solutions Inc. support SH4A UBC * * This file is subject to the terms and conditions of the GNU General Public * License. See the file "COPYING" in the main directory of this archive @@ -11,6 +12,41 @@ #ifndef __ASM_CPU_SH4_UBC_H #define __ASM_CPU_SH4_UBC_H +#if defined(CONFIG_CPU_SH4A) +#define UBC_CBR0 0xff200000 +#define UBC_CRR0 0xff200004 +#define UBC_CAR0 0xff200008 +#define UBC_CAMR0 0xff20000c +#define UBC_CBR1 0xff200020 +#define UBC_CRR1 0xff200024 +#define UBC_CAR1 0xff200028 +#define UBC_CAMR1 0xff20002c +#define UBC_CDR1 0xff200030 +#define UBC_CDMR1 0xff200034 +#define UBC_CETR1 0xff200038 +#define UBC_CCMFR 0xff200600 +#define UBC_CBCR 0xff200620 + +/* CBR */ +#define UBC_CBR_AIE (0x01<<30) +#define UBC_CBR_ID_INST (0x01<<4) +#define UBC_CBR_RW_READ (0x01<<1) +#define UBC_CBR_CE (0x01) + +#define UBC_CBR_AIV_MASK (0x00FF0000) +#define UBC_CBR_AIV_SHIFT (16) +#define UBC_CBR_AIV_SET(asid) (((asid)< Date: Thu, 12 Oct 2006 03:28:28 +0000 Subject: [CIFS] Level 1 QPathInfo needed for proper OS2 support Signed-off-by: Steve French diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 7dd2f48..4a4fd2d 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -119,6 +119,7 @@ extern int CIFSFindClose(const int, struct cifsTconInfo *tcon, extern int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * findData, + int legacy /* whether to use old info level */, const struct nls_table *nls_codepage, int remap); extern int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 79a01d3..6f50f2b 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2969,6 +2969,7 @@ int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * pFindData, + int legacy /* old style infolevel */, const struct nls_table *nls_codepage, int remap) { /* level 263 SMB_QUERY_FILE_ALL_INFO */ @@ -3017,7 +3018,10 @@ QPathInfoRetry: byte_count = params + 1 /* pad */ ; pSMB->TotalParameterCount = cpu_to_le16(params); pSMB->ParameterCount = pSMB->TotalParameterCount; - pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO); + if(legacy) + pSMB->InformationLevel = cpu_to_le16(SMB_INFO_STANDARD); + else + pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO); pSMB->Reserved4 = 0; pSMB->hdr.smb_buf_length += byte_count; pSMB->ByteCount = cpu_to_le16(byte_count); @@ -3029,13 +3033,24 @@ QPathInfoRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 40)) + if (rc) /* BB add auto retry on EOPNOTSUPP? */ + rc = -EIO; + else if (!legacy && (pSMBr->ByteCount < 40)) rc = -EIO; /* bad smb */ + else if(legacy && (pSMBr->ByteCount < 24)) + rc = -EIO; /* 24 or 26 expected but we do not read last field */ else if (pFindData){ + int size; __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); + if(legacy) /* we do not read the last field, EAsize, fortunately + since it varies by subdialect and on Set vs. Get, is + two bytes or 4 bytes depending but we don't care here */ + size = sizeof(FILE_INFO_STANDARD); + else + size = sizeof(FILE_ALL_INFO); memcpy((char *) pFindData, (char *) &pSMBr->hdr.Protocol + - data_offset, sizeof (FILE_ALL_INFO)); + data_offset, size); } else rc = -ENOMEM; } diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 06dbce3..fe6d21f 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -338,6 +338,7 @@ int cifs_get_inode_info(struct inode **pinode, pfindData = (FILE_ALL_INFO *)buf; /* could do find first instead but this returns more info */ rc = CIFSSMBQPathInfo(xid, pTcon, search_path, pfindData, + 0 /* not legacy */, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); /* BB optimize code so we do not make the above call @@ -385,8 +386,10 @@ int cifs_get_inode_info(struct inode **pinode, /* get new inode */ if (*pinode == NULL) { *pinode = new_inode(sb); - if (*pinode == NULL) + if (*pinode == NULL) { + kfree(buf); return -ENOMEM; + } /* Is an i_ino of zero legal? Can we use that to check if the server supports returning inode numbers? Are there other sanity checks we can use to ensure that -- cgit v0.10.2 From 07646e217f473a3e6213f8228336a9046833d6aa Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 23:45:23 -0400 Subject: Lockdep: fix compile error in drivers/input/serio/serio.c lockdep_set_subclass() was missing in !LOCKDEP case Signed-off-by: Andrew Morton Signed-off-by: Dmitry Torokhov diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 14fec2a..819f08f 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -268,6 +268,8 @@ static inline int lockdep_internal(void) do { (void)(key); } while (0) #define lockdep_set_class_and_subclass(lock, key, sub) \ do { (void)(key); } while (0) +#define lockdep_set_subclass(lock, sub) do { } while (0) + # define INIT_LOCKDEP # define lockdep_reset() do { debug_locks = 1; } while (0) # define lockdep_free_key_range(start, size) do { } while (0) -- cgit v0.10.2 From 0a66045bcfd3a7ba5d1253f9f305b78bf636ac57 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 12 Oct 2006 01:06:23 -0400 Subject: Input: serio core - handle errors returned by device_bind_driver() Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index 480fdc5..211943f 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -118,6 +118,8 @@ static int serio_match_port(const struct serio_device_id *ids, struct serio *ser static void serio_bind_driver(struct serio *serio, struct serio_driver *drv) { + int error; + down_write(&serio_bus.subsys.rwsem); if (serio_match_port(drv->id_table, serio)) { @@ -126,9 +128,19 @@ static void serio_bind_driver(struct serio *serio, struct serio_driver *drv) serio->dev.driver = NULL; goto out; } - device_bind_driver(&serio->dev); + error = device_bind_driver(&serio->dev); + if (error) { + printk(KERN_WARNING + "serio: device_bind_driver() failed " + "for %s (%s) and %s, error: %d\n", + serio->phys, serio->name, + drv->description, error); + serio_disconnect_driver(serio); + serio->dev.driver = NULL; + goto out; + } } -out: + out: up_write(&serio_bus.subsys.rwsem); } -- cgit v0.10.2 From 23de1510e2468ea349354889097e018d4e8770c5 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 12 Oct 2006 01:06:34 -0400 Subject: Input: gameport core - handle errors returned by device_bind_driver() Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c index 3f47ae5..a0af97e 100644 --- a/drivers/input/gameport/gameport.c +++ b/drivers/input/gameport/gameport.c @@ -191,6 +191,8 @@ static void gameport_run_poll_handler(unsigned long d) static void gameport_bind_driver(struct gameport *gameport, struct gameport_driver *drv) { + int error; + down_write(&gameport_bus.subsys.rwsem); gameport->dev.driver = &drv->driver; @@ -198,8 +200,20 @@ static void gameport_bind_driver(struct gameport *gameport, struct gameport_driv gameport->dev.driver = NULL; goto out; } - device_bind_driver(&gameport->dev); -out: + + error = device_bind_driver(&gameport->dev); + if (error) { + printk(KERN_WARNING + "gameport: device_bind_driver() failed " + "for %s (%s) and %s, error: %d\n", + gameport->phys, gameport->name, + drv->description, error); + drv->disconnect(gameport); + gameport->dev.driver = NULL; + goto out; + } + + out: up_write(&gameport_bus.subsys.rwsem); } -- cgit v0.10.2 From b435fdcda126db42343b8055d04a0a27c229717b Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Thu, 12 Oct 2006 01:06:53 -0400 Subject: Input: fm801-gp - handle errors from pci_enable_device() Signed-off-by: Jeff Garzik Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/gameport/fm801-gp.c b/drivers/input/gameport/fm801-gp.c index 90de5af..1dec00e 100644 --- a/drivers/input/gameport/fm801-gp.c +++ b/drivers/input/gameport/fm801-gp.c @@ -82,17 +82,19 @@ static int __devinit fm801_gp_probe(struct pci_dev *pci, const struct pci_device { struct fm801_gp *gp; struct gameport *port; + int error; gp = kzalloc(sizeof(struct fm801_gp), GFP_KERNEL); port = gameport_allocate_port(); if (!gp || !port) { printk(KERN_ERR "fm801-gp: Memory allocation failed\n"); - kfree(gp); - gameport_free_port(port); - return -ENOMEM; + error = -ENOMEM; + goto err_out_free; } - pci_enable_device(pci); + error = pci_enable_device(pci); + if (error) + goto err_out_free; port->open = fm801_gp_open; #ifdef HAVE_COOKED @@ -108,9 +110,8 @@ static int __devinit fm801_gp_probe(struct pci_dev *pci, const struct pci_device if (!gp->res_port) { printk(KERN_DEBUG "fm801-gp: unable to grab region 0x%x-0x%x\n", port->io, port->io + 0x0f); - gameport_free_port(port); - kfree(gp); - return -EBUSY; + error = -EBUSY; + goto err_out_disable_dev; } pci_set_drvdata(pci, gp); @@ -119,6 +120,13 @@ static int __devinit fm801_gp_probe(struct pci_dev *pci, const struct pci_device gameport_register_port(port); return 0; + + err_out_disable_dev: + pci_disable_device(pci); + err_out_free: + gameport_free_port(port); + kfree(gp); + return error; } static void __devexit fm801_gp_remove(struct pci_dev *pci) -- cgit v0.10.2 From c2baeb0526277676d4cee022f663af880c22fe84 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 10 Oct 2006 14:17:05 -0700 Subject: [SPARC32]: pcic.c needs asm/irq_regs.h Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c index b4e50ae..207f1b6 100644 --- a/arch/sparc/kernel/pcic.c +++ b/arch/sparc/kernel/pcic.c @@ -34,6 +34,7 @@ #include #include #include +#include unsigned int pcic_pin_to_irq(unsigned int pin, char *name); -- cgit v0.10.2 From e3096de34c35683bc233f76d4e86562f1d782d0d Mon Sep 17 00:00:00 2001 From: Martin Habets Date: Tue, 10 Oct 2006 14:36:47 -0700 Subject: [SPARC32]: Mark srmmu_nocache_init as __init. Fix these 2.6.19-rc1 build warnings from modpost: WARNING: vmlinux - Section mismatch: reference to .init.text:__alloc_bootmem from .text between 'srmmu_nocache_init' (at offset 0x1a0f8) and 'srmmu_mmu_info' WARNING: vmlinux - Section mismatch: reference to .init.text:__alloc_bootmem from .text between 'srmmu_nocache_init' (at offset 0x1a118) and 'srmmu_mmu_info' WARNING: vmlinux - Section mismatch: reference to .init.text:srmmu_early_allocate_ptable_skeleton from .text between 'srmmu_nocache_init' (at offset 0x1a188) and 'srmmu_mmu_info' Signed-off-by: Martin Habets Signed-off-by: David S. Miller diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c index b27a506..0df7121 100644 --- a/arch/sparc/mm/srmmu.c +++ b/arch/sparc/mm/srmmu.c @@ -402,7 +402,7 @@ void srmmu_nocache_calcsize(void) srmmu_nocache_end = SRMMU_NOCACHE_VADDR + srmmu_nocache_size; } -void srmmu_nocache_init(void) +void __init srmmu_nocache_init(void) { unsigned int bitmap_bits; pgd_t *pgd; -- cgit v0.10.2 From eba8cefc789f6e51a79363604a7db1dba6a678cb Mon Sep 17 00:00:00 2001 From: Martin Habets Date: Tue, 10 Oct 2006 14:44:01 -0700 Subject: [SPARC32]: Fix sparc32 modpost warnings with sunzilog Fix this 2.6.19-rc1 build warnings from modpost: WARNING: vmlinux - Section mismatch: reference to .init.text:sunzilog_console_setup from .data between 'sunzilog_console' (at offset 0x8394) and 'devices_subsys' Signed-off-by: Martin Habets Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/setup.c b/arch/sparc/kernel/setup.c index 0251cab..f5ee1ac 100644 --- a/arch/sparc/kernel/setup.c +++ b/arch/sparc/kernel/setup.c @@ -121,16 +121,6 @@ static struct console prom_debug_console = { .index = -1, }; -int obp_system_intr(void) -{ - if (boot_flags & BOOTME_DEBUG) { - printk("OBP: system interrupted\n"); - prom_halt(); - return 1; - } - return 0; -} - /* * Process kernel command line switches that are specific to the * SPARC or that require special low-level processing. diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c index 9582874..cc8ad48 100644 --- a/arch/sparc64/kernel/setup.c +++ b/arch/sparc64/kernel/setup.c @@ -91,16 +91,6 @@ void kernel_enter_debugger(void) { } -int obp_system_intr(void) -{ - if (boot_flags & BOOTME_DEBUG) { - printk("OBP: system interrupted\n"); - prom_halt(); - return 1; - } - return 0; -} - /* * Process kernel command line switches that are specific to the * SPARC or that require special low-level processing. diff --git a/drivers/serial/sunzilog.c b/drivers/serial/sunzilog.c index 73dd2ee..b2cc703 100644 --- a/drivers/serial/sunzilog.c +++ b/drivers/serial/sunzilog.c @@ -1182,7 +1182,7 @@ static int __init sunzilog_console_setup(struct console *con, char *options) return 0; } -static struct console sunzilog_console = { +static struct console sunzilog_console_ops = { .name = "ttyS", .write = sunzilog_console_write, .device = uart_console_device, @@ -1208,10 +1208,10 @@ static inline struct console *SUNZILOG_CONSOLE(void) if (i == NUM_CHANNELS) return NULL; - sunzilog_console.index = i; + sunzilog_console_ops.index = i; sunzilog_port_table[i].flags |= SUNZILOG_FLAG_IS_CONS; - return &sunzilog_console; + return &sunzilog_console_ops; } #else -- cgit v0.10.2 From ab5da288ce6a526a0a730362b8c4e03c3a0b58d5 Mon Sep 17 00:00:00 2001 From: Martin Habets Date: Wed, 11 Oct 2006 14:58:30 -0700 Subject: [SPARC32]: Fix sparc32 modpost warnings. Fix these 2.6.19-rc1 build warnings from modpost: WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'core_kernel_text' (at offset 0x3e060) and '__kernel_text_address' WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'core_kernel_text' (at offset 0x3e064) and '__kernel_text_address' WARNING: vmlinux - Section mismatch: reference to .init.text:_einittext from .text between 'core_kernel_text' (at offset 0x3e07c) and '__kernel_text_address' WARNING: vmlinux - Section mismatch: reference to .init.text:_einittext from .text between 'core_kernel_text' (at offset 0x3e080) and '__kernel_text_address' WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'is_ksym_addr' (at offset 0x4b3a4) and 'kallsyms_expand_symbol' WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'is_ksym_addr' (at offset 0x4b3a8) and 'kallsyms_expand_symbol' WARNING: vmlinux - Section mismatch: reference to .init.text:_einittext from .text between 'is_ksym_addr' (at offset 0x4b3b4) and 'kallsyms_expand_symbol' WARNING: vmlinux - Section mismatch: reference to .init.text:_einittext from .text between 'is_ksym_addr' (at offset 0x4b3e4) and 'kallsyms_expand_symbol' WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'get_symbol_pos' (at offset 0x4b640) and 'kallsyms_lookup_size_offset' WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'get_symbol_pos' (at offset 0x4b644) and 'kallsyms_lookup_size_offset' WARNING: vmlinux - Section mismatch: reference to .init.text:_einittext from .text between 'get_symbol_pos' (at offset 0x4b654) and 'kallsyms_lookup_size_offset' WARNING: vmlinux - Section mismatch: reference to .init.text:_einittext from .text between 'get_symbol_pos' (at offset 0x4b658) and 'kallsyms_lookup_size_offset' WARNING: vmlinux - Section mismatch: reference to .init.text:_sinittext from .text between 'get_symbol_pos' (at offset 0x4b68c) and 'kallsyms_lookup_size_offset' The crux of the matter is that modpost only checks the relocatable sections. i386 vmlinux has none, so modpost does no checking on it (it does on the modules). However, sparc vmlinux has plenty of relocatable sections because it is being built with 'ld -r' (to allow for btfixup processing). So for sparc, modpost does do a lot of checking. Sure enough, running modpost on arch/sparc/boot/image yields no output (i.e. all is well). modpost.c check_sec_ref() has: /* We want to process only relocation sections and not .init */ if (sechdrs[i].sh_type == SHT_RELA) { // check here } else if (sechdrs[i].sh_type == SHT_REL) { // check here } Signed-off-by: Martin Habets Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 346c19a..1dd78c8 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -36,11 +36,11 @@ SECTIONS . = ALIGN(4096); __init_begin = .; + _sinittext = .; .init.text : { - _sinittext = .; *(.init.text) - _einittext = .; } + _einittext = .; __init_text_end = .; .init.data : { *(.init.data) } . = ALIGN(16); -- cgit v0.10.2 From ffb733c65000ee701294f7b80c4eca2a5f335637 Mon Sep 17 00:00:00 2001 From: "paul.moore@hp.com" Date: Wed, 4 Oct 2006 11:46:31 -0400 Subject: NetLabel: fix a cache race condition Testing revealed a problem with the NetLabel cache where a cached entry could be freed while in use by the LSM layer causing an oops and other problems. This patch fixes that problem by introducing a reference counter to the cache entry so that it is only freed when it is no longer in use. Signed-off-by: Paul Moore Signed-off-by: James Morris diff --git a/include/net/netlabel.h b/include/net/netlabel.h index c63a580..113337c 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -34,6 +34,7 @@ #include #include #include +#include /* * NetLabel - A management interface for maintaining network packet label @@ -106,6 +107,7 @@ int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info); /* LSM security attributes */ struct netlbl_lsm_cache { + atomic_t refcount; void (*free) (const void *data); void *data; }; @@ -117,7 +119,7 @@ struct netlbl_lsm_secattr { unsigned char *mls_cat; size_t mls_cat_len; - struct netlbl_lsm_cache cache; + struct netlbl_lsm_cache *cache; }; /* @@ -126,6 +128,43 @@ struct netlbl_lsm_secattr { /** + * netlbl_secattr_cache_alloc - Allocate and initialize a secattr cache + * @flags: the memory allocation flags + * + * Description: + * Allocate and initialize a netlbl_lsm_cache structure. Returns a pointer + * on success, NULL on failure. + * + */ +static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc(int flags) +{ + struct netlbl_lsm_cache *cache; + + cache = kzalloc(sizeof(*cache), flags); + if (cache) + atomic_set(&cache->refcount, 1); + return cache; +} + +/** + * netlbl_secattr_cache_free - Frees a netlbl_lsm_cache struct + * @cache: the struct to free + * + * Description: + * Frees @secattr including all of the internal buffers. + * + */ +static inline void netlbl_secattr_cache_free(struct netlbl_lsm_cache *cache) +{ + if (!atomic_dec_and_test(&cache->refcount)) + return; + + if (cache->free) + cache->free(cache->data); + kfree(cache); +} + +/** * netlbl_secattr_init - Initialize a netlbl_lsm_secattr struct * @secattr: the struct to initialize * @@ -143,20 +182,16 @@ static inline int netlbl_secattr_init(struct netlbl_lsm_secattr *secattr) /** * netlbl_secattr_destroy - Clears a netlbl_lsm_secattr struct * @secattr: the struct to clear - * @clear_cache: cache clear flag * * Description: * Destroys the @secattr struct, including freeing all of the internal buffers. - * If @clear_cache is true then free the cache fields, otherwise leave them - * intact. The struct must be reset with a call to netlbl_secattr_init() - * before reuse. + * The struct must be reset with a call to netlbl_secattr_init() before reuse. * */ -static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr, - u32 clear_cache) +static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr) { - if (clear_cache && secattr->cache.data != NULL && secattr->cache.free) - secattr->cache.free(secattr->cache.data); + if (secattr->cache) + netlbl_secattr_cache_free(secattr->cache); kfree(secattr->domain); kfree(secattr->mls_cat); } @@ -178,17 +213,14 @@ static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(int flags) /** * netlbl_secattr_free - Frees a netlbl_lsm_secattr struct * @secattr: the struct to free - * @clear_cache: cache clear flag * * Description: - * Frees @secattr including all of the internal buffers. If @clear_cache is - * true then free the cache fields, otherwise leave them intact. + * Frees @secattr including all of the internal buffers. * */ -static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr, - u32 clear_cache) +static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr) { - netlbl_secattr_destroy(secattr, clear_cache); + netlbl_secattr_destroy(secattr); kfree(secattr); } diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index a8e2e87..bde8cca 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -43,6 +43,7 @@ #include #include #include +#include #include struct cipso_v4_domhsh_entry { @@ -79,7 +80,7 @@ struct cipso_v4_map_cache_entry { unsigned char *key; size_t key_len; - struct netlbl_lsm_cache lsm_data; + struct netlbl_lsm_cache *lsm_data; u32 activity; struct list_head list; @@ -188,13 +189,14 @@ static void cipso_v4_doi_domhsh_free(struct rcu_head *entry) * @entry: the entry to free * * Description: - * This function frees the memory associated with a cache entry. + * This function frees the memory associated with a cache entry including the + * LSM cache data if there are no longer any users, i.e. reference count == 0. * */ static void cipso_v4_cache_entry_free(struct cipso_v4_map_cache_entry *entry) { - if (entry->lsm_data.free) - entry->lsm_data.free(entry->lsm_data.data); + if (entry->lsm_data) + netlbl_secattr_cache_free(entry->lsm_data); kfree(entry->key); kfree(entry); } @@ -315,8 +317,8 @@ static int cipso_v4_cache_check(const unsigned char *key, entry->key_len == key_len && memcmp(entry->key, key, key_len) == 0) { entry->activity += 1; - secattr->cache.free = entry->lsm_data.free; - secattr->cache.data = entry->lsm_data.data; + atomic_inc(&entry->lsm_data->refcount); + secattr->cache = entry->lsm_data; if (prev_entry == NULL) { spin_unlock_bh(&cipso_v4_cache[bkt].lock); return 0; @@ -383,8 +385,8 @@ int cipso_v4_cache_add(const struct sk_buff *skb, memcpy(entry->key, cipso_ptr, cipso_ptr_len); entry->key_len = cipso_ptr_len; entry->hash = cipso_v4_map_cache_hash(cipso_ptr, cipso_ptr_len); - entry->lsm_data.free = secattr->cache.free; - entry->lsm_data.data = secattr->cache.data; + atomic_inc(&secattr->cache->refcount); + entry->lsm_data = secattr->cache; bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETBITS - 1); spin_lock_bh(&cipso_v4_cache[bkt].lock); diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index 54fb7de..ff97110 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -200,7 +200,7 @@ void netlbl_cache_invalidate(void) int netlbl_cache_add(const struct sk_buff *skb, const struct netlbl_lsm_secattr *secattr) { - if (secattr->cache.data == NULL) + if (secattr->cache == NULL) return -ENOMSG; if (CIPSO_V4_OPTEXIST(skb)) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 0c219a1..bb2d2bc 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2172,7 +2172,12 @@ struct netlbl_cache { */ static void selinux_netlbl_cache_free(const void *data) { - struct netlbl_cache *cache = NETLBL_CACHE(data); + struct netlbl_cache *cache; + + if (data == NULL) + return; + + cache = NETLBL_CACHE(data); switch (cache->type) { case NETLBL_CACHE_T_MLS: ebitmap_destroy(&cache->data.mls_label.level[0].cat); @@ -2197,17 +2202,20 @@ static void selinux_netlbl_cache_add(struct sk_buff *skb, struct context *ctx) struct netlbl_lsm_secattr secattr; netlbl_secattr_init(&secattr); + secattr.cache = netlbl_secattr_cache_alloc(GFP_ATOMIC); + if (secattr.cache == NULL) + goto netlbl_cache_add_return; cache = kzalloc(sizeof(*cache), GFP_ATOMIC); if (cache == NULL) - goto netlbl_cache_add_failure; - secattr.cache.free = selinux_netlbl_cache_free; - secattr.cache.data = (void *)cache; + goto netlbl_cache_add_return; + secattr.cache->free = selinux_netlbl_cache_free; + secattr.cache->data = (void *)cache; cache->type = NETLBL_CACHE_T_MLS; if (ebitmap_cpy(&cache->data.mls_label.level[0].cat, &ctx->range.level[0].cat) != 0) - goto netlbl_cache_add_failure; + goto netlbl_cache_add_return; cache->data.mls_label.level[1].cat.highbit = cache->data.mls_label.level[0].cat.highbit; cache->data.mls_label.level[1].cat.node = @@ -2215,13 +2223,10 @@ static void selinux_netlbl_cache_add(struct sk_buff *skb, struct context *ctx) cache->data.mls_label.level[0].sens = ctx->range.level[0].sens; cache->data.mls_label.level[1].sens = ctx->range.level[0].sens; - if (netlbl_cache_add(skb, &secattr) != 0) - goto netlbl_cache_add_failure; - - return; + netlbl_cache_add(skb, &secattr); -netlbl_cache_add_failure: - netlbl_secattr_destroy(&secattr, 1); +netlbl_cache_add_return: + netlbl_secattr_destroy(&secattr); } /** @@ -2263,8 +2268,8 @@ static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb, POLICY_RDLOCK; - if (secattr->cache.data) { - cache = NETLBL_CACHE(secattr->cache.data); + if (secattr->cache) { + cache = NETLBL_CACHE(secattr->cache->data); switch (cache->type) { case NETLBL_CACHE_T_SID: *sid = cache->data.sid; @@ -2369,7 +2374,7 @@ static int selinux_netlbl_skbuff_getsid(struct sk_buff *skb, &secattr, base_sid, sid); - netlbl_secattr_destroy(&secattr, 0); + netlbl_secattr_destroy(&secattr); return rc; } @@ -2415,7 +2420,7 @@ static int selinux_netlbl_socket_setsid(struct socket *sock, u32 sid) if (rc == 0) sksec->nlbl_state = NLBL_LABELED; - netlbl_secattr_destroy(&secattr, 0); + netlbl_secattr_destroy(&secattr); netlbl_socket_setsid_return: POLICY_RDUNLOCK; @@ -2517,7 +2522,7 @@ void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock) sksec->sid, &nlbl_peer_sid) == 0) sksec->peer_sid = nlbl_peer_sid; - netlbl_secattr_destroy(&secattr, 0); + netlbl_secattr_destroy(&secattr); sksec->nlbl_state = NLBL_REQUIRE; -- cgit v0.10.2 From 388b24057f90ba109d4bf855006a8809c383eb76 Mon Sep 17 00:00:00 2001 From: "paul.moore@hp.com" Date: Thu, 5 Oct 2006 18:28:24 -0400 Subject: NetLabel: use SECINITSID_UNLABELED for a base SID This patch changes NetLabel to use SECINITSID_UNLABLELED as it's source of SELinux type information when generating a NetLabel context. Signed-off-by: Paul Moore Signed-off-by: James Morris diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index bb2d2bc..18274b0 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2336,7 +2336,7 @@ static int selinux_netlbl_secattr_to_sid(struct sk_buff *skb, selinux_netlbl_cache_add(skb, &ctx_new); ebitmap_destroy(&ctx_new.range.level[0].cat); } else { - *sid = SECINITSID_UNLABELED; + *sid = SECSID_NULL; rc = 0; } @@ -2519,7 +2519,7 @@ void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock) if (netlbl_sock_getattr(sk, &secattr) == 0 && selinux_netlbl_secattr_to_sid(NULL, &secattr, - sksec->sid, + SECINITSID_UNLABELED, &nlbl_peer_sid) == 0) sksec->peer_sid = nlbl_peer_sid; netlbl_secattr_destroy(&secattr); @@ -2552,9 +2552,6 @@ u32 selinux_netlbl_inet_conn_request(struct sk_buff *skb, u32 sock_sid) if (rc != 0) return SECSID_NULL; - if (peer_sid == SECINITSID_UNLABELED) - return SECSID_NULL; - return peer_sid; } @@ -2616,11 +2613,13 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec, u32 netlbl_sid; u32 recv_perm; - rc = selinux_netlbl_skbuff_getsid(skb, SECINITSID_NETMSG, &netlbl_sid); + rc = selinux_netlbl_skbuff_getsid(skb, + SECINITSID_UNLABELED, + &netlbl_sid); if (rc != 0) return rc; - if (netlbl_sid == SECINITSID_UNLABELED) + if (netlbl_sid == SECSID_NULL) return 0; switch (sksec->sclass) { @@ -2658,10 +2657,6 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec, u32 selinux_netlbl_socket_getpeersec_stream(struct socket *sock) { struct sk_security_struct *sksec = sock->sk->sk_security; - - if (sksec->peer_sid == SECINITSID_UNLABELED) - return SECSID_NULL; - return sksec->peer_sid; } @@ -2677,16 +2672,10 @@ u32 selinux_netlbl_socket_getpeersec_stream(struct socket *sock) u32 selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb) { int peer_sid; - struct sock *sk = skb->sk; - struct inode_security_struct *isec; - if (sk == NULL || sk->sk_socket == NULL) - return SECSID_NULL; - - isec = SOCK_INODE(sk->sk_socket)->i_security; - if (selinux_netlbl_skbuff_getsid(skb, isec->sid, &peer_sid) != 0) - return SECSID_NULL; - if (peer_sid == SECINITSID_UNLABELED) + if (selinux_netlbl_skbuff_getsid(skb, + SECINITSID_UNLABELED, + &peer_sid) != 0) return SECSID_NULL; return peer_sid; -- cgit v0.10.2 From 134b0fc544ba062498451611cb6f3e4454221b3d Mon Sep 17 00:00:00 2001 From: James Morris Date: Thu, 5 Oct 2006 15:42:27 -0500 Subject: IPsec: propagate security module errors up from flow_cache_lookup When a security module is loaded (in this case, SELinux), the security_xfrm_policy_lookup() hook can return an access denied permission (or other error). We were not handling that correctly, and in fact inverting the return logic and propagating a false "ok" back up to xfrm_lookup(), which then allowed packets to pass as if they were not associated with an xfrm policy. The way I was seeing the problem was when connecting via IPsec to a confined service on an SELinux box (vsftpd), which did not have the appropriate SELinux policy permissions to send packets via IPsec. The first SYNACK would be blocked, because of an uncached lookup via flow_cache_lookup(), which would fail to resolve an xfrm policy because the SELinux policy is checked at that point via the resolver. However, retransmitted SYNACKs would then find a cached flow entry when calling into flow_cache_lookup() with a null xfrm policy, which is interpreted by xfrm_lookup() as the packet not having any associated policy and similarly to the first case, allowing it to pass without transformation. The solution presented here is to first ensure that errno values are correctly propagated all the way back up through the various call chains from security_xfrm_policy_lookup(), and handled correctly. Then, flow_cache_lookup() is modified, so that if the policy resolver fails (typically a permission denied via the security module), the flow cache entry is killed rather than having a null policy assigned (which indicates that the packet can pass freely). This also forces any future lookups for the same flow to consult the security module (e.g. SELinux) for current security policy (rather than, say, caching the error on the flow cache entry). Signed-off-by: James Morris diff --git a/include/net/flow.h b/include/net/flow.h index ddf5f3c..3b44d72 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -97,7 +97,7 @@ struct flowi { #define FLOW_DIR_FWD 2 struct sock; -typedef void (*flow_resolve_t)(struct flowi *key, u16 family, u8 dir, +typedef int (*flow_resolve_t)(struct flowi *key, u16 family, u8 dir, void **objp, atomic_t **obj_refp); extern void *flow_cache_lookup(struct flowi *key, u16 family, u8 dir, diff --git a/net/core/flow.c b/net/core/flow.c index f23e7e3..b16d31a 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -85,6 +85,14 @@ static void flow_cache_new_hashrnd(unsigned long arg) add_timer(&flow_hash_rnd_timer); } +static void flow_entry_kill(int cpu, struct flow_cache_entry *fle) +{ + if (fle->object) + atomic_dec(fle->object_ref); + kmem_cache_free(flow_cachep, fle); + flow_count(cpu)--; +} + static void __flow_cache_shrink(int cpu, int shrink_to) { struct flow_cache_entry *fle, **flp; @@ -100,10 +108,7 @@ static void __flow_cache_shrink(int cpu, int shrink_to) } while ((fle = *flp) != NULL) { *flp = fle->next; - if (fle->object) - atomic_dec(fle->object_ref); - kmem_cache_free(flow_cachep, fle); - flow_count(cpu)--; + flow_entry_kill(cpu, fle); } } } @@ -220,24 +225,33 @@ void *flow_cache_lookup(struct flowi *key, u16 family, u8 dir, nocache: { + int err; void *obj; atomic_t *obj_ref; - resolver(key, family, dir, &obj, &obj_ref); + err = resolver(key, family, dir, &obj, &obj_ref); if (fle) { - fle->genid = atomic_read(&flow_cache_genid); - - if (fle->object) - atomic_dec(fle->object_ref); - - fle->object = obj; - fle->object_ref = obj_ref; - if (obj) - atomic_inc(fle->object_ref); + if (err) { + /* Force security policy check on next lookup */ + *head = fle->next; + flow_entry_kill(cpu, fle); + } else { + fle->genid = atomic_read(&flow_cache_genid); + + if (fle->object) + atomic_dec(fle->object_ref); + + fle->object = obj; + fle->object_ref = obj_ref; + if (obj) + atomic_inc(fle->object_ref); + } } local_bh_enable(); + if (err) + obj = ERR_PTR(err); return obj; } } diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 2a78616..fffdd34 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -883,30 +883,32 @@ out: } EXPORT_SYMBOL(xfrm_policy_walk); -/* Find policy to apply to this flow. */ - +/* + * Find policy to apply to this flow. + * + * Returns 0 if policy found, else an -errno. + */ static int xfrm_policy_match(struct xfrm_policy *pol, struct flowi *fl, u8 type, u16 family, int dir) { struct xfrm_selector *sel = &pol->selector; - int match; + int match, ret = -ESRCH; if (pol->family != family || pol->type != type) - return 0; + return ret; match = xfrm_selector_match(sel, fl, family); - if (match) { - if (!security_xfrm_policy_lookup(pol, fl->secid, dir)) - return 1; - } + if (match) + ret = security_xfrm_policy_lookup(pol, fl->secid, dir); - return 0; + return ret; } static struct xfrm_policy *xfrm_policy_lookup_bytype(u8 type, struct flowi *fl, u16 family, u8 dir) { + int err; struct xfrm_policy *pol, *ret; xfrm_address_t *daddr, *saddr; struct hlist_node *entry; @@ -922,7 +924,15 @@ static struct xfrm_policy *xfrm_policy_lookup_bytype(u8 type, struct flowi *fl, chain = policy_hash_direct(daddr, saddr, family, dir); ret = NULL; hlist_for_each_entry(pol, entry, chain, bydst) { - if (xfrm_policy_match(pol, fl, type, family, dir)) { + err = xfrm_policy_match(pol, fl, type, family, dir); + if (err) { + if (err == -ESRCH) + continue; + else { + ret = ERR_PTR(err); + goto fail; + } + } else { ret = pol; priority = ret->priority; break; @@ -930,36 +940,53 @@ static struct xfrm_policy *xfrm_policy_lookup_bytype(u8 type, struct flowi *fl, } chain = &xfrm_policy_inexact[dir]; hlist_for_each_entry(pol, entry, chain, bydst) { - if (xfrm_policy_match(pol, fl, type, family, dir) && - pol->priority < priority) { + err = xfrm_policy_match(pol, fl, type, family, dir); + if (err) { + if (err == -ESRCH) + continue; + else { + ret = ERR_PTR(err); + goto fail; + } + } else if (pol->priority < priority) { ret = pol; break; } } if (ret) xfrm_pol_hold(ret); +fail: read_unlock_bh(&xfrm_policy_lock); return ret; } -static void xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir, +static int xfrm_policy_lookup(struct flowi *fl, u16 family, u8 dir, void **objp, atomic_t **obj_refp) { struct xfrm_policy *pol; + int err = 0; #ifdef CONFIG_XFRM_SUB_POLICY pol = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_SUB, fl, family, dir); - if (pol) + if (IS_ERR(pol)) { + err = PTR_ERR(pol); + pol = NULL; + } + if (pol || err) goto end; #endif pol = xfrm_policy_lookup_bytype(XFRM_POLICY_TYPE_MAIN, fl, family, dir); - + if (IS_ERR(pol)) { + err = PTR_ERR(pol); + pol = NULL; + } #ifdef CONFIG_XFRM_SUB_POLICY end: #endif if ((*objp = (void *) pol) != NULL) *obj_refp = &pol->refcnt; + return err; } static inline int policy_to_flow_dir(int dir) @@ -1297,6 +1324,8 @@ restart: policy = flow_cache_lookup(fl, dst_orig->ops->family, dir, xfrm_policy_lookup); + if (IS_ERR(policy)) + return PTR_ERR(policy); } if (!policy) @@ -1343,6 +1372,10 @@ restart: fl, family, XFRM_POLICY_OUT); if (pols[1]) { + if (IS_ERR(pols[1])) { + err = PTR_ERR(pols[1]); + goto error; + } if (pols[1]->action == XFRM_POLICY_BLOCK) { err = -EPERM; goto error; @@ -1581,6 +1614,9 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, pol = flow_cache_lookup(&fl, family, fl_dir, xfrm_policy_lookup); + if (IS_ERR(pol)) + return 0; + if (!pol) { if (skb->sp && secpath_has_nontransport(skb->sp, 0, &xerr_idx)) { xfrm_secpath_reject(xerr_idx, skb, &fl); @@ -1599,6 +1635,8 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, &fl, family, XFRM_POLICY_IN); if (pols[1]) { + if (IS_ERR(pols[1])) + return 0; pols[1]->curlft.use_time = (unsigned long)xtime.tv_sec; npols ++; } -- cgit v0.10.2 From 5b368e61c2bcb2666bb66e2acf1d6d85ba6f474d Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Thu, 5 Oct 2006 15:42:18 -0500 Subject: IPsec: correct semantics for SELinux policy matching Currently when an IPSec policy rule doesn't specify a security context, it is assumed to be "unlabeled" by SELinux, and so the IPSec policy rule fails to match to a flow that it would otherwise match to, unless one has explicitly added an SELinux policy rule allowing the flow to "polmatch" to the "unlabeled" IPSec policy rules. In the absence of such an explicitly added SELinux policy rule, the IPSec policy rule fails to match and so the packet(s) flow in clear text without the otherwise applicable xfrm(s) applied. The above SELinux behavior violates the SELinux security notion of "deny by default" which should actually translate to "encrypt by default" in the above case. This was first reported by Evgeniy Polyakov and the way James Morris was seeing the problem was when connecting via IPsec to a confined service on an SELinux box (vsftpd), which did not have the appropriate SELinux policy permissions to send packets via IPsec. With this patch applied, SELinux "polmatching" of flows Vs. IPSec policy rules will only come into play when there's a explicit context specified for the IPSec policy rule (which also means there's corresponding SELinux policy allowing appropriate domains/flows to polmatch to this context). Secondly, when a security module is loaded (in this case, SELinux), the security_xfrm_policy_lookup() hook can return errors other than access denied, such as -EINVAL. We were not handling that correctly, and in fact inverting the return logic and propagating a false "ok" back up to xfrm_lookup(), which then allowed packets to pass as if they were not associated with an xfrm policy. The solution for this is to first ensure that errno values are correctly propagated all the way back up through the various call chains from security_xfrm_policy_lookup(), and handled correctly. Then, flow_cache_lookup() is modified, so that if the policy resolver fails (typically a permission denied via the security module), the flow cache entry is killed rather than having a null policy assigned (which indicates that the packet can pass freely). This also forces any future lookups for the same flow to consult the security module (e.g. SELinux) for current security policy (rather than, say, caching the error on the flow cache entry). This patch: Fix the selinux side of things. This makes sure SELinux polmatching of flow contexts to IPSec policy rules comes into play only when an explicit context is associated with the IPSec policy rule. Also, this no longer defaults the context of a socket policy to the context of the socket since the "no explicit context" case is now handled properly. Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris diff --git a/include/linux/security.h b/include/linux/security.h index 9b5fea8..b200b98 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -882,7 +882,8 @@ struct request_sock; * Check permission when a flow selects a xfrm_policy for processing * XFRMs on a packet. The hook is called when selecting either a * per-socket policy or a generic xfrm policy. - * Return 0 if permission is granted. + * Return 0 if permission is granted, -ESRCH otherwise, or -errno + * on other errors. * @xfrm_state_pol_flow_match: * @x contains the state to match. * @xp contains the policy to check for a match. @@ -891,6 +892,7 @@ struct request_sock; * @xfrm_flow_state_match: * @fl contains the flow key to match. * @xfrm points to the xfrm_state to match. + * @xp points to the xfrm_policy to match. * Return 1 if there is a match. * @xfrm_decode_session: * @skb points to skb to decode. @@ -1388,7 +1390,8 @@ struct security_operations { int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 fl_secid, u8 dir); int (*xfrm_state_pol_flow_match)(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); - int (*xfrm_flow_state_match)(struct flowi *fl, struct xfrm_state *xfrm); + int (*xfrm_flow_state_match)(struct flowi *fl, struct xfrm_state *xfrm, + struct xfrm_policy *xp); int (*xfrm_decode_session)(struct sk_buff *skb, u32 *secid, int ckall); #endif /* CONFIG_SECURITY_NETWORK_XFRM */ @@ -3120,11 +3123,6 @@ static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm return security_ops->xfrm_policy_alloc_security(xp, sec_ctx, NULL); } -static inline int security_xfrm_sock_policy_alloc(struct xfrm_policy *xp, struct sock *sk) -{ - return security_ops->xfrm_policy_alloc_security(xp, NULL, sk); -} - static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) { return security_ops->xfrm_policy_clone_security(old, new); @@ -3175,9 +3173,10 @@ static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x, return security_ops->xfrm_state_pol_flow_match(x, xp, fl); } -static inline int security_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm) +static inline int security_xfrm_flow_state_match(struct flowi *fl, + struct xfrm_state *xfrm, struct xfrm_policy *xp) { - return security_ops->xfrm_flow_state_match(fl, xfrm); + return security_ops->xfrm_flow_state_match(fl, xfrm, xp); } static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid) @@ -3197,11 +3196,6 @@ static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm return 0; } -static inline int security_xfrm_sock_policy_alloc(struct xfrm_policy *xp, struct sock *sk) -{ - return 0; -} - static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) { return 0; @@ -3249,7 +3243,7 @@ static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x, } static inline int security_xfrm_flow_state_match(struct flowi *fl, - struct xfrm_state *xfrm) + struct xfrm_state *xfrm, struct xfrm_policy *xp) { return 1; } diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 1e2a4dd..737fdb2 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -995,7 +995,8 @@ struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto, int create, unsigned short family); extern void xfrm_policy_flush(u8 type); extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol); -extern int xfrm_bundle_ok(struct xfrm_dst *xdst, struct flowi *fl, int family, int strict); +extern int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst, + struct flowi *fl, int family, int strict); extern void xfrm_init_pmtu(struct dst_entry *dst); extern wait_queue_head_t km_waitq; diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 7a7a001..1bed0cd 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -52,7 +52,7 @@ __xfrm4_find_bundle(struct flowi *fl, struct xfrm_policy *policy) xdst->u.rt.fl.fl4_dst == fl->fl4_dst && xdst->u.rt.fl.fl4_src == fl->fl4_src && xdst->u.rt.fl.fl4_tos == fl->fl4_tos && - xfrm_bundle_ok(xdst, fl, AF_INET, 0)) { + xfrm_bundle_ok(policy, xdst, fl, AF_INET, 0)) { dst_clone(dst); break; } diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 6a252e2..73cee2e 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -73,7 +73,7 @@ __xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy) xdst->u.rt6.rt6i_src.plen); if (ipv6_addr_equal(&xdst->u.rt6.rt6i_dst.addr, &fl_dst_prefix) && ipv6_addr_equal(&xdst->u.rt6.rt6i_src.addr, &fl_src_prefix) && - xfrm_bundle_ok(xdst, fl, AF_INET6, + xfrm_bundle_ok(policy, xdst, fl, AF_INET6, (xdst->u.rt6.rt6i_dst.plen != 128 || xdst->u.rt6.rt6i_src.plen != 128))) { dst_clone(dst); diff --git a/net/key/af_key.c b/net/key/af_key.c index ff98e70..20ff7cc 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -2928,11 +2928,6 @@ static struct xfrm_policy *pfkey_compile_policy(struct sock *sk, int opt, if (*dir) goto out; } - else { - *dir = security_xfrm_sock_policy_alloc(xp, sk); - if (*dir) - goto out; - } *dir = pol->sadb_x_policy_dir-1; return xp; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index fffdd34..695761f 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1744,7 +1744,7 @@ static struct dst_entry *xfrm_dst_check(struct dst_entry *dst, u32 cookie) static int stale_bundle(struct dst_entry *dst) { - return !xfrm_bundle_ok((struct xfrm_dst *)dst, NULL, AF_UNSPEC, 0); + return !xfrm_bundle_ok(NULL, (struct xfrm_dst *)dst, NULL, AF_UNSPEC, 0); } void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev) @@ -1866,7 +1866,8 @@ EXPORT_SYMBOL(xfrm_init_pmtu); * still valid. */ -int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family, int strict) +int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, + struct flowi *fl, int family, int strict) { struct dst_entry *dst = &first->u.dst; struct xfrm_dst *last; @@ -1883,7 +1884,7 @@ int xfrm_bundle_ok(struct xfrm_dst *first, struct flowi *fl, int family, int str if (fl && !xfrm_selector_match(&dst->xfrm->sel, fl, family)) return 0; - if (fl && !security_xfrm_flow_state_match(fl, dst->xfrm)) + if (fl && !security_xfrm_flow_state_match(fl, dst->xfrm, pol)) return 0; if (dst->xfrm->km.state != XFRM_STATE_VALID) return 0; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index d54b3a7..2b2e59d 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1992,15 +1992,6 @@ static struct xfrm_policy *xfrm_compile_policy(struct sock *sk, int opt, xp->type = XFRM_POLICY_TYPE_MAIN; copy_templates(xp, ut, nr); - if (!xp->security) { - int err = security_xfrm_sock_policy_alloc(xp, sk); - if (err) { - kfree(xp); - *dir = err; - return NULL; - } - } - *dir = p->dir; return xp; diff --git a/security/dummy.c b/security/dummy.c index aeee705..43874c1 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -881,7 +881,8 @@ static int dummy_xfrm_state_pol_flow_match(struct xfrm_state *x, return 1; } -static int dummy_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm) +static int dummy_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, + struct xfrm_policy *xp) { return 1; } diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 81eb598..526b280 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -19,7 +19,8 @@ int selinux_xfrm_state_delete(struct xfrm_state *x); int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); -int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm); +int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, + struct xfrm_policy *xp); /* diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 3e742b8..675b995 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -77,8 +77,8 @@ static inline int selinux_authorizable_xfrm(struct xfrm_state *x) */ int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir) { - int rc = 0; - u32 sel_sid = SECINITSID_UNLABELED; + int rc; + u32 sel_sid; struct xfrm_sec_ctx *ctx; /* Context sid is either set to label or ANY_ASSOC */ @@ -88,11 +88,21 @@ int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir) sel_sid = ctx->ctx_sid; } + else + /* + * All flows should be treated as polmatch'ing an + * otherwise applicable "non-labeled" policy. This + * would prevent inadvertent "leaks". + */ + return 0; rc = avc_has_perm(fl_secid, sel_sid, SECCLASS_ASSOCIATION, ASSOCIATION__POLMATCH, NULL); + if (rc == -EACCES) + rc = -ESRCH; + return rc; } @@ -108,15 +118,20 @@ int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy * u32 pol_sid; int err; - if (x->security) - state_sid = x->security->ctx_sid; - else - state_sid = SECINITSID_UNLABELED; - - if (xp->security) + if (xp->security) { + if (!x->security) + /* unlabeled SA and labeled policy can't match */ + return 0; + else + state_sid = x->security->ctx_sid; pol_sid = xp->security->ctx_sid; - else - pol_sid = SECINITSID_UNLABELED; + } else + if (x->security) + /* unlabeled policy and labeled SA can't match */ + return 0; + else + /* unlabeled policy and unlabeled SA match all flows */ + return 1; err = avc_has_perm(state_sid, pol_sid, SECCLASS_ASSOCIATION, ASSOCIATION__POLMATCH, @@ -125,7 +140,11 @@ int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy * if (err) return 0; - return selinux_xfrm_flow_state_match(fl, x); + err = avc_has_perm(fl->secid, state_sid, SECCLASS_ASSOCIATION, + ASSOCIATION__SENDTO, + NULL)? 0:1; + + return err; } /* @@ -133,12 +152,22 @@ int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy * * can use a given security association. */ -int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm) +int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, + struct xfrm_policy *xp) { int rc = 0; u32 sel_sid = SECINITSID_UNLABELED; struct xfrm_sec_ctx *ctx; + if (!xp->security) + if (!xfrm->security) + return 1; + else + return 0; + else + if (!xfrm->security) + return 0; + /* Context sid is either set to label or ANY_ASSOC */ if ((ctx = xfrm->security)) { if (!selinux_authorizable_ctx(ctx)) -- cgit v0.10.2 From 3bccfbc7a7ba4085817deae6e7c67daf0cbd045a Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Thu, 5 Oct 2006 15:42:35 -0500 Subject: IPsec: fix handling of errors for socket policies This treats the security errors encountered in the case of socket policy matching, the same as how these are treated in the case of main/sub policies, which is to return a full lookup failure. Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 695761f..7736b23 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1016,12 +1016,16 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc sk->sk_family); int err = 0; - if (match) - err = security_xfrm_policy_lookup(pol, fl->secid, policy_to_flow_dir(dir)); - - if (match && !err) - xfrm_pol_hold(pol); - else + if (match) { + err = security_xfrm_policy_lookup(pol, fl->secid, + policy_to_flow_dir(dir)); + if (!err) + xfrm_pol_hold(pol); + else if (err == -ESRCH) + pol = NULL; + else + pol = ERR_PTR(err); + } else pol = NULL; } read_unlock_bh(&xfrm_policy_lock); @@ -1313,8 +1317,11 @@ restart: pol_dead = 0; xfrm_nr = 0; - if (sk && sk->sk_policy[1]) + if (sk && sk->sk_policy[1]) { policy = xfrm_sk_policy_lookup(sk, XFRM_POLICY_OUT, fl); + if (IS_ERR(policy)) + return PTR_ERR(policy); + } if (!policy) { /* To accelerate a bit... */ @@ -1607,8 +1614,11 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, } pol = NULL; - if (sk && sk->sk_policy[dir]) + if (sk && sk->sk_policy[dir]) { pol = xfrm_sk_policy_lookup(sk, dir, &fl); + if (IS_ERR(pol)) + return 0; + } if (!pol) pol = flow_cache_lookup(&fl, family, fl_dir, -- cgit v0.10.2 From 6e8c751e07b34d73069e9333f67fbe5ffe31ec3a Mon Sep 17 00:00:00 2001 From: Chad Sellers Date: Fri, 6 Oct 2006 16:09:52 -0400 Subject: SELinux: Bug fix in polidydb_destroy This patch fixes two bugs in policydb_destroy. Two list pointers (policydb.ocontexts[i] and policydb.genfs) were not being reset to NULL when the lists they pointed to were being freed. This caused a problem when the initial policy load failed, as the policydb being destroyed was not a temporary new policydb that was thrown away, but rather was the global (active) policydb. Consequently, later functions, particularly sys_bind->selinux_socket_bind->security_node_sid and do_rw_proc->selinux_sysctl->selinux_proc_get_sid->security_genfs_sid tried to dereference memory that had previously been freed. Signed-off-by: Chad Sellers Signed-off-by: James Morris diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c index b188953..ba48961 100644 --- a/security/selinux/ss/policydb.c +++ b/security/selinux/ss/policydb.c @@ -618,6 +618,7 @@ void policydb_destroy(struct policydb *p) c = c->next; ocontext_destroy(ctmp,i); } + p->ocontexts[i] = NULL; } g = p->genfs; @@ -633,6 +634,7 @@ void policydb_destroy(struct policydb *p) g = g->next; kfree(gtmp); } + p->genfs = NULL; cond_policydb_destroy(p); -- cgit v0.10.2 From 331c4ee7faa4ee1e1404c872a139784753100498 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 9 Oct 2006 21:34:04 -0700 Subject: [SCTP]: Fix receive buffer accounting. When doing receiver buffer accounting, we always used skb->truesize. This is problematic when processing bundled DATA chunks because for every DATA chunk that could be small part of one large skb, we would charge the size of the entire skb. The new approach is to store the size of the DATA chunk we are accounting for in the sctp_ulpevent structure and use that stored value for accounting. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index ee68a31..764e3af 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -139,6 +139,7 @@ int sctp_inet_listen(struct socket *sock, int backlog); void sctp_write_space(struct sock *sk); unsigned int sctp_poll(struct file *file, struct socket *sock, poll_table *wait); +void sctp_sock_rfree(struct sk_buff *skb); /* * sctp/primitive.c @@ -444,6 +445,19 @@ static inline struct list_head *sctp_list_dequeue(struct list_head *list) return result; } +/* SCTP version of skb_set_owner_r. We need this one because + * of the way we have to do receive buffer accounting on bundled + * chunks. + */ +static inline void sctp_skb_set_owner_r(struct sk_buff *skb, struct sock *sk) +{ + struct sctp_ulpevent *event = sctp_skb2event(skb); + + skb->sk = sk; + skb->destructor = sctp_sock_rfree; + atomic_add(event->rmem_len, &sk->sk_rmem_alloc); +} + /* Tests if the list has one and only one entry. */ static inline int sctp_list_single_entry(struct list_head *head) { diff --git a/include/net/sctp/ulpevent.h b/include/net/sctp/ulpevent.h index 6c40cfc..1a4ddc1 100644 --- a/include/net/sctp/ulpevent.h +++ b/include/net/sctp/ulpevent.h @@ -63,6 +63,7 @@ struct sctp_ulpevent { __u32 cumtsn; int msg_flags; int iif; + unsigned int rmem_len; }; /* Retrieve the skb this event sits inside of. */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 3fe906d..9deec43 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5362,6 +5362,20 @@ static void sctp_wfree(struct sk_buff *skb) sctp_association_put(asoc); } +/* Do accounting for the receive space on the socket. + * Accounting for the association is done in ulpevent.c + * We set this as a destructor for the cloned data skbs so that + * accounting is done at the correct time. + */ +void sctp_sock_rfree(struct sk_buff *skb) +{ + struct sock *sk = skb->sk; + struct sctp_ulpevent *event = sctp_skb2event(skb); + + atomic_sub(event->rmem_len, &sk->sk_rmem_alloc); +} + + /* Helper function to wait for space in the sndbuf. */ static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p, size_t msg_len) @@ -5634,10 +5648,10 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, sctp_skb_for_each(skb, &oldsk->sk_receive_queue, tmp) { event = sctp_skb2event(skb); if (event->asoc == assoc) { - sock_rfree(skb); + sctp_sock_rfree(skb); __skb_unlink(skb, &oldsk->sk_receive_queue); __skb_queue_tail(&newsk->sk_receive_queue, skb); - skb_set_owner_r(skb, newsk); + sctp_skb_set_owner_r(skb, newsk); } } @@ -5665,10 +5679,10 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, sctp_skb_for_each(skb, &oldsp->pd_lobby, tmp) { event = sctp_skb2event(skb); if (event->asoc == assoc) { - sock_rfree(skb); + sctp_sock_rfree(skb); __skb_unlink(skb, &oldsp->pd_lobby); __skb_queue_tail(queue, skb); - skb_set_owner_r(skb, newsk); + sctp_skb_set_owner_r(skb, newsk); } } diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index ee236784..a015283 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -55,10 +55,13 @@ static void sctp_ulpevent_release_frag_data(struct sctp_ulpevent *event); /* Initialize an ULP event from an given skb. */ -SCTP_STATIC void sctp_ulpevent_init(struct sctp_ulpevent *event, int msg_flags) +SCTP_STATIC void sctp_ulpevent_init(struct sctp_ulpevent *event, + int msg_flags, + unsigned int len) { memset(event, 0, sizeof(struct sctp_ulpevent)); event->msg_flags = msg_flags; + event->rmem_len = len; } /* Create a new sctp_ulpevent. */ @@ -73,7 +76,7 @@ SCTP_STATIC struct sctp_ulpevent *sctp_ulpevent_new(int size, int msg_flags, goto fail; event = sctp_skb2event(skb); - sctp_ulpevent_init(event, msg_flags); + sctp_ulpevent_init(event, msg_flags, skb->truesize); return event; @@ -101,17 +104,16 @@ static inline void sctp_ulpevent_set_owner(struct sctp_ulpevent *event, sctp_association_hold((struct sctp_association *)asoc); skb = sctp_event2skb(event); event->asoc = (struct sctp_association *)asoc; - atomic_add(skb->truesize, &event->asoc->rmem_alloc); - skb_set_owner_r(skb, asoc->base.sk); + atomic_add(event->rmem_len, &event->asoc->rmem_alloc); + sctp_skb_set_owner_r(skb, asoc->base.sk); } /* A simple destructor to give up the reference to the association. */ static inline void sctp_ulpevent_release_owner(struct sctp_ulpevent *event) { struct sctp_association *asoc = event->asoc; - struct sk_buff *skb = sctp_event2skb(event); - atomic_sub(skb->truesize, &asoc->rmem_alloc); + atomic_sub(event->rmem_len, &asoc->rmem_alloc); sctp_association_put(asoc); } @@ -372,7 +374,7 @@ struct sctp_ulpevent *sctp_ulpevent_make_remote_error( /* Embed the event fields inside the cloned skb. */ event = sctp_skb2event(skb); - sctp_ulpevent_init(event, MSG_NOTIFICATION); + sctp_ulpevent_init(event, MSG_NOTIFICATION, skb->truesize); sre = (struct sctp_remote_error *) skb_push(skb, sizeof(struct sctp_remote_error)); @@ -464,7 +466,7 @@ struct sctp_ulpevent *sctp_ulpevent_make_send_failed( /* Embed the event fields inside the cloned skb. */ event = sctp_skb2event(skb); - sctp_ulpevent_init(event, MSG_NOTIFICATION); + sctp_ulpevent_init(event, MSG_NOTIFICATION, skb->truesize); ssf = (struct sctp_send_failed *) skb_push(skb, sizeof(struct sctp_send_failed)); @@ -682,8 +684,11 @@ struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc, /* Embed the event fields inside the cloned skb. */ event = sctp_skb2event(skb); - /* Initialize event with flags 0. */ - sctp_ulpevent_init(event, 0); + /* Initialize event with flags 0 and correct length + * Since this is a clone of the original skb, only account for + * the data of this chunk as other chunks will be accounted separately. + */ + sctp_ulpevent_init(event, 0, skb->len + sizeof(struct sk_buff)); sctp_ulpevent_receive_data(event, asoc); diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c index 575e556..e1d1442 100644 --- a/net/sctp/ulpqueue.c +++ b/net/sctp/ulpqueue.c @@ -309,7 +309,7 @@ static struct sctp_ulpevent *sctp_make_reassembled_event(struct sk_buff_head *qu if (!new) return NULL; /* try again later */ - new->sk = f_frag->sk; + sctp_skb_set_owner_r(new, f_frag->sk); skb_shinfo(new)->frag_list = pos; } else -- cgit v0.10.2 From 6aa2551cf135f1d246d31482adc8c679eeea3a83 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 9 Oct 2006 21:34:26 -0700 Subject: [SCTP]: Fix the RX queue size shown in /proc/net/sctp/assocs output. Show the true receive buffer usage. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller diff --git a/net/sctp/proc.c b/net/sctp/proc.c index a356d8d..7f49e76 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -344,7 +344,7 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) assoc, sk, sctp_sk(sk)->type, sk->sk_state, assoc->state, hash, assoc->assoc_id, assoc->sndbuf_used, - (sk->sk_rcvbuf - assoc->rwnd), + atomic_read(&assoc->rmem_alloc), sock_i_uid(sk), sock_i_ino(sk), epb->bind_addr.port, assoc->peer.port); -- cgit v0.10.2 From effee6a00034a8d83a6dea6d221820d87364ac21 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 9 Oct 2006 21:42:14 -0700 Subject: [NET]: File descriptor loss while receiving SCM_RIGHTS If more than one file descriptor was sent with an SCM_RIGHTS message, and on the receiving end, after installing a nonzero (but not all) file descritpors the process runs out of fds, then the already installed fds will be lost (userspace will have no way of knowing about them). The following patch makes sure, that at least the already installed fds are sent to userspace. It doesn't solve the issue of losing file descriptors in case of an EFAULT on the userspace buffer. Signed-off-by: Miklos Szeredi Signed-off-by: David S. Miller diff --git a/net/compat.c b/net/compat.c index d5d69fa..52d32f1 100644 --- a/net/compat.c +++ b/net/compat.c @@ -285,8 +285,7 @@ void scm_detach_fds_compat(struct msghdr *kmsg, struct scm_cookie *scm) if (i > 0) { int cmlen = CMSG_COMPAT_LEN(i * sizeof(int)); - if (!err) - err = put_user(SOL_SOCKET, &cm->cmsg_level); + err = put_user(SOL_SOCKET, &cm->cmsg_level); if (!err) err = put_user(SCM_RIGHTS, &cm->cmsg_type); if (!err) diff --git a/net/core/scm.c b/net/core/scm.c index 649d01e..271cf06 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -245,8 +245,7 @@ void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm) if (i > 0) { int cmlen = CMSG_LEN(i*sizeof(int)); - if (!err) - err = put_user(SOL_SOCKET, &cm->cmsg_level); + err = put_user(SOL_SOCKET, &cm->cmsg_level); if (!err) err = put_user(SCM_RIGHTS, &cm->cmsg_type); if (!err) -- cgit v0.10.2 From 989e5b96e1af399296e2d1a34ca4a5aea1cf6d63 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 10 Oct 2006 14:47:44 -0700 Subject: [IPV6]: Seperate sit driver to extra module This patch removes the driver of the IPv6-in-IPv4 tunnel driver (sit) from the IPv6 module. It adds an option to Kconfig which makes it possible to compile it as a seperate module. Signed-off-by: Joerg Roedel Signed-off-by: David S. Miller diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index a460e81..ef5eaad 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -153,6 +153,19 @@ config INET6_XFRM_MODE_ROUTEOPTIMIZATION ---help--- Support for MIPv6 route optimization mode. +config IPV6_SIT + tristate "IPv6: IPv6-in-IPv4 tunnel (SIT driver)" + depends on IPV6 + default y + ---help--- + Tunneling means encapsulating data of one protocol type within + another protocol and sending it over a channel that understands the + encapsulating protocol. This driver implements encapsulation of IPv6 + into IPv4 packets. This is useful if you want to connect two IPv6 + networks over an IPv4-only path. + + Saying M here will produce a module called sit.ko. If unsure, say Y. + config IPV6_TUNNEL tristate "IPv6: IPv6-in-IPv6 tunnel" select INET6_TUNNEL diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 87274e4..addcc01 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -4,7 +4,7 @@ obj-$(CONFIG_IPV6) += ipv6.o -ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o sit.o \ +ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o raw.o \ protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ exthdrs.o sysctl_net_ipv6.o datagram.o proc.o \ @@ -29,6 +29,7 @@ obj-$(CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION) += xfrm6_mode_ro.o obj-$(CONFIG_INET6_XFRM_MODE_BEET) += xfrm6_mode_beet.o obj-$(CONFIG_NETFILTER) += netfilter/ +obj-$(CONFIG_IPV6_SIT) += sit.o obj-$(CONFIG_IPV6_TUNNEL) += ip6_tunnel.o obj-y += exthdrs_core.o diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index e94eccb..858cae2 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -850,7 +850,6 @@ static int __init inet6_init(void) err = addrconf_init(); if (err) goto addrconf_fail; - sit_init(); /* Init v6 extension headers. */ ipv6_rthdr_init(); @@ -927,7 +926,6 @@ static void __exit inet6_exit(void) mip6_fini(); #endif /* Cleanup code parts. */ - sit_cleanup(); ip6_flowlabel_cleanup(); addrconf_cleanup(); ip6_route_cleanup(); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 836eecd..dc5765b 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -850,3 +850,6 @@ int __init sit_init(void) inet_del_protocol(&sit_protocol, IPPROTO_IPV6); goto out; } + +module_init(sit_init); +module_exit(sit_cleanup); -- cgit v0.10.2 From 0be669bb37693103c15e64610454f8f431a38feb Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 10 Oct 2006 14:49:53 -0700 Subject: [IPV6]: Seperate sit driver to extra module (addrconf.c changes) This patch contains the changes to net/ipv6/addrconf.c to remove sit specific code if the sit driver is not selected. Signed-off-by: Joerg Roedel Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index e03c33b..b312a5f 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -396,8 +396,10 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) ndev->regen_timer.data = (unsigned long) ndev; if ((dev->flags&IFF_LOOPBACK) || dev->type == ARPHRD_TUNNEL || - dev->type == ARPHRD_NONE || - dev->type == ARPHRD_SIT) { +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) + dev->type == ARPHRD_SIT || +#endif + dev->type == ARPHRD_NONE) { printk(KERN_INFO "%s: Disabled Privacy Extensions\n", dev->name); @@ -1546,8 +1548,10 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev, This thing is done here expecting that the whole class of non-broadcast devices need not cloning. */ +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) if (dev->type == ARPHRD_SIT && (dev->flags & IFF_POINTOPOINT)) cfg.fc_flags |= RTF_NONEXTHOP; +#endif ip6_route_add(&cfg); } @@ -1569,6 +1573,7 @@ static void addrconf_add_mroute(struct net_device *dev) ip6_route_add(&cfg); } +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) static void sit_route_add(struct net_device *dev) { struct fib6_config cfg = { @@ -1582,6 +1587,7 @@ static void sit_route_add(struct net_device *dev) /* prefix length - 96 bits "::d.d.d.d" */ ip6_route_add(&cfg); } +#endif static void addrconf_add_lroute(struct net_device *dev) { @@ -1852,6 +1858,7 @@ int addrconf_set_dstaddr(void __user *arg) if (dev == NULL) goto err_exit; +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) if (dev->type == ARPHRD_SIT) { struct ifreq ifr; mm_segment_t oldfs; @@ -1881,6 +1888,7 @@ int addrconf_set_dstaddr(void __user *arg) err = dev_open(dev); } } +#endif err_exit: rtnl_unlock(); @@ -2010,6 +2018,7 @@ int addrconf_del_ifaddr(void __user *arg) return err; } +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) static void sit_add_v4_addrs(struct inet6_dev *idev) { struct inet6_ifaddr * ifp; @@ -2078,6 +2087,7 @@ static void sit_add_v4_addrs(struct inet6_dev *idev) } } } +#endif static void init_loopback(struct net_device *dev) { @@ -2141,6 +2151,7 @@ static void addrconf_dev_config(struct net_device *dev) addrconf_add_linklocal(idev, &addr); } +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) static void addrconf_sit_config(struct net_device *dev) { struct inet6_dev *idev; @@ -2166,6 +2177,7 @@ static void addrconf_sit_config(struct net_device *dev) } else sit_route_add(dev); } +#endif static inline int ipv6_inherit_linklocal(struct inet6_dev *idev, struct net_device *link_dev) @@ -2260,9 +2272,11 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, } switch(dev->type) { +#if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE) case ARPHRD_SIT: addrconf_sit_config(dev); break; +#endif case ARPHRD_TUNNEL6: addrconf_ip6_tnl_config(dev); break; -- cgit v0.10.2 From 4244f8a9f86a6d6e820b4cb53835c15c56d41aff Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 10 Oct 2006 19:40:50 -0700 Subject: [TCP]: Use TCPOLEN_TSTAMP_ALIGNED macro instead of magic number. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index c83938b..f745095 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -578,7 +578,7 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, struct tcphdr *th = skb->h.th; struct { struct tcphdr th; - u32 tsopt[3]; + u32 tsopt[TCPOLEN_TSTAMP_ALIGNED >> 2]; } rep; struct ip_reply_arg arg; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 3b65754..34e004b 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -653,7 +653,7 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 int tot_len = sizeof(struct tcphdr); if (ts) - tot_len += 3*4; + tot_len += TCPOLEN_TSTAMP_ALIGNED; buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len, GFP_ATOMIC); -- cgit v0.10.2 From 496c98dff8e353880299168d36fa082d6fba5237 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 10 Oct 2006 19:41:21 -0700 Subject: [NET]: Use hton{l,s}() for non-initializers. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index f5fba05..d5b5dec 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -611,8 +611,8 @@ static int ipgre_rcv(struct sk_buff *skb) * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header */ if (flags == 0 && - skb->protocol == __constant_htons(ETH_P_WCCP)) { - skb->protocol = __constant_htons(ETH_P_IP); + skb->protocol == htons(ETH_P_WCCP)) { + skb->protocol = htons(ETH_P_IP); if ((*(h + offset) & 0xF0) != 0x40) offset += 4; } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 9a253fa..f22536e 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -273,10 +273,10 @@ static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp, __u32 tstamp) { if (tp->rx_opt.tstamp_ok) { - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | - (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | - TCPOLEN_TIMESTAMP); + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_TIMESTAMP << 8) | + TCPOLEN_TIMESTAMP); *ptr++ = htonl(tstamp); *ptr++ = htonl(tp->rx_opt.ts_recent); } @@ -325,18 +325,27 @@ static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack, *ptr++ = htonl((TCPOPT_MSS << 24) | (TCPOLEN_MSS << 16) | mss); if (ts) { if(sack) - *ptr++ = __constant_htonl((TCPOPT_SACK_PERM << 24) | (TCPOLEN_SACK_PERM << 16) | - (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); + *ptr++ = htonl((TCPOPT_SACK_PERM << 24) | + (TCPOLEN_SACK_PERM << 16) | + (TCPOPT_TIMESTAMP << 8) | + TCPOLEN_TIMESTAMP); else - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_TIMESTAMP << 8) | + TCPOLEN_TIMESTAMP); *ptr++ = htonl(tstamp); /* TSVAL */ *ptr++ = htonl(ts_recent); /* TSECR */ } else if(sack) - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_SACK_PERM << 8) | TCPOLEN_SACK_PERM); + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_SACK_PERM << 8) | + TCPOLEN_SACK_PERM); if (offer_wscale) - *ptr++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_WINDOW << 16) | (TCPOLEN_WINDOW << 8) | (wscale)); + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_WINDOW << 16) | + (TCPOLEN_WINDOW << 8) | + (wscale)); } /* This routine actually transmits TCP packets queued in by -- cgit v0.10.2 From 9469c7b4aa210ce94c6e7208cfadbd0aca0ebe08 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 10 Oct 2006 19:41:46 -0700 Subject: [NET]: Use typesafe inet_twsk() inline function instead of cast. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index bf692c1..7e746c4 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -311,7 +311,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info) } if (sk->sk_state == DCCP_TIME_WAIT) { - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); return; } @@ -614,7 +614,7 @@ static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) bh_lock_sock(nsk); return nsk; } - inet_twsk_put((struct inet_timewait_sock *)nsk); + inet_twsk_put(inet_twsk(nsk)); return NULL; } @@ -980,7 +980,7 @@ discard_and_relse: goto discard_it; do_time_wait: - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); goto no_dccp_socket; } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 7a47399..7171a78 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -285,7 +285,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, } if (sk->sk_state == DCCP_TIME_WAIT) { - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); return; } @@ -663,7 +663,7 @@ static struct sock *dccp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) bh_lock_sock(nsk); return nsk; } - inet_twsk_put((struct inet_timewait_sock *)nsk); + inet_twsk_put(inet_twsk(nsk)); return NULL; } @@ -1109,7 +1109,7 @@ discard_and_relse: goto discard_it; do_time_wait: - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); goto no_dccp_socket; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index f745095..6bbd985 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -355,7 +355,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) return; } if (sk->sk_state == TCP_TIME_WAIT) { - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); return; } @@ -960,7 +960,7 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) bh_lock_sock(nsk); return nsk; } - inet_twsk_put((struct inet_timewait_sock *)nsk); + inet_twsk_put(inet_twsk(nsk)); return NULL; } @@ -1154,26 +1154,24 @@ discard_and_relse: do_time_wait: if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { - inet_twsk_put((struct inet_timewait_sock *) sk); + inet_twsk_put(inet_twsk(sk)); goto discard_it; } if (skb->len < (th->doff << 2) || tcp_checksum_complete(skb)) { TCP_INC_STATS_BH(TCP_MIB_INERRS); - inet_twsk_put((struct inet_timewait_sock *) sk); + inet_twsk_put(inet_twsk(sk)); goto discard_it; } - switch (tcp_timewait_state_process((struct inet_timewait_sock *)sk, - skb, th)) { + switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) { case TCP_TW_SYN: { struct sock *sk2 = inet_lookup_listener(&tcp_hashinfo, skb->nh.iph->daddr, th->dest, inet_iif(skb)); if (sk2) { - inet_twsk_deschedule((struct inet_timewait_sock *)sk, - &tcp_death_row); - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_deschedule(inet_twsk(sk), &tcp_death_row); + inet_twsk_put(inet_twsk(sk)); sk = sk2; goto process; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 34e004b..4c2a7c0 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -329,7 +329,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, } if (sk->sk_state == TCP_TIME_WAIT) { - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); return; } @@ -749,7 +749,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) bh_lock_sock(nsk); return nsk; } - inet_twsk_put((struct inet_timewait_sock *)nsk); + inet_twsk_put(inet_twsk(nsk)); return NULL; } @@ -1283,18 +1283,17 @@ discard_and_relse: do_time_wait: if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); goto discard_it; } if (skb->len < (th->doff<<2) || tcp_checksum_complete(skb)) { TCP_INC_STATS_BH(TCP_MIB_INERRS); - inet_twsk_put((struct inet_timewait_sock *)sk); + inet_twsk_put(inet_twsk(sk)); goto discard_it; } - switch (tcp_timewait_state_process((struct inet_timewait_sock *)sk, - skb, th)) { + switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) { case TCP_TW_SYN: { struct sock *sk2; -- cgit v0.10.2 From 42b6785eeb40fe3e9dab9981b6e3231a77c7c2f6 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 10 Oct 2006 19:42:09 -0700 Subject: [NET]: Introduce protocol-specific destructor for time-wait sockets. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index 6d14c22..5f48748 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -196,6 +196,7 @@ static inline void inet_twsk_put(struct inet_timewait_sock *tw) { if (atomic_dec_and_test(&tw->tw_refcnt)) { struct module *owner = tw->tw_prot->owner; + twsk_destructor((struct sock *)tw); #ifdef SOCK_REFCNT_DEBUG printk(KERN_DEBUG "%s timewait_sock %p released\n", tw->tw_prot->name, tw); diff --git a/include/net/timewait_sock.h b/include/net/timewait_sock.h index 2544281..be293d7 100644 --- a/include/net/timewait_sock.h +++ b/include/net/timewait_sock.h @@ -19,6 +19,7 @@ struct timewait_sock_ops { unsigned int twsk_obj_size; int (*twsk_unique)(struct sock *sk, struct sock *sktw, void *twp); + void (*twsk_destructor)(struct sock *sk); }; static inline int twsk_unique(struct sock *sk, struct sock *sktw, void *twp) @@ -28,4 +29,10 @@ static inline int twsk_unique(struct sock *sk, struct sock *sktw, void *twp) return 0; } +static inline void twsk_destructor(struct sock *sk) +{ + if (sk->sk_prot->twsk_prot->twsk_destructor != NULL) + sk->sk_prot->twsk_prot->twsk_destructor(sk); +} + #endif /* _TIMEWAIT_SOCK_H */ -- cgit v0.10.2 From 8238b218ec883adb44d710960a031c76105274cd Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 12 Oct 2006 00:49:15 -0700 Subject: [NET]: Do not memcmp() over pad bytes of struct flowi. They are not necessarily initialized to zero by the compiler, for example when using run-time initializers of automatic on-stack variables. Noticed by Eric Dumazet and Patrick McHardy. Signed-off-by: David S. Miller diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index dd0761e..a2a43d8 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -267,9 +267,14 @@ static void dn_dst_link_failure(struct sk_buff *skb) static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { - return memcmp(&fl1->nl_u.dn_u, &fl2->nl_u.dn_u, sizeof(fl1->nl_u.dn_u)) == 0 && - fl1->oif == fl2->oif && - fl1->iif == fl2->iif; + return ((fl1->nl_u.dn_u.daddr ^ fl2->nl_u.dn_u.daddr) | + (fl1->nl_u.dn_u.saddr ^ fl2->nl_u.dn_u.saddr) | +#ifdef CONFIG_IP_ROUTE_FWMARK + (fl1->nl_u.dn_u.fwmark ^ fl2->nl_u.dn_u.fwmark) | +#endif + (fl1->nl_u.dn_u.scope ^ fl2->nl_u.dn_u.scope) | + (fl1->oif ^ fl2->oif) | + (fl1->iif ^ fl2->iif)) == 0; } static int dn_insert_route(struct dn_route *rt, unsigned hash, struct dn_route **rp) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index c41ddba..925ee4d 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -566,9 +566,15 @@ static inline u32 rt_score(struct rtable *rt) static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { - return memcmp(&fl1->nl_u.ip4_u, &fl2->nl_u.ip4_u, sizeof(fl1->nl_u.ip4_u)) == 0 && - fl1->oif == fl2->oif && - fl1->iif == fl2->iif; + return ((fl1->nl_u.ip4_u.daddr ^ fl2->nl_u.ip4_u.daddr) | + (fl1->nl_u.ip4_u.saddr ^ fl2->nl_u.ip4_u.saddr) | +#ifdef CONFIG_IP_ROUTE_FWMARK + (fl1->nl_u.ip4_u.fwmark ^ fl2->nl_u.ip4_u.fwmark) | +#endif + (*(u16 *)&fl1->nl_u.ip4_u.tos ^ + *(u16 *)&fl2->nl_u.ip4_u.tos) | + (fl1->oif ^ fl2->oif) | + (fl1->iif ^ fl2->iif)) == 0; } #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED -- cgit v0.10.2 From 52c41a3224666d252d34597b580f1b6d4dc440e7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 01:48:20 -0700 Subject: [DECNET]: Fix sfuzz hanging on 2.6.18 Dave Jones wrote: > sfuzz D 724EF62A 2828 28717 28691 (NOTLB) > cd69fe98 00000082 0000012d 724ef62a 0001971a 00000010 00000007 df6d22b0 > dfd81080 725bbc5e 0001971a 000cc634 00000001 df6d23bc c140e260 00000202 > de1d5ba0 cd69fea0 de1d5ba0 00000000 00000000 de1d5b60 de1d5b8c de1d5ba0 > Call Trace: > [] lock_sock+0x75/0xa6 > [] dn_getname+0x18/0x5f [decnet] > [] sys_getsockname+0x5c/0xb0 > [] sys_socketcall+0xef/0x261 > [] syscall_call+0x7/0xb > DWARF2 unwinder stuck at syscall_call+0x7/0xb > > I wonder if the plethora of lockdep related changes inadvertantly broke something? Looks like unbalanced locking. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 70e0273..3456cd3 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -1178,8 +1178,10 @@ static int dn_getname(struct socket *sock, struct sockaddr *uaddr,int *uaddr_len if (peer) { if ((sock->state != SS_CONNECTED && sock->state != SS_CONNECTING) && - scp->accept_mode == ACC_IMMED) + scp->accept_mode == ACC_IMMED) { + release_sock(sk); return -ENOTCONN; + } memcpy(sa, &scp->peer, sizeof(struct sockaddr_dn)); } else { -- cgit v0.10.2 From b974179abef7cd680b80bd7c7042802bdd6f0eb6 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 01:50:30 -0700 Subject: [RTNETLINK]: Fix use of wrong skb in do_getlink() skb is the netlink query, nskb is the reply message. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 221e403..02f3c79 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -602,7 +602,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) goto errout; } - err = rtnl_unicast(skb, NETLINK_CB(skb).pid); + err = rtnl_unicast(nskb, NETLINK_CB(skb).pid); errout: kfree(iw_buf); dev_put(dev); -- cgit v0.10.2 From 30bdbe397bf58131a91fd836f60972442bed0544 Mon Sep 17 00:00:00 2001 From: Akinbou Mita Date: Thu, 12 Oct 2006 01:52:05 -0700 Subject: [PKT_SCHED] sch_htb: use rb_first() cleanup Use rb_first() to get first entry in rb tree. Signed-off-by: Akinbou Mita Signed-off-by: David S. Miller diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index bb3ddd4..9b9c555 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -786,11 +786,10 @@ static long htb_do_events(struct htb_sched *q, int level) for (i = 0; i < 500; i++) { struct htb_class *cl; long diff; - struct rb_node *p = q->wait_pq[level].rb_node; + struct rb_node *p = rb_first(&q->wait_pq[level]); + if (!p) return 0; - while (p->rb_left) - p = p->rb_left; cl = rb_entry(p, struct htb_class, pq_node); if (time_after(cl->pq_key, q->jiffies)) { -- cgit v0.10.2 From 2b1191af683d16a899c2b81b87b605841ceffdec Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 9 Oct 2006 13:04:35 +0200 Subject: [PATCH] elevator: elevator_type member not used elevator_type field in elevator_type structure is useless: it isn't used anywhere in kernel sources. Signed-off-by: Vasily Tarasov Signed-off-by: Jens Axboe diff --git a/include/linux/elevator.h b/include/linux/elevator.h index b3370ef..2fa9f11 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -70,7 +70,6 @@ struct elevator_type { struct list_head list; struct elevator_ops ops; - struct elevator_type *elevator_type; struct elv_fs_entry *elevator_attrs; char elevator_name[ELV_NAME_MAX]; struct module *elevator_owner; -- cgit v0.10.2 From a22b169df1b9f259391cf3b8ad8bfeea3d7be3f1 Mon Sep 17 00:00:00 2001 From: Vasily Tarasov Date: Wed, 11 Oct 2006 09:24:27 +0200 Subject: [PATCH] block layer: elevator_find function cleanup We can easily produce search through the elevator list without introducing additional elevator_type variable. Signed-off-by: Vasily Tarasov Signed-off-by: Jens Axboe diff --git a/block/elevator.c b/block/elevator.c index 487dd3d..d8030a8 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -93,21 +93,18 @@ static inline int elv_try_merge(struct request *__rq, struct bio *bio) static struct elevator_type *elevator_find(const char *name) { - struct elevator_type *e = NULL; + struct elevator_type *e; struct list_head *entry; list_for_each(entry, &elv_list) { - struct elevator_type *__e; - __e = list_entry(entry, struct elevator_type, list); + e = list_entry(entry, struct elevator_type, list); - if (!strcmp(__e->elevator_name, name)) { - e = __e; - break; - } + if (!strcmp(e->elevator_name, name)) + return e; } - return e; + return NULL; } static void elevator_put(struct elevator_type *e) -- cgit v0.10.2 From e6e80f294c2688302f41959c57acfee7e131e489 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 11 Oct 2006 10:03:09 +0200 Subject: [PATCH] splice: fix pipe_to_file() ->prepare_write() error path Don't jump to the unlock+release path, we already did that. Signed-off-by: Jens Axboe diff --git a/fs/splice.c b/fs/splice.c index 13e92dd..a567010 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -607,7 +607,7 @@ find_page: ret = -ENOMEM; page = page_cache_alloc_cold(mapping); if (unlikely(!page)) - goto out_nomem; + goto out_ret; /* * This will also lock the page @@ -666,7 +666,7 @@ find_page: if (sd->pos + this_len > isize) vmtruncate(mapping->host, isize); - goto out; + goto out_ret; } if (buf->page != page) { @@ -698,7 +698,7 @@ find_page: out: page_cache_release(page); unlock_page(page); -out_nomem: +out_ret: return ret; } -- cgit v0.10.2 From c5841642242e9ae817275e09b36b298456dc17d2 Mon Sep 17 00:00:00 2001 From: Vasily Tarasov Date: Wed, 11 Oct 2006 13:26:30 +0200 Subject: [PATCH] block layer: elv_iosched_show should get elv_list_lock elv_iosched_show function iterates other elv_list, hence elv_list_lock should be got. Signed-off-by: Vasily Tarasov Signed-off-by: Vasily Tarasov diff --git a/block/elevator.c b/block/elevator.c index d8030a8..8ccd163 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -1085,7 +1085,7 @@ ssize_t elv_iosched_show(request_queue_t *q, char *name) struct list_head *entry; int len = 0; - spin_lock_irq(q->queue_lock); + spin_lock_irq(&elv_list_lock); list_for_each(entry, &elv_list) { struct elevator_type *__e; @@ -1095,7 +1095,7 @@ ssize_t elv_iosched_show(request_queue_t *q, char *name) else len += sprintf(name+len, "%s ", __e->elevator_name); } - spin_unlock_irq(q->queue_lock); + spin_unlock_irq(&elv_list_lock); len += sprintf(len+name, "\n"); return len; -- cgit v0.10.2 From cea2885a2e989d1dc19af1fc991717b33b7d1456 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 12 Oct 2006 15:08:45 +0200 Subject: [PATCH] ide-cd: fix breakage with internally queued commands We still need to maintain a private PC style command, since it isn't completely unified with REQ_TYPE_BLOCK_PC yet. Signed-off-by: Jens Axboe diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 69bbb62..e7513e5 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -597,7 +597,7 @@ static void cdrom_prepare_request(ide_drive_t *drive, struct request *rq) struct cdrom_info *cd = drive->driver_data; ide_init_drive_cmd(rq); - rq->cmd_type = REQ_TYPE_BLOCK_PC; + rq->cmd_type = REQ_TYPE_ATA_PC; rq->rq_disk = cd->disk; } @@ -2023,7 +2023,8 @@ ide_do_rw_cdrom (ide_drive_t *drive, struct request *rq, sector_t block) } info->last_block = block; return action; - } else if (rq->cmd_type == REQ_TYPE_SENSE) { + } else if (rq->cmd_type == REQ_TYPE_SENSE || + rq->cmd_type == REQ_TYPE_ATA_PC) { return cdrom_do_packet_command(drive); } else if (blk_pc_request(rq)) { return cdrom_do_block_pc(drive, rq); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 26f7856..d370d2c 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -157,6 +157,7 @@ enum rq_cmd_type_bits { REQ_TYPE_ATA_CMD, REQ_TYPE_ATA_TASK, REQ_TYPE_ATA_TASKFILE, + REQ_TYPE_ATA_PC, }; /* -- cgit v0.10.2 From d58cdfb89ce0c6bd5f81ae931a984ef298dbda20 Mon Sep 17 00:00:00 2001 From: Vasily Tarasov Date: Thu, 12 Oct 2006 15:09:51 +0200 Subject: [PATCH] block layer: ioprio_best function fix Currently ioprio_best function first checks wethere aioprio or bioprio equals IOPRIO_CLASS_NONE (ioprio_valid() macros does that) and if it is so it returns bioprio/aioprio appropriately. Thus the next four lines, that set aclass/bclass to IOPRIO_CLASS_BE, if aclass/bclass == IOPRIO_CLASS_NONE, are never executed. The second problem: if aioprio from class IOPRIO_CLASS_NONE and bioprio from class IOPRIO_CLASS_IDLE are passed to ioprio_best function, it will return IOPRIO_CLASS_IDLE. It means that during __make_request we can merge two requests and set the priority of merged request to IDLE, while one of the initial requests originates from a process with NONE (default) priority. So we can get a situation when a process with default ioprio will experience IO starvation, while there is no process from real-time class in the system. Just removing ioprio_valid check should correct situation. Signed-off-by: Vasily Tarasov Signed-off-by: Jens Axboe diff --git a/fs/ioprio.c b/fs/ioprio.c index 6dc6721..89e8da1 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -150,11 +150,6 @@ int ioprio_best(unsigned short aprio, unsigned short bprio) unsigned short aclass = IOPRIO_PRIO_CLASS(aprio); unsigned short bclass = IOPRIO_PRIO_CLASS(bprio); - if (!ioprio_valid(aprio)) - return bprio; - if (!ioprio_valid(bprio)) - return aprio; - if (aclass == IOPRIO_CLASS_NONE) aclass = IOPRIO_CLASS_BE; if (bclass == IOPRIO_CLASS_NONE) -- cgit v0.10.2 From 994bd4f9f5a065ead4a92435fdd928ac7fd33809 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 11 Oct 2006 22:44:46 -0600 Subject: [PATCH] x86_64 irq: Properly update vector_irq This patch fixes my one line thinko where I was clearing the vector_irq entries on the wrong cpus. Signed-off-by: Eric W. Biederman Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index c3cdcab..44b55f8 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -660,7 +660,7 @@ next: } if (old_vector >= 0) { int old_cpu; - for_each_cpu_mask(old_cpu, domain) + for_each_cpu_mask(old_cpu, irq_domain[irq]) per_cpu(vector_irq, old_cpu)[old_vector] = -1; } for_each_cpu_mask(new_cpu, domain) -- cgit v0.10.2 From 230a03950ecd63bc613c6adeffbe9049189d9f05 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 12 Oct 2006 15:07:55 +0000 Subject: [CIFS] cifs Kconfig: don't select CONNECTOR `select' is a bit obnoxious: the option keeps on coming back and it's hard to work out what to do to make it go away again. The use of `depends on' is preferred (although it has usability problems too..) Signed-off-by: Andrew Morton Signed-off-by: Steve French diff --git a/fs/Kconfig b/fs/Kconfig index 5305816..6865a33 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1827,7 +1827,7 @@ config CIFS_EXPERIMENTAL config CIFS_UPCALL bool "Kerberos/SPNEGO advanced session setup (EXPERIMENTAL)" depends on CIFS_EXPERIMENTAL - select CONNECTOR + depends on CONNECTOR help Enables an upcall mechanism for CIFS which will be used to contact userspace helper utilities to provide SPNEGO packaged Kerberos -- cgit v0.10.2 From 8770c018da7bbaa3b41371abc401b2aa7e76a71a Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 12 Oct 2006 17:24:52 +0200 Subject: [PATCH] ide-cd: one more missing REQ_TYPE_CMD_ATA check Signed-off-by: Jens Axboe diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index e7513e5..bddfebd 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -716,7 +716,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret) ide_error(drive, "request sense failure", stat); return 1; - } else if (blk_pc_request(rq)) { + } else if (blk_pc_request(rq) || rq->cmd_type == REQ_TYPE_ATA_PC) { /* All other functions, except for READ. */ unsigned long flags; -- cgit v0.10.2 From d103e164bee2f21d0efe7d713cbbb0a443ba480d Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Oct 2006 17:49:24 +0000 Subject: [CIFS] Workaround incomplete byte length returned by some servers on small SMB responses Signed-off-by: Steve French diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 4a4fd2d..f1f8225 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -55,7 +55,7 @@ extern int SendReceiveBlockingLock(const unsigned int /* xid */ , struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , int * /* bytes returned */); -extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); +extern int checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length); extern int is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); extern int is_size_safe_to_change(struct cifsInodeInfo *); extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *); diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index ca6e9b1..bbc9cd3 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -418,26 +418,42 @@ checkSMBhdr(struct smb_hdr *smb, __u16 mid) } int -checkSMB(struct smb_hdr *smb, __u16 mid, int length) +checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length) { __u32 len = smb->smb_buf_length; __u32 clc_len; /* calculated length */ cFYI(0, ("checkSMB Length: 0x%x, smb_buf_length: 0x%x", length, len)); - if (((unsigned int)length < 2 + sizeof (struct smb_hdr)) || - (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4)) { - if ((unsigned int)length < 2 + sizeof (struct smb_hdr)) { - if (((unsigned int)length >= - sizeof (struct smb_hdr) - 1) + + if (length < 2 + sizeof (struct smb_hdr)) { + if ((length >= sizeof (struct smb_hdr) - 1) && (smb->Status.CifsError != 0)) { - smb->WordCount = 0; - /* some error cases do not return wct and bcc */ + smb->WordCount = 0; + /* some error cases do not return wct and bcc */ + return 0; + } else if ((length == sizeof(struct smb_hdr) + 1) && + (smb->WordCount == 0)) { + char * tmp = (char *)smb; + /* Need to work around a bug in two servers here */ + /* First, check if the part of bcc they sent was zero */ + if (tmp[sizeof(struct smb_hdr)] == 0) { + /* some servers return only half of bcc + * on simple responses (wct, bcc both zero) + * in particular have seen this on + * ulogoffX and FindClose. This leaves + * one byte of bcc potentially unitialized + */ + /* zero rest of bcc */ + tmp[sizeof(struct smb_hdr)+1] = 0; return 0; - } else { - cERROR(1, ("Length less than smb header size")); } + cERROR(1,("rcvd invalid byte count (bcc)")); + } else { + cERROR(1, ("Length less than smb header size")); } - if (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) - cERROR(1, ("smb length greater than MaxBufSize, mid=%d", + return 1; + } + if (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + cERROR(1, ("smb length greater than MaxBufSize, mid=%d", smb->Mid)); return 1; } @@ -446,7 +462,7 @@ checkSMB(struct smb_hdr *smb, __u16 mid, int length) return 1; clc_len = smbCalcSize_LE(smb); - if(4 + len != (unsigned int)length) { + if(4 + len != length) { cERROR(1, ("Length read does not match RFC1001 length %d",len)); return 1; } -- cgit v0.10.2 From 78b93f2db11df75ada5d04962ba631940b0a1398 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 12 Oct 2006 19:00:35 +0100 Subject: [PATCH] fixing includes in alpha_ksyms.c kernel_execve() fallout Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/arch/alpha/kernel/alpha_ksyms.c b/arch/alpha/kernel/alpha_ksyms.c index 692809e..e9762a3 100644 --- a/arch/alpha/kernel/alpha_ksyms.c +++ b/arch/alpha/kernel/alpha_ksyms.c @@ -12,7 +12,7 @@ #include #include -#include +#include /* these are C runtime functions with special calling conventions: */ extern void __divl (void); -- cgit v0.10.2 From 734c4c67393893f61b39bcdfe1e94f72111c56d6 Mon Sep 17 00:00:00 2001 From: Ravikiran Thirumalai Date: Thu, 12 Oct 2006 12:17:52 -0700 Subject: [PATCH] Fix build breakage with CONFIG_X86_VSMP Kernel build breaks with CONFIG_X86_VSMP. Probably due to some header file cleanups in 2.6.19-rc1. Signed-off-by: Ravikiran Thirumalai Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/vsmp.c b/arch/x86_64/kernel/vsmp.c index 044e852..414caf0 100644 --- a/arch/x86_64/kernel/vsmp.c +++ b/arch/x86_64/kernel/vsmp.c @@ -14,6 +14,7 @@ #include #include #include +#include static int __init vsmp_init(void) { -- cgit v0.10.2 From 2db02c0fe8519bd5985c6280896f4d719a6ae801 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 28 Sep 2006 09:35:27 +0200 Subject: [WATCHDOG] includes for sample watchdog program. Add missing includes to sample watchdog program. Signed-off-by: Jean Delvare Signed-off-by: Wim Van Sebroeck diff --git a/Documentation/watchdog/src/watchdog-simple.c b/Documentation/watchdog/src/watchdog-simple.c index 85cf17c..47801bc 100644 --- a/Documentation/watchdog/src/watchdog-simple.c +++ b/Documentation/watchdog/src/watchdog-simple.c @@ -1,4 +1,6 @@ +#include #include +#include #include int main(int argc, const char *argv[]) { -- cgit v0.10.2 From 1baaf0b424fe611a99cf3e2e59e84df0561d679a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 12 Oct 2006 19:07:59 +0100 Subject: [PATCH] more kernel_execve() fallout (sbus) drivers/sbus/char stuff using kernel_execve() needs linux/syscalls.h now; includes trimmed, while we are at it. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/sbus/char/bbc_envctrl.c b/drivers/sbus/char/bbc_envctrl.c index d27e4f6..0d3660c 100644 --- a/drivers/sbus/char/bbc_envctrl.c +++ b/drivers/sbus/char/bbc_envctrl.c @@ -4,10 +4,8 @@ * Copyright (C) 2001 David S. Miller (davem@redhat.com) */ -#include #include -#include -#include +#include #include #include #include diff --git a/drivers/sbus/char/envctrl.c b/drivers/sbus/char/envctrl.c index 728a133..6b6a855 100644 --- a/drivers/sbus/char/envctrl.c +++ b/drivers/sbus/char/envctrl.c @@ -20,16 +20,12 @@ */ #include -#include +#include #include -#include #include #include -#include #include -#include -#include -#include +#include #include #include -- cgit v0.10.2 From 733b736c91dd2c556f35dffdcf77e667cf10cefc Mon Sep 17 00:00:00 2001 From: Arnaud Patard Date: Thu, 12 Oct 2006 22:33:31 +0200 Subject: r8169: fix infinite loop during hotplug Bug reported for PCMCIA. Signed-off-by: Arnaud Patard Signed-off-by: Francois Romieu diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index c7309e9..c2c9a86 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -2702,6 +2702,7 @@ static void rtl8169_down(struct net_device *dev) struct rtl8169_private *tp = netdev_priv(dev); void __iomem *ioaddr = tp->mmio_addr; unsigned int poll_locked = 0; + unsigned int intrmask; rtl8169_delete_timer(dev); @@ -2740,8 +2741,11 @@ core_down: * 2) dev->change_mtu * -> rtl8169_poll can not be issued again and re-enable the * interruptions. Let's simply issue the IRQ down sequence again. + * + * No loop if hotpluged or major error (0xffff). */ - if (RTL_R16(IntrMask)) + intrmask = RTL_R16(IntrMask); + if (intrmask && (intrmask != 0xffff)) goto core_down; rtl8169_tx_clear(tp); -- cgit v0.10.2 From 52ae7b7935a079aaba25da98fe90772d04109f26 Mon Sep 17 00:00:00 2001 From: Russell Cattelan Date: Mon, 9 Oct 2006 12:11:54 -0500 Subject: [GFS2] Fix a size calculation error Fix a size calculation error. The size was incorrect being computed as a negative length and then being passed to an unsigned parameter. This in turn would cause the allocator to think it needed enough meta data to store a gigabyte file for every file created. Signed-off-by: Russell Cattelan Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 4fb743f..bdf56cf 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -370,15 +370,17 @@ static int gfs2_prepare_write(struct file *file, struct page *page, loff_t pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + from; loff_t end = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; struct gfs2_alloc *al; + unsigned int write_len = to - from; + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|GL_AOP, &ip->i_gh); error = gfs2_glock_nq_m_atime(1, &ip->i_gh); if (error) goto out_uninit; - gfs2_write_calc_reserv(ip, to - from, &data_blocks, &ind_blocks); + gfs2_write_calc_reserv(ip, write_len, &data_blocks, &ind_blocks); - error = gfs2_write_alloc_required(ip, pos, from - to, &alloc_required); + error = gfs2_write_alloc_required(ip, pos, write_len, &alloc_required); if (error) goto out_unlock; -- cgit v0.10.2 From f5c54804d9e3bb23d8924af09d9ca1c8de9560b6 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 10 Oct 2006 13:45:15 -0400 Subject: [GFS2] Fix uninitialised variable This fixes a bug where, in certain cases an uninitialised variable could cause a dereference of a NULL pointer in gfs2_commit_write(). Also a typo in a comment is fixed at the same time. Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index bdf56cf..99c9337 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -385,6 +385,7 @@ static int gfs2_prepare_write(struct file *file, struct page *page, goto out_unlock; + ip->i_alloc.al_requested = 0; if (alloc_required) { al = gfs2_alloc_get(ip); diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index 9eedfd1..b01e0cf 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -32,7 +32,7 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd); struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip); static inline void gfs2_alloc_put(struct gfs2_inode *ip) { - return; /* Se we can see where ip->i_alloc is used */ + return; /* So we can see where ip->i_alloc is used */ } int gfs2_inplace_reserve_i(struct gfs2_inode *ip, -- cgit v0.10.2 From 1ee48af22ed6dcddea8cdf93c7f2a268cbcf0d56 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 8 Oct 2006 04:30:48 +0200 Subject: [DLM] Kconfig: don't show an empty DLM menu Don't show an empty "Distributed Lock Manager" menu if IP_SCTP=n. Reported by Dmytro Bagrii in kernel Bugzilla #7268. Signed-off-by: Adrian Bunk Signed-off-by: David Teigland Signed-off-by: Patrick Caulfield Signed-off-by: Steven Whitehouse diff --git a/fs/dlm/Kconfig b/fs/dlm/Kconfig index 490f85b..81b2c64 100644 --- a/fs/dlm/Kconfig +++ b/fs/dlm/Kconfig @@ -1,10 +1,9 @@ menu "Distributed Lock Manager" - depends on INET && EXPERIMENTAL + depends on INET && IP_SCTP && EXPERIMENTAL config DLM tristate "Distributed Lock Manager (DLM)" depends on IPV6 || IPV6=n - depends on IP_SCTP select CONFIGFS_FS help A general purpose distributed lock manager for kernel or userspace -- cgit v0.10.2 From fe1a698ffef5af546dd4a8cd6a1f2f202491c4ef Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 11 Oct 2006 13:34:59 -0400 Subject: [GFS2] Fix bug where lock not held The log lock needs to be held when manipulating the counter for the number of free journal blocks. Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 554fe5b..72eec65 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -569,16 +569,15 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl) else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle) log_write_header(sdp, 0, PULL); lops_after_commit(sdp, ai); - sdp->sd_log_head = sdp->sd_log_flush_head; + gfs2_log_lock(sdp); + sdp->sd_log_head = sdp->sd_log_flush_head; sdp->sd_log_blks_free -= sdp->sd_log_num_hdrs; - sdp->sd_log_blks_reserved = 0; sdp->sd_log_commited_buf = 0; sdp->sd_log_num_hdrs = 0; sdp->sd_log_commited_revoke = 0; - gfs2_log_lock(sdp); if (!list_empty(&ai->ai_ail1_list)) { list_add(&ai->ai_list, &sdp->sd_ail1_list); ai = NULL; -- cgit v0.10.2 From c312c4fdc88514dd9522b7858eb879e610aeb9b1 Mon Sep 17 00:00:00 2001 From: Russell Cattelan Date: Thu, 12 Oct 2006 09:23:41 -0400 Subject: [GFS2] Pass the correct value to kunmap_atomic Pass kaddr rather than (incorrect) struct page to kunmap_atomic. Signed-off-by: Russell Cattelan Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 881e337..ab6d111 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -492,7 +492,7 @@ static int gfs2_check_magic(struct buffer_head *bh) ptr = kaddr + bh_offset(bh); if (*ptr == cpu_to_be32(GFS2_MAGIC)) rv = 1; - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); return rv; } @@ -626,7 +626,7 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp) memcpy(bh->b_data, kaddr + bh_offset(bd2->bd_bh), sdp->sd_sb.sb_bsize); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); *(__be32 *)bh->b_data = 0; } else { bh = gfs2_log_fake_buf(sdp, bd2->bd_bh); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 99c9337..e0599fe 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -162,7 +162,7 @@ static int zero_readpage(struct page *page) kaddr = kmap_atomic(page, KM_USER0); memset(kaddr, 0, PAGE_CACHE_SIZE); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); @@ -195,7 +195,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), ip->i_di.di_size); memset(kaddr + ip->i_di.di_size, 0, PAGE_CACHE_SIZE - ip->i_di.di_size); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); brelse(dibh); @@ -485,7 +485,7 @@ static int gfs2_commit_write(struct file *file, struct page *page, kaddr = kmap_atomic(page, KM_USER0); memcpy(dibh->b_data + sizeof(struct gfs2_dinode) + from, kaddr + from, to - from); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); -- cgit v0.10.2 From 4c5e1b1a8c3f591b21f09001d6748296ddff33b8 Mon Sep 17 00:00:00 2001 From: Patrick Caulfield Date: Thu, 12 Oct 2006 10:41:22 +0100 Subject: [DLM] fix iovec length in recvmsg The DLM always passes the iovec length as 1, this is wrong when the circular buffer wraps round. Signed-Off-By: Patrick Caulfield Signed-off-by: Steven Whitehouse diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 7bcea7c..867f93d 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -548,7 +548,7 @@ static int receive_from_sock(void) } len = iov[0].iov_len + iov[1].iov_len; - r = ret = kernel_recvmsg(sctp_con.sock, &msg, iov, 1, len, + r = ret = kernel_recvmsg(sctp_con.sock, &msg, iov, msg.msg_iovlen, len, MSG_NOSIGNAL | MSG_DONTWAIT); if (ret <= 0) goto out_close; -- cgit v0.10.2 From 370298e2e6f513bc4a9e9445eeed060d8c31f1ca Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Thu, 12 Oct 2006 15:40:05 -0400 Subject: [GFS2] Update git tree name/location The plan is to have two trees. One for bug fixes to be sent on a regular basis (-fixes) and another called -nmw which will contain items queued for the next merge window (hence the name). The -nmw tree will come & go according to need. Signed-off-by: Steven Whitehouse diff --git a/MAINTAINERS b/MAINTAINERS index 931e6e4..1b5430a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -905,7 +905,8 @@ P: David Teigland M: teigland@redhat.com L: cluster-devel@redhat.com W: http://sources.redhat.com/cluster/ -T: git kernel.org:/pub/scm/linux/kernel/git/steve/gfs-2.6.git +T: git kernel.org:/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes.git +T: git kernel.org:/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw.git S: Supported DAVICOM FAST ETHERNET (DMFE) NETWORK DRIVER @@ -1188,7 +1189,8 @@ P: Steven Whitehouse M: swhiteho@redhat.com L: cluster-devel@redhat.com W: http://sources.redhat.com/cluster/ -T: git kernel.org:/pub/scm/linux/kernel/git/steve/gfs-2.6.git +T: git kernel.org:/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes.git +T: git kernel.org:/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw.git S: Supported GIGASET ISDN DRIVERS -- cgit v0.10.2 From 1a4e15a04ec69cb3552f4120079f5472377df5f7 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 12 Oct 2006 21:33:51 +0000 Subject: [CIFS] Missing flags2 for DFS Partly suggested by Igor Mammedov Signed-off-by: Steve French diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 6f50f2b..5dc5a96 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -3675,6 +3675,14 @@ getDFSRetry: strncpy(pSMB->RequestFileName, searchName, name_len); } + if(ses->server) { + if(ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + pSMB->hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; + } + + pSMB->hdr.Uid = ses->Suid; + params = 2 /* level */ + name_len /*includes null */ ; pSMB->TotalDataCount = 0; pSMB->DataCount = 0; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index c96f3ed..1d176910 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3219,7 +3219,9 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, } /* else do not bother copying these informational fields */ } - if(smb_buffer_response->WordCount == 3) + if((smb_buffer_response->WordCount == 3) || + (smb_buffer_response->WordCount == 7)) + /* field is in same location */ tcon->Flags = le16_to_cpu(pSMBr->OptionalSupport); else tcon->Flags = 0; -- cgit v0.10.2 From c771746ef6ad64357897a90da42908d5c800a2c5 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Thu, 12 Oct 2006 22:21:16 -0500 Subject: [VOYAGER] fix genirq mess The implementation of genirq in x86 completely broke voyager (and presumably visws). Since it's plugged into so much of the x86 infrastructure, you can't expect it to work unconverted. This patch introduces a voyager IRQ handler type and switches voyager to the genirq infrastructure. Signed-off-by: James Bottomley diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c index d42422f..2e73f35 100644 --- a/arch/i386/mach-voyager/voyager_smp.c +++ b/arch/i386/mach-voyager/voyager_smp.c @@ -85,8 +85,8 @@ static int ack_QIC_CPI(__u8 cpi); static void ack_special_QIC_CPI(__u8 cpi); static void ack_VIC_CPI(__u8 cpi); static void send_CPI_allbutself(__u8 cpi); -static void enable_vic_irq(unsigned int irq); -static void disable_vic_irq(unsigned int irq); +static void mask_vic_irq(unsigned int irq); +static void unmask_vic_irq(unsigned int irq); static unsigned int startup_vic_irq(unsigned int irq); static void enable_local_vic_irq(unsigned int irq); static void disable_local_vic_irq(unsigned int irq); @@ -205,15 +205,12 @@ ack_CPI(__u8 cpi) /* The VIC IRQ descriptors -- these look almost identical to the * 8259 IRQs except that masks and things must be kept per processor */ -static struct hw_interrupt_type vic_irq_type = { - .typename = "VIC-level", - .startup = startup_vic_irq, - .shutdown = disable_vic_irq, - .enable = enable_vic_irq, - .disable = disable_vic_irq, - .ack = before_handle_vic_irq, - .end = after_handle_vic_irq, - .set_affinity = set_vic_irq_affinity, +static struct irq_chip vic_chip = { + .name = "VIC", + .startup = startup_vic_irq, + .mask = mask_vic_irq, + .unmask = unmask_vic_irq, + .set_affinity = set_vic_irq_affinity, }; /* used to count up as CPUs are brought on line (starts at 0) */ @@ -1397,6 +1394,17 @@ setup_profiling_timer(unsigned int multiplier) return 0; } +/* This is a bit of a mess, but forced on us by the genirq changes + * there's no genirq handler that really does what voyager wants + * so hack it up with the simple IRQ handler */ +static void fastcall +handle_vic_irq(unsigned int irq, struct irq_desc *desc) +{ + before_handle_vic_irq(irq); + handle_simple_irq(irq, desc); + after_handle_vic_irq(irq); +} + /* The CPIs are handled in the per cpu 8259s, so they must be * enabled to be received: FIX: enabling the CPIs in the early @@ -1433,7 +1441,7 @@ smp_intr_init(void) * This is for later: first 16 correspond to PC IRQs; next 16 * are Primary MC IRQs and final 16 are Secondary MC IRQs */ for(i = 0; i < 48; i++) - irq_desc[i].chip = &vic_irq_type; + set_irq_chip_and_handler(i, &vic_chip, handle_vic_irq); } /* send a CPI at level cpi to a set of cpus in cpuset (set 1 bit per @@ -1531,7 +1539,7 @@ ack_VIC_CPI(__u8 cpi) static unsigned int startup_vic_irq(unsigned int irq) { - enable_vic_irq(irq); + unmask_vic_irq(irq); return 0; } @@ -1558,7 +1566,7 @@ startup_vic_irq(unsigned int irq) * adjust their masks accordingly. */ static void -enable_vic_irq(unsigned int irq) +unmask_vic_irq(unsigned int irq) { /* linux doesn't to processor-irq affinity, so enable on * all CPUs we know about */ @@ -1567,7 +1575,7 @@ enable_vic_irq(unsigned int irq) __u32 processorList = 0; unsigned long flags; - VDEBUG(("VOYAGER: enable_vic_irq(%d) CPU%d affinity 0x%lx\n", + VDEBUG(("VOYAGER: unmask_vic_irq(%d) CPU%d affinity 0x%lx\n", irq, cpu, cpu_irq_affinity[cpu])); spin_lock_irqsave(&vic_irq_lock, flags); for_each_online_cpu(real_cpu) { @@ -1591,7 +1599,7 @@ enable_vic_irq(unsigned int irq) } static void -disable_vic_irq(unsigned int irq) +mask_vic_irq(unsigned int irq) { /* lazy disable, do nothing */ } @@ -1819,7 +1827,7 @@ set_vic_irq_affinity(unsigned int irq, cpumask_t mask) * disabled again as it comes in (voyager lazy disable). If * the affinity map is tightened to disable the interrupt on a * cpu, it will be pushed off when it comes in */ - enable_vic_irq(irq); + unmask_vic_irq(irq); } static void -- cgit v0.10.2 From 58f07943b0ef1e59cbf9a45cdc727048d224637f Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Thu, 12 Oct 2006 22:23:18 -0500 Subject: [VOYAGER] fix up attribute packed specifiers in voyager.h The old style (attribute on each structure entry) never really worked. Move it to an attribute per structure Signed-off-by: James Bottomley diff --git a/include/asm-i386/voyager.h b/include/asm-i386/voyager.h index e74c54aa..5b27838 100644 --- a/include/asm-i386/voyager.h +++ b/include/asm-i386/voyager.h @@ -118,33 +118,33 @@ typedef struct voyager_module { } voyager_module_t; typedef struct voyager_eeprom_hdr { - __u8 module_id[4] __attribute__((packed)); - __u8 version_id __attribute__((packed)); - __u8 config_id __attribute__((packed)); - __u16 boundry_id __attribute__((packed)); /* boundary scan id */ - __u16 ee_size __attribute__((packed)); /* size of EEPROM */ - __u8 assembly[11] __attribute__((packed)); /* assembly # */ - __u8 assembly_rev __attribute__((packed)); /* assembly rev */ - __u8 tracer[4] __attribute__((packed)); /* tracer number */ - __u16 assembly_cksum __attribute__((packed)); /* asm checksum */ - __u16 power_consump __attribute__((packed)); /* pwr requirements */ - __u16 num_asics __attribute__((packed)); /* number of asics */ - __u16 bist_time __attribute__((packed)); /* min. bist time */ - __u16 err_log_offset __attribute__((packed)); /* error log offset */ - __u16 scan_path_offset __attribute__((packed));/* scan path offset */ - __u16 cct_offset __attribute__((packed)); - __u16 log_length __attribute__((packed)); /* length of err log */ - __u16 xsum_end __attribute__((packed)); /* offset to end of + __u8 module_id[4]; + __u8 version_id; + __u8 config_id; + __u16 boundry_id; /* boundary scan id */ + __u16 ee_size; /* size of EEPROM */ + __u8 assembly[11]; /* assembly # */ + __u8 assembly_rev; /* assembly rev */ + __u8 tracer[4]; /* tracer number */ + __u16 assembly_cksum; /* asm checksum */ + __u16 power_consump; /* pwr requirements */ + __u16 num_asics; /* number of asics */ + __u16 bist_time; /* min. bist time */ + __u16 err_log_offset; /* error log offset */ + __u16 scan_path_offset;/* scan path offset */ + __u16 cct_offset; + __u16 log_length; /* length of err log */ + __u16 xsum_end; /* offset to end of checksum */ - __u8 reserved[4] __attribute__((packed)); - __u8 sflag __attribute__((packed)); /* starting sentinal */ - __u8 part_number[13] __attribute__((packed)); /* prom part number */ - __u8 version[10] __attribute__((packed)); /* version number */ - __u8 signature[8] __attribute__((packed)); - __u16 eeprom_chksum __attribute__((packed)); - __u32 data_stamp_offset __attribute__((packed)); - __u8 eflag __attribute__((packed)); /* ending sentinal */ -} voyager_eprom_hdr_t; + __u8 reserved[4]; + __u8 sflag; /* starting sentinal */ + __u8 part_number[13]; /* prom part number */ + __u8 version[10]; /* version number */ + __u8 signature[8]; + __u16 eeprom_chksum; + __u32 data_stamp_offset; + __u8 eflag ; /* ending sentinal */ +} __attribute__((packed)) voyager_eprom_hdr_t; @@ -155,30 +155,30 @@ typedef struct voyager_eeprom_hdr { * in the module EPROMs. We really only care about the IDs and * offsets */ typedef struct voyager_sp_table { - __u8 asic_id __attribute__((packed)); - __u8 bypass_flag __attribute__((packed)); - __u16 asic_data_offset __attribute__((packed)); - __u16 config_data_offset __attribute__((packed)); -} voyager_sp_table_t; + __u8 asic_id; + __u8 bypass_flag; + __u16 asic_data_offset; + __u16 config_data_offset; +} __attribute__((packed)) voyager_sp_table_t; typedef struct voyager_jtag_table { - __u8 icode[4] __attribute__((packed)); - __u8 runbist[4] __attribute__((packed)); - __u8 intest[4] __attribute__((packed)); - __u8 samp_preld[4] __attribute__((packed)); - __u8 ireg_len __attribute__((packed)); -} voyager_jtt_t; + __u8 icode[4]; + __u8 runbist[4]; + __u8 intest[4]; + __u8 samp_preld[4]; + __u8 ireg_len; +} __attribute__((packed)) voyager_jtt_t; typedef struct voyager_asic_data_table { - __u8 jtag_id[4] __attribute__((packed)); - __u16 length_bsr __attribute__((packed)); - __u16 length_bist_reg __attribute__((packed)); - __u32 bist_clk __attribute__((packed)); - __u16 subaddr_bits __attribute__((packed)); - __u16 seed_bits __attribute__((packed)); - __u16 sig_bits __attribute__((packed)); - __u16 jtag_offset __attribute__((packed)); -} voyager_at_t; + __u8 jtag_id[4]; + __u16 length_bsr; + __u16 length_bist_reg; + __u32 bist_clk; + __u16 subaddr_bits; + __u16 seed_bits; + __u16 sig_bits; + __u16 jtag_offset; +} __attribute__((packed)) voyager_at_t; /* Voyager Interrupt Controller (VIC) registers */ @@ -328,52 +328,52 @@ struct voyager_bios_info { #define NUMBER_OF_POS_REGS 8 typedef struct { - __u8 MC_Slot __attribute__((packed)); - __u8 POS_Values[NUMBER_OF_POS_REGS] __attribute__((packed)); -} MC_SlotInformation_t; + __u8 MC_Slot; + __u8 POS_Values[NUMBER_OF_POS_REGS]; +} __attribute__((packed)) MC_SlotInformation_t; struct QuadDescription { - __u8 Type __attribute__((packed)); /* for type 0 (DYADIC or MONADIC) all fields + __u8 Type; /* for type 0 (DYADIC or MONADIC) all fields * will be zero except for slot */ - __u8 StructureVersion __attribute__((packed)); - __u32 CPI_BaseAddress __attribute__((packed)); - __u32 LARC_BankSize __attribute__((packed)); - __u32 LocalMemoryStateBits __attribute__((packed)); - __u8 Slot __attribute__((packed)); /* Processor slots 1 - 4 */ -}; + __u8 StructureVersion; + __u32 CPI_BaseAddress; + __u32 LARC_BankSize; + __u32 LocalMemoryStateBits; + __u8 Slot; /* Processor slots 1 - 4 */ +} __attribute__((packed)); struct ProcBoardInfo { - __u8 Type __attribute__((packed)); - __u8 StructureVersion __attribute__((packed)); - __u8 NumberOfBoards __attribute__((packed)); - struct QuadDescription QuadData[MAX_PROCESSOR_BOARDS] __attribute__((packed)); -}; + __u8 Type; + __u8 StructureVersion; + __u8 NumberOfBoards; + struct QuadDescription QuadData[MAX_PROCESSOR_BOARDS]; +} __attribute__((packed)); struct CacheDescription { - __u8 Level __attribute__((packed)); - __u32 TotalSize __attribute__((packed)); - __u16 LineSize __attribute__((packed)); - __u8 Associativity __attribute__((packed)); - __u8 CacheType __attribute__((packed)); - __u8 WriteType __attribute__((packed)); - __u8 Number_CPUs_SharedBy __attribute__((packed)); - __u8 Shared_CPUs_Hardware_IDs[MAX_SHARED_CPUS] __attribute__((packed)); + __u8 Level; + __u32 TotalSize; + __u16 LineSize; + __u8 Associativity; + __u8 CacheType; + __u8 WriteType; + __u8 Number_CPUs_SharedBy; + __u8 Shared_CPUs_Hardware_IDs[MAX_SHARED_CPUS]; -}; +} __attribute__((packed)); struct CPU_Description { - __u8 CPU_HardwareId __attribute__((packed)); - char *FRU_String __attribute__((packed)); - __u8 NumberOfCacheLevels __attribute__((packed)); - struct CacheDescription CacheLevelData[MAX_CACHE_LEVELS] __attribute__((packed)); -}; + __u8 CPU_HardwareId; + char *FRU_String; + __u8 NumberOfCacheLevels; + struct CacheDescription CacheLevelData[MAX_CACHE_LEVELS]; +} __attribute__((packed)); struct CPU_Info { - __u8 Type __attribute__((packed)); - __u8 StructureVersion __attribute__((packed)); - __u8 NumberOf_CPUs __attribute__((packed)); - struct CPU_Description CPU_Data[MAX_CPUS] __attribute__((packed)); -}; + __u8 Type; + __u8 StructureVersion; + __u8 NumberOf_CPUs; + struct CPU_Description CPU_Data[MAX_CPUS]; +} __attribute__((packed)); /* -- cgit v0.10.2 From 81c06b10bcd4c7e8c88b4b425c55402b1d65fd0e Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Thu, 12 Oct 2006 22:25:03 -0500 Subject: [VOYAGER] fix up ptregs removal mess Apparently whoever converted voyager never actually checked that the patch would compile ... Remove as much of the pt_regs references as possible and move the remaining ones into line with what's in x86 generic. Signed-off-by: James Bottomley diff --git a/arch/i386/mach-voyager/voyager_basic.c b/arch/i386/mach-voyager/voyager_basic.c index c639d30..8fe7e45 100644 --- a/arch/i386/mach-voyager/voyager_basic.c +++ b/arch/i386/mach-voyager/voyager_basic.c @@ -44,7 +44,7 @@ struct voyager_SUS *voyager_SUS = NULL; #ifdef CONFIG_SMP static void -voyager_dump(int dummy1, struct pt_regs *dummy2, struct tty_struct *dummy3) +voyager_dump(int dummy1, struct tty_struct *dummy3) { /* get here via a sysrq */ voyager_smp_dump(); @@ -166,7 +166,7 @@ voyager_memory_detect(int region, __u32 *start, __u32 *length) * off the timer tick to the SMP code, since the VIC doesn't have an * internal timer (The QIC does, but that's another story). */ void -voyager_timer_interrupt(struct pt_regs *regs) +voyager_timer_interrupt(void) { if((jiffies & 0x3ff) == 0) { @@ -202,7 +202,7 @@ voyager_timer_interrupt(struct pt_regs *regs) } } #ifdef CONFIG_SMP - smp_vic_timer_interrupt(regs); + smp_vic_timer_interrupt(); #endif } diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c index 2e73f35..f3fea2a 100644 --- a/arch/i386/mach-voyager/voyager_smp.c +++ b/arch/i386/mach-voyager/voyager_smp.c @@ -1141,9 +1141,9 @@ smp_apic_timer_interrupt(struct pt_regs *regs) fastcall void smp_qic_timer_interrupt(struct pt_regs *regs) { - ack_QIC_CPI(QIC_TIMER_CPI); struct pt_regs *old_regs = set_irq_regs(regs); - wrapper_smp_local_timer_interrupt(void); + ack_QIC_CPI(QIC_TIMER_CPI); + wrapper_smp_local_timer_interrupt(); set_irq_regs(old_regs); } @@ -1267,12 +1267,10 @@ smp_send_stop(void) /* this function is triggered in time.c when a clock tick fires * we need to re-broadcast the tick to all CPUs */ void -smp_vic_timer_interrupt(struct pt_regs *regs) +smp_vic_timer_interrupt(void) { - struct pt_regs *old_regs = set_irq_regs(regs); send_CPI_allbutself(VIC_TIMER_CPI); smp_local_timer_interrupt(); - set_irq_regs(old_regs); } /* local (per CPU) timer interrupt. It does both profiling and @@ -1307,7 +1305,7 @@ smp_local_timer_interrupt(void) per_cpu(prof_counter, cpu); } - update_process_times(user_mode_vm(irq_regs)); + update_process_times(user_mode_vm(get_irq_regs())); } if( ((1< Date: Fri, 13 Oct 2006 04:13:16 +0200 Subject: [PATCH] Get core dump code to work... The file based core dump code was broken by pipe changes - a relative llseek returns the absolute file position on success, not the relative one, so dump_seek() always failed when invoked with non-zero current position. Only success/failure can be tested with relative lseek, we have to trust kernel that on success we've got right file offset. With this fix in place I have finally real core files instead of 1KB fragments... Signed-off-by: Petr Vandrovec [ Cleaned it up a bit while here - use SEEK_CUR instead of hardcoding 1 ] Signed-off-by: Linus Torvalds diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 06435f36..5610061 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1152,7 +1152,7 @@ static int dump_write(struct file *file, const void *addr, int nr) static int dump_seek(struct file *file, loff_t off) { if (file->f_op->llseek && file->f_op->llseek != no_llseek) { - if (file->f_op->llseek(file, off, 1) != off) + if (file->f_op->llseek(file, off, SEEK_CUR) < 0) return 0; } else { char *buf = (char *)get_zeroed_page(GFP_KERNEL); -- cgit v0.10.2 From 99a10a60ba9bedcf5d70ef81414d3e03816afa3f Mon Sep 17 00:00:00 2001 From: Franck Bui-Huu Date: Thu, 12 Oct 2006 21:06:33 +0200 Subject: [PATCH] Fix up mmap_kmem vma->vm_pgoff is an pfn _offset_ relatif to the begining of the memory start. The previous code was doing at first: vma->vm_pgoff << PAGE_SHIFT which results into a wrong physical address since some platforms have a physical mem start that can be different from 0. After that the previous call __pa() on this wrong physical address, however __pa() is used to convert a _virtual_ address into a physical one. This patch rewrites this convertion. It calculates the pfn of PAGE_OFFSET which is the pfn of the mem start then it adds the vma->vm_pgoff to it. It also uses virt_to_phys() instead of __pa() since the latter shouldn't be used by drivers. Signed-off-by: Franck Bui-Huu Signed-off-by: Linus Torvalds diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 6511012..a89cb52 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -292,8 +292,8 @@ static int mmap_kmem(struct file * file, struct vm_area_struct * vma) { unsigned long pfn; - /* Turn a kernel-virtual address into a physical page frame */ - pfn = __pa((u64)vma->vm_pgoff << PAGE_SHIFT) >> PAGE_SHIFT; + /* Turn a pfn offset into an absolute pfn */ + pfn = PFN_DOWN(virt_to_phys((void *)PAGE_OFFSET)) + vma->vm_pgoff; /* * RED-PEN: on some architectures there is more mapped memory -- cgit v0.10.2 From 1b95817dd76084c10cb8dc899f429bcc85454741 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 12 Oct 2006 19:10:04 +0100 Subject: [PATCH] uml shouldn't do HEADERS_CHECK The lack of asm-um/Kbuild is deliberate. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 8fd2dbf..77491e3 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -373,6 +373,7 @@ config FORCED_INLINING config HEADERS_CHECK bool "Run 'make headers_check' when building vmlinux" + depends on !UML help This option will extract the user-visible kernel headers whenever building the kernel, and will run basic sanity checks on them to -- cgit v0.10.2 From 6569345abbcdbd1e13915a3bc5033c98817d0ec6 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 12 Oct 2006 11:01:30 -0700 Subject: [PATCH] thermal throttle: sysfs error checking Get rid of warning in the thermal throttling code about not checking sysfs return values. Signed-off-by: Stephen Hemminger Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c index 4f43047..2d8703b 100644 --- a/arch/i386/kernel/cpu/mcheck/therm_throt.c +++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c @@ -110,17 +110,15 @@ int therm_throt_process(int curr) #ifdef CONFIG_SYSFS /* Add/Remove thermal_throttle interface for CPU device */ -static __cpuinit int thermal_throttle_add_dev(struct sys_device * sys_dev) +static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) { - sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group); - return 0; + return sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group); } #ifdef CONFIG_HOTPLUG_CPU -static __cpuinit int thermal_throttle_remove_dev(struct sys_device * sys_dev) +static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev) { - sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group); - return 0; + return sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group); } /* Mutex protecting device creation against CPU hotplug */ @@ -133,12 +131,14 @@ static __cpuinit int thermal_throttle_cpu_callback(struct notifier_block *nfb, { unsigned int cpu = (unsigned long)hcpu; struct sys_device *sys_dev; + int err; sys_dev = get_cpu_sysdev(cpu); mutex_lock(&therm_cpu_lock); switch (action) { case CPU_ONLINE: - thermal_throttle_add_dev(sys_dev); + err = thermal_throttle_add_dev(sys_dev); + WARN_ON(err); break; case CPU_DEAD: thermal_throttle_remove_dev(sys_dev); @@ -157,6 +157,7 @@ static struct notifier_block thermal_throttle_cpu_notifier = static __init int thermal_throttle_init_device(void) { unsigned int cpu = 0; + int err; if (!atomic_read(&therm_throt_en)) return 0; @@ -167,8 +168,10 @@ static __init int thermal_throttle_init_device(void) mutex_lock(&therm_cpu_lock); #endif /* connect live CPUs to sysfs */ - for_each_online_cpu(cpu) - thermal_throttle_add_dev(get_cpu_sysdev(cpu)); + for_each_online_cpu(cpu) { + err = thermal_throttle_add_dev(get_cpu_sysdev(cpu)); + WARN_ON(err); + } #ifdef CONFIG_HOTPLUG_CPU mutex_unlock(&therm_cpu_lock); #endif -- cgit v0.10.2 From e0fafda36a2fc5ecf8d11771f6c01c523b2a1fd8 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 12 Oct 2006 23:18:53 +0200 Subject: [PATCH] m68knommu: sync syscalls with m68k m68knommu: sync syscalls with m68k Signed-Off-By: Geert Uytterhoeven Signed-Off-By: Greg Ungerer Signed-off-by: Linus Torvalds diff --git a/arch/m68knommu/kernel/syscalltable.S b/arch/m68knommu/kernel/syscalltable.S index 617e43e..4603f4f 100644 --- a/arch/m68knommu/kernel/syscalltable.S +++ b/arch/m68knommu/kernel/syscalltable.S @@ -296,10 +296,39 @@ ENTRY(sys_call_table) .long sys_mq_notify /* 275 */ .long sys_mq_getsetattr .long sys_waitid - .long sys_ni_syscall /* sys_setaltroot */ - .long sys_ni_syscall /* sys_add_key */ - .long sys_ni_syscall /* 280 */ /* sys_request_key */ - .long sys_ni_syscall /* sys_keyctl */ + .long sys_ni_syscall /* for sys_vserver */ + .long sys_add_key + .long sys_request_key /* 280 */ + .long sys_keyctl + .long sys_ioprio_set + .long sys_ioprio_get + .long sys_inotify_init + .long sys_inotify_add_watch /* 285 */ + .long sys_inotify_rm_watch + .long sys_migrate_pages + .long sys_openat + .long sys_mkdirat + .long sys_mknodat /* 290 */ + .long sys_fchownat + .long sys_futimesat + .long sys_fstatat64 + .long sys_unlinkat + .long sys_renameat /* 295 */ + .long sys_linkat + .long sys_symlinkat + .long sys_readlinkat + .long sys_fchmodat + .long sys_faccessat /* 300 */ + .long sys_ni_syscall /* Reserved for pselect6 */ + .long sys_ni_syscall /* Reserved for ppoll */ + .long sys_unshare + .long sys_set_robust_list + .long sys_get_robust_list /* 305 */ + .long sys_splice + .long sys_sync_file_range + .long sys_tee + .long sys_vmsplice + .long sys_move_pages /* 310 */ .rept NR_syscalls-(.-sys_call_table)/4 .long sys_ni_syscall diff --git a/include/asm-m68knommu/unistd.h b/include/asm-m68knommu/unistd.h index daafb5d..ebaf031 100644 --- a/include/asm-m68knommu/unistd.h +++ b/include/asm-m68knommu/unistd.h @@ -281,14 +281,43 @@ #define __NR_mq_notify 275 #define __NR_mq_getsetattr 276 #define __NR_waitid 277 -#define __NR_sys_setaltroot 278 +#define __NR_vserver 278 #define __NR_add_key 279 #define __NR_request_key 280 #define __NR_keyctl 281 - +#define __NR_ioprio_set 282 +#define __NR_ioprio_get 283 +#define __NR_inotify_init 284 +#define __NR_inotify_add_watch 285 +#define __NR_inotify_rm_watch 286 +#define __NR_migrate_pages 287 +#define __NR_openat 288 +#define __NR_mkdirat 289 +#define __NR_mknodat 290 +#define __NR_fchownat 291 +#define __NR_futimesat 292 +#define __NR_fstatat64 293 +#define __NR_unlinkat 294 +#define __NR_renameat 295 +#define __NR_linkat 296 +#define __NR_symlinkat 297 +#define __NR_readlinkat 298 +#define __NR_fchmodat 299 +#define __NR_faccessat 300 +#define __NR_pselect6 301 +#define __NR_ppoll 302 +#define __NR_unshare 303 +#define __NR_set_robust_list 304 +#define __NR_get_robust_list 305 +#define __NR_splice 306 +#define __NR_sync_file_range 307 +#define __NR_tee 308 +#define __NR_vmsplice 309 +#define __NR_move_pages 310 + #ifdef __KERNEL__ -#define NR_syscalls 282 +#define NR_syscalls 311 #include /* user-visible error numbers are in the range -1 - -MAX_ERRNO: see -- cgit v0.10.2 From 0e7af8d04ecb4f6ba8cd1f731f036a004ad0e174 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 13 Oct 2006 16:04:23 +0100 Subject: [PATCH] Fix headers_check for O= builds; disable automatic check on UML. * make header_check work with O= Signed-off-by: David Woodhouse Signed-off-by: Linus Torvalds diff --git a/Makefile b/Makefile index f242829..eebf310 100644 --- a/Makefile +++ b/Makefile @@ -742,7 +742,7 @@ endif # ifdef CONFIG_KALLSYMS # vmlinux image - including updated kernel symbols vmlinux: $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) $(kallsyms.o) FORCE ifdef CONFIG_HEADERS_CHECK - $(Q)$(MAKE) headers_check + $(Q)$(MAKE) -f $(srctree)/Makefile headers_check endif $(call if_changed_rule,vmlinux__) $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost $@ @@ -935,7 +935,7 @@ headers_install_all: include/linux/version.h scripts_basic FORCE PHONY += headers_install headers_install: include/linux/version.h scripts_basic FORCE - @if [ ! -r include/asm-$(ARCH)/Kbuild ]; then \ + @if [ ! -r $(srctree)/include/asm-$(ARCH)/Kbuild ]; then \ echo '*** Error: Headers not exportable for this architecture ($(ARCH))'; \ exit 1 ; fi $(Q)$(MAKE) $(build)=scripts scripts/unifdef -- cgit v0.10.2 From b8a3ad5b53918787f4708ad9dfe90d2557cc78dd Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 13 Oct 2006 08:42:10 -0700 Subject: Include proper header file for PFN_DOWN() The recent commit (99a10a60ba9bedcf5d70ef81414d3e03816afa3f) to fix up mmap_kmem() broke compiles because it used PFN_DOWN() without including . Signed-off-by: Linus Torvalds diff --git a/drivers/char/mem.c b/drivers/char/mem.c index a89cb52..5547337 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include -- cgit v0.10.2 From b4bd8c66435a8cdf8c90334fb3b517a23ff2ab95 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 13 Oct 2006 09:25:04 -0700 Subject: Linux 2.6.19-rc2 That was slightly more painful than really necessary.. diff --git a/Makefile b/Makefile index eebf310..703d40a 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 19 -EXTRAVERSION =-rc1 +EXTRAVERSION =-rc2 NAME=Avast! A bilge rat! # *DOCUMENTATION* -- cgit v0.10.2 From 8f6cff98477edbcd8ae4976734ba7edd07bdd244 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Fri, 13 Oct 2006 12:42:36 -0500 Subject: JFS: pageno needs to be long diRead and diWrite are representing the page number as an unsigned int. This causes file system corruption on volumes larger than 16TB. Signed-off-by: Dave Kleikamp diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 489a3d6..ee9b473 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -318,7 +318,7 @@ int diRead(struct inode *ip) struct inomap *imap; int block_offset; int inodes_left; - uint pageno; + unsigned long pageno; int rel_inode; jfs_info("diRead: ino = %ld", ip->i_ino); @@ -606,7 +606,7 @@ int diWrite(tid_t tid, struct inode *ip) int block_offset; int inodes_left; struct metapage *mp; - uint pageno; + unsigned long pageno; int rel_inode; int dioffset; struct inode *ipimap; -- cgit v0.10.2 From bdcff3458f5448fac585a6174ad9342f361b5135 Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 26 Sep 2006 17:49:30 +0200 Subject: [WATCHDOG] Atmel AT91RM9200 rename. The new Atmel AT91SAM9261 and AT91SAM9260 processors use a different internal watchdog peripheral. This watchdog driver is therefore AT91RM9200-specific. This patch renames at91_wdt.c to at91rm9200_wdt.c, and changes the name of the configuration option. Signed-off-by: Andrew Victor Signed-off-by: Wim Van Sebroeck diff --git a/arch/arm/configs/at91rm9200dk_defconfig b/arch/arm/configs/at91rm9200dk_defconfig index c82e466..b430414 100644 --- a/arch/arm/configs/at91rm9200dk_defconfig +++ b/arch/arm/configs/at91rm9200dk_defconfig @@ -577,7 +577,7 @@ CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers # # CONFIG_SOFT_WATCHDOG is not set -CONFIG_AT91_WATCHDOG=y +CONFIG_AT91RM9200_WATCHDOG=y # # USB-based Watchdog Cards diff --git a/arch/arm/configs/at91rm9200ek_defconfig b/arch/arm/configs/at91rm9200ek_defconfig index b983fc5..d96fc83 100644 --- a/arch/arm/configs/at91rm9200ek_defconfig +++ b/arch/arm/configs/at91rm9200ek_defconfig @@ -558,7 +558,7 @@ CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers # # CONFIG_SOFT_WATCHDOG is not set -CONFIG_AT91_WATCHDOG=y +CONFIG_AT91RM9200_WATCHDOG=y # # USB-based Watchdog Cards diff --git a/arch/arm/configs/csb337_defconfig b/arch/arm/configs/csb337_defconfig index a2d6fd3..20e6825 100644 --- a/arch/arm/configs/csb337_defconfig +++ b/arch/arm/configs/csb337_defconfig @@ -615,7 +615,7 @@ CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers # # CONFIG_SOFT_WATCHDOG is not set -CONFIG_AT91_WATCHDOG=y +CONFIG_AT91RM9200_WATCHDOG=y # # USB-based Watchdog Cards diff --git a/arch/arm/configs/csb637_defconfig b/arch/arm/configs/csb637_defconfig index 2a1ac6c..df8595a 100644 --- a/arch/arm/configs/csb637_defconfig +++ b/arch/arm/configs/csb637_defconfig @@ -615,7 +615,7 @@ CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers # # CONFIG_SOFT_WATCHDOG is not set -CONFIG_AT91_WATCHDOG=y +CONFIG_AT91RM9200_WATCHDOG=y # # USB-based Watchdog Cards diff --git a/arch/arm/configs/kafa_defconfig b/arch/arm/configs/kafa_defconfig index 54fcd75..a4cdafc 100644 --- a/arch/arm/configs/kafa_defconfig +++ b/arch/arm/configs/kafa_defconfig @@ -560,7 +560,7 @@ CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers # # CONFIG_SOFT_WATCHDOG is not set -CONFIG_AT91_WATCHDOG=y +CONFIG_AT91RM9200_WATCHDOG=y # CONFIG_NVRAM is not set # CONFIG_DTLK is not set # CONFIG_R3964 is not set diff --git a/arch/arm/configs/onearm_defconfig b/arch/arm/configs/onearm_defconfig index cb1d94f..9b9f215 100644 --- a/arch/arm/configs/onearm_defconfig +++ b/arch/arm/configs/onearm_defconfig @@ -607,7 +607,7 @@ CONFIG_WATCHDOG_NOWAYOUT=y # Watchdog Device Drivers # # CONFIG_SOFT_WATCHDOG is not set -CONFIG_AT91_WATCHDOG=y +CONFIG_AT91RM9200_WATCHDOG=y # # USB-based Watchdog Cards diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index 847a260..529f0a7 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -60,7 +60,7 @@ config SOFT_WATCHDOG # ARM Architecture -config AT91_WATCHDOG +config AT91RM9200_WATCHDOG tristate "AT91RM9200 watchdog" depends on WATCHDOG && ARCH_AT91RM9200 help diff --git a/drivers/char/watchdog/Makefile b/drivers/char/watchdog/Makefile index ee34741..3644049 100644 --- a/drivers/char/watchdog/Makefile +++ b/drivers/char/watchdog/Makefile @@ -23,7 +23,7 @@ obj-$(CONFIG_WDTPCI) += wdt_pci.o obj-$(CONFIG_USBPCWATCHDOG) += pcwd_usb.o # ARM Architecture -obj-$(CONFIG_AT91_WATCHDOG) += at91_wdt.o +obj-$(CONFIG_AT91RM9200_WATCHDOG) += at91rm9200_wdt.o obj-$(CONFIG_OMAP_WATCHDOG) += omap_wdt.o obj-$(CONFIG_21285_WATCHDOG) += wdt285.o obj-$(CONFIG_977_WATCHDOG) += wdt977.o diff --git a/drivers/char/watchdog/at91_wdt.c b/drivers/char/watchdog/at91_wdt.c deleted file mode 100644 index 4e7a114..0000000 --- a/drivers/char/watchdog/at91_wdt.c +++ /dev/null @@ -1,287 +0,0 @@ -/* - * Watchdog driver for Atmel AT91RM9200 (Thunder) - * - * Copyright (C) 2003 SAN People (Pty) Ltd - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - - -#define WDT_DEFAULT_TIME 5 /* seconds */ -#define WDT_MAX_TIME 256 /* seconds */ - -static int wdt_time = WDT_DEFAULT_TIME; -static int nowayout = WATCHDOG_NOWAYOUT; - -module_param(wdt_time, int, 0); -MODULE_PARM_DESC(wdt_time, "Watchdog time in seconds. (default="__MODULE_STRING(WDT_DEFAULT_TIME) ")"); - -#ifdef CONFIG_WATCHDOG_NOWAYOUT -module_param(nowayout, int, 0); -MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")"); -#endif - - -static unsigned long at91wdt_busy; - -/* ......................................................................... */ - -/* - * Disable the watchdog. - */ -static void inline at91_wdt_stop(void) -{ - at91_sys_write(AT91_ST_WDMR, AT91_ST_EXTEN); -} - -/* - * Enable and reset the watchdog. - */ -static void inline at91_wdt_start(void) -{ - at91_sys_write(AT91_ST_WDMR, AT91_ST_EXTEN | AT91_ST_RSTEN | (((65536 * wdt_time) >> 8) & AT91_ST_WDV)); - at91_sys_write(AT91_ST_CR, AT91_ST_WDRST); -} - -/* - * Reload the watchdog timer. (ie, pat the watchdog) - */ -static void inline at91_wdt_reload(void) -{ - at91_sys_write(AT91_ST_CR, AT91_ST_WDRST); -} - -/* ......................................................................... */ - -/* - * Watchdog device is opened, and watchdog starts running. - */ -static int at91_wdt_open(struct inode *inode, struct file *file) -{ - if (test_and_set_bit(0, &at91wdt_busy)) - return -EBUSY; - - at91_wdt_start(); - return nonseekable_open(inode, file); -} - -/* - * Close the watchdog device. - * If CONFIG_WATCHDOG_NOWAYOUT is NOT defined then the watchdog is also - * disabled. - */ -static int at91_wdt_close(struct inode *inode, struct file *file) -{ - if (!nowayout) - at91_wdt_stop(); /* Disable the watchdog when file is closed */ - - clear_bit(0, &at91wdt_busy); - return 0; -} - -/* - * Change the watchdog time interval. - */ -static int at91_wdt_settimeout(int new_time) -{ - /* - * All counting occurs at SLOW_CLOCK / 128 = 0.256 Hz - * - * Since WDV is a 16-bit counter, the maximum period is - * 65536 / 0.256 = 256 seconds. - */ - if ((new_time <= 0) || (new_time > WDT_MAX_TIME)) - return -EINVAL; - - /* Set new watchdog time. It will be used when at91_wdt_start() is called. */ - wdt_time = new_time; - return 0; -} - -static struct watchdog_info at91_wdt_info = { - .identity = "at91 watchdog", - .options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING, -}; - -/* - * Handle commands from user-space. - */ -static int at91_wdt_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg) -{ - void __user *argp = (void __user *)arg; - int __user *p = argp; - int new_value; - - switch(cmd) { - case WDIOC_KEEPALIVE: - at91_wdt_reload(); /* pat the watchdog */ - return 0; - - case WDIOC_GETSUPPORT: - return copy_to_user(argp, &at91_wdt_info, sizeof(at91_wdt_info)) ? -EFAULT : 0; - - case WDIOC_SETTIMEOUT: - if (get_user(new_value, p)) - return -EFAULT; - - if (at91_wdt_settimeout(new_value)) - return -EINVAL; - - /* Enable new time value */ - at91_wdt_start(); - - /* Return current value */ - return put_user(wdt_time, p); - - case WDIOC_GETTIMEOUT: - return put_user(wdt_time, p); - - case WDIOC_GETSTATUS: - case WDIOC_GETBOOTSTATUS: - return put_user(0, p); - - case WDIOC_SETOPTIONS: - if (get_user(new_value, p)) - return -EFAULT; - - if (new_value & WDIOS_DISABLECARD) - at91_wdt_stop(); - if (new_value & WDIOS_ENABLECARD) - at91_wdt_start(); - return 0; - - default: - return -ENOTTY; - } -} - -/* - * Pat the watchdog whenever device is written to. - */ -static ssize_t at91_wdt_write(struct file *file, const char *data, size_t len, loff_t *ppos) -{ - at91_wdt_reload(); /* pat the watchdog */ - return len; -} - -/* ......................................................................... */ - -static const struct file_operations at91wdt_fops = { - .owner = THIS_MODULE, - .llseek = no_llseek, - .ioctl = at91_wdt_ioctl, - .open = at91_wdt_open, - .release = at91_wdt_close, - .write = at91_wdt_write, -}; - -static struct miscdevice at91wdt_miscdev = { - .minor = WATCHDOG_MINOR, - .name = "watchdog", - .fops = &at91wdt_fops, -}; - -static int __init at91wdt_probe(struct platform_device *pdev) -{ - int res; - - if (at91wdt_miscdev.dev) - return -EBUSY; - at91wdt_miscdev.dev = &pdev->dev; - - res = misc_register(&at91wdt_miscdev); - if (res) - return res; - - printk("AT91 Watchdog Timer enabled (%d seconds%s)\n", wdt_time, nowayout ? ", nowayout" : ""); - return 0; -} - -static int __exit at91wdt_remove(struct platform_device *pdev) -{ - int res; - - res = misc_deregister(&at91wdt_miscdev); - if (!res) - at91wdt_miscdev.dev = NULL; - - return res; -} - -static void at91wdt_shutdown(struct platform_device *pdev) -{ - at91_wdt_stop(); -} - -#ifdef CONFIG_PM - -static int at91wdt_suspend(struct platform_device *pdev, pm_message_t message) -{ - at91_wdt_stop(); - return 0; -} - -static int at91wdt_resume(struct platform_device *pdev) -{ - if (at91wdt_busy) - at91_wdt_start(); - return 0; -} - -#else -#define at91wdt_suspend NULL -#define at91wdt_resume NULL -#endif - -static struct platform_driver at91wdt_driver = { - .probe = at91wdt_probe, - .remove = __exit_p(at91wdt_remove), - .shutdown = at91wdt_shutdown, - .suspend = at91wdt_suspend, - .resume = at91wdt_resume, - .driver = { - .name = "at91_wdt", - .owner = THIS_MODULE, - }, -}; - -static int __init at91_wdt_init(void) -{ - /* Check that the heartbeat value is within range; if not reset to the default */ - if (at91_wdt_settimeout(wdt_time)) { - at91_wdt_settimeout(WDT_DEFAULT_TIME); - pr_info("at91_wdt: wdt_time value must be 1 <= wdt_time <= 256, using %d\n", wdt_time); - } - - return platform_driver_register(&at91wdt_driver); -} - -static void __exit at91_wdt_exit(void) -{ - platform_driver_unregister(&at91wdt_driver); -} - -module_init(at91_wdt_init); -module_exit(at91_wdt_exit); - -MODULE_AUTHOR("Andrew Victor"); -MODULE_DESCRIPTION("Watchdog driver for Atmel AT91RM9200"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); diff --git a/drivers/char/watchdog/at91rm9200_wdt.c b/drivers/char/watchdog/at91rm9200_wdt.c new file mode 100644 index 0000000..4e7a114 --- /dev/null +++ b/drivers/char/watchdog/at91rm9200_wdt.c @@ -0,0 +1,287 @@ +/* + * Watchdog driver for Atmel AT91RM9200 (Thunder) + * + * Copyright (C) 2003 SAN People (Pty) Ltd + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +#define WDT_DEFAULT_TIME 5 /* seconds */ +#define WDT_MAX_TIME 256 /* seconds */ + +static int wdt_time = WDT_DEFAULT_TIME; +static int nowayout = WATCHDOG_NOWAYOUT; + +module_param(wdt_time, int, 0); +MODULE_PARM_DESC(wdt_time, "Watchdog time in seconds. (default="__MODULE_STRING(WDT_DEFAULT_TIME) ")"); + +#ifdef CONFIG_WATCHDOG_NOWAYOUT +module_param(nowayout, int, 0); +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")"); +#endif + + +static unsigned long at91wdt_busy; + +/* ......................................................................... */ + +/* + * Disable the watchdog. + */ +static void inline at91_wdt_stop(void) +{ + at91_sys_write(AT91_ST_WDMR, AT91_ST_EXTEN); +} + +/* + * Enable and reset the watchdog. + */ +static void inline at91_wdt_start(void) +{ + at91_sys_write(AT91_ST_WDMR, AT91_ST_EXTEN | AT91_ST_RSTEN | (((65536 * wdt_time) >> 8) & AT91_ST_WDV)); + at91_sys_write(AT91_ST_CR, AT91_ST_WDRST); +} + +/* + * Reload the watchdog timer. (ie, pat the watchdog) + */ +static void inline at91_wdt_reload(void) +{ + at91_sys_write(AT91_ST_CR, AT91_ST_WDRST); +} + +/* ......................................................................... */ + +/* + * Watchdog device is opened, and watchdog starts running. + */ +static int at91_wdt_open(struct inode *inode, struct file *file) +{ + if (test_and_set_bit(0, &at91wdt_busy)) + return -EBUSY; + + at91_wdt_start(); + return nonseekable_open(inode, file); +} + +/* + * Close the watchdog device. + * If CONFIG_WATCHDOG_NOWAYOUT is NOT defined then the watchdog is also + * disabled. + */ +static int at91_wdt_close(struct inode *inode, struct file *file) +{ + if (!nowayout) + at91_wdt_stop(); /* Disable the watchdog when file is closed */ + + clear_bit(0, &at91wdt_busy); + return 0; +} + +/* + * Change the watchdog time interval. + */ +static int at91_wdt_settimeout(int new_time) +{ + /* + * All counting occurs at SLOW_CLOCK / 128 = 0.256 Hz + * + * Since WDV is a 16-bit counter, the maximum period is + * 65536 / 0.256 = 256 seconds. + */ + if ((new_time <= 0) || (new_time > WDT_MAX_TIME)) + return -EINVAL; + + /* Set new watchdog time. It will be used when at91_wdt_start() is called. */ + wdt_time = new_time; + return 0; +} + +static struct watchdog_info at91_wdt_info = { + .identity = "at91 watchdog", + .options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING, +}; + +/* + * Handle commands from user-space. + */ +static int at91_wdt_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg) +{ + void __user *argp = (void __user *)arg; + int __user *p = argp; + int new_value; + + switch(cmd) { + case WDIOC_KEEPALIVE: + at91_wdt_reload(); /* pat the watchdog */ + return 0; + + case WDIOC_GETSUPPORT: + return copy_to_user(argp, &at91_wdt_info, sizeof(at91_wdt_info)) ? -EFAULT : 0; + + case WDIOC_SETTIMEOUT: + if (get_user(new_value, p)) + return -EFAULT; + + if (at91_wdt_settimeout(new_value)) + return -EINVAL; + + /* Enable new time value */ + at91_wdt_start(); + + /* Return current value */ + return put_user(wdt_time, p); + + case WDIOC_GETTIMEOUT: + return put_user(wdt_time, p); + + case WDIOC_GETSTATUS: + case WDIOC_GETBOOTSTATUS: + return put_user(0, p); + + case WDIOC_SETOPTIONS: + if (get_user(new_value, p)) + return -EFAULT; + + if (new_value & WDIOS_DISABLECARD) + at91_wdt_stop(); + if (new_value & WDIOS_ENABLECARD) + at91_wdt_start(); + return 0; + + default: + return -ENOTTY; + } +} + +/* + * Pat the watchdog whenever device is written to. + */ +static ssize_t at91_wdt_write(struct file *file, const char *data, size_t len, loff_t *ppos) +{ + at91_wdt_reload(); /* pat the watchdog */ + return len; +} + +/* ......................................................................... */ + +static const struct file_operations at91wdt_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .ioctl = at91_wdt_ioctl, + .open = at91_wdt_open, + .release = at91_wdt_close, + .write = at91_wdt_write, +}; + +static struct miscdevice at91wdt_miscdev = { + .minor = WATCHDOG_MINOR, + .name = "watchdog", + .fops = &at91wdt_fops, +}; + +static int __init at91wdt_probe(struct platform_device *pdev) +{ + int res; + + if (at91wdt_miscdev.dev) + return -EBUSY; + at91wdt_miscdev.dev = &pdev->dev; + + res = misc_register(&at91wdt_miscdev); + if (res) + return res; + + printk("AT91 Watchdog Timer enabled (%d seconds%s)\n", wdt_time, nowayout ? ", nowayout" : ""); + return 0; +} + +static int __exit at91wdt_remove(struct platform_device *pdev) +{ + int res; + + res = misc_deregister(&at91wdt_miscdev); + if (!res) + at91wdt_miscdev.dev = NULL; + + return res; +} + +static void at91wdt_shutdown(struct platform_device *pdev) +{ + at91_wdt_stop(); +} + +#ifdef CONFIG_PM + +static int at91wdt_suspend(struct platform_device *pdev, pm_message_t message) +{ + at91_wdt_stop(); + return 0; +} + +static int at91wdt_resume(struct platform_device *pdev) +{ + if (at91wdt_busy) + at91_wdt_start(); + return 0; +} + +#else +#define at91wdt_suspend NULL +#define at91wdt_resume NULL +#endif + +static struct platform_driver at91wdt_driver = { + .probe = at91wdt_probe, + .remove = __exit_p(at91wdt_remove), + .shutdown = at91wdt_shutdown, + .suspend = at91wdt_suspend, + .resume = at91wdt_resume, + .driver = { + .name = "at91_wdt", + .owner = THIS_MODULE, + }, +}; + +static int __init at91_wdt_init(void) +{ + /* Check that the heartbeat value is within range; if not reset to the default */ + if (at91_wdt_settimeout(wdt_time)) { + at91_wdt_settimeout(WDT_DEFAULT_TIME); + pr_info("at91_wdt: wdt_time value must be 1 <= wdt_time <= 256, using %d\n", wdt_time); + } + + return platform_driver_register(&at91wdt_driver); +} + +static void __exit at91_wdt_exit(void) +{ + platform_driver_unregister(&at91wdt_driver); +} + +module_init(at91_wdt_init); +module_exit(at91_wdt_exit); + +MODULE_AUTHOR("Andrew Victor"); +MODULE_DESCRIPTION("Watchdog driver for Atmel AT91RM9200"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); -- cgit v0.10.2 From 76dc82ab57236105285fd8520895c1404b8b952f Mon Sep 17 00:00:00 2001 From: Steven Toth Date: Sat, 30 Sep 2006 00:43:58 -0300 Subject: V4L/DVB (4692): Add WinTV-HVR3000 DVB-T support The WinTV-HVR3000 is currently defined for analog support only. This patch adds full DVB-T support. (DVB-S support will be added soon) Signed-off-by: Steven Toth Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/Documentation/video4linux/CARDLIST.cx88 b/Documentation/video4linux/CARDLIST.cx88 index 126e59d9..8755b3e 100644 --- a/Documentation/video4linux/CARDLIST.cx88 +++ b/Documentation/video4linux/CARDLIST.cx88 @@ -51,7 +51,7 @@ 50 -> NPG Tech Real TV FM Top 10 [14f1:0842] 51 -> WinFast DTV2000 H [107d:665e] 52 -> Geniatech DVB-S [14f1:0084] - 53 -> Hauppauge WinTV-HVR3000 TriMode Analog/DVB-S/DVB-T [0070:1404] + 53 -> Hauppauge WinTV-HVR3000 TriMode Analog/DVB-S/DVB-T [0070:1404,0070:1400,0070:1401,0070:1402] 54 -> Norwood Micro TV Tuner 55 -> Shenzhen Tungsten Ages Tech TE-DTV-250 / Swann OEM [c180:c980] 56 -> Hauppauge WinTV-HVR1300 DVB-T/Hybrid MPEG Encoder [0070:9600,0070:9601,0070:9602] diff --git a/drivers/media/video/cx88/cx88-cards.c b/drivers/media/video/cx88/cx88-cards.c index af71d42..f764a57 100644 --- a/drivers/media/video/cx88/cx88-cards.c +++ b/drivers/media/video/cx88/cx88-cards.c @@ -1230,6 +1230,7 @@ struct cx88_board cx88_boards[] = { .vmux = 2, .gpio0 = 0x84bf, }}, + .mpeg = CX88_MPEG_DVB, }, [CX88_BOARD_NORWOOD_MICRO] = { .name = "Norwood Micro TV Tuner", @@ -1590,6 +1591,18 @@ struct cx88_subid cx88_subids[] = { .subvendor = 0x0070, .subdevice = 0x9000, .card = CX88_BOARD_HAUPPAUGE_DVB_T1, + },{ + .subvendor = 0x0070, + .subdevice = 0x1400, + .card = CX88_BOARD_HAUPPAUGE_HVR3000, + },{ + .subvendor = 0x0070, + .subdevice = 0x1401, + .card = CX88_BOARD_HAUPPAUGE_HVR3000, + },{ + .subvendor = 0x0070, + .subdevice = 0x1402, + .card = CX88_BOARD_HAUPPAUGE_HVR3000, }, }; const unsigned int cx88_idcount = ARRAY_SIZE(cx88_subids); @@ -1633,7 +1646,15 @@ static void hauppauge_eeprom(struct cx88_core *core, u8 *eeprom_data) /* Make sure we support the board model */ switch (tv.model) { + case 14009: /* WinTV-HVR3000 (Retail, IR, b/panel video, 3.5mm audio in) */ + case 14019: /* WinTV-HVR3000 (Retail, IR Blaster, b/panel video, 3.5mm audio in) */ + case 14029: /* WinTV-HVR3000 (Retail, IR, b/panel video, 3.5mm audio in - 880 bridge) */ + case 14109: /* WinTV-HVR3000 (Retail, IR, b/panel video, 3.5mm audio in - low profile) */ + case 14129: /* WinTV-HVR3000 (Retail, IR, b/panel video, 3.5mm audio in - 880 bridge - LP) */ + case 14559: /* WinTV-HVR3000 (OEM, no IR, b/panel video, 3.5mm audio in) */ case 14569: /* WinTV-HVR3000 (OEM, no IR, no back panel video) */ + case 14659: /* WinTV-HVR3000 (OEM, no IR, b/panel video, RCA audio in - Low profile) */ + case 14669: /* WinTV-HVR3000 (OEM, no IR, no b/panel video - Low profile) */ case 28552: /* WinTV-PVR 'Roslyn' (No IR) */ case 34519: /* WinTV-PCI-FM */ case 90002: /* Nova-T-PCI (9002) */ diff --git a/drivers/media/video/cx88/cx88-dvb.c b/drivers/media/video/cx88/cx88-dvb.c index bd0c879..0ef13e7 100644 --- a/drivers/media/video/cx88/cx88-dvb.c +++ b/drivers/media/video/cx88/cx88-dvb.c @@ -315,15 +315,22 @@ static struct cx22702_config hauppauge_novat_config = { .demod_address = 0x43, .output_mode = CX22702_SERIAL_OUTPUT, }; + static struct cx22702_config hauppauge_hvr1100_config = { .demod_address = 0x63, .output_mode = CX22702_SERIAL_OUTPUT, }; + static struct cx22702_config hauppauge_hvr1300_config = { .demod_address = 0x63, .output_mode = CX22702_SERIAL_OUTPUT, }; +static struct cx22702_config hauppauge_hvr3000_config = { + .demod_address = 0x63, + .output_mode = CX22702_SERIAL_OUTPUT, +}; + static int or51132_set_ts_param(struct dvb_frontend* fe, int is_punctured) { @@ -558,6 +565,16 @@ static int dvb_register(struct cx8802_dev *dev) &dvb_pll_fmd1216me); } break; + case CX88_BOARD_HAUPPAUGE_HVR3000: + dev->dvb.frontend = dvb_attach(cx22702_attach, + &hauppauge_hvr3000_config, + &dev->core->i2c_adap); + if (dev->dvb.frontend != NULL) { + dvb_attach(dvb_pll_attach, dev->dvb.frontend, 0x61, + &dev->core->i2c_adap, + &dvb_pll_fmd1216me); + } + break; case CX88_BOARD_DVICO_FUSIONHDTV_DVB_T_PLUS: dev->dvb.frontend = dvb_attach(mt352_attach, &dvico_fusionhdtv, diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c index 83ebf7a..ee48995 100644 --- a/drivers/media/video/cx88/cx88-input.c +++ b/drivers/media/video/cx88/cx88-input.c @@ -196,6 +196,7 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci) case CX88_BOARD_HAUPPAUGE_NOVASPLUS_S1: case CX88_BOARD_HAUPPAUGE_HVR1100: case CX88_BOARD_HAUPPAUGE_HVR1300: + case CX88_BOARD_HAUPPAUGE_HVR3000: ir_codes = ir_codes_hauppauge_new; ir_type = IR_TYPE_RC5; ir->sampling = 1; @@ -419,6 +420,7 @@ void cx88_ir_irq(struct cx88_core *core) case CX88_BOARD_HAUPPAUGE_NOVASPLUS_S1: case CX88_BOARD_HAUPPAUGE_HVR1100: case CX88_BOARD_HAUPPAUGE_HVR1300: + case CX88_BOARD_HAUPPAUGE_HVR3000: ircode = ir_decode_biphase(ir->samples, ir->scount, 5, 7); ir_dprintk("biphase decoded: %x\n", ircode); if ((ircode & 0xfffff000) != 0x3000) -- cgit v0.10.2 From 7844d7561307d6f8b0dd18b91f4dc6cff53848b4 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 6 Oct 2006 17:12:00 -0300 Subject: V4L/DVB (4725): Fix vivi compile on parisc parisc (and several other architectures) don't have a dma_address in their sg list. Use the macro instead. Signed-off-by: Matthew Wilcox Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/vivi.c b/drivers/media/video/vivi.c index e7c01d5..3c8dc72 100644 --- a/drivers/media/video/vivi.c +++ b/drivers/media/video/vivi.c @@ -272,7 +272,7 @@ static void gen_line(struct sg_to_addr to_addr[],int inipos,int pages,int wmax, /* Get first addr pointed to pixel position */ oldpg=get_addr_pos(pos,pages,to_addr); - pg=pfn_to_page(to_addr[oldpg].sg->dma_address >> PAGE_SHIFT); + pg=pfn_to_page(sg_dma_address(to_addr[oldpg].sg) >> PAGE_SHIFT); basep = kmap_atomic(pg, KM_BOUNCE_READ)+to_addr[oldpg].sg->offset; /* We will just duplicate the second pixel at the packet */ @@ -287,7 +287,7 @@ static void gen_line(struct sg_to_addr to_addr[],int inipos,int pages,int wmax, for (color=0;color<4;color++) { pgpos=get_addr_pos(pos,pages,to_addr); if (pgpos!=oldpg) { - pg=pfn_to_page(to_addr[pgpos].sg->dma_address >> PAGE_SHIFT); + pg=pfn_to_page(sg_dma_address(to_addr[pgpos].sg) >> PAGE_SHIFT); kunmap_atomic(basep, KM_BOUNCE_READ); basep= kmap_atomic(pg, KM_BOUNCE_READ)+to_addr[pgpos].sg->offset; oldpg=pgpos; @@ -339,8 +339,8 @@ static void gen_line(struct sg_to_addr to_addr[],int inipos,int pages,int wmax, for (color=0;color<4;color++) { pgpos=get_addr_pos(pos,pages,to_addr); if (pgpos!=oldpg) { - pg=pfn_to_page(to_addr[pgpos]. - sg->dma_address + pg=pfn_to_page(sg_dma_address( + to_addr[pgpos].sg) >> PAGE_SHIFT); kunmap_atomic(basep, KM_BOUNCE_READ); @@ -386,7 +386,7 @@ static void vivi_fillbuff(struct vivi_dev *dev,struct vivi_buffer *buf) struct timeval ts; /* Test if DMA mapping is ready */ - if (!vb->dma.sglist[0].dma_address) + if (!sg_dma_address(&vb->dma.sglist[0])) return; prep_to_addr(to_addr,vb); @@ -783,7 +783,7 @@ static int vivi_map_sg(void *dev, struct scatterlist *sg, int nents, for (i = 0; i < nents; i++ ) { BUG_ON(!sg[i].page); - sg[i].dma_address = page_to_phys(sg[i].page) + sg[i].offset; + sg_dma_address(&sg[i]) = page_to_phys(sg[i].page) + sg[i].offset; } return nents; -- cgit v0.10.2 From 2e7cf3ea5acc7ed57b8883cc6d35ffc06a5c95fa Mon Sep 17 00:00:00 2001 From: Hartmut Hackmann Date: Fri, 6 Oct 2006 19:45:23 -0300 Subject: V4L/DVB (4727): Support status readout for saa713x based FM radio This patch adds readout for stereo and signal level for saa713x cards which use the saa713x as FM demodulator. These are many cards based on saa7133, tda8290 and tda8275a. FM channel search should work now. Signed-off-by: Hartmut Hackmann Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c index 203302f..830617e 100644 --- a/drivers/media/video/saa7134/saa7134-video.c +++ b/drivers/media/video/saa7134/saa7134-video.c @@ -2248,7 +2248,11 @@ static int radio_do_ioctl(struct inode *inode, struct file *file, t->type = V4L2_TUNER_RADIO; saa7134_i2c_call_clients(dev, VIDIOC_G_TUNER, t); - + if (dev->input->amux == TV) { + t->signal = 0xf800 - ((saa_readb(0x581) & 0x1f) << 11); + t->rxsubchans = (saa_readb(0x529) & 0x08) ? + V4L2_TUNER_SUB_STEREO : V4L2_TUNER_SUB_MONO; + } return 0; } case VIDIOC_S_TUNER: -- cgit v0.10.2 From e0abc8cd54f5ac65465918f32f286218aa33e8c0 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Sat, 7 Oct 2006 16:22:10 -0300 Subject: V4L/DVB (4729): Fix VIDIOC_G_FMT for NTSC in cx25840. VIDIOC_G_FMT returned the sliced VBI types in the wrong lines for NTSC (three lines too low). Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/cx25840/cx25840-vbi.c b/drivers/media/video/cx25840/cx25840-vbi.c index 48014a2..f85f208 100644 --- a/drivers/media/video/cx25840/cx25840-vbi.c +++ b/drivers/media/video/cx25840/cx25840-vbi.c @@ -235,6 +235,7 @@ int cx25840_vbi(struct i2c_client *client, unsigned int cmd, void *arg) 0, 0, V4L2_SLICED_VPS, 0, 0, /* 9 */ 0, 0, 0, 0 }; + int is_pal = !(cx25840_get_v4lstd(client) & V4L2_STD_525_60); int i; fmt = arg; @@ -246,13 +247,25 @@ int cx25840_vbi(struct i2c_client *client, unsigned int cmd, void *arg) if ((cx25840_read(client, 0x404) & 0x10) == 0) break; - for (i = 7; i <= 23; i++) { - u8 v = cx25840_read(client, 0x424 + i - 7); + if (is_pal) { + for (i = 7; i <= 23; i++) { + u8 v = cx25840_read(client, 0x424 + i - 7); + + svbi->service_lines[0][i] = lcr2vbi[v >> 4]; + svbi->service_lines[1][i] = lcr2vbi[v & 0xf]; + svbi->service_set |= + svbi->service_lines[0][i] | svbi->service_lines[1][i]; + } + } + else { + for (i = 10; i <= 21; i++) { + u8 v = cx25840_read(client, 0x424 + i - 10); - svbi->service_lines[0][i] = lcr2vbi[v >> 4]; - svbi->service_lines[1][i] = lcr2vbi[v & 0xf]; - svbi->service_set |= - svbi->service_lines[0][i] | svbi->service_lines[1][i]; + svbi->service_lines[0][i] = lcr2vbi[v >> 4]; + svbi->service_lines[1][i] = lcr2vbi[v & 0xf]; + svbi->service_set |= + svbi->service_lines[0][i] | svbi->service_lines[1][i]; + } } break; } -- cgit v0.10.2 From 180958febfb8d32da1d4fee13868e03be0cb931a Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Sat, 7 Oct 2006 16:10:53 -0300 Subject: V4L/DVB (4731a): Kconfig: restore pvrusb2 menu items Looks like the pvrusb2 menu items were accidentally removed in git commit 1450e6bedc58c731617d99b4670070ed3ccc91b4 This patch restores the menu items so that the pvrusb2 driver can be built. Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index afb734d..fbe5b61 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -677,6 +677,8 @@ config VIDEO_M32R_AR_M64278 menu "V4L USB devices" depends on USB && VIDEO_DEV +source "drivers/media/video/pvrusb2/Kconfig" + source "drivers/media/video/em28xx/Kconfig" source "drivers/media/video/usbvideo/Kconfig" -- cgit v0.10.2 From 934765b8e2f211aec119dbdd9feea6d3f2ffaf7e Mon Sep 17 00:00:00 2001 From: Uwe Bugla Date: Fri, 6 Oct 2006 13:12:48 -0300 Subject: V4L/DVB (4732): Fix spelling error in Kconfig help text for DVB_CORE_ATTACH Signed-off-by: Uwe Bugla Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/dvb-core/Kconfig b/drivers/media/dvb/dvb-core/Kconfig index e46eae3..1990eda 100644 --- a/drivers/media/dvb/dvb-core/Kconfig +++ b/drivers/media/dvb/dvb-core/Kconfig @@ -19,6 +19,6 @@ config DVB_CORE_ATTACH allow the card drivers to only load the frontend modules they require. This saves several KBytes of memory. - Note: You will need moudule-init-tools v3.2 or later for this feature. + Note: You will need module-init-tools v3.2 or later for this feature. If unsure say Y. -- cgit v0.10.2 From fc13d929cc7af3c0da09ea2b6d23465b933e279d Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Mon, 9 Oct 2006 05:17:09 -0300 Subject: V4L/DVB (4733): Tda10086: fix frontend selection for dvb_attach Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/frontends/tda10086.h b/drivers/media/dvb/frontends/tda10086.h index e8061db..18457ad 100644 --- a/drivers/media/dvb/frontends/tda10086.h +++ b/drivers/media/dvb/frontends/tda10086.h @@ -35,7 +35,16 @@ struct tda10086_config u8 invert; }; +#if defined(CONFIG_DVB_TDA10086) || defined(CONFIG_DVB_TDA10086_MODULE) extern struct dvb_frontend* tda10086_attach(const struct tda10086_config* config, struct i2c_adapter* i2c); +#else +static inline struct dvb_frontend* tda10086_attach(const struct tda10086_config* config, + struct i2c_adapter* i2c) +{ + printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __FUNCTION__); + return NULL; +} +#endif // CONFIG_DVB_TDA10086 #endif // TDA10086_H -- cgit v0.10.2 From c071fab453f7b181c49d92d06d936bb243ef1932 Mon Sep 17 00:00:00 2001 From: Michael Krufky Date: Mon, 9 Oct 2006 05:17:45 -0300 Subject: V4L/DVB (4734): Tda826x: fix frontend selection for dvb_attach Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/frontends/tda826x.h b/drivers/media/dvb/frontends/tda826x.h index 3307607..83998c0 100644 --- a/drivers/media/dvb/frontends/tda826x.h +++ b/drivers/media/dvb/frontends/tda826x.h @@ -35,6 +35,19 @@ * @param has_loopthrough Set to 1 if the card has a loopthrough RF connector. * @return FE pointer on success, NULL on failure. */ -extern struct dvb_frontend *tda826x_attach(struct dvb_frontend *fe, int addr, struct i2c_adapter *i2c, int has_loopthrough); - -#endif +#if defined(CONFIG_DVB_TDA826X) || defined(CONFIG_DVB_TDA826X_MODULE) +extern struct dvb_frontend* tda826x_attach(struct dvb_frontend *fe, int addr, + struct i2c_adapter *i2c, + int has_loopthrough); +#else +static inline struct dvb_frontend* tda826x_attach(struct dvb_frontend *fe, + int addr, + struct i2c_adapter *i2c, + int has_loopthrough) +{ + printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __FUNCTION__); + return NULL; +} +#endif // CONFIG_DVB_TDA826X + +#endif // __DVB_TDA826X_H__ -- cgit v0.10.2 From 626ae83bb24927ca015503448f0199842ae2e8da Mon Sep 17 00:00:00 2001 From: Amit Choudhary Date: Mon, 9 Oct 2006 15:50:10 -0300 Subject: V4L/DVB (4738): Bt8xx/dvb-bt8xx.c: check kmalloc() return value. Check the return value of kmalloc() in function frontend_init(), in file drivers/media/dvb/bt8xx/dvb-bt8xx.c. Signed-off-by: Amit Choudhary Signed-off-by: Manu Abraham Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/bt8xx/dvb-bt8xx.c b/drivers/media/dvb/bt8xx/dvb-bt8xx.c index fb6c4cc..14e69a7 100644 --- a/drivers/media/dvb/bt8xx/dvb-bt8xx.c +++ b/drivers/media/dvb/bt8xx/dvb-bt8xx.c @@ -665,6 +665,10 @@ static void frontend_init(struct dvb_bt8xx_card *card, u32 type) case BTTV_BOARD_TWINHAN_DST: /* DST is not a frontend driver !!! */ state = (struct dst_state *) kmalloc(sizeof (struct dst_state), GFP_KERNEL); + if (!state) { + printk("dvb_bt8xx: No memory\n"); + break; + } /* Setup the Card */ state->config = &dst_config; state->i2c = card->i2c_adapter; -- cgit v0.10.2 From fc2fa31f4eaa53995593ced14c73f2cf63dcfa17 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?P=E1draig=20Brady?= Date: Mon, 9 Oct 2006 08:02:17 -0300 Subject: V4L/DVB (4739): SECAM support for saa7113 into saa7115 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Without the attached trivial patch, the saa7113 is set up for PAL when SECAM is selected and hence will see only show black and white for SECAM signals. Tested the patch against the saa7115 module in linux-2.6.17 with a Pinnacle 50e USB tuner (em28xx). Signed-off-by: Pádraig Brady Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/saa7115.c b/drivers/media/video/saa7115.c index 974179d..c5719f7 100644 --- a/drivers/media/video/saa7115.c +++ b/drivers/media/video/saa7115.c @@ -960,6 +960,8 @@ static void saa711x_set_v4lstd(struct i2c_client *client, v4l2_std_id std) reg |= 0x10; } else if (std == V4L2_STD_NTSC_M_JP) { reg |= 0x40; + } else if (std == V4L2_STD_SECAM) { + reg |= 0x50; } saa711x_write(client, R_0E_CHROMA_CNTL_1, reg); } else { -- cgit v0.10.2 From 474ce78130ba37cb50e620c538ab3ffe6c582ba6 Mon Sep 17 00:00:00 2001 From: Enrico Scholz Date: Mon, 9 Oct 2006 16:27:05 -0300 Subject: V4L/DVB (4740): Fixed an if-block to avoid floating with debug-messages The dbgarg() macro in videodev.c contains some printk() statements where only the first one is influenced by an if-statement. This causes floating with debug-messages which is fixed by this patch by adding a '{ ... }' pair. Signed-off-by: Enrico Scholz Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c index 479a067..98de872 100644 --- a/drivers/media/video/videodev.c +++ b/drivers/media/video/videodev.c @@ -17,10 +17,11 @@ */ #define dbgarg(cmd, fmt, arg...) \ - if (vfd->debug & V4L2_DEBUG_IOCTL_ARG) \ + if (vfd->debug & V4L2_DEBUG_IOCTL_ARG) { \ printk (KERN_DEBUG "%s: ", vfd->name); \ v4l_printk_ioctl(cmd); \ - printk (KERN_DEBUG "%s: " fmt, vfd->name, ## arg); + printk (KERN_DEBUG "%s: " fmt, vfd->name, ## arg); \ + } #define dbgarg2(fmt, arg...) \ if (vfd->debug & V4L2_DEBUG_IOCTL_ARG) \ -- cgit v0.10.2 From 2444a2fca488fa8e362895a4ca9fdc51f497282a Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 10 Oct 2006 15:09:43 -0300 Subject: V4L/DVB (4741): {ov511,stv680}: handle sysfs errors Signed-off-by: Jeff Garzik Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/ov511.c b/drivers/media/video/ov511.c index ce4886f..b4db2cb 100644 --- a/drivers/media/video/ov511.c +++ b/drivers/media/video/ov511.c @@ -5648,17 +5648,49 @@ static ssize_t show_exposure(struct class_device *cd, char *buf) } static CLASS_DEVICE_ATTR(exposure, S_IRUGO, show_exposure, NULL); -static void ov_create_sysfs(struct video_device *vdev) +static int ov_create_sysfs(struct video_device *vdev) { - video_device_create_file(vdev, &class_device_attr_custom_id); - video_device_create_file(vdev, &class_device_attr_model); - video_device_create_file(vdev, &class_device_attr_bridge); - video_device_create_file(vdev, &class_device_attr_sensor); - video_device_create_file(vdev, &class_device_attr_brightness); - video_device_create_file(vdev, &class_device_attr_saturation); - video_device_create_file(vdev, &class_device_attr_contrast); - video_device_create_file(vdev, &class_device_attr_hue); - video_device_create_file(vdev, &class_device_attr_exposure); + int rc; + + rc = video_device_create_file(vdev, &class_device_attr_custom_id); + if (rc) goto err; + rc = video_device_create_file(vdev, &class_device_attr_model); + if (rc) goto err_id; + rc = video_device_create_file(vdev, &class_device_attr_bridge); + if (rc) goto err_model; + rc = video_device_create_file(vdev, &class_device_attr_sensor); + if (rc) goto err_bridge; + rc = video_device_create_file(vdev, &class_device_attr_brightness); + if (rc) goto err_sensor; + rc = video_device_create_file(vdev, &class_device_attr_saturation); + if (rc) goto err_bright; + rc = video_device_create_file(vdev, &class_device_attr_contrast); + if (rc) goto err_sat; + rc = video_device_create_file(vdev, &class_device_attr_hue); + if (rc) goto err_contrast; + rc = video_device_create_file(vdev, &class_device_attr_exposure); + if (rc) goto err_hue; + + return 0; + +err_hue: + video_device_remove_file(vdev, &class_device_attr_hue); +err_contrast: + video_device_remove_file(vdev, &class_device_attr_contrast); +err_sat: + video_device_remove_file(vdev, &class_device_attr_saturation); +err_bright: + video_device_remove_file(vdev, &class_device_attr_brightness); +err_sensor: + video_device_remove_file(vdev, &class_device_attr_sensor); +err_bridge: + video_device_remove_file(vdev, &class_device_attr_bridge); +err_model: + video_device_remove_file(vdev, &class_device_attr_model); +err_id: + video_device_remove_file(vdev, &class_device_attr_custom_id); +err: + return rc; } /**************************************************************************** @@ -5817,7 +5849,11 @@ ov51x_probe(struct usb_interface *intf, const struct usb_device_id *id) ov->vdev->minor); usb_set_intfdata(intf, ov); - ov_create_sysfs(ov->vdev); + if (ov_create_sysfs(ov->vdev)) { + err("ov_create_sysfs failed"); + goto error; + } + return 0; error: diff --git a/drivers/media/video/stv680.c b/drivers/media/video/stv680.c index 87e1130..6d1ef1e 100644 --- a/drivers/media/video/stv680.c +++ b/drivers/media/video/stv680.c @@ -516,16 +516,45 @@ stv680_file(frames_read, framecount, "%d\n"); stv680_file(packets_dropped, dropped, "%d\n"); stv680_file(decoding_errors, error, "%d\n"); -static void stv680_create_sysfs_files(struct video_device *vdev) +static int stv680_create_sysfs_files(struct video_device *vdev) { - video_device_create_file(vdev, &class_device_attr_model); - video_device_create_file(vdev, &class_device_attr_in_use); - video_device_create_file(vdev, &class_device_attr_streaming); - video_device_create_file(vdev, &class_device_attr_palette); - video_device_create_file(vdev, &class_device_attr_frames_total); - video_device_create_file(vdev, &class_device_attr_frames_read); - video_device_create_file(vdev, &class_device_attr_packets_dropped); - video_device_create_file(vdev, &class_device_attr_decoding_errors); + int rc; + + rc = video_device_create_file(vdev, &class_device_attr_model); + if (rc) goto err; + rc = video_device_create_file(vdev, &class_device_attr_in_use); + if (rc) goto err_model; + rc = video_device_create_file(vdev, &class_device_attr_streaming); + if (rc) goto err_inuse; + rc = video_device_create_file(vdev, &class_device_attr_palette); + if (rc) goto err_stream; + rc = video_device_create_file(vdev, &class_device_attr_frames_total); + if (rc) goto err_pal; + rc = video_device_create_file(vdev, &class_device_attr_frames_read); + if (rc) goto err_framtot; + rc = video_device_create_file(vdev, &class_device_attr_packets_dropped); + if (rc) goto err_framread; + rc = video_device_create_file(vdev, &class_device_attr_decoding_errors); + if (rc) goto err_dropped; + + return 0; + +err_dropped: + video_device_remove_file(vdev, &class_device_attr_packets_dropped); +err_framread: + video_device_remove_file(vdev, &class_device_attr_frames_read); +err_framtot: + video_device_remove_file(vdev, &class_device_attr_frames_total); +err_pal: + video_device_remove_file(vdev, &class_device_attr_palette); +err_stream: + video_device_remove_file(vdev, &class_device_attr_streaming); +err_inuse: + video_device_remove_file(vdev, &class_device_attr_in_use); +err_model: + video_device_remove_file(vdev, &class_device_attr_model); +err: + return rc; } static void stv680_remove_sysfs_files(struct video_device *vdev) @@ -1418,9 +1447,13 @@ static int stv680_probe (struct usb_interface *intf, const struct usb_device_id PDEBUG (0, "STV(i): registered new video device: video%d", stv680->vdev->minor); usb_set_intfdata (intf, stv680); - stv680_create_sysfs_files(stv680->vdev); + retval = stv680_create_sysfs_files(stv680->vdev); + if (retval) + goto error_unreg; return 0; +error_unreg: + video_unregister_device(stv680->vdev); error_vdev: video_device_release(stv680->vdev); error: -- cgit v0.10.2 From c12e3be0860652ed1e15c9442adcba44317211d1 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Fri, 13 Oct 2006 07:17:32 -0300 Subject: V4L/DVB (4742): Drivers/media/video: handle sysfs errors Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/et61x251/et61x251_core.c b/drivers/media/video/et61x251/et61x251_core.c index bc544cc..f786ab1 100644 --- a/drivers/media/video/et61x251/et61x251_core.c +++ b/drivers/media/video/et61x251/et61x251_core.c @@ -973,16 +973,32 @@ static CLASS_DEVICE_ATTR(i2c_val, S_IRUGO | S_IWUSR, et61x251_show_i2c_val, et61x251_store_i2c_val); -static void et61x251_create_sysfs(struct et61x251_device* cam) +static int et61x251_create_sysfs(struct et61x251_device* cam) { struct video_device *v4ldev = cam->v4ldev; + int rc; - video_device_create_file(v4ldev, &class_device_attr_reg); - video_device_create_file(v4ldev, &class_device_attr_val); + rc = video_device_create_file(v4ldev, &class_device_attr_reg); + if (rc) goto err; + rc = video_device_create_file(v4ldev, &class_device_attr_val); + if (rc) goto err_reg; if (cam->sensor.sysfs_ops) { - video_device_create_file(v4ldev, &class_device_attr_i2c_reg); - video_device_create_file(v4ldev, &class_device_attr_i2c_val); + rc = video_device_create_file(v4ldev, &class_device_attr_i2c_reg); + if (rc) goto err_val; + rc = video_device_create_file(v4ldev, &class_device_attr_i2c_val); + if (rc) goto err_i2c_reg; } + + return 0; + +err_i2c_reg: + video_device_remove_file(v4ldev, &class_device_attr_i2c_reg); +err_val: + video_device_remove_file(v4ldev, &class_device_attr_val); +err_reg: + video_device_remove_file(v4ldev, &class_device_attr_reg); +err: + return rc; } #endif /* CONFIG_VIDEO_ADV_DEBUG */ @@ -2534,7 +2550,9 @@ et61x251_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) dev_nr = (dev_nr < ET61X251_MAX_DEVICES-1) ? dev_nr+1 : 0; #ifdef CONFIG_VIDEO_ADV_DEBUG - et61x251_create_sysfs(cam); + err = et61x251_create_sysfs(cam); + if (err) + goto fail2; DBG(2, "Optional device control through 'sysfs' interface ready"); #endif @@ -2544,6 +2562,13 @@ et61x251_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) return 0; +#ifdef CONFIG_VIDEO_ADV_DEBUG +fail2: + video_nr[dev_nr] = -1; + dev_nr = (dev_nr < ET61X251_MAX_DEVICES-1) ? dev_nr+1 : 0; + mutex_unlock(&cam->dev_mutex); + video_unregister_device(cam->v4ldev); +#endif fail: if (cam) { kfree(cam->control_buffer); diff --git a/drivers/media/video/pwc/pwc-if.c b/drivers/media/video/pwc/pwc-if.c index c77b85c..46c1148 100644 --- a/drivers/media/video/pwc/pwc-if.c +++ b/drivers/media/video/pwc/pwc-if.c @@ -1024,12 +1024,25 @@ static ssize_t show_snapshot_button_status(struct class_device *class_dev, char static CLASS_DEVICE_ATTR(button, S_IRUGO | S_IWUSR, show_snapshot_button_status, NULL); -static void pwc_create_sysfs_files(struct video_device *vdev) +static int pwc_create_sysfs_files(struct video_device *vdev) { struct pwc_device *pdev = video_get_drvdata(vdev); - if (pdev->features & FEATURE_MOTOR_PANTILT) - video_device_create_file(vdev, &class_device_attr_pan_tilt); - video_device_create_file(vdev, &class_device_attr_button); + int rc; + + rc = video_device_create_file(vdev, &class_device_attr_button); + if (rc) + goto err; + if (pdev->features & FEATURE_MOTOR_PANTILT) { + rc = video_device_create_file(vdev,&class_device_attr_pan_tilt); + if (rc) goto err_button; + } + + return 0; + +err_button: + video_device_remove_file(vdev, &class_device_attr_button); +err: + return rc; } static void pwc_remove_sysfs_files(struct video_device *vdev) @@ -1408,7 +1421,7 @@ static int usb_pwc_probe(struct usb_interface *intf, const struct usb_device_id struct usb_device *udev = interface_to_usbdev(intf); struct pwc_device *pdev = NULL; int vendor_id, product_id, type_id; - int i, hint; + int i, hint, rc; int features = 0; int video_nr = -1; /* default: use next available device */ char serial_number[30], *name; @@ -1709,9 +1722,8 @@ static int usb_pwc_probe(struct usb_interface *intf, const struct usb_device_id i = video_register_device(pdev->vdev, VFL_TYPE_GRABBER, video_nr); if (i < 0) { PWC_ERROR("Failed to register as video device (%d).\n", i); - video_device_release(pdev->vdev); /* Drip... drip... drip... */ - kfree(pdev); /* Oops, no memory leaks please */ - return -EIO; + rc = i; + goto err; } else { PWC_INFO("Registered as /dev/video%d.\n", pdev->vdev->minor & 0x3F); @@ -1723,13 +1735,24 @@ static int usb_pwc_probe(struct usb_interface *intf, const struct usb_device_id PWC_DEBUG_PROBE("probe() function returning struct at 0x%p.\n", pdev); usb_set_intfdata (intf, pdev); - pwc_create_sysfs_files(pdev->vdev); + rc = pwc_create_sysfs_files(pdev->vdev); + if (rc) + goto err_unreg; /* Set the leds off */ pwc_set_leds(pdev, 0, 0); pwc_camera_power(pdev, 0); return 0; + +err_unreg: + if (hint < MAX_DEV_HINTS) + device_hint[hint].pdev = NULL; + video_unregister_device(pdev->vdev); +err: + video_device_release(pdev->vdev); /* Drip... drip... drip... */ + kfree(pdev); /* Oops, no memory leaks please */ + return rc; } /* The user janked out the cable... */ diff --git a/drivers/media/video/sn9c102/sn9c102_core.c b/drivers/media/video/sn9c102/sn9c102_core.c index 3e0ff8a..a4702d3 100644 --- a/drivers/media/video/sn9c102/sn9c102_core.c +++ b/drivers/media/video/sn9c102/sn9c102_core.c @@ -1240,23 +1240,53 @@ static CLASS_DEVICE_ATTR(frame_header, S_IRUGO, sn9c102_show_frame_header, NULL); -static void sn9c102_create_sysfs(struct sn9c102_device* cam) +static int sn9c102_create_sysfs(struct sn9c102_device* cam) { struct video_device *v4ldev = cam->v4ldev; + int rc; + + rc = video_device_create_file(v4ldev, &class_device_attr_reg); + if (rc) goto err; + rc = video_device_create_file(v4ldev, &class_device_attr_val); + if (rc) goto err_reg; + rc = video_device_create_file(v4ldev, &class_device_attr_frame_header); + if (rc) goto err_val; - video_device_create_file(v4ldev, &class_device_attr_reg); - video_device_create_file(v4ldev, &class_device_attr_val); - video_device_create_file(v4ldev, &class_device_attr_frame_header); - if (cam->bridge == BRIDGE_SN9C101 || cam->bridge == BRIDGE_SN9C102) - video_device_create_file(v4ldev, &class_device_attr_green); - else if (cam->bridge == BRIDGE_SN9C103) { - video_device_create_file(v4ldev, &class_device_attr_blue); - video_device_create_file(v4ldev, &class_device_attr_red); - } if (cam->sensor.sysfs_ops) { - video_device_create_file(v4ldev, &class_device_attr_i2c_reg); - video_device_create_file(v4ldev, &class_device_attr_i2c_val); + rc = video_device_create_file(v4ldev, &class_device_attr_i2c_reg); + if (rc) goto err_frhead; + rc = video_device_create_file(v4ldev, &class_device_attr_i2c_val); + if (rc) goto err_i2c_reg; + } + + if (cam->bridge == BRIDGE_SN9C101 || cam->bridge == BRIDGE_SN9C102) { + rc = video_device_create_file(v4ldev, &class_device_attr_green); + if (rc) goto err_i2c_val; + } else if (cam->bridge == BRIDGE_SN9C103) { + rc = video_device_create_file(v4ldev, &class_device_attr_blue); + if (rc) goto err_i2c_val; + rc = video_device_create_file(v4ldev, &class_device_attr_red); + if (rc) goto err_blue; } + + return 0; + +err_blue: + video_device_remove_file(v4ldev, &class_device_attr_blue); +err_i2c_val: + if (cam->sensor.sysfs_ops) + video_device_remove_file(v4ldev, &class_device_attr_i2c_val); +err_i2c_reg: + if (cam->sensor.sysfs_ops) + video_device_remove_file(v4ldev, &class_device_attr_i2c_reg); +err_frhead: + video_device_remove_file(v4ldev, &class_device_attr_frame_header); +err_val: + video_device_remove_file(v4ldev, &class_device_attr_val); +err_reg: + video_device_remove_file(v4ldev, &class_device_attr_reg); +err: + return rc; } #endif /* CONFIG_VIDEO_ADV_DEBUG */ @@ -2809,10 +2839,7 @@ sn9c102_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) DBG(1, "V4L2 device registration failed"); if (err == -ENFILE && video_nr[dev_nr] == -1) DBG(1, "Free /dev/videoX node not found"); - video_nr[dev_nr] = -1; - dev_nr = (dev_nr < SN9C102_MAX_DEVICES-1) ? dev_nr+1 : 0; - mutex_unlock(&cam->dev_mutex); - goto fail; + goto fail2; } DBG(2, "V4L2 device registered as /dev/video%d", cam->v4ldev->minor); @@ -2823,7 +2850,9 @@ sn9c102_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) dev_nr = (dev_nr < SN9C102_MAX_DEVICES-1) ? dev_nr+1 : 0; #ifdef CONFIG_VIDEO_ADV_DEBUG - sn9c102_create_sysfs(cam); + err = sn9c102_create_sysfs(cam); + if (err) + goto fail3; DBG(2, "Optional device control through 'sysfs' interface ready"); #endif @@ -2833,6 +2862,14 @@ sn9c102_usb_probe(struct usb_interface* intf, const struct usb_device_id* id) return 0; +#ifdef CONFIG_VIDEO_ADV_DEBUG +fail3: + video_unregister_device(cam->v4ldev); +#endif +fail2: + video_nr[dev_nr] = -1; + dev_nr = (dev_nr < SN9C102_MAX_DEVICES-1) ? dev_nr+1 : 0; + mutex_unlock(&cam->dev_mutex); fail: if (cam) { kfree(cam->control_buffer); -- cgit v0.10.2 From 83427ac5d643308ccb36e05d525949952bdedc27 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Fri, 13 Oct 2006 07:51:16 -0300 Subject: V4L/DVB (4743): Fix oops in VIDIOC_G_PARM The call to v4l2_std_construct() in the VIDIOC_G_PARM handler treats vfd->current_norm as if it were an index - but it's not. The result is an oops if the driver has no vidioc_g_parm() method defined. Here's the fix. Signed-off-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c index 98de872..d424a41 100644 --- a/drivers/media/video/videodev.c +++ b/drivers/media/video/videodev.c @@ -1288,6 +1288,7 @@ static int __video_do_ioctl(struct inode *inode, struct file *file, ret=vfd->vidioc_g_parm(file, fh, p); } else { struct v4l2_standard s; + int i; if (!vfd->tvnormsize) { printk (KERN_WARNING "%s: no TV norms defined!\n", @@ -1298,8 +1299,14 @@ static int __video_do_ioctl(struct inode *inode, struct file *file, if (p->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) return -EINVAL; - v4l2_video_std_construct(&s, vfd->tvnorms[vfd->current_norm].id, - vfd->tvnorms[vfd->current_norm].name); + for (i = 0; i < vfd->tvnormsize; i++) + if (vfd->tvnorms[i].id == vfd->current_norm) + break; + if (i >= vfd->tvnormsize) + return -EINVAL; + + v4l2_video_std_construct(&s, vfd->current_norm, + vfd->tvnorms[i].name); memset(p,0,sizeof(*p)); -- cgit v0.10.2 From 1d3e6bdaa8b4b068d378ab58679c334e433496cd Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Thu, 12 Oct 2006 15:45:33 -0300 Subject: V4L/DVB (4744): The Samsung TCPN2121P30A does not have a tda9887 Contrary to all expections the Samsung TCPN2121P30A tuner does NOT have a tda9887. Remove the tda9887 flag from the tuner definition. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/tuner-types.c b/drivers/media/video/tuner-types.c index 8fff642..7816823 100644 --- a/drivers/media/video/tuner-types.c +++ b/drivers/media/video/tuner-types.c @@ -1046,7 +1046,6 @@ static struct tuner_params tuner_samsung_tcpn_2121p30a_params[] = { .type = TUNER_PARAM_TYPE_NTSC, .ranges = tuner_samsung_tcpn_2121p30a_ntsc_ranges, .count = ARRAY_SIZE(tuner_samsung_tcpn_2121p30a_ntsc_ranges), - .has_tda9887 = 1, }, }; -- cgit v0.10.2 From 5011915cbb139a331c083e65a61c82e9174f9813 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Fri, 13 Oct 2006 05:12:42 -0300 Subject: V4L/DVB (4746): HM12 is YUV 4:2:0, not YUV 4:1:1 Fix comment in videodev2.h Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h index c5fdf62..df5c465 100644 --- a/include/linux/videodev2.h +++ b/include/linux/videodev2.h @@ -243,7 +243,7 @@ struct v4l2_pix_format #define V4L2_PIX_FMT_YUV420 v4l2_fourcc('Y','U','1','2') /* 12 YUV 4:2:0 */ #define V4L2_PIX_FMT_YYUV v4l2_fourcc('Y','Y','U','V') /* 16 YUV 4:2:2 */ #define V4L2_PIX_FMT_HI240 v4l2_fourcc('H','I','2','4') /* 8 8-bit color */ -#define V4L2_PIX_FMT_HM12 v4l2_fourcc('H','M','1','2') /* 8 YUV 4:1:1 16x16 macroblocks */ +#define V4L2_PIX_FMT_HM12 v4l2_fourcc('H','M','1','2') /* 8 YUV 4:2:0 16x16 macroblocks */ /* see http://www.siliconimaging.com/RGB%20Bayer.htm */ #define V4L2_PIX_FMT_SBGGR8 v4l2_fourcc('B','A','8','1') /* 8 BGBG.. GRGR.. */ -- cgit v0.10.2 From 6a74216c4590e4d322a45e1085f3553b1fb07f06 Mon Sep 17 00:00:00 2001 From: Patrick Boettcher Date: Fri, 13 Oct 2006 11:33:26 -0300 Subject: V4L/DVB (4748): Fixed oops for Nova-T USB2 When using the remote control with the Nova-T USB there was an Oops because of the recent DVB-USB-Adapter change. Signed-off-by: Patrick Boettcher Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/dvb-usb/dibusb.h b/drivers/media/dvb/dvb-usb/dibusb.h index 5153fb9..b607810 100644 --- a/drivers/media/dvb/dvb-usb/dibusb.h +++ b/drivers/media/dvb/dvb-usb/dibusb.h @@ -99,7 +99,9 @@ struct dibusb_state { struct dib_fe_xfer_ops ops; int mt2060_present; +}; +struct dibusb_device_state { /* for RC5 remote control */ int old_toggle; int last_repeat_count; diff --git a/drivers/media/dvb/dvb-usb/nova-t-usb2.c b/drivers/media/dvb/dvb-usb/nova-t-usb2.c index a9219bf..a58874c 100644 --- a/drivers/media/dvb/dvb-usb/nova-t-usb2.c +++ b/drivers/media/dvb/dvb-usb/nova-t-usb2.c @@ -75,7 +75,7 @@ static int nova_t_rc_query(struct dvb_usb_device *d, u32 *event, int *state) u8 key[5],cmd[2] = { DIBUSB_REQ_POLL_REMOTE, 0x35 }, data,toggle,custom; u16 raw; int i; - struct dibusb_state *st = d->priv; + struct dibusb_device_state *st = d->priv; dvb_usb_generic_rw(d,cmd,2,key,5,0); @@ -184,6 +184,7 @@ static struct dvb_usb_device_properties nova_t_properties = { .size_of_priv = sizeof(struct dibusb_state), } }, + .size_of_priv = sizeof(struct dibusb_device_state), .power_ctrl = dibusb2_0_power_ctrl, .read_mac_address = nova_t_read_mac_address, -- cgit v0.10.2 From 5570dd02ca7fb2e28d32516fae05031d48711aa5 Mon Sep 17 00:00:00 2001 From: Patrick Boettcher Date: Fri, 13 Oct 2006 11:35:12 -0300 Subject: V4L/DVB (4750): AGC command1/2 is board specific Added config-struct-parameter to take board-specific AGC command 1 and 2 into account. Signed-off-by: Patrick Boettcher Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/dvb-usb/dibusb-common.c b/drivers/media/dvb/dvb-usb/dibusb-common.c index fd3a990..5143e42 100644 --- a/drivers/media/dvb/dvb-usb/dibusb-common.c +++ b/drivers/media/dvb/dvb-usb/dibusb-common.c @@ -169,7 +169,7 @@ EXPORT_SYMBOL(dibusb_read_eeprom_byte); // Config Adjacent channels Perf -cal22 static struct dibx000_agc_config dib3000p_mt2060_agc_config = { .band_caps = BAND_VHF | BAND_UHF, - .setup = (0 << 15) | (0 << 14) | (1 << 13) | (1 << 12) | (29 << 0), + .setup = (1 << 8) | (5 << 5) | (1 << 4) | (1 << 3) | (0 << 2) | (2 << 0), .agc1_max = 48497, .agc1_min = 23593, @@ -196,10 +196,14 @@ static struct dib3000mc_config stk3000p_dib3000p_config = { .ln_adc_level = 0x1cc7, .output_mpeg2_in_188_bytes = 1, + + .agc_command1 = 1, + .agc_command2 = 1, }; static struct dibx000_agc_config dib3000p_panasonic_agc_config = { - .setup = (0 << 15) | (0 << 14) | (1 << 13) | (1 << 12) | (29 << 0), + .band_caps = BAND_VHF | BAND_UHF, + .setup = (1 << 8) | (5 << 5) | (1 << 4) | (1 << 3) | (0 << 2) | (2 << 0), .agc1_max = 56361, .agc1_min = 22282, @@ -226,6 +230,9 @@ static struct dib3000mc_config mod3000p_dib3000p_config = { .ln_adc_level = 0x1cc7, .output_mpeg2_in_188_bytes = 1, + + .agc_command1 = 1, + .agc_command2 = 1, }; int dibusb_dib3000mc_frontend_attach(struct dvb_usb_adapter *adap) diff --git a/drivers/media/dvb/frontends/dib3000mc.c b/drivers/media/dvb/frontends/dib3000mc.c index ccc813b..3561a77 100644 --- a/drivers/media/dvb/frontends/dib3000mc.c +++ b/drivers/media/dvb/frontends/dib3000mc.c @@ -345,7 +345,7 @@ static int dib3000mc_init(struct dvb_frontend *demod) /* agc */ dib3000mc_write_word(state, 36, state->cfg->max_time); - dib3000mc_write_word(state, 37, agc->setup); + dib3000mc_write_word(state, 37, (state->cfg->agc_command1 << 13) | (state->cfg->agc_command2 << 12) | (0x1d << 0)); dib3000mc_write_word(state, 38, state->cfg->pwm3_value); dib3000mc_write_word(state, 39, state->cfg->ln_adc_level); diff --git a/drivers/media/dvb/frontends/dib3000mc.h b/drivers/media/dvb/frontends/dib3000mc.h index b198cd5..0d6fdef 100644 --- a/drivers/media/dvb/frontends/dib3000mc.h +++ b/drivers/media/dvb/frontends/dib3000mc.h @@ -28,6 +28,9 @@ struct dib3000mc_config { u16 max_time; u16 ln_adc_level; + u8 agc_command1 :1; + u8 agc_command2 :1; + u8 mobile_mode; u8 output_mpeg2_in_188_bytes; -- cgit v0.10.2 From 288f3ad406460f03642a41bb945826891a7b866f Mon Sep 17 00:00:00 2001 From: Marek W Date: Mon, 14 Aug 2006 22:37:20 -0700 Subject: ACPI: asus_acpi: W3000 support Add support for W3000 (W3V) and indirectly fixes an issue with kmilo under KDE (it was triggering excessive LCD read error messages by querying asus_acpi module) allowing people (I am probably the only one who tested this) with W3000 to run kmilo. Cc: Karol Kozimor Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/asus_acpi.c b/drivers/acpi/asus_acpi.c index e9ee4c5..ebc033f 100644 --- a/drivers/acpi/asus_acpi.c +++ b/drivers/acpi/asus_acpi.c @@ -138,6 +138,7 @@ struct asus_hotk { S2x, //S200 (J1 reported), Victor MP-XP7210 W1N, //W1000N W5A, //W5A + W3V, //W3030V xxN, //M2400N, M3700N, M5200N, M6800N, S1300N, S5200N //(Centrino) END_MODEL @@ -376,6 +377,17 @@ static struct model_data model_conf[END_MODEL] = { .display_get = "\\ADVG"}, { + .name = "W3V", + .mt_mled = "MLED", + .mt_wled = "WLED", + .mt_lcd_switch = xxN_PREFIX "_Q10", + .lcd_status = "\\BKLT", + .brightness_set = "SPLV", + .brightness_get = "GPLV", + .display_set = "SDSP", + .display_get = "\\INFB"}, + + { .name = "xxN", .mt_mled = "MLED", /* WLED present, but not controlled by ACPI */ @@ -1097,6 +1109,8 @@ static int asus_model_match(char *model) return A4G; else if (strncmp(model, "W1N", 3) == 0) return W1N; + else if (strncmp(model, "W3V", 3) == 0) + return W3V; else if (strncmp(model, "W5A", 3) == 0) return W5A; else @@ -1200,9 +1214,10 @@ static int asus_hotk_get_info(void) hotk->methods->mt_wled = NULL; /* L5D's WLED is not controlled by ACPI */ else if (strncmp(string, "M2N", 3) == 0 || + strncmp(string, "W3V", 3) == 0 || strncmp(string, "S1N", 3) == 0) hotk->methods->mt_wled = "WLED"; - /* M2N and S1N have a usable WLED */ + /* M2N, S1N and W3V have a usable WLED */ else if (asus_info) { if (strncmp(asus_info->oem_table_id, "L1", 2) == 0) hotk->methods->mled_status = NULL; -- cgit v0.10.2 From 4d6bd5ea4ec4991901a8cf5a586babef68e1fa3f Mon Sep 17 00:00:00 2001 From: Stefan Schmidt Date: Fri, 22 Sep 2006 12:19:14 +0200 Subject: ACPI: ibm_acpi: Remove experimental status for brightness and volume. The brightness and volume features from ibm-acpi are stable. The experimental flag is no longer needed. Signed-off-by: Stefan Schmidt Acked-by: Borislav Deianov Signed-off-by: Len Brown diff --git a/drivers/acpi/ibm_acpi.c b/drivers/acpi/ibm_acpi.c index 15fc124..003a987 100644 --- a/drivers/acpi/ibm_acpi.c +++ b/drivers/acpi/ibm_acpi.c @@ -1702,13 +1702,11 @@ static struct ibm_struct ibms[] = { .name = "brightness", .read = brightness_read, .write = brightness_write, - .experimental = 1, }, { .name = "volume", .read = volume_read, .write = volume_write, - .experimental = 1, }, { .name = "fan", -- cgit v0.10.2 From 24f7ff0af855ece60064a2532d8b316df02983c6 Mon Sep 17 00:00:00 2001 From: Stefan Schmidt Date: Fri, 22 Sep 2006 12:19:15 +0200 Subject: ACPI: ibm_acpi: Update documentation for brightness and volume. Document the change of the experimental flag for brightness and volume. Signed-off-by: Stefan Schmidt Acked-by: Borislav Deianov Signed-off-by: Len Brown diff --git a/Documentation/ibm-acpi.txt b/Documentation/ibm-acpi.txt index 71aa403..1672590 100644 --- a/Documentation/ibm-acpi.txt +++ b/Documentation/ibm-acpi.txt @@ -30,8 +30,8 @@ detailed description): - ACPI sounds - temperature sensors - Experimental: embedded controller register dump - - Experimental: LCD brightness control - - Experimental: volume control + - LCD brightness control + - Volume control - Experimental: fan speed, fan enable/disable A compatibility table by model and feature is maintained on the web @@ -523,13 +523,8 @@ registers contain the current battery capacity, etc. If you experiment with this, do send me your results (including some complete dumps with a description of the conditions when they were taken.) -EXPERIMENTAL: LCD brightness control -- /proc/acpi/ibm/brightness ------------------------------------------------------------------ - -This feature is marked EXPERIMENTAL because the implementation -directly accesses hardware registers and may not work as expected. USE -WITH CAUTION! To use this feature, you need to supply the -experimental=1 parameter when loading the module. +LCD brightness control -- /proc/acpi/ibm/brightness +--------------------------------------------------- This feature allows software control of the LCD brightness on ThinkPad models which don't have a hardware brightness slider. The available @@ -542,13 +537,8 @@ commands are: The number range is 0 to 7, although not all of them may be distinct. The current brightness level is shown in the file. -EXPERIMENTAL: Volume control -- /proc/acpi/ibm/volume ------------------------------------------------------ - -This feature is marked EXPERIMENTAL because the implementation -directly accesses hardware registers and may not work as expected. USE -WITH CAUTION! To use this feature, you need to supply the -experimental=1 parameter when loading the module. +Volume control -- /proc/acpi/ibm/volume +--------------------------------------- This feature allows volume control on ThinkPad models which don't have a hardware volume knob. The available commands are: -- cgit v0.10.2 From 28b779d127d3038ff83f42259d135a063b7cd848 Mon Sep 17 00:00:00 2001 From: Stefan Schmidt Date: Fri, 22 Sep 2006 12:19:16 +0200 Subject: ACPI: ibm_acpi: Documentation the wan feature. Document the wan feature Jeremy Fitzhardinge added to ibm_acpi. Signed-off-by: Stefan Schmidt Acked-by: Borislav Deianov Signed-off-by: Len Brown diff --git a/Documentation/ibm-acpi.txt b/Documentation/ibm-acpi.txt index 1672590..00b8cf39 100644 --- a/Documentation/ibm-acpi.txt +++ b/Documentation/ibm-acpi.txt @@ -33,6 +33,7 @@ detailed description): - LCD brightness control - Volume control - Experimental: fan speed, fan enable/disable + - Experimental: WAN enable and disable A compatibility table by model and feature is maintained on the web site, http://ibm-acpi.sf.net/. I appreciate any success or failure @@ -601,6 +602,23 @@ with the following command: echo 'level ' > /proc/acpi/ibm/thermal +EXPERIMENTAL: WAN -- /proc/acpi/ibm/wan +--------------------------------------- + +This feature is marked EXPERIMENTAL because the implementation +directly accesses hardware registers and may not work as expected. USE +WITH CAUTION! To use this feature, you need to supply the +experimental=1 parameter when loading the module. + +This feature shows the presence and current state of a WAN (Sierra +Wireless EV-DO) device. If WAN is installed, the following commands can +be used: + + echo enable > /proc/acpi/ibm/wan + echo disable > /proc/acpi/ibm/wan + +It was tested on a Lenovo Thinkpad X60. It should probably work on other +Thinkpad models which come with this module installed. Multiple Commands, Module Parameters ------------------------------------ -- cgit v0.10.2 From 963497c12acb4d43caa9751b9291b014eea51a1a Mon Sep 17 00:00:00 2001 From: "Lebedev, Vladimir P" Date: Tue, 5 Sep 2006 19:49:13 +0400 Subject: ACPI: sbs: check for NULL device pointer Signed-off-by: Len Brown diff --git a/drivers/acpi/sbs.c b/drivers/acpi/sbs.c index 62bef0b..79f38f0 100644 --- a/drivers/acpi/sbs.c +++ b/drivers/acpi/sbs.c @@ -1685,10 +1685,16 @@ static int acpi_sbs_add(struct acpi_device *device) int acpi_sbs_remove(struct acpi_device *device, int type) { - struct acpi_sbs *sbs = (struct acpi_sbs *)acpi_driver_data(device); + struct acpi_sbs *sbs = NULL; int id; - if (!device || !sbs) { + if (!device) { + return -EINVAL; + } + + sbs = (struct acpi_sbs *)acpi_driver_data(device); + + if (!sbs) { return -EINVAL; } -- cgit v0.10.2 From 3cd5b87d96db503f69a5892b8f5350d356d18969 Mon Sep 17 00:00:00 2001 From: "Lebedev, Vladimir P" Date: Tue, 5 Sep 2006 19:59:22 +0400 Subject: ACPI: sbs: fix module_param() initializers Signed-off-by: Len Brown diff --git a/drivers/acpi/sbs.c b/drivers/acpi/sbs.c index 79f38f0..8908a97 100644 --- a/drivers/acpi/sbs.c +++ b/drivers/acpi/sbs.c @@ -98,11 +98,11 @@ static int update_info_mode = UPDATE_INFO_MODE; static int update_time = UPDATE_TIME; static int update_time2 = UPDATE_TIME2; -module_param(capacity_mode, int, CAPACITY_UNIT); -module_param(update_mode, int, UPDATE_MODE); -module_param(update_info_mode, int, UPDATE_INFO_MODE); -module_param(update_time, int, UPDATE_TIME); -module_param(update_time2, int, UPDATE_TIME2); +module_param(capacity_mode, int, 0); +module_param(update_mode, int, 0); +module_param(update_info_mode, int, 0); +module_param(update_time, int, 0); +module_param(update_time2, int, 0); static int acpi_sbs_add(struct acpi_device *device); static int acpi_sbs_remove(struct acpi_device *device, int type); -- cgit v0.10.2 From 991528d7348667924176f3e29addea0675298944 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Mon, 25 Sep 2006 16:28:13 -0700 Subject: ACPI: Processor native C-states using MWAIT Intel processors starting with the Core Duo support support processor native C-state using the MWAIT instruction. Refer: Intel Architecture Software Developer's Manual http://www.intel.com/design/Pentium4/manuals/253668.htm Platform firmware exports the support for Native C-state to OS using ACPI _PDC and _CST methods. Refer: Intel Processor Vendor-Specific ACPI: Interface Specification http://www.intel.com/technology/iapc/acpi/downloads/302223.htm With Processor Native C-state, we use 'MWAIT' instruction on the processor to enter different C-states (C1, C2, C3). We won't use the special IO ports to enter C-state and no SMM mode etc required to enter C-state. Overall this will mean better C-state support. One major advantage of using MWAIT for all C-states is, with this and "treat interrupt as break event" feature of MWAIT, we can now get accurate timing for the time spent in C1, C2, .. states. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/arch/i386/kernel/acpi/cstate.c b/arch/i386/kernel/acpi/cstate.c index 25db49e..20563e5 100644 --- a/arch/i386/kernel/acpi/cstate.c +++ b/arch/i386/kernel/acpi/cstate.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include @@ -41,5 +42,124 @@ void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags, flags->bm_check = 1; } } - EXPORT_SYMBOL(acpi_processor_power_init_bm_check); + +/* The code below handles cstate entry with monitor-mwait pair on Intel*/ + +struct cstate_entry_s { + struct { + unsigned int eax; + unsigned int ecx; + } states[ACPI_PROCESSOR_MAX_POWER]; +}; +static struct cstate_entry_s *cpu_cstate_entry; /* per CPU ptr */ + +static short mwait_supported[ACPI_PROCESSOR_MAX_POWER]; + +#define MWAIT_SUBSTATE_MASK (0xf) +#define MWAIT_SUBSTATE_SIZE (4) + +#define CPUID_MWAIT_LEAF (5) +#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1) +#define CPUID5_ECX_INTERRUPT_BREAK (0x2) + +#define MWAIT_ECX_INTERRUPT_BREAK (0x1) + +#define NATIVE_CSTATE_BEYOND_HALT (2) + +int acpi_processor_ffh_cstate_probe(unsigned int cpu, + struct acpi_processor_cx *cx, struct acpi_power_register *reg) +{ + struct cstate_entry_s *percpu_entry; + struct cpuinfo_x86 *c = cpu_data + cpu; + + cpumask_t saved_mask; + int retval; + unsigned int eax, ebx, ecx, edx; + unsigned int edx_part; + unsigned int cstate_type; /* C-state type and not ACPI C-state type */ + unsigned int num_cstate_subtype; + + if (!cpu_cstate_entry || c->cpuid_level < CPUID_MWAIT_LEAF ) + return -1; + + if (reg->bit_offset != NATIVE_CSTATE_BEYOND_HALT) + return -1; + + percpu_entry = per_cpu_ptr(cpu_cstate_entry, cpu); + percpu_entry->states[cx->index].eax = 0; + percpu_entry->states[cx->index].ecx = 0; + + /* Make sure we are running on right CPU */ + saved_mask = current->cpus_allowed; + retval = set_cpus_allowed(current, cpumask_of_cpu(cpu)); + if (retval) + return -1; + + cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &edx); + + /* Check whether this particular cx_type (in CST) is supported or not */ + cstate_type = (cx->address >> MWAIT_SUBSTATE_SIZE) + 1; + edx_part = edx >> (cstate_type * MWAIT_SUBSTATE_SIZE); + num_cstate_subtype = edx_part & MWAIT_SUBSTATE_MASK; + + retval = 0; + if (num_cstate_subtype < (cx->address & MWAIT_SUBSTATE_MASK)) { + retval = -1; + goto out; + } + + /* mwait ecx extensions INTERRUPT_BREAK should be supported for C2/C3 */ + if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) || + !(ecx & CPUID5_ECX_INTERRUPT_BREAK)) { + retval = -1; + goto out; + } + percpu_entry->states[cx->index].ecx = MWAIT_ECX_INTERRUPT_BREAK; + + /* Use the hint in CST */ + percpu_entry->states[cx->index].eax = cx->address; + + if (!mwait_supported[cstate_type]) { + mwait_supported[cstate_type] = 1; + printk(KERN_DEBUG "Monitor-Mwait will be used to enter C-%d " + "state\n", cx->type); + } + +out: + set_cpus_allowed(current, saved_mask); + return retval; +} +EXPORT_SYMBOL_GPL(acpi_processor_ffh_cstate_probe); + +void acpi_processor_ffh_cstate_enter(struct acpi_processor_cx *cx) +{ + unsigned int cpu = smp_processor_id(); + struct cstate_entry_s *percpu_entry; + + percpu_entry = per_cpu_ptr(cpu_cstate_entry, cpu); + mwait_idle_with_hints(percpu_entry->states[cx->index].eax, + percpu_entry->states[cx->index].ecx); +} +EXPORT_SYMBOL_GPL(acpi_processor_ffh_cstate_enter); + +static int __init ffh_cstate_init(void) +{ + struct cpuinfo_x86 *c = &boot_cpu_data; + if (c->x86_vendor != X86_VENDOR_INTEL) + return -1; + + cpu_cstate_entry = alloc_percpu(struct cstate_entry_s); + return 0; +} + +static void __exit ffh_cstate_exit(void) +{ + if (cpu_cstate_entry) { + free_percpu(cpu_cstate_entry); + cpu_cstate_entry = NULL; + } +} + +arch_initcall(ffh_cstate_init); +__exitcall(ffh_cstate_exit); diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index b0a0780..57d3759 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -236,20 +236,28 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait); * We execute MONITOR against need_resched and enter optimized wait state * through MWAIT. Whenever someone changes need_resched, we would be woken * up from MWAIT (without an IPI). + * + * New with Core Duo processors, MWAIT can take some hints based on CPU + * capability. */ -static void mwait_idle(void) +void mwait_idle_with_hints(unsigned long eax, unsigned long ecx) { - local_irq_enable(); - - while (!need_resched()) { + if (!need_resched()) { __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); - if (need_resched()) - break; - __mwait(0, 0); + if (!need_resched()) + __mwait(eax, ecx); } } +/* Default MONITOR/MWAIT with no hints, used for default C1 state */ +static void mwait_idle(void) +{ + local_irq_enable(); + while (!need_resched()) + mwait_idle_with_hints(0, 0); +} + void __devinit select_idle_routine(const struct cpuinfo_x86 *c) { if (cpu_has(c, X86_FEATURE_MWAIT)) { diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 5e95b25..49f7fac 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -238,20 +238,28 @@ void cpu_idle (void) * We execute MONITOR against need_resched and enter optimized wait state * through MWAIT. Whenever someone changes need_resched, we would be woken * up from MWAIT (without an IPI). + * + * New with Core Duo processors, MWAIT can take some hints based on CPU + * capability. */ -static void mwait_idle(void) +void mwait_idle_with_hints(unsigned long eax, unsigned long ecx) { - local_irq_enable(); - - while (!need_resched()) { + if (!need_resched()) { __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); - if (need_resched()) - break; - __mwait(0, 0); + if (!need_resched()) + __mwait(eax, ecx); } } +/* Default MONITOR/MWAIT with no hints, used for default C1 state */ +static void mwait_idle(void) +{ + local_irq_enable(); + while (!need_resched()) + mwait_idle_with_hints(0,0); +} + void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) { static int printed; diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 0a395fc..429a39d 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -219,6 +219,23 @@ static void acpi_safe_halt(void) static atomic_t c3_cpu_count; +/* Common C-state entry for C2, C3, .. */ +static void acpi_cstate_enter(struct acpi_processor_cx *cstate) +{ + if (cstate->space_id == ACPI_CSTATE_FFH) { + /* Call into architectural FFH based C-state */ + acpi_processor_ffh_cstate_enter(cstate); + } else { + int unused; + /* IO port based C-state */ + inb(cstate->address); + /* Dummy wait op - must do something useless after P_LVL2 read + because chipsets cannot guarantee that STPCLK# signal + gets asserted in time to freeze execution properly. */ + unused = inl(acpi_fadt.xpm_tmr_blk.address); + } +} + static void acpi_processor_idle(void) { struct acpi_processor *pr = NULL; @@ -361,11 +378,7 @@ static void acpi_processor_idle(void) /* Get start time (ticks) */ t1 = inl(acpi_fadt.xpm_tmr_blk.address); /* Invoke C2 */ - inb(cx->address); - /* Dummy wait op - must do something useless after P_LVL2 read - because chipsets cannot guarantee that STPCLK# signal - gets asserted in time to freeze execution properly. */ - t2 = inl(acpi_fadt.xpm_tmr_blk.address); + acpi_cstate_enter(cx); /* Get end time (ticks) */ t2 = inl(acpi_fadt.xpm_tmr_blk.address); @@ -401,9 +414,7 @@ static void acpi_processor_idle(void) /* Get start time (ticks) */ t1 = inl(acpi_fadt.xpm_tmr_blk.address); /* Invoke C3 */ - inb(cx->address); - /* Dummy wait op (see above) */ - t2 = inl(acpi_fadt.xpm_tmr_blk.address); + acpi_cstate_enter(cx); /* Get end time (ticks) */ t2 = inl(acpi_fadt.xpm_tmr_blk.address); if (pr->flags.bm_check) { @@ -628,20 +639,16 @@ static int acpi_processor_get_power_info_fadt(struct acpi_processor *pr) return 0; } -static int acpi_processor_get_power_info_default_c1(struct acpi_processor *pr) +static int acpi_processor_get_power_info_default(struct acpi_processor *pr) { - - /* Zero initialize all the C-states info. */ - memset(pr->power.states, 0, sizeof(pr->power.states)); - - /* set the first C-State to C1 */ - pr->power.states[ACPI_STATE_C1].type = ACPI_STATE_C1; - - /* the C0 state only exists as a filler in our array, - * and all processors need to support C1 */ + if (!pr->power.states[ACPI_STATE_C1].valid) { + /* set the first C-State to C1 */ + /* all processors need to support C1 */ + pr->power.states[ACPI_STATE_C1].type = ACPI_STATE_C1; + pr->power.states[ACPI_STATE_C1].valid = 1; + } + /* the C0 state only exists as a filler in our array */ pr->power.states[ACPI_STATE_C0].valid = 1; - pr->power.states[ACPI_STATE_C1].valid = 1; - return 0; } @@ -658,12 +665,7 @@ static int acpi_processor_get_power_info_cst(struct acpi_processor *pr) if (nocst) return -ENODEV; - current_count = 1; - - /* Zero initialize C2 onwards and prepare for fresh CST lookup */ - for (i = 2; i < ACPI_PROCESSOR_MAX_POWER; i++) - memset(&(pr->power.states[i]), 0, - sizeof(struct acpi_processor_cx)); + current_count = 0; status = acpi_evaluate_object(pr->handle, "_CST", NULL, &buffer); if (ACPI_FAILURE(status)) { @@ -718,22 +720,39 @@ static int acpi_processor_get_power_info_cst(struct acpi_processor *pr) (reg->space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) continue; - cx.address = (reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE) ? - 0 : reg->address; - /* There should be an easy way to extract an integer... */ obj = (union acpi_object *)&(element->package.elements[1]); if (obj->type != ACPI_TYPE_INTEGER) continue; cx.type = obj->integer.value; - - if ((cx.type != ACPI_STATE_C1) && - (reg->space_id != ACPI_ADR_SPACE_SYSTEM_IO)) - continue; - - if ((cx.type < ACPI_STATE_C2) || (cx.type > ACPI_STATE_C3)) - continue; + /* + * Some buggy BIOSes won't list C1 in _CST - + * Let acpi_processor_get_power_info_default() handle them later + */ + if (i == 1 && cx.type != ACPI_STATE_C1) + current_count++; + + cx.address = reg->address; + cx.index = current_count + 1; + + cx.space_id = ACPI_CSTATE_SYSTEMIO; + if (reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE) { + if (acpi_processor_ffh_cstate_probe + (pr->id, &cx, reg) == 0) { + cx.space_id = ACPI_CSTATE_FFH; + } else if (cx.type != ACPI_STATE_C1) { + /* + * C1 is a special case where FIXED_HARDWARE + * can be handled in non-MWAIT way as well. + * In that case, save this _CST entry info. + * That is, we retain space_id of SYSTEM_IO for + * halt based C1. + * Otherwise, ignore this info and continue. + */ + continue; + } + } obj = (union acpi_object *)&(element->package.elements[2]); if (obj->type != ACPI_TYPE_INTEGER) @@ -938,12 +957,18 @@ static int acpi_processor_get_power_info(struct acpi_processor *pr) /* NOTE: the idle thread may not be running while calling * this function */ - /* Adding C1 state */ - acpi_processor_get_power_info_default_c1(pr); + /* Zero initialize all the C-states info. */ + memset(pr->power.states, 0, sizeof(pr->power.states)); + result = acpi_processor_get_power_info_cst(pr); if (result == -ENODEV) acpi_processor_get_power_info_fadt(pr); + if (result) + return result; + + acpi_processor_get_power_info_default(pr); + pr->power.count = acpi_processor_power_verify(pr); /* diff --git a/include/acpi/pdc_intel.h b/include/acpi/pdc_intel.h index c5472be..e72bfdd 100644 --- a/include/acpi/pdc_intel.h +++ b/include/acpi/pdc_intel.h @@ -13,6 +13,7 @@ #define ACPI_PDC_SMP_C_SWCOORD (0x0040) #define ACPI_PDC_SMP_T_SWCOORD (0x0080) #define ACPI_PDC_C_C1_FFH (0x0100) +#define ACPI_PDC_C_C2C3_FFH (0x0200) #define ACPI_PDC_EST_CAPABILITY_SMP (ACPI_PDC_SMP_C1PT | \ ACPI_PDC_C_C1_HALT | \ @@ -23,8 +24,10 @@ ACPI_PDC_SMP_P_SWCOORD | \ ACPI_PDC_P_FFH) -#define ACPI_PDC_C_CAPABILITY_SMP (ACPI_PDC_SMP_C2C3 | \ - ACPI_PDC_SMP_C1PT | \ - ACPI_PDC_C_C1_HALT) +#define ACPI_PDC_C_CAPABILITY_SMP (ACPI_PDC_SMP_C2C3 | \ + ACPI_PDC_SMP_C1PT | \ + ACPI_PDC_C_C1_HALT | \ + ACPI_PDC_C_C1_FFH | \ + ACPI_PDC_C_C2C3_FFH) #endif /* __PDC_INTEL_H__ */ diff --git a/include/acpi/processor.h b/include/acpi/processor.h index 9dd5b75..7798d2a 100644 --- a/include/acpi/processor.h +++ b/include/acpi/processor.h @@ -29,6 +29,9 @@ #define DOMAIN_COORD_TYPE_SW_ANY 0xfd #define DOMAIN_COORD_TYPE_HW_ALL 0xfe +#define ACPI_CSTATE_SYSTEMIO (0) +#define ACPI_CSTATE_FFH (1) + /* Power Management */ struct acpi_processor_cx; @@ -58,6 +61,8 @@ struct acpi_processor_cx { u8 valid; u8 type; u32 address; + u8 space_id; + u8 index; u32 latency; u32 latency_ticks; u32 power; @@ -206,6 +211,9 @@ void arch_acpi_processor_init_pdc(struct acpi_processor *pr); #ifdef ARCH_HAS_POWER_INIT void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags, unsigned int cpu); +int acpi_processor_ffh_cstate_probe(unsigned int cpu, + struct acpi_processor_cx *cx, struct acpi_power_register *reg); +void acpi_processor_ffh_cstate_enter(struct acpi_processor_cx *cstate); #else static inline void acpi_processor_power_init_bm_check(struct acpi_processor_flags @@ -214,6 +222,16 @@ static inline void acpi_processor_power_init_bm_check(struct flags->bm_check = 1; return; } +static inline int acpi_processor_ffh_cstate_probe(unsigned int cpu, + struct acpi_processor_cx *cx, struct acpi_power_register *reg) +{ + return -1; +} +static inline void acpi_processor_ffh_cstate_enter( + struct acpi_processor_cx *cstate) +{ + return; +} #endif /* in processor_perflib.c */ diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index 2277127..e0ddca9 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -306,6 +306,8 @@ static inline void __mwait(unsigned long eax, unsigned long ecx) : :"a" (eax), "c" (ecx)); } +extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx); + /* from system description table in BIOS. Mostly for MCA use, but others may find it useful. */ extern unsigned int machine_id; diff --git a/include/asm-x86_64/processor.h b/include/asm-x86_64/processor.h index de9c314..cef17e0 100644 --- a/include/asm-x86_64/processor.h +++ b/include/asm-x86_64/processor.h @@ -475,6 +475,8 @@ static inline void __mwait(unsigned long eax, unsigned long ecx) : :"a" (eax), "c" (ecx)); } +extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx); + #define stack_current() \ ({ \ struct thread_info *ti; \ -- cgit v0.10.2 From d7a76e4cb3b4469b1eccb6204c053e3ebcd4c196 Mon Sep 17 00:00:00 2001 From: Lennart Poettering Date: Tue, 5 Sep 2006 12:12:24 -0400 Subject: ACPI: consolidate functions in acpi ec driver Unify the following functions: acpi_ec_poll_read() acpi_ec_poll_write() acpi_ec_poll_query() acpi_ec_intr_read() acpi_ec_intr_write() acpi_ec_intr_query() into: acpi_ec_poll_transaction() acpi_ec_intr_transaction() These new functions take as arguments an ACPI EC command, a few bytes to write to the EC data register and a buffer for a few bytes to read from the EC data register. The old _read(), _write(), _query() are just special cases of these functions. Then unified the code in acpi_ec_poll_transaction() and acpi_ec_intr_transaction() a little more. Both functions are now just wrappers around the new acpi_ec_transaction_unlocked() function. The latter contains the EC access logic, the two original function now just do their special way of locking and call the the new function for the actual work. This saves a lot of very similar code. The primary reason for doing this, however, is that my driver for MSI 270 laptops needs to issue some non-standard EC commands in a safe way. Due to this I added a new exported function similar to ec_write()/ec_write() which is called ec_transaction() and is essentially just a wrapper around acpi_ec_{poll,intr}_transaction(). Signed-off-by: Lennart Poettering Acked-by: Luming Yu Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index e5d7963..a0dcbad 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -122,12 +122,12 @@ union acpi_ec { static int acpi_ec_poll_wait(union acpi_ec *ec, u8 event); static int acpi_ec_intr_wait(union acpi_ec *ec, unsigned int event); -static int acpi_ec_poll_read(union acpi_ec *ec, u8 address, u32 * data); -static int acpi_ec_intr_read(union acpi_ec *ec, u8 address, u32 * data); -static int acpi_ec_poll_write(union acpi_ec *ec, u8 address, u8 data); -static int acpi_ec_intr_write(union acpi_ec *ec, u8 address, u8 data); -static int acpi_ec_poll_query(union acpi_ec *ec, u32 * data); -static int acpi_ec_intr_query(union acpi_ec *ec, u32 * data); +static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len); +static int acpi_ec_intr_transaction(union acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len); static void acpi_ec_gpe_poll_query(void *ec_cxt); static void acpi_ec_gpe_intr_query(void *ec_cxt); static u32 acpi_ec_gpe_poll_handler(void *data); @@ -302,110 +302,95 @@ end: } #endif /* ACPI_FUTURE_USAGE */ -static int acpi_ec_read(union acpi_ec *ec, u8 address, u32 * data) +static int acpi_ec_transaction(union acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { if (acpi_ec_poll_mode) - return acpi_ec_poll_read(ec, address, data); + return acpi_ec_poll_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); else - return acpi_ec_intr_read(ec, address, data); + return acpi_ec_intr_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); +} +static int acpi_ec_read(union acpi_ec *ec, u8 address, u32 * data) +{ + int result; + u8 d; + result = acpi_ec_transaction(ec, ACPI_EC_COMMAND_READ, &address, 1, &d, 1); + *data = d; + return result; } static int acpi_ec_write(union acpi_ec *ec, u8 address, u8 data) { - if (acpi_ec_poll_mode) - return acpi_ec_poll_write(ec, address, data); - else - return acpi_ec_intr_write(ec, address, data); + u8 wdata[2] = { address, data }; + return acpi_ec_transaction(ec, ACPI_EC_COMMAND_WRITE, wdata, 2, NULL, 0); } -static int acpi_ec_poll_read(union acpi_ec *ec, u8 address, u32 * data) + +static int acpi_ec_transaction_unlocked(union acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { - acpi_status status = AE_OK; - int result = 0; - u32 glk = 0; + int result; + acpi_hw_low_level_write(8, command, &ec->common.command_addr); - if (!ec || !data) - return -EINVAL; + result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + if (result) + return result; - *data = 0; + for (; wdata_len > 0; wdata_len --) { - if (ec->common.global_lock) { - status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); - if (ACPI_FAILURE(status)) - return -ENODEV; - } + acpi_hw_low_level_write(8, *(wdata++), &ec->common.data_addr); - if (down_interruptible(&ec->poll.sem)) { - result = -ERESTARTSYS; - goto end_nosem; - } - - acpi_hw_low_level_write(8, ACPI_EC_COMMAND_READ, - &ec->common.command_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (result) - goto end; + result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + if (result) + return result; + } - acpi_hw_low_level_write(8, address, &ec->common.data_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); - if (result) - goto end; - acpi_hw_low_level_read(8, data, &ec->common.data_addr); + for (; rdata_len > 0; rdata_len --) { + u32 d; - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Read [%02x] from address [%02x]\n", - *data, address)); + result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); + if (result) + return result; - end: - up(&ec->poll.sem); -end_nosem: - if (ec->common.global_lock) - acpi_release_global_lock(glk); + acpi_hw_low_level_read(8, &d, &ec->common.data_addr); + *(rdata++) = (u8) d; + } - return result; + return 0; } -static int acpi_ec_poll_write(union acpi_ec *ec, u8 address, u8 data) +static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { - int result = 0; acpi_status status = AE_OK; + int result; u32 glk = 0; - - if (!ec) + if (!ec || (wdata_len && !wdata) || (rdata_len && !rdata)) return -EINVAL; + if (rdata) + memset(rdata, 0, rdata_len); + if (ec->common.global_lock) { status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); if (ACPI_FAILURE(status)) return -ENODEV; - } + } if (down_interruptible(&ec->poll.sem)) { result = -ERESTARTSYS; goto end_nosem; } - - acpi_hw_low_level_write(8, ACPI_EC_COMMAND_WRITE, - &ec->common.command_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (result) - goto end; - acpi_hw_low_level_write(8, address, &ec->common.data_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (result) - goto end; - - acpi_hw_low_level_write(8, data, &ec->common.data_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (result) - goto end; - - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Wrote [%02x] to address [%02x]\n", - data, address)); - - end: + result = acpi_ec_transaction_unlocked(ec, command, + wdata, wdata_len, + rdata, rdata_len); up(&ec->poll.sem); + end_nosem: if (ec->common.global_lock) acpi_release_global_lock(glk); @@ -413,16 +398,18 @@ end_nosem: return result; } -static int acpi_ec_intr_read(union acpi_ec *ec, u8 address, u32 * data) +static int acpi_ec_intr_transaction(union acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { - int status = 0; + int status; u32 glk; - - if (!ec || !data) + if (!ec || (wdata_len && !wdata) || (rdata_len && !rdata)) return -EINVAL; - *data = 0; + if (rdata) + memset(rdata, 0, rdata_len); if (ec->common.global_lock) { status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); @@ -438,72 +425,12 @@ static int acpi_ec_intr_read(union acpi_ec *ec, u8 address, u32 * data) printk(KERN_DEBUG PREFIX "read EC, IB not empty\n"); goto end; } - acpi_hw_low_level_write(8, ACPI_EC_COMMAND_READ, - &ec->common.command_addr); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (status) { - printk(KERN_DEBUG PREFIX "read EC, IB not empty\n"); - } - - acpi_hw_low_level_write(8, address, &ec->common.data_addr); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); - if (status) { - printk(KERN_DEBUG PREFIX "read EC, OB not full\n"); - goto end; - } - acpi_hw_low_level_read(8, data, &ec->common.data_addr); - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Read [%02x] from address [%02x]\n", - *data, address)); - - end: - up(&ec->intr.sem); - - if (ec->common.global_lock) - acpi_release_global_lock(glk); - - return status; -} - -static int acpi_ec_intr_write(union acpi_ec *ec, u8 address, u8 data) -{ - int status = 0; - u32 glk; - - - if (!ec) - return -EINVAL; - - if (ec->common.global_lock) { - status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); - if (ACPI_FAILURE(status)) - return -ENODEV; - } - - WARN_ON(in_interrupt()); - down(&ec->intr.sem); - - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (status) { - printk(KERN_DEBUG PREFIX "write EC, IB not empty\n"); - } - acpi_hw_low_level_write(8, ACPI_EC_COMMAND_WRITE, - &ec->common.command_addr); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (status) { - printk(KERN_DEBUG PREFIX "write EC, IB not empty\n"); - } - - acpi_hw_low_level_write(8, address, &ec->common.data_addr); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (status) { - printk(KERN_DEBUG PREFIX "write EC, IB not empty\n"); - } - acpi_hw_low_level_write(8, data, &ec->common.data_addr); - - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Wrote [%02x] to address [%02x]\n", - data, address)); + status = acpi_ec_transaction_unlocked(ec, command, + wdata, wdata_len, + rdata, rdata_len); +end: up(&ec->intr.sem); if (ec->common.global_lock) @@ -554,106 +481,44 @@ int ec_write(u8 addr, u8 val) EXPORT_SYMBOL(ec_write); -static int acpi_ec_query(union acpi_ec *ec, u32 * data) -{ - if (acpi_ec_poll_mode) - return acpi_ec_poll_query(ec, data); - else - return acpi_ec_intr_query(ec, data); -} -static int acpi_ec_poll_query(union acpi_ec *ec, u32 * data) +extern int ec_transaction(u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { - int result = 0; - acpi_status status = AE_OK; - u32 glk = 0; - - - if (!ec || !data) - return -EINVAL; - - *data = 0; - - if (ec->common.global_lock) { - status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); - if (ACPI_FAILURE(status)) - return -ENODEV; - } - - /* - * Query the EC to find out which _Qxx method we need to evaluate. - * Note that successful completion of the query causes the ACPI_EC_SCI - * bit to be cleared (and thus clearing the interrupt source). - */ - if (down_interruptible(&ec->poll.sem)) { - result = -ERESTARTSYS; - goto end_nosem; - } - - acpi_hw_low_level_write(8, ACPI_EC_COMMAND_QUERY, - &ec->common.command_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); - if (result) - goto end; + union acpi_ec *ec; - acpi_hw_low_level_read(8, data, &ec->common.data_addr); - if (!*data) - result = -ENODATA; + if (!first_ec) + return -ENODEV; - end: - up(&ec->poll.sem); -end_nosem: - if (ec->common.global_lock) - acpi_release_global_lock(glk); + ec = acpi_driver_data(first_ec); - return result; + return acpi_ec_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); } -static int acpi_ec_intr_query(union acpi_ec *ec, u32 * data) -{ - int status = 0; - u32 glk; +EXPORT_SYMBOL(ec_transaction); - if (!ec || !data) - return -EINVAL; - *data = 0; +static int acpi_ec_query(union acpi_ec *ec, u32 * data) { + int result; + u8 d; - if (ec->common.global_lock) { - status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); - if (ACPI_FAILURE(status)) - return -ENODEV; - } + if (!ec || !data) + return -EINVAL; - down(&ec->intr.sem); + /* + * Query the EC to find out which _Qxx method we need to evaluate. + * Note that successful completion of the query causes the ACPI_EC_SCI + * bit to be cleared (and thus clearing the interrupt source). + */ - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (status) { - printk(KERN_DEBUG PREFIX "query EC, IB not empty\n"); - goto end; - } - /* - * Query the EC to find out which _Qxx method we need to evaluate. - * Note that successful completion of the query causes the ACPI_EC_SCI - * bit to be cleared (and thus clearing the interrupt source). - */ - acpi_hw_low_level_write(8, ACPI_EC_COMMAND_QUERY, - &ec->common.command_addr); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); - if (status) { - printk(KERN_DEBUG PREFIX "query EC, OB not full\n"); - goto end; - } - - acpi_hw_low_level_read(8, data, &ec->common.data_addr); - if (!*data) - status = -ENODATA; + result = acpi_ec_transaction(ec, ACPI_EC_COMMAND_QUERY, NULL, 0, &d, 1); + if (result) + return result; - end: - up(&ec->intr.sem); + if (!d) + return -ENODATA; - if (ec->common.global_lock) - acpi_release_global_lock(glk); - - return status; + *data = d; + return 0; } /* -------------------------------------------------------------------------- diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 88b5dfd..2b0c955 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -494,6 +494,9 @@ void acpi_pci_unregister_driver(struct acpi_pci_driver *driver); extern int ec_read(u8 addr, u8 *val); extern int ec_write(u8 addr, u8 val); +extern int ec_transaction(u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len); #endif /*CONFIG_ACPI_EC*/ -- cgit v0.10.2 From 7c6db5e51227761f42c6ac8260753f5c24dc1dde Mon Sep 17 00:00:00 2001 From: "Denis M. Sadykov" Date: Tue, 26 Sep 2006 19:50:33 +0400 Subject: ACPI: EC: Remove unnecessary delay added by previous transation patch. Remove unnecessary delay (50 ms) while reading data from EC in interrupt mode. Signed-off-by: Alexey Y. Starikovskiy Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index a0dcbad..b6f935d 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -100,7 +100,7 @@ union acpi_ec { struct acpi_generic_address command_addr; struct acpi_generic_address data_addr; unsigned long global_lock; - unsigned int expect_event; + u8 expect_event; atomic_t leaving_burst; /* 0 : No, 1 : Yes, 2: abort */ atomic_t pending_gpe; struct semaphore sem; @@ -121,7 +121,7 @@ union acpi_ec { }; static int acpi_ec_poll_wait(union acpi_ec *ec, u8 event); -static int acpi_ec_intr_wait(union acpi_ec *ec, unsigned int event); +static int acpi_ec_intr_wait(union acpi_ec *ec, u8 event); static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len); @@ -161,6 +161,22 @@ static u32 acpi_ec_read_status(union acpi_ec *ec) return status; } +static int acpi_ec_check_status(u32 status, u8 event) { + + switch (event) { + case ACPI_EC_EVENT_OBF: + if (status & ACPI_EC_FLAG_OBF) + return 1; + case ACPI_EC_EVENT_IBE: + if (!(status & ACPI_EC_FLAG_IBF)) + return 1; + default: + break; + } + + return 0; +} + static int acpi_ec_wait(union acpi_ec *ec, u8 event) { if (acpi_ec_poll_mode) @@ -203,47 +219,28 @@ static int acpi_ec_poll_wait(union acpi_ec *ec, u8 event) return -ETIME; } -static int acpi_ec_intr_wait(union acpi_ec *ec, unsigned int event) -{ - int result = 0; +static int acpi_ec_intr_wait(union acpi_ec *ec, u8 event) +{ + long time_left; ec->intr.expect_event = event; - smp_mb(); - switch (event) { - case ACPI_EC_EVENT_IBE: - if (~acpi_ec_read_status(ec) & ACPI_EC_FLAG_IBF) { + if (acpi_ec_check_status(acpi_ec_read_status(ec), event)) { ec->intr.expect_event = 0; return 0; - } - break; - default: - break; } - result = wait_event_timeout(ec->intr.wait, - !ec->intr.expect_event, - msecs_to_jiffies(ACPI_EC_DELAY)); + time_left = wait_event_timeout(ec->intr.wait, !ec->intr.expect_event, + msecs_to_jiffies(ACPI_EC_DELAY)); ec->intr.expect_event = 0; - smp_mb(); - - /* - * Verify that the event in question has actually happened by - * querying EC status. Do the check even if operation timed-out - * to make sure that we did not miss interrupt. - */ - switch (event) { - case ACPI_EC_EVENT_OBF: - if (acpi_ec_read_status(ec) & ACPI_EC_FLAG_OBF) - return 0; - break; - - case ACPI_EC_EVENT_IBE: - if (~acpi_ec_read_status(ec) & ACPI_EC_FLAG_IBF) - return 0; - break; + if (time_left <= 0) { + if (acpi_ec_check_status(acpi_ec_read_status(ec), event)) { + return 0; + } + } else { + return 0; } return -ETIME; @@ -293,7 +290,7 @@ int acpi_ec_leave_burst_mode(union acpi_ec *ec) goto end; acpi_hw_low_level_write(8, ACPI_EC_BURST_DISABLE, &ec->common.command_addr); acpi_ec_wait(ec, ACPI_EC_FLAG_IBF); - } + } atomic_set(&ec->intr.leaving_burst, 1); return 0; end: @@ -333,32 +330,32 @@ static int acpi_ec_transaction_unlocked(union acpi_ec *ec, u8 command, acpi_hw_low_level_write(8, command, &ec->common.command_addr); - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (result) - return result; - - for (; wdata_len > 0; wdata_len --) { - - acpi_hw_low_level_write(8, *(wdata++), &ec->common.data_addr); + for (; wdata_len > 0; wdata_len --) { + result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + if (result) + return result; - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); - if (result) - return result; + acpi_hw_low_level_write(8, *(wdata++), &ec->common.data_addr); } + if (command == ACPI_EC_COMMAND_WRITE) { + result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + if (result) + return result; + } - for (; rdata_len > 0; rdata_len --) { - u32 d; + for (; rdata_len > 0; rdata_len --) { + u32 d; - result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); - if (result) - return result; + result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); + if (result) + return result; - acpi_hw_low_level_read(8, &d, &ec->common.data_addr); - *(rdata++) = (u8) d; - } + acpi_hw_low_level_read(8, &d, &ec->common.data_addr); + *(rdata++) = (u8) d; + } - return 0; + return 0; } static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, -- cgit v0.10.2 From 703959d47e887a29dc58123c05aa0ffcbbfa131d Mon Sep 17 00:00:00 2001 From: "Denis M. Sadykov" Date: Tue, 26 Sep 2006 19:50:33 +0400 Subject: ACPI: EC: Remove unused variables and duplicated code Signed-off-by: Alexey Y. Starikovskiy Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index b6f935d..c816b4e 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -45,203 +45,162 @@ ACPI_MODULE_NAME("acpi_ec") #define ACPI_EC_DRIVER_NAME "ACPI Embedded Controller Driver" #define ACPI_EC_DEVICE_NAME "Embedded Controller" #define ACPI_EC_FILE_INFO "info" + +/* EC status register */ #define ACPI_EC_FLAG_OBF 0x01 /* Output buffer full */ #define ACPI_EC_FLAG_IBF 0x02 /* Input buffer full */ #define ACPI_EC_FLAG_BURST 0x10 /* burst mode */ #define ACPI_EC_FLAG_SCI 0x20 /* EC-SCI occurred */ -#define ACPI_EC_EVENT_OBF 0x01 /* Output buffer full */ -#define ACPI_EC_EVENT_IBE 0x02 /* Input buffer empty */ -#define ACPI_EC_DELAY 50 /* Wait 50ms max. during EC ops */ -#define ACPI_EC_UDELAY_GLK 1000 /* Wait 1ms max. to get global lock */ -#define ACPI_EC_UDELAY 100 /* Poll @ 100us increments */ -#define ACPI_EC_UDELAY_COUNT 1000 /* Wait 10ms max. during EC ops */ + +/* EC commands */ #define ACPI_EC_COMMAND_READ 0x80 #define ACPI_EC_COMMAND_WRITE 0x81 #define ACPI_EC_BURST_ENABLE 0x82 #define ACPI_EC_BURST_DISABLE 0x83 #define ACPI_EC_COMMAND_QUERY 0x84 -#define EC_POLL 0xFF -#define EC_INTR 0x00 + +/* EC events */ +enum { + ACPI_EC_EVENT_OBF_1 = 1, /* Output buffer full */ + ACPI_EC_EVENT_IBF_0, /* Input buffer empty */ +}; + +#define ACPI_EC_DELAY 50 /* Wait 50ms max. during EC ops */ +#define ACPI_EC_UDELAY_GLK 1000 /* Wait 1ms max. to get global lock */ +#define ACPI_EC_UDELAY 100 /* Poll @ 100us increments */ +#define ACPI_EC_UDELAY_COUNT 1000 /* Wait 10ms max. during EC ops */ + +enum { + EC_INTR = 1, /* Output buffer full */ + EC_POLL, /* Input buffer empty */ +}; + static int acpi_ec_remove(struct acpi_device *device, int type); static int acpi_ec_start(struct acpi_device *device); static int acpi_ec_stop(struct acpi_device *device, int type); -static int acpi_ec_intr_add(struct acpi_device *device); -static int acpi_ec_poll_add(struct acpi_device *device); +static int acpi_ec_add(struct acpi_device *device); static struct acpi_driver acpi_ec_driver = { .name = ACPI_EC_DRIVER_NAME, .class = ACPI_EC_CLASS, .ids = ACPI_EC_HID, .ops = { - .add = acpi_ec_intr_add, + .add = acpi_ec_add, .remove = acpi_ec_remove, .start = acpi_ec_start, .stop = acpi_ec_stop, }, }; -union acpi_ec { - struct { - u32 mode; - acpi_handle handle; - unsigned long uid; - unsigned long gpe_bit; - struct acpi_generic_address status_addr; - struct acpi_generic_address command_addr; - struct acpi_generic_address data_addr; - unsigned long global_lock; - } common; - - struct { - u32 mode; - acpi_handle handle; - unsigned long uid; - unsigned long gpe_bit; - struct acpi_generic_address status_addr; - struct acpi_generic_address command_addr; - struct acpi_generic_address data_addr; - unsigned long global_lock; - u8 expect_event; - atomic_t leaving_burst; /* 0 : No, 1 : Yes, 2: abort */ - atomic_t pending_gpe; - struct semaphore sem; - wait_queue_head_t wait; - } intr; - - struct { - u32 mode; - acpi_handle handle; - unsigned long uid; - unsigned long gpe_bit; - struct acpi_generic_address status_addr; - struct acpi_generic_address command_addr; - struct acpi_generic_address data_addr; - unsigned long global_lock; - struct semaphore sem; - } poll; +struct acpi_ec { + acpi_handle handle; + unsigned long uid; + unsigned long gpe_bit; + struct acpi_generic_address status_addr; + struct acpi_generic_address command_addr; + struct acpi_generic_address data_addr; + unsigned long global_lock; + struct semaphore sem; + unsigned int expect_event; + atomic_t leaving_burst; /* 0 : No, 1 : Yes, 2: abort */ + wait_queue_head_t wait; }; -static int acpi_ec_poll_wait(union acpi_ec *ec, u8 event); -static int acpi_ec_intr_wait(union acpi_ec *ec, u8 event); -static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, +/* If we find an EC via the ECDT, we need to keep a ptr to its context */ +static struct acpi_ec *ec_ecdt; + +/* External interfaces use first EC only, so remember */ +static struct acpi_device *first_ec; +static int acpi_ec_mode = EC_INTR; + +static int acpi_ec_poll_transaction(struct acpi_ec *ec, u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len); -static int acpi_ec_intr_transaction(union acpi_ec *ec, u8 command, +static int acpi_ec_intr_transaction(struct acpi_ec *ec, u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len); static void acpi_ec_gpe_poll_query(void *ec_cxt); static void acpi_ec_gpe_intr_query(void *ec_cxt); static u32 acpi_ec_gpe_poll_handler(void *data); static u32 acpi_ec_gpe_intr_handler(void *data); -static acpi_status __init -acpi_fake_ecdt_poll_callback(acpi_handle handle, - u32 Level, void *context, void **retval); - -static acpi_status __init -acpi_fake_ecdt_intr_callback(acpi_handle handle, - u32 Level, void *context, void **retval); - -static int __init acpi_ec_poll_get_real_ecdt(void); -static int __init acpi_ec_intr_get_real_ecdt(void); -/* If we find an EC via the ECDT, we need to keep a ptr to its context */ -static union acpi_ec *ec_ecdt; - -/* External interfaces use first EC only, so remember */ -static struct acpi_device *first_ec; -static int acpi_ec_poll_mode = EC_INTR; /* -------------------------------------------------------------------------- Transaction Management -------------------------------------------------------------------------- */ -static u32 acpi_ec_read_status(union acpi_ec *ec) +static u32 acpi_ec_read_status(struct acpi_ec *ec) { u32 status = 0; - acpi_hw_low_level_read(8, &status, &ec->common.status_addr); + acpi_hw_low_level_read(8, &status, &ec->status_addr); return status; } -static int acpi_ec_check_status(u32 status, u8 event) { - - switch (event) { - case ACPI_EC_EVENT_OBF: - if (status & ACPI_EC_FLAG_OBF) - return 1; - case ACPI_EC_EVENT_IBE: - if (!(status & ACPI_EC_FLAG_IBF)) - return 1; - default: - break; - } +static u32 acpi_ec_read_data(struct acpi_ec *ec) +{ + u32 data = 0; - return 0; + acpi_hw_low_level_read(8, &data, &ec->data_addr); + return data; } -static int acpi_ec_wait(union acpi_ec *ec, u8 event) +static void acpi_ec_write_cmd(struct acpi_ec *ec, u32 command) { - if (acpi_ec_poll_mode) - return acpi_ec_poll_wait(ec, event); - else - return acpi_ec_intr_wait(ec, event); + acpi_hw_low_level_write(8, command, &ec->command_addr); } -static int acpi_ec_poll_wait(union acpi_ec *ec, u8 event) +static void acpi_ec_write_data(struct acpi_ec *ec, u32 data) { - u32 acpi_ec_status = 0; - u32 i = ACPI_EC_UDELAY_COUNT; + acpi_hw_low_level_write(8, data, &ec->data_addr); +} - if (!ec) - return -EINVAL; +static int acpi_ec_check_status(u32 status, u8 event) { - /* Poll the EC status register waiting for the event to occur. */ switch (event) { - case ACPI_EC_EVENT_OBF: - do { - acpi_hw_low_level_read(8, &acpi_ec_status, - &ec->common.status_addr); - if (acpi_ec_status & ACPI_EC_FLAG_OBF) - return 0; - udelay(ACPI_EC_UDELAY); - } while (--i > 0); + case ACPI_EC_EVENT_OBF_1: + if (status & ACPI_EC_FLAG_OBF) + return 1; break; - case ACPI_EC_EVENT_IBE: - do { - acpi_hw_low_level_read(8, &acpi_ec_status, - &ec->common.status_addr); - if (!(acpi_ec_status & ACPI_EC_FLAG_IBF)) - return 0; - udelay(ACPI_EC_UDELAY); - } while (--i > 0); + case ACPI_EC_EVENT_IBF_0: + if (!(status & ACPI_EC_FLAG_IBF)) + return 1; break; default: - return -EINVAL; + break; } - return -ETIME; + return 0; } -static int acpi_ec_intr_wait(union acpi_ec *ec, u8 event) +static int acpi_ec_wait(struct acpi_ec *ec, u8 event) { - long time_left; - - ec->intr.expect_event = event; + int i = (acpi_ec_mode == EC_POLL) ? ACPI_EC_UDELAY_COUNT : 0; + long time_left; + ec->expect_event = event; if (acpi_ec_check_status(acpi_ec_read_status(ec), event)) { - ec->intr.expect_event = 0; - return 0; + ec->expect_event = 0; + return 0; } - time_left = wait_event_timeout(ec->intr.wait, !ec->intr.expect_event, - msecs_to_jiffies(ACPI_EC_DELAY)); - - ec->intr.expect_event = 0; - if (time_left <= 0) { - if (acpi_ec_check_status(acpi_ec_read_status(ec), event)) { + do { + if (acpi_ec_mode == EC_POLL) { + udelay(ACPI_EC_UDELAY); + } else { + time_left = wait_event_timeout(ec->wait, + !ec->expect_event, + msecs_to_jiffies(ACPI_EC_DELAY)); + if (time_left > 0) { + ec->expect_event = 0; return 0; + } } - } else { - return 0; - } + if (acpi_ec_check_status(acpi_ec_read_status(ec), event)) { + ec->expect_event = 0; + return 0; + } + } while (--i > 0); + + ec->expect_event = 0; return -ETIME; } @@ -251,64 +210,63 @@ static int acpi_ec_intr_wait(union acpi_ec *ec, u8 event) * Note: samsung nv5000 doesn't work with ec burst mode. * http://bugzilla.kernel.org/show_bug.cgi?id=4980 */ -int acpi_ec_enter_burst_mode(union acpi_ec *ec) +int acpi_ec_enter_burst_mode(struct acpi_ec *ec) { u32 tmp = 0; - int status = 0; + u32 status = 0; status = acpi_ec_read_status(ec); if (status != -EINVAL && !(status & ACPI_EC_FLAG_BURST)) { - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if (status) goto end; - acpi_hw_low_level_write(8, ACPI_EC_BURST_ENABLE, - &ec->common.command_addr); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); - acpi_hw_low_level_read(8, &tmp, &ec->common.data_addr); + acpi_ec_write_cmd(ec, ACPI_EC_BURST_ENABLE); + status = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF_1); + tmp = acpi_ec_read_data(ec); if (tmp != 0x90) { /* Burst ACK byte */ return -EINVAL; } } - atomic_set(&ec->intr.leaving_burst, 0); + atomic_set(&ec->leaving_burst, 0); return 0; - end: - ACPI_EXCEPTION ((AE_INFO, status, "EC wait, burst mode"); + end: + ACPI_EXCEPTION((AE_INFO, status, "EC wait, burst mode")); return -1; } -int acpi_ec_leave_burst_mode(union acpi_ec *ec) +int acpi_ec_leave_burst_mode(struct acpi_ec *ec) { - int status = 0; + u32 status = 0; status = acpi_ec_read_status(ec); if (status != -EINVAL && (status & ACPI_EC_FLAG_BURST)){ - status = acpi_ec_wait(ec, ACPI_EC_FLAG_IBF); + status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if(status) goto end; - acpi_hw_low_level_write(8, ACPI_EC_BURST_DISABLE, &ec->common.command_addr); - acpi_ec_wait(ec, ACPI_EC_FLAG_IBF); + acpi_ec_write_cmd(ec, ACPI_EC_BURST_DISABLE); + acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); } - atomic_set(&ec->intr.leaving_burst, 1); + atomic_set(&ec->leaving_burst, 1); return 0; -end: - ACPI_EXCEPTION((AE_INFO, status, "EC leave burst mode"); + end: + ACPI_EXCEPTION((AE_INFO, status, "EC leave burst mode")); return -1; } #endif /* ACPI_FUTURE_USAGE */ -static int acpi_ec_transaction(union acpi_ec *ec, u8 command, +static int acpi_ec_transaction(struct acpi_ec *ec, u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len) { - if (acpi_ec_poll_mode) + if (acpi_ec_mode == EC_POLL) return acpi_ec_poll_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); else return acpi_ec_intr_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); } -static int acpi_ec_read(union acpi_ec *ec, u8 address, u32 * data) +static int acpi_ec_read(struct acpi_ec *ec, u8 address, u32 * data) { int result; u8 d; @@ -316,30 +274,30 @@ static int acpi_ec_read(union acpi_ec *ec, u8 address, u32 * data) *data = d; return result; } -static int acpi_ec_write(union acpi_ec *ec, u8 address, u8 data) +static int acpi_ec_write(struct acpi_ec *ec, u8 address, u8 data) { u8 wdata[2] = { address, data }; return acpi_ec_transaction(ec, ACPI_EC_COMMAND_WRITE, wdata, 2, NULL, 0); } -static int acpi_ec_transaction_unlocked(union acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len) +static int acpi_ec_transaction_unlocked(struct acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { int result; - acpi_hw_low_level_write(8, command, &ec->common.command_addr); + acpi_ec_write_cmd(ec, command); for (; wdata_len > 0; wdata_len --) { - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if (result) return result; - acpi_hw_low_level_write(8, *(wdata++), &ec->common.data_addr); + acpi_ec_write_data(ec, *(wdata++)); } if (command == ACPI_EC_COMMAND_WRITE) { - result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if (result) return result; } @@ -347,18 +305,18 @@ static int acpi_ec_transaction_unlocked(union acpi_ec *ec, u8 command, for (; rdata_len > 0; rdata_len --) { u32 d; - result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF); + result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF_1); if (result) return result; - acpi_hw_low_level_read(8, &d, &ec->common.data_addr); + d = acpi_ec_read_data(ec); *(rdata++) = (u8) d; } return 0; } -static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, +static int acpi_ec_poll_transaction(struct acpi_ec *ec, u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len) { @@ -372,13 +330,13 @@ static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, if (rdata) memset(rdata, 0, rdata_len); - if (ec->common.global_lock) { + if (ec->global_lock) { status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); if (ACPI_FAILURE(status)) return -ENODEV; } - if (down_interruptible(&ec->poll.sem)) { + if (down_interruptible(&ec->sem)) { result = -ERESTARTSYS; goto end_nosem; } @@ -386,16 +344,16 @@ static int acpi_ec_poll_transaction(union acpi_ec *ec, u8 command, result = acpi_ec_transaction_unlocked(ec, command, wdata, wdata_len, rdata, rdata_len); - up(&ec->poll.sem); + up(&ec->sem); end_nosem: - if (ec->common.global_lock) + if (ec->global_lock) acpi_release_global_lock(glk); return result; } -static int acpi_ec_intr_transaction(union acpi_ec *ec, u8 command, +static int acpi_ec_intr_transaction(struct acpi_ec *ec, u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len) { @@ -408,18 +366,18 @@ static int acpi_ec_intr_transaction(union acpi_ec *ec, u8 command, if (rdata) memset(rdata, 0, rdata_len); - if (ec->common.global_lock) { + if (ec->global_lock) { status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); if (ACPI_FAILURE(status)) return -ENODEV; } WARN_ON(in_interrupt()); - down(&ec->intr.sem); + down(&ec->sem); - status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBE); + status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if (status) { - printk(KERN_DEBUG PREFIX "read EC, IB not empty\n"); + ACPI_EXCEPTION((AE_INFO, status, "read EC, IB not empty")); goto end; } @@ -428,9 +386,9 @@ static int acpi_ec_intr_transaction(union acpi_ec *ec, u8 command, rdata, rdata_len); end: - up(&ec->intr.sem); + up(&ec->sem); - if (ec->common.global_lock) + if (ec->global_lock) acpi_release_global_lock(glk); return status; @@ -441,7 +399,7 @@ end: */ int ec_read(u8 addr, u8 * val) { - union acpi_ec *ec; + struct acpi_ec *ec; int err; u32 temp_data; @@ -463,7 +421,7 @@ EXPORT_SYMBOL(ec_read); int ec_write(u8 addr, u8 val) { - union acpi_ec *ec; + struct acpi_ec *ec; int err; if (!first_ec) @@ -482,7 +440,7 @@ extern int ec_transaction(u8 command, const u8 *wdata, unsigned wdata_len, u8 *rdata, unsigned rdata_len) { - union acpi_ec *ec; + struct acpi_ec *ec; if (!first_ec) return -ENODEV; @@ -494,7 +452,7 @@ extern int ec_transaction(u8 command, EXPORT_SYMBOL(ec_transaction); -static int acpi_ec_query(union acpi_ec *ec, u32 * data) { +static int acpi_ec_query(struct acpi_ec *ec, u32 * data) { int result; u8 d; @@ -529,7 +487,7 @@ union acpi_ec_query_data { static void acpi_ec_gpe_query(void *ec_cxt) { - if (acpi_ec_poll_mode) + if (acpi_ec_mode == EC_POLL) acpi_ec_gpe_poll_query(ec_cxt); else acpi_ec_gpe_intr_query(ec_cxt); @@ -537,7 +495,7 @@ static void acpi_ec_gpe_query(void *ec_cxt) static void acpi_ec_gpe_poll_query(void *ec_cxt) { - union acpi_ec *ec = (union acpi_ec *)ec_cxt; + struct acpi_ec *ec = (struct acpi_ec *)ec_cxt; u32 value = 0; static char object_name[5] = { '_', 'Q', '0', '0', '\0' }; const char hex[] = { '0', '1', '2', '3', '4', '5', '6', '7', @@ -548,11 +506,11 @@ static void acpi_ec_gpe_poll_query(void *ec_cxt) if (!ec_cxt) goto end; - if (down_interruptible (&ec->poll.sem)) { + if (down_interruptible (&ec->sem)) { return; } - acpi_hw_low_level_read(8, &value, &ec->common.command_addr); - up(&ec->poll.sem); + value = acpi_ec_read_status(ec); + up(&ec->sem); /* TBD: Implement asynch events! * NOTE: All we care about are EC-SCI's. Other EC events are @@ -571,14 +529,14 @@ static void acpi_ec_gpe_poll_query(void *ec_cxt) ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Evaluating %s\n", object_name)); - acpi_evaluate_object(ec->common.handle, object_name, NULL, NULL); + acpi_evaluate_object(ec->handle, object_name, NULL, NULL); end: - acpi_enable_gpe(NULL, ec->common.gpe_bit, ACPI_NOT_ISR); + acpi_enable_gpe(NULL, ec->gpe_bit, ACPI_NOT_ISR); } static void acpi_ec_gpe_intr_query(void *ec_cxt) { - union acpi_ec *ec = (union acpi_ec *)ec_cxt; + struct acpi_ec *ec = (struct acpi_ec *)ec_cxt; u32 value; int result = -ENODATA; static char object_name[5] = { '_', 'Q', '0', '0', '\0' }; @@ -598,15 +556,14 @@ static void acpi_ec_gpe_intr_query(void *ec_cxt) ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Evaluating %s\n", object_name)); - acpi_evaluate_object(ec->common.handle, object_name, NULL, NULL); + acpi_evaluate_object(ec->handle, object_name, NULL, NULL); end: - atomic_dec(&ec->intr.pending_gpe); return; } static u32 acpi_ec_gpe_handler(void *data) { - if (acpi_ec_poll_mode) + if (acpi_ec_mode == EC_POLL) return acpi_ec_gpe_poll_handler(data); else return acpi_ec_gpe_intr_handler(data); @@ -614,12 +571,12 @@ static u32 acpi_ec_gpe_handler(void *data) static u32 acpi_ec_gpe_poll_handler(void *data) { acpi_status status = AE_OK; - union acpi_ec *ec = (union acpi_ec *)data; + struct acpi_ec *ec = (struct acpi_ec *)data; if (!ec) return ACPI_INTERRUPT_NOT_HANDLED; - acpi_disable_gpe(NULL, ec->common.gpe_bit, ACPI_ISR); + acpi_disable_gpe(NULL, ec->gpe_bit, ACPI_ISR); status = acpi_os_execute(OSL_EC_POLL_HANDLER, acpi_ec_gpe_query, ec); @@ -632,39 +589,38 @@ static u32 acpi_ec_gpe_intr_handler(void *data) { acpi_status status = AE_OK; u32 value; - union acpi_ec *ec = (union acpi_ec *)data; + struct acpi_ec *ec = (struct acpi_ec *)data; if (!ec) return ACPI_INTERRUPT_NOT_HANDLED; - acpi_clear_gpe(NULL, ec->common.gpe_bit, ACPI_ISR); + acpi_clear_gpe(NULL, ec->gpe_bit, ACPI_ISR); value = acpi_ec_read_status(ec); - switch (ec->intr.expect_event) { - case ACPI_EC_EVENT_OBF: + switch (ec->expect_event) { + case ACPI_EC_EVENT_OBF_1: if (!(value & ACPI_EC_FLAG_OBF)) break; - ec->intr.expect_event = 0; - wake_up(&ec->intr.wait); + ec->expect_event = 0; + wake_up(&ec->wait); break; - case ACPI_EC_EVENT_IBE: + case ACPI_EC_EVENT_IBF_0: if ((value & ACPI_EC_FLAG_IBF)) break; - ec->intr.expect_event = 0; - wake_up(&ec->intr.wait); + ec->expect_event = 0; + wake_up(&ec->wait); break; default: break; } if (value & ACPI_EC_FLAG_SCI) { - atomic_add(1, &ec->intr.pending_gpe); status = acpi_os_execute(OSL_EC_BURST_HANDLER, acpi_ec_gpe_query, ec); return status == AE_OK ? ACPI_INTERRUPT_HANDLED : ACPI_INTERRUPT_NOT_HANDLED; } - acpi_enable_gpe(NULL, ec->common.gpe_bit, ACPI_ISR); + acpi_enable_gpe(NULL, ec->gpe_bit, ACPI_ISR); return status == AE_OK ? ACPI_INTERRUPT_HANDLED : ACPI_INTERRUPT_NOT_HANDLED; } @@ -695,7 +651,7 @@ acpi_ec_space_handler(u32 function, void *handler_context, void *region_context) { int result = 0; - union acpi_ec *ec = NULL; + struct acpi_ec *ec = NULL; u64 temp = *value; acpi_integer f_v = 0; int i = 0; @@ -705,12 +661,10 @@ acpi_ec_space_handler(u32 function, return AE_BAD_PARAMETER; if (bit_width != 8 && acpi_strict) { - printk(KERN_WARNING PREFIX - "acpi_ec_space_handler: bit_width should be 8\n"); return AE_BAD_PARAMETER; } - ec = (union acpi_ec *)handler_context; + ec = (struct acpi_ec *)handler_context; next_byte: switch (function) { @@ -767,20 +721,20 @@ static struct proc_dir_entry *acpi_ec_dir; static int acpi_ec_read_info(struct seq_file *seq, void *offset) { - union acpi_ec *ec = (union acpi_ec *)seq->private; + struct acpi_ec *ec = (struct acpi_ec *)seq->private; if (!ec) goto end; seq_printf(seq, "gpe bit: 0x%02x\n", - (u32) ec->common.gpe_bit); + (u32) ec->gpe_bit); seq_printf(seq, "ports: 0x%02x, 0x%02x\n", - (u32) ec->common.status_addr.address, - (u32) ec->common.data_addr.address); + (u32) ec->status_addr.address, + (u32) ec->data_addr.address); seq_printf(seq, "use global lock: %s\n", - ec->common.global_lock ? "yes" : "no"); - acpi_enable_gpe(NULL, ec->common.gpe_bit, ACPI_NOT_ISR); + ec->global_lock ? "yes" : "no"); + acpi_enable_gpe(NULL, ec->gpe_bit, ACPI_NOT_ISR); end: return 0; @@ -791,7 +745,7 @@ static int acpi_ec_info_open_fs(struct inode *inode, struct file *file) return single_open(file, acpi_ec_read_info, PDE(inode)->data); } -static const struct file_operations acpi_ec_info_ops = { +static struct file_operations acpi_ec_info_ops = { .open = acpi_ec_info_open_fs, .read = seq_read, .llseek = seq_lseek, @@ -840,101 +794,35 @@ static int acpi_ec_remove_fs(struct acpi_device *device) Driver Interface -------------------------------------------------------------------------- */ -static int acpi_ec_poll_add(struct acpi_device *device) +static int acpi_ec_add(struct acpi_device *device) { int result = 0; acpi_status status = AE_OK; - union acpi_ec *ec = NULL; + struct acpi_ec *ec = NULL; if (!device) return -EINVAL; - ec = kmalloc(sizeof(union acpi_ec), GFP_KERNEL); + ec = kmalloc(sizeof(struct acpi_ec), GFP_KERNEL); if (!ec) return -ENOMEM; - memset(ec, 0, sizeof(union acpi_ec)); - - ec->common.handle = device->handle; - ec->common.uid = -1; - init_MUTEX(&ec->poll.sem); - strcpy(acpi_device_name(device), ACPI_EC_DEVICE_NAME); - strcpy(acpi_device_class(device), ACPI_EC_CLASS); - acpi_driver_data(device) = ec; - - /* Use the global lock for all EC transactions? */ - acpi_evaluate_integer(ec->common.handle, "_GLK", NULL, - &ec->common.global_lock); - - /* XXX we don't test uids, because on some boxes ecdt uid = 0, see: - http://bugzilla.kernel.org/show_bug.cgi?id=6111 */ - if (ec_ecdt) { - acpi_remove_address_space_handler(ACPI_ROOT_OBJECT, - ACPI_ADR_SPACE_EC, - &acpi_ec_space_handler); - - acpi_remove_gpe_handler(NULL, ec_ecdt->common.gpe_bit, - &acpi_ec_gpe_handler); - - kfree(ec_ecdt); - } - - /* Get GPE bit assignment (EC events). */ - /* TODO: Add support for _GPE returning a package */ - status = - acpi_evaluate_integer(ec->common.handle, "_GPE", NULL, - &ec->common.gpe_bit); - if (ACPI_FAILURE(status)) { - ACPI_EXCEPTION((AE_INFO, status, "Obtaining GPE bit")); - result = -ENODEV; - goto end; + memset(ec, 0, sizeof(struct acpi_ec)); + + ec->handle = device->handle; + ec->uid = -1; + init_MUTEX(&ec->sem); + if (acpi_ec_mode == EC_INTR) { + atomic_set(&ec->leaving_burst, 1); + init_waitqueue_head(&ec->wait); } - - result = acpi_ec_add_fs(device); - if (result) - goto end; - - printk(KERN_INFO PREFIX "%s [%s] (gpe %d) polling mode.\n", - acpi_device_name(device), acpi_device_bid(device), - (u32) ec->common.gpe_bit); - - if (!first_ec) - first_ec = device; - - end: - if (result) - kfree(ec); - - return result; -} -static int acpi_ec_intr_add(struct acpi_device *device) -{ - int result = 0; - acpi_status status = AE_OK; - union acpi_ec *ec = NULL; - - - if (!device) - return -EINVAL; - - ec = kmalloc(sizeof(union acpi_ec), GFP_KERNEL); - if (!ec) - return -ENOMEM; - memset(ec, 0, sizeof(union acpi_ec)); - - ec->common.handle = device->handle; - ec->common.uid = -1; - atomic_set(&ec->intr.pending_gpe, 0); - atomic_set(&ec->intr.leaving_burst, 1); - init_MUTEX(&ec->intr.sem); - init_waitqueue_head(&ec->intr.wait); strcpy(acpi_device_name(device), ACPI_EC_DEVICE_NAME); strcpy(acpi_device_class(device), ACPI_EC_CLASS); acpi_driver_data(device) = ec; /* Use the global lock for all EC transactions? */ - acpi_evaluate_integer(ec->common.handle, "_GLK", NULL, - &ec->common.global_lock); + acpi_evaluate_integer(ec->handle, "_GLK", NULL, + &ec->global_lock); /* XXX we don't test uids, because on some boxes ecdt uid = 0, see: http://bugzilla.kernel.org/show_bug.cgi?id=6111 */ @@ -943,7 +831,7 @@ static int acpi_ec_intr_add(struct acpi_device *device) ACPI_ADR_SPACE_EC, &acpi_ec_space_handler); - acpi_remove_gpe_handler(NULL, ec_ecdt->common.gpe_bit, + acpi_remove_gpe_handler(NULL, ec_ecdt->gpe_bit, &acpi_ec_gpe_handler); kfree(ec_ecdt); @@ -952,10 +840,10 @@ static int acpi_ec_intr_add(struct acpi_device *device) /* Get GPE bit assignment (EC events). */ /* TODO: Add support for _GPE returning a package */ status = - acpi_evaluate_integer(ec->common.handle, "_GPE", NULL, - &ec->common.gpe_bit); + acpi_evaluate_integer(ec->handle, "_GPE", NULL, + &ec->gpe_bit); if (ACPI_FAILURE(status)) { - printk(KERN_ERR PREFIX "Obtaining GPE bit assignment\n"); + ACPI_EXCEPTION((AE_INFO, status, "Obtaining GPE bit assignment")); result = -ENODEV; goto end; } @@ -964,14 +852,14 @@ static int acpi_ec_intr_add(struct acpi_device *device) if (result) goto end; - printk(KERN_INFO PREFIX "%s [%s] (gpe %d) interrupt mode.\n", + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "%s [%s] (gpe %d) interrupt mode.", acpi_device_name(device), acpi_device_bid(device), - (u32) ec->common.gpe_bit); + (u32) ec->gpe_bit)); if (!first_ec) first_ec = device; - end: + end: if (result) kfree(ec); @@ -980,7 +868,7 @@ static int acpi_ec_intr_add(struct acpi_device *device) static int acpi_ec_remove(struct acpi_device *device, int type) { - union acpi_ec *ec = NULL; + struct acpi_ec *ec = NULL; if (!device) @@ -998,7 +886,7 @@ static int acpi_ec_remove(struct acpi_device *device, int type) static acpi_status acpi_ec_io_ports(struct acpi_resource *resource, void *context) { - union acpi_ec *ec = (union acpi_ec *)context; + struct acpi_ec *ec = (struct acpi_ec *)context; struct acpi_generic_address *addr; if (resource->type != ACPI_RESOURCE_TYPE_IO) { @@ -1010,10 +898,10 @@ acpi_ec_io_ports(struct acpi_resource *resource, void *context) * the second address region returned is the status/command * port. */ - if (ec->common.data_addr.register_bit_width == 0) { - addr = &ec->common.data_addr; - } else if (ec->common.command_addr.register_bit_width == 0) { - addr = &ec->common.command_addr; + if (ec->data_addr.register_bit_width == 0) { + addr = &ec->data_addr; + } else if (ec->command_addr.register_bit_width == 0) { + addr = &ec->command_addr; } else { return AE_CTRL_TERMINATE; } @@ -1029,7 +917,7 @@ acpi_ec_io_ports(struct acpi_resource *resource, void *context) static int acpi_ec_start(struct acpi_device *device) { acpi_status status = AE_OK; - union acpi_ec *ec = NULL; + struct acpi_ec *ec = NULL; if (!device) @@ -1043,39 +931,40 @@ static int acpi_ec_start(struct acpi_device *device) /* * Get I/O port addresses. Convert to GAS format. */ - status = acpi_walk_resources(ec->common.handle, METHOD_NAME__CRS, + status = acpi_walk_resources(ec->handle, METHOD_NAME__CRS, acpi_ec_io_ports, ec); if (ACPI_FAILURE(status) - || ec->common.command_addr.register_bit_width == 0) { - printk(KERN_ERR PREFIX "Error getting I/O port addresses\n"); + || ec->command_addr.register_bit_width == 0) { + ACPI_EXCEPTION((AE_INFO, status, + "Error getting I/O port addresses")); return -ENODEV; } - ec->common.status_addr = ec->common.command_addr; + ec->status_addr = ec->command_addr; - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "gpe=0x%02x, ports=0x%2x,0x%2x\n", - (u32) ec->common.gpe_bit, - (u32) ec->common.command_addr.address, - (u32) ec->common.data_addr.address)); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "gpe=0x%02x, ports=0x%2x,0x%2x", + (u32) ec->gpe_bit, + (u32) ec->command_addr.address, + (u32) ec->data_addr.address)); /* * Install GPE handler */ - status = acpi_install_gpe_handler(NULL, ec->common.gpe_bit, + status = acpi_install_gpe_handler(NULL, ec->gpe_bit, ACPI_GPE_EDGE_TRIGGERED, &acpi_ec_gpe_handler, ec); if (ACPI_FAILURE(status)) { return -ENODEV; } - acpi_set_gpe_type(NULL, ec->common.gpe_bit, ACPI_GPE_TYPE_RUNTIME); - acpi_enable_gpe(NULL, ec->common.gpe_bit, ACPI_NOT_ISR); + acpi_set_gpe_type(NULL, ec->gpe_bit, ACPI_GPE_TYPE_RUNTIME); + acpi_enable_gpe(NULL, ec->gpe_bit, ACPI_NOT_ISR); - status = acpi_install_address_space_handler(ec->common.handle, + status = acpi_install_address_space_handler(ec->handle, ACPI_ADR_SPACE_EC, &acpi_ec_space_handler, &acpi_ec_space_setup, ec); if (ACPI_FAILURE(status)) { - acpi_remove_gpe_handler(NULL, ec->common.gpe_bit, + acpi_remove_gpe_handler(NULL, ec->gpe_bit, &acpi_ec_gpe_handler); return -ENODEV; } @@ -1086,7 +975,7 @@ static int acpi_ec_start(struct acpi_device *device) static int acpi_ec_stop(struct acpi_device *device, int type) { acpi_status status = AE_OK; - union acpi_ec *ec = NULL; + struct acpi_ec *ec = NULL; if (!device) @@ -1094,14 +983,14 @@ static int acpi_ec_stop(struct acpi_device *device, int type) ec = acpi_driver_data(device); - status = acpi_remove_address_space_handler(ec->common.handle, + status = acpi_remove_address_space_handler(ec->handle, ACPI_ADR_SPACE_EC, &acpi_ec_space_handler); if (ACPI_FAILURE(status)) return -ENODEV; status = - acpi_remove_gpe_handler(NULL, ec->common.gpe_bit, + acpi_remove_gpe_handler(NULL, ec->gpe_bit, &acpi_ec_gpe_handler); if (ACPI_FAILURE(status)) return -ENODEV; @@ -1113,76 +1002,33 @@ static acpi_status __init acpi_fake_ecdt_callback(acpi_handle handle, u32 Level, void *context, void **retval) { - - if (acpi_ec_poll_mode) - return acpi_fake_ecdt_poll_callback(handle, - Level, context, retval); - else - return acpi_fake_ecdt_intr_callback(handle, - Level, context, retval); -} - -static acpi_status __init -acpi_fake_ecdt_poll_callback(acpi_handle handle, - u32 Level, void *context, void **retval) -{ - acpi_status status; - - status = acpi_walk_resources(handle, METHOD_NAME__CRS, - acpi_ec_io_ports, ec_ecdt); - if (ACPI_FAILURE(status)) - return status; - ec_ecdt->common.status_addr = ec_ecdt->common.command_addr; - - ec_ecdt->common.uid = -1; - acpi_evaluate_integer(handle, "_UID", NULL, &ec_ecdt->common.uid); - - status = - acpi_evaluate_integer(handle, "_GPE", NULL, - &ec_ecdt->common.gpe_bit); - if (ACPI_FAILURE(status)) - return status; - init_MUTEX(&ec_ecdt->poll.sem); - ec_ecdt->common.global_lock = TRUE; - ec_ecdt->common.handle = handle; - - printk(KERN_INFO PREFIX "GPE=0x%02x, ports=0x%2x, 0x%2x\n", - (u32) ec_ecdt->common.gpe_bit, - (u32) ec_ecdt->common.command_addr.address, - (u32) ec_ecdt->common.data_addr.address); - - return AE_CTRL_TERMINATE; -} - -static acpi_status __init -acpi_fake_ecdt_intr_callback(acpi_handle handle, - u32 Level, void *context, void **retval) -{ acpi_status status; - init_MUTEX(&ec_ecdt->intr.sem); - init_waitqueue_head(&ec_ecdt->intr.wait); + init_MUTEX(&ec_ecdt->sem); + if (acpi_ec_mode == EC_INTR) { + init_waitqueue_head(&ec_ecdt->wait); + } status = acpi_walk_resources(handle, METHOD_NAME__CRS, acpi_ec_io_ports, ec_ecdt); if (ACPI_FAILURE(status)) return status; - ec_ecdt->common.status_addr = ec_ecdt->common.command_addr; + ec_ecdt->status_addr = ec_ecdt->command_addr; - ec_ecdt->common.uid = -1; - acpi_evaluate_integer(handle, "_UID", NULL, &ec_ecdt->common.uid); + ec_ecdt->uid = -1; + acpi_evaluate_integer(handle, "_UID", NULL, &ec_ecdt->uid); status = acpi_evaluate_integer(handle, "_GPE", NULL, - &ec_ecdt->common.gpe_bit); + &ec_ecdt->gpe_bit); if (ACPI_FAILURE(status)) return status; - ec_ecdt->common.global_lock = TRUE; - ec_ecdt->common.handle = handle; + ec_ecdt->global_lock = TRUE; + ec_ecdt->handle = handle; - printk(KERN_INFO PREFIX "GPE=0x%02x, ports=0x%2x, 0x%2x\n", - (u32) ec_ecdt->common.gpe_bit, - (u32) ec_ecdt->common.command_addr.address, - (u32) ec_ecdt->common.data_addr.address); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "GPE=0x%02x, ports=0x%2x, 0x%2x", + (u32) ec_ecdt->gpe_bit, + (u32) ec_ecdt->command_addr.address, + (u32) ec_ecdt->data_addr.address)); return AE_CTRL_TERMINATE; } @@ -1202,14 +1048,14 @@ static int __init acpi_ec_fake_ecdt(void) acpi_status status; int ret = 0; - printk(KERN_INFO PREFIX "Try to make an fake ECDT\n"); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Try to make an fake ECDT")); - ec_ecdt = kmalloc(sizeof(union acpi_ec), GFP_KERNEL); + ec_ecdt = kmalloc(sizeof(struct acpi_ec), GFP_KERNEL); if (!ec_ecdt) { ret = -ENOMEM; goto error; } - memset(ec_ecdt, 0, sizeof(union acpi_ec)); + memset(ec_ecdt, 0, sizeof(struct acpi_ec)); status = acpi_get_devices(ACPI_EC_HID, acpi_fake_ecdt_callback, NULL, NULL); @@ -1217,24 +1063,16 @@ static int __init acpi_ec_fake_ecdt(void) kfree(ec_ecdt); ec_ecdt = NULL; ret = -ENODEV; + ACPI_EXCEPTION((AE_INFO, status, "Can't make an fake ECDT")); goto error; } return 0; - error: - printk(KERN_ERR PREFIX "Can't make an fake ECDT\n"); + error: return ret; } static int __init acpi_ec_get_real_ecdt(void) { - if (acpi_ec_poll_mode) - return acpi_ec_poll_get_real_ecdt(); - else - return acpi_ec_intr_get_real_ecdt(); -} - -static int __init acpi_ec_poll_get_real_ecdt(void) -{ acpi_status status; struct acpi_table_ecdt *ecdt_ptr; @@ -1244,80 +1082,37 @@ static int __init acpi_ec_poll_get_real_ecdt(void) if (ACPI_FAILURE(status)) return -ENODEV; - printk(KERN_INFO PREFIX "Found ECDT\n"); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Found ECDT")); /* * Generate a temporary ec context to use until the namespace is scanned */ - ec_ecdt = kmalloc(sizeof(union acpi_ec), GFP_KERNEL); + ec_ecdt = kmalloc(sizeof(struct acpi_ec), GFP_KERNEL); if (!ec_ecdt) return -ENOMEM; - memset(ec_ecdt, 0, sizeof(union acpi_ec)); - - ec_ecdt->common.command_addr = ecdt_ptr->ec_control; - ec_ecdt->common.status_addr = ecdt_ptr->ec_control; - ec_ecdt->common.data_addr = ecdt_ptr->ec_data; - ec_ecdt->common.gpe_bit = ecdt_ptr->gpe_bit; - init_MUTEX(&ec_ecdt->poll.sem); - /* use the GL just to be safe */ - ec_ecdt->common.global_lock = TRUE; - ec_ecdt->common.uid = ecdt_ptr->uid; + memset(ec_ecdt, 0, sizeof(struct acpi_ec)); - status = - acpi_get_handle(NULL, ecdt_ptr->ec_id, &ec_ecdt->common.handle); - if (ACPI_FAILURE(status)) { - goto error; + init_MUTEX(&ec_ecdt->sem); + if (acpi_ec_mode == EC_INTR) { + init_waitqueue_head(&ec_ecdt->wait); } - - return 0; - error: - printk(KERN_ERR PREFIX "Could not use ECDT\n"); - kfree(ec_ecdt); - ec_ecdt = NULL; - - return -ENODEV; -} - -static int __init acpi_ec_intr_get_real_ecdt(void) -{ - acpi_status status; - struct acpi_table_ecdt *ecdt_ptr; - - status = acpi_get_firmware_table("ECDT", 1, ACPI_LOGICAL_ADDRESSING, - (struct acpi_table_header **) - &ecdt_ptr); - if (ACPI_FAILURE(status)) - return -ENODEV; - - printk(KERN_INFO PREFIX "Found ECDT\n"); - - /* - * Generate a temporary ec context to use until the namespace is scanned - */ - ec_ecdt = kmalloc(sizeof(union acpi_ec), GFP_KERNEL); - if (!ec_ecdt) - return -ENOMEM; - memset(ec_ecdt, 0, sizeof(union acpi_ec)); - - init_MUTEX(&ec_ecdt->intr.sem); - init_waitqueue_head(&ec_ecdt->intr.wait); - ec_ecdt->common.command_addr = ecdt_ptr->ec_control; - ec_ecdt->common.status_addr = ecdt_ptr->ec_control; - ec_ecdt->common.data_addr = ecdt_ptr->ec_data; - ec_ecdt->common.gpe_bit = ecdt_ptr->gpe_bit; + ec_ecdt->command_addr = ecdt_ptr->ec_control; + ec_ecdt->status_addr = ecdt_ptr->ec_control; + ec_ecdt->data_addr = ecdt_ptr->ec_data; + ec_ecdt->gpe_bit = ecdt_ptr->gpe_bit; /* use the GL just to be safe */ - ec_ecdt->common.global_lock = TRUE; - ec_ecdt->common.uid = ecdt_ptr->uid; + ec_ecdt->global_lock = TRUE; + ec_ecdt->uid = ecdt_ptr->uid; status = - acpi_get_handle(NULL, ecdt_ptr->ec_id, &ec_ecdt->common.handle); + acpi_get_handle(NULL, ecdt_ptr->ec_id, &ec_ecdt->handle); if (ACPI_FAILURE(status)) { goto error; } return 0; - error: - printk(KERN_ERR PREFIX "Could not use ECDT\n"); + error: + ACPI_EXCEPTION((AE_INFO, status, "Could not use ECDT")); kfree(ec_ecdt); ec_ecdt = NULL; @@ -1342,14 +1137,14 @@ int __init acpi_ec_ecdt_probe(void) /* * Install GPE handler */ - status = acpi_install_gpe_handler(NULL, ec_ecdt->common.gpe_bit, + status = acpi_install_gpe_handler(NULL, ec_ecdt->gpe_bit, ACPI_GPE_EDGE_TRIGGERED, &acpi_ec_gpe_handler, ec_ecdt); if (ACPI_FAILURE(status)) { goto error; } - acpi_set_gpe_type(NULL, ec_ecdt->common.gpe_bit, ACPI_GPE_TYPE_RUNTIME); - acpi_enable_gpe(NULL, ec_ecdt->common.gpe_bit, ACPI_NOT_ISR); + acpi_set_gpe_type(NULL, ec_ecdt->gpe_bit, ACPI_GPE_TYPE_RUNTIME); + acpi_enable_gpe(NULL, ec_ecdt->gpe_bit, ACPI_NOT_ISR); status = acpi_install_address_space_handler(ACPI_ROOT_OBJECT, ACPI_ADR_SPACE_EC, @@ -1357,7 +1152,7 @@ int __init acpi_ec_ecdt_probe(void) &acpi_ec_space_setup, ec_ecdt); if (ACPI_FAILURE(status)) { - acpi_remove_gpe_handler(NULL, ec_ecdt->common.gpe_bit, + acpi_remove_gpe_handler(NULL, ec_ecdt->gpe_bit, &acpi_ec_gpe_handler); goto error; } @@ -1365,7 +1160,7 @@ int __init acpi_ec_ecdt_probe(void) return 0; error: - printk(KERN_ERR PREFIX "Could not use ECDT\n"); + ACPI_EXCEPTION((AE_INFO, status, "Could not use ECDT")); kfree(ec_ecdt); ec_ecdt = NULL; @@ -1424,13 +1219,13 @@ static int __init acpi_ec_set_intr_mode(char *str) return 0; if (intr) { - acpi_ec_poll_mode = EC_INTR; - acpi_ec_driver.ops.add = acpi_ec_intr_add; + acpi_ec_mode = EC_INTR; } else { - acpi_ec_poll_mode = EC_POLL; - acpi_ec_driver.ops.add = acpi_ec_poll_add; + acpi_ec_mode = EC_POLL; } - printk(KERN_INFO PREFIX "EC %s mode.\n", intr ? "interrupt" : "polling"); + acpi_ec_driver.ops.add = acpi_ec_add; + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "EC %s mode.\n", intr ? "interrupt" : "polling")); + return 1; } -- cgit v0.10.2 From 3576cf619b73d850f5b21375609645f221e6270f Mon Sep 17 00:00:00 2001 From: "Denis M. Sadykov" Date: Tue, 26 Sep 2006 19:50:33 +0400 Subject: ACPI: EC: Unify poll and interrupt mode transaction functions Signed-off-by: Alexey Y. Starikovskiy Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index c816b4e..9c7fce6 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -112,12 +112,6 @@ static struct acpi_ec *ec_ecdt; static struct acpi_device *first_ec; static int acpi_ec_mode = EC_INTR; -static int acpi_ec_poll_transaction(struct acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len); -static int acpi_ec_intr_transaction(struct acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len); static void acpi_ec_gpe_poll_query(void *ec_cxt); static void acpi_ec_gpe_intr_query(void *ec_cxt); static u32 acpi_ec_gpe_poll_handler(void *data); @@ -257,32 +251,9 @@ int acpi_ec_leave_burst_mode(struct acpi_ec *ec) } #endif /* ACPI_FUTURE_USAGE */ -static int acpi_ec_transaction(struct acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len) -{ - if (acpi_ec_mode == EC_POLL) - return acpi_ec_poll_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); - else - return acpi_ec_intr_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); -} -static int acpi_ec_read(struct acpi_ec *ec, u8 address, u32 * data) -{ - int result; - u8 d; - result = acpi_ec_transaction(ec, ACPI_EC_COMMAND_READ, &address, 1, &d, 1); - *data = d; - return result; -} -static int acpi_ec_write(struct acpi_ec *ec, u8 address, u8 data) -{ - u8 wdata[2] = { address, data }; - return acpi_ec_transaction(ec, ACPI_EC_COMMAND_WRITE, wdata, 2, NULL, 0); -} - static int acpi_ec_transaction_unlocked(struct acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len) + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { int result; @@ -292,9 +263,8 @@ static int acpi_ec_transaction_unlocked(struct acpi_ec *ec, u8 command, result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if (result) return result; - acpi_ec_write_data(ec, *(wdata++)); - } + } if (command == ACPI_EC_COMMAND_WRITE) { result = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); @@ -316,46 +286,9 @@ static int acpi_ec_transaction_unlocked(struct acpi_ec *ec, u8 command, return 0; } -static int acpi_ec_poll_transaction(struct acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len) -{ - acpi_status status = AE_OK; - int result; - u32 glk = 0; - - if (!ec || (wdata_len && !wdata) || (rdata_len && !rdata)) - return -EINVAL; - - if (rdata) - memset(rdata, 0, rdata_len); - - if (ec->global_lock) { - status = acpi_acquire_global_lock(ACPI_EC_UDELAY_GLK, &glk); - if (ACPI_FAILURE(status)) - return -ENODEV; - } - - if (down_interruptible(&ec->sem)) { - result = -ERESTARTSYS; - goto end_nosem; - } - - result = acpi_ec_transaction_unlocked(ec, command, - wdata, wdata_len, - rdata, rdata_len); - up(&ec->sem); - -end_nosem: - if (ec->global_lock) - acpi_release_global_lock(glk); - - return result; -} - -static int acpi_ec_intr_transaction(struct acpi_ec *ec, u8 command, - const u8 *wdata, unsigned wdata_len, - u8 *rdata, unsigned rdata_len) +static int acpi_ec_transaction(struct acpi_ec *ec, u8 command, + const u8 *wdata, unsigned wdata_len, + u8 *rdata, unsigned rdata_len) { int status; u32 glk; @@ -371,13 +304,11 @@ static int acpi_ec_intr_transaction(struct acpi_ec *ec, u8 command, if (ACPI_FAILURE(status)) return -ENODEV; } - - WARN_ON(in_interrupt()); down(&ec->sem); status = acpi_ec_wait(ec, ACPI_EC_EVENT_IBF_0); if (status) { - ACPI_EXCEPTION((AE_INFO, status, "read EC, IB not empty")); + printk(KERN_DEBUG PREFIX "read EC, IB not empty\n"); goto end; } @@ -394,6 +325,23 @@ end: return status; } +static int acpi_ec_read(struct acpi_ec *ec, u8 address, u32 * data) +{ + int result; + u8 d; + + result = acpi_ec_transaction(ec, ACPI_EC_COMMAND_READ, + &address, 1, &d, 1); + *data = d; + return result; +} +static int acpi_ec_write(struct acpi_ec *ec, u8 address, u8 data) +{ + u8 wdata[2] = { address, data }; + return acpi_ec_transaction(ec, ACPI_EC_COMMAND_WRITE, + wdata, 2, NULL, 0); +} + /* * Externally callable EC access functions. For now, assume 1 EC only */ @@ -447,13 +395,13 @@ extern int ec_transaction(u8 command, ec = acpi_driver_data(first_ec); - return acpi_ec_transaction(ec, command, wdata, wdata_len, rdata, rdata_len); + return acpi_ec_transaction(ec, command, wdata, + wdata_len, rdata, rdata_len); } -EXPORT_SYMBOL(ec_transaction); - -static int acpi_ec_query(struct acpi_ec *ec, u32 * data) { - int result; +static int acpi_ec_query(struct acpi_ec *ec, u32 * data) +{ + int result; u8 d; if (!ec || !data) -- cgit v0.10.2 From 8e0341ba791cc72c643340b0d8119141ae5a80c5 Mon Sep 17 00:00:00 2001 From: "Denis M. Sadykov" Date: Tue, 26 Sep 2006 19:50:33 +0400 Subject: ACPI: EC: Unify poll and interrupt gpe handlers Signed-off-by: Alexey Y. Starikovskiy Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 9c7fce6..0f232e7 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -112,11 +112,6 @@ static struct acpi_ec *ec_ecdt; static struct acpi_device *first_ec; static int acpi_ec_mode = EC_INTR; -static void acpi_ec_gpe_poll_query(void *ec_cxt); -static void acpi_ec_gpe_intr_query(void *ec_cxt); -static u32 acpi_ec_gpe_poll_handler(void *data); -static u32 acpi_ec_gpe_intr_handler(void *data); - /* -------------------------------------------------------------------------- Transaction Management -------------------------------------------------------------------------- */ @@ -428,21 +423,13 @@ static int acpi_ec_query(struct acpi_ec *ec, u32 * data) Event Management -------------------------------------------------------------------------- */ -union acpi_ec_query_data { +struct acpi_ec_query_data { acpi_handle handle; u8 data; }; static void acpi_ec_gpe_query(void *ec_cxt) { - if (acpi_ec_mode == EC_POLL) - acpi_ec_gpe_poll_query(ec_cxt); - else - acpi_ec_gpe_intr_query(ec_cxt); -} - -static void acpi_ec_gpe_poll_query(void *ec_cxt) -{ struct acpi_ec *ec = (struct acpi_ec *)ec_cxt; u32 value = 0; static char object_name[5] = { '_', 'Q', '0', '0', '\0' }; @@ -454,18 +441,8 @@ static void acpi_ec_gpe_poll_query(void *ec_cxt) if (!ec_cxt) goto end; - if (down_interruptible (&ec->sem)) { - return; - } value = acpi_ec_read_status(ec); - up(&ec->sem); - /* TBD: Implement asynch events! - * NOTE: All we care about are EC-SCI's. Other EC events are - * handled via polling (yuck!). This is because some systems - * treat EC-SCIs as level (versus EDGE!) triggered, preventing - * a purely interrupt-driven approach (grumble, grumble). - */ if (!(value & ACPI_EC_FLAG_SCI)) goto end; @@ -475,96 +452,36 @@ static void acpi_ec_gpe_poll_query(void *ec_cxt) object_name[2] = hex[((value >> 4) & 0x0F)]; object_name[3] = hex[(value & 0x0F)]; - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Evaluating %s\n", object_name)); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Evaluating %s", object_name)); acpi_evaluate_object(ec->handle, object_name, NULL, NULL); end: acpi_enable_gpe(NULL, ec->gpe_bit, ACPI_NOT_ISR); } -static void acpi_ec_gpe_intr_query(void *ec_cxt) -{ - struct acpi_ec *ec = (struct acpi_ec *)ec_cxt; - u32 value; - int result = -ENODATA; - static char object_name[5] = { '_', 'Q', '0', '0', '\0' }; - const char hex[] = { '0', '1', '2', '3', '4', '5', '6', '7', - '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' - }; - - - if (acpi_ec_read_status(ec) & ACPI_EC_FLAG_SCI) - result = acpi_ec_query(ec, &value); - - if (result) - goto end; - - object_name[2] = hex[((value >> 4) & 0x0F)]; - object_name[3] = hex[(value & 0x0F)]; - - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Evaluating %s\n", object_name)); - - acpi_evaluate_object(ec->handle, object_name, NULL, NULL); - end: - return; -} static u32 acpi_ec_gpe_handler(void *data) { - if (acpi_ec_mode == EC_POLL) - return acpi_ec_gpe_poll_handler(data); - else - return acpi_ec_gpe_intr_handler(data); -} -static u32 acpi_ec_gpe_poll_handler(void *data) -{ - acpi_status status = AE_OK; - struct acpi_ec *ec = (struct acpi_ec *)data; - - if (!ec) - return ACPI_INTERRUPT_NOT_HANDLED; - - acpi_disable_gpe(NULL, ec->gpe_bit, ACPI_ISR); - - status = acpi_os_execute(OSL_EC_POLL_HANDLER, acpi_ec_gpe_query, ec); - - if (status == AE_OK) - return ACPI_INTERRUPT_HANDLED; - else - return ACPI_INTERRUPT_NOT_HANDLED; -} -static u32 acpi_ec_gpe_intr_handler(void *data) -{ acpi_status status = AE_OK; u32 value; + u8 exec_mode; struct acpi_ec *ec = (struct acpi_ec *)data; - if (!ec) - return ACPI_INTERRUPT_NOT_HANDLED; - acpi_clear_gpe(NULL, ec->gpe_bit, ACPI_ISR); value = acpi_ec_read_status(ec); - switch (ec->expect_event) { - case ACPI_EC_EVENT_OBF_1: - if (!(value & ACPI_EC_FLAG_OBF)) - break; - ec->expect_event = 0; - wake_up(&ec->wait); - break; - case ACPI_EC_EVENT_IBF_0: - if ((value & ACPI_EC_FLAG_IBF)) - break; - ec->expect_event = 0; - wake_up(&ec->wait); - break; - default: - break; + if (acpi_ec_mode == EC_INTR) { + if (acpi_ec_check_status(value, ec->expect_event)) { + ec->expect_event = 0; + wake_up(&ec->wait); + } + exec_mode = OSL_EC_BURST_HANDLER; + } else { + exec_mode = OSL_EC_POLL_HANDLER; } if (value & ACPI_EC_FLAG_SCI) { - status = acpi_os_execute(OSL_EC_BURST_HANDLER, - acpi_ec_gpe_query, ec); + status = acpi_os_execute(exec_mode, acpi_ec_gpe_query, ec); return status == AE_OK ? ACPI_INTERRUPT_HANDLED : ACPI_INTERRUPT_NOT_HANDLED; } -- cgit v0.10.2 From 6ffb221a82de962f31034b45d945e203a0f0500f Mon Sep 17 00:00:00 2001 From: "Denis M. Sadykov" Date: Tue, 26 Sep 2006 19:50:33 +0400 Subject: ACPI: EC: Simplify acpi_hw_low_level*() with inb()/outb(). Simplify acpi_hw_low_level_xxx() functions to inb() and outb(). Signed-off-by: Alexey Y. Starikovskiy Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 0f232e7..ae05e8c 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -72,7 +72,7 @@ enum { enum { EC_INTR = 1, /* Output buffer full */ - EC_POLL, /* Input buffer empty */ + EC_POLL, /* Input buffer empty */ }; static int acpi_ec_remove(struct acpi_device *device, int type); @@ -91,22 +91,20 @@ static struct acpi_driver acpi_ec_driver = { .stop = acpi_ec_stop, }, }; + +/* If we find an EC via the ECDT, we need to keep a ptr to its context */ struct acpi_ec { acpi_handle handle; unsigned long uid; unsigned long gpe_bit; - struct acpi_generic_address status_addr; - struct acpi_generic_address command_addr; - struct acpi_generic_address data_addr; + unsigned long command_addr; + unsigned long data_addr; unsigned long global_lock; struct semaphore sem; unsigned int expect_event; atomic_t leaving_burst; /* 0 : No, 1 : Yes, 2: abort */ wait_queue_head_t wait; -}; - -/* If we find an EC via the ECDT, we need to keep a ptr to its context */ -static struct acpi_ec *ec_ecdt; +} *ec_ecdt; /* External interfaces use first EC only, so remember */ static struct acpi_device *first_ec; @@ -116,34 +114,28 @@ static int acpi_ec_mode = EC_INTR; Transaction Management -------------------------------------------------------------------------- */ -static u32 acpi_ec_read_status(struct acpi_ec *ec) +static inline u8 acpi_ec_read_status(struct acpi_ec *ec) { - u32 status = 0; - - acpi_hw_low_level_read(8, &status, &ec->status_addr); - return status; + return inb(ec->command_addr); } -static u32 acpi_ec_read_data(struct acpi_ec *ec) +static inline u8 acpi_ec_read_data(struct acpi_ec *ec) { - u32 data = 0; - - acpi_hw_low_level_read(8, &data, &ec->data_addr); - return data; + return inb(ec->data_addr); } -static void acpi_ec_write_cmd(struct acpi_ec *ec, u32 command) +static inline void acpi_ec_write_cmd(struct acpi_ec *ec, u8 command) { - acpi_hw_low_level_write(8, command, &ec->command_addr); + outb(command, ec->command_addr); } -static void acpi_ec_write_data(struct acpi_ec *ec, u32 data) +static inline void acpi_ec_write_data(struct acpi_ec *ec, u8 data) { - acpi_hw_low_level_write(8, data, &ec->data_addr); + outb(data, ec->data_addr); } -static int acpi_ec_check_status(u32 status, u8 event) { - +static int acpi_ec_check_status(u8 status, u8 event) +{ switch (event) { case ACPI_EC_EVENT_OBF_1: if (status & ACPI_EC_FLAG_OBF) @@ -201,8 +193,8 @@ static int acpi_ec_wait(struct acpi_ec *ec, u8 event) */ int acpi_ec_enter_burst_mode(struct acpi_ec *ec) { - u32 tmp = 0; - u32 status = 0; + u8 tmp = 0; + u8 status = 0; status = acpi_ec_read_status(ec); @@ -227,7 +219,7 @@ int acpi_ec_enter_burst_mode(struct acpi_ec *ec) int acpi_ec_leave_burst_mode(struct acpi_ec *ec) { - u32 status = 0; + u8 status = 0; status = acpi_ec_read_status(ec); @@ -268,14 +260,11 @@ static int acpi_ec_transaction_unlocked(struct acpi_ec *ec, u8 command, } for (; rdata_len > 0; rdata_len --) { - u32 d; - result = acpi_ec_wait(ec, ACPI_EC_EVENT_OBF_1); if (result) return result; - d = acpi_ec_read_data(ec); - *(rdata++) = (u8) d; + *(rdata++) = acpi_ec_read_data(ec); } return 0; @@ -320,7 +309,7 @@ end: return status; } -static int acpi_ec_read(struct acpi_ec *ec, u8 address, u32 * data) +static int acpi_ec_read(struct acpi_ec *ec, u8 address, u8 *data) { int result; u8 d; @@ -330,6 +319,7 @@ static int acpi_ec_read(struct acpi_ec *ec, u8 address, u32 * data) *data = d; return result; } + static int acpi_ec_write(struct acpi_ec *ec, u8 address, u8 data) { u8 wdata[2] = { address, data }; @@ -340,11 +330,11 @@ static int acpi_ec_write(struct acpi_ec *ec, u8 address, u8 data) /* * Externally callable EC access functions. For now, assume 1 EC only */ -int ec_read(u8 addr, u8 * val) +int ec_read(u8 addr, u8 *val) { struct acpi_ec *ec; int err; - u32 temp_data; + u8 temp_data; if (!first_ec) return -ENODEV; @@ -394,7 +384,7 @@ extern int ec_transaction(u8 command, wdata_len, rdata, rdata_len); } -static int acpi_ec_query(struct acpi_ec *ec, u32 * data) +static int acpi_ec_query(struct acpi_ec *ec, u8 *data) { int result; u8 d; @@ -431,14 +421,10 @@ struct acpi_ec_query_data { static void acpi_ec_gpe_query(void *ec_cxt) { struct acpi_ec *ec = (struct acpi_ec *)ec_cxt; - u32 value = 0; - static char object_name[5] = { '_', 'Q', '0', '0', '\0' }; - const char hex[] = { '0', '1', '2', '3', '4', '5', '6', '7', - '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' - }; - + u8 value = 0; + static char object_name[8]; - if (!ec_cxt) + if (!ec) goto end; value = acpi_ec_read_status(ec); @@ -449,8 +435,7 @@ static void acpi_ec_gpe_query(void *ec_cxt) if (acpi_ec_query(ec, &value)) goto end; - object_name[2] = hex[((value >> 4) & 0x0F)]; - object_name[3] = hex[(value & 0x0F)]; + snprintf(object_name, 8, "_Q%2.2X", value); ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Evaluating %s", object_name)); @@ -463,8 +448,7 @@ static void acpi_ec_gpe_query(void *ec_cxt) static u32 acpi_ec_gpe_handler(void *data) { acpi_status status = AE_OK; - u32 value; - u8 exec_mode; + u8 value; struct acpi_ec *ec = (struct acpi_ec *)data; acpi_clear_gpe(NULL, ec->gpe_bit, ACPI_ISR); @@ -475,13 +459,10 @@ static u32 acpi_ec_gpe_handler(void *data) ec->expect_event = 0; wake_up(&ec->wait); } - exec_mode = OSL_EC_BURST_HANDLER; - } else { - exec_mode = OSL_EC_POLL_HANDLER; } if (value & ACPI_EC_FLAG_SCI) { - status = acpi_os_execute(exec_mode, acpi_ec_gpe_query, ec); + status = acpi_os_execute(OSL_EC_BURST_HANDLER, acpi_ec_gpe_query, ec); return status == AE_OK ? ACPI_INTERRUPT_HANDLED : ACPI_INTERRUPT_NOT_HANDLED; } @@ -535,7 +516,7 @@ acpi_ec_space_handler(u32 function, switch (function) { case ACPI_READ: temp = 0; - result = acpi_ec_read(ec, (u8) address, (u32 *) & temp); + result = acpi_ec_read(ec, (u8) address, (u8 *) &temp); break; case ACPI_WRITE: result = acpi_ec_write(ec, (u8) address, (u8) temp); @@ -595,8 +576,8 @@ static int acpi_ec_read_info(struct seq_file *seq, void *offset) seq_printf(seq, "gpe bit: 0x%02x\n", (u32) ec->gpe_bit); seq_printf(seq, "ports: 0x%02x, 0x%02x\n", - (u32) ec->status_addr.address, - (u32) ec->data_addr.address); + (u32) ec->command_addr, + (u32) ec->data_addr); seq_printf(seq, "use global lock: %s\n", ec->global_lock ? "yes" : "no"); acpi_enable_gpe(NULL, ec->gpe_bit, ACPI_NOT_ISR); @@ -752,7 +733,6 @@ static acpi_status acpi_ec_io_ports(struct acpi_resource *resource, void *context) { struct acpi_ec *ec = (struct acpi_ec *)context; - struct acpi_generic_address *addr; if (resource->type != ACPI_RESOURCE_TYPE_IO) { return AE_OK; @@ -763,19 +743,14 @@ acpi_ec_io_ports(struct acpi_resource *resource, void *context) * the second address region returned is the status/command * port. */ - if (ec->data_addr.register_bit_width == 0) { - addr = &ec->data_addr; - } else if (ec->command_addr.register_bit_width == 0) { - addr = &ec->command_addr; + if (ec->data_addr == 0) { + ec->data_addr = resource->data.io.minimum; + } else if (ec->command_addr == 0) { + ec->command_addr = resource->data.io.minimum; } else { return AE_CTRL_TERMINATE; } - addr->address_space_id = ACPI_ADR_SPACE_SYSTEM_IO; - addr->register_bit_width = 8; - addr->register_bit_offset = 0; - addr->address = resource->data.io.minimum; - return AE_OK; } @@ -798,19 +773,14 @@ static int acpi_ec_start(struct acpi_device *device) */ status = acpi_walk_resources(ec->handle, METHOD_NAME__CRS, acpi_ec_io_ports, ec); - if (ACPI_FAILURE(status) - || ec->command_addr.register_bit_width == 0) { + if (ACPI_FAILURE(status) || ec->command_addr == 0) { ACPI_EXCEPTION((AE_INFO, status, "Error getting I/O port addresses")); return -ENODEV; } - ec->status_addr = ec->command_addr; - - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "gpe=0x%02x, ports=0x%2x,0x%2x", - (u32) ec->gpe_bit, - (u32) ec->command_addr.address, - (u32) ec->data_addr.address)); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "gpe=0x%02lx, ports=0x%2lx,0x%2lx", + ec->gpe_bit, ec->command_addr, ec->data_addr)); /* * Install GPE handler @@ -877,7 +847,6 @@ acpi_fake_ecdt_callback(acpi_handle handle, acpi_ec_io_ports, ec_ecdt); if (ACPI_FAILURE(status)) return status; - ec_ecdt->status_addr = ec_ecdt->command_addr; ec_ecdt->uid = -1; acpi_evaluate_integer(handle, "_UID", NULL, &ec_ecdt->uid); @@ -890,10 +859,8 @@ acpi_fake_ecdt_callback(acpi_handle handle, ec_ecdt->global_lock = TRUE; ec_ecdt->handle = handle; - ACPI_DEBUG_PRINT((ACPI_DB_INFO, "GPE=0x%02x, ports=0x%2x, 0x%2x", - (u32) ec_ecdt->gpe_bit, - (u32) ec_ecdt->command_addr.address, - (u32) ec_ecdt->data_addr.address)); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "GPE=0x%02lx, ports=0x%2lx, 0x%2lx", + ec_ecdt->gpe_bit, ec_ecdt->command_addr, ec_ecdt->data_addr)); return AE_CTRL_TERMINATE; } @@ -961,9 +928,8 @@ static int __init acpi_ec_get_real_ecdt(void) if (acpi_ec_mode == EC_INTR) { init_waitqueue_head(&ec_ecdt->wait); } - ec_ecdt->command_addr = ecdt_ptr->ec_control; - ec_ecdt->status_addr = ecdt_ptr->ec_control; - ec_ecdt->data_addr = ecdt_ptr->ec_data; + ec_ecdt->command_addr = ecdt_ptr->ec_control.address; + ec_ecdt->data_addr = ecdt_ptr->ec_data.address; ec_ecdt->gpe_bit = ecdt_ptr->gpe_bit; /* use the GL just to be safe */ ec_ecdt->global_lock = TRUE; -- cgit v0.10.2 From ab9e43c640b2b7d6e296fc39dd8cbcb96f9ae393 Mon Sep 17 00:00:00 2001 From: Lennart Poettering Date: Tue, 3 Oct 2006 22:49:00 -0400 Subject: ACPI: EC: export ec_transaction() for msi-laptop driver Signed-off-by: Lennart Poettering Signed-off-by: Len Brown diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index ae05e8c..e6d4b08 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -384,6 +384,8 @@ extern int ec_transaction(u8 command, wdata_len, rdata, rdata_len); } +EXPORT_SYMBOL(ec_transaction); + static int acpi_ec_query(struct acpi_ec *ec, u8 *data) { int result; -- cgit v0.10.2 From 8c4c731a89ea6458001f48033f8988447736fb38 Mon Sep 17 00:00:00 2001 From: Lennart Poettering Date: Fri, 6 Oct 2006 01:27:02 -0400 Subject: MSI S270 Laptop support: backlight, wlan, bluetooth states Create a driver to support the platform-specific features of MSI S270 laptops (and maybe other MSI laptops). This driver implements a backlight device for controlling LCD brightness (/sys/class/backlight/msi-laptop-bl/). In addition it allows access to the WLAN and Bluetooth states through a platform driver (/sys/devices/platform/msi-laptop-pf/). Signed-off-by: Lennart Poettering Cc: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/MAINTAINERS b/MAINTAINERS index 1b5430a..a2b6d9f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1998,6 +1998,13 @@ M: rubini@ipvvis.unipv.it L: linux-kernel@vger.kernel.org S: Maintained +MSI LAPTOP SUPPORT +P: Lennart Poettering +M: mzxreary@0pointer.de +L: https://tango.0pointer.de/mailman/listinfo/s270-linux +W: http://0pointer.de/lennart/tchibo.html +S: Maintained + MTRR AND SIMILAR SUPPORT [i386] P: Richard Gooch M: rgooch@atnf.csiro.au diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index 3df0e7a..fa7acc2 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -57,4 +57,23 @@ config TIFM_7XX1 To compile this driver as a module, choose M here: the module will be called tifm_7xx1. +config MSI_LAPTOP + tristate "MSI Laptop Extras" + depends on X86 + depends on ACPI_EC + depends on BACKLIGHT_CLASS_DEVICE + ---help--- + This is a driver for laptops built by MSI (MICRO-STAR + INTERNATIONAL): + + MSI MegaBook S270 (MS-1013) + Cytron/TCM/Medion/Tchibo MD96100/SAM2000 + + It adds support for Bluetooth, WLAN and LCD brightness control. + + More information about this driver is available at + . + + If you have an MSI S270 laptop, say Y or M here. + endmenu diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile index d65ece7..9a91c1e 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -5,6 +5,7 @@ obj- := misc.o # Dummy rule to force built-in.o to be made obj-$(CONFIG_IBM_ASM) += ibmasm/ obj-$(CONFIG_HDPU_FEATURES) += hdpuftrs/ +obj-$(CONFIG_MSI_LAPTOP) += msi-laptop.o obj-$(CONFIG_LKDTM) += lkdtm.o obj-$(CONFIG_TIFM_CORE) += tifm_core.o obj-$(CONFIG_TIFM_7XX1) += tifm_7xx1.o diff --git a/drivers/misc/msi-laptop.c b/drivers/misc/msi-laptop.c new file mode 100644 index 0000000..fdb7153 --- /dev/null +++ b/drivers/misc/msi-laptop.c @@ -0,0 +1,395 @@ +/*-*-linux-c-*-*/ + +/* + Copyright (C) 2006 Lennart Poettering + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301, USA. + */ + +/* + * msi-laptop.c - MSI S270 laptop support. This laptop is sold under + * various brands, including "Cytron/TCM/Medion/Tchibo MD96100". + * + * This driver exports a few files in /sys/devices/platform/msi-laptop-pf/: + * + * lcd_level - Screen brightness: contains a single integer in the + * range 0..8. (rw) + * + * auto_brightness - Enable automatic brightness control: contains + * either 0 or 1. If set to 1 the hardware adjusts the screen + * brightness automatically when the power cord is + * plugged/unplugged. (rw) + * + * wlan - WLAN subsystem enabled: contains either 0 or 1. (ro) + * + * bluetooth - Bluetooth subsystem enabled: contains either 0 or 1 + * Please note that this file is constantly 0 if no Bluetooth + * hardware is available. (ro) + * + * In addition to these platform device attributes the driver + * registers itself in the Linux backlight control subsystem and is + * available to userspace under /sys/class/backlight/msi-laptop-bl/. + * + * This driver might work on other laptops produced by MSI. If you + * want to try it you can pass force=1 as argument to the module which + * will force it to load even when the DMI data doesn't identify the + * laptop as MSI S270. YMMV. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#define MSI_DRIVER_VERSION "0.5" + +#define MSI_LCD_LEVEL_MAX 9 + +#define MSI_EC_COMMAND_WIRELESS 0x10 +#define MSI_EC_COMMAND_LCD_LEVEL 0x11 + +static int force; +module_param(force, bool, 0); +MODULE_PARM_DESC(force, "Force driver load, ignore DMI data"); + +static int auto_brightness; +module_param(auto_brightness, int, 0); +MODULE_PARM_DESC(auto_brightness, "Enable automatic brightness control (0: disabled; 1: enabled; 2: don't touch)"); + +/* Hardware access */ + +static int set_lcd_level(int level) +{ + u8 buf[2]; + + if (level < 0 || level >= MSI_LCD_LEVEL_MAX) + return -EINVAL; + + buf[0] = 0x80; + buf[1] = (u8) (level*31); + + return ec_transaction(MSI_EC_COMMAND_LCD_LEVEL, buf, sizeof(buf), NULL, 0); +} + +static int get_lcd_level(void) +{ + u8 wdata = 0, rdata; + int result; + + result = ec_transaction(MSI_EC_COMMAND_LCD_LEVEL, &wdata, 1, &rdata, 1); + if (result < 0) + return result; + + return (int) rdata / 31; +} + +static int get_auto_brightness(void) +{ + u8 wdata = 4, rdata; + int result; + + result = ec_transaction(MSI_EC_COMMAND_LCD_LEVEL, &wdata, 1, &rdata, 1); + if (result < 0) + return result; + + return !!(rdata & 8); +} + +static int set_auto_brightness(int enable) +{ + u8 wdata[2], rdata; + int result; + + wdata[0] = 4; + + result = ec_transaction(MSI_EC_COMMAND_LCD_LEVEL, wdata, 1, &rdata, 1); + if (result < 0) + return result; + + wdata[0] = 0x84; + wdata[1] = (rdata & 0xF7) | (enable ? 8 : 0); + + return ec_transaction(MSI_EC_COMMAND_LCD_LEVEL, wdata, 2, NULL, 0); +} + +static int get_wireless_state(int *wlan, int *bluetooth) +{ + u8 wdata = 0, rdata; + int result; + + result = ec_transaction(MSI_EC_COMMAND_WIRELESS, &wdata, 1, &rdata, 1); + if (result < 0) + return -1; + + if (wlan) + *wlan = !!(rdata & 8); + + if (bluetooth) + *bluetooth = !!(rdata & 128); + + return 0; +} + +/* Backlight device stuff */ + +static int bl_get_brightness(struct backlight_device *b) +{ + return get_lcd_level(); +} + + +static int bl_update_status(struct backlight_device *b) +{ + return set_lcd_level(b->props->brightness); +} + +static struct backlight_properties msibl_props = { + .owner = THIS_MODULE, + .get_brightness = bl_get_brightness, + .update_status = bl_update_status, + .max_brightness = MSI_LCD_LEVEL_MAX-1, +}; + +static struct backlight_device *msibl_device; + +/* Platform device */ + +static ssize_t show_wlan(struct device *dev, + struct device_attribute *attr, char *buf) +{ + + int ret, enabled; + + ret = get_wireless_state(&enabled, NULL); + if (ret < 0) + return ret; + + return sprintf(buf, "%i\n", enabled); +} + +static ssize_t show_bluetooth(struct device *dev, + struct device_attribute *attr, char *buf) +{ + + int ret, enabled; + + ret = get_wireless_state(NULL, &enabled); + if (ret < 0) + return ret; + + return sprintf(buf, "%i\n", enabled); +} + +static ssize_t show_lcd_level(struct device *dev, + struct device_attribute *attr, char *buf) +{ + + int ret; + + ret = get_lcd_level(); + if (ret < 0) + return ret; + + return sprintf(buf, "%i\n", ret); +} + +static ssize_t store_lcd_level(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + + int level, ret; + + if (sscanf(buf, "%i", &level) != 1 || (level < 0 || level >= MSI_LCD_LEVEL_MAX)) + return -EINVAL; + + ret = set_lcd_level(level); + if (ret < 0) + return ret; + + return count; +} + +static ssize_t show_auto_brightness(struct device *dev, + struct device_attribute *attr, char *buf) +{ + + int ret; + + ret = get_auto_brightness(); + if (ret < 0) + return ret; + + return sprintf(buf, "%i\n", ret); +} + +static ssize_t store_auto_brightness(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + + int enable, ret; + + if (sscanf(buf, "%i", &enable) != 1 || (enable != (enable & 1))) + return -EINVAL; + + ret = set_auto_brightness(enable); + if (ret < 0) + return ret; + + return count; +} + +static DEVICE_ATTR(lcd_level, 0644, show_lcd_level, store_lcd_level); +static DEVICE_ATTR(auto_brightness, 0644, show_auto_brightness, store_auto_brightness); +static DEVICE_ATTR(bluetooth, 0444, show_bluetooth, NULL); +static DEVICE_ATTR(wlan, 0444, show_wlan, NULL); + +static struct attribute *msipf_attributes[] = { + &dev_attr_lcd_level.attr, + &dev_attr_auto_brightness.attr, + &dev_attr_bluetooth.attr, + &dev_attr_wlan.attr, + NULL +}; + +static struct attribute_group msipf_attribute_group = { + .attrs = msipf_attributes +}; + +static struct platform_driver msipf_driver = { + .driver = { + .name = "msi-laptop-pf", + .owner = THIS_MODULE, + } +}; + +static struct platform_device *msipf_device; + +/* Initialization */ + +static struct dmi_system_id __initdata msi_dmi_table[] = { + { + .ident = "MSI S270", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "MICRO-STAR INT'L CO.,LTD"), + DMI_MATCH(DMI_PRODUCT_NAME, "MS-1013"), + } + }, + { + .ident = "Medion MD96100", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "NOTEBOOK"), + DMI_MATCH(DMI_PRODUCT_NAME, "SAM2000"), + } + }, + { } +}; + + +static int __init msi_init(void) +{ + int ret; + + if (acpi_disabled) + return -ENODEV; + + if (!force && !dmi_check_system(msi_dmi_table)) + return -ENODEV; + + if (auto_brightness < 0 || auto_brightness > 2) + return -EINVAL; + + /* Register backlight stuff */ + + msibl_device = backlight_device_register("msi-laptop-bl", NULL, &msibl_props); + if (IS_ERR(msibl_device)) + return PTR_ERR(msibl_device); + + ret = platform_driver_register(&msipf_driver); + if (ret) + goto fail_backlight; + + /* Register platform stuff */ + + msipf_device = platform_device_alloc("msi-laptop-pf", -1); + if (!msipf_device) { + ret = -ENOMEM; + goto fail_platform_driver; + } + + ret = platform_device_add(msipf_device); + if (ret) + goto fail_platform_device1; + + ret = sysfs_create_group(&msipf_device->dev.kobj, &msipf_attribute_group); + if (ret) + goto fail_platform_device2; + + /* Disable automatic brightness control by default because + * this module was probably loaded to do brightness control in + * software. */ + + if (auto_brightness != 2) + set_auto_brightness(auto_brightness); + + printk(KERN_INFO "msi-laptop: driver "MSI_DRIVER_VERSION" successfully loaded.\n"); + + return 0; + +fail_platform_device2: + + platform_device_del(msipf_device); + +fail_platform_device1: + + platform_device_put(msipf_device); + +fail_platform_driver: + + platform_driver_unregister(&msipf_driver); + +fail_backlight: + + backlight_device_unregister(msibl_device); + + return ret; +} + +static void __exit msi_cleanup(void) +{ + + sysfs_remove_group(&msipf_device->dev.kobj, &msipf_attribute_group); + platform_device_unregister(msipf_device); + platform_driver_unregister(&msipf_driver); + backlight_device_unregister(msibl_device); + + /* Enable automatic brightness control again */ + if (auto_brightness != 2) + set_auto_brightness(1); + + printk(KERN_INFO "msi-laptop: driver unloaded.\n"); +} + +module_init(msi_init); +module_exit(msi_cleanup); + +MODULE_AUTHOR("Lennart Poettering"); +MODULE_DESCRIPTION("MSI Laptop Support"); +MODULE_VERSION(MSI_DRIVER_VERSION); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From fcfc638c6b1345b6646523dbab0065b36a868ffc Mon Sep 17 00:00:00 2001 From: "Alexey Y. Starikovskiy" Date: Tue, 26 Sep 2006 04:18:16 -0400 Subject: ACPI: Remove deferred execution from global lock acquire wakeup path On acquiring the ACPI global lock, if there were sleepers on the lock, we used to use acpi_os_execute() to defer a thread which would signal sleepers. Now just signal the semaphore directly. http://bugzilla.kernel.org/show_bug.cgi?id=5534#c159 Signed-off-by: Len Brown diff --git a/drivers/acpi/events/evmisc.c b/drivers/acpi/events/evmisc.c index 6eef4ef..ee2a10b 100644 --- a/drivers/acpi/events/evmisc.c +++ b/drivers/acpi/events/evmisc.c @@ -342,20 +342,8 @@ static u32 acpi_ev_global_lock_handler(void *context) if (acquired) { /* Got the lock, now wake all threads waiting for it */ - acpi_gbl_global_lock_acquired = TRUE; - - /* Run the Global Lock thread which will signal all waiting threads */ - - status = - acpi_os_execute(OSL_GLOBAL_LOCK_HANDLER, - acpi_ev_global_lock_thread, context); - if (ACPI_FAILURE(status)) { - ACPI_EXCEPTION((AE_INFO, status, - "Could not queue Global Lock thread")); - - return (ACPI_INTERRUPT_NOT_HANDLED); - } + acpi_ev_global_lock_thread(context); } return (ACPI_INTERRUPT_HANDLED); -- cgit v0.10.2 From 37605a6900f6b4d886d995751fcfeef88c4e462c Mon Sep 17 00:00:00 2001 From: "Alexey Y. Starikovskiy" Date: Tue, 26 Sep 2006 04:20:47 -0400 Subject: ACPI: created a dedicated workqueue for notify() execution http://bugzilla.kernel.org/show_bug.cgi?id=5534#c160 Signed-off-by: Len Brown diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index 068fe4f..c84286c 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -73,6 +73,7 @@ static unsigned int acpi_irq_irq; static acpi_osd_handler acpi_irq_handler; static void *acpi_irq_context; static struct workqueue_struct *kacpid_wq; +static struct workqueue_struct *kacpi_notify_wq; acpi_status acpi_os_initialize(void) { @@ -91,8 +92,9 @@ acpi_status acpi_os_initialize1(void) return AE_NULL_ENTRY; } kacpid_wq = create_singlethread_workqueue("kacpid"); + kacpi_notify_wq = create_singlethread_workqueue("kacpi_notify"); BUG_ON(!kacpid_wq); - + BUG_ON(!kacpi_notify_wq); return AE_OK; } @@ -104,6 +106,7 @@ acpi_status acpi_os_terminate(void) } destroy_workqueue(kacpid_wq); + destroy_workqueue(kacpi_notify_wq); return AE_OK; } @@ -566,10 +569,7 @@ void acpi_os_derive_pci_id(acpi_handle rhandle, /* upper bound */ static void acpi_os_execute_deferred(void *context) { - struct acpi_os_dpc *dpc = NULL; - - - dpc = (struct acpi_os_dpc *)context; + struct acpi_os_dpc *dpc = (struct acpi_os_dpc *)context; if (!dpc) { printk(KERN_ERR PREFIX "Invalid (NULL) context\n"); return; @@ -604,14 +604,12 @@ acpi_status acpi_os_execute(acpi_execute_type type, struct acpi_os_dpc *dpc; struct work_struct *task; - ACPI_FUNCTION_TRACE("os_queue_for_execution"); - ACPI_DEBUG_PRINT((ACPI_DB_EXEC, "Scheduling function [%p(%p)] for deferred execution.\n", function, context)); if (!function) - return_ACPI_STATUS(AE_BAD_PARAMETER); + return AE_BAD_PARAMETER; /* * Allocate/initialize DPC structure. Note that this memory will be @@ -624,26 +622,20 @@ acpi_status acpi_os_execute(acpi_execute_type type, * from the same memory. */ - dpc = - kmalloc(sizeof(struct acpi_os_dpc) + sizeof(struct work_struct), - GFP_ATOMIC); + dpc = kmalloc(sizeof(struct acpi_os_dpc) + + sizeof(struct work_struct), GFP_ATOMIC); if (!dpc) - return_ACPI_STATUS(AE_NO_MEMORY); - + return AE_NO_MEMORY; dpc->function = function; dpc->context = context; - task = (void *)(dpc + 1); INIT_WORK(task, acpi_os_execute_deferred, (void *)dpc); - - if (!queue_work(kacpid_wq, task)) { - ACPI_DEBUG_PRINT((ACPI_DB_ERROR, - "Call to queue_work() failed.\n")); - kfree(dpc); + if (!queue_work((type == OSL_NOTIFY_HANDLER)? + kacpi_notify_wq : kacpid_wq, task)) { status = AE_ERROR; + kfree(dpc); } - - return_ACPI_STATUS(status); + return status; } EXPORT_SYMBOL(acpi_os_execute); -- cgit v0.10.2 From 2fe6dffabb06bfa0591c8c490b092b458fba1f06 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Thu, 31 Aug 2006 14:15:54 +0200 Subject: ACPI: ibm_acpi: delete obsolete documentation As this module is now part of the kernel tree, there is no need for instructions on how to download it and build an external module. Signed-off-by: Pavel Machek Signed-off-by: Len Brown diff --git a/Documentation/ibm-acpi.txt b/Documentation/ibm-acpi.txt index 00b8cf39..e50595b 100644 --- a/Documentation/ibm-acpi.txt +++ b/Documentation/ibm-acpi.txt @@ -53,40 +53,7 @@ Installation If you are compiling this driver as included in the Linux kernel sources, simply enable the CONFIG_ACPI_IBM option (Power Management / -ACPI / IBM ThinkPad Laptop Extras). The rest of this section describes -how to install this driver when downloaded from the web site. - -First, you need to get a kernel with ACPI support up and running. -Please refer to http://acpi.sourceforge.net/ for help with this -step. How successful you will be depends a lot on you ThinkPad model, -the kernel you are using and any additional patches applied. The -kernel provided with your distribution may not be good enough. I -needed to compile a 2.6.7 kernel with the 20040715 ACPI patch to get -ACPI working reliably on my ThinkPad X40. Old ThinkPad models may not -be supported at all. - -Assuming you have the basic ACPI support working (e.g. you can see the -/proc/acpi directory), follow the following steps to install this -driver: - - - unpack the archive: - - tar xzvf ibm-acpi-x.y.tar.gz; cd ibm-acpi-x.y - - - compile the driver: - - make - - - install the module in your kernel modules directory: - - make install - - - load the module: - - modprobe ibm_acpi - -After loading the module, check the "dmesg" output for any error messages. - +ACPI / IBM ThinkPad Laptop Extras). Features -------- -- cgit v0.10.2 From 786f18c666d7202a86a8aa42a98783b115fe8739 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 23 Aug 2006 23:18:06 -0400 Subject: ACPI: fix potential OOPS in power driver with CONFIG_ACPI_DEBUG device was set to null and used before set in a debug printk Signed-off-by: Dmitry Torokhov Signed-off-by: Len Brown diff --git a/drivers/acpi/power.c b/drivers/acpi/power.c index fec225d..fe67a8a 100644 --- a/drivers/acpi/power.c +++ b/drivers/acpi/power.c @@ -216,10 +216,8 @@ static int acpi_power_off_device(acpi_handle handle) { int result = 0; acpi_status status = AE_OK; - struct acpi_device *device = NULL; struct acpi_power_resource *resource = NULL; - result = acpi_power_get_context(handle, &resource); if (result) return result; @@ -230,13 +228,13 @@ static int acpi_power_off_device(acpi_handle handle) if (resource->references) { ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Resource [%s] is still in use, dereferencing\n", - device->pnp.bus_id)); + resource->device->pnp.bus_id)); return 0; } if (resource->state == ACPI_POWER_RESOURCE_STATE_OFF) { ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Resource [%s] already off\n", - device->pnp.bus_id)); + resource->device->pnp.bus_id)); return 0; } @@ -251,8 +249,7 @@ static int acpi_power_off_device(acpi_handle handle) return -ENOEXEC; /* Update the power resource's _device_ power state */ - device = resource->device; - device->power.state = ACPI_STATE_D3; + resource->device->power.state = ACPI_STATE_D3; ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Resource [%s] turned off\n", resource->name)); -- cgit v0.10.2 From 168a328f30e9d1a8bc6ff55c0501e0bdc08bee60 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Thu, 24 Aug 2006 00:36:19 -0400 Subject: ACPI: acpi_pci_link_set() can allocate with either GFP_ATOMIC or GFP_KERNEL acpi_pci_link_set() allocates both with interrupts on and with interrupts off (resume-time), so check interrupts and decide on GFP_ATOMIC or GFP_KERNEL at run-time. Signed-off-by: Jiri Kosina Signed-off-by: Len Brown diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c index 7f3e7e7..d53bd98 100644 --- a/drivers/acpi/pci_link.c +++ b/drivers/acpi/pci_link.c @@ -307,7 +307,7 @@ static int acpi_pci_link_set(struct acpi_pci_link *link, int irq) if (!link || !irq) return -EINVAL; - resource = kmalloc(sizeof(*resource) + 1, GFP_ATOMIC); + resource = kmalloc(sizeof(*resource) + 1, irqs_disabled() ? GFP_ATOMIC: GFP_KERNEL); if (!resource) return -ENOMEM; -- cgit v0.10.2 From f4d2e2d87eac0338884b2c26f6bafed115dbac5e Mon Sep 17 00:00:00 2001 From: Len Brown Date: Thu, 14 Sep 2006 17:16:22 -0400 Subject: ACPI: update comments in motherboard.c Signed-off-by: Len Brown diff --git a/drivers/acpi/motherboard.c b/drivers/acpi/motherboard.c index ec6b7f9..2e17ec7 100644 --- a/drivers/acpi/motherboard.c +++ b/drivers/acpi/motherboard.c @@ -48,6 +48,12 @@ ACPI_MODULE_NAME("acpi_motherboard") * the io ports if they really know they can use it, while * still preventing hotplug PCI devices from using it. */ + +/* + * When CONFIG_PNP is enabled, pnp/system.c binds to PNP0C01 + * and PNP0C02, redundant with acpi_reserve_io_ranges(). + * But acpi_reserve_io_ranges() is necessary for !CONFIG_PNP. + */ static acpi_status acpi_reserve_io_ranges(struct acpi_resource *res, void *data) { struct resource *requested_res = NULL; -- cgit v0.10.2 From 7af8b66004fa827958b4871112e59a07db5b3f6b Mon Sep 17 00:00:00 2001 From: Pierre Ossman Date: Tue, 10 Oct 2006 14:20:31 -0700 Subject: ACPI: fix section for CPU init functions The ACPI processor init functions should be marked as __cpuinit as they use structures marked with __cpuinitdata. Signed-off-by: Pierre Ossman Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c index b13d644..1908e0d2 100644 --- a/drivers/acpi/processor_core.c +++ b/drivers/acpi/processor_core.c @@ -519,7 +519,7 @@ static int acpi_processor_get_info(struct acpi_processor *pr) static void *processor_device_array[NR_CPUS]; -static int acpi_processor_start(struct acpi_device *device) +static int __cpuinit acpi_processor_start(struct acpi_device *device) { int result = 0; acpi_status status = AE_OK; diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 0a395fc..8537c42 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -1105,7 +1105,7 @@ static struct notifier_block acpi_processor_latency_notifier = { .notifier_call = acpi_processor_latency_notify, }; -int acpi_processor_power_init(struct acpi_processor *pr, +int __cpuinit acpi_processor_power_init(struct acpi_processor *pr, struct acpi_device *device) { acpi_status status = 0; -- cgit v0.10.2 From a790b323fb1b73f9388426bf3b96f153d1c90d2c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 10 Oct 2006 14:20:32 -0700 Subject: ACPI: fix printk format warnings Fix printk format warnings in drivers/acpi: drivers/acpi/tables/tbget.c:326: warning: format '%X' expects type 'unsigned int', but argument 5 has type 'long unsigned int' drivers/acpi/tables/tbrsdt.c:189: warning: format '%X' expects type 'unsigned int', but argument 5 has type 'long unsigned int' Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/tables/tbget.c b/drivers/acpi/tables/tbget.c index 7856db7..11e2d44 100644 --- a/drivers/acpi/tables/tbget.c +++ b/drivers/acpi/tables/tbget.c @@ -324,7 +324,7 @@ acpi_tb_get_this_table(struct acpi_pointer *address, if (header->length < sizeof(struct acpi_table_header)) { ACPI_ERROR((AE_INFO, - "Table length (%X) is smaller than minimum (%X)", + "Table length (%X) is smaller than minimum (%zX)", header->length, sizeof(struct acpi_table_header))); return_ACPI_STATUS(AE_INVALID_TABLE_LENGTH); diff --git a/drivers/acpi/tables/tbrsdt.c b/drivers/acpi/tables/tbrsdt.c index 0ad3dbb..86a5fca 100644 --- a/drivers/acpi/tables/tbrsdt.c +++ b/drivers/acpi/tables/tbrsdt.c @@ -187,7 +187,7 @@ acpi_status acpi_tb_validate_rsdt(struct acpi_table_header *table_ptr) if (table_ptr->length < sizeof(struct acpi_table_header)) { ACPI_ERROR((AE_INFO, - "RSDT/XSDT length (%X) is smaller than minimum (%X)", + "RSDT/XSDT length (%X) is smaller than minimum (%zX)", table_ptr->length, sizeof(struct acpi_table_header))); -- cgit v0.10.2 From 281ea49b0c294649a6de47a6f8fbe5611137726b Mon Sep 17 00:00:00 2001 From: Kimball Murray Date: Tue, 10 Oct 2006 14:20:33 -0700 Subject: ACPI: SCI interrupt source override The Linux group at Stratus Technologies has come across an issue with SCI routing under ACPI. We were bitten by this when we made an x86_64 platform whose BIOS provides an Interrupt Source Override for the SCI itself. Apparently the override has no effect for the System Control Interrupt, and this appears to be because of the way the SCI is setup in the ACPI code. It does not handle the case where busirq != gsi. The code that sets up the SCI routing assumes that bus irq == global irq. So there is simply no provision for telling it otherwise. The attached patch provides this mechanism. This patch provided by David Bulkow, was tested on an i386 platform, which does not use the SCI override, and also on an x86_64 platform which does use an override. Signed-off-by: David Bulkow Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index 92f79cd..ab974ff 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -332,7 +332,7 @@ acpi_parse_ioapic(acpi_table_entry_header * header, const unsigned long end) /* * Parse Interrupt Source Override for the ACPI SCI */ -static void acpi_sci_ioapic_setup(u32 gsi, u16 polarity, u16 trigger) +static void acpi_sci_ioapic_setup(u32 bus_irq, u32 gsi, u16 polarity, u16 trigger) { if (trigger == 0) /* compatible SCI trigger is level */ trigger = 3; @@ -352,13 +352,13 @@ static void acpi_sci_ioapic_setup(u32 gsi, u16 polarity, u16 trigger) * If GSI is < 16, this will update its flags, * else it will create a new mp_irqs[] entry. */ - mp_override_legacy_irq(gsi, polarity, trigger, gsi); + mp_override_legacy_irq(bus_irq, polarity, trigger, gsi); /* * stash over-ride to indicate we've been here * and for later update of acpi_fadt */ - acpi_sci_override_gsi = gsi; + acpi_sci_override_gsi = bus_irq; return; } @@ -376,7 +376,7 @@ acpi_parse_int_src_ovr(acpi_table_entry_header * header, acpi_table_print_madt_entry(header); if (intsrc->bus_irq == acpi_fadt.sci_int) { - acpi_sci_ioapic_setup(intsrc->global_irq, + acpi_sci_ioapic_setup(intsrc->bus_irq, intsrc->global_irq, intsrc->flags.polarity, intsrc->flags.trigger); return 0; @@ -879,7 +879,7 @@ static int __init acpi_parse_madt_ioapic_entries(void) * pretend we got one so we can set the SCI flags. */ if (!acpi_sci_override_gsi) - acpi_sci_ioapic_setup(acpi_fadt.sci_int, 0, 0); + acpi_sci_ioapic_setup(acpi_fadt.sci_int, acpi_fadt.sci_int, 0, 0); /* Fill in identity legacy mapings where no override */ mp_config_acpi_legacy_irqs(); -- cgit v0.10.2 From 6311f0dac006032b82e3556874a1e18b31e80de2 Mon Sep 17 00:00:00 2001 From: Darren Jenkins Date: Tue, 10 Oct 2006 14:20:35 -0700 Subject: ACPI: asus_acpi: fix proc files parsing ICC complains about a "Pointless comparsion of unsigned interger with zero" @ line 760 & 808 of asus_acpi.c parse_arg() mentioned below returns -E but it's copied into unsigned variable... Signed-off-by: Darren Jenkins Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/asus_acpi.c b/drivers/acpi/asus_acpi.c index ebc033f..33eaca1 100644 --- a/drivers/acpi/asus_acpi.c +++ b/drivers/acpi/asus_acpi.c @@ -567,11 +567,11 @@ static int write_led(const char __user * buffer, unsigned long count, char *ledname, int ledmask, int invert) { - int value; + int rv, value; int led_out = 0; - count = parse_arg(buffer, count, &value); - if (count > 0) + rv = parse_arg(buffer, count, &value); + if (rv > 0) led_out = value ? 1 : 0; hotk->status = @@ -584,7 +584,7 @@ write_led(const char __user * buffer, unsigned long count, printk(KERN_WARNING "Asus ACPI: LED (%s) write failed\n", ledname); - return count; + return rv; } /* @@ -619,20 +619,20 @@ static int proc_write_ledd(struct file *file, const char __user * buffer, unsigned long count, void *data) { - int value; + int rv, value; - count = parse_arg(buffer, count, &value); - if (count > 0) { + rv = parse_arg(buffer, count, &value); + if (rv > 0) { if (!write_acpi_int (hotk->handle, hotk->methods->mt_ledd, value, NULL)) printk(KERN_WARNING "Asus ACPI: LED display write failed\n"); else hotk->ledd_status = (u32) value; - } else if (count < 0) + } else if (rv < 0) printk(KERN_WARNING "Asus ACPI: Error reading user input\n"); - return count; + return rv; } /* @@ -773,12 +773,12 @@ static int proc_write_lcd(struct file *file, const char __user * buffer, unsigned long count, void *data) { - int value; + int rv, value; - count = parse_arg(buffer, count, &value); - if (count > 0) + rv = parse_arg(buffer, count, &value); + if (rv > 0) set_lcd_state(value); - return count; + return rv; } static int read_brightness(void) @@ -842,18 +842,18 @@ static int proc_write_brn(struct file *file, const char __user * buffer, unsigned long count, void *data) { - int value; + int rv, value; - count = parse_arg(buffer, count, &value); - if (count > 0) { + rv = parse_arg(buffer, count, &value); + if (rv > 0) { value = (0 < value) ? ((15 < value) ? 15 : value) : 0; /* 0 <= value <= 15 */ set_brightness(value); - } else if (count < 0) { + } else if (rv < 0) { printk(KERN_WARNING "Asus ACPI: Error reading user input\n"); } - return count; + return rv; } static void set_display(int value) @@ -892,15 +892,15 @@ static int proc_write_disp(struct file *file, const char __user * buffer, unsigned long count, void *data) { - int value; + int rv, value; - count = parse_arg(buffer, count, &value); - if (count > 0) + rv = parse_arg(buffer, count, &value); + if (rv > 0) set_display(value); - else if (count < 0) + else if (rv < 0) printk(KERN_WARNING "Asus ACPI: Error reading user input\n"); - return count; + return rv; } typedef int (proc_readfunc) (char *page, char **start, off_t off, int count, -- cgit v0.10.2 From 6df05702f97f99e038ab817f4466386f6255f58d Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 10 Oct 2006 14:20:36 -0700 Subject: ACPI: asus_acpi: don't printk on writing garbage to proc files This reporting is useless (we errno anyway). Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/asus_acpi.c b/drivers/acpi/asus_acpi.c index 33eaca1..c7ac929 100644 --- a/drivers/acpi/asus_acpi.c +++ b/drivers/acpi/asus_acpi.c @@ -629,9 +629,7 @@ proc_write_ledd(struct file *file, const char __user * buffer, "Asus ACPI: LED display write failed\n"); else hotk->ledd_status = (u32) value; - } else if (rv < 0) - printk(KERN_WARNING "Asus ACPI: Error reading user input\n"); - + } return rv; } @@ -849,10 +847,7 @@ proc_write_brn(struct file *file, const char __user * buffer, value = (0 < value) ? ((15 < value) ? 15 : value) : 0; /* 0 <= value <= 15 */ set_brightness(value); - } else if (rv < 0) { - printk(KERN_WARNING "Asus ACPI: Error reading user input\n"); } - return rv; } @@ -897,9 +892,6 @@ proc_write_disp(struct file *file, const char __user * buffer, rv = parse_arg(buffer, count, &value); if (rv > 0) set_display(value); - else if (rv < 0) - printk(KERN_WARNING "Asus ACPI: Error reading user input\n"); - return rv; } -- cgit v0.10.2 From aeb1104814e1177b865eec4f4b364602f5fcb6d1 Mon Sep 17 00:00:00 2001 From: Eiichiro Oiwa Date: Mon, 2 Oct 2006 19:18:03 +0400 Subject: ACPICA: Fix incorrect handling of PCI Express Root Bridge _HID I could not get correct PCI Express bus number from the structure of acpi_object_extra. I always get zero as bus number regardless of bus location. I found that there is incorrect comparison with _HID (PNP0A08) in acpi/events/evrgnini.c and PCI Express _BBN method always fail. Therefore, we always get zero as PCI Express bus number. http://bugzilla.kernel.org/show_bug.cgi?id=7145 Signed-off-by: Bob Moore Signed-off-by: Alexey Starikovskiy Signed-off-by: Len Brown diff --git a/drivers/acpi/events/evrgnini.c b/drivers/acpi/events/evrgnini.c index 5b3c7a8..203d135 100644 --- a/drivers/acpi/events/evrgnini.c +++ b/drivers/acpi/events/evrgnini.c @@ -225,13 +225,12 @@ acpi_ev_pci_config_region_setup(acpi_handle handle, if (! (ACPI_STRNCMP (object_hID.value, PCI_ROOT_HID_STRING, - sizeof(PCI_ROOT_HID_STRING)) - || - !(ACPI_STRNCMP - (object_hID.value, - PCI_EXPRESS_ROOT_HID_STRING, - sizeof(PCI_EXPRESS_ROOT_HID_STRING))))) - { + sizeof(PCI_ROOT_HID_STRING))) + || + !(ACPI_STRNCMP + (object_hID.value, + PCI_EXPRESS_ROOT_HID_STRING, + sizeof(PCI_EXPRESS_ROOT_HID_STRING)))) { /* Install a handler for this PCI root bridge */ -- cgit v0.10.2 From 34c4415ab857dc6d51db08d62bcd45d4b8513bb6 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Tue, 10 Oct 2006 14:20:41 -0700 Subject: ACPI: check battery status on resume for un/plug events during sleep Add ->resume method to the ACPI battery handler to check if the battery state has changed during sleep. If yes, update the ACPI internal data structures for benefit of /proc/acpi/battery/. Signed-off-by: Jiri Kosina Cc: Stefan Seyfried Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Len Brown diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index 9810e2a..026e4075 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -64,6 +64,7 @@ extern void *acpi_unlock_battery_dir(struct proc_dir_entry *acpi_battery_dir); static int acpi_battery_add(struct acpi_device *device); static int acpi_battery_remove(struct acpi_device *device, int type); +static int acpi_battery_resume(struct acpi_device *device, int status); static struct acpi_driver acpi_battery_driver = { .name = ACPI_BATTERY_DRIVER_NAME, @@ -71,6 +72,7 @@ static struct acpi_driver acpi_battery_driver = { .ids = ACPI_BATTERY_HID, .ops = { .add = acpi_battery_add, + .resume = acpi_battery_resume, .remove = acpi_battery_remove, }, }; @@ -753,6 +755,18 @@ static int acpi_battery_remove(struct acpi_device *device, int type) return 0; } +/* this is needed to learn about changes made in suspended state */ +static int acpi_battery_resume(struct acpi_device *device, int state) +{ + struct acpi_battery *battery; + + if (!device) + return -EINVAL; + + battery = device->driver_data; + return acpi_battery_check(battery); +} + static int __init acpi_battery_init(void) { int result; -- cgit v0.10.2 From 69f0304e174c765c624d75b79c35e49b7ba67ed4 Mon Sep 17 00:00:00 2001 From: Liam Girdwood Date: Thu, 5 Oct 2006 13:22:42 +0100 Subject: [ARM] 3888/1: add pxa27x SSP FSRT register bit definition This patch adds a register bit definition for the pxa27x SSP port Frame Sync Relative Timing (FSRT) bit. Signed-off-by: Liam Girdwood Signed-off-by: Russell King diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h index f5cc65d..68731e0 100644 --- a/include/asm-arm/arch-pxa/pxa-regs.h +++ b/include/asm-arm/arch-pxa/pxa-regs.h @@ -1681,6 +1681,7 @@ #define SSSR_TINT (1 << 19) /* Receiver Time-out Interrupt */ #define SSSR_PINT (1 << 18) /* Peripheral Trailing Byte Interrupt */ +#define SSPSP_FSRT (1 << 25) /* Frame Sync Relative Timing */ #define SSPSP_DMYSTOP(x) (x << 23) /* Dummy Stop */ #define SSPSP_SFRMWDTH(x) (x << 16) /* Serial Frame Width */ #define SSPSP_SFRMDLY(x) (x << 9) /* Serial Frame Delay */ -- cgit v0.10.2 From 48e3becbee8906b44050c9300b668227b85f0c95 Mon Sep 17 00:00:00 2001 From: Kristoffer Ericson Date: Sun, 8 Oct 2006 19:52:30 +0100 Subject: [ARM] 3889/1: [Jornada7xx] Addition of correct SDRAM params into cpu-sa1110.c This adds correct sdram params for K4S281632B-1H and sets the jornada to use them by default. Signed-off-by: Kristoffer Ericson Signed-off-by: Russell King diff --git a/arch/arm/mach-sa1100/cpu-sa1110.c b/arch/arm/mach-sa1100/cpu-sa1110.c index 6395977..90a4130 100644 --- a/arch/arm/mach-sa1100/cpu-sa1110.c +++ b/arch/arm/mach-sa1100/cpu-sa1110.c @@ -82,6 +82,14 @@ static struct sdram_params sdram_tbl[] __initdata = { .twr = 9, .refresh = 64000, .cas_latency = 3, + }, { /* Samsung K4S281632B-1H */ + .name = "K4S281632b-1H", + .rows = 12, + .tck = 10, + .trp = 20, + .twr = 10, + .refresh = 64000, + .cas_latency = 3, }, { /* Samsung KM416S4030CT */ .name = "KM416S4030CT", .rows = 13, @@ -366,6 +374,8 @@ static int __init sa1110_clk_init(void) if (machine_is_h3100()) name = "KM416S4030CT"; + if (machine_is_jornada720()) + name = "K4S281632B-1H"; } sdram = sa1110_find_sdram(name); -- cgit v0.10.2 From 8e25b84e76a0b2c117218405818cadd591512ff8 Mon Sep 17 00:00:00 2001 From: Kristoffer Ericson Date: Sun, 8 Oct 2006 20:00:53 +0100 Subject: [ARM] 3890/1: [Jornada7xx] Addition of MCU commands into jornada720.h This adds relevant MCU commands for the j7xx chipset. Signed-off-by: Kristoffer Ericson Signed-off-by: Russell King diff --git a/include/asm-arm/arch-sa1100/jornada720.h b/include/asm-arm/arch-sa1100/jornada720.h index 1b8e8a3..3f37ca0 100644 --- a/include/asm-arm/arch-sa1100/jornada720.h +++ b/include/asm-arm/arch-sa1100/jornada720.h @@ -19,6 +19,20 @@ #define GPIO_JORNADA720_KEYBOARD_IRQ IRQ_GPIO0 #define GPIO_JORNADA720_MOUSE_IRQ IRQ_GPIO9 +/* MCU COMMANDS */ +#define MCU_GetBatteryData 0xc0 +#define MCU_GetScanKeyCode 0x90 +#define MCU_GetTouchSamples 0xa0 +#define MCU_GetContrast 0xD0 +#define MCU_SetContrast 0xD1 +#define MCU_GetBrightness 0xD2 +#define MCU_SetBrightness 0xD3 +#define MCU_ContrastOff 0xD8 +#define MCU_BrightnessOff 0xD9 +#define MCU_PWMOFF 0xDF +#define MCU_TxDummy 0x11 +#define MCU_ErrorCode 0x00 + #ifndef __ASSEMBLY__ void jornada720_mcu_init(void); -- cgit v0.10.2 From cbf40d3f04c2c76a58f1183bb4a9a82fefb842e3 Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Sat, 14 Oct 2006 20:18:47 +0200 Subject: [WATCHDOG] remove experimental on iTCO_wdt.c The iTCO_wdt.c driver has been tested enough. So we can remove the experimental classification. Signed-off-by: Wim Van Sebroeck diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig index 529f0a7..0187b11 100644 --- a/drivers/char/watchdog/Kconfig +++ b/drivers/char/watchdog/Kconfig @@ -316,13 +316,16 @@ config I8XX_TCO To compile this driver as a module, choose M here: the module will be called i8xx_tco. + Note: This driver will be removed in the near future. Please + use the Intel TCO Timer/Watchdog driver. + config ITCO_WDT - tristate "Intel TCO Timer/Watchdog (EXPERIMENTAL)" - depends on WATCHDOG && (X86 || IA64) && PCI && EXPERIMENTAL + tristate "Intel TCO Timer/Watchdog" + depends on WATCHDOG && (X86 || IA64) && PCI ---help--- Hardware driver for the intel TCO timer based watchdog devices. These drivers are included in the Intel 82801 I/O Controller - Hub family 'from ICH0 up to ICH7) and in the Intel 6300ESB + Hub family (from ICH0 up to ICH8) and in the Intel 6300ESB controller hub. The TCO (Total Cost of Ownership) timer is a watchdog timer @@ -590,7 +593,7 @@ config SH_WDT_MMAP help If you say Y here, user applications will be able to mmap the WDT/CPG registers. -# + # SPARC64 Architecture config WATCHDOG_CP1XXX -- cgit v0.10.2 From 2326eb985b8844f44e150489c76f5cb56fa381b4 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 15 Oct 2006 13:48:37 +0100 Subject: [ARM] Fix fallout from IRQ regs changes Some ARM platforms were still broken as a result of the IRQ register passing changes, mostly due to a missing linux/irq.h include. Signed-off-by: Russell King diff --git a/arch/arm/mach-footbridge/dc21285.c b/arch/arm/mach-footbridge/dc21285.c index fa5d497..1463330 100644 --- a/arch/arm/mach-footbridge/dc21285.c +++ b/arch/arm/mach-footbridge/dc21285.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c index ee80d62..142c33c 100644 --- a/arch/arm/mach-pxa/lubbock.c +++ b/arch/arm/mach-pxa/lubbock.c @@ -397,7 +397,7 @@ static void lubbock_mmc_poll(unsigned long data) if (LUB_IRQ_SET_CLR & (1 << 0)) mod_timer(&mmc_timer, jiffies + MMC_POLL_RATE); else { - (void) mmc_detect_int(LUBBOCK_SD_IRQ, (void *)data, NULL); + (void) mmc_detect_int(LUBBOCK_SD_IRQ, (void *)data); enable_irq(LUBBOCK_SD_IRQ); } } diff --git a/arch/arm/oprofile/op_model_xscale.c b/arch/arm/oprofile/op_model_xscale.c index 7899d3c..7c3289c 100644 --- a/arch/arm/oprofile/op_model_xscale.c +++ b/arch/arm/oprofile/op_model_xscale.c @@ -20,7 +20,8 @@ #include #include #include -#include +#include + #include #include "op_counter.h" -- cgit v0.10.2 From 36bd262b3f2ac723dadd20ce35539c8c738877f1 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 15 Oct 2006 13:50:02 +0100 Subject: [ARM] Fix Zaurii keyboard/touchscreen drivers The Zaurii drivers were partially fixed up for the IRQ register changes, but unfortunately missed some bits, resulting in build errors. Fix these. Signed-off-by: Russell King diff --git a/drivers/input/keyboard/corgikbd.c b/drivers/input/keyboard/corgikbd.c index cb70970..befdd60 100644 --- a/drivers/input/keyboard/corgikbd.c +++ b/drivers/input/keyboard/corgikbd.c @@ -207,7 +207,7 @@ static irqreturn_t corgikbd_interrupt(int irq, void *dev_id) static void corgikbd_timer_callback(unsigned long data) { struct corgikbd *corgikbd_data = (struct corgikbd *) data; - corgikbd_scankeyboard(corgikbd_data, NULL); + corgikbd_scankeyboard(corgikbd_data); } /* diff --git a/drivers/input/keyboard/locomokbd.c b/drivers/input/keyboard/locomokbd.c index fd33c9c..5788dbc 100644 --- a/drivers/input/keyboard/locomokbd.c +++ b/drivers/input/keyboard/locomokbd.c @@ -186,7 +186,7 @@ static irqreturn_t locomokbd_interrupt(int irq, void *dev_id) static void locomokbd_timer_callback(unsigned long data) { struct locomokbd *locomokbd = (struct locomokbd *) data; - locomokbd_scankeyboard(locomokbd, NULL); + locomokbd_scankeyboard(locomokbd); } static int locomokbd_probe(struct locomo_dev *dev) diff --git a/drivers/input/keyboard/spitzkbd.c b/drivers/input/keyboard/spitzkbd.c index 8b18c00..28b2748 100644 --- a/drivers/input/keyboard/spitzkbd.c +++ b/drivers/input/keyboard/spitzkbd.c @@ -257,7 +257,7 @@ static void spitzkbd_timer_callback(unsigned long data) { struct spitzkbd *spitzkbd_data = (struct spitzkbd *) data; - spitzkbd_scankeyboard(spitzkbd_data, NULL); + spitzkbd_scankeyboard(spitzkbd_data); } /* diff --git a/drivers/input/touchscreen/corgi_ts.c b/drivers/input/touchscreen/corgi_ts.c index ca79b22..66121f6 100644 --- a/drivers/input/touchscreen/corgi_ts.c +++ b/drivers/input/touchscreen/corgi_ts.c @@ -219,7 +219,7 @@ static void ts_interrupt_main(struct corgi_ts *corgi_ts, int isTimer) static void corgi_ts_timer(unsigned long data) { struct corgi_ts *corgits_data = (struct corgi_ts *) data; - ts_interrupt_main(corgits_data, 1, NULL); + ts_interrupt_main(corgits_data, 1); } static irqreturn_t ts_interrupt(int irq, void *dev_id) @@ -237,7 +237,7 @@ static int corgits_suspend(struct platform_device *dev, pm_message_t state) if (corgi_ts->pendown) { del_timer_sync(&corgi_ts->timer); corgi_ts->tc.pressure = 0; - new_data(corgi_ts, NULL); + new_data(corgi_ts); corgi_ts->pendown = 0; } corgi_ts->power_mode = PWR_MODE_SUSPEND; -- cgit v0.10.2 From 645408d1ffe9f27b176a88302c22420f301607db Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 14 Oct 2006 16:50:38 +0100 Subject: [PATCH] gfp_t in netlabel Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/include/net/netlabel.h b/include/net/netlabel.h index 113337c..12c214b 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -136,7 +136,7 @@ struct netlbl_lsm_secattr { * on success, NULL on failure. * */ -static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc(int flags) +static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc(gfp_t flags) { struct netlbl_lsm_cache *cache; -- cgit v0.10.2 From 733f99acc82543030ce0417e2f2201ddc63097af Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 14 Oct 2006 16:48:26 +0100 Subject: [PATCH] new cifs endianness bugs * missing cpu_to_le64() for ChangeTime (introduced by [CIFS] Legacy time handling for Win9x and OS/2 part 1) * missing le16_to_cpu() for DialectIndex (introduced by [CIFS] Do not send newer QFSInfo to legacy servers which can not support it) Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 5dc5a96..098790e 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -399,6 +399,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) struct TCP_Server_Info * server; u16 count; unsigned int secFlags; + u16 dialect; if(ses->server) server = ses->server; @@ -438,9 +439,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) if (rc != 0) goto neg_err_exit; - cFYI(1,("Dialect: %d", pSMBr->DialectIndex)); + dialect = le16_to_cpu(pSMBr->DialectIndex); + cFYI(1,("Dialect: %d", dialect)); /* Check wct = 1 error case */ - if((pSMBr->hdr.WordCount < 13) || (pSMBr->DialectIndex == BAD_PROT)) { + if((pSMBr->hdr.WordCount < 13) || (dialect == BAD_PROT)) { /* core returns wct = 1, but we do not ask for core - otherwise small wct just comes when dialect index is -1 indicating we could not negotiate a common dialect */ @@ -448,8 +450,8 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) goto neg_err_exit; #ifdef CONFIG_CIFS_WEAK_PW_HASH } else if((pSMBr->hdr.WordCount == 13) - && ((pSMBr->DialectIndex == LANMAN_PROT) - || (pSMBr->DialectIndex == LANMAN2_PROT))) { + && ((dialect == LANMAN_PROT) + || (dialect == LANMAN2_PROT))) { __s16 tmp; struct lanman_neg_rsp * rsp = (struct lanman_neg_rsp *)pSMBr; @@ -2943,7 +2945,7 @@ QInfRetry: ts.tv_nsec = 0; ts.tv_sec = time; /* decode time fields */ - pFinfo->ChangeTime = cifs_UnixTimeToNT(ts); + pFinfo->ChangeTime = cpu_to_le64(cifs_UnixTimeToNT(ts)); pFinfo->LastWriteTime = pFinfo->ChangeTime; pFinfo->LastAccessTime = 0; pFinfo->AllocationSize = -- cgit v0.10.2 From 6ce6b3aeeae75eee34670bcd42870ac839bfec4c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 14 Oct 2006 16:52:36 +0100 Subject: [PATCH] hp drivers/input stuff: C99 initializers, NULL noise removal, __user annotations Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/input/misc/hp_sdc_rtc.c b/drivers/input/misc/hp_sdc_rtc.c index 1be9639..ab4da79 100644 --- a/drivers/input/misc/hp_sdc_rtc.c +++ b/drivers/input/misc/hp_sdc_rtc.c @@ -60,7 +60,7 @@ static struct fasync_struct *hp_sdc_rtc_async_queue; static DECLARE_WAIT_QUEUE_HEAD(hp_sdc_rtc_wait); -static ssize_t hp_sdc_rtc_read(struct file *file, char *buf, +static ssize_t hp_sdc_rtc_read(struct file *file, char __user *buf, size_t count, loff_t *ppos); static int hp_sdc_rtc_ioctl(struct inode *inode, struct file *file, @@ -385,14 +385,14 @@ static int hp_sdc_rtc_set_i8042timer (struct timeval *setto, uint8_t setcmd) return 0; } -static ssize_t hp_sdc_rtc_read(struct file *file, char *buf, +static ssize_t hp_sdc_rtc_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { ssize_t retval; if (count < sizeof(unsigned long)) return -EINVAL; - retval = put_user(68, (unsigned long *)buf); + retval = put_user(68, (unsigned long __user *)buf); return retval; } @@ -696,7 +696,7 @@ static int __init hp_sdc_rtc_init(void) if ((ret = hp_sdc_request_timer_irq(&hp_sdc_rtc_isr))) return ret; misc_register(&hp_sdc_rtc_dev); - create_proc_read_entry ("driver/rtc", 0, 0, + create_proc_read_entry ("driver/rtc", 0, NULL, hp_sdc_rtc_read_proc, NULL); printk(KERN_INFO "HP i8042 SDC + MSM-58321 RTC support loaded " diff --git a/drivers/input/serio/hil_mlc.c b/drivers/input/serio/hil_mlc.c index bdfde04..49e11e2 100644 --- a/drivers/input/serio/hil_mlc.c +++ b/drivers/input/serio/hil_mlc.c @@ -391,23 +391,23 @@ static int hilse_operate(hil_mlc *mlc, int repoll) { } #define FUNC(funct, funct_arg, zero_rc, neg_rc, pos_rc) \ -{ HILSE_FUNC, { func: &funct }, funct_arg, zero_rc, neg_rc, pos_rc }, +{ HILSE_FUNC, { .func = funct }, funct_arg, zero_rc, neg_rc, pos_rc }, #define OUT(pack) \ -{ HILSE_OUT, { packet: pack }, 0, HILSEN_NEXT, HILSEN_DOZE, 0 }, +{ HILSE_OUT, { .packet = pack }, 0, HILSEN_NEXT, HILSEN_DOZE, 0 }, #define CTS \ -{ HILSE_CTS, { packet: 0 }, 0, HILSEN_NEXT | HILSEN_SCHED | HILSEN_BREAK, HILSEN_DOZE, 0 }, +{ HILSE_CTS, { .packet = 0 }, 0, HILSEN_NEXT | HILSEN_SCHED | HILSEN_BREAK, HILSEN_DOZE, 0 }, #define EXPECT(comp, to, got, got_wrong, timed_out) \ -{ HILSE_EXPECT, { packet: comp }, to, got, got_wrong, timed_out }, +{ HILSE_EXPECT, { .packet = comp }, to, got, got_wrong, timed_out }, #define EXPECT_LAST(comp, to, got, got_wrong, timed_out) \ -{ HILSE_EXPECT_LAST, { packet: comp }, to, got, got_wrong, timed_out }, +{ HILSE_EXPECT_LAST, { .packet = comp }, to, got, got_wrong, timed_out }, #define EXPECT_DISC(comp, to, got, got_wrong, timed_out) \ -{ HILSE_EXPECT_DISC, { packet: comp }, to, got, got_wrong, timed_out }, +{ HILSE_EXPECT_DISC, { .packet = comp }, to, got, got_wrong, timed_out }, #define IN(to, got, got_error, timed_out) \ -{ HILSE_IN, { packet: 0 }, to, got, got_error, timed_out }, +{ HILSE_IN, { .packet = 0 }, to, got, got_error, timed_out }, #define OUT_DISC(pack) \ -{ HILSE_OUT_DISC, { packet: pack }, 0, 0, 0, 0 }, +{ HILSE_OUT_DISC, { .packet = pack }, 0, 0, 0, 0 }, #define OUT_LAST(pack) \ -{ HILSE_OUT_LAST, { packet: pack }, 0, 0, 0, 0 }, +{ HILSE_OUT_LAST, { .packet = pack }, 0, 0, 0, 0 }, struct hilse_node hil_mlc_se[HILSEN_END] = { diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c index ba7b920..9907ad3 100644 --- a/drivers/input/serio/hp_sdc.c +++ b/drivers/input/serio/hp_sdc.c @@ -310,7 +310,7 @@ static void hp_sdc_tasklet(unsigned long foo) { * in tasklet/bh context. */ if (curr->act.irqhook) - curr->act.irqhook(0, 0, 0, 0); + curr->act.irqhook(0, NULL, 0, 0); } curr->actidx = curr->idx; curr->idx++; @@ -525,7 +525,7 @@ actdone: up(curr->act.semaphore); } else if (act & HP_SDC_ACT_CALLBACK) { - curr->act.irqhook(0,0,0,0); + curr->act.irqhook(0,NULL,0,0); } if (curr->idx >= curr->endidx) { /* This transaction is over. */ if (act & HP_SDC_ACT_DEALLOC) kfree(curr); -- cgit v0.10.2 From cbff67668d597da48f8bc48549a9630cbf968f34 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 14 Oct 2006 16:53:38 +0100 Subject: [PATCH] sun3_ioremap() prototype Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/include/asm-m68k/sun3mmu.h b/include/asm-m68k/sun3mmu.h index 6c8c17d..d8f17a0 100644 --- a/include/asm-m68k/sun3mmu.h +++ b/include/asm-m68k/sun3mmu.h @@ -4,6 +4,7 @@ #ifndef __SUN3_MMU_H__ #define __SUN3_MMU_H__ +#include #include #include @@ -160,7 +161,7 @@ static inline void sun3_put_context(unsigned char c) return; } -extern void *sun3_ioremap(unsigned long phys, unsigned long size, +extern void __iomem *sun3_ioremap(unsigned long phys, unsigned long size, unsigned long type); extern int sun3_map_test(unsigned long addr, char *val); -- cgit v0.10.2 From e5a301ee02e53acf000bb8331587129930bc2290 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 14 Oct 2006 16:51:49 +0100 Subject: [PATCH] serial167 __user annotations, NULL noise removal Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/char/serial167.c b/drivers/char/serial167.c index 461bfe0..3af7f09 100644 --- a/drivers/char/serial167.c +++ b/drivers/char/serial167.c @@ -839,7 +839,7 @@ shutdown(struct cyclades_port * info) local_irq_save(flags); if (info->xmit_buf){ free_page((unsigned long) info->xmit_buf); - info->xmit_buf = 0; + info->xmit_buf = NULL; } base_addr[CyCAR] = (u_char)channel; @@ -1354,7 +1354,7 @@ cy_unthrottle(struct tty_struct * tty) static int get_serial_info(struct cyclades_port * info, - struct serial_struct * retinfo) + struct serial_struct __user * retinfo) { struct serial_struct tmp; @@ -1376,7 +1376,7 @@ get_serial_info(struct cyclades_port * info, static int set_serial_info(struct cyclades_port * info, - struct serial_struct * new_info) + struct serial_struct __user * new_info) { struct serial_struct new_serial; struct cyclades_port old_info; @@ -1503,7 +1503,7 @@ send_break( struct cyclades_port * info, int duration) } /* send_break */ static int -get_mon_info(struct cyclades_port * info, struct cyclades_monitor * mon) +get_mon_info(struct cyclades_port * info, struct cyclades_monitor __user * mon) { if (copy_to_user(mon, &info->mon, sizeof(struct cyclades_monitor))) @@ -1516,7 +1516,7 @@ get_mon_info(struct cyclades_port * info, struct cyclades_monitor * mon) } static int -set_threshold(struct cyclades_port * info, unsigned long *arg) +set_threshold(struct cyclades_port * info, unsigned long __user *arg) { volatile unsigned char *base_addr = (u_char *)BASE_ADDR; unsigned long value; @@ -1533,7 +1533,7 @@ set_threshold(struct cyclades_port * info, unsigned long *arg) } static int -get_threshold(struct cyclades_port * info, unsigned long *value) +get_threshold(struct cyclades_port * info, unsigned long __user *value) { volatile unsigned char *base_addr = (u_char *)BASE_ADDR; int channel; @@ -1546,7 +1546,7 @@ get_threshold(struct cyclades_port * info, unsigned long *value) } static int -set_default_threshold(struct cyclades_port * info, unsigned long *arg) +set_default_threshold(struct cyclades_port * info, unsigned long __user *arg) { unsigned long value; @@ -1558,13 +1558,13 @@ set_default_threshold(struct cyclades_port * info, unsigned long *arg) } static int -get_default_threshold(struct cyclades_port * info, unsigned long *value) +get_default_threshold(struct cyclades_port * info, unsigned long __user *value) { return put_user(info->default_threshold,value); } static int -set_timeout(struct cyclades_port * info, unsigned long *arg) +set_timeout(struct cyclades_port * info, unsigned long __user *arg) { volatile unsigned char *base_addr = (u_char *)BASE_ADDR; int channel; @@ -1581,7 +1581,7 @@ set_timeout(struct cyclades_port * info, unsigned long *arg) } static int -get_timeout(struct cyclades_port * info, unsigned long *value) +get_timeout(struct cyclades_port * info, unsigned long __user *value) { volatile unsigned char *base_addr = (u_char *)BASE_ADDR; int channel; @@ -1601,7 +1601,7 @@ set_default_timeout(struct cyclades_port * info, unsigned long value) } static int -get_default_timeout(struct cyclades_port * info, unsigned long *value) +get_default_timeout(struct cyclades_port * info, unsigned long __user *value) { return put_user(info->default_timeout,value); } @@ -1613,6 +1613,7 @@ cy_ioctl(struct tty_struct *tty, struct file * file, unsigned long val; struct cyclades_port * info = (struct cyclades_port *)tty->driver_data; int ret_val = 0; + void __user *argp = (void __user *)arg; #ifdef SERIAL_DEBUG_OTHER printk("cy_ioctl %s, cmd = %x arg = %lx\n", tty->name, cmd, arg); /* */ @@ -1620,28 +1621,28 @@ cy_ioctl(struct tty_struct *tty, struct file * file, switch (cmd) { case CYGETMON: - ret_val = get_mon_info(info, (struct cyclades_monitor *)arg); + ret_val = get_mon_info(info, argp); break; case CYGETTHRESH: - ret_val = get_threshold(info, (unsigned long *)arg); + ret_val = get_threshold(info, argp); break; case CYSETTHRESH: - ret_val = set_threshold(info, (unsigned long *)arg); + ret_val = set_threshold(info, argp); break; case CYGETDEFTHRESH: - ret_val = get_default_threshold(info, (unsigned long *)arg); + ret_val = get_default_threshold(info, argp); break; case CYSETDEFTHRESH: - ret_val = set_default_threshold(info, (unsigned long *)arg); + ret_val = set_default_threshold(info, argp); break; case CYGETTIMEOUT: - ret_val = get_timeout(info, (unsigned long *)arg); + ret_val = get_timeout(info, argp); break; case CYSETTIMEOUT: - ret_val = set_timeout(info, (unsigned long *)arg); + ret_val = set_timeout(info, argp); break; case CYGETDEFTIMEOUT: - ret_val = get_default_timeout(info, (unsigned long *)arg); + ret_val = get_default_timeout(info, argp); break; case CYSETDEFTIMEOUT: ret_val = set_default_timeout(info, (unsigned long)arg); @@ -1664,21 +1665,20 @@ cy_ioctl(struct tty_struct *tty, struct file * file, /* The following commands are incompletely implemented!!! */ case TIOCGSOFTCAR: - ret_val = put_user(C_CLOCAL(tty) ? 1 : 0, (unsigned long *) arg); + ret_val = put_user(C_CLOCAL(tty) ? 1 : 0, (unsigned long __user *) argp); break; case TIOCSSOFTCAR: - ret_val = get_user(val, (unsigned long *) arg); + ret_val = get_user(val, (unsigned long __user *) argp); if (ret_val) break; tty->termios->c_cflag = ((tty->termios->c_cflag & ~CLOCAL) | (val ? CLOCAL : 0)); break; case TIOCGSERIAL: - ret_val = get_serial_info(info, (struct serial_struct *) arg); + ret_val = get_serial_info(info, argp); break; case TIOCSSERIAL: - ret_val = set_serial_info(info, - (struct serial_struct *) arg); + ret_val = set_serial_info(info, argp); break; default: ret_val = -ENOIOCTLCMD; @@ -1773,7 +1773,7 @@ cy_close(struct tty_struct * tty, struct file * filp) tty->driver->flush_buffer(tty); tty_ldisc_flush(tty); info->event = 0; - info->tty = 0; + info->tty = NULL; if (info->blocked_open) { if (info->close_delay) { msleep_interruptible(jiffies_to_msecs(info->close_delay)); @@ -2250,7 +2250,7 @@ scrn[1] = '\0'; info->card = index; info->line = port_num; info->flags = STD_COM_FLAGS; - info->tty = 0; + info->tty = NULL; info->xmit_fifo_size = 12; info->cor1 = CyPARITY_NONE|Cy_8_BITS; info->cor2 = CyETC; -- cgit v0.10.2 From a7a0d86f5aa40a2215e36fe21d7911cf718ba428 Mon Sep 17 00:00:00 2001 From: Petr Vandrovec Date: Fri, 13 Oct 2006 18:42:07 +0200 Subject: [PATCH] Fix core files so they make sense to gdb... It is silly to use non-static variable for writting zeroes to the file. And more seriously, foffset in core dump file dump function was incremented too much, so some parts of core dump were shifted by size of few phdrs and notes down, so although gdb was able to load that file, it did not make lot of sense - in my test case data pages were shifted down by about 900 bytes. Signed-off-by: Petr Vandrovec Signed-off-by: Linus Torvalds diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 5610061..79b05a1 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1220,7 +1220,7 @@ static int notesize(struct memelfnote *en) static int alignfile(struct file *file, loff_t *foffset) { - char buf[4] = { 0, }; + static const char buf[4] = { 0, }; DUMP_WRITE(buf, roundup(*foffset, 4) - *foffset, foffset); return 1; } @@ -1569,7 +1569,8 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) DUMP_WRITE(elf, sizeof(*elf)); offset += sizeof(*elf); /* Elf header */ - offset += (segs+1) * sizeof(struct elf_phdr); /* Program headers */ + offset += (segs + 1) * sizeof(struct elf_phdr); /* Program headers */ + foffset = offset; /* Write notes phdr entry */ { @@ -1586,8 +1587,6 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) DUMP_WRITE(&phdr, sizeof(phdr)); } - foffset = offset; - dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); /* Write program headers for segments dump */ @@ -1612,7 +1611,6 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) phdr.p_align = ELF_EXEC_PAGESIZE; DUMP_WRITE(&phdr, sizeof(phdr)); - foffset += sizeof(phdr); } #ifdef ELF_CORE_WRITE_EXTRA_PHDRS -- cgit v0.10.2 From 80c5606c3b45e0176c32d3108ade1e1cb0b954f3 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 15 Oct 2006 14:09:55 -0700 Subject: Fix VM_MAYEXEC calculation .. and clean up the file mapping code while at it. No point in having a "if (file)" repeated twice, and generally doing similar checks in two different sections of the same code Signed-off-by: Linus Torvalds diff --git a/mm/mmap.c b/mm/mmap.c index eea8eef..497e502 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -900,17 +900,6 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr, int accountable = 1; unsigned long charged = 0, reqprot = prot; - if (file) { - if (is_file_hugepages(file)) - accountable = 0; - - if (!file->f_op || !file->f_op->mmap) - return -ENODEV; - - if ((prot & PROT_EXEC) && - (file->f_vfsmnt->mnt_flags & MNT_NOEXEC)) - return -EPERM; - } /* * Does the application expect PROT_READ to imply PROT_EXEC? * @@ -1000,6 +989,16 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr, case MAP_PRIVATE: if (!(file->f_mode & FMODE_READ)) return -EACCES; + if (file->f_vfsmnt->mnt_flags & MNT_NOEXEC) { + if (vm_flags & VM_EXEC) + return -EPERM; + vm_flags &= ~VM_MAYEXEC; + } + if (is_file_hugepages(file)) + accountable = 0; + + if (!file->f_op || !file->f_op->mmap) + return -ENODEV; break; default: -- cgit v0.10.2 From 51018b0a3160d253283173c2f54f16746cee5852 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Sun, 15 Oct 2006 15:03:48 -0400 Subject: [PATCH] make UML compile (FC6/x86-64) I need this patch to get a UML kernel to compile. This is with the kernel headers in FC6 which are automatically generated from the kernel tree. Some headers are missing but those files don't need them. At least it appears so since the resuling kernel works fine. Tested on x86-64. Signed-off-by: Ulrich Drepper Signed-off-by: Linus Torvalds diff --git a/arch/um/include/kern_util.h b/arch/um/include/kern_util.h index 59cfa9e..cec9fcc 100644 --- a/arch/um/include/kern_util.h +++ b/arch/um/include/kern_util.h @@ -6,7 +6,6 @@ #ifndef __KERN_UTIL_H__ #define __KERN_UTIL_H__ -#include "linux/threads.h" #include "sysdep/ptrace.h" #include "sysdep/faultinfo.h" diff --git a/arch/um/sys-x86_64/stub_segv.c b/arch/um/sys-x86_64/stub_segv.c index 1c96702..652fa34 100644 --- a/arch/um/sys-x86_64/stub_segv.c +++ b/arch/um/sys-x86_64/stub_segv.c @@ -5,7 +5,6 @@ #include #include -#include #include #include "uml-config.h" #include "sysdep/sigcontext.h" -- cgit v0.10.2 From 18088748d2a493ce9f6adf0be7f833b04041807e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=E4nzer?= Date: Wed, 4 Oct 2006 14:56:44 +0200 Subject: [AGPGART] uninorth: Add module param 'aperture' for aperture size MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In contrast to most if not all PC BIOSes, OpenFirmware (OF) on PowerMacs with UniNorth bridges does not allow changing the aperture size. The size set up by OF is usually 16 MB, which is too low for graphics intensive environments. Hence, add a module parameter that allows changing the aperture size at driver initialization time. When the parameter is not specified, the default is 32 MB. Signed-off-by: Michel Dänzer Acked-by: Benjamin Herrenschmidt Signed-off-by: Dave Jones diff --git a/drivers/char/agp/uninorth-agp.c b/drivers/char/agp/uninorth-agp.c index 91b71e7..dffc193 100644 --- a/drivers/char/agp/uninorth-agp.c +++ b/drivers/char/agp/uninorth-agp.c @@ -27,32 +27,42 @@ static int uninorth_rev; static int is_u3; +static char __devinitdata *aperture = NULL; static int uninorth_fetch_size(void) { - int i; - u32 temp; - struct aper_size_info_32 *values; - - pci_read_config_dword(agp_bridge->dev, UNI_N_CFG_GART_BASE, &temp); - temp &= ~(0xfffff000); - values = A_SIZE_32(agp_bridge->driver->aperture_sizes); - - for (i = 0; i < agp_bridge->driver->num_aperture_sizes; i++) { - if (temp == values[i].size_value) { - agp_bridge->previous_size = - agp_bridge->current_size = (void *) (values + i); - agp_bridge->aperture_size_idx = i; - return values[i].size; + int i, size = 0; + struct aper_size_info_32 *values = + A_SIZE_32(agp_bridge->driver->aperture_sizes); + + if (aperture) { + char *save = aperture; + + size = memparse(aperture, &aperture) >> 20; + aperture = save; + + for (i = 0; i < agp_bridge->driver->num_aperture_sizes; i++) + if (size == values[i].size) + break; + + if (i == agp_bridge->driver->num_aperture_sizes) { + printk(KERN_ERR PFX "Invalid aperture size, using" + " default\n"); + size = 0; + aperture = NULL; } } - agp_bridge->previous_size = - agp_bridge->current_size = (void *) (values + 1); - agp_bridge->aperture_size_idx = 1; - return values[1].size; + if (!size) { + for (i = 0; i < agp_bridge->driver->num_aperture_sizes; i++) + if (values[i].size == 32) + break; + } - return 0; + agp_bridge->previous_size = + agp_bridge->current_size = (void *)(values + i); + agp_bridge->aperture_size_idx = i; + return values[i].size; } static void uninorth_tlbflush(struct agp_memory *mem) @@ -683,5 +693,11 @@ static void __exit agp_uninorth_cleanup(void) module_init(agp_uninorth_init); module_exit(agp_uninorth_cleanup); +module_param(aperture, charp, 0); +MODULE_PARM_DESC(aperture, + "Aperture size, must be power of two between 4MB and an\n" + "\t\tupper limit specific to the UniNorth revision.\n" + "\t\tDefault: 32M"); + MODULE_AUTHOR("Ben Herrenschmidt & Paul Mackerras"); MODULE_LICENSE("GPL"); -- cgit v0.10.2 From e5267b4b376cfbdc4518abcc68d5a7fffb505638 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 10 Oct 2006 15:14:12 +1000 Subject: [POWERPC] Don't crash on cell with 2 BEs when !CONFIG_NUMA The SPU code will crash if CONFIG_NUMA is not set and SPUs are found on a non-0 node. This workaround will ignore those SPEs and just print an message in the kernel log. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c index ccfd0c4..d0fb959 100644 --- a/arch/powerpc/platforms/cell/spu_base.c +++ b/arch/powerpc/platforms/cell/spu_base.c @@ -781,6 +781,17 @@ static int __init create_spu(struct device_node *spe) if (!spu) goto out; + spu->node = find_spu_node_id(spe); + if (spu->node >= MAX_NUMNODES) { + printk(KERN_WARNING "SPE %s on node %d ignored," + " node number too big\n", spe->full_name, spu->node); + printk(KERN_WARNING "Check if CONFIG_NUMA is enabled.\n"); + return -ENODEV; + } + spu->nid = of_node_to_nid(spe); + if (spu->nid == -1) + spu->nid = 0; + ret = spu_map_device(spu, spe); /* try old method */ if (ret) @@ -788,10 +799,6 @@ static int __init create_spu(struct device_node *spe) if (ret) goto out_free; - spu->node = find_spu_node_id(spe); - spu->nid = of_node_to_nid(spe); - if (spu->nid == -1) - spu->nid = 0; ret = spu_map_interrupts(spu, spe); if (ret) ret = spu_map_interrupts_old(spu, spe); -- cgit v0.10.2 From 654e4aee495bec1e4fc71ba1af25735da7cadc15 Mon Sep 17 00:00:00 2001 From: "Noguchi, Masato" Date: Tue, 10 Oct 2006 10:27:29 +0200 Subject: [POWERPC] spufs: fix support for read/write on cntl This fixes a memory leak introduced by "spufs: add support for read/write oncntl", which was missing a call to simple_attr_close. Signed-off-by: Masato Noguchi Signed-off-by: Arnd Bergmann Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c index e0d7300..0de8e11 100644 --- a/arch/powerpc/platforms/cell/spufs/file.c +++ b/arch/powerpc/platforms/cell/spufs/file.c @@ -246,6 +246,7 @@ static int spufs_cntl_open(struct inode *inode, struct file *file) static struct file_operations spufs_cntl_fops = { .open = spufs_cntl_open, + .release = simple_attr_close, .read = simple_attr_read, .write = simple_attr_write, .mmap = spufs_cntl_mmap, -- cgit v0.10.2 From ee4ea82c07fa8aa46037962ec3e8b40c5144b5eb Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Tue, 10 Oct 2006 11:53:26 -0500 Subject: [POWERPC] Add DOS partition table support to mpc834x_itx_defconfig The default configuration file for the MPC8349E-mITX reference board, mpc834x_itx_defconfig, did not include support for DOS partition table types. This support is necessary because the hard drive that comes with the ITX is formatted with this partition table type. Without this config option, no partitions on the drive can be mounted. Signed-off-by: Timur Tabi Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/configs/mpc834x_itx_defconfig b/arch/powerpc/configs/mpc834x_itx_defconfig index cd3535e..0561b73 100644 --- a/arch/powerpc/configs/mpc834x_itx_defconfig +++ b/arch/powerpc/configs/mpc834x_itx_defconfig @@ -1248,7 +1248,7 @@ CONFIG_PARTITION_ADVANCED=y # CONFIG_AMIGA_PARTITION is not set # CONFIG_ATARI_PARTITION is not set # CONFIG_MAC_PARTITION is not set -# CONFIG_MSDOS_PARTITION is not set +CONFIG_MSDOS_PARTITION=y # CONFIG_LDM_PARTITION is not set # CONFIG_SGI_PARTITION is not set # CONFIG_ULTRIX_PARTITION is not set -- cgit v0.10.2 From bb579cf1d413b930be0241987b848e0f0c1b292f Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Tue, 10 Oct 2006 14:37:29 -0700 Subject: [POWERPC] Off-by-one in /arch/ppc/platforms/mpc8* A find -iname \*.[ch] | xargs grep "> ARRAY_SIZE(" revealed several incorrect usages of ARRAY_SIZE in the mpc drivers. The last element in the array is always ARRAY_SIZE()-1, this patch modifies the bounds checks accordingly. Signed-off-by: Eric Sesterhenn Cc: Vitaly Bordug Signed-off-by: Andrew Morton Signed-off-by: Paul Mackerras diff --git a/arch/ppc/platforms/mpc8272ads_setup.c b/arch/ppc/platforms/mpc8272ads_setup.c index d7b3a6a..1f9ea368 100644 --- a/arch/ppc/platforms/mpc8272ads_setup.c +++ b/arch/ppc/platforms/mpc8272ads_setup.c @@ -196,7 +196,7 @@ static void __init mpc8272ads_fixup_enet_pdata(struct platform_device *pdev, bd_t* bi = (void*)__res; int fs_no = fsid_fcc1+pdev->id-1; - if(fs_no > ARRAY_SIZE(mpc82xx_enet_pdata)) { + if(fs_no >= ARRAY_SIZE(mpc82xx_enet_pdata)) { return; } @@ -222,7 +222,7 @@ static void mpc8272ads_fixup_uart_pdata(struct platform_device *pdev, int id = fs_uart_id_scc2fsid(idx); /* no need to alter anything if console */ - if ((id <= num) && (!pdev->dev.platform_data)) { + if ((id < num) && (!pdev->dev.platform_data)) { pinfo = &mpc8272_uart_pdata[id]; pinfo->uart_clk = bd->bi_intfreq; pdev->dev.platform_data = pinfo; diff --git a/arch/ppc/platforms/mpc866ads_setup.c b/arch/ppc/platforms/mpc866ads_setup.c index 5f130dc..e95d2c1 100644 --- a/arch/ppc/platforms/mpc866ads_setup.c +++ b/arch/ppc/platforms/mpc866ads_setup.c @@ -259,7 +259,7 @@ static void mpc866ads_fixup_enet_pdata(struct platform_device *pdev, int fs_no) /* Get pointer to Communication Processor */ cp = cpmp; - if(fs_no > ARRAY_SIZE(mpc8xx_enet_pdata)) { + if(fs_no >= ARRAY_SIZE(mpc8xx_enet_pdata)) { printk(KERN_ERR"No network-suitable #%d device on bus", fs_no); return; } @@ -305,7 +305,7 @@ static void __init mpc866ads_fixup_uart_pdata(struct platform_device *pdev, int id = fs_uart_id_smc2fsid(idx); /* no need to alter anything if console */ - if ((id <= num) && (!pdev->dev.platform_data)) { + if ((id < num) && (!pdev->dev.platform_data)) { pinfo = &mpc866_uart_pdata[id]; pinfo->uart_clk = bd->bi_intfreq; pdev->dev.platform_data = pinfo; diff --git a/arch/ppc/platforms/mpc885ads_setup.c b/arch/ppc/platforms/mpc885ads_setup.c index 0229314..f8161f3 100644 --- a/arch/ppc/platforms/mpc885ads_setup.c +++ b/arch/ppc/platforms/mpc885ads_setup.c @@ -263,7 +263,7 @@ static void mpc885ads_fixup_enet_pdata(struct platform_device *pdev, int fs_no) char *e; int i; - if(fs_no > ARRAY_SIZE(mpc8xx_enet_pdata)) { + if(fs_no >= ARRAY_SIZE(mpc8xx_enet_pdata)) { printk(KERN_ERR"No network-suitable #%d device on bus", fs_no); return; } @@ -371,7 +371,7 @@ static void __init mpc885ads_fixup_uart_pdata(struct platform_device *pdev, int id = fs_uart_id_smc2fsid(idx); /* no need to alter anything if console */ - if ((id <= num) && (!pdev->dev.platform_data)) { + if ((id < num) && (!pdev->dev.platform_data)) { pinfo = &mpc885_uart_pdata[id]; pinfo->uart_clk = bd->bi_intfreq; pdev->dev.platform_data = pinfo; -- cgit v0.10.2 From 7d2bd30f67e231920091d55d0fffc3a4bf43e68a Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Wed, 11 Oct 2006 02:06:59 -0500 Subject: [POWERPC] ppc: Add missing calls to set_irq_regs In the timer_interrupt we were not calling set_irq_regs() and if we are profiling we will end up calling get_irq_regs(). This causes bad things to happen. Signed-off-by: Kumar Gala Signed-off-by: Paul Mackerras diff --git a/arch/ppc/kernel/time.c b/arch/ppc/kernel/time.c index d4b2cf7..18ee851 100644 --- a/arch/ppc/kernel/time.c +++ b/arch/ppc/kernel/time.c @@ -62,6 +62,7 @@ #include #include #include +#include #include @@ -129,6 +130,7 @@ void wakeup_decrementer(void) */ void timer_interrupt(struct pt_regs * regs) { + struct pt_regs *old_regs; int next_dec; unsigned long cpu = smp_processor_id(); unsigned jiffy_stamp = last_jiffy_stamp(cpu); @@ -137,6 +139,7 @@ void timer_interrupt(struct pt_regs * regs) if (atomic_read(&ppc_n_lost_interrupts) != 0) do_IRQ(regs); + old_regs = set_irq_regs(regs); irq_enter(); while ((next_dec = tb_ticks_per_jiffy - tb_delta(&jiffy_stamp)) <= 0) { @@ -188,6 +191,7 @@ void timer_interrupt(struct pt_regs * regs) ppc_md.heartbeat(); irq_exit(); + set_irq_regs(old_regs); } /* -- cgit v0.10.2 From f5a37b066165f9938a93653f6940f3dc85ce751d Mon Sep 17 00:00:00 2001 From: Li Yang Date: Wed, 11 Oct 2006 19:04:22 +0800 Subject: [POWERPC] Fix MPC8360EMDS PB board support MPC8360EMDS PB support is broken as some code was missing in last submission. This patch adds missing code and makes MPC8360EMDS PB support working. Signed-off-by: Li Yang Signed-off-by: Kim Phillips Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/83xx/Kconfig b/arch/powerpc/platforms/83xx/Kconfig index 0975e94..7edb6b4 100644 --- a/arch/powerpc/platforms/83xx/Kconfig +++ b/arch/powerpc/platforms/83xx/Kconfig @@ -32,6 +32,13 @@ config MPC834x_ITX Be aware that PCI initialization is the bootloader's responsiblilty. +config MPC8360E_PB + bool "Freescale MPC8360E PB" + select DEFAULT_UIMAGE + select QUICC_ENGINE + help + This option enables support for the MPC836x EMDS Processor Board. + endchoice config PPC_MPC832x @@ -46,4 +53,10 @@ config MPC834x select PPC_INDIRECT_PCI default y if MPC834x_SYS || MPC834x_ITX +config PPC_MPC836x + bool + select PPC_UDBG_16550 + select PPC_INDIRECT_PCI + default y if MPC8360E_PB + endmenu diff --git a/arch/powerpc/platforms/83xx/Makefile b/arch/powerpc/platforms/83xx/Makefile index 9387a11..e60fd75 100644 --- a/arch/powerpc/platforms/83xx/Makefile +++ b/arch/powerpc/platforms/83xx/Makefile @@ -5,3 +5,4 @@ obj-y := misc.o obj-$(CONFIG_PCI) += pci.o obj-$(CONFIG_MPC834x_SYS) += mpc834x_sys.o obj-$(CONFIG_MPC834x_ITX) += mpc834x_itx.o +obj-$(CONFIG_MPC8360E_PB) += mpc8360e_pb.o diff --git a/arch/powerpc/platforms/83xx/mpc8360e_pb.c b/arch/powerpc/platforms/83xx/mpc8360e_pb.c index c019190..1a523c8 100644 --- a/arch/powerpc/platforms/83xx/mpc8360e_pb.c +++ b/arch/powerpc/platforms/83xx/mpc8360e_pb.c @@ -30,6 +30,7 @@ #include #include +#include #include #include #include @@ -141,6 +142,24 @@ static void __init mpc8360_sys_setup_arch(void) #endif } +static int __init mpc8360_declare_of_platform_devices(void) +{ + struct device_node *np; + + for (np = NULL; (np = of_find_compatible_node(np, "network", + "ucc_geth")) != NULL;) { + int ucc_num; + char bus_id[BUS_ID_SIZE]; + + ucc_num = *((uint *) get_property(np, "device-id", NULL)) - 1; + snprintf(bus_id, BUS_ID_SIZE, "ucc_geth.%u", ucc_num); + of_platform_device_create(np, bus_id, NULL); + } + + return 0; +} +device_initcall(mpc8360_declare_of_platform_devices); + void __init mpc8360_sys_init_IRQ(void) { -- cgit v0.10.2 From 8ba738c2bb8fb83c0c99f680444d3ffd4c178ca2 Mon Sep 17 00:00:00 2001 From: Li Yang Date: Wed, 11 Oct 2006 19:27:57 +0800 Subject: [POWERPC] Add Makefile entry for MPC832x_mds support Add missing entry in Makefile for MPC832x MDS support. It also change white space to tab in MPC8360 entry. Signed-off-by: Li Yang Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/83xx/Makefile b/arch/powerpc/platforms/83xx/Makefile index e60fd75..f1aa7e2 100644 --- a/arch/powerpc/platforms/83xx/Makefile +++ b/arch/powerpc/platforms/83xx/Makefile @@ -5,4 +5,5 @@ obj-y := misc.o obj-$(CONFIG_PCI) += pci.o obj-$(CONFIG_MPC834x_SYS) += mpc834x_sys.o obj-$(CONFIG_MPC834x_ITX) += mpc834x_itx.o -obj-$(CONFIG_MPC8360E_PB) += mpc8360e_pb.o +obj-$(CONFIG_MPC8360E_PB) += mpc8360e_pb.o +obj-$(CONFIG_MPC832x_MDS) += mpc832x_mds.o -- cgit v0.10.2 From 60b2a46cd60c54bd6551ddfa01f0aab08ca58a5d Mon Sep 17 00:00:00 2001 From: Randy Vinson Date: Thu, 12 Oct 2006 13:36:23 -0700 Subject: [POWERPC] Fix IO Window Updates on P2P bridges. When update_bridge_base() updates the IO window on a PCI-to-PCI bridge, it fails to zero the upper 16 bits of the base and limit registers if the window size is less than 64K. This fixes it. Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/pci_32.c b/arch/powerpc/kernel/pci_32.c index 9b49f86..0d9ff72 100644 --- a/arch/powerpc/kernel/pci_32.c +++ b/arch/powerpc/kernel/pci_32.c @@ -441,14 +441,14 @@ update_bridge_base(struct pci_bus *bus, int i) end = res->end - off; io_base_lo = (start >> 8) & PCI_IO_RANGE_MASK; io_limit_lo = (end >> 8) & PCI_IO_RANGE_MASK; - if (end > 0xffff) { - pci_write_config_word(dev, PCI_IO_BASE_UPPER16, - start >> 16); - pci_write_config_word(dev, PCI_IO_LIMIT_UPPER16, - end >> 16); + if (end > 0xffff) io_base_lo |= PCI_IO_RANGE_TYPE_32; - } else + else io_base_lo |= PCI_IO_RANGE_TYPE_16; + pci_write_config_word(dev, PCI_IO_BASE_UPPER16, + start >> 16); + pci_write_config_word(dev, PCI_IO_LIMIT_UPPER16, + end >> 16); pci_write_config_byte(dev, PCI_IO_BASE, io_base_lo); pci_write_config_byte(dev, PCI_IO_LIMIT, io_limit_lo); -- cgit v0.10.2 From 6c4841c2b6c32a134f9f36e5e08857138cc12b10 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Fri, 13 Oct 2006 11:41:00 +1000 Subject: [POWERPC] Never panic when taking altivec exceptions from userspace At the moment we rely on a cpu feature bit or a firmware property to detect altivec. If we dont have either of these and the cpu does in fact support altivec we can cause a panic from userspace. It seems safer to always send a signal if we manage to get an 0xf20 exception from userspace. Signed-off-by: Anton Blanchard Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index d9f10f2f..5ed4c2c 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -900,14 +900,13 @@ void kernel_fp_unavailable_exception(struct pt_regs *regs) void altivec_unavailable_exception(struct pt_regs *regs) { -#if !defined(CONFIG_ALTIVEC) if (user_mode(regs)) { /* A user program has executed an altivec instruction, but this kernel doesn't support altivec. */ _exception(SIGILL, regs, ILL_ILLOPC, regs->nip); return; } -#endif + printk(KERN_EMERG "Unrecoverable VMX/Altivec Unavailable Exception " "%lx at %lx\n", regs->trap, regs->nip); die("Unrecoverable VMX/Altivec Unavailable Exception", regs, SIGABRT); -- cgit v0.10.2 From 99f48610252b736908fa5bdea505a480368308d6 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Fri, 13 Oct 2006 12:13:12 +1000 Subject: [POWERPC] POWER6 has 6 PMCs Change ->num_pmcs to match the number of PMCs in POWER6. Signed-off-by: Anton Blanchard Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 47a613c..95382f9 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -268,7 +268,7 @@ struct cpu_spec cpu_specs[] = { .cpu_user_features = COMMON_USER_POWER6, .icache_bsize = 128, .dcache_bsize = 128, - .num_pmcs = 8, + .num_pmcs = 6, .oprofile_cpu_type = "ppc64/power6", .oprofile_type = PPC_OPROFILE_POWER4, .oprofile_mmcra_sihv = POWER6_MMCRA_SIHV, -- cgit v0.10.2 From 00ae36de49cc718d4122e1c8aac96fd1a5a2553c Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Fri, 13 Oct 2006 12:17:16 +1000 Subject: [POWERPC] Better check in show_instructions Instead of just checking that an address is in the right range, use the provided __kernel_text_address() helper which covers both the kernel and module text sections. Signed-off-by: Anton Blanchard Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 7b2f645..f3d4dd5 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -341,13 +341,6 @@ struct task_struct *__switch_to(struct task_struct *prev, static int instructions_to_print = 16; -#ifdef CONFIG_PPC64 -#define BAD_PC(pc) ((REGION_ID(pc) != KERNEL_REGION_ID) && \ - (REGION_ID(pc) != VMALLOC_REGION_ID)) -#else -#define BAD_PC(pc) ((pc) < KERNELBASE) -#endif - static void show_instructions(struct pt_regs *regs) { int i; @@ -366,7 +359,8 @@ static void show_instructions(struct pt_regs *regs) * bad address because the pc *should* only be a * kernel address. */ - if (BAD_PC(pc) || __get_user(instr, (unsigned int __user *)pc)) { + if (!__kernel_text_address(pc) || + __get_user(instr, (unsigned int __user *)pc)) { printk("XXXXXXXX "); } else { if (regs->nip == pc) -- cgit v0.10.2 From 284a940675a64df253e3dffc60b09bb4bbb149e4 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Fri, 13 Oct 2006 12:26:57 +1000 Subject: [POWERPC] Check for offline nodes in pci NUMA code During boot we bring up all memory and cpu nodes. Normally a PCI device will be in one of these online nodes, however in some weird setups it may not. We have only seen this in the lab but we may as well check for the case and fallback to -1 (all nodes). Signed-off-by: Anton Blanchard Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c index 78d3c0f..9bae8a5 100644 --- a/arch/powerpc/kernel/pci_64.c +++ b/arch/powerpc/kernel/pci_64.c @@ -199,8 +199,14 @@ struct pci_controller * pcibios_alloc_controller(struct device_node *dev) pci_setup_pci_controller(phb); phb->arch_data = dev; phb->is_dynamic = mem_init_done; - if (dev) - PHB_SET_NODE(phb, of_node_to_nid(dev)); + if (dev) { + int nid = of_node_to_nid(dev); + + if (nid < 0 || !node_online(nid)) + nid = -1; + + PHB_SET_NODE(phb, nid); + } return phb; } -- cgit v0.10.2 From be60358e948484e472c88bf8a2d77e644ff5ed83 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 13:53:32 -0700 Subject: [DECNET]: Use correct config option for routing by fwmark in compare_keys() Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index a2a43d8..491429c 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -269,7 +269,7 @@ static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { return ((fl1->nl_u.dn_u.daddr ^ fl2->nl_u.dn_u.daddr) | (fl1->nl_u.dn_u.saddr ^ fl2->nl_u.dn_u.saddr) | -#ifdef CONFIG_IP_ROUTE_FWMARK +#ifdef CONFIG_DECNET_ROUTE_FWMARK (fl1->nl_u.dn_u.fwmark ^ fl2->nl_u.dn_u.fwmark) | #endif (fl1->nl_u.dn_u.scope ^ fl2->nl_u.dn_u.scope) | -- cgit v0.10.2 From f64ad5bb044326c6ebc6535d661c1abe78a0e5f2 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 14:07:52 -0700 Subject: [NETFILTER]: fix cut-and-paste error in exit functions Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index db9b896..39e1175 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -68,7 +68,7 @@ static int __init xt_nfqueue_init(void) static void __exit xt_nfqueue_fini(void) { - xt_register_targets(xt_nfqueue_target, ARRAY_SIZE(xt_nfqueue_target)); + xt_unregister_targets(xt_nfqueue_target, ARRAY_SIZE(xt_nfqueue_target)); } module_init(xt_nfqueue_init); diff --git a/net/netfilter/xt_connmark.c b/net/netfilter/xt_connmark.c index 92a5726..a8f0305 100644 --- a/net/netfilter/xt_connmark.c +++ b/net/netfilter/xt_connmark.c @@ -147,7 +147,7 @@ static int __init xt_connmark_init(void) static void __exit xt_connmark_fini(void) { - xt_register_matches(xt_connmark_match, ARRAY_SIZE(xt_connmark_match)); + xt_unregister_matches(xt_connmark_match, ARRAY_SIZE(xt_connmark_match)); } module_init(xt_connmark_init); -- cgit v0.10.2 From f603b6ec50faabbabde53ae2e2ce774968524c40 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 14:08:11 -0700 Subject: [NETFILTER]: arp_tables: missing unregistration on module unload Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 17e1a68..0849f1c 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1196,6 +1196,8 @@ err1: static void __exit arp_tables_fini(void) { nf_unregister_sockopt(&arpt_sockopts); + xt_unregister_target(&arpt_error_target); + xt_unregister_target(&arpt_standard_target); xt_proto_fini(NF_ARP); } -- cgit v0.10.2 From a9f54596fa20be3edefaa0b24c8714edb945eeaa Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 14:08:26 -0700 Subject: [NETFILTER]: ipt_ECN/ipt_TOS: fix incorrect checksum update Even though the tos field is only a single byte large, the values need to be converted to net-endian for the checkum update so they are in the corrent byte position. Also fix incorrect endian annotations. Reported by Stephane Chazelas Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c index 12a818a..1aa4517 100644 --- a/net/ipv4/netfilter/ipt_ECN.c +++ b/net/ipv4/netfilter/ipt_ECN.c @@ -28,7 +28,7 @@ static inline int set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo) { struct iphdr *iph = (*pskb)->nh.iph; - __be16 oldtos; + u_int16_t oldtos; if ((iph->tos & IPT_ECN_IP_MASK) != (einfo->ip_ect & IPT_ECN_IP_MASK)) { if (!skb_make_writable(pskb, sizeof(struct iphdr))) @@ -37,8 +37,8 @@ set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo) oldtos = iph->tos; iph->tos &= ~IPT_ECN_IP_MASK; iph->tos |= (einfo->ip_ect & IPT_ECN_IP_MASK); - iph->check = nf_csum_update(oldtos ^ htons(0xFFFF), iph->tos, - iph->check); + iph->check = nf_csum_update(htons(oldtos) ^ htons(0xFFFF), + htons(iph->tos), iph->check); } return 1; } diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c index 6b8b14c..83b80b3 100644 --- a/net/ipv4/netfilter/ipt_TOS.c +++ b/net/ipv4/netfilter/ipt_TOS.c @@ -30,7 +30,7 @@ target(struct sk_buff **pskb, { const struct ipt_tos_target_info *tosinfo = targinfo; struct iphdr *iph = (*pskb)->nh.iph; - __be16 oldtos; + u_int16_t oldtos; if ((iph->tos & IPTOS_TOS_MASK) != tosinfo->tos) { if (!skb_make_writable(pskb, sizeof(struct iphdr))) @@ -38,8 +38,8 @@ target(struct sk_buff **pskb, iph = (*pskb)->nh.iph; oldtos = iph->tos; iph->tos = (iph->tos & IPTOS_PREC_MASK) | tosinfo->tos; - iph->check = nf_csum_update(oldtos ^ htons(0xFFFF), iph->tos, - iph->check); + iph->check = nf_csum_update(htons(oldtos) ^ htons(0xFFFF), + htons(iph->tos), iph->check); } return IPT_CONTINUE; } -- cgit v0.10.2 From c08de5d5308ae0d20290344551ddd9cea8ded661 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 14:08:41 -0700 Subject: [NETFILTER]: xt_CONNSECMARK: fix Kconfig dependencies MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CONNSECMARK needs conntrack, add missing dependency to fix linking error with CONNSECMARK=y and CONNTRACK=m. Reported by Toralf Förster . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index ce94732..f619c65 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -209,7 +209,9 @@ config NETFILTER_XT_TARGET_SECMARK config NETFILTER_XT_TARGET_CONNSECMARK tristate '"CONNSECMARK" target support' - depends on NETFILTER_XTABLES && (NF_CONNTRACK_SECMARK || IP_NF_CONNTRACK_SECMARK) + depends on NETFILTER_XTABLES && \ + ((NF_CONNTRACK && NF_CONNTRACK_SECMARK) || \ + (IP_NF_CONNTRACK && IP_NF_CONNTRACK_SECMARK)) help The CONNSECMARK target copies security markings from packets to connections, and restores security markings from connections -- cgit v0.10.2 From 82b985434c5e9ceee772ca206889a856a7a4a200 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 12 Oct 2006 14:08:55 -0700 Subject: [NETFILTER]: Update MAINTAINERS entry Patches should go to myself CC netfilter-devel. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index a2b6d9f..208da3c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2049,11 +2049,13 @@ P: Marc Boucher P: James Morris P: Harald Welte P: Jozsef Kadlecsik -M: coreteam@netfilter.org +P: Patrick McHardy +M: kaber@trash.net +L: netfilter-devel@lists.netfilter.org +L: netfilter@lists.netfilter.org +L: coreteam@netfilter.org W: http://www.netfilter.org/ W: http://www.iptables.org/ -L: netfilter@lists.netfilter.org -L: netfilter-devel@lists.netfilter.org S: Supported NETLABEL -- cgit v0.10.2 From 9ea8cfd6aa74e710f0cb0731ecb9dee53fbebfb9 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 12 Oct 2006 14:09:16 -0700 Subject: [NETFILTER]: ctnetlink: Remove debugging messages Remove (compilation-breaking) debugging messages introduced at early development stage. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 53b6dff..262d0d4 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -44,13 +44,6 @@ MODULE_LICENSE("GPL"); static char __initdata version[] = "0.90"; -#if 0 -#define DEBUGP printk -#else -#define DEBUGP(format, args...) -#endif - - static inline int ctnetlink_dump_tuples_proto(struct sk_buff *skb, const struct ip_conntrack_tuple *tuple, @@ -398,7 +391,6 @@ nfattr_failure: static int ctnetlink_done(struct netlink_callback *cb) { - DEBUGP("entered %s\n", __FUNCTION__); if (cb->args[1]) ip_conntrack_put((struct ip_conntrack *)cb->args[1]); return 0; @@ -411,9 +403,6 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) struct ip_conntrack_tuple_hash *h; struct list_head *i; - DEBUGP("entered %s, last bucket=%lu id=%u\n", __FUNCTION__, - cb->args[0], *id); - read_lock_bh(&ip_conntrack_lock); last = (struct ip_conntrack *)cb->args[1]; for (; cb->args[0] < ip_conntrack_htable_size; cb->args[0]++) { @@ -452,7 +441,6 @@ out: if (last) ip_conntrack_put(last); - DEBUGP("leaving, last bucket=%lu id=%u\n", cb->args[0], *id); return skb->len; } @@ -466,8 +454,6 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct ip_conntrack_tuple *tuple) { struct nfattr *tb[CTA_IP_MAX]; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_IP_MAX, attr); if (nfattr_bad_size(tb, CTA_IP_MAX, cta_min_ip)) @@ -481,8 +467,6 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct ip_conntrack_tuple *tuple) return -EINVAL; tuple->dst.ip = *(__be32 *)NFA_DATA(tb[CTA_IP_V4_DST-1]); - DEBUGP("leaving\n"); - return 0; } @@ -503,8 +487,6 @@ ctnetlink_parse_tuple_proto(struct nfattr *attr, struct ip_conntrack_protocol *proto; int ret = 0; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_PROTO_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) @@ -531,8 +513,6 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct ip_conntrack_tuple *tuple, struct nfattr *tb[CTA_TUPLE_MAX]; int err; - DEBUGP("entered %s\n", __FUNCTION__); - memset(tuple, 0, sizeof(*tuple)); nfattr_parse_nested(tb, CTA_TUPLE_MAX, cda[type-1]); @@ -557,10 +537,6 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct ip_conntrack_tuple *tuple, else tuple->dst.dir = IP_CT_DIR_ORIGINAL; - DUMP_TUPLE(tuple); - - DEBUGP("leaving\n"); - return 0; } @@ -577,8 +553,6 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, struct nfattr *tb[CTA_PROTONAT_MAX]; struct ip_nat_protocol *npt; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_PROTONAT_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTONAT_MAX, cta_min_protonat)) @@ -597,7 +571,6 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, ip_nat_proto_put(npt); - DEBUGP("leaving\n"); return 0; } @@ -613,8 +586,6 @@ ctnetlink_parse_nat(struct nfattr *nat, struct nfattr *tb[CTA_NAT_MAX]; int err; - DEBUGP("entered %s\n", __FUNCTION__); - memset(range, 0, sizeof(*range)); nfattr_parse_nested(tb, CTA_NAT_MAX, nat); @@ -640,7 +611,6 @@ ctnetlink_parse_nat(struct nfattr *nat, if (err < 0) return err; - DEBUGP("leaving\n"); return 0; } #endif @@ -650,8 +620,6 @@ ctnetlink_parse_help(struct nfattr *attr, char **helper_name) { struct nfattr *tb[CTA_HELP_MAX]; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_HELP_MAX, attr); if (!tb[CTA_HELP_NAME-1]) @@ -679,8 +647,6 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, struct ip_conntrack *ct; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_MAX, cta_min)) return -EINVAL; @@ -698,10 +664,8 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, return err; h = ip_conntrack_find_get(&tuple, NULL); - if (!h) { - DEBUGP("tuple not found in conntrack hash\n"); + if (!h) return -ENOENT; - } ct = tuplehash_to_ctrack(h); @@ -716,7 +680,6 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, ct->timeout.function((unsigned long)ct); ip_conntrack_put(ct); - DEBUGP("leaving\n"); return 0; } @@ -731,8 +694,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, struct sk_buff *skb2 = NULL; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nlh->nlmsg_flags & NLM_F_DUMP) { struct nfgenmsg *msg = NLMSG_DATA(nlh); u32 rlen; @@ -770,11 +731,9 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, return err; h = ip_conntrack_find_get(&tuple, NULL); - if (!h) { - DEBUGP("tuple not found in conntrack hash"); + if (!h) return -ENOENT; - } - DEBUGP("tuple found\n"); + ct = tuplehash_to_ctrack(h); err = -ENOMEM; @@ -795,7 +754,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) goto out; - DEBUGP("leaving\n"); return 0; free: @@ -866,8 +824,6 @@ ctnetlink_change_helper(struct ip_conntrack *ct, struct nfattr *cda[]) char *helpname; int err; - DEBUGP("entered %s\n", __FUNCTION__); - /* don't change helper of sibling connections */ if (ct->master) return -EINVAL; @@ -938,8 +894,6 @@ ctnetlink_change_conntrack(struct ip_conntrack *ct, struct nfattr *cda[]) { int err; - DEBUGP("entered %s\n", __FUNCTION__); - if (cda[CTA_HELP-1]) { err = ctnetlink_change_helper(ct, cda); if (err < 0) @@ -969,7 +923,6 @@ ctnetlink_change_conntrack(struct ip_conntrack *ct, struct nfattr *cda[]) ct->mark = ntohl(*(__be32 *)NFA_DATA(cda[CTA_MARK-1])); #endif - DEBUGP("all done\n"); return 0; } @@ -981,8 +934,6 @@ ctnetlink_create_conntrack(struct nfattr *cda[], struct ip_conntrack *ct; int err = -EINVAL; - DEBUGP("entered %s\n", __FUNCTION__); - ct = ip_conntrack_alloc(otuple, rtuple); if (ct == NULL || IS_ERR(ct)) return -ENOMEM; @@ -1017,7 +968,6 @@ ctnetlink_create_conntrack(struct nfattr *cda[], if (ct->helper) ip_conntrack_helper_put(ct->helper); - DEBUGP("conntrack with id %u inserted\n", ct->id); return 0; err: @@ -1033,8 +983,6 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, struct ip_conntrack_tuple_hash *h = NULL; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_MAX, cta_min)) return -EINVAL; @@ -1058,7 +1006,6 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, if (h == NULL) { write_unlock_bh(&ip_conntrack_lock); - DEBUGP("no such conntrack, create new\n"); err = -ENOENT; if (nlh->nlmsg_flags & NLM_F_CREATE) err = ctnetlink_create_conntrack(cda, &otuple, &rtuple); @@ -1074,7 +1021,6 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, /* We manipulate the conntrack inside the global conntrack table lock, * so there's no need to increase the refcount */ - DEBUGP("conntrack found\n"); err = -EEXIST; if (!(nlh->nlmsg_flags & NLM_F_EXCL)) err = ctnetlink_change_conntrack(tuplehash_to_ctrack(h), cda); @@ -1249,8 +1195,6 @@ ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) struct list_head *i; u_int32_t *id = (u_int32_t *) &cb->args[0]; - DEBUGP("entered %s, last id=%llu\n", __FUNCTION__, *id); - read_lock_bh(&ip_conntrack_lock); list_for_each_prev(i, &ip_conntrack_expect_list) { exp = (struct ip_conntrack_expect *) i; @@ -1266,8 +1210,6 @@ ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) out: read_unlock_bh(&ip_conntrack_lock); - DEBUGP("leaving, last id=%llu\n", *id); - return skb->len; } @@ -1285,8 +1227,6 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, struct sk_buff *skb2; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_EXPECT_MAX, cta_min_exp)) return -EINVAL; @@ -1437,8 +1377,6 @@ ctnetlink_create_expect(struct nfattr *cda[]) struct ip_conntrack *ct; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - /* caller guarantees that those three CTA_EXPECT_* exist */ err = ctnetlink_parse_tuple(cda, &tuple, CTA_EXPECT_TUPLE); if (err < 0) @@ -1490,8 +1428,6 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, struct ip_conntrack_expect *exp; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_EXPECT_MAX, cta_min_exp)) return -EINVAL; @@ -1520,8 +1456,6 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, err = ctnetlink_change_expect(exp, cda); write_unlock_bh(&ip_conntrack_lock); - DEBUGP("leaving\n"); - return err; } diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 1721f7c..bd0156a 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -47,13 +47,6 @@ MODULE_LICENSE("GPL"); static char __initdata version[] = "0.93"; -#if 0 -#define DEBUGP printk -#else -#define DEBUGP(format, args...) -#endif - - static inline int ctnetlink_dump_tuples_proto(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple, @@ -410,7 +403,6 @@ static int ctnetlink_done(struct netlink_callback *cb) { if (cb->args[1]) nf_ct_put((struct nf_conn *)cb->args[1]); - DEBUGP("entered %s\n", __FUNCTION__); return 0; } @@ -425,9 +417,6 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) struct nfgenmsg *nfmsg = NLMSG_DATA(cb->nlh); u_int8_t l3proto = nfmsg->nfgen_family; - DEBUGP("entered %s, last bucket=%lu id=%u\n", __FUNCTION__, - cb->args[0], *id); - read_lock_bh(&nf_conntrack_lock); last = (struct nf_conn *)cb->args[1]; for (; cb->args[0] < nf_conntrack_htable_size; cb->args[0]++) { @@ -471,7 +460,6 @@ out: if (last) nf_ct_put(last); - DEBUGP("leaving, last bucket=%lu id=%u\n", cb->args[0], *id); return skb->len; } @@ -482,8 +470,6 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct nf_conntrack_tuple *tuple) struct nf_conntrack_l3proto *l3proto; int ret = 0; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_IP_MAX, attr); l3proto = nf_ct_l3proto_find_get(tuple->src.l3num); @@ -493,8 +479,6 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct nf_conntrack_tuple *tuple) nf_ct_l3proto_put(l3proto); - DEBUGP("leaving\n"); - return ret; } @@ -510,8 +494,6 @@ ctnetlink_parse_tuple_proto(struct nfattr *attr, struct nf_conntrack_protocol *proto; int ret = 0; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_PROTO_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) @@ -538,8 +520,6 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct nf_conntrack_tuple *tuple, struct nfattr *tb[CTA_TUPLE_MAX]; int err; - DEBUGP("entered %s\n", __FUNCTION__); - memset(tuple, 0, sizeof(*tuple)); nfattr_parse_nested(tb, CTA_TUPLE_MAX, cda[type-1]); @@ -566,10 +546,6 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct nf_conntrack_tuple *tuple, else tuple->dst.dir = IP_CT_DIR_ORIGINAL; - NF_CT_DUMP_TUPLE(tuple); - - DEBUGP("leaving\n"); - return 0; } @@ -586,8 +562,6 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, struct nfattr *tb[CTA_PROTONAT_MAX]; struct ip_nat_protocol *npt; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_PROTONAT_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTONAT_MAX, cta_min_protonat)) @@ -606,7 +580,6 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, ip_nat_proto_put(npt); - DEBUGP("leaving\n"); return 0; } @@ -622,8 +595,6 @@ ctnetlink_parse_nat(struct nfattr *nat, struct nfattr *tb[CTA_NAT_MAX]; int err; - DEBUGP("entered %s\n", __FUNCTION__); - memset(range, 0, sizeof(*range)); nfattr_parse_nested(tb, CTA_NAT_MAX, nat); @@ -649,7 +620,6 @@ ctnetlink_parse_nat(struct nfattr *nat, if (err < 0) return err; - DEBUGP("leaving\n"); return 0; } #endif @@ -659,8 +629,6 @@ ctnetlink_parse_help(struct nfattr *attr, char **helper_name) { struct nfattr *tb[CTA_HELP_MAX]; - DEBUGP("entered %s\n", __FUNCTION__); - nfattr_parse_nested(tb, CTA_HELP_MAX, attr); if (!tb[CTA_HELP_NAME-1]) @@ -690,8 +658,6 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, u_int8_t u3 = nfmsg->nfgen_family; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_MAX, cta_min)) return -EINVAL; @@ -709,10 +675,8 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, return err; h = nf_conntrack_find_get(&tuple, NULL); - if (!h) { - DEBUGP("tuple not found in conntrack hash\n"); + if (!h) return -ENOENT; - } ct = nf_ct_tuplehash_to_ctrack(h); @@ -727,7 +691,6 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, ct->timeout.function((unsigned long)ct); nf_ct_put(ct); - DEBUGP("leaving\n"); return 0; } @@ -744,8 +707,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, u_int8_t u3 = nfmsg->nfgen_family; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nlh->nlmsg_flags & NLM_F_DUMP) { u32 rlen; @@ -779,11 +740,9 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, return err; h = nf_conntrack_find_get(&tuple, NULL); - if (!h) { - DEBUGP("tuple not found in conntrack hash"); + if (!h) return -ENOENT; - } - DEBUGP("tuple found\n"); + ct = nf_ct_tuplehash_to_ctrack(h); err = -ENOMEM; @@ -804,7 +763,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) goto out; - DEBUGP("leaving\n"); return 0; free: @@ -876,8 +834,6 @@ ctnetlink_change_helper(struct nf_conn *ct, struct nfattr *cda[]) char *helpname; int err; - DEBUGP("entered %s\n", __FUNCTION__); - if (!help) { /* FIXME: we need to reallocate and rehash */ return -EBUSY; @@ -954,8 +910,6 @@ ctnetlink_change_conntrack(struct nf_conn *ct, struct nfattr *cda[]) { int err; - DEBUGP("entered %s\n", __FUNCTION__); - if (cda[CTA_HELP-1]) { err = ctnetlink_change_helper(ct, cda); if (err < 0) @@ -985,7 +939,6 @@ ctnetlink_change_conntrack(struct nf_conn *ct, struct nfattr *cda[]) ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1])); #endif - DEBUGP("all done\n"); return 0; } @@ -997,8 +950,6 @@ ctnetlink_create_conntrack(struct nfattr *cda[], struct nf_conn *ct; int err = -EINVAL; - DEBUGP("entered %s\n", __FUNCTION__); - ct = nf_conntrack_alloc(otuple, rtuple); if (ct == NULL || IS_ERR(ct)) return -ENOMEM; @@ -1028,7 +979,6 @@ ctnetlink_create_conntrack(struct nfattr *cda[], add_timer(&ct->timeout); nf_conntrack_hash_insert(ct); - DEBUGP("conntrack with id %u inserted\n", ct->id); return 0; err: @@ -1046,8 +996,6 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, u_int8_t u3 = nfmsg->nfgen_family; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_MAX, cta_min)) return -EINVAL; @@ -1071,7 +1019,6 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, if (h == NULL) { write_unlock_bh(&nf_conntrack_lock); - DEBUGP("no such conntrack, create new\n"); err = -ENOENT; if (nlh->nlmsg_flags & NLM_F_CREATE) err = ctnetlink_create_conntrack(cda, &otuple, &rtuple); @@ -1087,7 +1034,6 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, /* We manipulate the conntrack inside the global conntrack table lock, * so there's no need to increase the refcount */ - DEBUGP("conntrack found\n"); err = -EEXIST; if (!(nlh->nlmsg_flags & NLM_F_EXCL)) err = ctnetlink_change_conntrack(nf_ct_tuplehash_to_ctrack(h), cda); @@ -1268,8 +1214,6 @@ ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) struct nfgenmsg *nfmsg = NLMSG_DATA(cb->nlh); u_int8_t l3proto = nfmsg->nfgen_family; - DEBUGP("entered %s, last id=%llu\n", __FUNCTION__, *id); - read_lock_bh(&nf_conntrack_lock); list_for_each_prev(i, &nf_conntrack_expect_list) { exp = (struct nf_conntrack_expect *) i; @@ -1287,8 +1231,6 @@ ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) out: read_unlock_bh(&nf_conntrack_lock); - DEBUGP("leaving, last id=%llu\n", *id); - return skb->len; } @@ -1308,8 +1250,6 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, u_int8_t u3 = nfmsg->nfgen_family; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_EXPECT_MAX, cta_min_exp)) return -EINVAL; @@ -1460,8 +1400,6 @@ ctnetlink_create_expect(struct nfattr *cda[], u_int8_t u3) struct nf_conn_help *help; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - /* caller guarantees that those three CTA_EXPECT_* exist */ err = ctnetlink_parse_tuple(cda, &tuple, CTA_EXPECT_TUPLE, u3); if (err < 0) @@ -1516,8 +1454,6 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, u_int8_t u3 = nfmsg->nfgen_family; int err = 0; - DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_bad_size(cda, CTA_EXPECT_MAX, cta_min_exp)) return -EINVAL; @@ -1546,8 +1482,6 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, err = ctnetlink_change_expect(exp, cda); write_unlock_bh(&nf_conntrack_lock); - DEBUGP("leaving\n"); - return err; } -- cgit v0.10.2 From 1a620698c29b5e18150ec04ace0609fb07d08d3e Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 12 Oct 2006 14:45:38 -0700 Subject: [BRIDGE]: flush forwarding table when device carrier off Flush the forwarding table when carrier is lost. This helps for availability because we don't want to forward to a downed device and new packets may come in on other links. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 3a73b8c..d9f0486 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -128,7 +128,10 @@ void br_fdb_cleanup(unsigned long _data) mod_timer(&br->gc_timer, jiffies + HZ/10); } -void br_fdb_delete_by_port(struct net_bridge *br, struct net_bridge_port *p) + +void br_fdb_delete_by_port(struct net_bridge *br, + const struct net_bridge_port *p, + int do_all) { int i; @@ -142,6 +145,8 @@ void br_fdb_delete_by_port(struct net_bridge *br, struct net_bridge_port *p) if (f->dst != p) continue; + if (f->is_static && !do_all) + continue; /* * if multiple ports all have the same device address * then when one port is deleted, assign diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index b1211d534..f753c40 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -163,7 +163,7 @@ static void del_nbp(struct net_bridge_port *p) br_stp_disable_port(p); spin_unlock_bh(&br->lock); - br_fdb_delete_by_port(br, p); + br_fdb_delete_by_port(br, p, 1); list_del_rcu(&p->list); @@ -448,7 +448,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) return 0; err2: - br_fdb_delete_by_port(br, p); + br_fdb_delete_by_port(br, p, 1); err1: kobject_del(&p->kobj); err0: diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index c491fb2..74258d8 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -143,7 +143,7 @@ extern void br_fdb_changeaddr(struct net_bridge_port *p, const unsigned char *newaddr); extern void br_fdb_cleanup(unsigned long arg); extern void br_fdb_delete_by_port(struct net_bridge *br, - struct net_bridge_port *p); + const struct net_bridge_port *p, int do_all); extern struct net_bridge_fdb_entry *__br_fdb_get(struct net_bridge *br, const unsigned char *addr); extern struct net_bridge_fdb_entry *br_fdb_get(struct net_bridge *br, diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c index 14cd025..d294224 100644 --- a/net/bridge/br_stp_if.c +++ b/net/bridge/br_stp_if.c @@ -113,6 +113,8 @@ void br_stp_disable_port(struct net_bridge_port *p) del_timer(&p->forward_delay_timer); del_timer(&p->hold_timer); + br_fdb_delete_by_port(br, p, 0); + br_configuration_update(br); br_port_state_selection(br); -- cgit v0.10.2 From 044a68ed8a692f643cf3c0a54c380a922584f34f Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Wed, 11 Oct 2006 19:10:47 -0400 Subject: NetLabel: only deref the CIPSOv4 standard map fields when using standard mapping Fix several places in the CIPSO code where it was dereferencing fields which did not have valid pointers by moving those pointer dereferences into code blocks where the pointers are valid. Signed-off-by: Paul Moore Signed-off-by: James Morris diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index bde8cca..d19c9ac 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -773,13 +773,15 @@ static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def, { int cat = -1; u32 bitmap_len_bits = bitmap_len * 8; - u32 cipso_cat_size = doi_def->map.std->cat.cipso_size; - u32 *cipso_array = doi_def->map.std->cat.cipso; + u32 cipso_cat_size; + u32 *cipso_array; switch (doi_def->type) { case CIPSO_V4_MAP_PASS: return 0; case CIPSO_V4_MAP_STD: + cipso_cat_size = doi_def->map.std->cat.cipso_size; + cipso_array = doi_def->map.std->cat.cipso; for (;;) { cat = cipso_v4_bitmap_walk(bitmap, bitmap_len_bits, @@ -825,8 +827,8 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, u32 net_spot_max = 0; u32 host_clen_bits = host_cat_len * 8; u32 net_clen_bits = net_cat_len * 8; - u32 host_cat_size = doi_def->map.std->cat.local_size; - u32 *host_cat_array = doi_def->map.std->cat.local; + u32 host_cat_size; + u32 *host_cat_array; switch (doi_def->type) { case CIPSO_V4_MAP_PASS: @@ -838,6 +840,8 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, memcpy(net_cat, host_cat, net_spot_max); return net_spot_max; case CIPSO_V4_MAP_STD: + host_cat_size = doi_def->map.std->cat.local_size; + host_cat_array = doi_def->map.std->cat.local; for (;;) { host_spot = cipso_v4_bitmap_walk(host_cat, host_clen_bits, @@ -893,8 +897,8 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def, int net_spot = -1; u32 net_clen_bits = net_cat_len * 8; u32 host_clen_bits = host_cat_len * 8; - u32 net_cat_size = doi_def->map.std->cat.cipso_size; - u32 *net_cat_array = doi_def->map.std->cat.cipso; + u32 net_cat_size; + u32 *net_cat_array; switch (doi_def->type) { case CIPSO_V4_MAP_PASS: @@ -903,6 +907,8 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def, memcpy(host_cat, net_cat, net_cat_len); return net_cat_len; case CIPSO_V4_MAP_STD: + net_cat_size = doi_def->map.std->cat.cipso_size; + net_cat_array = doi_def->map.std->cat.cipso; for (;;) { net_spot = cipso_v4_bitmap_walk(net_cat, net_clen_bits, -- cgit v0.10.2 From bf0edf39296097f20c5fcc4919ed7d339194bd75 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Wed, 11 Oct 2006 19:10:48 -0400 Subject: NetLabel: better error handling involving mls_export_cat() Upon inspection it looked like the error handling for mls_export_cat() was rather poor. This patch addresses this by NULL'ing out kfree()'d pointers before returning and checking the return value of the function everywhere it is called. Signed-off-by: Paul Moore Signed-off-by: James Morris diff --git a/security/selinux/ss/ebitmap.c b/security/selinux/ss/ebitmap.c index cfed1d3..d539346 100644 --- a/security/selinux/ss/ebitmap.c +++ b/security/selinux/ss/ebitmap.c @@ -93,11 +93,15 @@ int ebitmap_export(const struct ebitmap *src, size_t bitmap_byte; unsigned char bitmask; + if (src->highbit == 0) { + *dst = NULL; + *dst_len = 0; + return 0; + } + bitmap_len = src->highbit / 8; if (src->highbit % 7) bitmap_len += 1; - if (bitmap_len == 0) - return -EINVAL; bitmap = kzalloc((bitmap_len & ~(sizeof(MAPTYPE) - 1)) + sizeof(MAPTYPE), diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c index c713af2..2cca8e2 100644 --- a/security/selinux/ss/mls.c +++ b/security/selinux/ss/mls.c @@ -640,8 +640,13 @@ int mls_export_cat(const struct context *context, { int rc = -EPERM; - if (!selinux_mls_enabled) + if (!selinux_mls_enabled) { + *low = NULL; + *low_len = 0; + *high = NULL; + *high_len = 0; return 0; + } if (low != NULL) { rc = ebitmap_export(&context->range.level[0].cat, @@ -661,10 +666,16 @@ int mls_export_cat(const struct context *context, return 0; export_cat_failure: - if (low != NULL) + if (low != NULL) { kfree(*low); - if (high != NULL) + *low = NULL; + *low_len = 0; + } + if (high != NULL) { kfree(*high); + *high = NULL; + *high_len = 0; + } return rc; } diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 18274b0..b1f6fb3 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2399,31 +2399,33 @@ static int selinux_netlbl_socket_setsid(struct socket *sock, u32 sid) if (!ss_initialized) return 0; + netlbl_secattr_init(&secattr); + POLICY_RDLOCK; ctx = sidtab_search(&sidtab, sid); if (ctx == NULL) goto netlbl_socket_setsid_return; - netlbl_secattr_init(&secattr); secattr.domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1], GFP_ATOMIC); mls_export_lvl(ctx, &secattr.mls_lvl, NULL); secattr.mls_lvl_vld = 1; - mls_export_cat(ctx, - &secattr.mls_cat, - &secattr.mls_cat_len, - NULL, - NULL); + rc = mls_export_cat(ctx, + &secattr.mls_cat, + &secattr.mls_cat_len, + NULL, + NULL); + if (rc != 0) + goto netlbl_socket_setsid_return; rc = netlbl_socket_setattr(sock, &secattr); if (rc == 0) sksec->nlbl_state = NLBL_LABELED; - netlbl_secattr_destroy(&secattr); - netlbl_socket_setsid_return: POLICY_RDUNLOCK; + netlbl_secattr_destroy(&secattr); return rc; } -- cgit v0.10.2 From ea614d7f4fb2d436b7a5ee490d1011615f6b38d5 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Wed, 11 Oct 2006 19:10:49 -0400 Subject: NetLabel: the CIPSOv4 passthrough mapping does not pass categories correctly The CIPSO passthrough mapping had a problem when sending categories which would cause no or incorrect categories to be sent on the wire with a packet. This patch fixes the problem which was a simple off-by-one bug. Signed-off-by: Paul Moore Signed-off-by: James Morris diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index d19c9ac..e2077a3 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -832,8 +832,8 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, switch (doi_def->type) { case CIPSO_V4_MAP_PASS: - net_spot_max = host_cat_len - 1; - while (net_spot_max > 0 && host_cat[net_spot_max] == 0) + net_spot_max = host_cat_len; + while (net_spot_max > 0 && host_cat[net_spot_max - 1] == 0) net_spot_max--; if (net_spot_max > net_cat_len) return -EINVAL; -- cgit v0.10.2 From 4663afe2c848e2abc8791202beecf40684f13eb4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 12 Oct 2006 21:21:06 -0700 Subject: [NET]: reduce sizeof(struct inet_peer), cleanup, change in peer_check_expire() 1) shrink struct inet_peer on 64 bits platforms. diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h index 925573f..f13cc0c 100644 --- a/include/net/inetpeer.h +++ b/include/net/inetpeer.h @@ -19,7 +19,7 @@ struct inet_peer { struct inet_peer *avl_left, *avl_right; struct inet_peer *unused_next, **unused_prevp; - unsigned long dtime; /* the time of last use of not + __u32 dtime; /* the time of last use of not * referenced entries */ atomic_t refcnt; __be32 v4daddr; /* peer's address */ @@ -35,21 +35,8 @@ void inet_initpeers(void) __init; /* can be called with or without local BH being disabled */ struct inet_peer *inet_getpeer(__be32 daddr, int create); -extern spinlock_t inet_peer_unused_lock; -extern struct inet_peer **inet_peer_unused_tailp; /* can be called from BH context or outside */ -static inline void inet_putpeer(struct inet_peer *p) -{ - spin_lock_bh(&inet_peer_unused_lock); - if (atomic_dec_and_test(&p->refcnt)) { - p->unused_prevp = inet_peer_unused_tailp; - p->unused_next = NULL; - *inet_peer_unused_tailp = p; - inet_peer_unused_tailp = &p->unused_next; - p->dtime = jiffies; - } - spin_unlock_bh(&inet_peer_unused_lock); -} +extern void inet_putpeer(struct inet_peer *p); extern spinlock_t inet_peer_idlock; /* can be called with or without local BH being disabled */ diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index 2b1a54b..f072f38 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -94,10 +94,8 @@ int inet_peer_minttl = 120 * HZ; /* TTL under high load: 120 sec */ int inet_peer_maxttl = 10 * 60 * HZ; /* usual time to live: 10 min */ static struct inet_peer *inet_peer_unused_head; -/* Exported for inet_putpeer inline function. */ -struct inet_peer **inet_peer_unused_tailp = &inet_peer_unused_head; -DEFINE_SPINLOCK(inet_peer_unused_lock); -#define PEER_MAX_CLEANUP_WORK 30 +static struct inet_peer **inet_peer_unused_tailp = &inet_peer_unused_head; +static DEFINE_SPINLOCK(inet_peer_unused_lock); static void peer_check_expire(unsigned long dummy); static DEFINE_TIMER(peer_periodic_timer, peer_check_expire, 0, 0); @@ -340,7 +338,8 @@ static int cleanup_once(unsigned long ttl) spin_lock_bh(&inet_peer_unused_lock); p = inet_peer_unused_head; if (p != NULL) { - if (time_after(p->dtime + ttl, jiffies)) { + __u32 delta = (__u32)jiffies - p->dtime; + if (delta < ttl) { /* Do not prune fresh entries. */ spin_unlock_bh(&inet_peer_unused_lock); return -1; @@ -432,7 +431,7 @@ out_free: /* Called with local BH disabled. */ static void peer_check_expire(unsigned long dummy) { - int i; + unsigned long now = jiffies; int ttl; if (peer_total >= inet_peer_threshold) @@ -441,7 +440,10 @@ static void peer_check_expire(unsigned long dummy) ttl = inet_peer_maxttl - (inet_peer_maxttl - inet_peer_minttl) / HZ * peer_total / inet_peer_threshold * HZ; - for (i = 0; i < PEER_MAX_CLEANUP_WORK && !cleanup_once(ttl); i++); + while (!cleanup_once(ttl)) { + if (jiffies != now) + break; + } /* Trigger the timer after inet_peer_gc_mintime .. inet_peer_gc_maxtime * interval depending on the total number of entries (more entries, @@ -455,3 +457,16 @@ static void peer_check_expire(unsigned long dummy) peer_total / inet_peer_threshold * HZ; add_timer(&peer_periodic_timer); } + +void inet_putpeer(struct inet_peer *p) +{ + spin_lock_bh(&inet_peer_unused_lock); + if (atomic_dec_and_test(&p->refcnt)) { + p->unused_prevp = inet_peer_unused_tailp; + p->unused_next = NULL; + *inet_peer_unused_tailp = p; + inet_peer_unused_tailp = &p->unused_next; + p->dtime = (__u32)jiffies; + } + spin_unlock_bh(&inet_peer_unused_lock); +} -- cgit v0.10.2 From 918049f0135854a1583f9b3b88f44dbf2b027329 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 12 Oct 2006 22:03:24 -0700 Subject: [XFRM]: Fix xfrm_state_num going negative. Missing counter bump when hashing in a new ACQ xfrm_state. Now that we have two spots to do the hash grow check, break it out into a helper function. Signed-off-by: David S. Miller diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 39b8bf3..84bbf84 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -614,6 +614,14 @@ out: return x; } +static void xfrm_hash_grow_check(int have_hash_collision) +{ + if (have_hash_collision && + (xfrm_state_hmask + 1) < xfrm_state_hashmax && + xfrm_state_num > xfrm_state_hmask) + schedule_work(&xfrm_hash_work); +} + static void __xfrm_state_insert(struct xfrm_state *x) { unsigned int h; @@ -642,10 +650,7 @@ static void __xfrm_state_insert(struct xfrm_state *x) xfrm_state_num++; - if (x->bydst.next != NULL && - (xfrm_state_hmask + 1) < xfrm_state_hashmax && - xfrm_state_num > xfrm_state_hmask) - schedule_work(&xfrm_hash_work); + xfrm_hash_grow_check(x->bydst.next != NULL); } /* xfrm_state_lock is held */ @@ -753,6 +758,10 @@ static struct xfrm_state *__find_acq_core(unsigned short family, u8 mode, u32 re h = xfrm_src_hash(daddr, saddr, family); hlist_add_head(&x->bysrc, xfrm_state_bysrc+h); wake_up(&km_waitq); + + xfrm_state_num++; + + xfrm_hash_grow_check(x->bydst.next != NULL); } return x; -- cgit v0.10.2 From adaa70bbdfbc725e485179b06c8b23a20fbb7952 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 13 Oct 2006 15:01:03 -0700 Subject: [IPv6] rules: Use RT6_LOOKUP_F_HAS_SADDR and fix source based selectors Fixes rt6_lookup() to provide the source address in the flow and sets RT6_LOOKUP_F_HAS_SADDR whenever it is present in the flow. Avoids unnecessary prefix comparisons by checking for a prefix length first. Fixes the rule logic to not match packets if a source selector has been specified but no source address is available. Thanks to Kim Nordlund for working on this patch with me. Signed-off-by: Thomas Graf Acked-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index d8c1057..1896ecb 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -117,12 +117,15 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) { struct fib6_rule *r = (struct fib6_rule *) rule; - if (!ipv6_prefix_equal(&fl->fl6_dst, &r->dst.addr, r->dst.plen)) + if (r->dst.plen && + !ipv6_prefix_equal(&fl->fl6_dst, &r->dst.addr, r->dst.plen)) return 0; - if ((flags & RT6_LOOKUP_F_HAS_SADDR) && - !ipv6_prefix_equal(&fl->fl6_src, &r->src.addr, r->src.plen)) - return 0; + if (r->src.plen) { + if (!(flags & RT6_LOOKUP_F_HAS_SADDR) || + !ipv6_prefix_equal(&fl->fl6_src, &r->src.addr, r->src.plen)) + return 0; + } if (r->tclass && r->tclass != ((ntohl(fl->fl6_flowlabel) >> 20) & 0xff)) return 0; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d6b4b4f..a1b0f07 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -529,13 +529,17 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, .nl_u = { .ip6_u = { .daddr = *daddr, - /* TODO: saddr */ }, }, }; struct dst_entry *dst; int flags = strict ? RT6_LOOKUP_F_IFACE : 0; + if (saddr) { + memcpy(&fl.fl6_src, saddr, sizeof(*saddr)); + flags |= RT6_LOOKUP_F_HAS_SADDR; + } + dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup); if (dst->error == 0) return (struct rt6_info *) dst; @@ -697,6 +701,7 @@ out2: void ip6_route_input(struct sk_buff *skb) { struct ipv6hdr *iph = skb->nh.ipv6h; + int flags = RT6_LOOKUP_F_HAS_SADDR; struct flowi fl = { .iif = skb->dev->ifindex, .nl_u = { @@ -711,7 +716,9 @@ void ip6_route_input(struct sk_buff *skb) }, .proto = iph->nexthdr, }; - int flags = rt6_need_strict(&iph->daddr) ? RT6_LOOKUP_F_IFACE : 0; + + if (rt6_need_strict(&iph->daddr)) + flags |= RT6_LOOKUP_F_IFACE; skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input); } @@ -794,6 +801,9 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) if (rt6_need_strict(&fl->fl6_dst)) flags |= RT6_LOOKUP_F_IFACE; + if (!ipv6_addr_any(&fl->fl6_src)) + flags |= RT6_LOOKUP_F_HAS_SADDR; + return fib6_rule_lookup(fl, flags, ip6_pol_route_output); } @@ -1345,6 +1355,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, struct in6_addr *gateway, struct net_device *dev) { + int flags = RT6_LOOKUP_F_HAS_SADDR; struct ip6rd_flowi rdfl = { .fl = { .oif = dev->ifindex, @@ -1357,7 +1368,9 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, }, .gateway = *gateway, }; - int flags = rt6_need_strict(dest) ? RT6_LOOKUP_F_IFACE : 0; + + if (rt6_need_strict(dest)) + flags |= RT6_LOOKUP_F_IFACE; return (struct rt6_info *)fib6_rule_lookup((struct flowi *)&rdfl, flags, __ip6_route_redirect); } -- cgit v0.10.2 From f1a95859a86fcdfd94f8b6dc3255d70d037e1caf Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Fri, 13 Oct 2006 15:03:34 -0700 Subject: [IPV6]: Remove bogus WARN_ON in Proxy-NA handling. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 0304b5f..41a8a5f 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -967,8 +967,6 @@ static void ndisc_recv_na(struct sk_buff *skb) ipv6_devconf.forwarding && ipv6_devconf.proxy_ndp && pneigh_lookup(&nd_tbl, &msg->target, dev, 0)) { /* XXX: idev->cnf.prixy_ndp */ - WARN_ON(skb->dst != NULL && - ((struct rt6_info *)skb->dst)->rt6i_idev); goto out; } -- cgit v0.10.2 From 39c850863d5e36e72ecf9bc3537ec717bcce97fd Mon Sep 17 00:00:00 2001 From: Jan Dittmer Date: Fri, 13 Oct 2006 15:05:53 -0700 Subject: [IPV6] sit: Add missing MODULE_LICENSE This is missing the MODULE_LICENSE statements and taints the kernel upon loading. License is obvious from the beginning of the file. Signed-off-by: Jan Dittmer Signed-off-by: Joerg Roedel Signed-off-by: David S. Miller diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index dc5765b..b481a4d 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -853,3 +853,4 @@ int __init sit_init(void) module_init(sit_init); module_exit(sit_cleanup); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From e9c5702e3c5558dade169949abd730173e87ef9c Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:30:22 +0200 Subject: [Bluetooth] Fix compat ioctl for BNEP, CMTP and HIDP There exists no attempt do deal with the fact that a structure with a uint32_t followed by a pointer is going to be different for 32-bit and 64-bit userspace. Any 32-bit process trying to use it will be failing with -EFAULT if it's lucky; suffering from having data dumped at a random address if it's not. Signed-off-by: David Woodhouse Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/bnep/sock.c b/net/bluetooth/bnep/sock.c index 28c5583..5d9d6f1 100644 --- a/net/bluetooth/bnep/sock.c +++ b/net/bluetooth/bnep/sock.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #include @@ -146,24 +147,56 @@ static int bnep_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long return 0; } +#ifdef CONFIG_COMPAT +static int bnep_sock_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) +{ + if (cmd == BNEPGETCONNLIST) { + struct bnep_connlist_req cl; + uint32_t uci; + int err; + + if (get_user(cl.cnum, (uint32_t __user *) arg) || + get_user(uci, (u32 __user *) (arg + 4))) + return -EFAULT; + + cl.ci = compat_ptr(uci); + + if (cl.cnum <= 0) + return -EINVAL; + + err = bnep_get_connlist(&cl); + + if (!err && put_user(cl.cnum, (uint32_t __user *) arg)) + err = -EFAULT; + + return err; + } + + return bnep_sock_ioctl(sock, cmd, arg); +} +#endif + static const struct proto_ops bnep_sock_ops = { - .family = PF_BLUETOOTH, - .owner = THIS_MODULE, - .release = bnep_sock_release, - .ioctl = bnep_sock_ioctl, - .bind = sock_no_bind, - .getname = sock_no_getname, - .sendmsg = sock_no_sendmsg, - .recvmsg = sock_no_recvmsg, - .poll = sock_no_poll, - .listen = sock_no_listen, - .shutdown = sock_no_shutdown, - .setsockopt = sock_no_setsockopt, - .getsockopt = sock_no_getsockopt, - .connect = sock_no_connect, - .socketpair = sock_no_socketpair, - .accept = sock_no_accept, - .mmap = sock_no_mmap + .family = PF_BLUETOOTH, + .owner = THIS_MODULE, + .release = bnep_sock_release, + .ioctl = bnep_sock_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = bnep_sock_compat_ioctl, +#endif + .bind = sock_no_bind, + .getname = sock_no_getname, + .sendmsg = sock_no_sendmsg, + .recvmsg = sock_no_recvmsg, + .poll = sock_no_poll, + .listen = sock_no_listen, + .shutdown = sock_no_shutdown, + .setsockopt = sock_no_setsockopt, + .getsockopt = sock_no_getsockopt, + .connect = sock_no_connect, + .socketpair = sock_no_socketpair, + .accept = sock_no_accept, + .mmap = sock_no_mmap }; static struct proto bnep_proto = { diff --git a/net/bluetooth/cmtp/sock.c b/net/bluetooth/cmtp/sock.c index 10ad7fd..0547edd 100644 --- a/net/bluetooth/cmtp/sock.c +++ b/net/bluetooth/cmtp/sock.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -137,11 +138,43 @@ static int cmtp_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long return -EINVAL; } +#ifdef CONFIG_COMPAT +static int cmtp_sock_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) +{ + if (cmd == CMTPGETCONNLIST) { + struct cmtp_connlist_req cl; + uint32_t uci; + int err; + + if (get_user(cl.cnum, (uint32_t __user *) arg) || + get_user(uci, (u32 __user *) (arg + 4))) + return -EFAULT; + + cl.ci = compat_ptr(uci); + + if (cl.cnum <= 0) + return -EINVAL; + + err = cmtp_get_connlist(&cl); + + if (!err && put_user(cl.cnum, (uint32_t __user *) arg)) + err = -EFAULT; + + return err; + } + + return cmtp_sock_ioctl(sock, cmd, arg); +} +#endif + static const struct proto_ops cmtp_sock_ops = { .family = PF_BLUETOOTH, .owner = THIS_MODULE, .release = cmtp_sock_release, .ioctl = cmtp_sock_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = cmtp_sock_compat_ioctl, +#endif .bind = sock_no_bind, .getname = sock_no_getname, .sendmsg = sock_no_sendmsg, diff --git a/net/bluetooth/hidp/sock.c b/net/bluetooth/hidp/sock.c index 099646e..6242446 100644 --- a/net/bluetooth/hidp/sock.c +++ b/net/bluetooth/hidp/sock.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include "hidp.h" @@ -143,11 +144,88 @@ static int hidp_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long return -EINVAL; } +#ifdef CONFIG_COMPAT +struct compat_hidp_connadd_req { + int ctrl_sock; // Connected control socket + int intr_sock; // Connteted interrupt socket + __u16 parser; + __u16 rd_size; + compat_uptr_t rd_data; + __u8 country; + __u8 subclass; + __u16 vendor; + __u16 product; + __u16 version; + __u32 flags; + __u32 idle_to; + char name[128]; +}; + +static int hidp_sock_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) +{ + if (cmd == HIDPGETCONNLIST) { + struct hidp_connlist_req cl; + uint32_t uci; + int err; + + if (get_user(cl.cnum, (uint32_t __user *) arg) || + get_user(uci, (u32 __user *) (arg + 4))) + return -EFAULT; + + cl.ci = compat_ptr(uci); + + if (cl.cnum <= 0) + return -EINVAL; + + err = hidp_get_connlist(&cl); + + if (!err && put_user(cl.cnum, (uint32_t __user *) arg)) + err = -EFAULT; + + return err; + } else if (cmd == HIDPCONNADD) { + struct compat_hidp_connadd_req ca; + struct hidp_connadd_req __user *uca; + + uca = compat_alloc_user_space(sizeof(*uca)); + + if (copy_from_user(&ca, (void *) arg, sizeof(ca))) + return -EFAULT; + + if (put_user(ca.ctrl_sock, &uca->ctrl_sock) || + put_user(ca.intr_sock, &uca->intr_sock) || + put_user(ca.parser, &uca->parser) || + put_user(ca.rd_size, &uca->parser) || + put_user(compat_ptr(ca.rd_data), &uca->rd_data) || + put_user(ca.country, &uca->country) || + put_user(ca.subclass, &uca->subclass) || + put_user(ca.vendor, &uca->vendor) || + put_user(ca.product, &uca->product) || + put_user(ca.version, &uca->version) || + put_user(ca.flags, &uca->flags) || + put_user(ca.idle_to, &uca->idle_to) || + copy_to_user(&uca->name[0], &ca.name[0], 128)) + return -EFAULT; + + arg = (unsigned long) uca; + + /* Fall through. We don't actually write back any _changes_ + to the structure anyway, so there's no need to copy back + into the original compat version */ + } + + return hidp_sock_ioctl(sock, cmd, arg); +} +#endif + static const struct proto_ops hidp_sock_ops = { .family = PF_BLUETOOTH, .owner = THIS_MODULE, .release = hidp_sock_release, .ioctl = hidp_sock_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = hidp_sock_compat_ioctl, +#endif .bind = sock_no_bind, .getname = sock_no_getname, .sendmsg = sock_no_sendmsg, -- cgit v0.10.2 From df5c37ea9a3820a9515d8204450aec00f48f8f88 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:30:45 +0200 Subject: [Bluetooth] Handle return values from driver core functions Some return values of the driver core register and create functions are not handled and so might cause unexpected problems. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 989b22d..c7e079d 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -242,10 +242,14 @@ static void add_conn(void *data) struct hci_conn *conn = data; int i; - device_register(&conn->dev); + if (device_register(&conn->dev) < 0) { + BT_ERR("Failed to register connection device"); + return; + } for (i = 0; conn_attrs[i]; i++) - device_create_file(&conn->dev, conn_attrs[i]); + if (device_create_file(&conn->dev, conn_attrs[i]) < 0) + BT_ERR("Failed to create connection attribute"); } void hci_conn_add_sysfs(struct hci_conn *conn) @@ -312,7 +316,8 @@ int hci_register_sysfs(struct hci_dev *hdev) return err; for (i = 0; bt_attrs[i]; i++) - device_create_file(dev, bt_attrs[i]); + if (device_create_file(dev, bt_attrs[i]) < 0) + BT_ERR("Failed to create device attribute"); return 0; } diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index d56f60b..66fc621 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -2216,7 +2216,8 @@ static int __init l2cap_init(void) goto error; } - class_create_file(bt_class, &class_attr_l2cap); + if (class_create_file(bt_class, &class_attr_l2cap) < 0) + BT_ERR("Failed to create L2CAP info file"); BT_INFO("L2CAP ver %s", VERSION); BT_INFO("L2CAP socket layer initialized"); diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 468df3b..ddc4e9d 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -2058,7 +2058,8 @@ static int __init rfcomm_init(void) kernel_thread(rfcomm_run, NULL, CLONE_KERNEL); - class_create_file(bt_class, &class_attr_rfcomm_dlc); + if (class_create_file(bt_class, &class_attr_rfcomm_dlc) < 0) + BT_ERR("Failed to create RFCOMM info file"); rfcomm_init_sockets(); diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 220fee0..530cc41 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -944,7 +944,8 @@ int __init rfcomm_init_sockets(void) if (err < 0) goto error; - class_create_file(bt_class, &class_attr_rfcomm); + if (class_create_file(bt_class, &class_attr_rfcomm) < 0) + BT_ERR("Failed to create RFCOMM info file"); BT_INFO("RFCOMM socket layer initialized"); diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 7714a2e..14b0f69 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -967,7 +967,8 @@ static int __init sco_init(void) goto error; } - class_create_file(bt_class, &class_attr_sco); + if (class_create_file(bt_class, &class_attr_sco) < 0) + BT_ERR("Failed to create SCO info file"); BT_INFO("SCO (Voice Link) ver %s", VERSION); BT_INFO("SCO socket layer initialized"); -- cgit v0.10.2 From e9c4bec63eac001651d6d30239dd4175cc3698ef Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:30:50 +0200 Subject: [Bluetooth] Make use of virtual devices tree The Bluetooth subsystem currently uses a platform device for devices with no parent. It is a better idea to use the new virtual devices tree for these. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index c7e079d..954eb74 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -299,11 +299,7 @@ int hci_register_sysfs(struct hci_dev *hdev) BT_DBG("%p name %s type %d", hdev, hdev->name, hdev->type); dev->class = bt_class; - - if (hdev->parent) - dev->parent = hdev->parent; - else - dev->parent = &bt_platform->dev; + dev->parent = hdev->parent; strlcpy(dev->bus_id, hdev->name, BUS_ID_SIZE); -- cgit v0.10.2 From 4c67bc74f016b0d360b8573e18969c0ff7926974 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:30:56 +0200 Subject: [Bluetooth] Support concurrent connect requests Most Bluetooth chips don't support concurrent connect requests, because this would involve a multiple baseband page with only one radio. In the case an upper layer like L2CAP requests a concurrent connect these chips return the error "Command Disallowed" for the second request. If this happens it the responsibility of the Bluetooth core to queue the request and try again after the previous connect attempt has been completed. Signed-off-by: Marcel Holtmann diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index df22efc..c0fc396 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -153,6 +153,7 @@ struct hci_conn { __u8 mode; __u8 type; __u8 out; + __u8 attempt; __u8 dev_class[3]; __u8 features[8]; __u16 interval; @@ -289,6 +290,22 @@ static inline struct hci_conn *hci_conn_hash_lookup_ba(struct hci_dev *hdev, return NULL; } +static inline struct hci_conn *hci_conn_hash_lookup_state(struct hci_dev *hdev, + __u8 type, __u16 state) +{ + struct hci_conn_hash *h = &hdev->conn_hash; + struct list_head *p; + struct hci_conn *c; + + list_for_each(p, &h->list) { + c = list_entry(p, struct hci_conn, list); + if (c->type == type && c->state == state) + return c; + } + return NULL; +} + +void hci_acl_connect(struct hci_conn *conn); void hci_acl_disconn(struct hci_conn *conn, __u8 reason); void hci_add_sco(struct hci_conn *conn, __u16 handle); diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 305a099..a91fee4 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -48,7 +48,7 @@ #define BT_DBG(D...) #endif -#define VERSION "2.10" +#define VERSION "2.11" /* Bluetooth sockets */ #define BT_MAX_PROTO 8 diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 90e3a28..6cd5711 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -51,7 +51,7 @@ #define BT_DBG(D...) #endif -static void hci_acl_connect(struct hci_conn *conn) +void hci_acl_connect(struct hci_conn *conn) { struct hci_dev *hdev = conn->hdev; struct inquiry_entry *ie; @@ -63,6 +63,8 @@ static void hci_acl_connect(struct hci_conn *conn) conn->out = 1; conn->link_mode = HCI_LM_MASTER; + conn->attempt++; + memset(&cp, 0, sizeof(cp)); bacpy(&cp.bdaddr, &conn->dst); cp.pscan_rep_mode = 0x02; @@ -80,7 +82,7 @@ static void hci_acl_connect(struct hci_conn *conn) cp.role_switch = 0x01; else cp.role_switch = 0x00; - + hci_send_cmd(hdev, OGF_LINK_CTL, OCF_CREATE_CONN, sizeof(cp), &cp); } diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index d43d0c8..65f0948 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -414,9 +414,12 @@ static inline void hci_cs_create_conn(struct hci_dev *hdev, __u8 status) if (status) { if (conn && conn->state == BT_CONNECT) { - conn->state = BT_CLOSED; - hci_proto_connect_cfm(conn, status); - hci_conn_del(conn); + if (status != 0x0c || conn->attempt > 2) { + conn->state = BT_CLOSED; + hci_proto_connect_cfm(conn, status); + hci_conn_del(conn); + } else + conn->state = BT_CONNECT2; } } else { if (!conn) { @@ -728,7 +731,7 @@ static inline void hci_conn_request_evt(struct hci_dev *hdev, struct sk_buff *sk static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) { struct hci_ev_conn_complete *ev = (struct hci_ev_conn_complete *) skb->data; - struct hci_conn *conn; + struct hci_conn *conn, *pend; BT_DBG("%s", hdev->name); @@ -801,6 +804,10 @@ static inline void hci_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *s if (ev->status) hci_conn_del(conn); + pend = hci_conn_hash_lookup_state(hdev, ACL_LINK, BT_CONNECT2); + if (pend) + hci_acl_connect(pend); + hci_dev_unlock(hdev); } -- cgit v0.10.2 From e86070c83e9110e89800274385c013db602b1444 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:31:00 +0200 Subject: [Bluetooth] Disconnect HID interrupt channel first The Bluetooth HID specification demands that the interrupt channel shall be disconnected first. This is needed to pass the qualification tests. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 03b5dad..b2d6da6 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -510,11 +510,11 @@ static int hidp_session(void *arg) if (intr_sk->sk_state != BT_CONNECTED) wait_event_timeout(*(ctrl_sk->sk_sleep), (ctrl_sk->sk_state == BT_CLOSED), HZ); - fput(session->ctrl_sock->file); + fput(session->intr_sock->file); wait_event_timeout(*(intr_sk->sk_sleep), (intr_sk->sk_state == BT_CLOSED), HZ); - fput(session->intr_sock->file); + fput(session->ctrl_sock->file); __hidp_unlink_session(session); -- cgit v0.10.2 From b2cfcd75df77b80d9cc3fa84190a350dfa79eb93 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:31:05 +0200 Subject: [Bluetooth] Fix reference count when connection lookup fails When the connection lookup for the device structure fails, the reference count for the HCI device needs to be decremented. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c index 2312d05..4d3424c 100644 --- a/net/bluetooth/bnep/core.c +++ b/net/bluetooth/bnep/core.c @@ -528,12 +528,10 @@ static struct device *bnep_get_device(struct bnep_session *session) return NULL; conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst); - if (!conn) - return NULL; hci_dev_put(hdev); - return &conn->dev; + return conn ? &conn->dev : NULL; } int bnep_add_connection(struct bnep_connadd_req *req, struct socket *sock) diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index b2d6da6..9a562cf 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -541,12 +541,10 @@ static struct device *hidp_get_device(struct hidp_session *session) return NULL; conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst); - if (!conn) - return NULL; hci_dev_put(hdev); - return &conn->dev; + return conn ? &conn->dev : NULL; } static inline void hidp_setup_input(struct hidp_session *session, struct hidp_connadd_req *req) diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index 1958ad1..5833b87 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -172,12 +172,10 @@ static struct device *rfcomm_get_device(struct rfcomm_dev *dev) return NULL; conn = hci_conn_hash_lookup_ba(hdev, ACL_LINK, &dev->dst); - if (!conn) - return NULL; hci_dev_put(hdev); - return &conn->dev; + return conn ? &conn->dev : NULL; } static int rfcomm_dev_add(struct rfcomm_dev_req *req, struct rfcomm_dlc *dlc) -- cgit v0.10.2 From cb19d9ea2ce2bcbe291d3d48e3501dc4f33ba627 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:31:10 +0200 Subject: [Bluetooth] Check if DLC is still attached to the TTY If the DLC device is no longer attached to the TTY device, then it makes no sense to go through with changing the termios settings. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index 5833b87..b8e3a5f 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -765,6 +765,9 @@ static void rfcomm_tty_set_termios(struct tty_struct *tty, struct termios *old) BT_DBG("tty %p termios %p", tty, old); + if (!dev) + return; + /* Handle turning off CRTSCTS */ if ((old->c_cflag & CRTSCTS) && !(new->c_cflag & CRTSCTS)) BT_DBG("Turning off CRTSCTS unsupported"); -- cgit v0.10.2 From 74da626a1098640ddc40c0e3481c0cd41e8ec1e9 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:31:14 +0200 Subject: [Bluetooth] Add locking for bt_proto array manipulation The bt_proto array needs to be protected by some kind of locking to prevent a race condition between bt_sock_create and bt_sock_register. And in addition all calls to sk_alloc need to be made GFP_ATOMIC now. Signed-off-by: Masatake YAMATO Signed-off-by: Frederik Deweerdt Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index a91fee4..67df99e 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -53,36 +53,51 @@ /* Bluetooth sockets */ #define BT_MAX_PROTO 8 static struct net_proto_family *bt_proto[BT_MAX_PROTO]; +static DEFINE_RWLOCK(bt_proto_lock); int bt_sock_register(int proto, struct net_proto_family *ops) { + int err = 0; + if (proto < 0 || proto >= BT_MAX_PROTO) return -EINVAL; + write_lock(&bt_proto_lock); + if (bt_proto[proto]) - return -EEXIST; + err = -EEXIST; + else + bt_proto[proto] = ops; - bt_proto[proto] = ops; - return 0; + write_unlock(&bt_proto_lock); + + return err; } EXPORT_SYMBOL(bt_sock_register); int bt_sock_unregister(int proto) { + int err = 0; + if (proto < 0 || proto >= BT_MAX_PROTO) return -EINVAL; + write_lock(&bt_proto_lock); + if (!bt_proto[proto]) - return -ENOENT; + err = -ENOENT; + else + bt_proto[proto] = NULL; - bt_proto[proto] = NULL; - return 0; + write_unlock(&bt_proto_lock); + + return err; } EXPORT_SYMBOL(bt_sock_unregister); static int bt_sock_create(struct socket *sock, int proto) { - int err = 0; + int err; if (proto < 0 || proto >= BT_MAX_PROTO) return -EINVAL; @@ -92,11 +107,18 @@ static int bt_sock_create(struct socket *sock, int proto) request_module("bt-proto-%d", proto); } #endif + err = -EPROTONOSUPPORT; + + read_lock(&bt_proto_lock); + if (bt_proto[proto] && try_module_get(bt_proto[proto]->owner)) { err = bt_proto[proto]->create(sock, proto); module_put(bt_proto[proto]->owner); } + + read_unlock(&bt_proto_lock); + return err; } diff --git a/net/bluetooth/bnep/sock.c b/net/bluetooth/bnep/sock.c index 5d9d6f1..5563db1 100644 --- a/net/bluetooth/bnep/sock.c +++ b/net/bluetooth/bnep/sock.c @@ -214,7 +214,7 @@ static int bnep_sock_create(struct socket *sock, int protocol) if (sock->type != SOCK_RAW) return -ESOCKTNOSUPPORT; - sk = sk_alloc(PF_BLUETOOTH, GFP_KERNEL, &bnep_proto, 1); + sk = sk_alloc(PF_BLUETOOTH, GFP_ATOMIC, &bnep_proto, 1); if (!sk) return -ENOMEM; diff --git a/net/bluetooth/cmtp/sock.c b/net/bluetooth/cmtp/sock.c index 0547edd..53295d3 100644 --- a/net/bluetooth/cmtp/sock.c +++ b/net/bluetooth/cmtp/sock.c @@ -205,7 +205,7 @@ static int cmtp_sock_create(struct socket *sock, int protocol) if (sock->type != SOCK_RAW) return -ESOCKTNOSUPPORT; - sk = sk_alloc(PF_BLUETOOTH, GFP_KERNEL, &cmtp_proto, 1); + sk = sk_alloc(PF_BLUETOOTH, GFP_ATOMIC, &cmtp_proto, 1); if (!sk) return -ENOMEM; diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 1a35d34..f26a9eb 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -618,7 +618,7 @@ static int hci_sock_create(struct socket *sock, int protocol) sock->ops = &hci_sock_ops; - sk = sk_alloc(PF_BLUETOOTH, GFP_KERNEL, &hci_sk_proto, 1); + sk = sk_alloc(PF_BLUETOOTH, GFP_ATOMIC, &hci_sk_proto, 1); if (!sk) return -ENOMEM; diff --git a/net/bluetooth/hidp/sock.c b/net/bluetooth/hidp/sock.c index 6242446..407fba4 100644 --- a/net/bluetooth/hidp/sock.c +++ b/net/bluetooth/hidp/sock.c @@ -256,7 +256,7 @@ static int hidp_sock_create(struct socket *sock, int protocol) if (sock->type != SOCK_RAW) return -ESOCKTNOSUPPORT; - sk = sk_alloc(PF_BLUETOOTH, GFP_KERNEL, &hidp_proto, 1); + sk = sk_alloc(PF_BLUETOOTH, GFP_ATOMIC, &hidp_proto, 1); if (!sk) return -ENOMEM; diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index 66fc621..2b3dcb8 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -559,7 +559,7 @@ static int l2cap_sock_create(struct socket *sock, int protocol) sock->ops = &l2cap_sock_ops; - sk = l2cap_sock_alloc(sock, protocol, GFP_KERNEL); + sk = l2cap_sock_alloc(sock, protocol, GFP_ATOMIC); if (!sk) return -ENOMEM; diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 530cc41..544d65b 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -336,7 +336,8 @@ static int rfcomm_sock_create(struct socket *sock, int protocol) sock->ops = &rfcomm_sock_ops; - if (!(sk = rfcomm_sock_alloc(sock, protocol, GFP_KERNEL))) + sk = rfcomm_sock_alloc(sock, protocol, GFP_ATOMIC); + if (!sk) return -ENOMEM; rfcomm_sock_init(sk, NULL); diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 14b0f69..5d13d4f 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -452,7 +452,8 @@ static int sco_sock_create(struct socket *sock, int protocol) sock->ops = &sco_sock_ops; - if (!(sk = sco_sock_alloc(sock, protocol, GFP_KERNEL))) + sk = sco_sock_alloc(sock, protocol, GFP_ATOMIC); + if (!sk) return -ENOMEM; sco_sock_init(sk, NULL); -- cgit v0.10.2 From 3f5306927d800306ebba542438cfdf1a1c418376 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 15 Oct 2006 17:31:19 +0200 Subject: [Bluetooth] Use work queue to trigger URB submission The bcm203x firmware loading driver uses a timer to trigger the URB submission. It is better to use a work queue instead. Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/bcm203x.c b/drivers/bluetooth/bcm203x.c index 67cdda4..5167517 100644 --- a/drivers/bluetooth/bcm203x.c +++ b/drivers/bluetooth/bcm203x.c @@ -29,7 +29,6 @@ #include #include #include -#include #include #include @@ -43,7 +42,7 @@ #define BT_DBG(D...) #endif -#define VERSION "1.0" +#define VERSION "1.1" static int ignore = 0; @@ -72,7 +71,7 @@ struct bcm203x_data { unsigned long state; - struct timer_list timer; + struct work_struct work; struct urb *urb; unsigned char *buffer; @@ -105,7 +104,7 @@ static void bcm203x_complete(struct urb *urb) data->state = BCM203X_SELECT_MEMORY; - mod_timer(&data->timer, jiffies + (HZ / 10)); + schedule_work(&data->work); break; case BCM203X_SELECT_MEMORY: @@ -158,9 +157,9 @@ static void bcm203x_complete(struct urb *urb) } } -static void bcm203x_timer(unsigned long user_data) +static void bcm203x_work(void *user_data) { - struct bcm203x_data *data = (struct bcm203x_data *) user_data; + struct bcm203x_data *data = user_data; if (usb_submit_urb(data->urb, GFP_ATOMIC) < 0) BT_ERR("Can't submit URB"); @@ -247,13 +246,11 @@ static int bcm203x_probe(struct usb_interface *intf, const struct usb_device_id release_firmware(firmware); - init_timer(&data->timer); - data->timer.function = bcm203x_timer; - data->timer.data = (unsigned long) data; + INIT_WORK(&data->work, bcm203x_work, (void *) data); usb_set_intfdata(intf, data); - mod_timer(&data->timer, jiffies + HZ); + schedule_work(&data->work); return 0; } -- cgit v0.10.2 From 43518407d57f1b685f5a9f1a981734ce66a21f76 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 16 Oct 2006 21:28:58 +1000 Subject: [CRYPTO] api: Select cryptomgr where needed Since cryptomgr is the only way to construct algorithm instances for now it makes sense to let the templates depend on it as otherwise it may be left off inadvertently. Signed-off-by: Herbert Xu diff --git a/crypto/Kconfig b/crypto/Kconfig index 1e2f39c..cbae839 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -27,7 +27,6 @@ config CRYPTO_HASH config CRYPTO_MANAGER tristate "Cryptographic algorithm manager" select CRYPTO_ALGAPI - default m help Create default cryptographic template instantiations such as cbc(aes). @@ -35,6 +34,7 @@ config CRYPTO_MANAGER config CRYPTO_HMAC tristate "HMAC support" select CRYPTO_HASH + select CRYPTO_MANAGER help HMAC: Keyed-Hashing for Message Authentication (RFC2104). This is required for IPSec. @@ -131,6 +131,7 @@ config CRYPTO_TGR192 config CRYPTO_ECB tristate "ECB support" select CRYPTO_BLKCIPHER + select CRYPTO_MANAGER default m help ECB: Electronic CodeBook mode @@ -140,6 +141,7 @@ config CRYPTO_ECB config CRYPTO_CBC tristate "CBC support" select CRYPTO_BLKCIPHER + select CRYPTO_MANAGER default m help CBC: Cipher Block Chaining mode -- cgit v0.10.2 From 9d90dafdb1f0e3c2b69fa8d3fbe99649127c8fa4 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:28:44 +0100 Subject: [PATCH] rio: fix array checking Found by an analysis tool and reported to the list. Fix is simple enough Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds diff --git a/drivers/char/rio/rioctrl.c b/drivers/char/rio/rioctrl.c index 052e812..7ce7761 100644 --- a/drivers/char/rio/rioctrl.c +++ b/drivers/char/rio/rioctrl.c @@ -662,7 +662,7 @@ int riocontrol(struct rio_info *p, dev_t dev, int cmd, unsigned long arg, int su p->RIOError.Error = COPYIN_FAILED; return -EFAULT; } - if (portStats.port >= RIO_PORTS) { + if (portStats.port < 0 || portStats.port >= RIO_PORTS) { p->RIOError.Error = PORT_NUMBER_OUT_OF_RANGE; return -ENXIO; } @@ -702,7 +702,7 @@ int riocontrol(struct rio_info *p, dev_t dev, int cmd, unsigned long arg, int su p->RIOError.Error = COPYIN_FAILED; return -EFAULT; } - if (portStats.port >= RIO_PORTS) { + if (portStats.port < 0 || portStats.port >= RIO_PORTS) { p->RIOError.Error = PORT_NUMBER_OUT_OF_RANGE; return -ENXIO; } -- cgit v0.10.2 From 3a42bb223f61fbd755d6e61b9b50b9681d68fcae Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:31:02 +0100 Subject: [PATCH] ide: add sanity checking to ide taskfile ioctl Without this the user can feed in bogus values and get very bogus results. Security impact is minimal as this ioctl isn't available to unpriviledged processes anyway. Reported to the l/k list and found with an auditing tool. Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c index 1d0470c..30175c7 100644 --- a/drivers/ide/ide-taskfile.c +++ b/drivers/ide/ide-taskfile.c @@ -524,8 +524,8 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg) task_ioreg_t *hobsptr = args.hobRegister; int err = 0; int tasksize = sizeof(struct ide_task_request_s); - int taskin = 0; - int taskout = 0; + unsigned int taskin = 0; + unsigned int taskout = 0; u8 io_32bit = drive->io_32bit; char __user *buf = (char __user *)arg; @@ -538,8 +538,13 @@ int ide_taskfile_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg) return -EFAULT; } - taskout = (int) req_task->out_size; - taskin = (int) req_task->in_size; + taskout = req_task->out_size; + taskin = req_task->in_size; + + if (taskin > 65536 || taskout > 65536) { + err = -EINVAL; + goto abort; + } if (taskout) { int outtotal = tasksize; -- cgit v0.10.2 From 8741ca71a3f626a56595b88200ebf952ce77ceef Mon Sep 17 00:00:00 2001 From: Andrey Mirkin Date: Mon, 16 Oct 2006 12:08:43 +0400 Subject: [PATCH] scsi: megaraid_{mm,mbox}: 64-bit DMA capability fix It is known that 2 LSI Logic MegaRAID SATA RAID Controllers (150-4 and 150-6) don't support 64-bit DMA. Unfortunately currently this check is wrong and driver sets 64-bit DMA mode for these devices. Signed-off-by: Andrey Mirkin Acked-by: Vasily Averin Signed-off-by: Linus Torvalds diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c index c0edb66..7bac86d 100644 --- a/drivers/scsi/megaraid/megaraid_mbox.c +++ b/drivers/scsi/megaraid/megaraid_mbox.c @@ -884,7 +884,7 @@ megaraid_init_mbox(adapter_t *adapter) if (((magic64 == HBA_SIGNATURE_64_BIT) && ((adapter->pdev->subsystem_device != - PCI_SUBSYS_ID_MEGARAID_SATA_150_6) || + PCI_SUBSYS_ID_MEGARAID_SATA_150_6) && (adapter->pdev->subsystem_device != PCI_SUBSYS_ID_MEGARAID_SATA_150_4))) || (adapter->pdev->vendor == PCI_VENDOR_ID_LSI_LOGIC && -- cgit v0.10.2 From 29da7eb0ec69245c6e9b4eb5bdaa04af685f5c4f Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 16 Oct 2006 14:10:49 +0100 Subject: [PATCH] FRV: Use the correct preemption primitives in kmap_atomic() and co Use inc/dec_preempt_count() rather than preempt_enable/disable() and manually add in the compiler barriers that were provided by the latter. This makes FRV consistent with other archs. Furthermore, the compiler barrier effects are now there unconditionally - at least as far as preemption is concerned - because we don't want the compiler moving memory accesses out of the section of code in which the mapping is in force - in effect the kmap_atomic() must imply a LOCK-class barrier and the kunmap_atomic() must imply an UNLOCK-class barrier to the compiler. Signed-off-by: David Howells Acked-by: Peter Zijlstra Signed-off-by: Linus Torvalds diff --git a/include/asm-frv/highmem.h b/include/asm-frv/highmem.h index e2247c2..0f390f4 100644 --- a/include/asm-frv/highmem.h +++ b/include/asm-frv/highmem.h @@ -82,11 +82,11 @@ extern struct page *kmap_atomic_to_page(void *ptr); dampr = paddr | xAMPRx_L | xAMPRx_M | xAMPRx_S | xAMPRx_SS_16Kb | xAMPRx_V; \ \ if (type != __KM_CACHE) \ - asm volatile("movgs %0,dampr"#ampr :: "r"(dampr)); \ + asm volatile("movgs %0,dampr"#ampr :: "r"(dampr) : "memory"); \ else \ asm volatile("movgs %0,iampr"#ampr"\n" \ "movgs %0,dampr"#ampr"\n" \ - :: "r"(dampr) \ + :: "r"(dampr) : "memory" \ ); \ \ asm("movsg damlr"#ampr",%0" : "=r"(damlr)); \ @@ -104,7 +104,7 @@ extern struct page *kmap_atomic_to_page(void *ptr); asm volatile("movgs %0,tplr \n" \ "movgs %1,tppr \n" \ "tlbpr %0,gr0,#2,#1" \ - : : "r"(damlr), "r"(dampr)); \ + : : "r"(damlr), "r"(dampr) : "memory"); \ \ /*printk("TLB: SECN sl=%d L=%08lx P=%08lx\n", slot, damlr, dampr);*/ \ \ @@ -115,7 +115,7 @@ static inline void *kmap_atomic(struct page *page, enum km_type type) { unsigned long paddr; - preempt_disable(); + inc_preempt_count(); paddr = page_to_phys(page); switch (type) { @@ -138,16 +138,16 @@ static inline void *kmap_atomic(struct page *page, enum km_type type) } } -#define __kunmap_atomic_primary(type, ampr) \ -do { \ - asm volatile("movgs gr0,dampr"#ampr"\n"); \ - if (type == __KM_CACHE) \ - asm volatile("movgs gr0,iampr"#ampr"\n"); \ +#define __kunmap_atomic_primary(type, ampr) \ +do { \ + asm volatile("movgs gr0,dampr"#ampr"\n" ::: "memory"); \ + if (type == __KM_CACHE) \ + asm volatile("movgs gr0,iampr"#ampr"\n" ::: "memory"); \ } while(0) -#define __kunmap_atomic_secondary(slot, vaddr) \ -do { \ - asm volatile("tlbpr %0,gr0,#4,#1" : : "r"(vaddr)); \ +#define __kunmap_atomic_secondary(slot, vaddr) \ +do { \ + asm volatile("tlbpr %0,gr0,#4,#1" : : "r"(vaddr) : "memory"); \ } while(0) static inline void kunmap_atomic(void *kvaddr, enum km_type type) @@ -170,7 +170,8 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) default: BUG(); } - preempt_enable(); + dec_preempt_count(); + preempt_check_resched(); } #endif /* !__ASSEMBLY__ */ -- cgit v0.10.2 From 39af114377bf80d2a88e47be33d578d1fa9b0775 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 16 Oct 2006 09:01:46 -0700 Subject: [PATCH] fix epoll_pwait when EPOLL=n Fixes http://bugzilla.kernel.org/show_bug.cgi?id=7371 sys_epoll_pwait needs to be listed as a conditional (weak) entry point for CONFIG_EPOLL=n. Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 7a3b2e75..0e53314 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -49,6 +49,7 @@ cond_syscall(compat_sys_get_robust_list); cond_syscall(sys_epoll_create); cond_syscall(sys_epoll_ctl); cond_syscall(sys_epoll_wait); +cond_syscall(sys_epoll_pwait); cond_syscall(sys_semget); cond_syscall(sys_semop); cond_syscall(sys_semtimedop); -- cgit v0.10.2 From d986a27413aad10574f7211524de6a529870d134 Mon Sep 17 00:00:00 2001 From: Henrik Kretzschmar Date: Tue, 10 Oct 2006 14:26:01 -0700 Subject: RDMA/amso1100: pci_module_init() conversion pci_module_init() convertion in amso1100 driver. Signed-off-by: Henrik Kretzschmar Signed-off-by: Andrew Morton Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/amso1100/c2.c b/drivers/infiniband/hw/amso1100/c2.c index dc1ebea..9e7bd94 100644 --- a/drivers/infiniband/hw/amso1100/c2.c +++ b/drivers/infiniband/hw/amso1100/c2.c @@ -1243,7 +1243,7 @@ static struct pci_driver c2_pci_driver = { static int __init c2_init_module(void) { - return pci_module_init(&c2_pci_driver); + return pci_register_driver(&c2_pci_driver); } static void __exit c2_exit_module(void) -- cgit v0.10.2 From fb7711e71ea7cd0d3e77e969df59162388c8a1f9 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 10 Oct 2006 14:26:02 -0700 Subject: RDMA/amso1100: Fix a NULL dereference in error path This patch fixes a NULL dereference spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Acked-by: Steve Wise Acked-by: Tom Tucker Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c index e37c568..30409e1 100644 --- a/drivers/infiniband/hw/amso1100/c2_rnic.c +++ b/drivers/infiniband/hw/amso1100/c2_rnic.c @@ -150,8 +150,8 @@ static int c2_rnic_query(struct c2_dev *c2dev, struct ib_device_attr *props) (struct c2wr_rnic_query_rep *) (unsigned long) (vq_req->reply_msg); if (!reply) err = -ENOMEM; - - err = c2_errno(reply); + else + err = c2_errno(reply); if (err) goto bail2; -- cgit v0.10.2 From 6ef93dddfe11a72ab98a37ac4ef20ad681b008b0 Mon Sep 17 00:00:00 2001 From: Robert Walsh Date: Tue, 10 Oct 2006 14:55:45 -0700 Subject: IB/ipath: Initialize diagpkt file on device init only Don't attempt to set up the diagpkt device in the module init code. Instead, wait until a piece of hardware is initialized. Fixes a problem when loading the ib_ipath module when no InfiniPath hardware is present: modprobe would go into the D state and stay there. Signed-off-by: Robert Walsh Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/ipath/ipath_diag.c b/drivers/infiniband/hw/ipath/ipath_diag.c index 29958b6..28c087b 100644 --- a/drivers/infiniband/hw/ipath/ipath_diag.c +++ b/drivers/infiniband/hw/ipath/ipath_diag.c @@ -67,19 +67,54 @@ static struct file_operations diag_file_ops = { .release = ipath_diag_release }; +static ssize_t ipath_diagpkt_write(struct file *fp, + const char __user *data, + size_t count, loff_t *off); + +static struct file_operations diagpkt_file_ops = { + .owner = THIS_MODULE, + .write = ipath_diagpkt_write, +}; + +static atomic_t diagpkt_count = ATOMIC_INIT(0); +static struct cdev *diagpkt_cdev; +static struct class_device *diagpkt_class_dev; + int ipath_diag_add(struct ipath_devdata *dd) { char name[16]; + int ret = 0; + + if (atomic_inc_return(&diagpkt_count) == 1) { + ret = ipath_cdev_init(IPATH_DIAGPKT_MINOR, + "ipath_diagpkt", &diagpkt_file_ops, + &diagpkt_cdev, &diagpkt_class_dev); + + if (ret) { + ipath_dev_err(dd, "Couldn't create ipath_diagpkt " + "device: %d", ret); + goto done; + } + } snprintf(name, sizeof(name), "ipath_diag%d", dd->ipath_unit); - return ipath_cdev_init(IPATH_DIAG_MINOR_BASE + dd->ipath_unit, name, - &diag_file_ops, &dd->diag_cdev, - &dd->diag_class_dev); + ret = ipath_cdev_init(IPATH_DIAG_MINOR_BASE + dd->ipath_unit, name, + &diag_file_ops, &dd->diag_cdev, + &dd->diag_class_dev); + if (ret) + ipath_dev_err(dd, "Couldn't create %s device: %d", + name, ret); + +done: + return ret; } void ipath_diag_remove(struct ipath_devdata *dd) { + if (atomic_dec_and_test(&diagpkt_count)) + ipath_cdev_cleanup(&diagpkt_cdev, &diagpkt_class_dev); + ipath_cdev_cleanup(&dd->diag_cdev, &dd->diag_class_dev); } @@ -275,30 +310,6 @@ bail: return ret; } -static ssize_t ipath_diagpkt_write(struct file *fp, - const char __user *data, - size_t count, loff_t *off); - -static struct file_operations diagpkt_file_ops = { - .owner = THIS_MODULE, - .write = ipath_diagpkt_write, -}; - -static struct cdev *diagpkt_cdev; -static struct class_device *diagpkt_class_dev; - -int __init ipath_diagpkt_add(void) -{ - return ipath_cdev_init(IPATH_DIAGPKT_MINOR, - "ipath_diagpkt", &diagpkt_file_ops, - &diagpkt_cdev, &diagpkt_class_dev); -} - -void __exit ipath_diagpkt_remove(void) -{ - ipath_cdev_cleanup(&diagpkt_cdev, &diagpkt_class_dev); -} - /** * ipath_diagpkt_write - write an IB packet * @fp: the diag data device file pointer diff --git a/drivers/infiniband/hw/ipath/ipath_driver.c b/drivers/infiniband/hw/ipath/ipath_driver.c index 12cefa6..b4ffaa7 100644 --- a/drivers/infiniband/hw/ipath/ipath_driver.c +++ b/drivers/infiniband/hw/ipath/ipath_driver.c @@ -2005,18 +2005,8 @@ static int __init infinipath_init(void) goto bail_group; } - ret = ipath_diagpkt_add(); - if (ret < 0) { - printk(KERN_ERR IPATH_DRV_NAME ": Unable to create " - "diag data device: error %d\n", -ret); - goto bail_ipathfs; - } - goto bail; -bail_ipathfs: - ipath_exit_ipathfs(); - bail_group: ipath_driver_remove_group(&ipath_driver.driver); diff --git a/drivers/infiniband/hw/ipath/ipath_kernel.h b/drivers/infiniband/hw/ipath/ipath_kernel.h index 7c43669..06d5020 100644 --- a/drivers/infiniband/hw/ipath/ipath_kernel.h +++ b/drivers/infiniband/hw/ipath/ipath_kernel.h @@ -869,9 +869,6 @@ int ipath_device_create_group(struct device *, struct ipath_devdata *); void ipath_device_remove_group(struct device *, struct ipath_devdata *); int ipath_expose_reset(struct device *); -int ipath_diagpkt_add(void); -void ipath_diagpkt_remove(void); - int ipath_init_ipathfs(void); void ipath_exit_ipathfs(void); int ipathfs_add_device(struct ipath_devdata *); -- cgit v0.10.2 From 4d99bfac9d5ce53b383d3c8279b917050be4e06c Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 16 Oct 2006 19:59:43 +0200 Subject: [PATCH] CONFIG_TELCLOCK depends on X86 The telecom clock driver for MPBL0010 ATCA SBC depends on X86 Signed-off-by: Geert Uytterhoeven Acked-by: Mark Gross Signed-off-by: Linus Torvalds diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 0e6f35f..39a9f8c 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -1046,7 +1046,7 @@ source "drivers/char/tpm/Kconfig" config TELCLOCK tristate "Telecom clock driver for MPBL0010 ATCA SBC" - depends on EXPERIMENTAL + depends on EXPERIMENTAL && X86 default n help The telecom clock device is specific to the MPBL0010 ATCA computer and -- cgit v0.10.2 From 0f6f65f607b6d516fa001e8cdf5a2618c81372f5 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 16 Oct 2006 21:14:51 +0100 Subject: [ARM] Update mach-types Signed-off-by: Russell King diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types index b02af1d..579c69a 100644 --- a/arch/arm/tools/mach-types +++ b/arch/arm/tools/mach-types @@ -4,7 +4,7 @@ # # Up to date versions of this file can be obtained from: # -# http://www.arm.linux.org.uk/developer/machines/?action=download +# http://www.arm.linux.org.uk/developer/machines/download.php # # Please do not send patches to this file; it is automatically generated! # To add an entry into this database, please see Documentation/arm/README, @@ -12,7 +12,7 @@ # # http://www.arm.linux.org.uk/developer/machines/?action=new # -# Last update: Sat Sep 23 13:20:43 2006 +# Last update: Mon Oct 16 21:13:36 2006 # # machine_is_xxx CONFIG_xxxx MACH_TYPE_xxx number # @@ -1157,3 +1157,17 @@ adsturboxb MACH_ADSTURBOXB ADSTURBOXB 1143 oti4110 MACH_OTI4110 OTI4110 1144 hme_pxa MACH_HME_PXA HME_PXA 1145 deisterdca MACH_DEISTERDCA DEISTERDCA 1146 +ces_ssem2 MACH_CES_SSEM2 CES_SSEM2 1147 +ces_mtr MACH_CES_MTR CES_MTR 1148 +tds_avng_sbc MACH_TDS_AVNG_SBC TDS_AVNG_SBC 1149 +everest MACH_EVEREST EVEREST 1150 +pnx4010 MACH_PNX4010 PNX4010 1151 +oxnas MACH_OXNAS OXNAS 1152 +fiori MACH_FIORI FIORI 1153 +ml1200 MACH_ML1200 ML1200 1154 +cactus MACH_CACTUS CACTUS 1155 +nb2xxx MACH_NB2XXX NB2XXX 1156 +hw6900 MACH_HW6900 HW6900 1157 +cdcs_quoll MACH_CDCS_QUOLL CDCS_QUOLL 1158 +quicksilver MACH_QUICKSILVER QUICKSILVER 1159 +uplat926 MACH_UPLAT926 UPLAT926 1160 -- cgit v0.10.2 From 3693ec670b3bb4d11295856bea3592dd8f37f9a5 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Tue, 26 Sep 2006 13:22:41 -0500 Subject: [PATCH] bcm43xx: fix race condition in periodic work handler There is a potential race condition in the periodic_work_handler routine of bcm43xx-softmac. In addition to fixing this condition, the size of code is reduced by moving the mutex lock outside the if. Signed-off-by: Larry Finger Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index bad3452..0f047d4 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -3164,12 +3164,12 @@ static void bcm43xx_periodic_work_handler(void *d) u32 savedirqs = 0; int badness; + mutex_lock(&bcm->mutex); badness = estimate_periodic_work_badness(bcm->periodic_state); if (badness > BADNESS_LIMIT) { /* Periodic work will take a long time, so we want it to * be preemtible. */ - mutex_lock(&bcm->mutex); netif_tx_disable(bcm->net_dev); spin_lock_irqsave(&bcm->irq_lock, flags); bcm43xx_mac_suspend(bcm); @@ -3182,7 +3182,6 @@ static void bcm43xx_periodic_work_handler(void *d) /* Periodic work should take short time, so we want low * locking overhead. */ - mutex_lock(&bcm->mutex); spin_lock_irqsave(&bcm->irq_lock, flags); } -- cgit v0.10.2 From 7c28ad2d83ecc637237fe684659a6afbce0bb2a8 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Wed, 27 Sep 2006 15:26:33 +0300 Subject: [PATCH] softmac: Fix WX and association related races This fixes some race conditions in the WirelessExtension handling and association handling code. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_leds.c b/drivers/net/wireless/bcm43xx/bcm43xx_leds.c index c3f90c8..2ddbec6 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_leds.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_leds.c @@ -242,7 +242,7 @@ void bcm43xx_leds_update(struct bcm43xx_private *bcm, int activity) //TODO break; case BCM43xx_LED_ASSOC: - if (bcm->softmac->associated) + if (bcm->softmac->associnfo.associated) turn_on = 1; break; #ifdef CONFIG_BCM43XX_DEBUG diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c index 9b7b15c..d27016f 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c @@ -847,7 +847,7 @@ static struct iw_statistics *bcm43xx_get_wireless_stats(struct net_device *net_d unsigned long flags; wstats = &bcm->stats.wstats; - if (!mac->associated) { + if (!mac->associnfo.associated) { wstats->miss.beacon = 0; // bcm->ieee->ieee_stats.tx_retry_limit_exceeded = 0; // FIXME: should this be cleared here? wstats->discard.retries = 0; diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h index 425b3a5..617b672 100644 --- a/include/net/ieee80211softmac.h +++ b/include/net/ieee80211softmac.h @@ -63,13 +63,11 @@ struct ieee80211softmac_wpa { /* * Information about association - * - * Do we need a lock for this? - * We only ever use this structure inlined - * into our global struct. I've used its lock, - * but maybe we need a local one here? */ struct ieee80211softmac_assoc_info { + + struct mutex mutex; + /* * This is the requested ESSID. It is written * only by the WX handlers. @@ -99,12 +97,13 @@ struct ieee80211softmac_assoc_info { * * bssfixed is used for SIOCSIWAP. */ - u8 static_essid:1, - short_preamble_available:1, - associating:1, - assoc_wait:1, - bssvalid:1, - bssfixed:1; + u8 static_essid; + u8 short_preamble_available; + u8 associating; + u8 associated; + u8 assoc_wait; + u8 bssvalid; + u8 bssfixed; /* Scan retries remaining */ int scan_retry; @@ -229,12 +228,10 @@ struct ieee80211softmac_device { /* private stuff follows */ /* this lock protects this structure */ spinlock_t lock; - - /* couple of flags */ - u8 scanning:1, /* protects scanning from being done multiple times at once */ - associated:1, - running:1; - + + u8 running; /* SoftMAC started? */ + u8 scanning; + struct ieee80211softmac_scaninfo *scaninfo; struct ieee80211softmac_assoc_info associnfo; struct ieee80211softmac_bss_info bssinfo; @@ -250,7 +247,7 @@ struct ieee80211softmac_device { /* we need to keep a list of network structs we copied */ struct list_head network_list; - + /* This must be the last item so that it points to the data * allocated beyond this structure by alloc_ieee80211 */ u8 priv[0]; @@ -295,7 +292,7 @@ static inline u8 ieee80211softmac_suggest_txrate(struct ieee80211softmac_device { struct ieee80211softmac_txrates *txrates = &mac->txrates; - if (!mac->associated) + if (!mac->associnfo.associated) return txrates->mgt_mcast_rate; /* We are associated, sending unicast frame */ diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c index 589f6d2..cf51c87 100644 --- a/net/ieee80211/softmac/ieee80211softmac_assoc.c +++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c @@ -48,7 +48,7 @@ ieee80211softmac_assoc(struct ieee80211softmac_device *mac, struct ieee80211soft dprintk(KERN_INFO PFX "sent association request!\n"); spin_lock_irqsave(&mac->lock, flags); - mac->associated = 0; /* just to make sure */ + mac->associnfo.associated = 0; /* just to make sure */ /* Set a timer for timeout */ /* FIXME: make timeout configurable */ @@ -62,24 +62,22 @@ ieee80211softmac_assoc_timeout(void *d) { struct ieee80211softmac_device *mac = (struct ieee80211softmac_device *)d; struct ieee80211softmac_network *n; - unsigned long flags; - spin_lock_irqsave(&mac->lock, flags); + mutex_lock(&mac->associnfo.mutex); /* we might race against ieee80211softmac_handle_assoc_response, * so make sure only one of us does something */ - if (!mac->associnfo.associating) { - spin_unlock_irqrestore(&mac->lock, flags); - return; - } + if (!mac->associnfo.associating) + goto out; mac->associnfo.associating = 0; mac->associnfo.bssvalid = 0; - mac->associated = 0; + mac->associnfo.associated = 0; n = ieee80211softmac_get_network_by_bssid_locked(mac, mac->associnfo.bssid); - spin_unlock_irqrestore(&mac->lock, flags); dprintk(KERN_INFO PFX "assoc request timed out!\n"); ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT, n); +out: + mutex_unlock(&mac->associnfo.mutex); } void @@ -93,7 +91,7 @@ ieee80211softmac_disassoc(struct ieee80211softmac_device *mac) netif_carrier_off(mac->dev); - mac->associated = 0; + mac->associnfo.associated = 0; mac->associnfo.bssvalid = 0; mac->associnfo.associating = 0; ieee80211softmac_init_bss(mac); @@ -107,7 +105,7 @@ ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reas { struct ieee80211softmac_network *found; - if (mac->associnfo.bssvalid && mac->associated) { + if (mac->associnfo.bssvalid && mac->associnfo.associated) { found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid); if (found) ieee80211softmac_send_mgt_frame(mac, found, IEEE80211_STYPE_DISASSOC, reason); @@ -196,17 +194,18 @@ ieee80211softmac_assoc_work(void *d) int bssvalid; unsigned long flags; + mutex_lock(&mac->associnfo.mutex); + + if (!mac->associnfo.associating) + goto out; + /* ieee80211_disassoc might clear this */ bssvalid = mac->associnfo.bssvalid; /* meh */ - if (mac->associated) + if (mac->associnfo.associated) ieee80211softmac_send_disassoc_req(mac, WLAN_REASON_DISASSOC_STA_HAS_LEFT); - spin_lock_irqsave(&mac->lock, flags); - mac->associnfo.associating = 1; - spin_unlock_irqrestore(&mac->lock, flags); - /* try to find the requested network in our list, if we found one already */ if (bssvalid || mac->associnfo.bssfixed) found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid); @@ -260,10 +259,8 @@ ieee80211softmac_assoc_work(void *d) if (!found) { if (mac->associnfo.scan_retry > 0) { - spin_lock_irqsave(&mac->lock, flags); mac->associnfo.scan_retry--; - spin_unlock_irqrestore(&mac->lock, flags); - + /* We know of no such network. Let's scan. * NB: this also happens if we had no memory to copy the network info... * Maybe we can hope to have more memory after scanning finishes ;) @@ -272,19 +269,17 @@ ieee80211softmac_assoc_work(void *d) ieee80211softmac_notify(mac->dev, IEEE80211SOFTMAC_EVENT_SCAN_FINISHED, ieee80211softmac_assoc_notify_scan, NULL); if (ieee80211softmac_start_scan(mac)) dprintk(KERN_INFO PFX "Associate: failed to initiate scan. Is device up?\n"); - return; + goto out; } else { - spin_lock_irqsave(&mac->lock, flags); mac->associnfo.associating = 0; - mac->associated = 0; - spin_unlock_irqrestore(&mac->lock, flags); + mac->associnfo.associated = 0; dprintk(KERN_INFO PFX "Unable to find matching network after scan!\n"); /* reset the retry counter for the next user request since we * break out and don't reschedule ourselves after this point. */ mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT; ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND, NULL); - return; + goto out; } } @@ -297,7 +292,7 @@ ieee80211softmac_assoc_work(void *d) /* copy the ESSID for displaying it */ mac->associnfo.associate_essid.len = found->essid.len; memcpy(mac->associnfo.associate_essid.data, found->essid.data, IW_ESSID_MAX_SIZE + 1); - + /* we found a network! authenticate (if necessary) and associate to it. */ if (found->authenticating) { dprintk(KERN_INFO PFX "Already requested authentication, waiting...\n"); @@ -305,7 +300,7 @@ ieee80211softmac_assoc_work(void *d) mac->associnfo.assoc_wait = 1; ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL); } - return; + goto out; } if (!found->authenticated && !found->authenticating) { /* This relies on the fact that _auth_req only queues the work, @@ -321,11 +316,14 @@ ieee80211softmac_assoc_work(void *d) mac->associnfo.assoc_wait = 0; ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, found); } - return; + goto out; } /* finally! now we can start associating */ mac->associnfo.assoc_wait = 0; ieee80211softmac_assoc(mac, found); + +out: + mutex_unlock(&mac->associnfo.mutex); } /* call this to do whatever is necessary when we're associated */ @@ -341,7 +339,7 @@ ieee80211softmac_associated(struct ieee80211softmac_device *mac, mac->bssinfo.supported_rates = net->supported_rates; ieee80211softmac_recalc_txrates(mac); - mac->associated = 1; + mac->associnfo.associated = 1; mac->associnfo.short_preamble_available = (cap & WLAN_CAPABILITY_SHORT_PREAMBLE) != 0; @@ -421,7 +419,7 @@ ieee80211softmac_handle_assoc_response(struct net_device * dev, dprintk(KERN_INFO PFX "associating failed (reason: 0x%x)!\n", status); mac->associnfo.associating = 0; mac->associnfo.bssvalid = 0; - mac->associated = 0; + mac->associnfo.associated = 0; ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, network); } diff --git a/net/ieee80211/softmac/ieee80211softmac_io.c b/net/ieee80211/softmac/ieee80211softmac_io.c index 82bfddb..e8419cf 100644 --- a/net/ieee80211/softmac/ieee80211softmac_io.c +++ b/net/ieee80211/softmac/ieee80211softmac_io.c @@ -475,8 +475,13 @@ int ieee80211softmac_handle_beacon(struct net_device *dev, { struct ieee80211softmac_device *mac = ieee80211_priv(dev); - if (mac->associated && memcmp(network->bssid, mac->associnfo.bssid, ETH_ALEN) == 0) - ieee80211softmac_process_erp(mac, network->erp_value); + /* This might race, but we don't really care and it's not worth + * adding heavyweight locking in this fastpath. + */ + if (mac->associnfo.associated) { + if (memcmp(network->bssid, mac->associnfo.bssid, ETH_ALEN) == 0) + ieee80211softmac_process_erp(mac, network->erp_value); + } return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c index addea1c..33aff4f 100644 --- a/net/ieee80211/softmac/ieee80211softmac_module.c +++ b/net/ieee80211/softmac/ieee80211softmac_module.c @@ -57,6 +57,7 @@ struct net_device *alloc_ieee80211softmac(int sizeof_priv) INIT_LIST_HEAD(&softmac->network_list); INIT_LIST_HEAD(&softmac->events); + mutex_init(&softmac->associnfo.mutex); INIT_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work, softmac); INIT_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout, softmac); softmac->start_scan = ieee80211softmac_start_scan_implementation; diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c index 2aa779d..23068a8 100644 --- a/net/ieee80211/softmac/ieee80211softmac_wx.c +++ b/net/ieee80211/softmac/ieee80211softmac_wx.c @@ -73,13 +73,14 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev, struct ieee80211softmac_network *n; struct ieee80211softmac_auth_queue_item *authptr; int length = 0; - unsigned long flags; + + mutex_lock(&sm->associnfo.mutex); /* Check if we're already associating to this or another network * If it's another network, cancel and start over with our new network * If it's our network, ignore the change, we're already doing it! */ - if((sm->associnfo.associating || sm->associated) && + if((sm->associnfo.associating || sm->associnfo.associated) && (data->essid.flags && data->essid.length)) { /* Get the associating network */ n = ieee80211softmac_get_network_by_bssid(sm, sm->associnfo.bssid); @@ -87,10 +88,9 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev, !memcmp(n->essid.data, extra, n->essid.len)) { dprintk(KERN_INFO PFX "Already associating or associated to "MAC_FMT"\n", MAC_ARG(sm->associnfo.bssid)); - return 0; + goto out; } else { dprintk(KERN_INFO PFX "Canceling existing associate request!\n"); - spin_lock_irqsave(&sm->lock,flags); /* Cancel assoc work */ cancel_delayed_work(&sm->associnfo.work); /* We don't have to do this, but it's a little cleaner */ @@ -98,14 +98,13 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev, cancel_delayed_work(&authptr->work); sm->associnfo.bssvalid = 0; sm->associnfo.bssfixed = 0; - spin_unlock_irqrestore(&sm->lock,flags); flush_scheduled_work(); + sm->associnfo.associating = 0; + sm->associnfo.associated = 0; } } - spin_lock_irqsave(&sm->lock, flags); - sm->associnfo.static_essid = 0; sm->associnfo.assoc_wait = 0; @@ -121,10 +120,12 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev, * If applicable, we have already copied the data in */ sm->associnfo.req_essid.len = length; + sm->associnfo.associating = 1; /* queue lower level code to do work (if necessary) */ schedule_work(&sm->associnfo.work); +out: + mutex_unlock(&sm->associnfo.mutex); - spin_unlock_irqrestore(&sm->lock, flags); return 0; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_essid); @@ -136,10 +137,8 @@ ieee80211softmac_wx_get_essid(struct net_device *net_dev, char *extra) { struct ieee80211softmac_device *sm = ieee80211_priv(net_dev); - unsigned long flags; - /* avoid getting inconsistent information */ - spin_lock_irqsave(&sm->lock, flags); + mutex_lock(&sm->associnfo.mutex); /* If all fails, return ANY (empty) */ data->essid.length = 0; data->essid.flags = 0; /* active */ @@ -152,12 +151,13 @@ ieee80211softmac_wx_get_essid(struct net_device *net_dev, } /* If we're associating/associated, return that */ - if (sm->associated || sm->associnfo.associating) { + if (sm->associnfo.associated || sm->associnfo.associating) { data->essid.length = sm->associnfo.associate_essid.len; data->essid.flags = 1; /* active */ memcpy(extra, sm->associnfo.associate_essid.data, sm->associnfo.associate_essid.len); } - spin_unlock_irqrestore(&sm->lock, flags); + mutex_unlock(&sm->associnfo.mutex); + return 0; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_essid); @@ -322,15 +322,15 @@ ieee80211softmac_wx_get_wap(struct net_device *net_dev, { struct ieee80211softmac_device *mac = ieee80211_priv(net_dev); int err = 0; - unsigned long flags; - spin_lock_irqsave(&mac->lock, flags); + mutex_lock(&mac->associnfo.mutex); if (mac->associnfo.bssvalid) memcpy(data->ap_addr.sa_data, mac->associnfo.bssid, ETH_ALEN); else memset(data->ap_addr.sa_data, 0xff, ETH_ALEN); data->ap_addr.sa_family = ARPHRD_ETHER; - spin_unlock_irqrestore(&mac->lock, flags); + mutex_unlock(&mac->associnfo.mutex); + return err; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_wap); @@ -342,28 +342,27 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev, char *extra) { struct ieee80211softmac_device *mac = ieee80211_priv(net_dev); - unsigned long flags; /* sanity check */ if (data->ap_addr.sa_family != ARPHRD_ETHER) { return -EINVAL; } - spin_lock_irqsave(&mac->lock, flags); + mutex_lock(&mac->associnfo.mutex); if (is_broadcast_ether_addr(data->ap_addr.sa_data)) { /* the bssid we have is not to be fixed any longer, * and we should reassociate to the best AP. */ mac->associnfo.bssfixed = 0; /* force reassociation */ mac->associnfo.bssvalid = 0; - if (mac->associated) + if (mac->associnfo.associated) schedule_work(&mac->associnfo.work); } else if (is_zero_ether_addr(data->ap_addr.sa_data)) { /* the bssid we have is no longer fixed */ mac->associnfo.bssfixed = 0; } else { if (!memcmp(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN)) { - if (mac->associnfo.associating || mac->associated) { + if (mac->associnfo.associating || mac->associnfo.associated) { /* bssid unchanged and associated or associating - just return */ goto out; } @@ -378,7 +377,8 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev, } out: - spin_unlock_irqrestore(&mac->lock, flags); + mutex_unlock(&mac->associnfo.mutex); + return 0; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_wap); @@ -394,7 +394,8 @@ ieee80211softmac_wx_set_genie(struct net_device *dev, int err = 0; char *buf; int i; - + + mutex_lock(&mac->associnfo.mutex); spin_lock_irqsave(&mac->lock, flags); /* bleh. shouldn't be locked for that kmalloc... */ @@ -432,6 +433,8 @@ ieee80211softmac_wx_set_genie(struct net_device *dev, out: spin_unlock_irqrestore(&mac->lock, flags); + mutex_unlock(&mac->associnfo.mutex); + return err; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_genie); @@ -446,7 +449,8 @@ ieee80211softmac_wx_get_genie(struct net_device *dev, unsigned long flags; int err = 0; int space = wrqu->data.length; - + + mutex_lock(&mac->associnfo.mutex); spin_lock_irqsave(&mac->lock, flags); wrqu->data.length = 0; @@ -459,6 +463,8 @@ ieee80211softmac_wx_get_genie(struct net_device *dev, err = -E2BIG; } spin_unlock_irqrestore(&mac->lock, flags); + mutex_lock(&mac->associnfo.mutex); + return err; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_genie); @@ -473,10 +479,13 @@ ieee80211softmac_wx_set_mlme(struct net_device *dev, struct iw_mlme *mlme = (struct iw_mlme *)extra; u16 reason = cpu_to_le16(mlme->reason_code); struct ieee80211softmac_network *net; + int err = -EINVAL; + + mutex_lock(&mac->associnfo.mutex); if (memcmp(mac->associnfo.bssid, mlme->addr.sa_data, ETH_ALEN)) { printk(KERN_DEBUG PFX "wx_set_mlme: requested operation on net we don't use\n"); - return -EINVAL; + goto out; } switch (mlme->cmd) { @@ -484,14 +493,22 @@ ieee80211softmac_wx_set_mlme(struct net_device *dev, net = ieee80211softmac_get_network_by_bssid_locked(mac, mlme->addr.sa_data); if (!net) { printk(KERN_DEBUG PFX "wx_set_mlme: we should know the net here...\n"); - return -EINVAL; + goto out; } return ieee80211softmac_deauth_req(mac, net, reason); case IW_MLME_DISASSOC: ieee80211softmac_send_disassoc_req(mac, reason); - return 0; + mac->associnfo.associated = 0; + mac->associnfo.associating = 0; + err = 0; + goto out; default: - return -EOPNOTSUPP; + err = -EOPNOTSUPP; } + +out: + mutex_unlock(&mac->associnfo.mutex); + + return err; } EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_mlme); -- cgit v0.10.2 From 16bfa676a7cc64695f7e9694c380ebd26c461ae5 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Thu, 28 Sep 2006 00:10:55 -0500 Subject: [PATCH] bcm43xx-softmac: check returned value from pci_enable_device Linus's tree now has a configuration option that prints a warning whenever the returned value of any routine is ignored. This patch fixes the only such warning for bcm43xx. Signed-off-by: Larry Finger Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index 0f047d4..7776b5c 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -4207,7 +4207,11 @@ static int bcm43xx_resume(struct pci_dev *pdev) dprintk(KERN_INFO PFX "Resuming...\n"); pci_set_power_state(pdev, 0); - pci_enable_device(pdev); + err = pci_enable_device(pdev); + if (err) { + printk(KERN_ERR PFX "Failure with pci_enable_device!\n"); + return err; + } pci_restore_state(pdev); bcm43xx_chipset_attach(bcm); -- cgit v0.10.2 From 8da81e52b743edac0bfbb7d0c1286f919b2f209b Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Mon, 2 Oct 2006 23:48:54 -0500 Subject: [PATCH] bcm43xx-softmac: Fix system hang for x86-64 with >1GB RAM The bcm43xx-softmac software currently fails when running on x86_64 systems with more than 1GB RAM and one of the card variants with 30-bit DMA addressing. This patch uses the address extension bits in the hardware to set the correct DMA mask for the specific card in use. Signed-off-by: Larry Finger Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_dma.c b/drivers/net/wireless/bcm43xx/bcm43xx_dma.c index 76e3aed..978ed09 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_dma.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_dma.c @@ -705,11 +705,30 @@ int bcm43xx_dma_init(struct bcm43xx_private *bcm) struct bcm43xx_dmaring *ring; int err = -ENOMEM; int dma64 = 0; - u32 sbtmstatehi; + u64 mask = bcm43xx_get_supported_dma_mask(bcm); + int nobits; - sbtmstatehi = bcm43xx_read32(bcm, BCM43xx_CIR_SBTMSTATEHIGH); - if (sbtmstatehi & BCM43xx_SBTMSTATEHIGH_DMA64BIT) + if (mask == DMA_64BIT_MASK) { dma64 = 1; + nobits = 64; + } else if (mask == DMA_32BIT_MASK) + nobits = 32; + else + nobits = 30; + err = pci_set_dma_mask(bcm->pci_dev, mask); + err |= pci_set_consistent_dma_mask(bcm->pci_dev, mask); + if (err) { +#ifdef CONFIG_BCM43XX_PIO + printk(KERN_WARNING PFX "DMA not supported on this device." + " Falling back to PIO.\n"); + bcm->__using_pio = 1; + return -ENOSYS; +#else + printk(KERN_ERR PFX "FATAL: DMA not supported and PIO not configured. " + "Please recompile the driver with PIO support.\n"); + return -ENODEV; +#endif /* CONFIG_BCM43XX_PIO */ + } /* setup TX DMA channels. */ ring = bcm43xx_setup_dmaring(bcm, 0, 1, dma64); @@ -755,8 +774,7 @@ int bcm43xx_dma_init(struct bcm43xx_private *bcm) dma->rx_ring3 = ring; } - dprintk(KERN_INFO PFX "%s DMA initialized\n", - dma64 ? "64-bit" : "32-bit"); + dprintk(KERN_INFO PFX "%d-bit DMA initialized\n", nobits); err = 0; out: return err; diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_dma.h b/drivers/net/wireless/bcm43xx/bcm43xx_dma.h index e04bcad..ea16078 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_dma.h +++ b/drivers/net/wireless/bcm43xx/bcm43xx_dma.h @@ -314,6 +314,23 @@ int bcm43xx_dma_tx(struct bcm43xx_private *bcm, struct ieee80211_txb *txb); void bcm43xx_dma_rx(struct bcm43xx_dmaring *ring); +/* Helper function that returns the dma mask for this device. */ +static inline +u64 bcm43xx_get_supported_dma_mask(struct bcm43xx_private *bcm) +{ + int dma64 = bcm43xx_read32(bcm, BCM43xx_CIR_SBTMSTATEHIGH) & + BCM43xx_SBTMSTATEHIGH_DMA64BIT; + u16 mmio_base = bcm43xx_dmacontroller_base(dma64, 0); + u32 mask = BCM43xx_DMA32_TXADDREXT_MASK; + + if (dma64) + return DMA_64BIT_MASK; + bcm43xx_write32(bcm, mmio_base + BCM43xx_DMA32_TXCTL, mask); + if (bcm43xx_read32(bcm, mmio_base + BCM43xx_DMA32_TXCTL) & mask) + return DMA_32BIT_MASK; + return DMA_30BIT_MASK; +} + #else /* CONFIG_BCM43XX_DMA */ diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index 7776b5c..a94c6d8 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -2925,10 +2925,13 @@ static int bcm43xx_wireless_core_init(struct bcm43xx_private *bcm, bcm43xx_write16(bcm, 0x043C, 0x000C); if (active_wlcore) { - if (bcm43xx_using_pio(bcm)) + if (bcm43xx_using_pio(bcm)) { err = bcm43xx_pio_init(bcm); - else + } else { err = bcm43xx_dma_init(bcm); + if (err == -ENOSYS) + err = bcm43xx_pio_init(bcm); + } if (err) goto err_chip_cleanup; } @@ -3992,8 +3995,6 @@ static int bcm43xx_init_private(struct bcm43xx_private *bcm, struct net_device *net_dev, struct pci_dev *pci_dev) { - int err; - bcm43xx_set_status(bcm, BCM43xx_STAT_UNINIT); bcm->ieee = netdev_priv(net_dev); bcm->softmac = ieee80211_priv(net_dev); @@ -4011,22 +4012,8 @@ static int bcm43xx_init_private(struct bcm43xx_private *bcm, (void (*)(unsigned long))bcm43xx_interrupt_tasklet, (unsigned long)bcm); tasklet_disable_nosync(&bcm->isr_tasklet); - if (modparam_pio) { + if (modparam_pio) bcm->__using_pio = 1; - } else { - err = pci_set_dma_mask(pci_dev, DMA_30BIT_MASK); - err |= pci_set_consistent_dma_mask(pci_dev, DMA_30BIT_MASK); - if (err) { -#ifdef CONFIG_BCM43XX_PIO - printk(KERN_WARNING PFX "DMA not supported. Falling back to PIO.\n"); - bcm->__using_pio = 1; -#else - printk(KERN_ERR PFX "FATAL: DMA not supported and PIO not configured. " - "Recompile the driver with PIO support, please.\n"); - return -ENODEV; -#endif /* CONFIG_BCM43XX_PIO */ - } - } bcm->rts_threshold = BCM43xx_DEFAULT_RTS_THRESHOLD; /* default to sw encryption for now */ -- cgit v0.10.2 From 431aca5a18f15f61cc51c466073928c4f9565fe4 Mon Sep 17 00:00:00 2001 From: Florin Malita Date: Tue, 10 Oct 2006 16:46:30 -0400 Subject: [PATCH] airo.c: check returned values create_proc_entry() can fail and return NULL in setup_proc_entry(), the result must be checked before dereferencing. (Coverity ID 1443) init_wifidev() & setup_proc_entry() can also fail in _init_airo_card(). This adds the checks & cleanup code and removes some whitespace. Signed-off-by: Florin Malita Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index 0a33c8a..9d5427a 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -2897,6 +2897,8 @@ static struct net_device *_init_airo_card( unsigned short irq, int port, goto err_out_map; } ai->wifidev = init_wifidev(ai, dev); + if (!ai->wifidev) + goto err_out_reg; set_bit(FLAG_REGISTERED,&ai->flags); airo_print_info(dev->name, "MAC enabled %x:%x:%x:%x:%x:%x", @@ -2908,11 +2910,18 @@ static struct net_device *_init_airo_card( unsigned short irq, int port, for( i = 0; i < MAX_FIDS; i++ ) ai->fids[i] = transmit_allocate(ai,AIRO_DEF_MTU,i>=MAX_FIDS/2); - setup_proc_entry( dev, dev->priv ); /* XXX check for failure */ + if (setup_proc_entry(dev, dev->priv) < 0) + goto err_out_wifi; + netif_start_queue(dev); SET_MODULE_OWNER(dev); return dev; +err_out_wifi: + unregister_netdev(ai->wifidev); + free_netdev(ai->wifidev); +err_out_reg: + unregister_netdev(dev); err_out_map: if (test_bit(FLAG_MPI,&ai->flags) && pci) { pci_free_consistent(pci, PCI_SHARED_LEN, ai->shared, ai->shared_dma); @@ -4495,91 +4504,128 @@ static int setup_proc_entry( struct net_device *dev, apriv->proc_entry = create_proc_entry(apriv->proc_name, S_IFDIR|airo_perm, airo_entry); - apriv->proc_entry->uid = proc_uid; - apriv->proc_entry->gid = proc_gid; - apriv->proc_entry->owner = THIS_MODULE; + if (!apriv->proc_entry) + goto fail; + apriv->proc_entry->uid = proc_uid; + apriv->proc_entry->gid = proc_gid; + apriv->proc_entry->owner = THIS_MODULE; /* Setup the StatsDelta */ entry = create_proc_entry("StatsDelta", S_IFREG | (S_IRUGO&proc_perm), apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_stats_delta; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_statsdelta_ops); /* Setup the Stats */ entry = create_proc_entry("Stats", S_IFREG | (S_IRUGO&proc_perm), apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_stats; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_stats_ops); /* Setup the Status */ entry = create_proc_entry("Status", S_IFREG | (S_IRUGO&proc_perm), apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_status; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_status_ops); /* Setup the Config */ entry = create_proc_entry("Config", S_IFREG | proc_perm, apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_config; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_config_ops); /* Setup the SSID */ entry = create_proc_entry("SSID", S_IFREG | proc_perm, apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_ssid; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_SSID_ops); /* Setup the APList */ entry = create_proc_entry("APList", S_IFREG | proc_perm, apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_aplist; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_APList_ops); /* Setup the BSSList */ entry = create_proc_entry("BSSList", S_IFREG | proc_perm, apriv->proc_entry); + if (!entry) + goto fail_bsslist; entry->uid = proc_uid; entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_BSSList_ops); /* Setup the WepKey */ entry = create_proc_entry("WepKey", S_IFREG | proc_perm, apriv->proc_entry); - entry->uid = proc_uid; - entry->gid = proc_gid; + if (!entry) + goto fail_wepkey; + entry->uid = proc_uid; + entry->gid = proc_gid; entry->data = dev; - entry->owner = THIS_MODULE; + entry->owner = THIS_MODULE; SETPROC_OPS(entry, proc_wepkey_ops); return 0; + +fail_wepkey: + remove_proc_entry("BSSList", apriv->proc_entry); +fail_bsslist: + remove_proc_entry("APList", apriv->proc_entry); +fail_aplist: + remove_proc_entry("SSID", apriv->proc_entry); +fail_ssid: + remove_proc_entry("Config", apriv->proc_entry); +fail_config: + remove_proc_entry("Status", apriv->proc_entry); +fail_status: + remove_proc_entry("Stats", apriv->proc_entry); +fail_stats: + remove_proc_entry("StatsDelta", apriv->proc_entry); +fail_stats_delta: + remove_proc_entry(apriv->proc_name, airo_entry); +fail: + return -ENOMEM; } static int takedown_proc_entry( struct net_device *dev, -- cgit v0.10.2 From 7e4e8d99c2288a490a0806b9cb40016913312cfe Mon Sep 17 00:00:00 2001 From: Jean Tourrilhes Date: Tue, 10 Oct 2006 14:45:44 -0700 Subject: [PATCH] orinoco: fix WE-21 buffer overflow This patch fixes the Orinoco driver overflow issue with WE-21. Cc: Valdis Kletnieks Cc: Pavel Roskin Signed-off-by: Andrew Morton Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c index b779c7d..336caba 100644 --- a/drivers/net/wireless/orinoco.c +++ b/drivers/net/wireless/orinoco.c @@ -2457,6 +2457,7 @@ void free_orinocodev(struct net_device *dev) /* Wireless extensions */ /********************************************************************/ +/* Return : < 0 -> error code ; >= 0 -> length */ static int orinoco_hw_get_essid(struct orinoco_private *priv, int *active, char buf[IW_ESSID_MAX_SIZE+1]) { @@ -2501,9 +2502,9 @@ static int orinoco_hw_get_essid(struct orinoco_private *priv, int *active, len = le16_to_cpu(essidbuf.len); BUG_ON(len > IW_ESSID_MAX_SIZE); - memset(buf, 0, IW_ESSID_MAX_SIZE+1); + memset(buf, 0, IW_ESSID_MAX_SIZE); memcpy(buf, p, len); - buf[len] = '\0'; + err = len; fail_unlock: orinoco_unlock(priv, &flags); @@ -3027,17 +3028,18 @@ static int orinoco_ioctl_getessid(struct net_device *dev, if (netif_running(dev)) { err = orinoco_hw_get_essid(priv, &active, essidbuf); - if (err) + if (err < 0) return err; + erq->length = err; } else { if (orinoco_lock(priv, &flags) != 0) return -EBUSY; - memcpy(essidbuf, priv->desired_essid, IW_ESSID_MAX_SIZE + 1); + memcpy(essidbuf, priv->desired_essid, IW_ESSID_MAX_SIZE); + erq->length = strlen(priv->desired_essid); orinoco_unlock(priv, &flags); } erq->flags = 1; - erq->length = strlen(essidbuf); return 0; } @@ -3075,10 +3077,10 @@ static int orinoco_ioctl_getnick(struct net_device *dev, if (orinoco_lock(priv, &flags) != 0) return -EBUSY; - memcpy(nickbuf, priv->nick, IW_ESSID_MAX_SIZE+1); + memcpy(nickbuf, priv->nick, IW_ESSID_MAX_SIZE); orinoco_unlock(priv, &flags); - nrq->length = strlen(nickbuf); + nrq->length = strlen(priv->nick); return 0; } -- cgit v0.10.2 From 683f8c9e00d2aa911382186ca891bd221efaea74 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Tue, 10 Oct 2006 14:45:45 -0700 Subject: [PATCH] zd1201: Possible NULL dereference If we enter the if(!zd) and set free to 1, we dereference zd in the exit code. Signed-off-by: Eric Sesterhenn Signed-off-by: Andrew Morton Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/zd1201.c b/drivers/net/wireless/zd1201.c index 30057a3..36b29ff 100644 --- a/drivers/net/wireless/zd1201.c +++ b/drivers/net/wireless/zd1201.c @@ -193,10 +193,8 @@ static void zd1201_usbrx(struct urb *urb) struct sk_buff *skb; unsigned char type; - if (!zd) { - free = 1; - goto exit; - } + if (!zd) + return; switch(urb->status) { case -EILSEQ: -- cgit v0.10.2 From 53077944f119808df3d1c6be07241f17a87e7c28 Mon Sep 17 00:00:00 2001 From: Jean Tourrilhes Date: Tue, 10 Oct 2006 14:45:46 -0700 Subject: [PATCH] wireless: More WE-21 potential overflows... After the Orinoco issue, I did an audit of other drivers for the same issue. Three drivers were NULL terminating the ESSID, which could cause an overflow in WE-21 when the ESSID has maximum size. Signed-off-by: Jean Tourrilhes Signed-off-by: Andrew Morton Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index 9d5427a..e0710fa 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -5970,7 +5970,6 @@ static int airo_get_essid(struct net_device *dev, /* Get the current SSID */ memcpy(extra, status_rid.SSID, status_rid.SSIDlen); - extra[status_rid.SSIDlen] = '\0'; /* If none, we may want to get the one that was set */ /* Push it out ! */ diff --git a/drivers/net/wireless/atmel.c b/drivers/net/wireless/atmel.c index 31eed85..0c07b8b 100644 --- a/drivers/net/wireless/atmel.c +++ b/drivers/net/wireless/atmel.c @@ -1678,11 +1678,9 @@ static int atmel_get_essid(struct net_device *dev, /* Get the current SSID */ if (priv->new_SSID_size != 0) { memcpy(extra, priv->new_SSID, priv->new_SSID_size); - extra[priv->new_SSID_size] = '\0'; dwrq->length = priv->new_SSID_size; } else { memcpy(extra, priv->SSID, priv->SSID_size); - extra[priv->SSID_size] = '\0'; dwrq->length = priv->SSID_size; } diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c index 0b381d7..7fbfc9e 100644 --- a/drivers/net/wireless/ray_cs.c +++ b/drivers/net/wireless/ray_cs.c @@ -1198,7 +1198,6 @@ static int ray_get_essid(struct net_device *dev, /* Get the essid that was set */ memcpy(extra, local->sparm.b5.a_current_ess_id, IW_ESSID_MAX_SIZE); - extra[IW_ESSID_MAX_SIZE] = '\0'; /* Push it out ! */ dwrq->length = strlen(extra); -- cgit v0.10.2 From 5bb85f18087b10a908bd793e9fd3ccd63aebb724 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Tue, 10 Oct 2006 14:45:46 -0700 Subject: [PATCH] airo: check if need to freeze The airo driver used to break out of while loop if there were any signals pending. Since it no longer checks for signals, it at least needs to check if it needs to be frozen. Signed-off-by: Dave Kleikamp Cc: Jean Tourrilhes Cc: Sukadev Bhattiprolu Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index e0710fa..efcdaf1 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -3098,7 +3098,8 @@ static int airo_thread(void *data) { set_bit(JOB_AUTOWEP, &ai->jobs); break; } - if (!kthread_should_stop()) { + if (!kthread_should_stop() && + !freezing(current)) { unsigned long wake_at; if (!ai->expires || !ai->scan_timeout) { wake_at = max(ai->expires, @@ -3110,7 +3111,8 @@ static int airo_thread(void *data) { schedule_timeout(wake_at - jiffies); continue; } - } else if (!kthread_should_stop()) { + } else if (!kthread_should_stop() && + !freezing(current)) { schedule(); continue; } -- cgit v0.10.2 From 1f5c23e2c10d642a23aa3ebb449670a5184b6aab Mon Sep 17 00:00:00 2001 From: Arthur Kepner Date: Mon, 16 Oct 2006 20:22:35 -0700 Subject: IB/mthca: Use mmiowb after doorbell ring We discovered a problem when running IPoIB applications on multiple CPUs on an Altix system. Many messages such as: ib_mthca 0002:01:00.0: SQ 000014 full (19941644 head, 19941707 tail, 64 max, 0 nreq) appear in syslog, and the driver wedges up. Apparently this is because writes to the doorbells from different CPUs reach the device out of order. The following patch adds mmiowb() calls after doorbell rings to ensure the doorbell writes are ordered. Signed-off-by: Arthur Kepner Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/mthca/mthca_cq.c b/drivers/infiniband/hw/mthca/mthca_cq.c index e393681..149b369 100644 --- a/drivers/infiniband/hw/mthca/mthca_cq.c +++ b/drivers/infiniband/hw/mthca/mthca_cq.c @@ -39,6 +39,8 @@ #include #include +#include + #include #include "mthca_dev.h" @@ -210,6 +212,11 @@ static inline void update_cons_index(struct mthca_dev *dev, struct mthca_cq *cq, mthca_write64(doorbell, dev->kar + MTHCA_CQ_DOORBELL, MTHCA_GET_DOORBELL_LOCK(&dev->doorbell_lock)); + /* + * Make sure doorbells don't leak out of CQ spinlock + * and reach the HCA out of order: + */ + mmiowb(); } } diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 5e5c58b..6a7822e 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -39,6 +39,8 @@ #include #include +#include + #include #include #include @@ -1732,6 +1734,11 @@ out: mthca_write64(doorbell, dev->kar + MTHCA_SEND_DOORBELL, MTHCA_GET_DOORBELL_LOCK(&dev->doorbell_lock)); + /* + * Make sure doorbells don't leak out of SQ spinlock + * and reach the HCA out of order: + */ + mmiowb(); } qp->sq.next_ind = ind; @@ -1851,6 +1858,12 @@ out: qp->rq.next_ind = ind; qp->rq.head += nreq; + /* + * Make sure doorbells don't leak out of RQ spinlock and reach + * the HCA out of order: + */ + mmiowb(); + spin_unlock_irqrestore(&qp->rq.lock, flags); return err; } @@ -2112,6 +2125,12 @@ out: MTHCA_GET_DOORBELL_LOCK(&dev->doorbell_lock)); } + /* + * Make sure doorbells don't leak out of SQ spinlock and reach + * the HCA out of order: + */ + mmiowb(); + spin_unlock_irqrestore(&qp->sq.lock, flags); return err; } diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index 92a72f5..f5d7677 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -35,6 +35,8 @@ #include #include +#include + #include "mthca_dev.h" #include "mthca_cmd.h" #include "mthca_memfree.h" @@ -595,6 +597,12 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, MTHCA_GET_DOORBELL_LOCK(&dev->doorbell_lock)); } + /* + * Make sure doorbells don't leak out of SRQ spinlock and + * reach the HCA out of order: + */ + mmiowb(); + spin_unlock_irqrestore(&srq->lock, flags); return err; } -- cgit v0.10.2 From 6684e59aa3cf6cb7ebf04ea2953198500c93b0a9 Mon Sep 17 00:00:00 2001 From: Laurent Riffard Date: Thu, 12 Oct 2006 00:17:36 +0200 Subject: [PATCH] sotftmac: fix a slab corruption in WEP restricted key association Fix a slab corruption in ieee80211softmac_auth(). The size of a buffer was miscomputed. see http://bugzilla.kernel.org/show_bug.cgi?id=7245 Acked-by: Daniel Drake Signed-off-by: Laurent Riffard Signed-off-by: John W. Linville diff --git a/net/ieee80211/softmac/ieee80211softmac_io.c b/net/ieee80211/softmac/ieee80211softmac_io.c index e8419cf..b969310 100644 --- a/net/ieee80211/softmac/ieee80211softmac_io.c +++ b/net/ieee80211/softmac/ieee80211softmac_io.c @@ -304,7 +304,7 @@ ieee80211softmac_auth(struct ieee80211_auth **pkt, 2 + /* Auth Transaction Seq */ 2 + /* Status Code */ /* Challenge Text IE */ - is_shared_response ? 0 : 1 + 1 + net->challenge_len + (is_shared_response ? 1 + 1 + net->challenge_len : 0) ); if (unlikely((*pkt) == NULL)) return 0; -- cgit v0.10.2 From 107d5a72f2c6a6819b66eebcb0281c7a67b6baaa Mon Sep 17 00:00:00 2001 From: Brent Casavant Date: Tue, 17 Oct 2006 00:09:24 -0700 Subject: [PATCH] ioc4: Remove SN2 feature and config dependencies The SGI PCI-RT card, based on the SGI IOC4 chip, will be made available on Altix XE (x86_64) platforms in the near future. As such dependencies on SN2-specific features and config dependencies need to be removed. This patch updates the Kconfig files to remove the config dependency, and updates the IOC4 bus speed detection routine to use universally available time interfaces instead of mmtimer. Signed-off-by: Brent Casavant Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/sn/Kconfig b/drivers/sn/Kconfig index a347316..34d4fde 100644 --- a/drivers/sn/Kconfig +++ b/drivers/sn/Kconfig @@ -7,16 +7,15 @@ menu "SN Devices" config SGI_IOC4 tristate "SGI IOC4 Base IO support" - depends on MMTIMER default m ---help--- - This option enables basic support for the SGI IOC4-based Base IO - controller card. This option does not enable any specific - functions on such a card, but provides necessary infrastructure - for other drivers to utilize. + This option enables basic support for the IOC4 chip on certain + SGI IO controller cards (IO9, IO10, and PCI-RT). This option + does not enable any specific functions on such a card, but provides + necessary infrastructure for other drivers to utilize. - If you have an SGI Altix with an IOC4-based - I/O controller say Y. Otherwise say N. + If you have an SGI Altix with an IOC4-based card say Y. + Otherwise say N. config SGI_IOC3 tristate "SGI IOC3 Base IO support" diff --git a/drivers/sn/ioc4.c b/drivers/sn/ioc4.c index 8562821..83d2e90 100644 --- a/drivers/sn/ioc4.c +++ b/drivers/sn/ioc4.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (C) 2005-2006 Silicon Graphics, Inc. All Rights Reserved. */ /* This file contains the master driver module for use by SGI IOC4 subdrivers. @@ -29,9 +29,9 @@ #include #include #include -#include -#include +#include #include +#include #include #include #include @@ -43,7 +43,7 @@ /* Tweakable values */ /* PCI bus speed detection/calibration */ -#define IOC4_CALIBRATE_COUNT 63 /* Calibration cycle period */ +#define IOC4_CALIBRATE_COUNT 63 /* Calibration cycle period */ #define IOC4_CALIBRATE_CYCLES 256 /* Average over this many cycles */ #define IOC4_CALIBRATE_DISCARD 2 /* Discard first few cycles */ #define IOC4_CALIBRATE_LOW_MHZ 25 /* Lower bound on bus speed sanity */ @@ -143,11 +143,11 @@ ioc4_unregister_submodule(struct ioc4_submodule *is) static void ioc4_clock_calibrate(struct ioc4_driver_data *idd) { - extern unsigned long sn_rtc_cycles_per_second; union ioc4_int_out int_out; union ioc4_gpcr gpcr; unsigned int state, last_state = 1; - uint64_t start = 0, end, period; + struct timespec start_ts, end_ts; + uint64_t start, end, period; unsigned int count = 0; /* Enable output */ @@ -175,30 +175,28 @@ ioc4_clock_calibrate(struct ioc4_driver_data *idd) if (!last_state && state) { count++; if (count == IOC4_CALIBRATE_END) { - end = rtc_time(); + ktime_get_ts(&end_ts); break; } else if (count == IOC4_CALIBRATE_DISCARD) - start = rtc_time(); + ktime_get_ts(&start_ts); } last_state = state; } while (1); /* Calculation rearranged to preserve intermediate precision. * Logically: - * 1. "end - start" gives us number of RTC cycles over all the - * square wave cycles measured. - * 2. Divide by number of square wave cycles to get number of - * RTC cycles per square wave cycle. + * 1. "end - start" gives us the measurement period over all + * the square wave cycles. + * 2. Divide by number of square wave cycles to get the period + * of a square wave cycle. * 3. Divide by 2*(int_out.fields.count+1), which is the formula * by which the IOC4 generates the square wave, to get the - * number of RTC cycles per IOC4 INT_OUT count. - * 4. Divide by sn_rtc_cycles_per_second to get seconds per - * count. - * 5. Multiply by 1E9 to get nanoseconds per count. + * period of an IOC4 INT_OUT count. */ - period = ((end - start) * 1000000000) / - (IOC4_CALIBRATE_CYCLES * 2 * (IOC4_CALIBRATE_COUNT + 1) - * sn_rtc_cycles_per_second); + end = end_ts.tv_sec * NSEC_PER_SEC + end_ts.tv_nsec; + start = start_ts.tv_sec * NSEC_PER_SEC + start_ts.tv_nsec; + period = (end - start) / + (IOC4_CALIBRATE_CYCLES * 2 * (IOC4_CALIBRATE_COUNT + 1)); /* Bounds check the result. */ if (period > IOC4_CALIBRATE_LOW_LIMIT || -- cgit v0.10.2 From 59f148005cfd3d41537a4b872c266213d5fe4dc6 Mon Sep 17 00:00:00 2001 From: Brent Casavant Date: Tue, 17 Oct 2006 00:09:25 -0700 Subject: [PATCH] ioc4: Enable build on non-SN2 The SGI PCI-RT card, based on the SGI IOC4 chip, will be made available on Altix XE (x86_64) platforms in the near future. As such it is now a misnomer for the IOC4 base device driver to live under drivers/sn, and would complicate builds for non-SN2. This patch moves the IOC4 base driver code from drivers/sn to drivers/misc, and updates the associated Makefiles and Kconfig files to allow building on non-SN2 configs. Due to the resulting change in link order, it is now necessary to use late_initcall() for IOC4 subdriver initialization. [akpm@osdl.org: __udivdi3 fix] [akpm@osdl.org: fix default in Kconfig] Acked-by: Pat Gefre Acked-by: Jeremy Higdon Signed-off-by: Brent Casavant Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/Kconfig b/drivers/Kconfig index 263e86d..f394634 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -14,6 +14,10 @@ source "drivers/pnp/Kconfig" source "drivers/block/Kconfig" +# misc before ide - BLK_DEV_SGIIOC4 depends on SGI_IOC4 + +source "drivers/misc/Kconfig" + source "drivers/ide/Kconfig" source "drivers/scsi/Kconfig" @@ -52,8 +56,6 @@ source "drivers/w1/Kconfig" source "drivers/hwmon/Kconfig" -source "drivers/misc/Kconfig" - source "drivers/mfd/Kconfig" source "drivers/media/Kconfig" diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c index f3fe287..244f7eb 100644 --- a/drivers/ide/pci/sgiioc4.c +++ b/drivers/ide/pci/sgiioc4.c @@ -774,7 +774,7 @@ ioc4_ide_exit(void) ioc4_unregister_submodule(&ioc4_ide_submodule); } -module_init(ioc4_ide_init); +late_initcall(ioc4_ide_init); /* Call only after IDE init is done */ module_exit(ioc4_ide_exit); MODULE_AUTHOR("Aniket Malatpure/Jeremy Higdon"); diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index fa7acc2..b6c045d 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -28,6 +28,17 @@ config IBM_ASM If unsure, say N. +config SGI_IOC4 + tristate "SGI IOC4 Base IO support" + ---help--- + This option enables basic support for the IOC4 chip on certain + SGI IO controller cards (IO9, IO10, and PCI-RT). This option + does not enable any specific functions on such a card, but provides + necessary infrastructure for other drivers to utilize. + + If you have an SGI Altix with an IOC4-based card say Y. + Otherwise say N. + config TIFM_CORE tristate "TI Flash Media interface support (EXPERIMENTAL)" depends on EXPERIMENTAL diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile index 9a91c1e..c9e98ab 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -9,3 +9,4 @@ obj-$(CONFIG_MSI_LAPTOP) += msi-laptop.o obj-$(CONFIG_LKDTM) += lkdtm.o obj-$(CONFIG_TIFM_CORE) += tifm_core.o obj-$(CONFIG_TIFM_7XX1) += tifm_7xx1.o +obj-$(CONFIG_SGI_IOC4) += ioc4.o diff --git a/drivers/misc/ioc4.c b/drivers/misc/ioc4.c new file mode 100644 index 0000000..1c3c14a --- /dev/null +++ b/drivers/misc/ioc4.c @@ -0,0 +1,473 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2005-2006 Silicon Graphics, Inc. All Rights Reserved. + */ + +/* This file contains the master driver module for use by SGI IOC4 subdrivers. + * + * It allocates any resources shared between multiple subdevices, and + * provides accessor functions (where needed) and the like for those + * resources. It also provides a mechanism for the subdevice modules + * to support loading and unloading. + * + * Non-shared resources (e.g. external interrupt A_INT_OUT register page + * alias, serial port and UART registers) are handled by the subdevice + * modules themselves. + * + * This is all necessary because IOC4 is not implemented as a multi-function + * PCI device, but an amalgamation of disparate registers for several + * types of device (ATA, serial, external interrupts). The normal + * resource management in the kernel doesn't have quite the right interfaces + * to handle this situation (e.g. multiple modules can't claim the same + * PCI ID), thus this IOC4 master module. + */ + +#include +#include +#include +#include +#include +#include +#include + +/*************** + * Definitions * + ***************/ + +/* Tweakable values */ + +/* PCI bus speed detection/calibration */ +#define IOC4_CALIBRATE_COUNT 63 /* Calibration cycle period */ +#define IOC4_CALIBRATE_CYCLES 256 /* Average over this many cycles */ +#define IOC4_CALIBRATE_DISCARD 2 /* Discard first few cycles */ +#define IOC4_CALIBRATE_LOW_MHZ 25 /* Lower bound on bus speed sanity */ +#define IOC4_CALIBRATE_HIGH_MHZ 75 /* Upper bound on bus speed sanity */ +#define IOC4_CALIBRATE_DEFAULT_MHZ 66 /* Assumed if sanity check fails */ + +/************************ + * Submodule management * + ************************/ + +static DEFINE_MUTEX(ioc4_mutex); + +static LIST_HEAD(ioc4_devices); +static LIST_HEAD(ioc4_submodules); + +/* Register an IOC4 submodule */ +int +ioc4_register_submodule(struct ioc4_submodule *is) +{ + struct ioc4_driver_data *idd; + + mutex_lock(&ioc4_mutex); + list_add(&is->is_list, &ioc4_submodules); + + /* Initialize submodule for each IOC4 */ + if (!is->is_probe) + goto out; + + list_for_each_entry(idd, &ioc4_devices, idd_list) { + if (is->is_probe(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule %s probe failed " + "for pci_dev %s", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + out: + mutex_unlock(&ioc4_mutex); + return 0; +} + +/* Unregister an IOC4 submodule */ +void +ioc4_unregister_submodule(struct ioc4_submodule *is) +{ + struct ioc4_driver_data *idd; + + mutex_lock(&ioc4_mutex); + list_del(&is->is_list); + + /* Remove submodule for each IOC4 */ + if (!is->is_remove) + goto out; + + list_for_each_entry(idd, &ioc4_devices, idd_list) { + if (is->is_remove(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule %s remove failed " + "for pci_dev %s.\n", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + out: + mutex_unlock(&ioc4_mutex); +} + +/********************* + * Device management * + *********************/ + +#define IOC4_CALIBRATE_LOW_LIMIT \ + (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_LOW_MHZ) +#define IOC4_CALIBRATE_HIGH_LIMIT \ + (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_HIGH_MHZ) +#define IOC4_CALIBRATE_DEFAULT \ + (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_DEFAULT_MHZ) + +#define IOC4_CALIBRATE_END \ + (IOC4_CALIBRATE_CYCLES + IOC4_CALIBRATE_DISCARD) + +#define IOC4_INT_OUT_MODE_TOGGLE 0x7 /* Toggle INT_OUT every COUNT+1 ticks */ + +/* Determines external interrupt output clock period of the PCI bus an + * IOC4 is attached to. This value can be used to determine the PCI + * bus speed. + * + * IOC4 has a design feature that various internal timers are derived from + * the PCI bus clock. This causes IOC4 device drivers to need to take the + * bus speed into account when setting various register values (e.g. INT_OUT + * register COUNT field, UART divisors, etc). Since this information is + * needed by several subdrivers, it is determined by the main IOC4 driver, + * even though the following code utilizes external interrupt registers + * to perform the speed calculation. + */ +static void +ioc4_clock_calibrate(struct ioc4_driver_data *idd) +{ + union ioc4_int_out int_out; + union ioc4_gpcr gpcr; + unsigned int state, last_state = 1; + struct timespec start_ts, end_ts; + uint64_t start, end, period; + unsigned int count = 0; + + /* Enable output */ + gpcr.raw = 0; + gpcr.fields.dir = IOC4_GPCR_DIR_0; + gpcr.fields.int_out_en = 1; + writel(gpcr.raw, &idd->idd_misc_regs->gpcr_s.raw); + + /* Reset to power-on state */ + writel(0, &idd->idd_misc_regs->int_out.raw); + mmiowb(); + + /* Set up square wave */ + int_out.raw = 0; + int_out.fields.count = IOC4_CALIBRATE_COUNT; + int_out.fields.mode = IOC4_INT_OUT_MODE_TOGGLE; + int_out.fields.diag = 0; + writel(int_out.raw, &idd->idd_misc_regs->int_out.raw); + mmiowb(); + + /* Check square wave period averaged over some number of cycles */ + do { + int_out.raw = readl(&idd->idd_misc_regs->int_out.raw); + state = int_out.fields.int_out; + if (!last_state && state) { + count++; + if (count == IOC4_CALIBRATE_END) { + ktime_get_ts(&end_ts); + break; + } else if (count == IOC4_CALIBRATE_DISCARD) + ktime_get_ts(&start_ts); + } + last_state = state; + } while (1); + + /* Calculation rearranged to preserve intermediate precision. + * Logically: + * 1. "end - start" gives us the measurement period over all + * the square wave cycles. + * 2. Divide by number of square wave cycles to get the period + * of a square wave cycle. + * 3. Divide by 2*(int_out.fields.count+1), which is the formula + * by which the IOC4 generates the square wave, to get the + * period of an IOC4 INT_OUT count. + */ + end = end_ts.tv_sec * NSEC_PER_SEC + end_ts.tv_nsec; + start = start_ts.tv_sec * NSEC_PER_SEC + start_ts.tv_nsec; + period = (end - start) / + (IOC4_CALIBRATE_CYCLES * 2 * (IOC4_CALIBRATE_COUNT + 1)); + + /* Bounds check the result. */ + if (period > IOC4_CALIBRATE_LOW_LIMIT || + period < IOC4_CALIBRATE_HIGH_LIMIT) { + printk(KERN_INFO + "IOC4 %s: Clock calibration failed. Assuming" + "PCI clock is %d ns.\n", + pci_name(idd->idd_pdev), + IOC4_CALIBRATE_DEFAULT / IOC4_EXTINT_COUNT_DIVISOR); + period = IOC4_CALIBRATE_DEFAULT; + } else { + u64 ns = period; + + do_div(ns, IOC4_EXTINT_COUNT_DIVISOR); + printk(KERN_DEBUG + "IOC4 %s: PCI clock is %lld ns.\n", + pci_name(idd->idd_pdev), ns); + } + + /* Remember results. We store the extint clock period rather + * than the PCI clock period so that greater precision is + * retained. Divide by IOC4_EXTINT_COUNT_DIVISOR to get + * PCI clock period. + */ + idd->count_period = period; +} + +/* There are three variants of IOC4 cards: IO9, IO10, and PCI-RT. + * Each brings out different combinations of IOC4 signals, thus. + * the IOC4 subdrivers need to know to which we're attached. + * + * We look for the presence of a SCSI (IO9) or SATA (IO10) controller + * on the same PCI bus at slot number 3 to differentiate IO9 from IO10. + * If neither is present, it's a PCI-RT. + */ +static unsigned int +ioc4_variant(struct ioc4_driver_data *idd) +{ + struct pci_dev *pdev = NULL; + int found = 0; + + /* IO9: Look for a QLogic ISP 12160 at the same bus and slot 3. */ + do { + pdev = pci_get_device(PCI_VENDOR_ID_QLOGIC, + PCI_DEVICE_ID_QLOGIC_ISP12160, pdev); + if (pdev && + idd->idd_pdev->bus->number == pdev->bus->number && + 3 == PCI_SLOT(pdev->devfn)) + found = 1; + pci_dev_put(pdev); + } while (pdev && !found); + if (NULL != pdev) + return IOC4_VARIANT_IO9; + + /* IO10: Look for a Vitesse VSC 7174 at the same bus and slot 3. */ + pdev = NULL; + do { + pdev = pci_get_device(PCI_VENDOR_ID_VITESSE, + PCI_DEVICE_ID_VITESSE_VSC7174, pdev); + if (pdev && + idd->idd_pdev->bus->number == pdev->bus->number && + 3 == PCI_SLOT(pdev->devfn)) + found = 1; + pci_dev_put(pdev); + } while (pdev && !found); + if (NULL != pdev) + return IOC4_VARIANT_IO10; + + /* PCI-RT: No SCSI/SATA controller will be present */ + return IOC4_VARIANT_PCI_RT; +} + +/* Adds a new instance of an IOC4 card */ +static int +ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) +{ + struct ioc4_driver_data *idd; + struct ioc4_submodule *is; + uint32_t pcmd; + int ret; + + /* Enable IOC4 and take ownership of it */ + if ((ret = pci_enable_device(pdev))) { + printk(KERN_WARNING + "%s: Failed to enable IOC4 device for pci_dev %s.\n", + __FUNCTION__, pci_name(pdev)); + goto out; + } + pci_set_master(pdev); + + /* Set up per-IOC4 data */ + idd = kmalloc(sizeof(struct ioc4_driver_data), GFP_KERNEL); + if (!idd) { + printk(KERN_WARNING + "%s: Failed to allocate IOC4 data for pci_dev %s.\n", + __FUNCTION__, pci_name(pdev)); + ret = -ENODEV; + goto out_idd; + } + idd->idd_pdev = pdev; + idd->idd_pci_id = pci_id; + + /* Map IOC4 misc registers. These are shared between subdevices + * so the main IOC4 module manages them. + */ + idd->idd_bar0 = pci_resource_start(idd->idd_pdev, 0); + if (!idd->idd_bar0) { + printk(KERN_WARNING + "%s: Unable to find IOC4 misc resource " + "for pci_dev %s.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + ret = -ENODEV; + goto out_pci; + } + if (!request_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs), + "ioc4_misc")) { + printk(KERN_WARNING + "%s: Unable to request IOC4 misc region " + "for pci_dev %s.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + ret = -ENODEV; + goto out_pci; + } + idd->idd_misc_regs = ioremap(idd->idd_bar0, + sizeof(struct ioc4_misc_regs)); + if (!idd->idd_misc_regs) { + printk(KERN_WARNING + "%s: Unable to remap IOC4 misc region " + "for pci_dev %s.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + ret = -ENODEV; + goto out_misc_region; + } + + /* Failsafe portion of per-IOC4 initialization */ + + /* Detect card variant */ + idd->idd_variant = ioc4_variant(idd); + printk(KERN_INFO "IOC4 %s: %s card detected.\n", pci_name(pdev), + idd->idd_variant == IOC4_VARIANT_IO9 ? "IO9" : + idd->idd_variant == IOC4_VARIANT_PCI_RT ? "PCI-RT" : + idd->idd_variant == IOC4_VARIANT_IO10 ? "IO10" : "unknown"); + + /* Initialize IOC4 */ + pci_read_config_dword(idd->idd_pdev, PCI_COMMAND, &pcmd); + pci_write_config_dword(idd->idd_pdev, PCI_COMMAND, + pcmd | PCI_COMMAND_PARITY | PCI_COMMAND_SERR); + + /* Determine PCI clock */ + ioc4_clock_calibrate(idd); + + /* Disable/clear all interrupts. Need to do this here lest + * one submodule request the shared IOC4 IRQ, but interrupt + * is generated by a different subdevice. + */ + /* Disable */ + writel(~0, &idd->idd_misc_regs->other_iec.raw); + writel(~0, &idd->idd_misc_regs->sio_iec); + /* Clear (i.e. acknowledge) */ + writel(~0, &idd->idd_misc_regs->other_ir.raw); + writel(~0, &idd->idd_misc_regs->sio_ir); + + /* Track PCI-device specific data */ + idd->idd_serial_data = NULL; + pci_set_drvdata(idd->idd_pdev, idd); + + mutex_lock(&ioc4_mutex); + list_add_tail(&idd->idd_list, &ioc4_devices); + + /* Add this IOC4 to all submodules */ + list_for_each_entry(is, &ioc4_submodules, is_list) { + if (is->is_probe && is->is_probe(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule 0x%s probe failed " + "for pci_dev %s.\n", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + mutex_unlock(&ioc4_mutex); + + return 0; + +out_misc_region: + release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs)); +out_pci: + kfree(idd); +out_idd: + pci_disable_device(pdev); +out: + return ret; +} + +/* Removes a particular instance of an IOC4 card. */ +static void +ioc4_remove(struct pci_dev *pdev) +{ + struct ioc4_submodule *is; + struct ioc4_driver_data *idd; + + idd = pci_get_drvdata(pdev); + + /* Remove this IOC4 from all submodules */ + mutex_lock(&ioc4_mutex); + list_for_each_entry(is, &ioc4_submodules, is_list) { + if (is->is_remove && is->is_remove(idd)) { + printk(KERN_WARNING + "%s: IOC4 submodule 0x%s remove failed " + "for pci_dev %s.\n", + __FUNCTION__, module_name(is->is_owner), + pci_name(idd->idd_pdev)); + } + } + mutex_unlock(&ioc4_mutex); + + /* Release resources */ + iounmap(idd->idd_misc_regs); + if (!idd->idd_bar0) { + printk(KERN_WARNING + "%s: Unable to get IOC4 misc mapping for pci_dev %s. " + "Device removal may be incomplete.\n", + __FUNCTION__, pci_name(idd->idd_pdev)); + } + release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs)); + + /* Disable IOC4 and relinquish */ + pci_disable_device(pdev); + + /* Remove and free driver data */ + mutex_lock(&ioc4_mutex); + list_del(&idd->idd_list); + mutex_unlock(&ioc4_mutex); + kfree(idd); +} + +static struct pci_device_id ioc4_id_table[] = { + {PCI_VENDOR_ID_SGI, PCI_DEVICE_ID_SGI_IOC4, PCI_ANY_ID, + PCI_ANY_ID, 0x0b4000, 0xFFFFFF}, + {0} +}; + +static struct pci_driver ioc4_driver = { + .name = "IOC4", + .id_table = ioc4_id_table, + .probe = ioc4_probe, + .remove = ioc4_remove, +}; + +MODULE_DEVICE_TABLE(pci, ioc4_id_table); + +/********************* + * Module management * + *********************/ + +/* Module load */ +static int __devinit +ioc4_init(void) +{ + return pci_register_driver(&ioc4_driver); +} + +/* Module unload */ +static void __devexit +ioc4_exit(void) +{ + pci_unregister_driver(&ioc4_driver); +} + +module_init(ioc4_init); +module_exit(ioc4_exit); + +MODULE_AUTHOR("Brent Casavant - Silicon Graphics, Inc. "); +MODULE_DESCRIPTION("PCI driver master module for SGI IOC4 Base-IO Card"); +MODULE_LICENSE("GPL"); + +EXPORT_SYMBOL(ioc4_register_submodule); +EXPORT_SYMBOL(ioc4_unregister_submodule); diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index 98ce88d..ff4fa25 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -2935,7 +2935,7 @@ static void __devexit ioc4_serial_exit(void) uart_unregister_driver(&ioc4_uart_rs422); } -module_init(ioc4_serial_init); +late_initcall(ioc4_serial_init); /* Call only after tty init is done */ module_exit(ioc4_serial_exit); MODULE_AUTHOR("Pat Gefre - Silicon Graphics Inc. (SGI) "); diff --git a/drivers/sn/Kconfig b/drivers/sn/Kconfig index 34d4fde..c66ba9a 100644 --- a/drivers/sn/Kconfig +++ b/drivers/sn/Kconfig @@ -5,18 +5,6 @@ menu "SN Devices" depends on SGI_SN -config SGI_IOC4 - tristate "SGI IOC4 Base IO support" - default m - ---help--- - This option enables basic support for the IOC4 chip on certain - SGI IO controller cards (IO9, IO10, and PCI-RT). This option - does not enable any specific functions on such a card, but provides - necessary infrastructure for other drivers to utilize. - - If you have an SGI Altix with an IOC4-based card say Y. - Otherwise say N. - config SGI_IOC3 tristate "SGI IOC3 Base IO support" default m diff --git a/drivers/sn/Makefile b/drivers/sn/Makefile index 2cda011..693db8b 100644 --- a/drivers/sn/Makefile +++ b/drivers/sn/Makefile @@ -3,5 +3,4 @@ # # -obj-$(CONFIG_SGI_IOC4) += ioc4.o obj-$(CONFIG_SGI_IOC3) += ioc3.o diff --git a/drivers/sn/ioc4.c b/drivers/sn/ioc4.c deleted file mode 100644 index 83d2e90..0000000 --- a/drivers/sn/ioc4.c +++ /dev/null @@ -1,474 +0,0 @@ -/* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (C) 2005-2006 Silicon Graphics, Inc. All Rights Reserved. - */ - -/* This file contains the master driver module for use by SGI IOC4 subdrivers. - * - * It allocates any resources shared between multiple subdevices, and - * provides accessor functions (where needed) and the like for those - * resources. It also provides a mechanism for the subdevice modules - * to support loading and unloading. - * - * Non-shared resources (e.g. external interrupt A_INT_OUT register page - * alias, serial port and UART registers) are handled by the subdevice - * modules themselves. - * - * This is all necessary because IOC4 is not implemented as a multi-function - * PCI device, but an amalgamation of disparate registers for several - * types of device (ATA, serial, external interrupts). The normal - * resource management in the kernel doesn't have quite the right interfaces - * to handle this situation (e.g. multiple modules can't claim the same - * PCI ID), thus this IOC4 master module. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -/*************** - * Definitions * - ***************/ - -/* Tweakable values */ - -/* PCI bus speed detection/calibration */ -#define IOC4_CALIBRATE_COUNT 63 /* Calibration cycle period */ -#define IOC4_CALIBRATE_CYCLES 256 /* Average over this many cycles */ -#define IOC4_CALIBRATE_DISCARD 2 /* Discard first few cycles */ -#define IOC4_CALIBRATE_LOW_MHZ 25 /* Lower bound on bus speed sanity */ -#define IOC4_CALIBRATE_HIGH_MHZ 75 /* Upper bound on bus speed sanity */ -#define IOC4_CALIBRATE_DEFAULT_MHZ 66 /* Assumed if sanity check fails */ - -/************************ - * Submodule management * - ************************/ - -static DEFINE_MUTEX(ioc4_mutex); - -static LIST_HEAD(ioc4_devices); -static LIST_HEAD(ioc4_submodules); - -/* Register an IOC4 submodule */ -int -ioc4_register_submodule(struct ioc4_submodule *is) -{ - struct ioc4_driver_data *idd; - - mutex_lock(&ioc4_mutex); - list_add(&is->is_list, &ioc4_submodules); - - /* Initialize submodule for each IOC4 */ - if (!is->is_probe) - goto out; - - list_for_each_entry(idd, &ioc4_devices, idd_list) { - if (is->is_probe(idd)) { - printk(KERN_WARNING - "%s: IOC4 submodule %s probe failed " - "for pci_dev %s", - __FUNCTION__, module_name(is->is_owner), - pci_name(idd->idd_pdev)); - } - } - out: - mutex_unlock(&ioc4_mutex); - return 0; -} - -/* Unregister an IOC4 submodule */ -void -ioc4_unregister_submodule(struct ioc4_submodule *is) -{ - struct ioc4_driver_data *idd; - - mutex_lock(&ioc4_mutex); - list_del(&is->is_list); - - /* Remove submodule for each IOC4 */ - if (!is->is_remove) - goto out; - - list_for_each_entry(idd, &ioc4_devices, idd_list) { - if (is->is_remove(idd)) { - printk(KERN_WARNING - "%s: IOC4 submodule %s remove failed " - "for pci_dev %s.\n", - __FUNCTION__, module_name(is->is_owner), - pci_name(idd->idd_pdev)); - } - } - out: - mutex_unlock(&ioc4_mutex); -} - -/********************* - * Device management * - *********************/ - -#define IOC4_CALIBRATE_LOW_LIMIT \ - (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_LOW_MHZ) -#define IOC4_CALIBRATE_HIGH_LIMIT \ - (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_HIGH_MHZ) -#define IOC4_CALIBRATE_DEFAULT \ - (1000*IOC4_EXTINT_COUNT_DIVISOR/IOC4_CALIBRATE_DEFAULT_MHZ) - -#define IOC4_CALIBRATE_END \ - (IOC4_CALIBRATE_CYCLES + IOC4_CALIBRATE_DISCARD) - -#define IOC4_INT_OUT_MODE_TOGGLE 0x7 /* Toggle INT_OUT every COUNT+1 ticks */ - -/* Determines external interrupt output clock period of the PCI bus an - * IOC4 is attached to. This value can be used to determine the PCI - * bus speed. - * - * IOC4 has a design feature that various internal timers are derived from - * the PCI bus clock. This causes IOC4 device drivers to need to take the - * bus speed into account when setting various register values (e.g. INT_OUT - * register COUNT field, UART divisors, etc). Since this information is - * needed by several subdrivers, it is determined by the main IOC4 driver, - * even though the following code utilizes external interrupt registers - * to perform the speed calculation. - */ -static void -ioc4_clock_calibrate(struct ioc4_driver_data *idd) -{ - union ioc4_int_out int_out; - union ioc4_gpcr gpcr; - unsigned int state, last_state = 1; - struct timespec start_ts, end_ts; - uint64_t start, end, period; - unsigned int count = 0; - - /* Enable output */ - gpcr.raw = 0; - gpcr.fields.dir = IOC4_GPCR_DIR_0; - gpcr.fields.int_out_en = 1; - writel(gpcr.raw, &idd->idd_misc_regs->gpcr_s.raw); - - /* Reset to power-on state */ - writel(0, &idd->idd_misc_regs->int_out.raw); - mmiowb(); - - /* Set up square wave */ - int_out.raw = 0; - int_out.fields.count = IOC4_CALIBRATE_COUNT; - int_out.fields.mode = IOC4_INT_OUT_MODE_TOGGLE; - int_out.fields.diag = 0; - writel(int_out.raw, &idd->idd_misc_regs->int_out.raw); - mmiowb(); - - /* Check square wave period averaged over some number of cycles */ - do { - int_out.raw = readl(&idd->idd_misc_regs->int_out.raw); - state = int_out.fields.int_out; - if (!last_state && state) { - count++; - if (count == IOC4_CALIBRATE_END) { - ktime_get_ts(&end_ts); - break; - } else if (count == IOC4_CALIBRATE_DISCARD) - ktime_get_ts(&start_ts); - } - last_state = state; - } while (1); - - /* Calculation rearranged to preserve intermediate precision. - * Logically: - * 1. "end - start" gives us the measurement period over all - * the square wave cycles. - * 2. Divide by number of square wave cycles to get the period - * of a square wave cycle. - * 3. Divide by 2*(int_out.fields.count+1), which is the formula - * by which the IOC4 generates the square wave, to get the - * period of an IOC4 INT_OUT count. - */ - end = end_ts.tv_sec * NSEC_PER_SEC + end_ts.tv_nsec; - start = start_ts.tv_sec * NSEC_PER_SEC + start_ts.tv_nsec; - period = (end - start) / - (IOC4_CALIBRATE_CYCLES * 2 * (IOC4_CALIBRATE_COUNT + 1)); - - /* Bounds check the result. */ - if (period > IOC4_CALIBRATE_LOW_LIMIT || - period < IOC4_CALIBRATE_HIGH_LIMIT) { - printk(KERN_INFO - "IOC4 %s: Clock calibration failed. Assuming" - "PCI clock is %d ns.\n", - pci_name(idd->idd_pdev), - IOC4_CALIBRATE_DEFAULT / IOC4_EXTINT_COUNT_DIVISOR); - period = IOC4_CALIBRATE_DEFAULT; - } else { - printk(KERN_DEBUG - "IOC4 %s: PCI clock is %ld ns.\n", - pci_name(idd->idd_pdev), - period / IOC4_EXTINT_COUNT_DIVISOR); - } - - /* Remember results. We store the extint clock period rather - * than the PCI clock period so that greater precision is - * retained. Divide by IOC4_EXTINT_COUNT_DIVISOR to get - * PCI clock period. - */ - idd->count_period = period; -} - -/* There are three variants of IOC4 cards: IO9, IO10, and PCI-RT. - * Each brings out different combinations of IOC4 signals, thus. - * the IOC4 subdrivers need to know to which we're attached. - * - * We look for the presence of a SCSI (IO9) or SATA (IO10) controller - * on the same PCI bus at slot number 3 to differentiate IO9 from IO10. - * If neither is present, it's a PCI-RT. - */ -static unsigned int -ioc4_variant(struct ioc4_driver_data *idd) -{ - struct pci_dev *pdev = NULL; - int found = 0; - - /* IO9: Look for a QLogic ISP 12160 at the same bus and slot 3. */ - do { - pdev = pci_get_device(PCI_VENDOR_ID_QLOGIC, - PCI_DEVICE_ID_QLOGIC_ISP12160, pdev); - if (pdev && - idd->idd_pdev->bus->number == pdev->bus->number && - 3 == PCI_SLOT(pdev->devfn)) - found = 1; - pci_dev_put(pdev); - } while (pdev && !found); - if (NULL != pdev) - return IOC4_VARIANT_IO9; - - /* IO10: Look for a Vitesse VSC 7174 at the same bus and slot 3. */ - pdev = NULL; - do { - pdev = pci_get_device(PCI_VENDOR_ID_VITESSE, - PCI_DEVICE_ID_VITESSE_VSC7174, pdev); - if (pdev && - idd->idd_pdev->bus->number == pdev->bus->number && - 3 == PCI_SLOT(pdev->devfn)) - found = 1; - pci_dev_put(pdev); - } while (pdev && !found); - if (NULL != pdev) - return IOC4_VARIANT_IO10; - - /* PCI-RT: No SCSI/SATA controller will be present */ - return IOC4_VARIANT_PCI_RT; -} - -/* Adds a new instance of an IOC4 card */ -static int -ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) -{ - struct ioc4_driver_data *idd; - struct ioc4_submodule *is; - uint32_t pcmd; - int ret; - - /* Enable IOC4 and take ownership of it */ - if ((ret = pci_enable_device(pdev))) { - printk(KERN_WARNING - "%s: Failed to enable IOC4 device for pci_dev %s.\n", - __FUNCTION__, pci_name(pdev)); - goto out; - } - pci_set_master(pdev); - - /* Set up per-IOC4 data */ - idd = kmalloc(sizeof(struct ioc4_driver_data), GFP_KERNEL); - if (!idd) { - printk(KERN_WARNING - "%s: Failed to allocate IOC4 data for pci_dev %s.\n", - __FUNCTION__, pci_name(pdev)); - ret = -ENODEV; - goto out_idd; - } - idd->idd_pdev = pdev; - idd->idd_pci_id = pci_id; - - /* Map IOC4 misc registers. These are shared between subdevices - * so the main IOC4 module manages them. - */ - idd->idd_bar0 = pci_resource_start(idd->idd_pdev, 0); - if (!idd->idd_bar0) { - printk(KERN_WARNING - "%s: Unable to find IOC4 misc resource " - "for pci_dev %s.\n", - __FUNCTION__, pci_name(idd->idd_pdev)); - ret = -ENODEV; - goto out_pci; - } - if (!request_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs), - "ioc4_misc")) { - printk(KERN_WARNING - "%s: Unable to request IOC4 misc region " - "for pci_dev %s.\n", - __FUNCTION__, pci_name(idd->idd_pdev)); - ret = -ENODEV; - goto out_pci; - } - idd->idd_misc_regs = ioremap(idd->idd_bar0, - sizeof(struct ioc4_misc_regs)); - if (!idd->idd_misc_regs) { - printk(KERN_WARNING - "%s: Unable to remap IOC4 misc region " - "for pci_dev %s.\n", - __FUNCTION__, pci_name(idd->idd_pdev)); - ret = -ENODEV; - goto out_misc_region; - } - - /* Failsafe portion of per-IOC4 initialization */ - - /* Detect card variant */ - idd->idd_variant = ioc4_variant(idd); - printk(KERN_INFO "IOC4 %s: %s card detected.\n", pci_name(pdev), - idd->idd_variant == IOC4_VARIANT_IO9 ? "IO9" : - idd->idd_variant == IOC4_VARIANT_PCI_RT ? "PCI-RT" : - idd->idd_variant == IOC4_VARIANT_IO10 ? "IO10" : "unknown"); - - /* Initialize IOC4 */ - pci_read_config_dword(idd->idd_pdev, PCI_COMMAND, &pcmd); - pci_write_config_dword(idd->idd_pdev, PCI_COMMAND, - pcmd | PCI_COMMAND_PARITY | PCI_COMMAND_SERR); - - /* Determine PCI clock */ - ioc4_clock_calibrate(idd); - - /* Disable/clear all interrupts. Need to do this here lest - * one submodule request the shared IOC4 IRQ, but interrupt - * is generated by a different subdevice. - */ - /* Disable */ - writel(~0, &idd->idd_misc_regs->other_iec.raw); - writel(~0, &idd->idd_misc_regs->sio_iec); - /* Clear (i.e. acknowledge) */ - writel(~0, &idd->idd_misc_regs->other_ir.raw); - writel(~0, &idd->idd_misc_regs->sio_ir); - - /* Track PCI-device specific data */ - idd->idd_serial_data = NULL; - pci_set_drvdata(idd->idd_pdev, idd); - - mutex_lock(&ioc4_mutex); - list_add_tail(&idd->idd_list, &ioc4_devices); - - /* Add this IOC4 to all submodules */ - list_for_each_entry(is, &ioc4_submodules, is_list) { - if (is->is_probe && is->is_probe(idd)) { - printk(KERN_WARNING - "%s: IOC4 submodule 0x%s probe failed " - "for pci_dev %s.\n", - __FUNCTION__, module_name(is->is_owner), - pci_name(idd->idd_pdev)); - } - } - mutex_unlock(&ioc4_mutex); - - return 0; - -out_misc_region: - release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs)); -out_pci: - kfree(idd); -out_idd: - pci_disable_device(pdev); -out: - return ret; -} - -/* Removes a particular instance of an IOC4 card. */ -static void -ioc4_remove(struct pci_dev *pdev) -{ - struct ioc4_submodule *is; - struct ioc4_driver_data *idd; - - idd = pci_get_drvdata(pdev); - - /* Remove this IOC4 from all submodules */ - mutex_lock(&ioc4_mutex); - list_for_each_entry(is, &ioc4_submodules, is_list) { - if (is->is_remove && is->is_remove(idd)) { - printk(KERN_WARNING - "%s: IOC4 submodule 0x%s remove failed " - "for pci_dev %s.\n", - __FUNCTION__, module_name(is->is_owner), - pci_name(idd->idd_pdev)); - } - } - mutex_unlock(&ioc4_mutex); - - /* Release resources */ - iounmap(idd->idd_misc_regs); - if (!idd->idd_bar0) { - printk(KERN_WARNING - "%s: Unable to get IOC4 misc mapping for pci_dev %s. " - "Device removal may be incomplete.\n", - __FUNCTION__, pci_name(idd->idd_pdev)); - } - release_region(idd->idd_bar0, sizeof(struct ioc4_misc_regs)); - - /* Disable IOC4 and relinquish */ - pci_disable_device(pdev); - - /* Remove and free driver data */ - mutex_lock(&ioc4_mutex); - list_del(&idd->idd_list); - mutex_unlock(&ioc4_mutex); - kfree(idd); -} - -static struct pci_device_id ioc4_id_table[] = { - {PCI_VENDOR_ID_SGI, PCI_DEVICE_ID_SGI_IOC4, PCI_ANY_ID, - PCI_ANY_ID, 0x0b4000, 0xFFFFFF}, - {0} -}; - -static struct pci_driver ioc4_driver = { - .name = "IOC4", - .id_table = ioc4_id_table, - .probe = ioc4_probe, - .remove = ioc4_remove, -}; - -MODULE_DEVICE_TABLE(pci, ioc4_id_table); - -/********************* - * Module management * - *********************/ - -/* Module load */ -static int __devinit -ioc4_init(void) -{ - return pci_register_driver(&ioc4_driver); -} - -/* Module unload */ -static void __devexit -ioc4_exit(void) -{ - pci_unregister_driver(&ioc4_driver); -} - -module_init(ioc4_init); -module_exit(ioc4_exit); - -MODULE_AUTHOR("Brent Casavant - Silicon Graphics, Inc. "); -MODULE_DESCRIPTION("PCI driver master module for SGI IOC4 Base-IO Card"); -MODULE_LICENSE("GPL"); - -EXPORT_SYMBOL(ioc4_register_submodule); -EXPORT_SYMBOL(ioc4_unregister_submodule); -- cgit v0.10.2 From 623a43952abfad2d48f287d1fab07b2089d07554 Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Tue, 17 Oct 2006 00:09:27 -0700 Subject: [PATCH] synclink: remove PAGE_SIZE reference Remove reference to PAGE_SIZE that causes errors if PAGE_SIZE != 4096 Signed-off-by: Paul Fulghum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index f2864cc..06784ad 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -133,8 +133,8 @@ static MGSL_PARAMS default_params = { }; #define SHARED_MEM_ADDRESS_SIZE 0x40000 -#define BUFFERLISTSIZE (PAGE_SIZE) -#define DMABUFFERSIZE (PAGE_SIZE) +#define BUFFERLISTSIZE 4096 +#define DMABUFFERSIZE 4096 #define MAXRXFRAMES 7 typedef struct _DMABUFFERENTRY -- cgit v0.10.2 From ca268c691de95612981b93e58899c1d73fdb6b47 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 17 Oct 2006 00:09:28 -0700 Subject: [PATCH] lockdep: increase max allowed recursion depth In general, lockdep warnings are intended to be non-fatal, so I have put in various practical limits on internal data structure failure modes. We haven't had a /single/ lockdep-internal crash ever since lockdep went upstream [the unwinder crashes are outside of lockdep], and that's largely due to the good internal checks it does. Recursion within the dependency graph is currently limited to 20, that's probably not enough on some many-CPU boxes - this patch doubles it to 40. I have written the lockdep functions to have as small stackframes as possible, so 40 should be OK too. (The practical recursion limit should be somewhere between 100 and 200 entries. If we hit that then I'll change the algorithm to be iteration-based. Graph walking logic is so easy to program via recursion, so i'd like to keep recursion as long as possible.) Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 805a322..d1a3b2c 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -575,6 +575,8 @@ static noinline int print_circular_bug_tail(void) return 0; } +#define RECURSION_LIMIT 40 + static int noinline print_infinite_recursion_bug(void) { __raw_spin_unlock(&hash_lock); @@ -595,7 +597,7 @@ check_noncircular(struct lock_class *source, unsigned int depth) debug_atomic_inc(&nr_cyclic_check_recursions); if (depth > max_recursion_depth) max_recursion_depth = depth; - if (depth >= 20) + if (depth >= RECURSION_LIMIT) return print_infinite_recursion_bug(); /* * Check this lock's dependency list: @@ -645,7 +647,7 @@ find_usage_forwards(struct lock_class *source, unsigned int depth) if (depth > max_recursion_depth) max_recursion_depth = depth; - if (depth >= 20) + if (depth >= RECURSION_LIMIT) return print_infinite_recursion_bug(); debug_atomic_inc(&nr_find_usage_forwards_checks); @@ -684,7 +686,7 @@ find_usage_backwards(struct lock_class *source, unsigned int depth) if (depth > max_recursion_depth) max_recursion_depth = depth; - if (depth >= 20) + if (depth >= RECURSION_LIMIT) return print_infinite_recursion_bug(); debug_atomic_inc(&nr_find_usage_backwards_checks); -- cgit v0.10.2 From b9f5d8040b470e4f74e3266ece2969ef06d0f8aa Mon Sep 17 00:00:00 2001 From: Pierre Ossman Date: Tue, 17 Oct 2006 00:09:30 -0700 Subject: [PATCH] New MMC maintainer I will be taking over after Russell King as the new maintainer of the MMC layer. Signed-off-by: Pierre Ossman Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 208da3c..5305dd6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2012,8 +2012,11 @@ L: linux-kernel@vger.kernel.org W: http://www.atnf.csiro.au/~rgooch/linux/kernel-patches.html S: Maintained -MULTIMEDIA CARD (MMC) SUBSYSTEM -S: Orphan +MULTIMEDIA CARD (MMC) AND SECURE DIGITAL (SD) SUBSYSTEM +P: Pierre Ossman +M: drzeus-mmc@drzeus.cx +L: linux-kernel@vger.kernel.org +S: Maintained MULTISOUND SOUND DRIVER P: Andrew Veliath -- cgit v0.10.2 From 3f4a0b917ce72ef47e438d354c433eb645218e87 Mon Sep 17 00:00:00 2001 From: john stultz Date: Tue, 17 Oct 2006 00:09:32 -0700 Subject: [PATCH] i386 Time: Avoid PIT SMP lockups Avoid possible PIT livelock issues seen on SMP systems (and reported by Andi), by not allowing it as a clocksource on SMP boxes. However, since the PIT may no longer be present, we have to properly handle the cases where SMP systems have TSC skew and fall back from the TSC. Since the PIT isn't there, it would "fall back" to the TSC again. So this changes the jiffies rating to 1, and the TSC-bad rating value to 0. Thus you will get the following behavior priority on i386 systems: tsc [if present & stable] hpet [if present] cyclone [if present] acpi_pm [if present] pit [if UP] jiffies Rather then the current more complicated: tsc [if present & stable] hpet [if present] cyclone [if present] acpi_pm [if present] pit [if cpus < 4] tsc [if present & unstable] jiffies Signed-off-by: John Stultz Cc: Andi Kleen Cc: Ingo Molnar Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/i8253.c b/arch/i386/kernel/i8253.c index 477b24d..9a0060b 100644 --- a/arch/i386/kernel/i8253.c +++ b/arch/i386/kernel/i8253.c @@ -109,7 +109,7 @@ static struct clocksource clocksource_pit = { static int __init init_pit_clocksource(void) { - if (num_possible_cpus() > 4) /* PIT does not scale! */ + if (num_possible_cpus() > 1) /* PIT does not scale! */ return 0; clocksource_pit.mult = clocksource_hz2mult(CLOCK_TICK_RATE, 20); diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c index b8fa0a8..fbc9582 100644 --- a/arch/i386/kernel/tsc.c +++ b/arch/i386/kernel/tsc.c @@ -349,8 +349,8 @@ static int tsc_update_callback(void) int change = 0; /* check to see if we should switch to the safe clocksource: */ - if (clocksource_tsc.rating != 50 && check_tsc_unstable()) { - clocksource_tsc.rating = 50; + if (clocksource_tsc.rating != 0 && check_tsc_unstable()) { + clocksource_tsc.rating = 0; clocksource_reselect(); change = 1; } @@ -461,7 +461,7 @@ static int __init init_tsc_clocksource(void) clocksource_tsc.shift); /* lower the rating if we already know its unstable: */ if (check_tsc_unstable()) - clocksource_tsc.rating = 50; + clocksource_tsc.rating = 0; init_timer(&verify_tsc_freq_timer); verify_tsc_freq_timer.function = verify_tsc_freq; diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index 126bb30..a99b2a6 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -57,7 +57,7 @@ static cycle_t jiffies_read(void) struct clocksource clocksource_jiffies = { .name = "jiffies", - .rating = 0, /* lowest rating*/ + .rating = 1, /* lowest valid rating*/ .read = jiffies_read, .mask = 0xffffffff, /*32bits*/ .mult = NSEC_PER_JIFFY << JIFFIES_SHIFT, /* details above */ -- cgit v0.10.2 From 5eb30790d4ccd3409240a80eaf9ab76b4fb75fd8 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Tue, 17 Oct 2006 00:09:35 -0700 Subject: [PATCH] null dereference in fs/jbd2/journal.c This is Eric Sesterhenn's jbd patch applied to jbd2. Commit: 41716c7c21b15e7ecf14f0caf1eef3980707fb74 His words: Since commit d1807793e1e7e502e3dc047115e9dbc3b50e4534 we dereference a NULL pointer. Coverity id #1432. We set journal to NULL, and use it directly afterwards. Signed-off-by: Dave Kleikamp Cc: Eric Sesterhenn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 10db92c..c60f378 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -725,6 +725,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, __FUNCTION__); kfree(journal); journal = NULL; + goto out; } journal->j_dev = bdev; journal->j_fs_dev = fs_dev; @@ -735,7 +736,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, J_ASSERT(bh != NULL); journal->j_sb_buffer = bh; journal->j_superblock = (journal_superblock_t *)bh->b_data; - +out: return journal; } -- cgit v0.10.2 From a649fd9271773dd0f78e2b9f347bcceecb8827f9 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Oct 2006 00:09:36 -0700 Subject: [PATCH] invalidate: remove_mapping() fix If remove_mapping() failed to remove the page from its mapping, don't go and mark it not uptodate! Makes kernel go dead. (Actually, I don't think the ClearPageUptodate is needed there at all). Says Nick Piggin: "Right, it isn't needed because at this point the page is guaranteed by remove_mapping to have no references (except us) and cannot pick up any new ones because it is removed from pagecache. We can delete it." Signed-off-by: Andrew Morton Acked-by: Nick Piggin Signed-off-by: Linus Torvalds diff --git a/mm/truncate.c b/mm/truncate.c index 11ca480..e07b1e6 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -96,7 +96,6 @@ invalidate_complete_page(struct address_space *mapping, struct page *page) return 0; ret = remove_mapping(mapping, page); - ClearPageUptodate(page); return ret; } diff --git a/mm/vmscan.c b/mm/vmscan.c index eca7031..af73c14 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -378,6 +378,12 @@ static pageout_t pageout(struct page *page, struct address_space *mapping) return PAGE_CLEAN; } +/* + * Attempt to detach a locked page from its ->mapping. If it is dirty or if + * someone else has a ref on the page, abort and return 0. If it was + * successfully detached, return 1. Assumes the caller has a single ref on + * this page. + */ int remove_mapping(struct address_space *mapping, struct page *page) { BUG_ON(!PageLocked(page)); -- cgit v0.10.2 From e24650c2e744f99541125a5b023f0d02cad19d14 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Tue, 17 Oct 2006 00:09:38 -0700 Subject: [PATCH] md: fix /proc/mdstat refcounting I have seen mdadm oops after successfully unloading md module. This patch privents from unloading md module while mdadm is polling /proc/mdstat. Cc: Neil Brown Signed-off-by: Akinbou Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index 57fa64f..f7f1908 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -4912,6 +4912,7 @@ static unsigned int mdstat_poll(struct file *filp, poll_table *wait) } static struct file_operations md_seq_fops = { + .owner = THIS_MODULE, .open = md_seq_open, .read = seq_read, .llseek = seq_lseek, -- cgit v0.10.2 From ac08c26492a0ad4d94a25bd47d5630cd38337069 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Oct 2006 00:09:39 -0700 Subject: [PATCH] posix-cpu-timers: prevent signal delivery starvation The integer divisions in the timer accounting code can round the result down to 0. Adding 0 is without effect and the signal delivery stops. Clamp the division result to minimum 1 to avoid this. Problem was reported by Seongbae Park , who provided also an inital patch. Roland sayeth: I have had some more time to think about the problem, and to reproduce it using Toyo's test case. For the record, if my understanding of the problem is correct, this happens only in one very particular case. First, the expiry time has to be so soon that in cputime_t units (usually 1s/HZ ticks) it's < nthreads so the division yields zero. Second, it only affects each thread that is so new that its CPU time accumulation is zero so now+0 is still zero and ->it_*_expires winds up staying zero. For the VIRT and PROF clocks when cputime_t is tick granularity (or the SCHED clock on configurations where sched_clock's value only advances on clock ticks), this is not hard to arrange with new threads starting up and blocking before they accumulate a whole tick of CPU time. That's what happens in Toyo's test case. Note that in general it is fine for that division to round down to zero, and set each thread's expiry time to its "now" time. The problem only arises with thread's whose "now" value is still zero, so that now+0 winds up 0 and is interpreted as "not set" instead of ">= now". So it would be a sufficient and more precise fix to just use max(ticks, 1) inside the loop when setting each it_*_expires value. But, it does no harm to round the division up to one and always advance every thread's expiry time. If the thread didn't already fire timers for the expiry time of "now", there is no expectation that it will do so before the next tick anyway. So I followed Thomas's patch in lifting the max out of the loops. This patch also covers the reload cases, which are harder to write a test for (and I didn't try). I've tested it with Toyo's case and it fixes that. [toyoa@mvista.com: fix: min_t -> max_t] Signed-off-by: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Roland McGrath Cc: Daniel Walker Cc: Toyo Abe Cc: john stultz Cc: Roman Zippel Cc: Seongbae Park Cc: Peter Mattis Cc: Rohit Seth Cc: Martin Bligh Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 479b16b..7c3e1e6 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -88,6 +88,19 @@ static inline union cpu_time_count cpu_time_sub(const clockid_t which_clock, } /* + * Divide and limit the result to res >= 1 + * + * This is necessary to prevent signal delivery starvation, when the result of + * the division would be rounded down to 0. + */ +static inline cputime_t cputime_div_non_zero(cputime_t time, unsigned long div) +{ + cputime_t res = cputime_div(time, div); + + return max_t(cputime_t, res, 1); +} + +/* * Update expiry time from increment, and increase overrun count, * given the current clock sample. */ @@ -483,8 +496,8 @@ static void process_timer_rebalance(struct task_struct *p, BUG(); break; case CPUCLOCK_PROF: - left = cputime_div(cputime_sub(expires.cpu, val.cpu), - nthreads); + left = cputime_div_non_zero(cputime_sub(expires.cpu, val.cpu), + nthreads); do { if (likely(!(t->flags & PF_EXITING))) { ticks = cputime_add(prof_ticks(t), left); @@ -498,8 +511,8 @@ static void process_timer_rebalance(struct task_struct *p, } while (t != p); break; case CPUCLOCK_VIRT: - left = cputime_div(cputime_sub(expires.cpu, val.cpu), - nthreads); + left = cputime_div_non_zero(cputime_sub(expires.cpu, val.cpu), + nthreads); do { if (likely(!(t->flags & PF_EXITING))) { ticks = cputime_add(virt_ticks(t), left); @@ -515,6 +528,7 @@ static void process_timer_rebalance(struct task_struct *p, case CPUCLOCK_SCHED: nsleft = expires.sched - val.sched; do_div(nsleft, nthreads); + nsleft = max_t(unsigned long long, nsleft, 1); do { if (likely(!(t->flags & PF_EXITING))) { ns = t->sched_time + nsleft; @@ -1159,12 +1173,13 @@ static void check_process_timers(struct task_struct *tsk, prof_left = cputime_sub(prof_expires, utime); prof_left = cputime_sub(prof_left, stime); - prof_left = cputime_div(prof_left, nthreads); + prof_left = cputime_div_non_zero(prof_left, nthreads); virt_left = cputime_sub(virt_expires, utime); - virt_left = cputime_div(virt_left, nthreads); + virt_left = cputime_div_non_zero(virt_left, nthreads); if (sched_expires) { sched_left = sched_expires - sched_time; do_div(sched_left, nthreads); + sched_left = max_t(unsigned long long, sched_left, 1); } else { sched_left = 0; } -- cgit v0.10.2 From 0187f879ee8d4b914e74ffa3cc5df268311fc2d2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Oct 2006 00:09:41 -0700 Subject: [PATCH] PROC_NUMBUF is wrong Actually, the decimal representation of a 32-bit signed number can take 12 bytes, including the \0. And then some code adds a \n as well, so let's give it 13 bytes. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/base.c b/fs/proc/base.c index 82da55b..26a8f841 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -86,7 +86,7 @@ /* Worst case buffer size needed for holding an integer. */ -#define PROC_NUMBUF 10 +#define PROC_NUMBUF 13 struct pid_entry { int len; -- cgit v0.10.2 From aaa248f6c9c81b2683db7dbb0689cd5ed1c86d88 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 00:09:42 -0700 Subject: [PATCH] rename net_random to random32 Make net_random() more widely available by calling it random32 akpm: hopefully this will permit the removal of carta_random32. That needs confirmation from Stephane - this code looks somewhat more computationally expensive, and has a different (ie: callee-stateful) interface. [akpm@osdl.org: lots of build fixes, cleanups] Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller Cc: Stephane Eranian Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/net.h b/include/linux/net.h index c257f71..15c733b 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -19,6 +19,7 @@ #define _LINUX_NET_H #include +#include #include struct poll_table_struct; @@ -193,9 +194,9 @@ extern int sock_map_fd(struct socket *sock); extern struct socket *sockfd_lookup(int fd, int *err); #define sockfd_put(sock) fput(sock->file) extern int net_ratelimit(void); -extern unsigned long net_random(void); -extern void net_srandom(unsigned long); -extern void net_random_init(void); + +#define net_random() random32() +#define net_srandom(seed) srandom32(seed) extern int kernel_sendmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, size_t num, size_t len); diff --git a/include/linux/random.h b/include/linux/random.h index 5d6456b..0248b30 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -69,6 +69,9 @@ extern struct file_operations random_fops, urandom_fops; unsigned int get_random_int(void); unsigned long randomize_range(unsigned long start, unsigned long end, unsigned long len); +u32 random32(void); +void srandom32(u32 seed); + #endif /* __KERNEL___ */ #endif /* _LINUX_RANDOM_H */ diff --git a/lib/Makefile b/lib/Makefile index 59070db..4b8052f 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -12,7 +12,7 @@ lib-$(CONFIG_SMP) += cpumask.o lib-y += kobject.o kref.o kobject_uevent.o klist.o -obj-y += sort.o parser.o halfmd4.o iomap_copy.o debug_locks.o +obj-y += sort.o parser.o halfmd4.o iomap_copy.o debug_locks.o random32.o ifeq ($(CONFIG_DEBUG_KOBJECT),y) CFLAGS_kobject.o += -DDEBUG diff --git a/lib/random32.c b/lib/random32.c new file mode 100644 index 0000000..4a15ce5 --- /dev/null +++ b/lib/random32.c @@ -0,0 +1,142 @@ +/* + This is a maximally equidistributed combined Tausworthe generator + based on code from GNU Scientific Library 1.5 (30 Jun 2004) + + x_n = (s1_n ^ s2_n ^ s3_n) + + s1_{n+1} = (((s1_n & 4294967294) <<12) ^ (((s1_n <<13) ^ s1_n) >>19)) + s2_{n+1} = (((s2_n & 4294967288) << 4) ^ (((s2_n << 2) ^ s2_n) >>25)) + s3_{n+1} = (((s3_n & 4294967280) <<17) ^ (((s3_n << 3) ^ s3_n) >>11)) + + The period of this generator is about 2^88. + + From: P. L'Ecuyer, "Maximally Equidistributed Combined Tausworthe + Generators", Mathematics of Computation, 65, 213 (1996), 203--213. + + This is available on the net from L'Ecuyer's home page, + + http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme.ps + ftp://ftp.iro.umontreal.ca/pub/simulation/lecuyer/papers/tausme.ps + + There is an erratum in the paper "Tables of Maximally + Equidistributed Combined LFSR Generators", Mathematics of + Computation, 68, 225 (1999), 261--269: + http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme2.ps + + ... the k_j most significant bits of z_j must be non- + zero, for each j. (Note: this restriction also applies to the + computer code given in [4], but was mistakenly not mentioned in + that paper.) + + This affects the seeding procedure by imposing the requirement + s1 > 1, s2 > 7, s3 > 15. + +*/ + +#include +#include +#include +#include + +struct rnd_state { + u32 s1, s2, s3; +}; + +static DEFINE_PER_CPU(struct rnd_state, net_rand_state); + +static u32 __random32(struct rnd_state *state) +{ +#define TAUSWORTHE(s,a,b,c,d) ((s&c)<>b) + + state->s1 = TAUSWORTHE(state->s1, 13, 19, 4294967294UL, 12); + state->s2 = TAUSWORTHE(state->s2, 2, 25, 4294967288UL, 4); + state->s3 = TAUSWORTHE(state->s3, 3, 11, 4294967280UL, 17); + + return (state->s1 ^ state->s2 ^ state->s3); +} + +static void __set_random32(struct rnd_state *state, unsigned long s) +{ + if (s == 0) + s = 1; /* default seed is 1 */ + +#define LCG(n) (69069 * n) + state->s1 = LCG(s); + state->s2 = LCG(state->s1); + state->s3 = LCG(state->s2); + + /* "warm it up" */ + __random32(state); + __random32(state); + __random32(state); + __random32(state); + __random32(state); + __random32(state); +} + +/** + * random32 - pseudo random number generator + * + * A 32 bit pseudo-random number is generated using a fast + * algorithm suitable for simulation. This algorithm is NOT + * considered safe for cryptographic use. + */ +u32 random32(void) +{ + unsigned long r; + struct rnd_state *state = &get_cpu_var(net_rand_state); + r = __random32(state); + put_cpu_var(state); + return r; +} +EXPORT_SYMBOL(random32); + +/** + * srandom32 - add entropy to pseudo random number generator + * @seed: seed value + * + * Add some additional seeding to the random32() pool. + * Note: this pool is per cpu so it only affects current CPU. + */ +void srandom32(u32 entropy) +{ + struct rnd_state *state = &get_cpu_var(net_rand_state); + __set_random32(state, state->s1 ^ entropy); + put_cpu_var(state); +} +EXPORT_SYMBOL(srandom32); + +/* + * Generate some initially weak seeding values to allow + * to start the random32() engine. + */ +static int __init random32_init(void) +{ + int i; + + for_each_possible_cpu(i) { + struct rnd_state *state = &per_cpu(net_rand_state,i); + __set_random32(state, i + jiffies); + } + return 0; +} +core_initcall(random32_init); + +/* + * Generate better values after random number generator + * is fully initalized. + */ +static int __init random32_reseed(void) +{ + int i; + unsigned long seed; + + for_each_possible_cpu(i) { + struct rnd_state *state = &per_cpu(net_rand_state,i); + + get_random_bytes(&seed, sizeof(seed)); + __set_random32(state, seed); + } + return 0; +} +late_initcall(random32_reseed); diff --git a/net/core/dev.c b/net/core/dev.c index 4d891be..81c426a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3502,8 +3502,6 @@ static int __init net_dev_init(void) BUG_ON(!dev_boot_phase); - net_random_init(); - if (dev_proc_init()) goto out; diff --git a/net/core/utils.c b/net/core/utils.c index 94c5d76..d93fe64 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -30,119 +30,6 @@ #include #include -/* - This is a maximally equidistributed combined Tausworthe generator - based on code from GNU Scientific Library 1.5 (30 Jun 2004) - - x_n = (s1_n ^ s2_n ^ s3_n) - - s1_{n+1} = (((s1_n & 4294967294) <<12) ^ (((s1_n <<13) ^ s1_n) >>19)) - s2_{n+1} = (((s2_n & 4294967288) << 4) ^ (((s2_n << 2) ^ s2_n) >>25)) - s3_{n+1} = (((s3_n & 4294967280) <<17) ^ (((s3_n << 3) ^ s3_n) >>11)) - - The period of this generator is about 2^88. - - From: P. L'Ecuyer, "Maximally Equidistributed Combined Tausworthe - Generators", Mathematics of Computation, 65, 213 (1996), 203--213. - - This is available on the net from L'Ecuyer's home page, - - http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme.ps - ftp://ftp.iro.umontreal.ca/pub/simulation/lecuyer/papers/tausme.ps - - There is an erratum in the paper "Tables of Maximally - Equidistributed Combined LFSR Generators", Mathematics of - Computation, 68, 225 (1999), 261--269: - http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme2.ps - - ... the k_j most significant bits of z_j must be non- - zero, for each j. (Note: this restriction also applies to the - computer code given in [4], but was mistakenly not mentioned in - that paper.) - - This affects the seeding procedure by imposing the requirement - s1 > 1, s2 > 7, s3 > 15. - -*/ -struct nrnd_state { - u32 s1, s2, s3; -}; - -static DEFINE_PER_CPU(struct nrnd_state, net_rand_state); - -static u32 __net_random(struct nrnd_state *state) -{ -#define TAUSWORTHE(s,a,b,c,d) ((s&c)<>b) - - state->s1 = TAUSWORTHE(state->s1, 13, 19, 4294967294UL, 12); - state->s2 = TAUSWORTHE(state->s2, 2, 25, 4294967288UL, 4); - state->s3 = TAUSWORTHE(state->s3, 3, 11, 4294967280UL, 17); - - return (state->s1 ^ state->s2 ^ state->s3); -} - -static void __net_srandom(struct nrnd_state *state, unsigned long s) -{ - if (s == 0) - s = 1; /* default seed is 1 */ - -#define LCG(n) (69069 * n) - state->s1 = LCG(s); - state->s2 = LCG(state->s1); - state->s3 = LCG(state->s2); - - /* "warm it up" */ - __net_random(state); - __net_random(state); - __net_random(state); - __net_random(state); - __net_random(state); - __net_random(state); -} - - -unsigned long net_random(void) -{ - unsigned long r; - struct nrnd_state *state = &get_cpu_var(net_rand_state); - r = __net_random(state); - put_cpu_var(state); - return r; -} - - -void net_srandom(unsigned long entropy) -{ - struct nrnd_state *state = &get_cpu_var(net_rand_state); - __net_srandom(state, state->s1^entropy); - put_cpu_var(state); -} - -void __init net_random_init(void) -{ - int i; - - for_each_possible_cpu(i) { - struct nrnd_state *state = &per_cpu(net_rand_state,i); - __net_srandom(state, i+jiffies); - } -} - -static int net_random_reseed(void) -{ - int i; - unsigned long seed; - - for_each_possible_cpu(i) { - struct nrnd_state *state = &per_cpu(net_rand_state,i); - - get_random_bytes(&seed, sizeof(seed)); - __net_srandom(state, seed); - } - return 0; -} -late_initcall(net_random_reseed); - int net_msg_cost = 5*HZ; int net_msg_burst = 10; @@ -153,10 +40,7 @@ int net_ratelimit(void) { return __printk_ratelimit(net_msg_cost, net_msg_burst); } - -EXPORT_SYMBOL(net_random); EXPORT_SYMBOL(net_ratelimit); -EXPORT_SYMBOL(net_srandom); /* * Convert an ASCII string to binary IP. -- cgit v0.10.2 From 5c496374a72320279ddb86291ef709e090a5d531 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Oct 2006 00:09:44 -0700 Subject: [PATCH] remove carta_random32 This library function should be in obj-y and not in lib-y. But when we do that it clashes unpleasantly with the assembly-language implementation in the ia64 architecture. Instead of trying to fix it all up, just remove the generic carta_random32 in the expectation that the recently-made-generic random32() will suffice. If/when perfmon is migrated to random32, ia64's private carta_random32 implementation can also be removed. Cc: Stephane Eranian Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/Makefile b/lib/Makefile index 4b8052f..cf98fab 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -5,7 +5,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \ bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \ idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \ - sha1.o irq_regs.o carta_random32.o + sha1.o irq_regs.o lib-$(CONFIG_MMU) += ioremap.o lib-$(CONFIG_SMP) += cpumask.o diff --git a/lib/carta_random32.c b/lib/carta_random32.c deleted file mode 100644 index ca82df7..0000000 --- a/lib/carta_random32.c +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Copyright (c) 2002-2006 Hewlett-Packard Development Company, L.P. - * Contributed by David Mosberger-Tang - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of version 2 of the GNU General Public - * License as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA - * 02111-1307 USA - */ -#include -#include - -/* - * Fast, simple, yet decent quality random number generator based on - * a paper by David G. Carta ("Two Fast Implementations of the - * `Minimal Standard' Random Number Generator," Communications of the - * ACM, January, 1990). - */ -u64 carta_random32 (u64 seed) -{ -# define A 16807 -# define M ((u32) 1 << 31) - u64 s, prod = A * seed, p, q; - - p = (prod >> 31) & (M - 1); - q = (prod >> 0) & (M - 1); - s = p + q; - if (s >= M) - s -= M - 1; - return s; -} -EXPORT_SYMBOL_GPL(carta_random32); -- cgit v0.10.2 From 91828a405ae454a9503c41a7744f6ff877a80714 Mon Sep 17 00:00:00 2001 From: "David M. Grimes" Date: Tue, 17 Oct 2006 00:09:45 -0700 Subject: [PATCH] knfsd: add nfs-export support to tmpfs We need to encode a decode the 'file' part of a handle. We simply use the inode number and generation number to construct the filehandle. The generation number is the time when the file was created. As inode numbers cycle through the full 32 bits before being reused, there is no real chance of the same inum being allocated to different files in the same second so this is suitably unique. Using time-of-day rather than e.g. jiffies makes it less likely that the same filehandle can be created after a reboot. In order to be able to decode a filehandle we need to be able to lookup by inum, which means that the inode needs to be added to the inode hash table (tmpfs doesn't currently hash inodes as there is never a need to lookup by inum). To avoid overhead when not exporting, we only hash an inode when it is first exported. This requires a lock to ensure it isn't hashed twice. This code is separate from the patch posted in June06 from Atal Shargorodsky which provided the same functionality, but does borrow slightly from it. Locking comment: Most filesystems that hash their inodes do so at the point where the 'struct inode' is initialised, and that has suitable locking (I_NEW). Here in shmem, we are hashing the inode later, the first time we need an NFS file handle for it. We no longer have I_NEW to ensure only one thread tries to add it to the hash table. Cc: Atal Shargorodsky Cc: Gilad Ben-Yossef Signed-off-by: David M. Grimes Signed-off-by: Neil Brown Acked-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/shmem.c b/mm/shmem.c index bb8ca7e..b378f66 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1362,6 +1362,7 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev) inode->i_mapping->a_ops = &shmem_aops; inode->i_mapping->backing_dev_info = &shmem_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; + inode->i_generation = get_seconds(); info = SHMEM_I(inode); memset(info, 0, (char *)inode - (char *)info); spin_lock_init(&info->lock); @@ -1956,6 +1957,85 @@ static struct xattr_handler *shmem_xattr_handlers[] = { }; #endif +static struct dentry *shmem_get_parent(struct dentry *child) +{ + return ERR_PTR(-ESTALE); +} + +static int shmem_match(struct inode *ino, void *vfh) +{ + __u32 *fh = vfh; + __u64 inum = fh[2]; + inum = (inum << 32) | fh[1]; + return ino->i_ino == inum && fh[0] == ino->i_generation; +} + +static struct dentry *shmem_get_dentry(struct super_block *sb, void *vfh) +{ + struct dentry *de = NULL; + struct inode *inode; + __u32 *fh = vfh; + __u64 inum = fh[2]; + inum = (inum << 32) | fh[1]; + + inode = ilookup5(sb, (unsigned long)(inum+fh[0]), shmem_match, vfh); + if (inode) { + de = d_find_alias(inode); + iput(inode); + } + + return de? de: ERR_PTR(-ESTALE); +} + +static struct dentry *shmem_decode_fh(struct super_block *sb, __u32 *fh, + int len, int type, + int (*acceptable)(void *context, struct dentry *de), + void *context) +{ + if (len < 3) + return ERR_PTR(-ESTALE); + + return sb->s_export_op->find_exported_dentry(sb, fh, NULL, acceptable, + context); +} + +static int shmem_encode_fh(struct dentry *dentry, __u32 *fh, int *len, + int connectable) +{ + struct inode *inode = dentry->d_inode; + + if (*len < 3) + return 255; + + if (hlist_unhashed(&inode->i_hash)) { + /* Unfortunately insert_inode_hash is not idempotent, + * so as we hash inodes here rather than at creation + * time, we need a lock to ensure we only try + * to do it once + */ + static DEFINE_SPINLOCK(lock); + spin_lock(&lock); + if (hlist_unhashed(&inode->i_hash)) + __insert_inode_hash(inode, + inode->i_ino + inode->i_generation); + spin_unlock(&lock); + } + + fh[0] = inode->i_generation; + fh[1] = inode->i_ino; + fh[2] = ((__u64)inode->i_ino) >> 32; + + *len = 3; + return 1; +} + +static struct export_operations shmem_export_ops = { + .get_parent = shmem_get_parent, + .get_dentry = shmem_get_dentry, + .encode_fh = shmem_encode_fh, + .decode_fh = shmem_decode_fh, +}; + static int shmem_parse_options(char *options, int *mode, uid_t *uid, gid_t *gid, unsigned long *blocks, unsigned long *inodes, int *policy, nodemask_t *policy_nodes) @@ -2128,6 +2208,7 @@ static int shmem_fill_super(struct super_block *sb, &inodes, &policy, &policy_nodes)) return -EINVAL; } + sb->s_export_op = &shmem_export_ops; #else sb->s_flags |= MS_NOUSER; #endif -- cgit v0.10.2 From f9b2e97bea228739b74b541033b1119c5707200b Mon Sep 17 00:00:00 2001 From: Evgeniy Polyakov Date: Tue, 17 Oct 2006 00:09:47 -0700 Subject: [PATCH] w1 kconfig fix Remove dependency of w1 subsytem from connector, only w1_con must depend on it. With attached patch applied to vanilla 2.6.19-git things works fine. Signed-off-by: Evgeniy Polyakov Cc: Cc: Greg KH Cc: Roman Zippel Cc: "Randy.Dunlap" Cc: Adrian Bunk Acked-by: Jean Delvare Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/w1/Kconfig b/drivers/w1/Kconfig index 27c9d05..c287a9a 100644 --- a/drivers/w1/Kconfig +++ b/drivers/w1/Kconfig @@ -2,7 +2,6 @@ menu "Dallas's 1-wire bus" config W1 tristate "Dallas's 1-wire support" - depends on CONNECTOR ---help--- Dallas' 1-wire bus is useful to connect slow 1-pin devices such as iButtons and thermal sensors. -- cgit v0.10.2 From c430169e0c9f42f2cd27e0a6161e7ff4dc7e608d Mon Sep 17 00:00:00 2001 From: Francisco Larramendi Date: Tue, 17 Oct 2006 00:09:53 -0700 Subject: [PATCH] rtc-max6902: month conversion fix Fix October-only BCD-to-binary conversion bug: 0x08 -> 7 0x09 -> 8 0x10 -> 15 (!) 0x11 -> 19 Fixes http://bugzilla.kernel.org/show_bug.cgi?id=7361 Cc: Raphael Assenat Cc: Alessandro Zummo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-max6902.c b/drivers/rtc/rtc-max6902.c index 0b20dfa..d941707 100644 --- a/drivers/rtc/rtc-max6902.c +++ b/drivers/rtc/rtc-max6902.c @@ -136,7 +136,7 @@ static int max6902_get_datetime(struct device *dev, struct rtc_time *dt) dt->tm_min = BCD2BIN(chip->buf[2]); dt->tm_hour = BCD2BIN(chip->buf[3]); dt->tm_mday = BCD2BIN(chip->buf[4]); - dt->tm_mon = BCD2BIN(chip->buf[5] - 1); + dt->tm_mon = BCD2BIN(chip->buf[5]) - 1; dt->tm_wday = BCD2BIN(chip->buf[6]); dt->tm_year = BCD2BIN(chip->buf[7]); -- cgit v0.10.2 From 286e1ea3ac1ca4f503ebbb3020bdb0cbe6adffac Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Oct 2006 00:09:57 -0700 Subject: [PATCH] vmalloc(): don't pass __GFP_ZERO to slab A recent change to the vmalloc() code accidentally resulted in us passing __GFP_ZERO into the slab allocator. But we only wanted __GFP_ZERO for the actual pages whcih are being vmalloc()ed, and passing __GFP_ZERO into slab is not a rational thing to ask for. Cc: Jonathan Corbet Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 750ab6e..1133dd3 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -428,8 +428,11 @@ void *__vmalloc_area_node(struct vm_struct *area, gfp_t gfp_mask, if (array_size > PAGE_SIZE) { pages = __vmalloc_node(array_size, gfp_mask, PAGE_KERNEL, node); area->flags |= VM_VPAGES; - } else - pages = kmalloc_node(array_size, (gfp_mask & ~__GFP_HIGHMEM), node); + } else { + pages = kmalloc_node(array_size, + (gfp_mask & ~(__GFP_HIGHMEM | __GFP_ZERO)), + node); + } area->pages = pages; if (!area->pages) { remove_vm_area(area->addr); -- cgit v0.10.2 From 1fec74a9cda95772887c82ede5c0ac60f5be857e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Oct 2006 00:09:58 -0700 Subject: [PATCH] acpi_processor_latency_notifier(): UP warning fix drivers/acpi/processor_idle.c:1112: warning: 'smp_callback' defined but not used Cc: Len Brown Cc: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 526387d..e67144c 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -1108,6 +1108,7 @@ static const struct file_operations acpi_processor_power_fops = { .release = single_release, }; +#ifdef CONFIG_SMP static void smp_callback(void *v) { /* we already woke the CPU up, nothing more to do */ @@ -1129,6 +1130,7 @@ static int acpi_processor_latency_notify(struct notifier_block *b, static struct notifier_block acpi_processor_latency_notifier = { .notifier_call = acpi_processor_latency_notify, }; +#endif int __cpuinit acpi_processor_power_init(struct acpi_processor *pr, struct acpi_device *device) @@ -1146,7 +1148,9 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr, "ACPI: processor limited to max C-state %d\n", max_cstate); first_run++; +#ifdef CONFIG_SMP register_latency_notifier(&acpi_processor_latency_notifier); +#endif } if (!pr) @@ -1218,7 +1222,9 @@ int acpi_processor_power_exit(struct acpi_processor *pr, * copies of pm_idle before proceeding. */ cpu_idle_wait(); +#ifdef CONFIG_SMP unregister_latency_notifier(&acpi_processor_latency_notifier); +#endif } return 0; -- cgit v0.10.2 From c60099bfe3a5e6fa22a930627689b3769c52153f Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 17 Oct 2006 00:09:59 -0700 Subject: [PATCH] swsusp: fix memory leaks My fancy new swsusp IO code had a big memory leak. It's somewhat invisible because the whole mem_map[] gets overwritten after resume, but it can cause us to get low on memory during the actual suspend process. Cc: "Rafael J. Wysocki" Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 9b2ee53..1a3b0dd 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -425,7 +425,8 @@ static int submit(int rw, pgoff_t page_off, struct page *page, bio_set_pages_dirty(bio); bio_put(bio); } else { - get_page(page); + if (rw == READ) + get_page(page); /* These pages are freed later */ bio->bi_private = *bio_chain; *bio_chain = bio; submit_bio(rw | (1 << BIO_RW_SYNC), bio); -- cgit v0.10.2 From a4bb2cf1c3d30e7498e5561b22246b5bcbfe2e15 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 17 Oct 2006 00:10:00 -0700 Subject: [PATCH] drivers/char/specialix.c: fix the baud conversion Correct the following bugs introduced by commit 67cc0161ecc9ebee6eba4af6cbfdba028090b1b9: - remove one remaining and now incorrect baud_table[] usage - "baud +=" is no longer correct The former bug was spotted by the Coverity checker. Rolf Eike Beer spotted a bug in the initial version of my patch. Signed-off-by: Adrian Bunk Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index d0b88d0..7e1bd95 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -183,11 +183,6 @@ static int sx_poll = HZ; static struct tty_driver *specialix_driver; -static unsigned long baud_table[] = { - 0, 50, 75, 110, 134, 150, 200, 300, 600, 1200, 1800, 2400, 4800, - 9600, 19200, 38400, 57600, 115200, 0, -}; - static struct specialix_board sx_board[SX_NBOARD] = { { 0, SX_IOBASE1, 9, }, { 0, SX_IOBASE2, 11, }, @@ -1090,9 +1085,9 @@ static void sx_change_speed(struct specialix_board *bp, struct specialix_port *p if (baud == 38400) { if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI) - baud ++; + baud = 57600; if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI) - baud += 2; + baud = 115200; } if (!baud) { @@ -1150,11 +1145,9 @@ static void sx_change_speed(struct specialix_board *bp, struct specialix_port *p sx_out(bp, CD186x_RBPRL, tmp & 0xff); sx_out(bp, CD186x_TBPRL, tmp & 0xff); spin_unlock_irqrestore(&bp->lock, flags); - if (port->custom_divisor) { + if (port->custom_divisor) baud = (SX_OSCFREQ + port->custom_divisor/2) / port->custom_divisor; - baud = ( baud + 5 ) / 10; - } else - baud = (baud_table[baud] + 5) / 10; /* Estimated CPS */ + baud = (baud + 5) / 10; /* Estimated CPS */ /* Two timer ticks seems enough to wakeup something like SLIP driver */ tmp = ((baud + HZ/2) / HZ) * 2 - CD186x_NFIFO; -- cgit v0.10.2 From 308ba5fcf89b6e328f9290067181c1e4d772fdc9 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Tue, 17 Oct 2006 00:10:02 -0700 Subject: [PATCH] fix `make headers_install' Fix this: make[3]: *** No rule to make target `/mnt/md0/devel/linux-git/include/linux/version.h', needed by `/mnt/md0/devel/linux-git-obj/usr/include/linux/version.h'. Stop. make[2]: *** [linux] Error 2 make[1]: *** [headers_install] Error 2 Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/Makefile.headersinst b/scripts/Makefile.headersinst index 6a026f6..4241e0d 100644 --- a/scripts/Makefile.headersinst +++ b/scripts/Makefile.headersinst @@ -168,7 +168,7 @@ $(objhdr-y) $(header-y) $(unifdef-y): $(KBUILDFILES) $(call cmd,gen) else -$(objhdr-y) : $(INSTALL_HDR_PATH)/$(_dst)/%.h: $(srctree)/$(obj)/%.h $(KBUILDFILES) +$(objhdr-y) : $(INSTALL_HDR_PATH)/$(_dst)/%.h: $(objtree)/$(obj)/%.h $(KBUILDFILES) $(call cmd,o_hdr_install) $(header-y) : $(INSTALL_HDR_PATH)/$(_dst)/%.h: $(srctree)/$(obj)/%.h $(KBUILDFILES) -- cgit v0.10.2 From a460e745e8f9c75a0525ff94154a0629f9d3e05d Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 17 Oct 2006 00:10:03 -0700 Subject: [PATCH] genirq: clean up irq-flow-type naming Introduce desc->name and eliminate the handle_irq_name() hack. Add set_irq_chip_and_handler_name() to set the flow type and name at once. Signed-off-by: Ingo Molnar Acked-by: Thomas Gleixner Cc: "Eric W. Biederman" Cc: Matthew Wilcox Cc: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c index d53eafb..62996cd 100644 --- a/arch/i386/kernel/i8259.c +++ b/arch/i386/kernel/i8259.c @@ -113,7 +113,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c index 0612a33..c4ef801 100644 --- a/arch/x86_64/kernel/i8259.c +++ b/arch/x86_64/kernel/i8259.c @@ -178,7 +178,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) diff --git a/include/linux/irq.h b/include/linux/irq.h index c64f3cc..775f5a7 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -141,6 +141,7 @@ struct irq_chip { * @pending_mask: pending rebalanced interrupts * @dir: /proc/irq/ procfs entry * @affinity_entry: /proc/irq/smp_affinity procfs entry on SMP + * @name: flow handler name for /proc/interrupts output * * Pad this out to 32 bytes for cache and indexing reasons. */ @@ -165,8 +166,9 @@ struct irq_desc { cpumask_t pending_mask; #endif #ifdef CONFIG_PROC_FS - struct proc_dir_entry *dir; + struct proc_dir_entry *dir; #endif + const char *name; } ____cacheline_aligned; extern struct irq_desc irq_desc[NR_IRQS]; @@ -272,12 +274,6 @@ extern void fastcall handle_percpu_irq(unsigned int irq, struct irq_desc *desc); extern void fastcall handle_bad_irq(unsigned int irq, struct irq_desc *desc); /* - * Get a descriptive string for the highlevel handler, for - * /proc/interrupts output: - */ -extern const char *handle_irq_name(irq_flow_handler_t handle); - -/* * Monolithic do_IRQ implementation. * (is an explicit fastcall, because i386 4KSTACKS calls it from assembly) */ @@ -326,10 +322,12 @@ extern struct irq_chip no_irq_chip; extern struct irq_chip dummy_irq_chip; extern void -set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, - irq_flow_handler_t handle); +set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle, const char *name); + extern void -__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained); +__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name); /* * Set a highlevel flow handler for a given IRQ: @@ -337,7 +335,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained); static inline void set_irq_handler(unsigned int irq, irq_flow_handler_t handle) { - __set_irq_handler(irq, handle, 0); + __set_irq_handler(irq, handle, 0, NULL); } /* @@ -349,7 +347,7 @@ static inline void set_irq_chained_handler(unsigned int irq, irq_flow_handler_t handle) { - __set_irq_handler(irq, handle, 1); + __set_irq_handler(irq, handle, 1, NULL); } /* Handle dynamic irq creation and destruction */ diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 11c9969..2d0dc3e 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -499,7 +499,8 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc) #endif /* CONFIG_SMP */ void -__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained) +__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name) { struct irq_desc *desc; unsigned long flags; @@ -540,6 +541,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained) desc->depth = 1; } desc->handle_irq = handle; + desc->name = name; if (handle != handle_bad_irq && is_chained) { desc->status &= ~IRQ_DISABLED; @@ -555,30 +557,13 @@ set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, irq_flow_handler_t handle) { set_irq_chip(irq, chip); - __set_irq_handler(irq, handle, 0); + __set_irq_handler(irq, handle, 0, NULL); } -/* - * Get a descriptive string for the highlevel handler, for - * /proc/interrupts output: - */ -const char * -handle_irq_name(irq_flow_handler_t handle) +void +set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle, const char *name) { - if (handle == handle_level_irq) - return "level "; - if (handle == handle_fasteoi_irq) - return "fasteoi"; - if (handle == handle_edge_irq) - return "edge "; - if (handle == handle_simple_irq) - return "simple "; -#ifdef CONFIG_SMP - if (handle == handle_percpu_irq) - return "percpu "; -#endif - if (handle == handle_bad_irq) - return "bad "; - - return NULL; + set_irq_chip(irq, chip); + __set_irq_handler(irq, handle, 0, name); } -- cgit v0.10.2 From 48d1a7ea6373337985f27dc1c707649469df5827 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 17 Oct 2006 00:10:05 -0700 Subject: [PATCH] sx: fix user-visible typo (devic) Signed-off-by: Alexey Dobriyan Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/sx.c b/drivers/char/sx.c index 5fec626..cc10af0 100644 --- a/drivers/char/sx.c +++ b/drivers/char/sx.c @@ -2602,7 +2602,7 @@ static void __exit sx_exit (void) } } if (misc_deregister(&sx_fw_device) < 0) { - printk (KERN_INFO "sx: couldn't deregister firmware loader devic\n"); + printk (KERN_INFO "sx: couldn't deregister firmware loader device\n"); } sx_dprintk (SX_DEBUG_CLEANUP, "Cleaning up drivers (%d)\n", sx_initialized); if (sx_initialized) -- cgit v0.10.2 From 9ffbb9162312fd8113037cb3d94f787f06bbfa9a Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:06 -0700 Subject: [PATCH] fuse: fix hang on SMP Fuse didn't always call i_size_write() with i_mutex held which caused rare hangs on SMP/32bit. This bug has been present since fuse-2.2, well before being merged into mainline. The simplest solution is to protect i_size_write() with the per-connection spinlock. Using i_mutex for this purpose would require some restructuring of the code and I'm not even sure it's always safe to acquire i_mutex in all places i_size needs to be set. Since most of vmtruncate is already duplicated for other reasons, duplicate the remaining part as well, making all i_size_write() calls internal to fuse. Using i_size_write() was unnecessary in fuse_init_inode(), since this function is only called on a newly created locked inode. Reported by a few people over the years, but special thanks to Dana Henriksen who was persistent enough in helping me debug it. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 8605155..a8f65c1 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -935,14 +935,30 @@ static void iattr_to_fattr(struct iattr *iattr, struct fuse_setattr_in *arg) } } +static void fuse_vmtruncate(struct inode *inode, loff_t offset) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + int need_trunc; + + spin_lock(&fc->lock); + need_trunc = inode->i_size > offset; + i_size_write(inode, offset); + spin_unlock(&fc->lock); + + if (need_trunc) { + struct address_space *mapping = inode->i_mapping; + unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1); + truncate_inode_pages(mapping, offset); + } +} + /* * Set attributes, and at the same time refresh them. * * Truncation is slightly complicated, because the 'truncate' request * may fail, in which case we don't want to touch the mapping. - * vmtruncate() doesn't allow for this case. So do the rlimit - * checking by hand and call vmtruncate() only after the file has - * actually been truncated. + * vmtruncate() doesn't allow for this case, so do the rlimit checking + * and the actual truncation by hand. */ static int fuse_setattr(struct dentry *entry, struct iattr *attr) { @@ -993,12 +1009,8 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr) make_bad_inode(inode); err = -EIO; } else { - if (is_truncate) { - loff_t origsize = i_size_read(inode); - i_size_write(inode, outarg.attr.size); - if (origsize > outarg.attr.size) - vmtruncate(inode, outarg.attr.size); - } + if (is_truncate) + fuse_vmtruncate(inode, outarg.attr.size); fuse_change_attributes(inode, &outarg.attr); fi->i_time = time_to_jiffies(outarg.attr_valid, outarg.attr_valid_nsec); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 1836268..2bb5ace 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -481,8 +481,10 @@ static int fuse_commit_write(struct file *file, struct page *page, err = -EIO; if (!err) { pos += count; - if (pos > i_size_read(inode)) + spin_lock(&fc->lock); + if (pos > inode->i_size) i_size_write(inode, pos); + spin_unlock(&fc->lock); if (offset == 0 && to == PAGE_CACHE_SIZE) { clear_page_dirty(page); @@ -586,8 +588,12 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf, } fuse_put_request(fc, req); if (res > 0) { - if (write && pos > i_size_read(inode)) - i_size_write(inode, pos); + if (write) { + spin_lock(&fc->lock); + if (pos > inode->i_size) + i_size_write(inode, pos); + spin_unlock(&fc->lock); + } *ppos = pos; } fuse_invalidate_attr(inode); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 7d0a9ae..8e10616 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -109,6 +109,7 @@ static int fuse_remount_fs(struct super_block *sb, int *flags, char *data) void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) { + struct fuse_conn *fc = get_fuse_conn(inode); if (S_ISREG(inode->i_mode) && i_size_read(inode) != attr->size) invalidate_inode_pages(inode->i_mapping); @@ -117,7 +118,9 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) inode->i_nlink = attr->nlink; inode->i_uid = attr->uid; inode->i_gid = attr->gid; + spin_lock(&fc->lock); i_size_write(inode, attr->size); + spin_unlock(&fc->lock); inode->i_blocks = attr->blocks; inode->i_atime.tv_sec = attr->atime; inode->i_atime.tv_nsec = attr->atimensec; @@ -130,7 +133,7 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr) { inode->i_mode = attr->mode & S_IFMT; - i_size_write(inode, attr->size); + inode->i_size = attr->size; if (S_ISREG(inode->i_mode)) { fuse_init_common(inode); fuse_init_file_inode(inode); -- cgit v0.10.2 From 7762f5a0b709b415fda132258ad37b9f2a1db994 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:07 -0700 Subject: [PATCH] document i_size_write locking rules Unless someone reads the documentation for write_seqcount_{begin,end} it is not obvious, that i_size_write() needs locking. Especially, that lack of such locking can result in a system hang. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/fs.h b/include/linux/fs.h index 34406ed..661c7c5 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -656,7 +656,11 @@ static inline loff_t i_size_read(struct inode *inode) #endif } - +/* + * NOTE: unlike i_size_read(), i_size_write() does need locking around it + * (normally i_mutex), otherwise on 32bit/SMP an update of i_size_seqcount + * can be lost, resulting in subsequent i_size_read() calls spinning forever. + */ static inline void i_size_write(struct inode *inode, loff_t i_size) { #if BITS_PER_LONG==32 && defined(CONFIG_SMP) -- cgit v0.10.2 From 8da5ff23ce0a84d9845b01e6fe5047e17836bf5a Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:08 -0700 Subject: [PATCH] fuse: locking fix for nlookup An inode could be returned by independent parallel lookups, in this case an update of the lookup counter could be lost resulting in a memory leak in userspace. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index a8f65c1..7ecfe95 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -163,7 +163,9 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) fuse_send_forget(fc, req, outarg.nodeid, 1); return 0; } + spin_lock(&fc->lock); fi->nlookup ++; + spin_unlock(&fc->lock); } fuse_put_request(fc, req); if (err || (outarg.attr.mode ^ inode->i_mode) & S_IFMT) diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 8e10616..e911423 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -195,7 +195,9 @@ struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, } fi = get_fuse_inode(inode); + spin_lock(&fc->lock); fi->nlookup ++; + spin_unlock(&fc->lock); fuse_change_attributes(inode, attr); return inode; } -- cgit v0.10.2 From 265126ba9e1f8e217e61d1017c6609f76828aa7a Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:09 -0700 Subject: [PATCH] fuse: fix spurious BUG Fix a spurious BUG in an unlikely race, where at least three parallel lookups return the same inode, but with different file type. This has not yet been observed in real life. Allowing unlimited retries could delay fuse_iget() indefinitely, but this is really for the broken userspace filesystem to worry about. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index e911423..4ee8f72 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -172,7 +172,6 @@ struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, struct inode *inode; struct fuse_inode *fi; struct fuse_conn *fc = get_fuse_conn_super(sb); - int retried = 0; retry: inode = iget5_locked(sb, nodeid, fuse_inode_eq, fuse_inode_set, &nodeid); @@ -186,11 +185,9 @@ struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, fuse_init_inode(inode, attr); unlock_new_inode(inode); } else if ((inode->i_mode ^ attr->mode) & S_IFMT) { - BUG_ON(retried); /* Inode has changed type, any I/O on the old should fail */ make_bad_inode(inode); iput(inode); - retried = 1; goto retry; } -- cgit v0.10.2 From d2a85164aaa8d514ef5efbf5d05746e85dd13ddd Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:11 -0700 Subject: [PATCH] fuse: fix handling of moved directory Fuse considered it an error (EIO) if lookup returned a directory inode, to which a dentry already refered. This is because directory aliases are not allowed. But in a network filesystem this could happen legitimately, if a directory is moved on a remote client. This patch attempts to relax the restriction by trying to first evict the offending alias from the cache. If this fails, it still returns an error (EBUSY). A rarer situation is if an mkdir races with an indenpendent lookup, which finds the newly created directory already moved. In this situation the mkdir should return success, but that would be incorrect, since the dentry cannot be instantiated, so return EBUSY. Previously checking for a directory alias and instantiation of the dentry weren't done atomically in lookup/mkdir, hence two such calls racing with each other could create aliased directories. To prevent this introduce a new per-connection mutex: fuse_conn->inst_mutex, which is taken for instantiations with a directory inode. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 7ecfe95..9d0ef5e1 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -177,22 +177,6 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) return 1; } -/* - * Check if there's already a hashed alias of this directory inode. - * If yes, then lookup and mkdir must not create a new alias. - */ -static int dir_alias(struct inode *inode) -{ - if (S_ISDIR(inode->i_mode)) { - struct dentry *alias = d_find_alias(inode); - if (alias) { - dput(alias); - return 1; - } - } - return 0; -} - static int invalid_nodeid(u64 nodeid) { return !nodeid || nodeid == FUSE_ROOT_ID; @@ -208,6 +192,24 @@ static int valid_mode(int m) S_ISBLK(m) || S_ISFIFO(m) || S_ISSOCK(m); } +/* + * Add a directory inode to a dentry, ensuring that no other dentry + * refers to this inode. Called with fc->inst_mutex. + */ +static int fuse_d_add_directory(struct dentry *entry, struct inode *inode) +{ + struct dentry *alias = d_find_alias(inode); + if (alias) { + /* This tries to shrink the subtree below alias */ + fuse_invalidate_entry(alias); + dput(alias); + if (!list_empty(&inode->i_dentry)) + return -EBUSY; + } + d_add(entry, inode); + return 0; +} + static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, struct nameidata *nd) { @@ -243,11 +245,17 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, if (err && err != -ENOENT) return ERR_PTR(err); - if (inode && dir_alias(inode)) { - iput(inode); - return ERR_PTR(-EIO); - } - d_add(entry, inode); + if (inode && S_ISDIR(inode->i_mode)) { + mutex_lock(&fc->inst_mutex); + err = fuse_d_add_directory(entry, inode); + mutex_unlock(&fc->inst_mutex); + if (err) { + iput(inode); + return ERR_PTR(err); + } + } else + d_add(entry, inode); + entry->d_op = &fuse_dentry_operations; if (!err) fuse_change_timeout(entry, &outarg); @@ -403,12 +411,22 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, } fuse_put_request(fc, req); - if (dir_alias(inode)) { - iput(inode); - return -EIO; - } + if (S_ISDIR(inode->i_mode)) { + struct dentry *alias; + mutex_lock(&fc->inst_mutex); + alias = d_find_alias(inode); + if (alias) { + /* New directory must have moved since mkdir */ + mutex_unlock(&fc->inst_mutex); + dput(alias); + iput(inode); + return -EBUSY; + } + d_instantiate(entry, inode); + mutex_unlock(&fc->inst_mutex); + } else + d_instantiate(entry, inode); - d_instantiate(entry, inode); fuse_change_timeout(entry, &outarg); fuse_invalidate_attr(dir); return 0; diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 69c7750..91edb89 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -239,6 +239,9 @@ struct fuse_conn { /** Lock protecting accessess to members of this structure */ spinlock_t lock; + /** Mutex protecting against directory alias creation */ + struct mutex inst_mutex; + /** Refcount */ atomic_t count; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 4ee8f72..fc42035 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -379,6 +379,7 @@ static struct fuse_conn *new_conn(void) fc = kzalloc(sizeof(*fc), GFP_KERNEL); if (fc) { spin_lock_init(&fc->lock); + mutex_init(&fc->inst_mutex); atomic_set(&fc->count, 1); init_waitqueue_head(&fc->waitq); init_waitqueue_head(&fc->blocked_waitq); @@ -398,8 +399,10 @@ static struct fuse_conn *new_conn(void) void fuse_conn_put(struct fuse_conn *fc) { - if (atomic_dec_and_test(&fc->count)) + if (atomic_dec_and_test(&fc->count)) { + mutex_destroy(&fc->inst_mutex); kfree(fc); + } } struct fuse_conn *fuse_conn_get(struct fuse_conn *fc) -- cgit v0.10.2 From e956edd0523b6b48ed367c63b0c82d8f4c447a58 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:12 -0700 Subject: [PATCH] fuse: fix dereferencing dentry parent There's no locking for ->d_revalidate, so fuse_dentry_revalidate() should use dget_parent() instead of simply dereferencing ->d_parent. Due to topology changes in the directory tree the parent could become negative or be destroyed while being used. There hasn't been any reports about this yet. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 9d0ef5e1..cfc8f81 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -138,6 +138,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) struct fuse_entry_out outarg; struct fuse_conn *fc; struct fuse_req *req; + struct dentry *parent; /* Doesn't hurt to "reset" the validity timeout */ fuse_invalidate_entry_cache(entry); @@ -151,8 +152,10 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) if (IS_ERR(req)) return 0; - fuse_lookup_init(req, entry->d_parent->d_inode, entry, &outarg); + parent = dget_parent(entry); + fuse_lookup_init(req, parent->d_inode, entry, &outarg); request_send(fc, req); + dput(parent); err = req->out.h.error; /* Zero nodeid is same as -ENOENT */ if (!err && !outarg.nodeid) -- cgit v0.10.2 From dc730e173785e29b297aa605786c94adaffe2544 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 17 Oct 2006 00:10:13 -0700 Subject: [PATCH] knfsd: nfsd4: fix owner-override on open If a client creates a file using an open which sets the mode to 000, or if a chmod changes permissions after a file is opened, then situations may arise where an NFS client knows that some IO is permitted (because a process holds the file open), but the NFS server does not (because it doesn't know about the open, and only sees that the IO conflicts with the current mode of the file). As a hack to solve this problem, NFS servers normally allow the owner to override permissions on IO. The client can still enforce correct permissions-checking on open by performing an explicit access check. In NFSv4 the client can rely on the explicit on-the-wire open instead of an access check. Therefore we should not be allowing the owner to override permissions on an over-the-wire open! However, we should still allow the owner to override permissions in the case where the client is claiming an open that it already made either before a reboot, or while it was holding a delegation. Thanks to Jim Rees for reporting the bug. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 8333db1..a05d337 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -68,20 +68,18 @@ fh_dup2(struct svc_fh *dst, struct svc_fh *src) } static int -do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) +do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, int accmode) { - int accmode, status; + int status; if (open->op_truncate && !(open->op_share_access & NFS4_SHARE_ACCESS_WRITE)) return nfserr_inval; - accmode = MAY_NOP; if (open->op_share_access & NFS4_SHARE_ACCESS_READ) - accmode = MAY_READ; + accmode |= MAY_READ; if (open->op_share_deny & NFS4_SHARE_ACCESS_WRITE) accmode |= (MAY_WRITE | MAY_TRUNC); - accmode |= MAY_OWNER_OVERRIDE; status = fh_verify(rqstp, current_fh, S_IFREG, accmode); @@ -124,7 +122,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o &resfh.fh_handle.fh_base, resfh.fh_handle.fh_size); - status = do_open_permission(rqstp, current_fh, open); + status = do_open_permission(rqstp, current_fh, open, MAY_NOP); } fh_put(&resfh); @@ -155,7 +153,7 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ open->op_truncate = (open->op_iattr.ia_valid & ATTR_SIZE) && (open->op_iattr.ia_size == 0); - status = do_open_permission(rqstp, current_fh, open); + status = do_open_permission(rqstp, current_fh, open, MAY_OWNER_OVERRIDE); return status; } -- cgit v0.10.2 From 9801d8a39cfe6c34f39f9552a246a6bd002e735e Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 17 Oct 2006 00:10:14 -0700 Subject: [PATCH] knfsd: nfsd4: fix open permission checking We weren't actually checking for SHARE_ACCESS_WRITE, with the result that the owner could open a non-writeable file for write! Continue to allow DENY_WRITE only with write access. Thanks to Jim Rees for reporting the bug. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index a05d337..d1fac68 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -78,8 +78,10 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs if (open->op_share_access & NFS4_SHARE_ACCESS_READ) accmode |= MAY_READ; - if (open->op_share_deny & NFS4_SHARE_ACCESS_WRITE) + if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) accmode |= (MAY_WRITE | MAY_TRUNC); + if (open->op_share_deny & NFS4_SHARE_DENY_WRITE) + accmode |= MAY_WRITE; status = fh_verify(rqstp, current_fh, S_IFREG, accmode); -- cgit v0.10.2 From 0942176f4353ffebcd6e3f95abce9fd8e24f2cb1 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 17 Oct 2006 00:10:16 -0700 Subject: [PATCH] knfsd: nfsd4: Fix error handling in nfsd's callback client Coverity noticed that the error handling code in the NFSv4 callback client sets cb->cb_client to NULL, then calls rpc_shutdown_client with the NULL pointer. Coverity: #cid 1397 Signed-off-by: Chuck Lever Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index f6ca9fb..324a278 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -421,7 +421,7 @@ nfsd4_probe_callback(struct nfs4_client *clp) /* Create RPC client */ cb->cb_client = rpc_create(&args); - if (!cb->cb_client) { + if (IS_ERR(cb->cb_client)) { dprintk("NFSD: couldn't create callback client\n"); goto out_err; } @@ -448,10 +448,10 @@ nfsd4_probe_callback(struct nfs4_client *clp) out_rpciod: atomic_dec(&clp->cl_count); rpciod_down(); - cb->cb_client = NULL; out_clnt: rpc_shutdown_client(cb->cb_client); out_err: + cb->cb_client = NULL; dprintk("NFSD: warning: no callback path to client %.*s\n", (int)clp->cl_name.len, clp->cl_name.data); } -- cgit v0.10.2 From 4481d1038f4116f3f5c307d919e6dc815a3acbb9 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 17 Oct 2006 00:10:17 -0700 Subject: [PATCH] knfsd: Fix bug in recent lockd patches that can cause reclaim to fail When an nfs server shuts down, lockd needs to release all the locks even though the client still holds them. It should therefore not 'unmonitor' the clients, so that the files in nfs/sm will still be there when the nfs server restarts, so that those clients will be told to reclaim their locks. However the hosts are fully unmonitored, so statd may well remove the files. lockd has a test for 'sm_sticky' and avoid the unmonitor call if it is set, but it is currently not set. So set it when tearing down lockd. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 514f5f2..c5f9113 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -324,7 +324,16 @@ nlmsvc_same_host(struct nlm_host *host, struct nlm_host *other) static int nlmsvc_is_client(struct nlm_host *host, struct nlm_host *dummy) { - return host->h_server; + if (host->h_server) { + /* we are destroying locks even though the client + * hasn't asked us too, so don't unmonitor the + * client + */ + if (host->h_nsmhandle) + host->h_nsmhandle->sm_sticky = 1; + return 1; + } else + return 0; } /* -- cgit v0.10.2 From d343fce148a4eee24a907a05c4101d3268045aae Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 17 Oct 2006 00:10:18 -0700 Subject: [PATCH] knfsd: Allow lockd to drop replies as appropriate It is possible for the ->fopen callback from lockd into nfsd to find that an answer cannot be given straight away (an upcall is needed) and so the request has to be 'dropped', to be retried later. That error status is not currently propagated back. So: Change nlm_fopen to return nlm error codes (rather than a private protocol) and define a new nlm_drop_reply code. Cause nlm_drop_reply to cause the rpc request to get rpc_drop_reply when this error comes back. Cause svc_process to drop a request which returns a status of rpc_drop_reply. [akpm@osdl.org: fix warning storm] Cc: Marc Eshel Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index fa370f6..399ad11 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -96,7 +96,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now check for conflicting locks */ resp->status = nlmsvc_testlock(file, &argp->lock, &resp->lock); @@ -126,7 +126,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; #if 0 /* If supplied state doesn't match current state, we assume it's @@ -169,7 +169,7 @@ nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Try to cancel request. */ resp->status = nlmsvc_cancel_blocked(file, &argp->lock); @@ -202,7 +202,7 @@ nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to remove the lock */ resp->status = nlmsvc_unlock(file, &argp->lock); @@ -339,7 +339,7 @@ nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to create the share */ resp->status = nlmsvc_share_file(host, file, argp); @@ -372,7 +372,7 @@ nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to lock the file */ resp->status = nlmsvc_unshare_file(host, file, argp); diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 75b2c81..6a931f4 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -59,7 +59,7 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error; + u32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) @@ -88,6 +88,8 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, no_locks: if (host) nlm_release_host(host); + if (error) + return error; return nlm_lck_denied_nolocks; } @@ -122,7 +124,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now check for conflicting locks */ resp->status = cast_status(nlmsvc_testlock(file, &argp->lock, &resp->lock)); @@ -153,7 +155,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; #if 0 /* If supplied state doesn't match current state, we assume it's @@ -196,7 +198,7 @@ nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Try to cancel request. */ resp->status = cast_status(nlmsvc_cancel_blocked(file, &argp->lock)); @@ -229,7 +231,7 @@ nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to remove the lock */ resp->status = cast_status(nlmsvc_unlock(file, &argp->lock)); @@ -368,7 +370,7 @@ nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to create the share */ resp->status = cast_status(nlmsvc_share_file(host, file, argp)); @@ -401,7 +403,7 @@ nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to unshare the file */ resp->status = cast_status(nlmsvc_unshare_file(host, file, argp)); diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index c5f9113..7dac96e 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -135,12 +135,6 @@ out_unlock: out_free: kfree(file); -#ifdef CONFIG_LOCKD_V4 - if (nfserr == 1) - nfserr = nlm4_stale_fh; - else -#endif - nfserr = nlm_lck_denied; goto out_unlock; } diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c index 7b889ff..9b9e7e1 100644 --- a/fs/nfsd/lockd.c +++ b/fs/nfsd/lockd.c @@ -39,18 +39,20 @@ nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) fh_put(&fh); rqstp->rq_client = NULL; exp_readunlock(); - /* nlm and nfsd don't share error codes. - * we invent: 0 = no error - * 1 = stale file handle - * 2 = other error + /* We return nlm error codes as nlm doesn't know + * about nfsd, but nfsd does know about nlm.. */ switch (nfserr) { case nfs_ok: return 0; + case nfserr_dropit: + return nlm_drop_reply; +#ifdef CONFIG_LOCKD_V4 case nfserr_stale: - return 1; + return nlm4_stale_fh; +#endif default: - return 2; + return nlm_lck_denied; } } diff --git a/include/linux/lockd/bind.h b/include/linux/lockd/bind.h index 81e3a18..aa50d89 100644 --- a/include/linux/lockd/bind.h +++ b/include/linux/lockd/bind.h @@ -10,6 +10,11 @@ #define LINUX_LOCKD_BIND_H #include +/* need xdr-encoded error codes too, so... */ +#include +#ifdef CONFIG_LOCKD_V4 +#include +#endif /* Dummy declarations */ struct svc_rqst; diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h index bb0a0f1..66fdae3 100644 --- a/include/linux/lockd/xdr.h +++ b/include/linux/lockd/xdr.h @@ -13,6 +13,8 @@ #include #include +struct svc_rqst; + #define NLM_MAXCOOKIELEN 32 #define NLM_MAXSTRLEN 1024 @@ -22,6 +24,8 @@ #define nlm_lck_blocked __constant_htonl(NLM_LCK_BLOCKED) #define nlm_lck_denied_grace_period __constant_htonl(NLM_LCK_DENIED_GRACE_PERIOD) +#define nlm_drop_reply __constant_htonl(30000) + /* Lock info passed via NLM */ struct nlm_lock { char * caller; diff --git a/include/linux/sunrpc/msg_prot.h b/include/linux/sunrpc/msg_prot.h index 1e65f2d..606cb21 100644 --- a/include/linux/sunrpc/msg_prot.h +++ b/include/linux/sunrpc/msg_prot.h @@ -56,7 +56,9 @@ enum rpc_accept_stat { RPC_PROG_MISMATCH = 2, RPC_PROC_UNAVAIL = 3, RPC_GARBAGE_ARGS = 4, - RPC_SYSTEM_ERR = 5 + RPC_SYSTEM_ERR = 5, + /* internal use only */ + RPC_DROP_REPLY = 60000, }; enum rpc_reject_stat { diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 953723b..ac69e55 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -74,6 +74,7 @@ struct xdr_buf { #define rpc_proc_unavail __constant_htonl(RPC_PROC_UNAVAIL) #define rpc_garbage_args __constant_htonl(RPC_GARBAGE_ARGS) #define rpc_system_err __constant_htonl(RPC_SYSTEM_ERR) +#define rpc_drop_reply __constant_htonl(RPC_DROP_REPLY) #define rpc_auth_ok __constant_htonl(RPC_AUTH_OK) #define rpc_autherr_badcred __constant_htonl(RPC_AUTH_BADCRED) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 2807fa0..eb44ec9 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -828,6 +828,11 @@ svc_process(struct svc_rqst *rqstp) *statp = procp->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp); /* Encode reply */ + if (*statp == rpc_drop_reply) { + if (procp->pc_release) + procp->pc_release(rqstp, NULL, rqstp->rq_resp); + goto dropit; + } if (*statp == rpc_success && (xdr = procp->pc_encode) && !xdr(rqstp, resv->iov_base+resv->iov_len, rqstp->rq_resp)) { dprintk("svc: failed to encode reply\n"); -- cgit v0.10.2 From 58ff407bee5a55f9c1188a3f9d70ffc79485183c Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 17 Oct 2006 00:10:19 -0700 Subject: [PATCH] Fix IO error reporting on fsync() When IO error happens on metadata buffer, buffer is freed from memory and later fsync() is called, filesystems like ext2 fail to report EIO. We solve the problem by introducing a pointer to associated address space into the buffer_head. When a buffer is removed from a list of metadata buffers associated with an address space, IO error is transferred from the buffer to the address space, so that fsync can later report it. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/buffer.c b/fs/buffer.c index f65ef88..35527dc 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -452,6 +452,7 @@ static void end_buffer_async_write(struct buffer_head *bh, int uptodate) bdevname(bh->b_bdev, b)); } set_bit(AS_EIO, &page->mapping->flags); + set_buffer_write_io_error(bh); clear_buffer_uptodate(bh); SetPageError(page); } @@ -571,6 +572,10 @@ EXPORT_SYMBOL(mark_buffer_async_write); static inline void __remove_assoc_queue(struct buffer_head *bh) { list_del_init(&bh->b_assoc_buffers); + WARN_ON(!bh->b_assoc_map); + if (buffer_write_io_error(bh)) + set_bit(AS_EIO, &bh->b_assoc_map->flags); + bh->b_assoc_map = NULL; } int inode_has_buffers(struct inode *inode) @@ -669,6 +674,7 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) spin_lock(&buffer_mapping->private_lock); list_move_tail(&bh->b_assoc_buffers, &mapping->private_list); + bh->b_assoc_map = mapping; spin_unlock(&buffer_mapping->private_lock); } } @@ -765,7 +771,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) spin_lock(lock); while (!list_empty(list)) { bh = BH_ENTRY(list->next); - list_del_init(&bh->b_assoc_buffers); + __remove_assoc_queue(bh); if (buffer_dirty(bh) || buffer_locked(bh)) { list_add(&bh->b_assoc_buffers, &tmp); if (buffer_dirty(bh)) { @@ -786,7 +792,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) while (!list_empty(&tmp)) { bh = BH_ENTRY(tmp.prev); - __remove_assoc_queue(bh); + list_del_init(&bh->b_assoc_buffers); get_bh(bh); spin_unlock(lock); wait_on_buffer(bh); @@ -1167,6 +1173,7 @@ void __bforget(struct buffer_head *bh) spin_lock(&buffer_mapping->private_lock); list_del_init(&bh->b_assoc_buffers); + bh->b_assoc_map = NULL; spin_unlock(&buffer_mapping->private_lock); } __brelse(bh); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 131ffd3..5d9fb0e 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -69,6 +69,8 @@ struct buffer_head { bh_end_io_t *b_end_io; /* I/O completion */ void *b_private; /* reserved for b_end_io */ struct list_head b_assoc_buffers; /* associated with another mapping */ + struct address_space *b_assoc_map; /* mapping this buffer is + associated with */ atomic_t b_count; /* users using this buffer_head */ }; -- cgit v0.10.2 From 12fda16814bba05a84a49a1da25a069d6c249758 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 17 Oct 2006 00:10:20 -0700 Subject: [PATCH] drivers/led: handle sysfs errors Signed-off-by: Jeff Garzik Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c index aecbbe2..3c17112 100644 --- a/drivers/leds/led-class.c +++ b/drivers/leds/led-class.c @@ -91,6 +91,8 @@ EXPORT_SYMBOL_GPL(led_classdev_resume); */ int led_classdev_register(struct device *parent, struct led_classdev *led_cdev) { + int rc; + led_cdev->class_dev = class_device_create(leds_class, NULL, 0, parent, "%s", led_cdev->name); if (unlikely(IS_ERR(led_cdev->class_dev))) @@ -99,8 +101,10 @@ int led_classdev_register(struct device *parent, struct led_classdev *led_cdev) class_set_devdata(led_cdev->class_dev, led_cdev); /* register the attributes */ - class_device_create_file(led_cdev->class_dev, - &class_device_attr_brightness); + rc = class_device_create_file(led_cdev->class_dev, + &class_device_attr_brightness); + if (rc) + goto err_out; /* add to the list of leds */ write_lock(&leds_list_lock); @@ -110,16 +114,28 @@ int led_classdev_register(struct device *parent, struct led_classdev *led_cdev) #ifdef CONFIG_LEDS_TRIGGERS rwlock_init(&led_cdev->trigger_lock); - led_trigger_set_default(led_cdev); + rc = class_device_create_file(led_cdev->class_dev, + &class_device_attr_trigger); + if (rc) + goto err_out_led_list; - class_device_create_file(led_cdev->class_dev, - &class_device_attr_trigger); + led_trigger_set_default(led_cdev); #endif printk(KERN_INFO "Registered led device: %s\n", led_cdev->class_dev->class_id); return 0; + +#ifdef CONFIG_LEDS_TRIGGERS +err_out_led_list: + class_device_remove_file(led_cdev->class_dev, + &class_device_attr_brightness); + list_del(&led_cdev->node); +#endif +err_out: + class_device_unregister(led_cdev->class_dev); + return rc; } EXPORT_SYMBOL_GPL(led_classdev_register); diff --git a/drivers/leds/ledtrig-timer.c b/drivers/leds/ledtrig-timer.c index 179c287..29a8818a 100644 --- a/drivers/leds/ledtrig-timer.c +++ b/drivers/leds/ledtrig-timer.c @@ -123,6 +123,7 @@ static CLASS_DEVICE_ATTR(delay_off, 0644, led_delay_off_show, static void timer_trig_activate(struct led_classdev *led_cdev) { struct timer_trig_data *timer_data; + int rc; timer_data = kzalloc(sizeof(struct timer_trig_data), GFP_KERNEL); if (!timer_data) @@ -134,10 +135,21 @@ static void timer_trig_activate(struct led_classdev *led_cdev) timer_data->timer.function = led_timer_function; timer_data->timer.data = (unsigned long) led_cdev; - class_device_create_file(led_cdev->class_dev, + rc = class_device_create_file(led_cdev->class_dev, &class_device_attr_delay_on); - class_device_create_file(led_cdev->class_dev, + if (rc) goto err_out; + rc = class_device_create_file(led_cdev->class_dev, &class_device_attr_delay_off); + if (rc) goto err_out_delayon; + + return; + +err_out_delayon: + class_device_remove_file(led_cdev->class_dev, + &class_device_attr_delay_on); +err_out: + led_cdev->trigger_data = NULL; + kfree(timer_data); } static void timer_trig_deactivate(struct led_classdev *led_cdev) -- cgit v0.10.2 From 6b5f29675c6a1041aefc147271508bd56cf2b761 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 17 Oct 2006 00:10:22 -0700 Subject: [PATCH] I2O: handle a few sysfs errors Signed-off-by: Jeff Garzik Cc: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/message/i2o/bus-osm.c b/drivers/message/i2o/bus-osm.c index ac06f10..d96c687 100644 --- a/drivers/message/i2o/bus-osm.c +++ b/drivers/message/i2o/bus-osm.c @@ -80,18 +80,26 @@ static DEVICE_ATTR(scan, S_IWUSR, NULL, i2o_bus_store_scan); * @dev: device to verify if it is a I2O Bus Adapter device * * Because we want all Bus Adapters always return 0. + * Except when we fail. Then we are sad. * - * Returns 0. + * Returns 0, except when we fail to excel. */ static int i2o_bus_probe(struct device *dev) { struct i2o_device *i2o_dev = to_i2o_device(get_device(dev)); + int rc; - device_create_file(dev, &dev_attr_scan); + rc = device_create_file(dev, &dev_attr_scan); + if (rc) + goto err_out; osm_info("device added (TID: %03x)\n", i2o_dev->lct_data.tid); return 0; + +err_out: + put_device(dev); + return rc; }; /** diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 7bd4d85..91f95d1 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -325,13 +325,24 @@ static DEVICE_ATTR(product_id, S_IRUGO, i2o_exec_show_product_id, NULL); static int i2o_exec_probe(struct device *dev) { struct i2o_device *i2o_dev = to_i2o_device(dev); + int rc; - i2o_event_register(i2o_dev, &i2o_exec_driver, 0, 0xffffffff); + rc = i2o_event_register(i2o_dev, &i2o_exec_driver, 0, 0xffffffff); + if (rc) goto err_out; - device_create_file(dev, &dev_attr_vendor_id); - device_create_file(dev, &dev_attr_product_id); + rc = device_create_file(dev, &dev_attr_vendor_id); + if (rc) goto err_evtreg; + rc = device_create_file(dev, &dev_attr_product_id); + if (rc) goto err_vid; return 0; + +err_vid: + device_remove_file(dev, &dev_attr_vendor_id); +err_evtreg: + i2o_event_register(to_i2o_device(dev), &i2o_exec_driver, 0, 0); +err_out: + return rc; }; /** -- cgit v0.10.2 From eee44cca665aa1a5663e6a00c2bdfc275739dac5 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 17 Oct 2006 00:10:23 -0700 Subject: [PATCH] fs/partitions/check: add sysfs error handling Handle errors thrown in disk_sysfs_symlinks(), and propagate back to caller. The callers and associated functions don't do a real good job of handling kobject errors anyway (add_partition, register_disk, rescan_partitions), so this should do until something better comes along. Signed-off-by: Jeff Garzik Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 51c6a74..6fb4b61 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -376,18 +376,48 @@ static char *make_block_name(struct gendisk *disk) return name; } -static void disk_sysfs_symlinks(struct gendisk *disk) +static int disk_sysfs_symlinks(struct gendisk *disk) { struct device *target = get_device(disk->driverfs_dev); + int err; + char *disk_name = NULL; + if (target) { - char *disk_name = make_block_name(disk); - sysfs_create_link(&disk->kobj,&target->kobj,"device"); - if (disk_name) { - sysfs_create_link(&target->kobj,&disk->kobj,disk_name); - kfree(disk_name); + disk_name = make_block_name(disk); + if (!disk_name) { + err = -ENOMEM; + goto err_out; } + + err = sysfs_create_link(&disk->kobj, &target->kobj, "device"); + if (err) + goto err_out_disk_name; + + err = sysfs_create_link(&target->kobj, &disk->kobj, disk_name); + if (err) + goto err_out_dev_link; } - sysfs_create_link(&disk->kobj, &block_subsys.kset.kobj, "subsystem"); + + err = sysfs_create_link(&disk->kobj, &block_subsys.kset.kobj, + "subsystem"); + if (err) + goto err_out_disk_name_lnk; + + kfree(disk_name); + + return 0; + +err_out_disk_name_lnk: + if (target) { + sysfs_remove_link(&target->kobj, disk_name); +err_out_dev_link: + sysfs_remove_link(&disk->kobj, "device"); +err_out_disk_name: + kfree(disk_name); +err_out: + put_device(target); + } + return err; } /* Not exported, helper to add_disk(). */ @@ -406,7 +436,11 @@ void register_disk(struct gendisk *disk) *s = '!'; if ((err = kobject_add(&disk->kobj))) return; - disk_sysfs_symlinks(disk); + err = disk_sysfs_symlinks(disk); + if (err) { + kobject_del(&disk->kobj); + return; + } disk_sysfs_add_subdirs(disk); /* No minors to use for partitions */ -- cgit v0.10.2 From 6a15f46c1272afd3010259067451bf0df04f6511 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 17 Oct 2006 00:10:25 -0700 Subject: [PATCH] rtc: fix printk of 64-bit res on 32-bit platform With 64-bit resources on 32-bit platforms, the resource address might be larger than a void*. Fix printk to work regardless of resource size. Signed-off-by: Jeff Garzik Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-v3020.c b/drivers/rtc/rtc-v3020.c index 09b714f..3b58d3d 100644 --- a/drivers/rtc/rtc-v3020.c +++ b/drivers/rtc/rtc-v3020.c @@ -195,9 +195,9 @@ static int rtc_probe(struct platform_device *pdev) * are all disabled */ v3020_set_reg(chip, V3020_STATUS_0, 0x0); - dev_info(&pdev->dev, "Chip available at physical address 0x%p," + dev_info(&pdev->dev, "Chip available at physical address 0x%llx," "data connected to D%d\n", - (void*)pdev->resource[0].start, + (unsigned long long)pdev->resource[0].start, chip->leftshift); platform_set_drvdata(pdev, chip); -- cgit v0.10.2 From 3864c4894a7f4c03d69a90082a5bb0ab10e437ab Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 17 Oct 2006 00:10:26 -0700 Subject: [PATCH] lockdep: annotate i386 apm Lockdep doesn't like to enable interrupts when they are enabled already. BUG: warning at kernel/lockdep.c:1814/trace_hardirqs_on() (Not tainted) [] show_trace_log_lvl+0x58/0x16a [] show_trace+0xd/0x10 [] dump_stack+0x19/0x1b [] trace_hardirqs_on+0xa2/0x11e [] apm_bios_call_simple+0xcd/0xfd [] apm+0x92/0x5b1 [] kernel_thread_helper+0x5/0xb DWARF2 unwinder stuck at kernel_thread_helper+0x5/0xb Leftover inexact backtrace: [] show_trace+0xd/0x10 [] dump_stack+0x19/0x1b [] trace_hardirqs_on+0xa2/0x11e [] apm_bios_call_simple+0xcd/0xfd [] apm+0x92/0x5b1 [] kernel_thread_helper+0x5/0xb Signed-off-by: Peter Zijlstra Cc: Stephen Rothwell Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index b42f2d9..2af6585 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -540,11 +540,30 @@ static inline void apm_restore_cpus(cpumask_t mask) * Also, we KNOW that for the non error case of apm_bios_call, there * is no useful data returned in the low order 8 bits of eax. */ -#define APM_DO_CLI \ - if (apm_info.allow_ints) \ - local_irq_enable(); \ - else \ + +static inline unsigned long __apm_irq_save(void) +{ + unsigned long flags; + local_save_flags(flags); + if (apm_info.allow_ints) { + if (irqs_disabled_flags(flags)) + local_irq_enable(); + } else + local_irq_disable(); + + return flags; +} + +#define apm_irq_save(flags) \ + do { flags = __apm_irq_save(); } while (0) + +static inline void apm_irq_restore(unsigned long flags) +{ + if (irqs_disabled_flags(flags)) local_irq_disable(); + else if (irqs_disabled()) + local_irq_enable(); +} #ifdef APM_ZERO_SEGS # define APM_DECL_SEGS \ @@ -596,12 +615,11 @@ static u8 apm_bios_call(u32 func, u32 ebx_in, u32 ecx_in, save_desc_40 = gdt[0x40 / 8]; gdt[0x40 / 8] = bad_bios_desc; - local_save_flags(flags); - APM_DO_CLI; + apm_irq_save(flags); APM_DO_SAVE_SEGS; apm_bios_call_asm(func, ebx_in, ecx_in, eax, ebx, ecx, edx, esi); APM_DO_RESTORE_SEGS; - local_irq_restore(flags); + apm_irq_restore(flags); gdt[0x40 / 8] = save_desc_40; put_cpu(); apm_restore_cpus(cpus); @@ -640,12 +658,11 @@ static u8 apm_bios_call_simple(u32 func, u32 ebx_in, u32 ecx_in, u32 *eax) save_desc_40 = gdt[0x40 / 8]; gdt[0x40 / 8] = bad_bios_desc; - local_save_flags(flags); - APM_DO_CLI; + apm_irq_save(flags); APM_DO_SAVE_SEGS; error = apm_bios_call_simple_asm(func, ebx_in, ecx_in, eax); APM_DO_RESTORE_SEGS; - local_irq_restore(flags); + apm_irq_restore(flags); gdt[0x40 / 8] = save_desc_40; put_cpu(); apm_restore_cpus(cpus); -- cgit v0.10.2 From ea6f94dfe9db4d19a39e774cfafa5c9428a9fdbc Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Tue, 17 Oct 2006 00:10:27 -0700 Subject: [PATCH] rd: memory leak on rd_init() failure If RAM disk driver initialization fails due to blk_alloc_queue() faulure, the gendisk structs stored in rd_disks[] will not be freed completely. This patch resolves that memory leak case by doing alloc_disk() and blk_alloc_queue() at the same time. Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/rd.c b/drivers/block/rd.c index a3f64bf..485aa87 100644 --- a/drivers/block/rd.c +++ b/drivers/block/rd.c @@ -432,6 +432,12 @@ static int __init rd_init(void) rd_disks[i] = alloc_disk(1); if (!rd_disks[i]) goto out; + + rd_queue[i] = blk_alloc_queue(GFP_KERNEL); + if (!rd_queue[i]) { + put_disk(rd_disks[i]); + goto out; + } } if (register_blkdev(RAMDISK_MAJOR, "ramdisk")) { @@ -442,10 +448,6 @@ static int __init rd_init(void) for (i = 0; i < CONFIG_BLK_DEV_RAM_COUNT; i++) { struct gendisk *disk = rd_disks[i]; - rd_queue[i] = blk_alloc_queue(GFP_KERNEL); - if (!rd_queue[i]) - goto out_queue; - blk_queue_make_request(rd_queue[i], &rd_make_request); blk_queue_hardsect_size(rd_queue[i], rd_blocksize); @@ -466,8 +468,6 @@ static int __init rd_init(void) CONFIG_BLK_DEV_RAM_COUNT, rd_size, rd_blocksize); return 0; -out_queue: - unregister_blkdev(RAMDISK_MAJOR, "ramdisk"); out: while (i--) { put_disk(rd_disks[i]); -- cgit v0.10.2 From dabad0568a5935e9f4903f5fd1d8f22b1c7c88c7 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Tue, 17 Oct 2006 00:10:28 -0700 Subject: [PATCH] epca: prevent panic on tty_register_driver() failure Make epca fail on initialization failure instead of panic. Cc: "Digi International, Inc" Signed-off-by: Akinobu Mita Acked-by: Alan Cox Acked-by: Scott Kilau Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/epca.c b/drivers/char/epca.c index c3f9558..706733c 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -1157,6 +1157,7 @@ static int __init pc_init(void) int crd; struct board_info *bd; unsigned char board_id = 0; + int err = -ENOMEM; int pci_boards_found, pci_count; @@ -1164,13 +1165,11 @@ static int __init pc_init(void) pc_driver = alloc_tty_driver(MAX_ALLOC); if (!pc_driver) - return -ENOMEM; + goto out1; pc_info = alloc_tty_driver(MAX_ALLOC); - if (!pc_info) { - put_tty_driver(pc_driver); - return -ENOMEM; - } + if (!pc_info) + goto out2; /* ----------------------------------------------------------------------- If epca_setup has not been ran by LILO set num_cards to defaults; copy @@ -1370,11 +1369,17 @@ static int __init pc_init(void) } /* End for each card */ - if (tty_register_driver(pc_driver)) - panic("Couldn't register Digi PC/ driver"); + err = tty_register_driver(pc_driver); + if (err) { + printk(KERN_ERR "Couldn't register Digi PC/ driver"); + goto out3; + } - if (tty_register_driver(pc_info)) - panic("Couldn't register Digi PC/ info "); + err = tty_register_driver(pc_info); + if (err) { + printk(KERN_ERR "Couldn't register Digi PC/ info "); + goto out4; + } /* ------------------------------------------------------------------- Start up the poller to check for events on all enabled boards @@ -1385,6 +1390,15 @@ static int __init pc_init(void) mod_timer(&epca_timer, jiffies + HZ/25); return 0; +out4: + tty_unregister_driver(pc_driver); +out3: + put_tty_driver(pc_info); +out2: + put_tty_driver(pc_driver); +out1: + return err; + } /* End pc_init */ /* ------------------ Begin post_fep_init ---------------------- */ -- cgit v0.10.2 From 3f3fd3c055853d11295a1ec6cdc81e37e2182d16 Mon Sep 17 00:00:00 2001 From: Greg Banks Date: Tue, 17 Oct 2006 00:10:30 -0700 Subject: [PATCH] kbuild: allow multi-word $M in Makefile.modpost Some people want to do crazy things like pass multiple directories as the value of $(SUBDIRS) or $M. Mostly this kinda works, except that Makefile.modpost constructs a modpost commandline which fails modpost's argument parsing. This patch fixes that little wrinkle. Signed-off-by: Greg Banks Cc: Sam Ravnborg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index 6c5469b..65e0a79 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -44,7 +44,7 @@ include scripts/Kbuild.include include scripts/Makefile.lib kernelsymfile := $(objtree)/Module.symvers -modulesymfile := $(KBUILD_EXTMOD)/Module.symvers +modulesymfile := $(firstword $(KBUILD_EXTMOD))/Module.symvers # Step 1), find all modules listed in $(MODVERDIR)/ __modules := $(sort $(shell grep -h '\.ko' /dev/null $(wildcard $(MODVERDIR)/*.mod))) -- cgit v0.10.2 From 91b943ee4afa2037678dc1db30b89baef0e17090 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 17 Oct 2006 00:10:32 -0700 Subject: [PATCH] Add entry.S labels to tag file Add functions defined using ENTRY macro to the tags file. Signed-off-by: Aneesh Kumar K.V Cc: Sam Ravnborg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Makefile b/Makefile index 703d40a..62a1343 100644 --- a/Makefile +++ b/Makefile @@ -1319,7 +1319,8 @@ define xtags $(all-sources) | xargs $1 -a \ -I __initdata,__exitdata,__acquires,__releases \ -I EXPORT_SYMBOL,EXPORT_SYMBOL_GPL \ - --extra=+f --c-kinds=+px; \ + --extra=+f --c-kinds=+px \ + --regex-asm='/ENTRY\(([^)]*)\).*/\1/'; \ $(all-kconfigs) | xargs $1 -a \ --langdef=kconfig \ --language-force=kconfig \ -- cgit v0.10.2 From bea493a031fe3337f4fe5479e8e865513828ea76 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 17 Oct 2006 00:10:33 -0700 Subject: [PATCH] rt-mutex: fixup rt-mutex debug code BUG: warning at kernel/rtmutex-debug.c:125/rt_mutex_debug_task_free() (Not tainted) [] show_trace_log_lvl+0x58/0x16a [] show_trace+0xd/0x10 [] dump_stack+0x19/0x1b [] rt_mutex_debug_task_free+0x35/0x6a [] free_task+0x15/0x24 [] copy_process+0x12bd/0x1324 [] do_fork+0x42/0x113 [] sys_fork+0x19/0x1b [] syscall_call+0x7/0xb In copy_process(), dup_task_struct() also duplicates the ->pi_lock, ->pi_waiters and ->pi_blocked_on members. rt_mutex_debug_task_free() called from free_task() validates these members. However free_task() can be invoked before these members are reset for the new task. Move the initialization code before the first bail that can hit free_task(). Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/fork.c b/kernel/fork.c index 7dc6140..29ebb30 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -984,6 +984,8 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (!p) goto fork_out; + rt_mutex_init_task(p); + #ifdef CONFIG_TRACE_IRQFLAGS DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled); DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled); @@ -1088,8 +1090,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->lockdep_recursion = 0; #endif - rt_mutex_init_task(p); - #ifdef CONFIG_DEBUG_MUTEXES p->blocked_on = NULL; /* not blocked yet */ #endif -- cgit v0.10.2 From bd5349cfd2b9bbb10a3dbcd3fe5cbaabe0b2ab9e Mon Sep 17 00:00:00 2001 From: Neil Brown Date: Tue, 17 Oct 2006 00:10:35 -0700 Subject: [PATCH] Convert cpu hotplug notifiers to use raw_notifier instead of blocking_notifier The use of blocking notifier by _cpu_up and _cpu_down in cpu.c has two problem. 1/ An interaction with the workqueue notifier causes lockdep to spit a warning. 2/ A notifier could conceivable be added or removed while _cpu_up or _cpu_down are in process. As each notifier is called twice (prepare then commit/abort) this could be unhealthy. To fix to we simply take cpu_add_remove_lock while adding or removing notifiers to/from the list. This makes the 'blocking' usage unnecessary as all accesses to cpu_chain are now protected by cpu_add_remove_lock. So change "blocking" to "raw" in all relevant places. This fixes 1. Credit: Andrew Morton Cc: Rusty Russell Cc: Michal Piotrowski (reporter) Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/cpu.c b/kernel/cpu.c index 32c9662..27dd3ee 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -19,7 +19,7 @@ static DEFINE_MUTEX(cpu_add_remove_lock); static DEFINE_MUTEX(cpu_bitmask_lock); -static __cpuinitdata BLOCKING_NOTIFIER_HEAD(cpu_chain); +static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain); /* If set, cpu_up and cpu_down will return -EBUSY and do nothing. * Should always be manipulated under cpu_add_remove_lock @@ -68,7 +68,11 @@ EXPORT_SYMBOL_GPL(unlock_cpu_hotplug); /* Need to know about CPUs going up/down? */ int __cpuinit register_cpu_notifier(struct notifier_block *nb) { - return blocking_notifier_chain_register(&cpu_chain, nb); + int ret; + mutex_lock(&cpu_add_remove_lock); + ret = raw_notifier_chain_register(&cpu_chain, nb); + mutex_unlock(&cpu_add_remove_lock); + return ret; } #ifdef CONFIG_HOTPLUG_CPU @@ -77,7 +81,9 @@ EXPORT_SYMBOL(register_cpu_notifier); void unregister_cpu_notifier(struct notifier_block *nb) { - blocking_notifier_chain_unregister(&cpu_chain, nb); + mutex_lock(&cpu_add_remove_lock); + raw_notifier_chain_unregister(&cpu_chain, nb); + mutex_unlock(&cpu_add_remove_lock); } EXPORT_SYMBOL(unregister_cpu_notifier); @@ -126,7 +132,7 @@ static int _cpu_down(unsigned int cpu) if (!cpu_online(cpu)) return -EINVAL; - err = blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, + err = raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, (void *)(long)cpu); if (err == NOTIFY_BAD) { printk("%s: attempt to take down CPU %u failed\n", @@ -146,7 +152,7 @@ static int _cpu_down(unsigned int cpu) if (IS_ERR(p)) { /* CPU didn't die: tell everyone. Can't complain. */ - if (blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, + if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, (void *)(long)cpu) == NOTIFY_BAD) BUG(); @@ -169,7 +175,7 @@ static int _cpu_down(unsigned int cpu) put_cpu(); /* CPU is completely dead: tell everyone. Too late to complain. */ - if (blocking_notifier_call_chain(&cpu_chain, CPU_DEAD, + if (raw_notifier_call_chain(&cpu_chain, CPU_DEAD, (void *)(long)cpu) == NOTIFY_BAD) BUG(); @@ -206,7 +212,7 @@ static int __devinit _cpu_up(unsigned int cpu) if (cpu_online(cpu) || !cpu_present(cpu)) return -EINVAL; - ret = blocking_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); + ret = raw_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); if (ret == NOTIFY_BAD) { printk("%s: attempt to bring up CPU %u failed\n", __FUNCTION__, cpu); @@ -223,11 +229,11 @@ static int __devinit _cpu_up(unsigned int cpu) BUG_ON(!cpu_online(cpu)); /* Now call notifier in preparation. */ - blocking_notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); + raw_notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); out_notify: if (ret != 0) - blocking_notifier_call_chain(&cpu_chain, + raw_notifier_call_chain(&cpu_chain, CPU_UP_CANCELED, hcpu); return ret; -- cgit v0.10.2 From 0d9ba869e103d91d471146378ad85bf1fb8e74b4 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Tue, 17 Oct 2006 00:10:36 -0700 Subject: [PATCH] drivers/isdn/hysdn: save_flags()/cli(), restore_flags() replaced appropriately With Karsten Keil save_flags()/cli() pair is replaced with spin_lock_irqsave() and restore_flags() replaced with spin_unlock_irqrestore() Tested compile only using allmodconfig Signed-off-by: Amol Lad Acked-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/hysdn/boardergo.c b/drivers/isdn/hysdn/boardergo.c index 160f22f..8bbe33a 100644 --- a/drivers/isdn/hysdn/boardergo.c +++ b/drivers/isdn/hysdn/boardergo.c @@ -45,11 +45,10 @@ ergo_interrupt(int intno, void *dev_id) if (!card->irq_enabled) return IRQ_NONE; /* other device interrupting or irq switched off */ - save_flags(flags); - cli(); /* no further irqs allowed */ + spin_lock_irqsave(&card->hysdn_lock, flags); /* no further irqs allowed */ if (!(bytein(card->iobase + PCI9050_INTR_REG) & PCI9050_INTR_REG_STAT1)) { - restore_flags(flags); /* restore old state */ + spin_unlock_irqrestore(&card->hysdn_lock, flags); /* restore old state */ return IRQ_NONE; /* no interrupt requested by E1 */ } /* clear any pending ints on the board */ @@ -61,7 +60,7 @@ ergo_interrupt(int intno, void *dev_id) /* start kernel task immediately after leaving all interrupts */ if (!card->hw_lock) schedule_work(&card->irq_queue); - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); return IRQ_HANDLED; } /* ergo_interrupt */ @@ -83,10 +82,9 @@ ergo_irq_bh(hysdn_card * card) dpr = card->dpram; /* point to DPRAM */ - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); if (card->hw_lock) { - restore_flags(flags); /* hardware currently unavailable */ + spin_unlock_irqrestore(&card->hysdn_lock, flags); /* hardware currently unavailable */ return; } card->hw_lock = 1; /* we now lock the hardware */ @@ -120,7 +118,7 @@ ergo_irq_bh(hysdn_card * card) card->hw_lock = 0; /* free hardware again */ } while (again); /* until nothing more to do */ - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); } /* ergo_irq_bh */ @@ -137,8 +135,7 @@ ergo_stopcard(hysdn_card * card) #ifdef CONFIG_HYSDN_CAPI hycapi_capi_stop(card); #endif /* CONFIG_HYSDN_CAPI */ - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); val = bytein(card->iobase + PCI9050_INTR_REG); /* get actual value */ val &= ~(PCI9050_INTR_REG_ENPCI | PCI9050_INTR_REG_EN1); /* mask irq */ byteout(card->iobase + PCI9050_INTR_REG, val); @@ -147,7 +144,7 @@ ergo_stopcard(hysdn_card * card) card->state = CARD_STATE_UNUSED; card->err_log_state = ERRLOG_STATE_OFF; /* currently no log active */ - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); } /* ergo_stopcard */ /**************************************************************************/ @@ -162,12 +159,11 @@ ergo_set_errlog_state(hysdn_card * card, int on) card->err_log_state = ERRLOG_STATE_OFF; /* must be off */ return; } - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); if (((card->err_log_state == ERRLOG_STATE_OFF) && !on) || ((card->err_log_state == ERRLOG_STATE_ON) && on)) { - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); return; /* nothing to do */ } if (on) @@ -175,7 +171,7 @@ ergo_set_errlog_state(hysdn_card * card, int on) else card->err_log_state = ERRLOG_STATE_STOP; /* request stop */ - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); schedule_work(&card->irq_queue); } /* ergo_set_errlog_state */ @@ -356,8 +352,7 @@ ergo_waitpofready(struct HYSDN_CARD *card) if (card->debug_flags & LOG_POF_RECORD) hysdn_addlog(card, "ERGO: pof boot success"); - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); card->state = CARD_STATE_RUN; /* now card is running */ /* enable the cards interrupt */ @@ -370,7 +365,7 @@ ergo_waitpofready(struct HYSDN_CARD *card) dpr->ToHyInt = 1; dpr->ToPcInt = 1; /* interrupt to E1 for all cards */ - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); if ((hynet_enable & (1 << card->myid)) && (i = hysdn_net_create(card))) { @@ -448,6 +443,7 @@ ergo_inithardware(hysdn_card * card) card->waitpofready = ergo_waitpofready; card->set_errlog_state = ergo_set_errlog_state; INIT_WORK(&card->irq_queue, (void *) (void *) ergo_irq_bh, card); + card->hysdn_lock = SPIN_LOCK_UNLOCKED; return (0); } /* ergo_inithardware */ diff --git a/drivers/isdn/hysdn/hysdn_defs.h b/drivers/isdn/hysdn/hysdn_defs.h index 461e831..729df40 100644 --- a/drivers/isdn/hysdn/hysdn_defs.h +++ b/drivers/isdn/hysdn/hysdn_defs.h @@ -188,6 +188,8 @@ typedef struct HYSDN_CARD { /* init and deinit stopcard for booting, too */ void (*stopcard) (struct HYSDN_CARD *); void (*releasehardware) (struct HYSDN_CARD *); + + spinlock_t hysdn_lock; #ifdef CONFIG_HYSDN_CAPI struct hycapictrl_info { char cardname[32]; diff --git a/drivers/isdn/hysdn/hysdn_proclog.c b/drivers/isdn/hysdn/hysdn_proclog.c index c4301e8..fcd4992 100644 --- a/drivers/isdn/hysdn/hysdn_proclog.c +++ b/drivers/isdn/hysdn/hysdn_proclog.c @@ -116,8 +116,7 @@ put_log_buffer(hysdn_card * card, char *cp) strcpy(ib->log_start, cp); /* set output string */ ib->next = NULL; ib->proc_ctrl = pd; /* point to own control structure */ - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); ib->usage_cnt = pd->if_used; if (!pd->log_head) pd->log_head = ib; /* new head */ @@ -125,7 +124,7 @@ put_log_buffer(hysdn_card * card, char *cp) pd->log_tail->next = ib; /* follows existing messages */ pd->log_tail = ib; /* new tail */ i = pd->del_lock++; /* get lock state */ - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); /* delete old entrys */ if (!i) @@ -270,14 +269,13 @@ hysdn_log_open(struct inode *ino, struct file *filep) } else if ((filep->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_READ) { /* read access -> log/debug read */ - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); pd->if_used++; if (pd->log_head) filep->private_data = &pd->log_tail->next; else filep->private_data = &pd->log_head; - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); } else { /* simultaneous read/write access forbidden ! */ unlock_kernel(); return (-EPERM); /* no permission this time */ @@ -301,7 +299,7 @@ hysdn_log_close(struct inode *ino, struct file *filep) hysdn_card *card; int retval = 0; unsigned long flags; - + spinlock_t hysdn_lock = SPIN_LOCK_UNLOCKED; lock_kernel(); if ((filep->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_WRITE) { @@ -311,8 +309,7 @@ hysdn_log_close(struct inode *ino, struct file *filep) /* read access -> log/debug read, mark one further file as closed */ pd = NULL; - save_flags(flags); - cli(); + spin_lock_irqsave(&hysdn_lock, flags); inf = *((struct log_data **) filep->private_data); /* get first log entry */ if (inf) pd = (struct procdata *) inf->proc_ctrl; /* still entries there */ @@ -335,7 +332,7 @@ hysdn_log_close(struct inode *ino, struct file *filep) inf->usage_cnt--; /* decrement usage count for buffers */ inf = inf->next; } - restore_flags(flags); + spin_unlock_irqrestore(&hysdn_lock, flags); if (pd) if (pd->if_used <= 0) /* delete buffers if last file closed */ diff --git a/drivers/isdn/hysdn/hysdn_sched.c b/drivers/isdn/hysdn/hysdn_sched.c index 1c0d54a..1fadf01 100644 --- a/drivers/isdn/hysdn/hysdn_sched.c +++ b/drivers/isdn/hysdn/hysdn_sched.c @@ -155,8 +155,7 @@ hysdn_tx_cfgline(hysdn_card *card, unsigned char *line, unsigned short chan) if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg chan=%d len=%d", chan, strlen(line) + 1); - save_flags(flags); - cli(); + spin_lock_irqsave(&card->hysdn_lock, flags); while (card->async_busy) { sti(); @@ -165,7 +164,7 @@ hysdn_tx_cfgline(hysdn_card *card, unsigned char *line, unsigned short chan) msleep_interruptible(20); /* Timeout 20ms */ if (!--cnt) { - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); return (-ERR_ASYNC_TIME); /* timed out */ } cli(); @@ -194,13 +193,13 @@ hysdn_tx_cfgline(hysdn_card *card, unsigned char *line, unsigned short chan) msleep_interruptible(20); /* Timeout 20ms */ if (!--cnt) { - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); return (-ERR_ASYNC_TIME); /* timed out */ } cli(); } /* wait for buffer to become free again */ - restore_flags(flags); + spin_unlock_irqrestore(&card->hysdn_lock, flags); if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg data send"); -- cgit v0.10.2 From 078d396598401dbaa88d5f95ec45579f9d3dce0e Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Tue, 17 Oct 2006 00:10:37 -0700 Subject: [PATCH] drivers/isdn/isdnloop: save_flags()/cli(), restore_flags() replaced appropriately Signed-off-by: Amol Lad Acked-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/isdnloop/isdnloop.c b/drivers/isdn/isdnloop/isdnloop.c index fabbd46..23afba4 100644 --- a/drivers/isdn/isdnloop/isdnloop.c +++ b/drivers/isdn/isdnloop/isdnloop.c @@ -100,12 +100,11 @@ isdnloop_pollbchan(unsigned long data) isdnloop_bchan_send(card, 1); if (card->flags & (ISDNLOOP_FLAGS_B1ACTIVE | ISDNLOOP_FLAGS_B2ACTIVE)) { /* schedule b-channel polling again */ - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); card->rb_timer.expires = jiffies + ISDNLOOP_TIMER_BCREAD; add_timer(&card->rb_timer); card->flags |= ISDNLOOP_FLAGS_RBTIMER; - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } else card->flags &= ~ISDNLOOP_FLAGS_RBTIMER; } @@ -281,8 +280,7 @@ isdnloop_putmsg(isdnloop_card * card, unsigned char c) { ulong flags; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); *card->msg_buf_write++ = (c == 0xff) ? '\n' : c; if (card->msg_buf_write == card->msg_buf_read) { if (++card->msg_buf_read > card->msg_buf_end) @@ -290,7 +288,7 @@ isdnloop_putmsg(isdnloop_card * card, unsigned char c) } if (card->msg_buf_write > card->msg_buf_end) card->msg_buf_write = card->msg_buf; - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } /* @@ -372,21 +370,19 @@ isdnloop_polldchan(unsigned long data) if (!(card->flags & ISDNLOOP_FLAGS_RBTIMER)) { /* schedule b-channel polling */ card->flags |= ISDNLOOP_FLAGS_RBTIMER; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); del_timer(&card->rb_timer); card->rb_timer.function = isdnloop_pollbchan; card->rb_timer.data = (unsigned long) card; card->rb_timer.expires = jiffies + ISDNLOOP_TIMER_BCREAD; add_timer(&card->rb_timer); - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } /* schedule again */ - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); card->st_timer.expires = jiffies + ISDNLOOP_TIMER_DCREAD; add_timer(&card->st_timer); - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } /* @@ -416,8 +412,7 @@ isdnloop_sendbuf(int channel, struct sk_buff *skb, isdnloop_card * card) return 0; if (card->sndcount[channel] > ISDNLOOP_MAX_SQUEUE) return 0; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); nskb = dev_alloc_skb(skb->len); if (nskb) { memcpy(skb_put(nskb, len), skb->data, len); @@ -426,7 +421,7 @@ isdnloop_sendbuf(int channel, struct sk_buff *skb, isdnloop_card * card) } else len = 0; card->sndcount[channel] += len; - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } return len; } @@ -576,8 +571,7 @@ isdnloop_atimeout(isdnloop_card * card, int ch) unsigned long flags; char buf[60]; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); if (card->rcard) { isdnloop_fake(card->rcard[ch], "DDIS_I", card->rch[ch] + 1); card->rcard[ch]->rcard[card->rch[ch]] = NULL; @@ -587,7 +581,7 @@ isdnloop_atimeout(isdnloop_card * card, int ch) /* No user responding */ sprintf(buf, "CAU%s", isdnloop_unicause(card, 1, 3)); isdnloop_fake(card, buf, ch + 1); - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } /* @@ -622,8 +616,7 @@ isdnloop_start_ctimer(isdnloop_card * card, int ch) { unsigned long flags; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); init_timer(&card->c_timer[ch]); card->c_timer[ch].expires = jiffies + ISDNLOOP_TIMER_ALERTWAIT; if (ch) @@ -632,7 +625,7 @@ isdnloop_start_ctimer(isdnloop_card * card, int ch) card->c_timer[ch].function = isdnloop_atimeout0; card->c_timer[ch].data = (unsigned long) card; add_timer(&card->c_timer[ch]); - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } /* @@ -647,10 +640,9 @@ isdnloop_kill_ctimer(isdnloop_card * card, int ch) { unsigned long flags; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); del_timer(&card->c_timer[ch]); - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } static u_char si2bit[] = @@ -706,13 +698,12 @@ isdnloop_try_call(isdnloop_card * card, char *p, int lch, isdn_ctrl * cmd) } } if (num_match) { - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); /* channel idle? */ if (!(cc->rcard[ch])) { /* Check SI */ if (!(si2bit[cmd->parm.setup.si1] & cc->sil[ch])) { - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return 3; } /* ch is idle, si and number matches */ @@ -720,10 +711,10 @@ isdnloop_try_call(isdnloop_card * card, char *p, int lch, isdn_ctrl * cmd) cc->rch[ch] = lch; card->rcard[lch] = cc; card->rch[lch] = ch; - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return 0; } else { - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); /* num matches, but busy */ if (ch == 1) return 1; @@ -1027,8 +1018,7 @@ isdnloop_stopcard(isdnloop_card * card) unsigned long flags; isdn_ctrl cmd; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); if (card->flags & ISDNLOOP_FLAGS_RUNNING) { card->flags &= ~ISDNLOOP_FLAGS_RUNNING; del_timer(&card->st_timer); @@ -1039,7 +1029,7 @@ isdnloop_stopcard(isdnloop_card * card) cmd.driver = card->myid; card->interface.statcallb(&cmd); } - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); } /* @@ -1078,18 +1068,17 @@ isdnloop_start(isdnloop_card * card, isdnloop_sdef * sdefp) return -EBUSY; if (copy_from_user((char *) &sdef, (char *) sdefp, sizeof(sdef))) return -EFAULT; - save_flags(flags); - cli(); + spin_lock_irqsave(&card->isdnloop_lock, flags); switch (sdef.ptype) { case ISDN_PTYPE_EURO: if (isdnloop_fake(card, "DRV1.23EC-Q.931-CAPI-CNS-BASIS-20.02.96", -1)) { - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return -ENOMEM; } card->sil[0] = card->sil[1] = 4; if (isdnloop_fake(card, "TEI OK", 0)) { - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return -ENOMEM; } for (i = 0; i < 3; i++) @@ -1098,12 +1087,12 @@ isdnloop_start(isdnloop_card * card, isdnloop_sdef * sdefp) case ISDN_PTYPE_1TR6: if (isdnloop_fake(card, "DRV1.04TC-1TR6-CAPI-CNS-BASIS-29.11.95", -1)) { - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return -ENOMEM; } card->sil[0] = card->sil[1] = 4; if (isdnloop_fake(card, "TEI OK", 0)) { - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return -ENOMEM; } strcpy(card->s0num[0], sdef.num[0]); @@ -1111,7 +1100,7 @@ isdnloop_start(isdnloop_card * card, isdnloop_sdef * sdefp) card->s0num[2][0] = '\0'; break; default: - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); printk(KERN_WARNING "isdnloop: Illegal D-channel protocol %d\n", sdef.ptype); return -EINVAL; @@ -1122,7 +1111,7 @@ isdnloop_start(isdnloop_card * card, isdnloop_sdef * sdefp) card->st_timer.data = (unsigned long) card; add_timer(&card->st_timer); card->flags |= ISDNLOOP_FLAGS_RUNNING; - restore_flags(flags); + spin_unlock_irqrestore(&card->isdnloop_lock, flags); return 0; } @@ -1472,6 +1461,7 @@ isdnloop_initcard(char *id) skb_queue_head_init(&card->bqueue[i]); } skb_queue_head_init(&card->dqueue); + card->isdnloop_lock = SPIN_LOCK_UNLOCKED; card->next = cards; cards = card; if (!register_isdn(&card->interface)) { diff --git a/drivers/isdn/isdnloop/isdnloop.h b/drivers/isdn/isdnloop/isdnloop.h index d699fe5..0d458a8 100644 --- a/drivers/isdn/isdnloop/isdnloop.h +++ b/drivers/isdn/isdnloop/isdnloop.h @@ -94,6 +94,7 @@ typedef struct isdnloop_card { struct sk_buff_head bqueue[ISDNLOOP_BCH]; /* B-Channel queues */ struct sk_buff_head dqueue; /* D-Channel queue */ + spinlock_t isdnloop_lock; } isdnloop_card; /* -- cgit v0.10.2 From 04518bfe8eac2e82b476fb2b0093527adc2bc791 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 17 Oct 2006 00:10:39 -0700 Subject: [PATCH] ISDN: fix drivers, by handling errors thrown by ->readstat() This is a particularly ugly on-failure bug, possibly security, since the lack of error handling here is covering up another class of bug: failure to handle copy_to_user() return values. The I4L API function ->readstat() returns an integer, and by looking at several existing driver implementations, it is clear that a negative return value was meant to indicate an error. Given that several drivers already return a negative value indicating an errno-style error, the current code would blindly accept that [negative] value as a valid amount of bytes read. Obvious damage ensues. Correcting ->readstat() handling to properly notice errors fixes the existing code to work correctly on error, and enables future patches to more easily indicate errors during operation. Signed-off-by: Jeff Garzik Cc: Karsten Keil Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/i4l/isdn_common.c b/drivers/isdn/i4l/isdn_common.c index c3d79ee..69aee26 100644 --- a/drivers/isdn/i4l/isdn_common.c +++ b/drivers/isdn/i4l/isdn_common.c @@ -1134,9 +1134,12 @@ isdn_read(struct file *file, char __user *buf, size_t count, loff_t * off) if (dev->drv[drvidx]->interface->readstat) { if (count > dev->drv[drvidx]->stavail) count = dev->drv[drvidx]->stavail; - len = dev->drv[drvidx]->interface-> - readstat(buf, count, drvidx, - isdn_minor2chan(minor)); + len = dev->drv[drvidx]->interface->readstat(buf, count, + drvidx, isdn_minor2chan(minor)); + if (len < 0) { + retval = len; + goto out; + } } else { len = 0; } -- cgit v0.10.2 From 7786ce192fc4917fb9b789dd823476ff8fd6cf66 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 17 Oct 2006 00:10:40 -0700 Subject: [PATCH] ISDN: check for userspace copy faults Most of the ISDN ->readstat() implementations needed to check copy_to_user() and put_user() return values. Signed-off-by: Jeff Garzik Cc: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/capi/capidrv.c b/drivers/isdn/capi/capidrv.c index d10c8b8..b6f9476 100644 --- a/drivers/isdn/capi/capidrv.c +++ b/drivers/isdn/capi/capidrv.c @@ -1907,7 +1907,8 @@ static int if_readstat(u8 __user *buf, int len, int id, int channel) } for (p=buf, count=0; count < len; p++, count++) { - put_user(*card->q931_read++, p); + if (put_user(*card->q931_read++, p)) + return -EFAULT; if (card->q931_read > card->q931_end) card->q931_read = card->q931_buf; } diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c index e4823ab2..785b085 100644 --- a/drivers/isdn/hisax/config.c +++ b/drivers/isdn/hisax/config.c @@ -631,7 +631,8 @@ static int HiSax_readstatus(u_char __user *buf, int len, int id, int channel) count = cs->status_end - cs->status_read + 1; if (count >= len) count = len; - copy_to_user(p, cs->status_read, count); + if (copy_to_user(p, cs->status_read, count)) + return -EFAULT; cs->status_read += count; if (cs->status_read > cs->status_end) cs->status_read = cs->status_buf; @@ -642,7 +643,8 @@ static int HiSax_readstatus(u_char __user *buf, int len, int id, int channel) cnt = HISAX_STATUS_BUFSIZE; else cnt = count; - copy_to_user(p, cs->status_read, cnt); + if (copy_to_user(p, cs->status_read, cnt)) + return -EFAULT; p += cnt; cs->status_read += cnt % HISAX_STATUS_BUFSIZE; count -= cnt; diff --git a/drivers/isdn/icn/icn.c b/drivers/isdn/icn/icn.c index 6649f8b..730bbd0 100644 --- a/drivers/isdn/icn/icn.c +++ b/drivers/isdn/icn/icn.c @@ -1010,7 +1010,8 @@ icn_readstatus(u_char __user *buf, int len, icn_card * card) for (p = buf, count = 0; count < len; p++, count++) { if (card->msg_buf_read == card->msg_buf_write) return count; - put_user(*card->msg_buf_read++, p); + if (put_user(*card->msg_buf_read++, p)) + return -EFAULT; if (card->msg_buf_read > card->msg_buf_end) card->msg_buf_read = card->msg_buf; } diff --git a/drivers/isdn/isdnloop/isdnloop.c b/drivers/isdn/isdnloop/isdnloop.c index 23afba4..c3ae2ed 100644 --- a/drivers/isdn/isdnloop/isdnloop.c +++ b/drivers/isdn/isdnloop/isdnloop.c @@ -446,7 +446,8 @@ isdnloop_readstatus(u_char __user *buf, int len, isdnloop_card * card) for (p = buf, count = 0; count < len; p++, count++) { if (card->msg_buf_read == card->msg_buf_write) return count; - put_user(*card->msg_buf_read++, p); + if (put_user(*card->msg_buf_read++, p)) + return -EFAULT; if (card->msg_buf_read > card->msg_buf_end) card->msg_buf_read = card->msg_buf; } diff --git a/drivers/isdn/pcbit/drv.c b/drivers/isdn/pcbit/drv.c index 94f2148..6ead5e1 100644 --- a/drivers/isdn/pcbit/drv.c +++ b/drivers/isdn/pcbit/drv.c @@ -725,23 +725,27 @@ static int pcbit_stat(u_char __user *buf, int len, int driver, int channel) if (stat_st < stat_end) { - copy_to_user(buf, statbuf + stat_st, len); + if (copy_to_user(buf, statbuf + stat_st, len)) + return -EFAULT; stat_st += len; } else { if (len > STATBUF_LEN - stat_st) { - copy_to_user(buf, statbuf + stat_st, - STATBUF_LEN - stat_st); - copy_to_user(buf, statbuf, - len - (STATBUF_LEN - stat_st)); + if (copy_to_user(buf, statbuf + stat_st, + STATBUF_LEN - stat_st)) + return -EFAULT; + if (copy_to_user(buf, statbuf, + len - (STATBUF_LEN - stat_st))) + return -EFAULT; stat_st = len - (STATBUF_LEN - stat_st); } else { - copy_to_user(buf, statbuf + stat_st, len); + if (copy_to_user(buf, statbuf + stat_st, len)) + return -EFAULT; stat_st += len; -- cgit v0.10.2 From 7281c248f797723f66244b7ecef204620f664648 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:49:50 +0100 Subject: [ARM] switch to new pci_get_bus_and_slot API Signed-off-by: Alan Cox Signed-off-by: Russell King diff --git a/arch/arm/mach-ixp2000/ixdp2400.c b/arch/arm/mach-ixp2000/ixdp2400.c index a6f1480..9ee6383 100644 --- a/arch/arm/mach-ixp2000/ixdp2400.c +++ b/arch/arm/mach-ixp2000/ixdp2400.c @@ -133,11 +133,13 @@ static void ixdp2400_pci_postinit(void) struct pci_dev *dev; if (ixdp2x00_master_npu()) { - dev = pci_find_slot(1, IXDP2400_SLAVE_ENET_DEVFN); + dev = pci_get_bus_and_slot(1, IXDP2400_SLAVE_ENET_DEVFN); pci_remove_bus_device(dev); + pci_dev_put(dev) } else { - dev = pci_find_slot(1, IXDP2400_MASTER_ENET_DEVFN); + dev = pci_get_bus_and_slot(1, IXDP2400_MASTER_ENET_DEVFN); pci_remove_bus_device(dev); + pci_dev_put(dev) ixdp2x00_slave_pci_postinit(); } diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c index 91d36d9..70d247f 100644 --- a/arch/arm/mach-ixp2000/ixdp2800.c +++ b/arch/arm/mach-ixp2000/ixdp2800.c @@ -261,14 +261,16 @@ int __init ixdp2800_pci_init(void) pci_common_init(&ixdp2800_pci); if (ixdp2x00_master_npu()) { - dev = pci_find_slot(1, IXDP2800_SLAVE_ENET_DEVFN); + dev = pci_get_bus_and_slot(1, IXDP2800_SLAVE_ENET_DEVFN); pci_remove_bus_device(dev); + pci_dev_put(dev); ixdp2800_master_enable_slave(); ixdp2800_master_wait_for_slave_bus_scan(); } else { - dev = pci_find_slot(1, IXDP2800_MASTER_ENET_DEVFN); + dev = pci_get_bus_and_slot(1, IXDP2800_MASTER_ENET_DEVFN); pci_remove_bus_device(dev); + pci_dev_put(dev); } } diff --git a/arch/arm/mach-ixp2000/ixdp2x00.c b/arch/arm/mach-ixp2000/ixdp2x00.c index af48cb5..aa26550 100644 --- a/arch/arm/mach-ixp2000/ixdp2x00.c +++ b/arch/arm/mach-ixp2000/ixdp2x00.c @@ -241,11 +241,14 @@ void ixdp2x00_slave_pci_postinit(void) /* * Remove PMC device is there is one */ - if((dev = pci_find_slot(1, IXDP2X00_PMC_DEVFN))) + if((dev = pci_get_bus_and_slot(1, IXDP2X00_PMC_DEVFN))) { pci_remove_bus_device(dev); + pci_dev_put(dev); + } - dev = pci_find_slot(0, IXDP2X00_21555_DEVFN); + dev = pci_get_bus_and_slot(0, IXDP2X00_21555_DEVFN); pci_remove_bus_device(dev); + pci_dev_put(dev); } /************************************************************************** -- cgit v0.10.2 From 2bffc23a01a489ad46ba7d61a1a657cecec87cc8 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:17:18 -0700 Subject: sky2: MSI test is only a warning Some motherboards don't implement MSI correctly. The driver handles this but the warning is too verbose and overly cautious. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index c10e7f5..9e31efe 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -3326,9 +3326,8 @@ static int __devinit sky2_test_msi(struct sky2_hw *hw) if (!hw->msi_detected) { /* MSI test failed, go back to INTx mode */ - printk(KERN_WARNING PFX "%s: No interrupt was generated using MSI, " - "switching to INTx mode. Please report this failure to " - "the PCI maintainer and include system chipset information.\n", + printk(KERN_INFO PFX "%s: No interrupt generated using MSI, " + "switching to INTx mode.\n", pci_name(pdev)); err = -EOPNOTSUPP; @@ -3336,6 +3335,7 @@ static int __devinit sky2_test_msi(struct sky2_hw *hw) } sky2_write32(hw, B0_IMSK, 0); + sky2_read32(hw, B0_IMSK); free_irq(pdev->irq, hw); -- cgit v0.10.2 From e561a83be5c9cada5fa3733efdff67a2098a0c8e Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:20:51 -0700 Subject: sky2: turn of workaround timer The workaround timer is not needed in most systems with proper IRQ routing and by perodically waking up it adds to laptop power consumption. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 9e31efe..bd5ccae 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -96,9 +96,9 @@ static int disable_msi = 0; module_param(disable_msi, int, 0); MODULE_PARM_DESC(disable_msi, "Disable Message Signaled Interrupt (MSI)"); -static int idle_timeout = 100; +static int idle_timeout = 0; module_param(idle_timeout, int, 0); -MODULE_PARM_DESC(idle_timeout, "Idle timeout workaround for lost interrupts (ms)"); +MODULE_PARM_DESC(idle_timeout, "Watchdog timer for lost interrupts (ms)"); static const struct pci_device_id sky2_id_table[] = { { PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, 0x9000) }, -- cgit v0.10.2 From ebc646f681a6ad5a81989a6906832e82155df283 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:23:56 -0700 Subject: sky2: phy irq on shutdown When PHY is turned off on shutdown, it causes the IRQ to get stuck on. Make sure and disable the IRQ first, and if IRQ occurs when device is not running, don't access PHY because that will hang. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index bd5ccae..2747e2f 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -1499,6 +1499,11 @@ static int sky2_down(struct net_device *dev) /* Stop more packets from being queued */ netif_stop_queue(dev); + /* Disable port IRQ */ + imask = sky2_read32(hw, B0_IMSK); + imask &= ~portirq_msk[port]; + sky2_write32(hw, B0_IMSK, imask); + sky2_gmac_reset(hw, port); /* Stop transmitter */ @@ -1549,11 +1554,6 @@ static int sky2_down(struct net_device *dev) sky2_write8(hw, SK_REG(port, RX_GMF_CTRL_T), GMF_RST_SET); sky2_write8(hw, SK_REG(port, TX_GMF_CTRL_T), GMF_RST_SET); - /* Disable port IRQ */ - imask = sky2_read32(hw, B0_IMSK); - imask &= ~portirq_msk[port]; - sky2_write32(hw, B0_IMSK, imask); - sky2_phy_power(hw, port, 0); /* turn off LED's */ @@ -1750,13 +1750,13 @@ static void sky2_phy_intr(struct sky2_hw *hw, unsigned port) struct sky2_port *sky2 = netdev_priv(dev); u16 istatus, phystat; + if (!netif_running(dev)) + return; + spin_lock(&sky2->phy_lock); istatus = gm_phy_read(hw, port, PHY_MARV_INT_STAT); phystat = gm_phy_read(hw, port, PHY_MARV_PHY_STAT); - if (!netif_running(dev)) - goto out; - if (netif_msg_intr(sky2)) printk(KERN_INFO PFX "%s: phy interrupt status 0x%x 0x%x\n", sky2->netdev->name, istatus, phystat); -- cgit v0.10.2 From 709c6e7bb07411176ef9ef660242b1e59fc87a6f Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:04 -0700 Subject: sky2: fiber pause bits The advertisement bits for flow control are located in different location on fiber (1000baseX) Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 2747e2f..88b12e8 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -384,20 +384,31 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) adv |= PHY_M_AN_10_FD; if (sky2->advertising & ADVERTISED_10baseT_Half) adv |= PHY_M_AN_10_HD; + + /* desired flow control */ + if (sky2->tx_pause && sky2->rx_pause) /* both */ + adv |= PHY_M_AN_PC | PHY_M_AN_ASP; + else if (sky2->tx_pause) + adv |= PHY_M_AN_ASP; + else if (sky2->rx_pause) + adv |= PHY_M_AN_PC; + + } else { /* special defines for FIBER (88E1040S only) */ if (sky2->advertising & ADVERTISED_1000baseT_Full) adv |= PHY_M_AN_1000X_AFD; if (sky2->advertising & ADVERTISED_1000baseT_Half) adv |= PHY_M_AN_1000X_AHD; - } - /* Set Flow-control capabilities */ - if (sky2->tx_pause && sky2->rx_pause) - adv |= PHY_AN_PAUSE_CAP; /* symmetric */ - else if (sky2->rx_pause && !sky2->tx_pause) - adv |= PHY_AN_PAUSE_ASYM | PHY_AN_PAUSE_CAP; - else if (!sky2->rx_pause && sky2->tx_pause) - adv |= PHY_AN_PAUSE_ASYM; /* local */ + if (sky2->tx_pause && sky2->rx_pause) /* both */ + adv |= PHY_M_P_BOTH_MD_X; + else if (sky2->tx_pause) + adv |= PHY_M_P_ASYM_MD_X; + else if (sky2->rx_pause) + adv |= PHY_M_P_SYM_MD_X; + else + adv |= PHY_M_P_NO_PAUSE_X; + } /* Restart Auto-negotiation */ ctrl |= PHY_CT_ANE | PHY_CT_RE_CFG; -- cgit v0.10.2 From 0edea0f54e1e28bdc1ce6b02d5ca3c4c878cf959 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:07 -0700 Subject: sky2: advertising register 16 bits The advertising bits (from ethtool.h) fit in 16 bits. diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index 43d2acc..0a8d821 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -1860,7 +1860,7 @@ struct sky2_port { dma_addr_t rx_le_map; dma_addr_t tx_le_map; - u32 advertising; /* ADVERTISED_ bits */ + u16 advertising; /* ADVERTISED_ bits */ u16 speed; /* SPEED_1000, SPEED_100, ... */ u8 autoneg; /* AUTONEG_ENABLE, AUTONEG_DISABLE */ u8 duplex; /* DUPLEX_HALF, DUPLEX_FULL */ -- cgit v0.10.2 From 7c74ac1c236457e454804774e832046c1a7cc0bf Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:08 -0700 Subject: sky2: use duplex result bits The result of duplex negotiation is avaliable in the phy status register, so use that to simplify code and avoid rereading the PHY. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 88b12e8..6a594b0 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -1714,26 +1714,7 @@ static int sky2_autoneg_done(struct sky2_port *sky2, u16 aux) } sky2->speed = sky2_phy_speed(hw, aux); - if (sky2->speed == SPEED_1000) { - u16 ctl2 = gm_phy_read(hw, port, PHY_MARV_1000T_CTRL); - u16 lpa2 = gm_phy_read(hw, port, PHY_MARV_1000T_STAT); - if (lpa2 & PHY_B_1000S_MSF) { - printk(KERN_ERR PFX "%s: master/slave fault", - sky2->netdev->name); - return -1; - } - - if ((ctl2 & PHY_M_1000C_AFD) && (lpa2 & PHY_B_1000S_LP_FD)) - sky2->duplex = DUPLEX_FULL; - else - sky2->duplex = DUPLEX_HALF; - } else { - u16 adv = gm_phy_read(hw, port, PHY_MARV_AUNE_ADV); - if ((aux & adv) & PHY_AN_FULL) - sky2->duplex = DUPLEX_FULL; - else - sky2->duplex = DUPLEX_HALF; - } + sky2->duplex = (aux & PHY_M_PS_FULL_DUP) ? DUPLEX_FULL : DUPLEX_HALF; /* Pause bits are offset (9..8) */ if (hw->chip_id == CHIP_ID_YUKON_XL || hw->chip_id == CHIP_ID_YUKON_EC_U) -- cgit v0.10.2 From 7800fddcd05a7dc89276389b96664af4f7890ea7 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:10 -0700 Subject: sky2: don't reset PHY twice Don't need to reset PHY twice on startup. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 6a594b0..20a8c34 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -356,16 +356,7 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) gm_phy_write(hw, port, PHY_MARV_EXT_ADR, pg); } - ctrl = gm_phy_read(hw, port, PHY_MARV_CTRL); - if (sky2->autoneg == AUTONEG_DISABLE) - ctrl &= ~PHY_CT_ANE; - else - ctrl |= PHY_CT_ANE; - - ctrl |= PHY_CT_RESET; - gm_phy_write(hw, port, PHY_MARV_CTRL, ctrl); - - ctrl = 0; + ctrl = PHY_CT_RESET; ct1000 = 0; adv = PHY_AN_CSMA; reg = 0; @@ -450,8 +441,6 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_ON); else sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_OFF); - - ctrl |= PHY_CT_RESET; } gma_write16(hw, port, GM_GP_CTRL, reg); -- cgit v0.10.2 From 16ad91e1c686734aaa5664cd08af0b5e9bf3af61 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:13 -0700 Subject: sky2: flow control setting fixes The result of flow control negotiation should not limit the next negotiatition. If board is plugged into an old half duplex 10Mbit port, without pause, then replugged into a gigabit port, it should negotiate what is desired, not inherit that last negotiation. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 20a8c34..b8f2021 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -284,6 +284,31 @@ static void sky2_gmac_reset(struct sky2_hw *hw, unsigned port) gma_write16(hw, port, GM_RX_CTRL, reg); } +/* flow control to advertise bits */ +static const u16 copper_fc_adv[] = { + [FC_NONE] = 0, + [FC_TX] = PHY_M_AN_ASP, + [FC_RX] = PHY_M_AN_PC, + [FC_BOTH] = PHY_M_AN_PC | PHY_M_AN_ASP, +}; + +/* flow control to advertise bits when using 1000BaseX */ +static const u16 fiber_fc_adv[] = { + [FC_BOTH] = PHY_M_P_BOTH_MD_X, + [FC_TX] = PHY_M_P_ASYM_MD_X, + [FC_RX] = PHY_M_P_SYM_MD_X, + [FC_NONE] = PHY_M_P_NO_PAUSE_X, +}; + +/* flow control to GMA disable bits */ +static const u16 gm_fc_disable[] = { + [FC_NONE] = GM_GPCR_FC_RX_DIS | GM_GPCR_FC_TX_DIS, + [FC_TX] = GM_GPCR_FC_RX_DIS, + [FC_RX] = GM_GPCR_FC_TX_DIS, + [FC_BOTH] = 0, +}; + + static void sky2_phy_init(struct sky2_hw *hw, unsigned port) { struct sky2_port *sky2 = netdev_priv(hw->dev[port]); @@ -376,29 +401,14 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) if (sky2->advertising & ADVERTISED_10baseT_Half) adv |= PHY_M_AN_10_HD; - /* desired flow control */ - if (sky2->tx_pause && sky2->rx_pause) /* both */ - adv |= PHY_M_AN_PC | PHY_M_AN_ASP; - else if (sky2->tx_pause) - adv |= PHY_M_AN_ASP; - else if (sky2->rx_pause) - adv |= PHY_M_AN_PC; - - + adv |= copper_fc_adv[sky2->flow_mode]; } else { /* special defines for FIBER (88E1040S only) */ if (sky2->advertising & ADVERTISED_1000baseT_Full) adv |= PHY_M_AN_1000X_AFD; if (sky2->advertising & ADVERTISED_1000baseT_Half) adv |= PHY_M_AN_1000X_AHD; - if (sky2->tx_pause && sky2->rx_pause) /* both */ - adv |= PHY_M_P_BOTH_MD_X; - else if (sky2->tx_pause) - adv |= PHY_M_P_ASYM_MD_X; - else if (sky2->rx_pause) - adv |= PHY_M_P_SYM_MD_X; - else - adv |= PHY_M_P_NO_PAUSE_X; + adv |= fiber_fc_adv[sky2->flow_mode]; } /* Restart Auto-negotiation */ @@ -424,20 +434,14 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) if (sky2->duplex == DUPLEX_FULL) { reg |= GM_GPCR_DUP_FULL; ctrl |= PHY_CT_DUP_MD; - } else if (sky2->speed != SPEED_1000 && hw->chip_id != CHIP_ID_YUKON_EC_U) { - /* Turn off flow control for 10/100mbps */ - sky2->rx_pause = 0; - sky2->tx_pause = 0; - } + } else if (sky2->speed < SPEED_1000) + sky2->flow_mode = FC_NONE; - if (!sky2->rx_pause) - reg |= GM_GPCR_FC_RX_DIS; - if (!sky2->tx_pause) - reg |= GM_GPCR_FC_TX_DIS; + reg |= gm_fc_disable[sky2->flow_mode]; /* Forward pause packets to GMAC? */ - if (sky2->tx_pause || sky2->rx_pause) + if (sky2->flow_mode & FC_RX) sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_ON); else sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_OFF); @@ -1605,6 +1609,12 @@ static void sky2_link_up(struct sky2_port *sky2) struct sky2_hw *hw = sky2->hw; unsigned port = sky2->port; u16 reg; + static const char *fc_name[] = { + [FC_NONE] = "none", + [FC_TX] = "tx", + [FC_RX] = "rx", + [FC_BOTH] = "both", + }; /* enable Rx/Tx */ reg = gma_read16(hw, port, GM_GP_CTRL); @@ -1648,8 +1658,7 @@ static void sky2_link_up(struct sky2_port *sky2) "%s: Link is up at %d Mbps, %s duplex, flow control %s\n", sky2->netdev->name, sky2->speed, sky2->duplex == DUPLEX_FULL ? "full" : "half", - (sky2->tx_pause && sky2->rx_pause) ? "both" : - sky2->tx_pause ? "tx" : sky2->rx_pause ? "rx" : "none"); + fc_name[sky2->flow_status]); } static void sky2_link_down(struct sky2_port *sky2) @@ -1664,7 +1673,7 @@ static void sky2_link_down(struct sky2_port *sky2) reg &= ~(GM_GPCR_RX_ENA | GM_GPCR_TX_ENA); gma_write16(hw, port, GM_GP_CTRL, reg); - if (sky2->rx_pause && !sky2->tx_pause) { + if (sky2->flow_status == FC_RX) { /* restore Asymmetric Pause bit */ gm_phy_write(hw, port, PHY_MARV_AUNE_ADV, gm_phy_read(hw, port, PHY_MARV_AUNE_ADV) @@ -1683,6 +1692,14 @@ static void sky2_link_down(struct sky2_port *sky2) sky2_phy_init(hw, port); } +static enum flow_control sky2_flow(int rx, int tx) +{ + if (rx) + return tx ? FC_BOTH : FC_RX; + else + return tx ? FC_TX : FC_NONE; +} + static int sky2_autoneg_done(struct sky2_port *sky2, u16 aux) { struct sky2_hw *hw = sky2->hw; @@ -1709,14 +1726,14 @@ static int sky2_autoneg_done(struct sky2_port *sky2, u16 aux) if (hw->chip_id == CHIP_ID_YUKON_XL || hw->chip_id == CHIP_ID_YUKON_EC_U) aux >>= 6; - sky2->rx_pause = (aux & PHY_M_PS_RX_P_EN) != 0; - sky2->tx_pause = (aux & PHY_M_PS_TX_P_EN) != 0; + sky2->flow_status = sky2_flow(aux & PHY_M_PS_RX_P_EN, + aux & PHY_M_PS_TX_P_EN); - if (sky2->duplex == DUPLEX_HALF && sky2->speed != SPEED_1000 + if (sky2->duplex == DUPLEX_HALF && sky2->speed < SPEED_1000 && hw->chip_id != CHIP_ID_YUKON_EC_U) - sky2->rx_pause = sky2->tx_pause = 0; + sky2->flow_status = FC_NONE; - if (sky2->rx_pause || sky2->tx_pause) + if (aux & PHY_M_PS_RX_P_EN) sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_ON); else sky2_write8(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_OFF); @@ -2729,7 +2746,7 @@ static int sky2_nway_reset(struct net_device *dev) { struct sky2_port *sky2 = netdev_priv(dev); - if (sky2->autoneg != AUTONEG_ENABLE) + if (!netif_running(dev) || sky2->autoneg != AUTONEG_ENABLE) return -EINVAL; sky2_phy_reinit(sky2); @@ -2971,8 +2988,20 @@ static void sky2_get_pauseparam(struct net_device *dev, { struct sky2_port *sky2 = netdev_priv(dev); - ecmd->tx_pause = sky2->tx_pause; - ecmd->rx_pause = sky2->rx_pause; + switch (sky2->flow_mode) { + case FC_NONE: + ecmd->tx_pause = ecmd->rx_pause = 0; + break; + case FC_TX: + ecmd->tx_pause = 1, ecmd->rx_pause = 0; + break; + case FC_RX: + ecmd->tx_pause = 0, ecmd->rx_pause = 1; + break; + case FC_BOTH: + ecmd->tx_pause = ecmd->rx_pause = 1; + } + ecmd->autoneg = sky2->autoneg; } @@ -2982,10 +3011,10 @@ static int sky2_set_pauseparam(struct net_device *dev, struct sky2_port *sky2 = netdev_priv(dev); sky2->autoneg = ecmd->autoneg; - sky2->tx_pause = ecmd->tx_pause != 0; - sky2->rx_pause = ecmd->rx_pause != 0; + sky2->flow_mode = sky2_flow(ecmd->rx_pause, ecmd->tx_pause); - sky2_phy_reinit(sky2); + if (netif_running(dev)) + sky2_phy_reinit(sky2); return 0; } @@ -3215,8 +3244,8 @@ static __devinit struct net_device *sky2_init_netdev(struct sky2_hw *hw, /* Auto speed and flow control */ sky2->autoneg = AUTONEG_ENABLE; - sky2->tx_pause = 1; - sky2->rx_pause = 1; + sky2->flow_mode = FC_BOTH; + sky2->duplex = -1; sky2->speed = -1; sky2->advertising = sky2_supported_modes(hw); diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index 0a8d821..3f05492 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -1828,6 +1828,13 @@ struct rx_ring_info { dma_addr_t frag_addr[ETH_JUMBO_MTU >> PAGE_SHIFT]; }; +enum flow_control { + FC_NONE = 0, + FC_TX = 1, + FC_RX = 2, + FC_BOTH = 3, +}; + struct sky2_port { struct sky2_hw *hw; struct net_device *netdev; @@ -1864,9 +1871,9 @@ struct sky2_port { u16 speed; /* SPEED_1000, SPEED_100, ... */ u8 autoneg; /* AUTONEG_ENABLE, AUTONEG_DISABLE */ u8 duplex; /* DUPLEX_HALF, DUPLEX_FULL */ - u8 rx_pause; - u8 tx_pause; u8 rx_csum; + enum flow_control flow_mode; + enum flow_control flow_status; struct net_device_stats net_stats; -- cgit v0.10.2 From b6d7773462df13c105c19ab89706687ded839844 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:16 -0700 Subject: sky2: no message on rx fifo overflow Under high load it is possible to make the receiver FIFO get overloaded. The driver/hardware recover properly, so there is no reason to fill the log with lots of extra messages, just update counter. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index b8f2021..6eddd0f 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -2014,6 +2014,10 @@ oversize: error: ++sky2->net_stats.rx_errors; + if (status & GMR_FS_RX_FF_OV) { + sky2->net_stats.rx_fifo_errors++; + goto resubmit; + } if (netif_msg_rx_err(sky2) && net_ratelimit()) printk(KERN_INFO PFX "%s: rx error, status 0x%x length %d\n", @@ -2025,8 +2029,6 @@ error: sky2->net_stats.rx_frame_errors++; if (status & GMR_FS_CRC_ERR) sky2->net_stats.rx_crc_errors++; - if (status & GMR_FS_RX_FF_OV) - sky2->net_stats.rx_fifo_errors++; goto resubmit; } -- cgit v0.10.2 From 52c89cac6781dea0ee2426821cd3effae1a925d3 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:18 -0700 Subject: sky2: version 1.9 Mark version, this has been a lot of patches. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 6eddd0f..c193383 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -50,7 +50,7 @@ #include "sky2.h" #define DRV_NAME "sky2" -#define DRV_VERSION "1.9" +#define DRV_VERSION "1.10" #define PFX DRV_NAME " " /* -- cgit v0.10.2 From a052b52f4b6b77503af2647dc0c7415939d8232a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:23 -0700 Subject: sky2: accept multicast pause frames When using flow control, the PHY needs to accept multicast pause frames. Without this fix, these frames were getting discarded by the PHY before doing any flow control. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index c193383..67ecd66 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -2850,6 +2850,14 @@ static int sky2_set_mac_address(struct net_device *dev, void *p) return 0; } +static void inline sky2_add_filter(u8 filter[8], const u8 *addr) +{ + u32 bit; + + bit = ether_crc(ETH_ALEN, addr) & 63; + filter[bit >> 3] |= 1 << (bit & 7); +} + static void sky2_set_multicast(struct net_device *dev) { struct sky2_port *sky2 = netdev_priv(dev); @@ -2858,7 +2866,10 @@ static void sky2_set_multicast(struct net_device *dev) struct dev_mc_list *list = dev->mc_list; u16 reg; u8 filter[8]; + int rx_pause; + static const u8 pause_mc_addr[ETH_ALEN] = { 0x1, 0x80, 0xc2, 0x0, 0x0, 0x1 }; + rx_pause = (sky2->flow_status == FC_RX || sky2->flow_status == FC_BOTH); memset(filter, 0, sizeof(filter)); reg = gma_read16(hw, port, GM_RX_CTRL); @@ -2866,18 +2877,19 @@ static void sky2_set_multicast(struct net_device *dev) if (dev->flags & IFF_PROMISC) /* promiscuous */ reg &= ~(GM_RXCR_UCF_ENA | GM_RXCR_MCF_ENA); - else if ((dev->flags & IFF_ALLMULTI) || dev->mc_count > 16) /* all multicast */ + else if (dev->flags & IFF_ALLMULTI) memset(filter, 0xff, sizeof(filter)); - else if (dev->mc_count == 0) /* no multicast */ + else if (dev->mc_count == 0 && !rx_pause) reg &= ~GM_RXCR_MCF_ENA; else { int i; reg |= GM_RXCR_MCF_ENA; - for (i = 0; list && i < dev->mc_count; i++, list = list->next) { - u32 bit = ether_crc(ETH_ALEN, list->dmi_addr) & 0x3f; - filter[bit / 8] |= 1 << (bit % 8); - } + if (rx_pause) + sky2_add_filter(filter, pause_mc_addr); + + for (i = 0; list && i < dev->mc_count; i++, list = list->next) + sky2_add_filter(filter, list->dmi_addr); } gma_write16(hw, port, GM_MC_ADDR_H1, -- cgit v0.10.2 From 4e4bc305e16440ab38060d61fbcb7d774881d2f1 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 10:24:25 -0700 Subject: sky2: GMAC pause frame This reverts earlier change that attempted to fix flow control. Device needs to discard pause frames, otherwise it hangs after a while. Signed-off-by: Stephen Hemminger diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index 3f05492..6d2a23f 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -1576,7 +1576,7 @@ enum { GMR_FS_ANY_ERR = GMR_FS_RX_FF_OV | GMR_FS_CRC_ERR | GMR_FS_FRAGMENT | GMR_FS_LONG_ERR | - GMR_FS_MII_ERR | GMR_FS_BAD_FC | + GMR_FS_MII_ERR | GMR_FS_GOOD_FC | GMR_FS_BAD_FC | GMR_FS_UN_SIZE | GMR_FS_JABBER, }; -- cgit v0.10.2 From 41072a1be57f63bf83afc31c44d72de018d800fa Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Tue, 17 Oct 2006 13:47:40 -0400 Subject: [PATCH] zd1211rw: fix build-break caused by association race fix The break was caused by 7c28ad2d83ecc637237fe684659a6afbce0bb2a8. Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index 2d12837..a7d29bd 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -1099,7 +1099,7 @@ static void link_led_handler(void *p) int r; spin_lock_irq(&mac->lock); - is_associated = sm->associated != 0; + is_associated = sm->associnfo.associated != 0; spin_unlock_irq(&mac->lock); r = zd_chip_control_leds(chip, -- cgit v0.10.2 From 64f89798da35f43c6ef6afda0541e25034513458 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 17 Oct 2006 13:57:18 -0700 Subject: USB: revert EHCI VIA workaround patch This reverts 26f953fd884ea4879585287917f855c63c6b2666 which caused resume problems on the mac mini. Cc: David Brownell Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c index 23b95b2..34b7a31 100644 --- a/drivers/usb/host/ehci-dbg.c +++ b/drivers/usb/host/ehci-dbg.c @@ -754,7 +754,9 @@ show_registers (struct class_device *class_dev, char *buf) } if (ehci->reclaim) { - temp = scnprintf (next, size, "reclaim qh %p\n", ehci->reclaim); + temp = scnprintf (next, size, "reclaim qh %p%s\n", + ehci->reclaim, + ehci->reclaim_ready ? " ready" : ""); size -= temp; next += temp; } diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c index aac6ec5..9030994 100644 --- a/drivers/usb/host/ehci-hcd.c +++ b/drivers/usb/host/ehci-hcd.c @@ -111,7 +111,7 @@ static const char hcd_name [] = "ehci_hcd"; #define EHCI_TUNE_MULT_TT 1 #define EHCI_TUNE_FLS 2 /* (small) 256 frame schedule */ -#define EHCI_IAA_MSECS 10 /* arbitrary */ +#define EHCI_IAA_JIFFIES (HZ/100) /* arbitrary; ~10 msec */ #define EHCI_IO_JIFFIES (HZ/10) /* io watchdog > irq_thresh */ #define EHCI_ASYNC_JIFFIES (HZ/20) /* async idle timeout */ #define EHCI_SHRINK_JIFFIES (HZ/200) /* async qh unlink delay */ @@ -254,7 +254,6 @@ static void ehci_quiesce (struct ehci_hcd *ehci) /*-------------------------------------------------------------------------*/ -static void end_unlink_async (struct ehci_hcd *ehci); static void ehci_work(struct ehci_hcd *ehci); #include "ehci-hub.c" @@ -264,37 +263,25 @@ static void ehci_work(struct ehci_hcd *ehci); /*-------------------------------------------------------------------------*/ -static void ehci_iaa_watchdog (unsigned long param) +static void ehci_watchdog (unsigned long param) { struct ehci_hcd *ehci = (struct ehci_hcd *) param; unsigned long flags; - u32 status; spin_lock_irqsave (&ehci->lock, flags); - WARN_ON(!ehci->reclaim); - /* lost IAA irqs wedge things badly; seen first with a vt8235 */ + /* lost IAA irqs wedge things badly; seen with a vt8235 */ if (ehci->reclaim) { - status = readl (&ehci->regs->status); + u32 status = readl (&ehci->regs->status); if (status & STS_IAA) { ehci_vdbg (ehci, "lost IAA\n"); COUNT (ehci->stats.lost_iaa); writel (STS_IAA, &ehci->regs->status); - end_unlink_async (ehci); + ehci->reclaim_ready = 1; } } - spin_unlock_irqrestore (&ehci->lock, flags); -} - -static void ehci_watchdog (unsigned long param) -{ - struct ehci_hcd *ehci = (struct ehci_hcd *) param; - unsigned long flags; - - spin_lock_irqsave (&ehci->lock, flags); - - /* stop async processing after it's idled a bit */ + /* stop async processing after it's idled a bit */ if (test_bit (TIMER_ASYNC_OFF, &ehci->actions)) start_unlink_async (ehci, ehci->async); @@ -345,6 +332,8 @@ static void ehci_port_power (struct ehci_hcd *ehci, int is_on) static void ehci_work (struct ehci_hcd *ehci) { timer_action_done (ehci, TIMER_IO_WATCHDOG); + if (ehci->reclaim_ready) + end_unlink_async (ehci); /* another CPU may drop ehci->lock during a schedule scan while * it reports urb completions. this flag guards against bogus @@ -379,7 +368,6 @@ static void ehci_stop (struct usb_hcd *hcd) /* no more interrupts ... */ del_timer_sync (&ehci->watchdog); - del_timer_sync (&ehci->iaa_watchdog); spin_lock_irq(&ehci->lock); if (HC_IS_RUNNING (hcd->state)) @@ -426,10 +414,6 @@ static int ehci_init(struct usb_hcd *hcd) ehci->watchdog.function = ehci_watchdog; ehci->watchdog.data = (unsigned long) ehci; - init_timer(&ehci->iaa_watchdog); - ehci->iaa_watchdog.function = ehci_iaa_watchdog; - ehci->iaa_watchdog.data = (unsigned long) ehci; - /* * hw default: 1K periodic list heads, one per frame. * periodic_size can shrink by USBCMD update if hcc_params allows. @@ -446,6 +430,7 @@ static int ehci_init(struct usb_hcd *hcd) ehci->i_thresh = 2 + HCC_ISOC_THRES(hcc_params); ehci->reclaim = NULL; + ehci->reclaim_ready = 0; ehci->next_uframe = -1; /* @@ -619,7 +604,7 @@ static irqreturn_t ehci_irq (struct usb_hcd *hcd) /* complete the unlinking of some qh [4.15.2.3] */ if (status & STS_IAA) { COUNT (ehci->stats.reclaim); - end_unlink_async (ehci); + ehci->reclaim_ready = 1; bh = 1; } @@ -723,14 +708,10 @@ static int ehci_urb_enqueue ( static void unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh) { - // BUG_ON(qh->qh_state != QH_STATE_LINKED); - - /* failfast */ - if (!HC_IS_RUNNING (ehci_to_hcd(ehci)->state)) - end_unlink_async (ehci); - - /* defer till later if busy */ - else if (ehci->reclaim) { + /* if we need to use IAA and it's busy, defer */ + if (qh->qh_state == QH_STATE_LINKED + && ehci->reclaim + && HC_IS_RUNNING (ehci_to_hcd(ehci)->state)) { struct ehci_qh *last; for (last = ehci->reclaim; @@ -740,8 +721,12 @@ static void unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh) qh->qh_state = QH_STATE_UNLINK_WAIT; last->reclaim = qh; - /* start IAA cycle */ - } else + /* bypass IAA if the hc can't care */ + } else if (!HC_IS_RUNNING (ehci_to_hcd(ehci)->state) && ehci->reclaim) + end_unlink_async (ehci); + + /* something else might have unlinked the qh by now */ + if (qh->qh_state == QH_STATE_LINKED) start_unlink_async (ehci, qh); } @@ -763,19 +748,7 @@ static int ehci_urb_dequeue (struct usb_hcd *hcd, struct urb *urb) qh = (struct ehci_qh *) urb->hcpriv; if (!qh) break; - switch (qh->qh_state) { - case QH_STATE_LINKED: - case QH_STATE_COMPLETING: - unlink_async (ehci, qh); - break; - case QH_STATE_UNLINK: - case QH_STATE_UNLINK_WAIT: - /* already started */ - break; - case QH_STATE_IDLE: - WARN_ON(1); - break; - } + unlink_async (ehci, qh); break; case PIPE_INTERRUPT: @@ -867,7 +840,6 @@ rescan: unlink_async (ehci, qh); /* FALL THROUGH */ case QH_STATE_UNLINK: /* wait for hw to finish? */ - case QH_STATE_UNLINK_WAIT: idle_timeout: spin_unlock_irqrestore (&ehci->lock, flags); schedule_timeout_uninterruptible(1); diff --git a/drivers/usb/host/ehci-hub.c b/drivers/usb/host/ehci-hub.c index 2012213..1b20722 100644 --- a/drivers/usb/host/ehci-hub.c +++ b/drivers/usb/host/ehci-hub.c @@ -48,7 +48,7 @@ static int ehci_bus_suspend (struct usb_hcd *hcd) } ehci->command = readl (&ehci->regs->command); if (ehci->reclaim) - end_unlink_async (ehci); + ehci->reclaim_ready = 1; ehci_work(ehci); /* suspend any active/unsuspended ports, maybe allow wakeup */ diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c index 35e3fab..e51c1ed8 100644 --- a/drivers/usb/host/ehci-pci.c +++ b/drivers/usb/host/ehci-pci.c @@ -303,7 +303,7 @@ restart: /* emptying the schedule aborts any urbs */ spin_lock_irq(&ehci->lock); if (ehci->reclaim) - end_unlink_async (ehci); + ehci->reclaim_ready = 1; ehci_work(ehci); spin_unlock_irq(&ehci->lock); diff --git a/drivers/usb/host/ehci-q.c b/drivers/usb/host/ehci-q.c index 4632727..62e46dc 100644 --- a/drivers/usb/host/ehci-q.c +++ b/drivers/usb/host/ehci-q.c @@ -967,7 +967,7 @@ static void end_unlink_async (struct ehci_hcd *ehci) struct ehci_qh *qh = ehci->reclaim; struct ehci_qh *next; - iaa_watchdog_done (ehci); + timer_action_done (ehci, TIMER_IAA_WATCHDOG); // qh->hw_next = cpu_to_le32 (qh->qh_dma); qh->qh_state = QH_STATE_IDLE; @@ -977,6 +977,7 @@ static void end_unlink_async (struct ehci_hcd *ehci) /* other unlink(s) may be pending (in QH_STATE_UNLINK_WAIT) */ next = qh->reclaim; ehci->reclaim = next; + ehci->reclaim_ready = 0; qh->reclaim = NULL; qh_completions (ehci, qh); @@ -1051,10 +1052,11 @@ static void start_unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh) return; } + ehci->reclaim_ready = 0; cmd |= CMD_IAAD; writel (cmd, &ehci->regs->command); (void) readl (&ehci->regs->command); - iaa_watchdog_start (ehci); + timer_action (ehci, TIMER_IAA_WATCHDOG); } /*-------------------------------------------------------------------------*/ diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h index 6aac39f..bbc3082 100644 --- a/drivers/usb/host/ehci.h +++ b/drivers/usb/host/ehci.h @@ -58,6 +58,7 @@ struct ehci_hcd { /* one per controller */ /* async schedule support */ struct ehci_qh *async; struct ehci_qh *reclaim; + unsigned reclaim_ready : 1; unsigned scanning : 1; /* periodic schedule support */ @@ -80,7 +81,6 @@ struct ehci_hcd { /* one per controller */ struct dma_pool *itd_pool; /* itd per iso urb */ struct dma_pool *sitd_pool; /* sitd per split iso urb */ - struct timer_list iaa_watchdog; struct timer_list watchdog; unsigned long actions; unsigned stamp; @@ -114,21 +114,9 @@ static inline struct usb_hcd *ehci_to_hcd (struct ehci_hcd *ehci) } -static inline void -iaa_watchdog_start (struct ehci_hcd *ehci) -{ - WARN_ON(timer_pending(&ehci->iaa_watchdog)); - mod_timer (&ehci->iaa_watchdog, - jiffies + msecs_to_jiffies(EHCI_IAA_MSECS)); -} - -static inline void iaa_watchdog_done (struct ehci_hcd *ehci) -{ - del_timer (&ehci->iaa_watchdog); -} - enum ehci_timer_action { TIMER_IO_WATCHDOG, + TIMER_IAA_WATCHDOG, TIMER_ASYNC_SHRINK, TIMER_ASYNC_OFF, }; @@ -146,6 +134,9 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action) unsigned long t; switch (action) { + case TIMER_IAA_WATCHDOG: + t = EHCI_IAA_JIFFIES; + break; case TIMER_IO_WATCHDOG: t = EHCI_IO_JIFFIES; break; @@ -162,7 +153,8 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action) // async queue SHRINK often precedes IAA. while it's ready // to go OFF neither can matter, and afterwards the IO // watchdog stops unless there's still periodic traffic. - if (time_before_eq(t, ehci->watchdog.expires) + if (action != TIMER_IAA_WATCHDOG + && t > ehci->watchdog.expires && timer_pending (&ehci->watchdog)) return; mod_timer (&ehci->watchdog, t); -- cgit v0.10.2 From 3bbe486b361b317ac7103378ed3d1aab4779715e Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Tue, 17 Oct 2006 14:28:16 -0700 Subject: [IA64] perfmon fix for global IRQ fix Missed one piece of ia64 fallout from the global IRQ patch 7d12e780e003f93433d49ce78cfedf4b4c52adc5 Perfmon interrupt handler needs to use get_irq_regs() too. Acked-by: stephane eranian Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 281004f..3aaede0 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -5558,12 +5558,13 @@ report_spurious2: } static irqreturn_t -pfm_interrupt_handler(int irq, void *arg, struct pt_regs *regs) +pfm_interrupt_handler(int irq, void *arg) { unsigned long start_cycles, total_cycles; unsigned long min, max; int this_cpu; int ret; + struct pt_regs *regs = get_irq_regs(); this_cpu = get_cpu(); if (likely(!pfm_alt_intr_handler)) { -- cgit v0.10.2 From 8d32e3ae5972641ee9eb813e7a5c44a2b85d3694 Mon Sep 17 00:00:00 2001 From: Ping Cheng Date: Tue, 26 Sep 2006 13:34:47 -0700 Subject: USB: Wacom driver updates This fixes some issues with the current wacom driver due to the split of the driver into different pieces and adds support for the Intuos3 4x6 Signed-off-by: Ping Cheng Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/Makefile b/drivers/usb/input/Makefile index 295f459..71437db 100644 --- a/drivers/usb/input/Makefile +++ b/drivers/usb/input/Makefile @@ -3,7 +3,7 @@ # # Multipart objects. -wacom-objs := wacom_sys.o wacom_wac.o +wacom-objs := wacom_wac.o wacom_sys.o usbhid-objs := hid-core.o # Optional parts of multipart objects. diff --git a/drivers/usb/input/wacom.h b/drivers/usb/input/wacom.h index 7b3840e..1cf08f0 100644 --- a/drivers/usb/input/wacom.h +++ b/drivers/usb/input/wacom.h @@ -63,6 +63,7 @@ * v1.46 (pc) - Split wacom.c into wacom_sys.c and wacom_wac.c, * - where wacom_sys.c deals with system specific code, * - and wacom_wac.c deals with Wacom specific code + * - Support Intuos3 4x6 */ /* @@ -118,6 +119,7 @@ extern void wacom_input_sync(void *wcombo); extern void wacom_init_input_dev(struct input_dev *input_dev, struct wacom_wac *wacom_wac); extern void input_dev_g4(struct input_dev *input_dev, struct wacom_wac *wacom_wac); extern void input_dev_g(struct input_dev *input_dev, struct wacom_wac *wacom_wac); +extern void input_dev_i3s(struct input_dev *input_dev, struct wacom_wac *wacom_wac); extern void input_dev_i3(struct input_dev *input_dev, struct wacom_wac *wacom_wac); extern void input_dev_i(struct input_dev *input_dev, struct wacom_wac *wacom_wac); extern void input_dev_pl(struct input_dev *input_dev, struct wacom_wac *wacom_wac); diff --git a/drivers/usb/input/wacom_sys.c b/drivers/usb/input/wacom_sys.c index d233c37..3c27f9b 100644 --- a/drivers/usb/input/wacom_sys.c +++ b/drivers/usb/input/wacom_sys.c @@ -110,7 +110,7 @@ __u16 wacom_be16_to_cpu(unsigned char *data) __u16 wacom_le16_to_cpu(unsigned char *data) { __u16 value; - value = be16_to_cpu(*(__be16 *) data); + value = le16_to_cpu(*(__le16 *) data); return value; } @@ -143,7 +143,7 @@ void input_dev_g4(struct input_dev *input_dev, struct wacom_wac *wacom_wac) input_dev->evbit[0] |= BIT(EV_MSC); input_dev->mscbit[0] |= BIT(MSC_SERIAL); input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_FINGER); - input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3) | BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7); + input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_4); } void input_dev_g(struct input_dev *input_dev, struct wacom_wac *wacom_wac) @@ -155,11 +155,16 @@ void input_dev_g(struct input_dev *input_dev, struct wacom_wac *wacom_wac) input_set_abs_params(input_dev, ABS_DISTANCE, 0, wacom_wac->features->distance_max, 0, 0); } -void input_dev_i3(struct input_dev *input_dev, struct wacom_wac *wacom_wac) +void input_dev_i3s(struct input_dev *input_dev, struct wacom_wac *wacom_wac) { input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_FINGER); - input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3) | BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7); + input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3); input_set_abs_params(input_dev, ABS_RX, 0, 4097, 0, 0); +} + +void input_dev_i3(struct input_dev *input_dev, struct wacom_wac *wacom_wac) +{ + input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7); input_set_abs_params(input_dev, ABS_RY, 0, 4097, 0, 0); } @@ -244,7 +249,7 @@ static int wacom_probe(struct usb_interface *intf, const struct usb_device_id *i usb_fill_int_urb(wacom->irq, dev, usb_rcvintpipe(dev, endpoint->bEndpointAddress), wacom_wac->data, wacom_wac->features->pktlen, - wacom_wac->features->irq, wacom, endpoint->bInterval); + wacom_sys_irq, wacom, endpoint->bInterval); wacom->irq->transfer_dma = wacom->data_dma; wacom->irq->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; diff --git a/drivers/usb/input/wacom_wac.c b/drivers/usb/input/wacom_wac.c index aa31d22..92726fe 100644 --- a/drivers/usb/input/wacom_wac.c +++ b/drivers/usb/input/wacom_wac.c @@ -191,9 +191,9 @@ static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo) wacom_report_key(wcombo, BTN_LEFT, data[1] & 0x01); wacom_report_key(wcombo, BTN_RIGHT, data[1] & 0x02); if (wacom->features->type == WACOM_G4) - wacom_report_abs(wcombo, ABS_DISTANCE, data[6]); + wacom_report_abs(wcombo, ABS_DISTANCE, data[6] & 0x3f); else - wacom_report_abs(wcombo, ABS_DISTANCE, data[7]); + wacom_report_abs(wcombo, ABS_DISTANCE, data[7] & 0x3f); break; } } @@ -303,8 +303,9 @@ static int wacom_intuos_inout(struct wacom_wac *wacom, void *wcombo) wacom->tool[idx] = BTN_TOOL_PEN; } /* only large I3 support Lens Cursor */ - if(!((wacom->tool[idx] == BTN_TOOL_LENS) && - (wacom->features->type == INTUOS3))) { + if(!((wacom->tool[idx] == BTN_TOOL_LENS) + && ((wacom->features->type == INTUOS3) + || (wacom->features->type == INTUOS3S)))) { wacom_report_abs(wcombo, ABS_MISC, wacom->id[idx]); /* report tool id */ wacom_report_key(wcombo, wacom->tool[idx], 1); wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, wacom->serial[idx]); @@ -315,10 +316,14 @@ static int wacom_intuos_inout(struct wacom_wac *wacom, void *wcombo) /* Exit report */ if ((data[1] & 0xfe) == 0x80) { - wacom_report_key(wcombo, wacom->tool[idx], 0); - wacom_report_abs(wcombo, ABS_MISC, 0); /* reset tool id */ - wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, wacom->serial[idx]); - return 2; + if(!((wacom->tool[idx] == BTN_TOOL_LENS) + && ((wacom->features->type == INTUOS3) + || (wacom->features->type == INTUOS3S)))) { + wacom_report_key(wcombo, wacom->tool[idx], 0); + wacom_report_abs(wcombo, ABS_MISC, 0); /* reset tool id */ + wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, wacom->serial[idx]); + return 2; + } } return 0; } @@ -382,7 +387,8 @@ static int wacom_intuos_irq(struct wacom_wac *wacom, void *wcombo) wacom_report_abs(wcombo, ABS_RX, ((data[1] & 0x1f) << 8) | data[2]); wacom_report_abs(wcombo, ABS_RY, ((data[3] & 0x1f) << 8) | data[4]); - if((data[5] & 0x0f) | (data[6] & 0x0f) | (data[1] & 0x1f) | data[2]) + if((data[5] & 0x0f) | (data[6] & 0x0f) | (data[1] & 0x1f) | + data[2] | (data[3] & 0x1f) | data[4]) wacom_report_key(wcombo, wacom->tool[1], 1); else wacom_report_key(wcombo, wacom->tool[1], 0); @@ -432,7 +438,7 @@ static int wacom_intuos_irq(struct wacom_wac *wacom, void *wcombo) ((t - 1) / 2) : -t / 2); } - } else if (!(data[1] & 0x10) && wacom->features->type < INTUOS3) { + } else if (!(data[1] & 0x10) && wacom->features->type < INTUOS3S) { /* 4D mouse packet */ wacom_report_key(wcombo, BTN_LEFT, data[8] & 0x01); wacom_report_key(wcombo, BTN_MIDDLE, data[8] & 0x02); @@ -452,12 +458,12 @@ static int wacom_intuos_irq(struct wacom_wac *wacom, void *wcombo) - ((data[8] & 0x02) >> 1)); /* I3 2D mouse side buttons */ - if (wacom->features->type == INTUOS3) { + if (wacom->features->type >= INTUOS3S && wacom->features->type <= INTUOS3L) { wacom_report_key(wcombo, BTN_SIDE, data[8] & 0x40); wacom_report_key(wcombo, BTN_EXTRA, data[8] & 0x20); } - } else if (wacom->features->type < INTUOS3) { + } else if (wacom->features->type < INTUOS3S || wacom->features->type == INTUOS3L) { /* Lens cursor packets */ wacom_report_key(wcombo, BTN_LEFT, data[8] & 0x01); wacom_report_key(wcombo, BTN_MIDDLE, data[8] & 0x02); @@ -490,6 +496,7 @@ int wacom_wac_irq(struct wacom_wac *wacom_wac, void *wcombo) return (wacom_ptu_irq(wacom_wac, wcombo)); break; case INTUOS: + case INTUOS3S: case INTUOS3: case INTUOS3L: case CINTIQ: @@ -515,6 +522,8 @@ void wacom_init_input_dev(struct input_dev *input_dev, struct wacom_wac *wacom_w case CINTIQ: input_dev_i3(input_dev, wacom_wac); /* fall through */ + case INTUOS3S: + input_dev_i3s(input_dev, wacom_wac); case INTUOS: input_dev_i(input_dev, wacom_wac); break; @@ -530,49 +539,50 @@ void wacom_init_input_dev(struct input_dev *input_dev, struct wacom_wac *wacom_w } static struct wacom_features wacom_features[] = { - { "Wacom Penpartner", 7, 5040, 3780, 255, 32, PENPARTNER, wacom_sys_irq }, - { "Wacom Graphire", 8, 10206, 7422, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Graphire2 4x5", 8, 10206, 7422, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Graphire2 5x7", 8, 13918, 10206, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Graphire3", 8, 10208, 7424, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Graphire3 6x8", 8, 16704, 12064, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Graphire4 4x5", 8, 10208, 7424, 511, 32, WACOM_G4, wacom_sys_irq }, - { "Wacom Graphire4 6x8", 8, 16704, 12064, 511, 32, WACOM_G4, wacom_sys_irq }, - { "Wacom Volito", 8, 5104, 3712, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom PenStation2", 8, 3250, 2320, 255, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Volito2 4x5", 8, 5104, 3712, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Volito2 2x3", 8, 3248, 2320, 511, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom PenPartner2", 8, 3250, 2320, 255, 32, GRAPHIRE, wacom_sys_irq }, - { "Wacom Intuos 4x5", 10, 12700, 10600, 1023, 15, INTUOS, wacom_sys_irq}, - { "Wacom Intuos 6x8", 10, 20320, 16240, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos 9x12", 10, 30480, 24060, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos 12x12", 10, 30480, 31680, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos 12x18", 10, 45720, 31680, 1023, 15, INTUOS, wacom_sys_irq}, - { "Wacom PL400", 8, 5408, 4056, 255, 32, PL, wacom_sys_irq }, - { "Wacom PL500", 8, 6144, 4608, 255, 32, PL, wacom_sys_irq }, - { "Wacom PL600", 8, 6126, 4604, 255, 32, PL, wacom_sys_irq }, - { "Wacom PL600SX", 8, 6260, 5016, 255, 32, PL, wacom_sys_irq }, - { "Wacom PL550", 8, 6144, 4608, 511, 32, PL, wacom_sys_irq }, - { "Wacom PL800", 8, 7220, 5780, 511, 32, PL, wacom_sys_irq }, - { "Wacom PL700", 8, 6758, 5406, 511, 32, PL, wacom_sys_irq }, - { "Wacom PL510", 8, 6282, 4762, 511, 32, PL, wacom_sys_irq }, - { "Wacom DTU710", 8, 34080, 27660, 511, 32, PL, wacom_sys_irq }, - { "Wacom DTF521", 8, 6282, 4762, 511, 32, PL, wacom_sys_irq }, - { "Wacom DTF720", 8, 6858, 5506, 511, 32, PL, wacom_sys_irq }, - { "Wacom Cintiq Partner",8, 20480, 15360, 511, 32, PTU, wacom_sys_irq }, - { "Wacom Intuos2 4x5", 10, 12700, 10600, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos2 6x8", 10, 20320, 16240, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos2 9x12", 10, 30480, 24060, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos2 12x12", 10, 30480, 31680, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos2 12x18", 10, 45720, 31680, 1023, 15, INTUOS, wacom_sys_irq }, - { "Wacom Intuos3 4x5", 10, 25400, 20320, 1023, 15, INTUOS3, wacom_sys_irq }, - { "Wacom Intuos3 6x8", 10, 40640, 30480, 1023, 15, INTUOS3, wacom_sys_irq }, - { "Wacom Intuos3 9x12", 10, 60960, 45720, 1023, 15, INTUOS3, wacom_sys_irq }, - { "Wacom Intuos3 12x12", 10, 60960, 60960, 1023, 15, INTUOS3L, wacom_sys_irq }, - { "Wacom Intuos3 12x19", 10, 97536, 60960, 1023, 15, INTUOS3L, wacom_sys_irq }, - { "Wacom Intuos3 6x11", 10, 54204, 31750, 1023, 15, INTUOS3, wacom_sys_irq }, - { "Wacom Cintiq 21UX", 10, 87200, 65600, 1023, 15, CINTIQ, wacom_sys_irq }, - { "Wacom Intuos2 6x8", 10, 20320, 16240, 1023, 15, INTUOS, wacom_sys_irq }, + { "Wacom Penpartner", 7, 5040, 3780, 255, 0, PENPARTNER }, + { "Wacom Graphire", 8, 10206, 7422, 511, 63, GRAPHIRE }, + { "Wacom Graphire2 4x5", 8, 10206, 7422, 511, 63, GRAPHIRE }, + { "Wacom Graphire2 5x7", 8, 13918, 10206, 511, 63, GRAPHIRE }, + { "Wacom Graphire3", 8, 10208, 7424, 511, 63, GRAPHIRE }, + { "Wacom Graphire3 6x8", 8, 16704, 12064, 511, 63, GRAPHIRE }, + { "Wacom Graphire4 4x5", 8, 10208, 7424, 511, 63, WACOM_G4 }, + { "Wacom Graphire4 6x8", 8, 16704, 12064, 511, 63, WACOM_G4 }, + { "Wacom Volito", 8, 5104, 3712, 511, 0, GRAPHIRE }, + { "Wacom PenStation2", 8, 3250, 2320, 255, 0, GRAPHIRE }, + { "Wacom Volito2 4x5", 8, 5104, 3712, 511, 0, GRAPHIRE }, + { "Wacom Volito2 2x3", 8, 3248, 2320, 511, 0, GRAPHIRE }, + { "Wacom PenPartner2", 8, 3250, 2320, 255, 0, GRAPHIRE }, + { "Wacom Intuos 4x5", 10, 12700, 10600, 1023, 63, INTUOS }, + { "Wacom Intuos 6x8", 10, 20320, 16240, 1023, 63, INTUOS }, + { "Wacom Intuos 9x12", 10, 30480, 24060, 1023, 63, INTUOS }, + { "Wacom Intuos 12x12", 10, 30480, 31680, 1023, 63, INTUOS }, + { "Wacom Intuos 12x18", 10, 45720, 31680, 1023, 63, INTUOS }, + { "Wacom PL400", 8, 5408, 4056, 255, 0, PL }, + { "Wacom PL500", 8, 6144, 4608, 255, 0, PL }, + { "Wacom PL600", 8, 6126, 4604, 255, 0, PL }, + { "Wacom PL600SX", 8, 6260, 5016, 255, 0, PL }, + { "Wacom PL550", 8, 6144, 4608, 511, 0, PL }, + { "Wacom PL800", 8, 7220, 5780, 511, 0, PL }, + { "Wacom PL700", 8, 6758, 5406, 511, 0, PL }, + { "Wacom PL510", 8, 6282, 4762, 511, 0, PL }, + { "Wacom DTU710", 8, 34080, 27660, 511, 0, PL }, + { "Wacom DTF521", 8, 6282, 4762, 511, 0, PL }, + { "Wacom DTF720", 8, 6858, 5506, 511, 0, PL }, + { "Wacom Cintiq Partner",8, 20480, 15360, 511, 0, PTU }, + { "Wacom Intuos2 4x5", 10, 12700, 10600, 1023, 63, INTUOS }, + { "Wacom Intuos2 6x8", 10, 20320, 16240, 1023, 63, INTUOS }, + { "Wacom Intuos2 9x12", 10, 30480, 24060, 1023, 63, INTUOS }, + { "Wacom Intuos2 12x12", 10, 30480, 31680, 1023, 63, INTUOS }, + { "Wacom Intuos2 12x18", 10, 45720, 31680, 1023, 63, INTUOS }, + { "Wacom Intuos3 4x5", 10, 25400, 20320, 1023, 63, INTUOS3S }, + { "Wacom Intuos3 6x8", 10, 40640, 30480, 1023, 63, INTUOS3 }, + { "Wacom Intuos3 9x12", 10, 60960, 45720, 1023, 63, INTUOS3 }, + { "Wacom Intuos3 12x12", 10, 60960, 60960, 1023, 63, INTUOS3L }, + { "Wacom Intuos3 12x19", 10, 97536, 60960, 1023, 63, INTUOS3L }, + { "Wacom Intuos3 6x11", 10, 54204, 31750, 1023, 63, INTUOS3 }, + { "Wacom Intuos3 4x6", 10, 31496, 19685, 1023, 15, INTUOS3S }, + { "Wacom Cintiq 21UX", 10, 87200, 65600, 1023, 63, CINTIQ }, + { "Wacom Intuos2 6x8", 10, 20320, 16240, 1023, 63, INTUOS }, { } }; @@ -618,6 +628,7 @@ static struct usb_device_id wacom_ids[] = { { USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB3) }, { USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB4) }, { USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB5) }, + { USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB7) }, { USB_DEVICE(USB_VENDOR_ID_WACOM, 0x3F) }, { USB_DEVICE(USB_VENDOR_ID_WACOM, 0x47) }, { } diff --git a/drivers/usb/input/wacom_wac.h b/drivers/usb/input/wacom_wac.h index ceae7bf..a1d9ce0 100644 --- a/drivers/usb/input/wacom_wac.h +++ b/drivers/usb/input/wacom_wac.h @@ -20,6 +20,7 @@ enum { PTU, PL, INTUOS, + INTUOS3S, INTUOS3, INTUOS3L, CINTIQ, @@ -34,7 +35,6 @@ struct wacom_features { int pressure_max; int distance_max; int type; - usb_complete_t irq; }; struct wacom_wac { -- cgit v0.10.2 From 2920349d438ec08d2b1f6761c8b78b8d13fd1dee Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn <[mailto:snakebyte@gmx.de]> Date: Tue, 17 Oct 2006 14:46:30 -0700 Subject: USB: BUG_ON conversion for wacom.c this patch converts two if () BUG(); construct to BUG_ON(); which occupies less space, uses unlikely and is safer when BUG() is disabled. Signed-off-by: Eric Sesterhenn Acked-by: "Ping Cheng" Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/wacom_sys.c b/drivers/usb/input/wacom_sys.c index 3c27f9b..31379b8 100644 --- a/drivers/usb/input/wacom_sys.c +++ b/drivers/usb/input/wacom_sys.c @@ -223,8 +223,7 @@ static int wacom_probe(struct usb_interface *intf, const struct usb_device_id *i strlcat(wacom->phys, "/input0", sizeof(wacom->phys)); wacom_wac->features = get_wacom_feature(id); - if (wacom_wac->features->pktlen > 10) - BUG(); + BUG_ON(wacom_wac->features->pktlen > 10); input_dev->name = wacom_wac->features->name; wacom->wacom_wac = wacom_wac; -- cgit v0.10.2 From 2daa48729dfafd349c2a52520734de2edb9dc805 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Wed, 4 Oct 2006 09:56:44 -0700 Subject: USB: fix use after free in wacom_sys.c the following commit added a use after free http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=3D3bea733ab21247290bd552dd6a2cd3049af9adef Found by coverity (cid #1441) Signed-off-by: Eric Sesterhenn Signed-off-by: "Ping Cheng" Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/wacom_sys.c b/drivers/usb/input/wacom_sys.c index 31379b8..3498b89 100644 --- a/drivers/usb/input/wacom_sys.c +++ b/drivers/usb/input/wacom_sys.c @@ -282,8 +282,8 @@ static void wacom_disconnect(struct usb_interface *intf) input_unregister_device(wacom->dev); usb_free_urb(wacom->irq); usb_buffer_free(interface_to_usbdev(intf), 10, wacom->wacom_wac->data, wacom->data_dma); - kfree(wacom); kfree(wacom->wacom_wac); + kfree(wacom); } } -- cgit v0.10.2 From 931b0411ac296591643662b7a790d15d6e23d57e Mon Sep 17 00:00:00 2001 From: "Luiz Fernando N. Capitulino" Date: Tue, 3 Oct 2006 10:31:36 -0300 Subject: airprime: New device ID. Adds support for the verizon wireless Broadband Access, National Access V640 ExpressCard34 Qualcomm 3G CDMA. Reported by Maciej A. __enczykowski Signed-off-by: Luiz Fernando N. Capitulino Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/airprime.c b/drivers/usb/serial/airprime.c index 2c19f19..392a512 100644 --- a/drivers/usb/serial/airprime.c +++ b/drivers/usb/serial/airprime.c @@ -24,6 +24,7 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x1199, 0x0112) }, /* Sierra Wireless Aircard 580 */ { USB_DEVICE(0x1199, 0x0218) }, /* Sierra Wireless MC5720 */ { USB_DEVICE(0x1410, 0x1110) }, /* Novatel Wireless Merlin CDMA */ + { USB_DEVICE(0x1410, 0x1100) }, /* ExpressCard34 Qualcomm 3G CDMA */ { }, }; MODULE_DEVICE_TABLE(usb, id_table); -- cgit v0.10.2 From 91a9c9214e34c364bf15406aadb922787ae7129b Mon Sep 17 00:00:00 2001 From: Chris Malley Date: Tue, 3 Oct 2006 10:08:28 +0100 Subject: USB: Support for BT On-Air USB modem in cdc-acm.c The patch below is a necessary workaround to support the BT On-Air USB modem, which fails to initialise properly during normal probing thus: Sep 30 17:34:57 sled kernel: drivers/usb/class/cdc-acm.c: Zero length descriptor references Sep 30 17:34:57 sled kernel: cdc_acm: probe of 1-1.2:1.0 failed with error -22 Adding the patch below causes the probing section to be skipped, and the modem then initialises correctly. Signed-off-by: Chris Malley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index ec4d1d7..daecdf0 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1083,6 +1083,9 @@ static struct usb_device_id acm_ids[] = { { USB_DEVICE(0x0482, 0x0203), /* KYOCERA AH-K3001V */ .driver_info = NO_UNION_NORMAL, /* has no union descriptor */ }, + { USB_DEVICE(0x079b, 0x000f), /* BT On-Air USB MODEM */ + .driver_info = NO_UNION_NORMAL, /* has no union descriptor */ + }, { USB_DEVICE(0x0ace, 0x1608), /* ZyDAS 56K USB MODEM */ .driver_info = SINGLE_RX_URB, /* firmware bug */ }, -- cgit v0.10.2 From fbe2bafcb00b25265c2c869ba4615d6a5324b7f1 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Thu, 28 Sep 2006 23:36:04 +0200 Subject: USB: remove private debug macros from kaweth this kills the private debug macros from the kaweth driver. Signed-off-by: Oliver Neukum Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/kaweth.c b/drivers/usb/net/kaweth.c index 957d4ad..9ef9075 100644 --- a/drivers/usb/net/kaweth.c +++ b/drivers/usb/net/kaweth.c @@ -65,16 +65,6 @@ #undef DEBUG -#ifdef DEBUG -#define kaweth_dbg(format, arg...) printk(KERN_DEBUG __FILE__ ": " format "\n" ,##arg) -#else -#define kaweth_dbg(format, arg...) do {} while (0) -#endif -#define kaweth_err(format, arg...) printk(KERN_ERR __FILE__ ": " format "\n" ,##arg) -#define kaweth_info(format, arg...) printk(KERN_INFO __FILE__ ": " format "\n" , ##arg) -#define kaweth_warn(format, arg...) printk(KERN_WARNING __FILE__ ": " format "\n" , ##arg) - - #include "kawethfw.h" #define KAWETH_MTU 1514 @@ -265,17 +255,17 @@ static int kaweth_control(struct kaweth_device *kaweth, { struct usb_ctrlrequest *dr; - kaweth_dbg("kaweth_control()"); + dbg("kaweth_control()"); if(in_interrupt()) { - kaweth_dbg("in_interrupt()"); + dbg("in_interrupt()"); return -EBUSY; } dr = kmalloc(sizeof(struct usb_ctrlrequest), GFP_ATOMIC); if (!dr) { - kaweth_dbg("kmalloc() failed"); + dbg("kmalloc() failed"); return -ENOMEM; } @@ -300,7 +290,7 @@ static int kaweth_read_configuration(struct kaweth_device *kaweth) { int retval; - kaweth_dbg("Reading kaweth configuration"); + dbg("Reading kaweth configuration"); retval = kaweth_control(kaweth, usb_rcvctrlpipe(kaweth->dev, 0), @@ -322,7 +312,7 @@ static int kaweth_set_urb_size(struct kaweth_device *kaweth, __u16 urb_size) { int retval; - kaweth_dbg("Setting URB size to %d", (unsigned)urb_size); + dbg("Setting URB size to %d", (unsigned)urb_size); retval = kaweth_control(kaweth, usb_sndctrlpipe(kaweth->dev, 0), @@ -344,7 +334,7 @@ static int kaweth_set_sofs_wait(struct kaweth_device *kaweth, __u16 sofs_wait) { int retval; - kaweth_dbg("Set SOFS wait to %d", (unsigned)sofs_wait); + dbg("Set SOFS wait to %d", (unsigned)sofs_wait); retval = kaweth_control(kaweth, usb_sndctrlpipe(kaweth->dev, 0), @@ -367,7 +357,7 @@ static int kaweth_set_receive_filter(struct kaweth_device *kaweth, { int retval; - kaweth_dbg("Set receive filter to %d", (unsigned)receive_filter); + dbg("Set receive filter to %d", (unsigned)receive_filter); retval = kaweth_control(kaweth, usb_sndctrlpipe(kaweth->dev, 0), @@ -392,7 +382,7 @@ static int kaweth_download_firmware(struct kaweth_device *kaweth, __u8 type) { if(data_len > KAWETH_FIRMWARE_BUF_SIZE) { - kaweth_err("Firmware too big: %d", data_len); + err("Firmware too big: %d", data_len); return -ENOSPC; } @@ -403,13 +393,13 @@ static int kaweth_download_firmware(struct kaweth_device *kaweth, kaweth->firmware_buf[4] = type; kaweth->firmware_buf[5] = interrupt; - kaweth_dbg("High: %i, Low:%i", kaweth->firmware_buf[3], + dbg("High: %i, Low:%i", kaweth->firmware_buf[3], kaweth->firmware_buf[2]); - kaweth_dbg("Downloading firmware at %p to kaweth device at %p", + dbg("Downloading firmware at %p to kaweth device at %p", data, kaweth); - kaweth_dbg("Firmware length: %d", data_len); + dbg("Firmware length: %d", data_len); return kaweth_control(kaweth, usb_sndctrlpipe(kaweth->dev, 0), @@ -437,7 +427,7 @@ static int kaweth_trigger_firmware(struct kaweth_device *kaweth, kaweth->firmware_buf[6] = 0x00; kaweth->firmware_buf[7] = 0x00; - kaweth_dbg("Triggering firmware"); + dbg("Triggering firmware"); return kaweth_control(kaweth, usb_sndctrlpipe(kaweth->dev, 0), @@ -457,7 +447,7 @@ static int kaweth_reset(struct kaweth_device *kaweth) { int result; - kaweth_dbg("kaweth_reset(%p)", kaweth); + dbg("kaweth_reset(%p)", kaweth); result = kaweth_control(kaweth, usb_sndctrlpipe(kaweth->dev, 0), USB_REQ_SET_CONFIGURATION, @@ -470,7 +460,7 @@ static int kaweth_reset(struct kaweth_device *kaweth) mdelay(10); - kaweth_dbg("kaweth_reset() returns %d.",result); + dbg("kaweth_reset() returns %d.",result); return result; } @@ -568,7 +558,7 @@ static int kaweth_resubmit_rx_urb(struct kaweth_device *kaweth, kaweth->suspend_lowmem_rx = 1; schedule_delayed_work(&kaweth->lowmem_work, HZ/4); } - kaweth_err("resubmitting rx_urb %d failed", result); + err("resubmitting rx_urb %d failed", result); } else { kaweth->suspend_lowmem_rx = 0; } @@ -605,7 +595,7 @@ static void kaweth_usb_receive(struct urb *urb) return; if(urb->status && urb->status != -EREMOTEIO && count != 1) { - kaweth_err("%s RX status: %d count: %d packet_len: %d", + err("%s RX status: %d count: %d packet_len: %d", net->name, urb->status, count, @@ -616,9 +606,9 @@ static void kaweth_usb_receive(struct urb *urb) if(kaweth->net && (count > 2)) { if(pkt_len > (count - 2)) { - kaweth_err("Packet length too long for USB frame (pkt_len: %x, count: %x)",pkt_len, count); - kaweth_err("Packet len & 2047: %x", pkt_len & 2047); - kaweth_err("Count 2: %x", count2); + err("Packet length too long for USB frame (pkt_len: %x, count: %x)",pkt_len, count); + err("Packet len & 2047: %x", pkt_len & 2047); + err("Count 2: %x", count2); kaweth_resubmit_rx_urb(kaweth, GFP_ATOMIC); return; } @@ -655,7 +645,7 @@ static int kaweth_open(struct net_device *net) struct kaweth_device *kaweth = netdev_priv(net); int res; - kaweth_dbg("Opening network device."); + dbg("Opening network device."); res = kaweth_resubmit_rx_urb(kaweth, GFP_KERNEL); if (res) @@ -732,7 +722,7 @@ static void kaweth_usb_transmit_complete(struct urb *urb) if (unlikely(urb->status != 0)) if (urb->status != -ENOENT) - kaweth_dbg("%s: TX status %d.", kaweth->net->name, urb->status); + dbg("%s: TX status %d.", kaweth->net->name, urb->status); netif_wake_queue(kaweth->net); dev_kfree_skb_irq(skb); @@ -783,7 +773,7 @@ static int kaweth_start_xmit(struct sk_buff *skb, struct net_device *net) if((res = usb_submit_urb(kaweth->tx_urb, GFP_ATOMIC))) { - kaweth_warn("kaweth failed tx_urb %d", res); + warn("kaweth failed tx_urb %d", res); kaweth->stats.tx_errors++; netif_start_queue(net); @@ -812,7 +802,7 @@ static void kaweth_set_rx_mode(struct net_device *net) KAWETH_PACKET_FILTER_BROADCAST | KAWETH_PACKET_FILTER_MULTICAST; - kaweth_dbg("Setting Rx mode to %d", packet_filter_bitmap); + dbg("Setting Rx mode to %d", packet_filter_bitmap); netif_stop_queue(net); @@ -850,10 +840,10 @@ static void kaweth_async_set_rx_mode(struct kaweth_device *kaweth) KAWETH_CONTROL_TIMEOUT); if(result < 0) { - kaweth_err("Failed to set Rx mode: %d", result); + err("Failed to set Rx mode: %d", result); } else { - kaweth_dbg("Set Rx mode to %d", packet_filter_bitmap); + dbg("Set Rx mode to %d", packet_filter_bitmap); } } } @@ -874,7 +864,7 @@ static void kaweth_tx_timeout(struct net_device *net) { struct kaweth_device *kaweth = netdev_priv(net); - kaweth_warn("%s: Tx timed out. Resetting.", net->name); + warn("%s: Tx timed out. Resetting.", net->name); kaweth->stats.tx_errors++; net->trans_start = jiffies; @@ -895,15 +885,15 @@ static int kaweth_probe( const eth_addr_t bcast_addr = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF }; int result = 0; - kaweth_dbg("Kawasaki Device Probe (Device number:%d): 0x%4.4x:0x%4.4x:0x%4.4x", + dbg("Kawasaki Device Probe (Device number:%d): 0x%4.4x:0x%4.4x:0x%4.4x", dev->devnum, le16_to_cpu(dev->descriptor.idVendor), le16_to_cpu(dev->descriptor.idProduct), le16_to_cpu(dev->descriptor.bcdDevice)); - kaweth_dbg("Device at %p", dev); + dbg("Device at %p", dev); - kaweth_dbg("Descriptor length: %x type: %x", + dbg("Descriptor length: %x type: %x", (int)dev->descriptor.bLength, (int)dev->descriptor.bDescriptorType); @@ -918,7 +908,7 @@ static int kaweth_probe( spin_lock_init(&kaweth->device_lock); init_waitqueue_head(&kaweth->term_wait); - kaweth_dbg("Resetting."); + dbg("Resetting."); kaweth_reset(kaweth); @@ -928,17 +918,17 @@ static int kaweth_probe( */ if (le16_to_cpu(dev->descriptor.bcdDevice) >> 8) { - kaweth_info("Firmware present in device."); + info("Firmware present in device."); } else { /* Download the firmware */ - kaweth_info("Downloading firmware..."); + info("Downloading firmware..."); kaweth->firmware_buf = (__u8 *)__get_free_page(GFP_KERNEL); if ((result = kaweth_download_firmware(kaweth, kaweth_new_code, len_kaweth_new_code, 100, 2)) < 0) { - kaweth_err("Error downloading firmware (%d)", result); + err("Error downloading firmware (%d)", result); goto err_fw; } @@ -947,7 +937,7 @@ static int kaweth_probe( len_kaweth_new_code_fix, 100, 3)) < 0) { - kaweth_err("Error downloading firmware fix (%d)", result); + err("Error downloading firmware fix (%d)", result); goto err_fw; } @@ -956,7 +946,7 @@ static int kaweth_probe( len_kaweth_trigger_code, 126, 2)) < 0) { - kaweth_err("Error downloading trigger code (%d)", result); + err("Error downloading trigger code (%d)", result); goto err_fw; } @@ -966,18 +956,18 @@ static int kaweth_probe( len_kaweth_trigger_code_fix, 126, 3)) < 0) { - kaweth_err("Error downloading trigger code fix (%d)", result); + err("Error downloading trigger code fix (%d)", result); goto err_fw; } if ((result = kaweth_trigger_firmware(kaweth, 126)) < 0) { - kaweth_err("Error triggering firmware (%d)", result); + err("Error triggering firmware (%d)", result); goto err_fw; } /* Device will now disappear for a moment... */ - kaweth_info("Firmware loaded. I'll be back..."); + info("Firmware loaded. I'll be back..."); err_fw: free_page((unsigned long)kaweth->firmware_buf); free_netdev(netdev); @@ -987,14 +977,14 @@ err_fw: result = kaweth_read_configuration(kaweth); if(result < 0) { - kaweth_err("Error reading configuration (%d), no net device created", result); + err("Error reading configuration (%d), no net device created", result); goto err_free_netdev; } - kaweth_info("Statistics collection: %x", kaweth->configuration.statistics_mask); - kaweth_info("Multicast filter limit: %x", kaweth->configuration.max_multicast_filters & ((1 << 15) - 1)); - kaweth_info("MTU: %d", le16_to_cpu(kaweth->configuration.segment_size)); - kaweth_info("Read MAC address %2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x", + info("Statistics collection: %x", kaweth->configuration.statistics_mask); + info("Multicast filter limit: %x", kaweth->configuration.max_multicast_filters & ((1 << 15) - 1)); + info("MTU: %d", le16_to_cpu(kaweth->configuration.segment_size)); + info("Read MAC address %2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x", (int)kaweth->configuration.hw_addr[0], (int)kaweth->configuration.hw_addr[1], (int)kaweth->configuration.hw_addr[2], @@ -1005,17 +995,17 @@ err_fw: if(!memcmp(&kaweth->configuration.hw_addr, &bcast_addr, sizeof(bcast_addr))) { - kaweth_err("Firmware not functioning properly, no net device created"); + err("Firmware not functioning properly, no net device created"); goto err_free_netdev; } if(kaweth_set_urb_size(kaweth, KAWETH_BUF_SIZE) < 0) { - kaweth_dbg("Error setting URB size"); + dbg("Error setting URB size"); goto err_free_netdev; } if(kaweth_set_sofs_wait(kaweth, KAWETH_SOFS_TO_WAIT) < 0) { - kaweth_err("Error setting SOFS wait"); + err("Error setting SOFS wait"); goto err_free_netdev; } @@ -1025,11 +1015,11 @@ err_fw: KAWETH_PACKET_FILTER_MULTICAST); if(result < 0) { - kaweth_err("Error setting receive filter"); + err("Error setting receive filter"); goto err_free_netdev; } - kaweth_dbg("Initializing net device."); + dbg("Initializing net device."); kaweth->tx_urb = usb_alloc_urb(0, GFP_KERNEL); if (!kaweth->tx_urb) @@ -1086,13 +1076,13 @@ err_fw: SET_NETDEV_DEV(netdev, &intf->dev); if (register_netdev(netdev) != 0) { - kaweth_err("Error registering netdev."); + err("Error registering netdev."); goto err_intfdata; } - kaweth_info("kaweth interface created at %s", kaweth->net->name); + info("kaweth interface created at %s", kaweth->net->name); - kaweth_dbg("Kaweth probe returning."); + dbg("Kaweth probe returning."); return 0; @@ -1121,16 +1111,16 @@ static void kaweth_disconnect(struct usb_interface *intf) struct kaweth_device *kaweth = usb_get_intfdata(intf); struct net_device *netdev; - kaweth_info("Unregistering"); + info("Unregistering"); usb_set_intfdata(intf, NULL); if (!kaweth) { - kaweth_warn("unregistering non-existant device"); + warn("unregistering non-existant device"); return; } netdev = kaweth->net; - kaweth_dbg("Unregistering net device"); + dbg("Unregistering net device"); unregister_netdev(netdev); usb_free_urb(kaweth->rx_urb); @@ -1185,7 +1175,7 @@ static int usb_start_wait_urb(struct urb *urb, int timeout, int* actual_length) if (!wait_event_timeout(awd.wqh, awd.done, timeout)) { // timeout - kaweth_warn("usb_control/bulk_msg: timeout"); + warn("usb_control/bulk_msg: timeout"); usb_kill_urb(urb); // remove urb safely status = -ETIMEDOUT; } @@ -1234,7 +1224,7 @@ static int kaweth_internal_control_msg(struct usb_device *usb_dev, ****************************************************************/ static int __init kaweth_init(void) { - kaweth_dbg("Driver loading"); + dbg("Driver loading"); return usb_register(&kaweth_driver); } -- cgit v0.10.2 From 1a2ea1dfc4ee078841cd6406ebf6bf0c5a3d25e9 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Tue, 3 Oct 2006 10:30:52 +0200 Subject: USB: suspend/resume support for kaweth this adds support for suspend and resume to the kaweth driver. Signed-off-by: Oliver Neukum Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/kaweth.c b/drivers/usb/net/kaweth.c index 9ef9075..7c906a4 100644 --- a/drivers/usb/net/kaweth.c +++ b/drivers/usb/net/kaweth.c @@ -76,6 +76,9 @@ #define KAWETH_STATUS_BROKEN 0x0000001 #define KAWETH_STATUS_CLOSING 0x0000002 +#define KAWETH_STATUS_SUSPENDING 0x0000004 + +#define KAWETH_STATUS_BLOCKED (KAWETH_STATUS_CLOSING | KAWETH_STATUS_SUSPENDING) #define KAWETH_PACKET_FILTER_PROMISCUOUS 0x01 #define KAWETH_PACKET_FILTER_ALL_MULTICAST 0x02 @@ -102,6 +105,8 @@ #define STATE_MASK 0x40 #define STATE_SHIFT 5 +#define IS_BLOCKED(s) (s & KAWETH_STATUS_BLOCKED) + MODULE_AUTHOR("Michael Zappe , Stephane Alnet , Brad Hards and Oliver Neukum "); MODULE_DESCRIPTION("KL5USB101 USB Ethernet driver"); @@ -118,6 +123,8 @@ static int kaweth_internal_control_msg(struct usb_device *usb_dev, unsigned int pipe, struct usb_ctrlrequest *cmd, void *data, int len, int timeout); +static int kaweth_suspend(struct usb_interface *intf, pm_message_t message); +static int kaweth_resume(struct usb_interface *intf); /**************************************************************** * usb_device_id @@ -169,6 +176,8 @@ static struct usb_driver kaweth_driver = { .name = driver_name, .probe = kaweth_probe, .disconnect = kaweth_disconnect, + .suspend = kaweth_suspend, + .resume = kaweth_resume, .id_table = usb_klsi_table, }; @@ -212,6 +221,7 @@ struct kaweth_device int suspend_lowmem_rx; int suspend_lowmem_ctrl; int linkstate; + int opened; struct work_struct lowmem_work; struct usb_device *dev; @@ -524,7 +534,7 @@ static void kaweth_resubmit_tl(void *d) { struct kaweth_device *kaweth = (struct kaweth_device *)d; - if (kaweth->status | KAWETH_STATUS_CLOSING) + if (IS_BLOCKED(kaweth->status)) return; if (kaweth->suspend_lowmem_rx) @@ -591,8 +601,12 @@ static void kaweth_usb_receive(struct urb *urb) return; } - if (kaweth->status & KAWETH_STATUS_CLOSING) + spin_lock(&kaweth->device_lock); + if (IS_BLOCKED(kaweth->status)) { + spin_unlock(&kaweth->device_lock); return; + } + spin_unlock(&kaweth->device_lock); if(urb->status && urb->status != -EREMOTEIO && count != 1) { err("%s RX status: %d count: %d packet_len: %d", @@ -668,6 +682,7 @@ static int kaweth_open(struct net_device *net) usb_kill_urb(kaweth->rx_urb); return -EIO; } + kaweth->opened = 1; netif_start_queue(net); @@ -678,14 +693,8 @@ static int kaweth_open(struct net_device *net) /**************************************************************** * kaweth_close ****************************************************************/ -static int kaweth_close(struct net_device *net) +static void kaweth_kill_urbs(struct kaweth_device *kaweth) { - struct kaweth_device *kaweth = netdev_priv(net); - - netif_stop_queue(net); - - kaweth->status |= KAWETH_STATUS_CLOSING; - usb_kill_urb(kaweth->irq_urb); usb_kill_urb(kaweth->rx_urb); usb_kill_urb(kaweth->tx_urb); @@ -696,6 +705,21 @@ static int kaweth_close(struct net_device *net) we hit them again */ usb_kill_urb(kaweth->irq_urb); usb_kill_urb(kaweth->rx_urb); +} + +/**************************************************************** + * kaweth_close + ****************************************************************/ +static int kaweth_close(struct net_device *net) +{ + struct kaweth_device *kaweth = netdev_priv(net); + + netif_stop_queue(net); + kaweth->opened = 0; + + kaweth->status |= KAWETH_STATUS_CLOSING; + + kaweth_kill_urbs(kaweth); kaweth->status &= ~KAWETH_STATUS_CLOSING; @@ -742,6 +766,9 @@ static int kaweth_start_xmit(struct sk_buff *skb, struct net_device *net) kaweth_async_set_rx_mode(kaweth); netif_stop_queue(net); + if (IS_BLOCKED(kaweth->status)) { + goto skip; + } /* We now decide whether we can put our special header into the sk_buff */ if (skb_cloned(skb) || skb_headroom(skb) < 2) { @@ -774,6 +801,7 @@ static int kaweth_start_xmit(struct sk_buff *skb, struct net_device *net) if((res = usb_submit_urb(kaweth->tx_urb, GFP_ATOMIC))) { warn("kaweth failed tx_urb %d", res); +skip: kaweth->stats.tx_errors++; netif_start_queue(net); @@ -872,6 +900,42 @@ static void kaweth_tx_timeout(struct net_device *net) } /**************************************************************** + * kaweth_suspend + ****************************************************************/ +static int kaweth_suspend(struct usb_interface *intf, pm_message_t message) +{ + struct kaweth_device *kaweth = usb_get_intfdata(intf); + unsigned long flags; + + spin_lock_irqsave(&kaweth->device_lock, flags); + kaweth->status |= KAWETH_STATUS_SUSPENDING; + spin_unlock_irqrestore(&kaweth->device_lock, flags); + + kaweth_kill_urbs(kaweth); + return 0; +} + +/**************************************************************** + * kaweth_resume + ****************************************************************/ +static int kaweth_resume(struct usb_interface *intf) +{ + struct kaweth_device *kaweth = usb_get_intfdata(intf); + unsigned long flags; + + spin_lock_irqsave(&kaweth->device_lock, flags); + kaweth->status &= ~KAWETH_STATUS_SUSPENDING; + spin_unlock_irqrestore(&kaweth->device_lock, flags); + + if (!kaweth->opened) + return 0; + kaweth_resubmit_rx_urb(kaweth, GFP_NOIO); + kaweth_resubmit_int_urb(kaweth, GFP_NOIO); + + return 0; +} + +/**************************************************************** * kaweth_probe ****************************************************************/ static int kaweth_probe( -- cgit v0.10.2 From 8442ae00d47dad690ac1105b426274433dc672f8 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Mon, 2 Oct 2006 07:20:10 -0700 Subject: USB: ohci-pnx4008 build fixes The OHCI bus glue for the Philips PNX chips is missing a few calls. - Bus suspend/resume were wrongly omitted in the original submission. - Two new calls were added since that glue was submitted: * Root hub irq enable call * Shutdown hook for usbcore Plus usb_bus.hcpriv has now been removed from usbcore. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ohci-pnx4008.c b/drivers/usb/host/ohci-pnx4008.c index 82cb22f..2dbb774 100644 --- a/drivers/usb/host/ohci-pnx4008.c +++ b/drivers/usb/host/ohci-pnx4008.c @@ -262,6 +262,7 @@ static const struct hc_driver ohci_pnx4008_hc_driver = { */ .start = ohci_pnx4008_start, .stop = ohci_stop, + .shutdown = ohci_shutdown, /* * managing i/o requests and associated device resources @@ -280,7 +281,11 @@ static const struct hc_driver ohci_pnx4008_hc_driver = { */ .hub_status_data = ohci_hub_status_data, .hub_control = ohci_hub_control, - + .hub_irq_enable = ohci_rhsc_enable, +#ifdef CONFIG_PM + .bus_suspend = ohci_bus_suspend, + .bus_resume = ohci_bus_resume, +#endif .start_port_reset = ohci_start_port_reset, }; @@ -410,8 +415,6 @@ static int __devinit usb_hcd_pnx4008_probe(struct platform_device *pdev) goto out4; } - hcd->self.hcpriv = (void *)hcd; - pnx4008_start_hc(); platform_set_drvdata(pdev, hcd); ohci = hcd_to_ohci(hcd); -- cgit v0.10.2 From 9ca5346483ea2c2e8932268246d1d8746fe3bcaa Mon Sep 17 00:00:00 2001 From: matthieu castet Date: Tue, 3 Oct 2006 21:46:33 +0200 Subject: UEAGLE : be suspend friendly this patch avoid that the kernel thread block the suspend process. Some work is still need to recover after a resume. Signed-off-by: Matthieu Castet Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index f5434b1..68b5d0d 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -1173,6 +1173,9 @@ static int uea_kthread(void *data) ret = uea_stat(sc); if (ret != -EAGAIN) msleep(1000); + if (try_to_freeze()) + uea_err(INS_TO_USBDEV(sc), "suspend/resume not supported, " + "please unplug/replug your modem\n"); } uea_leaves(INS_TO_USBDEV(sc)); return ret; -- cgit v0.10.2 From 531a39bbab213209a9914e68809bcf8b60a54f47 Mon Sep 17 00:00:00 2001 From: matthieu castet Date: Tue, 3 Oct 2006 21:49:29 +0200 Subject: UEAGLE : use interruptible sleep this patch use wait_event_interruptible_timeout and msleep_interruptible beacause uninterruptible sleep (task state 'D') is counted as 1 towards load average, like running processes. Signed-off-by: Matthieu Castet Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index 68b5d0d..88585da 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -744,7 +744,7 @@ static inline void wake_up_cmv_ack(struct uea_softc *sc) static inline int wait_cmv_ack(struct uea_softc *sc) { - int ret = wait_event_timeout(sc->cmv_ack_wait, + int ret = wait_event_interruptible_timeout(sc->cmv_ack_wait, sc->cmv_ack, ACK_TIMEOUT); sc->cmv_ack = 0; @@ -1172,7 +1172,7 @@ static int uea_kthread(void *data) if (!ret) ret = uea_stat(sc); if (ret != -EAGAIN) - msleep(1000); + msleep_interruptible(1000); if (try_to_freeze()) uea_err(INS_TO_USBDEV(sc), "suspend/resume not supported, " "please unplug/replug your modem\n"); @@ -1600,7 +1600,7 @@ static int uea_heavy(struct usbatm_data *usbatm, struct usb_interface *intf) { struct uea_softc *sc = usbatm->driver_data; - wait_event(sc->sync_q, IS_OPERATIONAL(sc)); + wait_event_interruptible(sc->sync_q, IS_OPERATIONAL(sc)); return 0; -- cgit v0.10.2 From a7a0c9cd1f45c2cae38ebe0951246bf94399818a Mon Sep 17 00:00:00 2001 From: matthieu castet Date: Tue, 3 Oct 2006 21:44:11 +0200 Subject: UEAGLE : comestic changes Hi, this patch does some cosmetic changes : - dump firwmare version as soon as possible and export it on sysfs - hint about wrong cmv/dsp - Display a message to warn user when the modem is ready : it can help people to detect problems on the line without debug trace - Fix wrong indent - display modem type (pots/isdn) - increase version number Signed-off-by: Matthieu Castet Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index 88585da..5705298 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -68,7 +68,7 @@ #include "usbatm.h" -#define EAGLEUSBVERSION "ueagle 1.3" +#define EAGLEUSBVERSION "ueagle 1.4" /* @@ -80,14 +80,14 @@ dev_dbg(&(usb_dev)->dev, \ "[ueagle-atm dbg] %s: " format, \ __FUNCTION__, ##args); \ - } while (0) + } while (0) #define uea_vdbg(usb_dev, format, args...) \ do { \ if (debug >= 2) \ dev_dbg(&(usb_dev)->dev, \ "[ueagle-atm vdbg] " format, ##args); \ - } while (0) + } while (0) #define uea_enters(usb_dev) \ uea_vdbg(usb_dev, "entering %s\n", __FUNCTION__) @@ -218,8 +218,8 @@ enum { #define UEA_CHIP_VERSION(x) \ ((x)->driver_info & 0xf) -#define IS_ISDN(sc) \ - (le16_to_cpu(sc->usb_dev->descriptor.bcdDevice) & 0x80) +#define IS_ISDN(usb_dev) \ + (le16_to_cpu((usb_dev)->descriptor.bcdDevice) & 0x80) #define INS_TO_USBDEV(ins) ins->usb_dev @@ -625,12 +625,12 @@ static int request_dsp(struct uea_softc *sc) char *dsp_name; if (UEA_CHIP_VERSION(sc) == ADI930) { - if (IS_ISDN(sc)) + if (IS_ISDN(sc->usb_dev)) dsp_name = FW_DIR "DSP9i.bin"; else dsp_name = FW_DIR "DSP9p.bin"; } else { - if (IS_ISDN(sc)) + if (IS_ISDN(sc->usb_dev)) dsp_name = FW_DIR "DSPei.bin"; else dsp_name = FW_DIR "DSPep.bin"; @@ -885,7 +885,8 @@ static int uea_stat(struct uea_softc *sc) break; case 3: /* fail ... */ - uea_info(INS_TO_USBDEV(sc), "modem synchronization failed\n"); + uea_info(INS_TO_USBDEV(sc), "modem synchronization failed" + " (may be try other cmv/dsp)\n"); return -EAGAIN; case 4 ... 6: /* test state */ @@ -913,12 +914,6 @@ static int uea_stat(struct uea_softc *sc) release_firmware(sc->dsp_firm); sc->dsp_firm = NULL; } - - ret = uea_read_cmv(sc, SA_INFO, 10, &sc->stats.phy.firmid); - if (ret < 0) - return ret; - uea_info(INS_TO_USBDEV(sc), "ATU-R firmware version : %x\n", - sc->stats.phy.firmid); } /* always update it as atm layer could not be init when we switch to @@ -1033,9 +1028,9 @@ static int request_cmvs(struct uea_softc *sc, if (cmv_file[sc->modem_index] == NULL) { if (UEA_CHIP_VERSION(sc) == ADI930) - file = (IS_ISDN(sc)) ? "CMV9i.bin" : "CMV9p.bin"; + file = (IS_ISDN(sc->usb_dev)) ? "CMV9i.bin" : "CMV9p.bin"; else - file = (IS_ISDN(sc)) ? "CMVei.bin" : "CMVep.bin"; + file = (IS_ISDN(sc->usb_dev)) ? "CMVei.bin" : "CMVep.bin"; } else file = cmv_file[sc->modem_index]; @@ -1131,6 +1126,13 @@ static int uea_start_reset(struct uea_softc *sc) if (ret < 0) return ret; + /* Dump firmware version */ + ret = uea_read_cmv(sc, SA_INFO, 10, &sc->stats.phy.firmid); + if (ret < 0) + return ret; + uea_info(INS_TO_USBDEV(sc), "ATU-R firmware version : %x\n", + sc->stats.phy.firmid); + /* get options */ ret = len = request_cmvs(sc, &cmvs, &cmvs_fw); if (ret < 0) @@ -1147,6 +1149,8 @@ static int uea_start_reset(struct uea_softc *sc) /* Enter in R-ACT-REQ */ ret = uea_write_cmv(sc, SA_CNTL, 0, 2); uea_vdbg(INS_TO_USBDEV(sc), "Entering in R-ACT-REQ state\n"); + uea_info(INS_TO_USBDEV(sc), "Modem started, " + "waiting synchronization\n"); out: release_firmware(cmvs_fw); sc->reset = 0; @@ -1569,6 +1573,7 @@ UEA_ATTR(uscorr, 0); UEA_ATTR(dscorr, 0); UEA_ATTR(usunc, 0); UEA_ATTR(dsunc, 0); +UEA_ATTR(firmid, 0); /* Retrieve the device End System Identifier (MAC) */ @@ -1642,6 +1647,7 @@ static struct attribute *attrs[] = { &dev_attr_stat_dscorr.attr, &dev_attr_stat_usunc.attr, &dev_attr_stat_dsunc.attr, + &dev_attr_stat_firmid.attr, }; static struct attribute_group attr_grp = { .attrs = attrs, @@ -1756,10 +1762,10 @@ static int uea_probe(struct usb_interface *intf, const struct usb_device_id *id) struct usb_device *usb = interface_to_usbdev(intf); uea_enters(usb); - uea_info(usb, "ADSL device founded vid (%#X) pid (%#X) : %s\n", + uea_info(usb, "ADSL device founded vid (%#X) pid (%#X) : %s %s\n", le16_to_cpu(usb->descriptor.idVendor), le16_to_cpu(usb->descriptor.idProduct), - chip_name[UEA_CHIP_VERSION(id)]); + chip_name[UEA_CHIP_VERSION(id)], IS_ISDN(usb)?"isdn":"pots"); usb_reset_device(usb); -- cgit v0.10.2 From 762f007b05446f5c63268fb2c28646f28959ee4b Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Fri, 6 Oct 2006 07:23:11 +0200 Subject: USB: fix cdc-acm problems with hard irq? (inconsistent lock state) Signed-off-by: Jarek Poplawski Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index daecdf0..9a9012f 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -325,7 +325,7 @@ static void acm_rx_tasklet(unsigned long _acm) struct acm_rb *buf; struct tty_struct *tty = acm->tty; struct acm_ru *rcv; - //unsigned long flags; + unsigned long flags; int i = 0; dbg("Entering acm_rx_tasklet"); @@ -333,15 +333,15 @@ static void acm_rx_tasklet(unsigned long _acm) return; next_buffer: - spin_lock(&acm->read_lock); + spin_lock_irqsave(&acm->read_lock, flags); if (list_empty(&acm->filled_read_bufs)) { - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); goto urbs; } buf = list_entry(acm->filled_read_bufs.next, struct acm_rb, list); list_del(&buf->list); - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); dbg("acm_rx_tasklet: procesing buf 0x%p, size = %d", buf, buf->size); @@ -356,29 +356,29 @@ next_buffer: memmove(buf->base, buf->base + i, buf->size - i); buf->size -= i; spin_unlock(&acm->throttle_lock); - spin_lock(&acm->read_lock); + spin_lock_irqsave(&acm->read_lock, flags); list_add(&buf->list, &acm->filled_read_bufs); - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); return; } spin_unlock(&acm->throttle_lock); - spin_lock(&acm->read_lock); + spin_lock_irqsave(&acm->read_lock, flags); list_add(&buf->list, &acm->spare_read_bufs); - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); goto next_buffer; urbs: while (!list_empty(&acm->spare_read_bufs)) { - spin_lock(&acm->read_lock); + spin_lock_irqsave(&acm->read_lock, flags); if (list_empty(&acm->spare_read_urbs)) { - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); return; } rcv = list_entry(acm->spare_read_urbs.next, struct acm_ru, list); list_del(&rcv->list); - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); buf = list_entry(acm->spare_read_bufs.next, struct acm_rb, list); @@ -400,9 +400,9 @@ urbs: free-urbs-pool and resubmited ASAP */ if (usb_submit_urb(rcv->urb, GFP_ATOMIC) < 0) { list_add(&buf->list, &acm->spare_read_bufs); - spin_lock(&acm->read_lock); + spin_lock_irqsave(&acm->read_lock, flags); list_add(&rcv->list, &acm->spare_read_urbs); - spin_unlock(&acm->read_lock); + spin_unlock_irqrestore(&acm->read_lock, flags); return; } } -- cgit v0.10.2 From e4a20daa7b44ab9805979eb716f6bb7532bc67b9 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 4 Oct 2006 16:31:15 -0400 Subject: USB: unusual_devs entry for Nokia 6131 This patch (as796) adds an unusual_devs entry for the Nokia 6131, which doesn't like large transfer sizes. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index c9a8d50..0a846e4 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -182,6 +182,13 @@ UNUSUAL_DEV( 0x0421, 0x044e, 0x0100, 0x0100, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_IGNORE_RESIDUE | US_FL_FIX_CAPACITY ), +/* Reported by Bardur Arantsson */ +UNUSUAL_DEV( 0x0421, 0x047c, 0x0370, 0x0370, + "Nokia", + "6131", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_MAX_SECTORS_64 ), + /* Reported by Olaf Hering from novell bug #105878 */ UNUSUAL_DEV( 0x0424, 0x0fdc, 0x0210, 0x0210, "SMSC", -- cgit v0.10.2 From ccf40d62c70128990cf2e8775853cc3287cd7ce3 Mon Sep 17 00:00:00 2001 From: Duncan Sands Date: Thu, 5 Oct 2006 09:56:44 +0200 Subject: usbatm: fix tiny race If usbatm_do_heavy_init finishes before usbatm_heavy_init writes the pid, the disconnect method could shoot down the wrong process if the pid has been recycled. Signed-off-by: Duncan Sands Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/usbatm.c b/drivers/usb/atm/usbatm.c index 309073f..ab091fa 100644 --- a/drivers/usb/atm/usbatm.c +++ b/drivers/usb/atm/usbatm.c @@ -1001,6 +1001,7 @@ static int usbatm_do_heavy_init(void *arg) daemonize(instance->driver->driver_name); allow_signal(SIGTERM); + instance->thread_pid = get_current()->pid; complete(&instance->thread_started); @@ -1025,10 +1026,6 @@ static int usbatm_heavy_init(struct usbatm_data *instance) return ret; } - mutex_lock(&instance->serialize); - instance->thread_pid = ret; - mutex_unlock(&instance->serialize); - wait_for_completion(&instance->thread_started); return 0; -- cgit v0.10.2 From 6a4f1b41357d2bd65d39f7a5d44e92f69daaf04b Mon Sep 17 00:00:00 2001 From: Duncan Sands Date: Thu, 5 Oct 2006 10:40:55 +0200 Subject: speedtch: "extended reach" The speedtouch modem setup code was reverse engineered many years ago from a prehistoric windows driver. Less ancient windows drivers, even those from a few years ago, perform extra initialization steps which this patch adds to the linux driver. David Woodhouse observed that this initialization along with the firmware bin/sachu3/zzzlp2.eni from the driver at http://www.speedtouch.co.uk/downloads/330/301/UK3012%20Extended.zip improves line sync speeds by about 20%. He provided the original patch, which I've modified to use symbolic names (BMaxDSL, ModemMode, ModemOption) rather than magic numbers. These names may not seem like much of an improvement (after all, what is "ModemOption" exactly?), but they do have one big advantage: they are the names used in the windows registry. I've made them available as module parameters. Thanks are due to Aurelio Arroyo, who noticed the relationship between these magic numbers and the entries in Phonebook.ini. Signed-off-by: Duncan Sands Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/speedtch.c b/drivers/usb/atm/speedtch.c index 7c7b507..c870c80 100644 --- a/drivers/usb/atm/speedtch.c +++ b/drivers/usb/atm/speedtch.c @@ -55,7 +55,6 @@ static const char speedtch_driver_name[] = "speedtch"; #define OFFSET_d 9 /* size 4 */ #define OFFSET_e 13 /* size 1 */ #define OFFSET_f 14 /* size 1 */ -#define TOTAL 15 #define SIZE_7 1 #define SIZE_b 8 @@ -79,6 +78,18 @@ static int dl_512_first = DEFAULT_DL_512_FIRST; static int enable_isoc = DEFAULT_ENABLE_ISOC; static int sw_buffering = DEFAULT_SW_BUFFERING; +#define DEFAULT_B_MAX_DSL 8128 +#define DEFAULT_MODEM_MODE 11 +#define MODEM_OPTION_LENGTH 16 +static const unsigned char DEFAULT_MODEM_OPTION[MODEM_OPTION_LENGTH] = { + 0x10, 0x00, 0x00, 0x00, 0x20, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 +}; + +static unsigned int BMaxDSL = DEFAULT_B_MAX_DSL; +static unsigned char ModemMode = DEFAULT_MODEM_MODE; +static unsigned char ModemOption[MODEM_OPTION_LENGTH]; +static int num_ModemOption; + module_param(altsetting, uint, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(altsetting, "Alternative setting for data interface (bulk_default: " @@ -100,6 +111,17 @@ MODULE_PARM_DESC(sw_buffering, "Enable software buffering (default: " __MODULE_STRING(DEFAULT_SW_BUFFERING) ")"); +module_param(BMaxDSL, uint, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(BMaxDSL, + "default: " __MODULE_STRING(DEFAULT_B_MAX_DSL)); + +module_param(ModemMode, byte, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(ModemMode, + "default: " __MODULE_STRING(DEFAULT_MODEM_MODE)); + +module_param_array(ModemOption, byte, &num_ModemOption, S_IRUGO); +MODULE_PARM_DESC(ModemOption, "default: 0x10,0x00,0x00,0x00,0x20"); + #define INTERFACE_DATA 1 #define ENDPOINT_INT 0x81 #define ENDPOINT_BULK_DATA 0x07 @@ -108,10 +130,17 @@ MODULE_PARM_DESC(sw_buffering, #define hex2int(c) ( (c >= '0') && (c <= '9') ? (c - '0') : ((c & 0xf) + 9) ) +struct speedtch_params { + unsigned int altsetting; + unsigned int BMaxDSL; + unsigned char ModemMode; + unsigned char ModemOption[MODEM_OPTION_LENGTH]; +}; + struct speedtch_instance_data { struct usbatm_data *usbatm; - unsigned int altsetting; + struct speedtch_params params; /* set in probe, constant afterwards */ struct work_struct status_checker; @@ -123,7 +152,7 @@ struct speedtch_instance_data { struct urb *int_urb; unsigned char int_data[16]; - unsigned char scratch_buffer[TOTAL]; + unsigned char scratch_buffer[16]; }; /*************** @@ -186,6 +215,34 @@ static void speedtch_test_sequence(struct speedtch_instance_data *instance) 0x01, 0x40, 0x04, 0x00, buf, 3, CTRL_TIMEOUT); if (ret < 0) usb_warn(usbatm, "%s failed on URB150: %d\n", __func__, ret); + + /* Extra initialisation in recent drivers - gives higher speeds */ + + /* URBext1 */ + buf[0] = instance->params.ModemMode; + ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), + 0x01, 0x40, 0x11, 0x00, buf, 1, CTRL_TIMEOUT); + if (ret < 0) + usb_warn(usbatm, "%s failed on URBext1: %d\n", __func__, ret); + + /* URBext2 */ + /* This seems to be the one which actually triggers the higher sync + rate -- it does require the new firmware too, although it works OK + with older firmware */ + ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), + 0x01, 0x40, 0x14, 0x00, + instance->params.ModemOption, + MODEM_OPTION_LENGTH, CTRL_TIMEOUT); + if (ret < 0) + usb_warn(usbatm, "%s failed on URBext2: %d\n", __func__, ret); + + /* URBext3 */ + buf[0] = instance->params.BMaxDSL & 0xff; + buf[1] = instance->params.BMaxDSL >> 8; + ret = usb_control_msg(usb_dev, usb_sndctrlpipe(usb_dev, 0), + 0x01, 0x40, 0x12, 0x00, buf, 2, CTRL_TIMEOUT); + if (ret < 0) + usb_warn(usbatm, "%s failed on URBext3: %d\n", __func__, ret); } static int speedtch_upload_firmware(struct speedtch_instance_data *instance, @@ -285,8 +342,8 @@ static int speedtch_upload_firmware(struct speedtch_instance_data *instance, because we're in our own kernel thread anyway. */ msleep_interruptible(1000); - if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, instance->altsetting)) < 0) { - usb_err(usbatm, "%s: setting interface to %d failed (%d)!\n", __func__, instance->altsetting, ret); + if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, instance->params.altsetting)) < 0) { + usb_err(usbatm, "%s: setting interface to %d failed (%d)!\n", __func__, instance->params.altsetting, ret); goto out_free; } @@ -372,7 +429,7 @@ static int speedtch_read_status(struct speedtch_instance_data *instance) unsigned char *buf = instance->scratch_buffer; int ret; - memset(buf, 0, TOTAL); + memset(buf, 0, 16); ret = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0), 0x12, 0xc0, 0x07, 0x00, buf + OFFSET_7, SIZE_7, @@ -746,17 +803,21 @@ static int speedtch_bind(struct usbatm_data *usbatm, instance->usbatm = usbatm; - /* altsetting and enable_isoc may change at any moment, so take a snapshot */ - instance->altsetting = altsetting; + /* module parameters may change at any moment, so take a snapshot */ + instance->params.altsetting = altsetting; + instance->params.BMaxDSL = BMaxDSL; + instance->params.ModemMode = ModemMode; + memcpy(instance->params.ModemOption, DEFAULT_MODEM_OPTION, MODEM_OPTION_LENGTH); + memcpy(instance->params.ModemOption, ModemOption, num_ModemOption); use_isoc = enable_isoc; - if (instance->altsetting) - if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, instance->altsetting)) < 0) { - usb_err(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, instance->altsetting, ret); - instance->altsetting = 0; /* fall back to default */ + if (instance->params.altsetting) + if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, instance->params.altsetting)) < 0) { + usb_err(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, instance->params.altsetting, ret); + instance->params.altsetting = 0; /* fall back to default */ } - if (!instance->altsetting && use_isoc) + if (!instance->params.altsetting && use_isoc) if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, DEFAULT_ISOC_ALTSETTING)) < 0) { usb_dbg(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, DEFAULT_ISOC_ALTSETTING, ret); use_isoc = 0; /* fall back to bulk */ @@ -783,14 +844,14 @@ static int speedtch_bind(struct usbatm_data *usbatm, usb_info(usbatm, "isochronous transfer not supported - using bulk\n"); } - if (!use_isoc && !instance->altsetting) + if (!use_isoc && !instance->params.altsetting) if ((ret = usb_set_interface(usb_dev, INTERFACE_DATA, DEFAULT_BULK_ALTSETTING)) < 0) { usb_err(usbatm, "%s: setting interface to %2d failed (%d)!\n", __func__, DEFAULT_BULK_ALTSETTING, ret); goto fail_free; } - if (!instance->altsetting) - instance->altsetting = use_isoc ? DEFAULT_ISOC_ALTSETTING : DEFAULT_BULK_ALTSETTING; + if (!instance->params.altsetting) + instance->params.altsetting = use_isoc ? DEFAULT_ISOC_ALTSETTING : DEFAULT_BULK_ALTSETTING; usbatm->flags |= (use_isoc ? UDSL_USE_ISOC : 0); -- cgit v0.10.2 From 44960af1b6ab3e8fd23dc134fcf7862caf42936b Mon Sep 17 00:00:00 2001 From: Duncan Sands Date: Thu, 5 Oct 2006 11:05:50 +0200 Subject: cxacru: add the ZTE ZXDSL 852 From http://doc.ubuntu-fr.org/materiel/zxdsl852. Signed-off-by: Duncan Sands Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/cxacru.c b/drivers/usb/atm/cxacru.c index 3892a9e..e656563 100644 --- a/drivers/usb/atm/cxacru.c +++ b/drivers/usb/atm/cxacru.c @@ -793,6 +793,9 @@ static const struct usb_device_id cxacru_usb_ids[] = { { /* V = Conexant P = ADSL modem */ USB_DEVICE(0x0572, 0xcb06), .driver_info = (unsigned long) &cxacru_cb00 }, + { /* V = Conexant P = ADSL modem (ZTE ZXDSL 852) */ + USB_DEVICE(0x0572, 0xcb07), .driver_info = (unsigned long) &cxacru_cb00 + }, { /* V = Olitec P = ADSL modem version 2 */ USB_DEVICE(0x08e3, 0x0100), .driver_info = (unsigned long) &cxacru_cafe }, -- cgit v0.10.2 From 516077c1ee8a4a47cc41634a29954b636f3975ea Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Thu, 5 Oct 2006 09:04:11 +0200 Subject: USB: fix suspend support for usblp this implements suspend support for usblp. According to the CUPS people ENODEV will make CUPS retry the job. Thus it is returned in the runtime case. My printer survives suspend/resume cycles with it. Signed-off-by: Oliver Neukum Signed-off-by: Vojtech Pavlik Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c index a161d70..809d465 100644 --- a/drivers/usb/class/usblp.c +++ b/drivers/usb/class/usblp.c @@ -154,6 +154,7 @@ struct usblp { unsigned char used; /* True if open */ unsigned char present; /* True if not disconnected */ unsigned char bidir; /* interface is bidirectional */ + unsigned char sleeping; /* interface is suspended */ unsigned char *device_id_string; /* IEEE 1284 DEVICE ID string (ptr) */ /* first 2 bytes are (big-endian) length */ }; @@ -183,6 +184,7 @@ static void usblp_dump(struct usblp *usblp) { dbg("quirks=%d", usblp->quirks); dbg("used=%d", usblp->used); dbg("bidir=%d", usblp->bidir); + dbg("sleeping=%d", usblp->sleeping); dbg("device_id_string=\"%s\"", usblp->device_id_string ? usblp->device_id_string + 2 : @@ -338,6 +340,20 @@ static int usblp_check_status(struct usblp *usblp, int err) return newerr; } +static int handle_bidir (struct usblp *usblp) +{ + if (usblp->bidir && usblp->used && !usblp->sleeping) { + usblp->readcount = 0; + usblp->readurb->dev = usblp->dev; + if (usb_submit_urb(usblp->readurb, GFP_KERNEL) < 0) { + usblp->used = 0; + return -EIO; + } + } + + return 0; +} + /* * File op functions. */ @@ -390,14 +406,9 @@ static int usblp_open(struct inode *inode, struct file *file) usblp->writeurb->status = 0; usblp->readurb->status = 0; - if (usblp->bidir) { - usblp->readcount = 0; - usblp->readurb->dev = usblp->dev; - if (usb_submit_urb(usblp->readurb, GFP_KERNEL) < 0) { - retval = -EIO; - usblp->used = 0; - file->private_data = NULL; - } + if (handle_bidir(usblp) < 0) { + file->private_data = NULL; + retval = -EIO; } out: mutex_unlock (&usblp_mutex); @@ -460,6 +471,11 @@ static long usblp_ioctl(struct file *file, unsigned int cmd, unsigned long arg) goto done; } + if (usblp->sleeping) { + retval = -ENODEV; + goto done; + } + dbg("usblp_ioctl: cmd=0x%x (%c nr=%d len=%d dir=%d)", cmd, _IOC_TYPE(cmd), _IOC_NR(cmd), _IOC_SIZE(cmd), _IOC_DIR(cmd) ); @@ -658,6 +674,11 @@ static ssize_t usblp_write(struct file *file, const char __user *buffer, size_t return -ENODEV; } + if (usblp->sleeping) { + up (&usblp->sem); + return writecount ? writecount : -ENODEV; + } + if (usblp->writeurb->status != 0) { if (usblp->quirks & USBLP_QUIRK_BIDIR) { if (!usblp->wcomplete) @@ -749,6 +770,11 @@ static ssize_t usblp_read(struct file *file, char __user *buffer, size_t count, goto done; } + if (usblp->sleeping) { + count = -ENODEV; + goto done; + } + if (usblp->readurb->status) { err("usblp%d: error %d reading from printer", usblp->minor, usblp->readurb->status); @@ -1167,6 +1193,41 @@ static void usblp_disconnect(struct usb_interface *intf) mutex_unlock (&usblp_mutex); } +static int usblp_suspend (struct usb_interface *intf, pm_message_t message) +{ + struct usblp *usblp = usb_get_intfdata (intf); + + /* this races against normal access and open */ + mutex_lock (&usblp_mutex); + down (&usblp->sem); + /* we take no more IO */ + usblp->sleeping = 1; + /* we wait for anything printing */ + wait_event (usblp->wait, usblp->wcomplete || !usblp->present); + usblp_unlink_urbs(usblp); + up (&usblp->sem); + mutex_unlock (&usblp_mutex); + + return 0; +} + +static int usblp_resume (struct usb_interface *intf) +{ + struct usblp *usblp = usb_get_intfdata (intf); + int r; + + mutex_lock (&usblp_mutex); + down (&usblp->sem); + + usblp->sleeping = 0; + r = handle_bidir (usblp); + + up (&usblp->sem); + mutex_unlock (&usblp_mutex); + + return r; +} + static struct usb_device_id usblp_ids [] = { { USB_DEVICE_INFO(7, 1, 1) }, { USB_DEVICE_INFO(7, 1, 2) }, @@ -1183,6 +1244,8 @@ static struct usb_driver usblp_driver = { .name = "usblp", .probe = usblp_probe, .disconnect = usblp_disconnect, + .suspend = usblp_suspend, + .resume = usblp_resume, .id_table = usblp_ids, }; -- cgit v0.10.2 From 96a518928e1fd00a6d0eb344f420ea82aeec8ab9 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 9 Oct 2006 12:24:49 -0700 Subject: USB: ftdi-elan: fix sparse warnings Deleted some unused code that could do bad things on non-x86 platforms. Also fixed some minor formatting errors. Thanks to Al Viro for pointing out the sparse errors. Cc: Tony Olech Cc: Al Viro Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/misc/ftdi-elan.c b/drivers/usb/misc/ftdi-elan.c index 0eb26a2..37d1f4e 100644 --- a/drivers/usb/misc/ftdi-elan.c +++ b/drivers/usb/misc/ftdi-elan.c @@ -1186,11 +1186,8 @@ static ssize_t ftdi_elan_write(struct file *file, int retval = 0; struct urb *urb; char *buf; - char data[30 *3 + 4]; - char *d = data; - const char __user *s = user_buffer; - int m = (sizeof(data) - 1) / 3; - struct usb_ftdi *ftdi = (struct usb_ftdi *)file->private_data; + struct usb_ftdi *ftdi = file->private_data; + if (ftdi->disconnected > 0) { return -ENODEV; } @@ -1220,27 +1217,18 @@ static ssize_t ftdi_elan_write(struct file *file, if (retval) { dev_err(&ftdi->udev->dev, "failed submitting write urb, error %" "d\n", retval); - goto error_4; + goto error_3; } usb_free_urb(urb); - exit:; - if (count > m) { - int I = m - 1; - while (I-- > 0) { - d += sprintf(d, " %02X", 0x000000FF & *s++); - } - d += sprintf(d, " .."); - } else { - int I = count; - while (I-- > 0) { - d += sprintf(d, " %02X", 0x000000FF & *s++); - } - } + +exit: return count; - error_4: error_3:usb_buffer_free(ftdi->udev, count, buf, - urb->transfer_dma); - error_2:usb_free_urb(urb); - error_1:return retval; +error_3: + usb_buffer_free(ftdi->udev, count, buf, urb->transfer_dma); +error_2: + usb_free_urb(urb); +error_1: + return retval; } static struct file_operations ftdi_elan_fops = { -- cgit v0.10.2 From b62df4516981745d4b5de01ceec1d65a9174a524 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 10 Oct 2006 10:54:00 -0400 Subject: UHCI: workaround for Asus motherboard This patch (as798) adds a workaround to uhci-hcd. At least one Asus motherboard is wired in such a way that any device attached to a suspended UHCI controller will prevent the system from entering suspend-to-RAM by immediately waking it up. The only way around the problem is to turn the controller off instead of suspending it. This fixes Bugzilla #6193. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c index 45ee692..226bf3d 100644 --- a/drivers/usb/host/uhci-hcd.c +++ b/drivers/usb/host/uhci-hcd.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include @@ -196,12 +197,42 @@ static int resume_detect_interrupts_are_broken(struct uhci_hcd *uhci) return 0; } +static int remote_wakeup_is_broken(struct uhci_hcd *uhci) +{ + static struct dmi_system_id broken_wakeup_table[] = { + { + .ident = "Asus A7V8X", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK"), + DMI_MATCH(DMI_BOARD_NAME, "A7V8X"), + DMI_MATCH(DMI_BOARD_VERSION, "REV 1.xx"), + } + }, + { } + }; + int port; + + /* One of Asus's motherboards has a bug which causes it to + * wake up immediately from suspend-to-RAM if any of the ports + * are connected. In such cases we will not set EGSM. + */ + if (dmi_check_system(broken_wakeup_table)) { + for (port = 0; port < uhci->rh_numports; ++port) { + if (inw(uhci->io_addr + USBPORTSC1 + port * 2) & + USBPORTSC_CCS) + return 1; + } + } + + return 0; +} + static void suspend_rh(struct uhci_hcd *uhci, enum uhci_rh_state new_state) __releases(uhci->lock) __acquires(uhci->lock) { int auto_stop; - int int_enable; + int int_enable, egsm_enable; auto_stop = (new_state == UHCI_RH_AUTO_STOPPED); dev_dbg(&uhci_to_hcd(uhci)->self.root_hub->dev, @@ -217,15 +248,18 @@ __acquires(uhci->lock) } /* Enable resume-detect interrupts if they work. - * Then enter Global Suspend mode, still configured. + * Then enter Global Suspend mode if _it_ works, still configured. */ + egsm_enable = USBCMD_EGSM; uhci->working_RD = 1; int_enable = USBINTR_RESUME; - if (resume_detect_interrupts_are_broken(uhci)) { + if (remote_wakeup_is_broken(uhci)) + egsm_enable = 0; + if (resume_detect_interrupts_are_broken(uhci) || !egsm_enable) uhci->working_RD = int_enable = 0; - } + outw(int_enable, uhci->io_addr + USBINTR); - outw(USBCMD_EGSM | USBCMD_CF, uhci->io_addr + USBCMD); + outw(egsm_enable | USBCMD_CF, uhci->io_addr + USBCMD); mb(); udelay(5); -- cgit v0.10.2 From c40fd5ea565587c05b0e2c49c02cad2c35fd85c6 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 10 Oct 2006 11:55:47 -0400 Subject: usbcore: fix refcount bug in endpoint removal This patch (as799) fixes a nasty refcount error in the USB endpoint class. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/endpoint.c b/drivers/usb/core/endpoint.c index 3ebb901..40ba76a 100644 --- a/drivers/usb/core/endpoint.c +++ b/drivers/usb/core/endpoint.c @@ -282,8 +282,6 @@ void usb_remove_ep_files(struct usb_host_endpoint *endpoint) sysfs_remove_group(&endpoint->ep_dev->dev.kobj, &ep_dev_attr_grp); device_unregister(&endpoint->ep_dev->dev); endpoint->ep_dev = NULL; + destroy_endpoint_class(); } - destroy_endpoint_class(); } - - -- cgit v0.10.2 From d5477c11111467e19787f00d3cab20fb48c2699e Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 10 Oct 2006 11:56:26 -0400 Subject: usbcore: fix endpoint device creation This patch (as800) straightens out the USB endpoint class device creation routine, fixing a refcount bug in the process. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/endpoint.c b/drivers/usb/core/endpoint.c index 40ba76a..3b2d137 100644 --- a/drivers/usb/core/endpoint.c +++ b/drivers/usb/core/endpoint.c @@ -223,7 +223,7 @@ int usb_create_ep_files(struct device *parent, ep_dev = kzalloc(sizeof(*ep_dev), GFP_KERNEL); if (!ep_dev) { retval = -ENOMEM; - goto exit; + goto error_alloc; } /* fun calculation to determine the minor of this endpoint */ @@ -241,33 +241,31 @@ int usb_create_ep_files(struct device *parent, retval = device_register(&ep_dev->dev); if (retval) - goto error; + goto error_register; retval = sysfs_create_group(&ep_dev->dev.kobj, &ep_dev_attr_grp); if (retval) goto error_group; - endpoint->ep_dev = ep_dev; - /* create the symlink to the old-style "ep_XX" directory */ sprintf(name, "ep_%02x", endpoint->desc.bEndpointAddress); - retval = sysfs_create_link(&parent->kobj, - &endpoint->ep_dev->dev.kobj, name); + retval = sysfs_create_link(&parent->kobj, &ep_dev->dev.kobj, name); if (retval) goto error_link; -exit: + endpoint->ep_dev = ep_dev; return retval; error_link: sysfs_remove_group(&ep_dev->dev.kobj, &ep_dev_attr_grp); - error_group: device_unregister(&ep_dev->dev); - endpoint->ep_dev = NULL; destroy_endpoint_class(); return retval; -error: + +error_register: kfree(ep_dev); +error_alloc: destroy_endpoint_class(); +exit: return retval; } -- cgit v0.10.2 From f8ac232ad7388bfff680b26e84b3ac63889d1cea Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sun, 8 Oct 2006 16:02:00 +0400 Subject: USB: drivers/usb/net/*: use BUILD_BUG_ON Signed-off-by: Alexey Dobriyan Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/cdc_ether.c b/drivers/usb/net/cdc_ether.c index 82ce035..f6971b8 100644 --- a/drivers/usb/net/cdc_ether.c +++ b/drivers/usb/net/cdc_ether.c @@ -498,7 +498,7 @@ static struct usb_driver cdc_driver = { static int __init cdc_init(void) { - BUG_ON((sizeof(((struct usbnet *)0)->data) + BUILD_BUG_ON((sizeof(((struct usbnet *)0)->data) < sizeof(struct cdc_state))); return usb_register(&cdc_driver); diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index 24bd348..af6d061 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -1225,7 +1225,7 @@ EXPORT_SYMBOL_GPL(usbnet_resume); static int __init usbnet_init(void) { /* compiler should optimize this out */ - BUG_ON (sizeof (((struct sk_buff *)0)->cb) + BUILD_BUG_ON (sizeof (((struct sk_buff *)0)->cb) < sizeof (struct skb_data)); random_ether_addr(node_id); -- cgit v0.10.2 From 27d39e2627dc7493f554bc0549d8c63953762478 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 9 Oct 2006 18:09:33 +0900 Subject: usb devio: handle class_device_create() error This patch adds missing class_device_create() error check, and makes notifier return NOTIFY_BAD. Signed-off-by: Akinobu Mita Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c index 724822c..fed92be 100644 --- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -1588,15 +1588,18 @@ const struct file_operations usbfs_device_file_operations = { .release = usbdev_release, }; -static void usbdev_add(struct usb_device *dev) +static int usbdev_add(struct usb_device *dev) { int minor = ((dev->bus->busnum-1) * 128) + (dev->devnum-1); dev->class_dev = class_device_create(usb_device_class, NULL, MKDEV(USB_DEVICE_MAJOR, minor), &dev->dev, "usbdev%d.%d", dev->bus->busnum, dev->devnum); + if (IS_ERR(dev->class_dev)) + return PTR_ERR(dev->class_dev); dev->class_dev->class_data = dev; + return 0; } static void usbdev_remove(struct usb_device *dev) @@ -1609,7 +1612,8 @@ static int usbdev_notify(struct notifier_block *self, unsigned long action, { switch (action) { case USB_DEVICE_ADD: - usbdev_add(dev); + if (usbdev_add(dev)) + return NOTIFY_BAD; break; case USB_DEVICE_REMOVE: usbdev_remove(dev); -- cgit v0.10.2 From 2a36d7083438ccb607055abae633f39495a99947 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 9 Oct 2006 00:08:00 +0200 Subject: USB: driver for mcs7830 (aka DeLOCK) USB ethernet adapter This driver adds support for the DeLOCK USB ethernet adapter and potentially others based on the MosChip MCS7830 chip. It is based on the usbnet and asix drivers as well as the original device driver provided by MosChip, which in turn was based on the usbnet driver. It has been tested successfully on an OHCI, but interestingly there seems to be a problem with the mcs7830 when connected to the ICH6/EHCI in my thinkpad: it keeps receiving lots of broken packets in the RX interrupt. The problem goes away when I'm using an active USB hub, so I assume it's not related to the device driver, but rather to the hardware. Signed-off-by: David Brownell Signed-off-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/Kconfig b/drivers/usb/net/Kconfig index 0540596..454a186 100644 --- a/drivers/usb/net/Kconfig +++ b/drivers/usb/net/Kconfig @@ -207,6 +207,14 @@ config USB_NET_PLUSB Choose this option if you're using a host-to-host cable with one of these chips. +config USB_NET_MCS7830 + tristate "MosChip MCS7830 based Ethernet adapters" + depends on USB_USBNET + help + Choose this option if you're using a 10/100 Ethernet USB2 + adapter based on the MosChip 7830 controller. This includes + adapters marketed under the DeLOCK brand. + config USB_NET_RNDIS_HOST tristate "Host for RNDIS devices (EXPERIMENTAL)" depends on USB_USBNET && EXPERIMENTAL diff --git a/drivers/usb/net/Makefile b/drivers/usb/net/Makefile index 160f19d..7b51964 100644 --- a/drivers/usb/net/Makefile +++ b/drivers/usb/net/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_USB_NET_PLUSB) += plusb.o obj-$(CONFIG_USB_NET_RNDIS_HOST) += rndis_host.o obj-$(CONFIG_USB_NET_CDC_SUBSET) += cdc_subset.o obj-$(CONFIG_USB_NET_ZAURUS) += zaurus.o +obj-$(CONFIG_USB_NET_MCS7830) += mcs7830.o obj-$(CONFIG_USB_USBNET) += usbnet.o ifeq ($(CONFIG_USB_DEBUG),y) diff --git a/drivers/usb/net/mcs7830.c b/drivers/usb/net/mcs7830.c new file mode 100644 index 0000000..0266090 --- /dev/null +++ b/drivers/usb/net/mcs7830.c @@ -0,0 +1,525 @@ +/* + * MosChips MCS7830 based USB 2.0 Ethernet Devices + * + * based on usbnet.c, asix.c and the vendor provided mcs7830 driver + * + * Copyright (C) 2006 Arnd Bergmann + * Copyright (C) 2003-2005 David Hollis + * Copyright (C) 2005 Phil Chang + * Copyright (c) 2002-2003 TiVo Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "usbnet.h" + +/* requests */ +#define MCS7830_RD_BMREQ (USB_DIR_IN | USB_TYPE_VENDOR | \ + USB_RECIP_DEVICE) +#define MCS7830_WR_BMREQ (USB_DIR_OUT | USB_TYPE_VENDOR | \ + USB_RECIP_DEVICE) +#define MCS7830_RD_BREQ 0x0E +#define MCS7830_WR_BREQ 0x0D + +#define MCS7830_CTRL_TIMEOUT 1000 +#define MCS7830_MAX_MCAST 64 + +#define MCS7830_VENDOR_ID 0x9710 +#define MCS7830_PRODUCT_ID 0x7830 + +#define MCS7830_MII_ADVERTISE (ADVERTISE_PAUSE_CAP | ADVERTISE_100FULL | \ + ADVERTISE_100HALF | ADVERTISE_10FULL | \ + ADVERTISE_10HALF | ADVERTISE_CSMA) + +/* HIF_REG_XX coressponding index value */ +enum { + HIF_REG_MULTICAST_HASH = 0x00, + HIF_REG_PACKET_GAP1 = 0x08, + HIF_REG_PACKET_GAP2 = 0x09, + HIF_REG_PHY_DATA = 0x0a, + HIF_REG_PHY_CMD1 = 0x0c, + HIF_REG_PHY_CMD1_READ = 0x40, + HIF_REG_PHY_CMD1_WRITE = 0x20, + HIF_REG_PHY_CMD1_PHYADDR = 0x01, + HIF_REG_PHY_CMD2 = 0x0d, + HIF_REG_PHY_CMD2_PEND_FLAG_BIT = 0x80, + HIF_REG_PHY_CMD2_READY_FLAG_BIT = 0x40, + HIF_REG_CONFIG = 0x0e, + HIF_REG_CONFIG_CFG = 0x80, + HIF_REG_CONFIG_SPEED100 = 0x40, + HIF_REG_CONFIG_FULLDUPLEX_ENABLE = 0x20, + HIF_REG_CONFIG_RXENABLE = 0x10, + HIF_REG_CONFIG_TXENABLE = 0x08, + HIF_REG_CONFIG_SLEEPMODE = 0x04, + HIF_REG_CONFIG_ALLMULTICAST = 0x02, + HIF_REG_CONFIG_PROMISCIOUS = 0x01, + HIF_REG_ETHERNET_ADDR = 0x0f, + HIF_REG_22 = 0x15, + HIF_REG_PAUSE_THRESHOLD = 0x16, + HIF_REG_PAUSE_THRESHOLD_DEFAULT = 0, +}; + +struct mcs7830_data { + u8 multi_filter[8]; + u8 config; +}; + +static const char driver_name[] = "MOSCHIP usb-ethernet driver"; + +static int mcs7830_get_reg(struct usbnet *dev, u16 index, u16 size, void *data) +{ + struct usb_device *xdev = dev->udev; + int ret; + + ret = usb_control_msg(xdev, usb_rcvctrlpipe(xdev, 0), MCS7830_RD_BREQ, + MCS7830_RD_BMREQ, 0x0000, index, data, + size, msecs_to_jiffies(MCS7830_CTRL_TIMEOUT)); + return ret; +} + +static int mcs7830_set_reg(struct usbnet *dev, u16 index, u16 size, void *data) +{ + struct usb_device *xdev = dev->udev; + int ret; + + ret = usb_control_msg(xdev, usb_sndctrlpipe(xdev, 0), MCS7830_WR_BREQ, + MCS7830_WR_BMREQ, 0x0000, index, data, + size, msecs_to_jiffies(MCS7830_CTRL_TIMEOUT)); + return ret; +} + +static void mcs7830_async_cmd_callback(struct urb *urb) +{ + struct usb_ctrlrequest *req = (struct usb_ctrlrequest *)urb->context; + + if (urb->status < 0) + printk(KERN_DEBUG "mcs7830_async_cmd_callback() failed with %d", + urb->status); + + kfree(req); + usb_free_urb(urb); +} + +static void mcs7830_set_reg_async(struct usbnet *dev, u16 index, u16 size, void *data) +{ + struct usb_ctrlrequest *req; + int ret; + struct urb *urb; + + urb = usb_alloc_urb(0, GFP_ATOMIC); + if (!urb) { + dev_dbg(&dev->udev->dev, "Error allocating URB " + "in write_cmd_async!"); + return; + } + + req = kmalloc(sizeof *req, GFP_ATOMIC); + if (!req) { + dev_err(&dev->udev->dev, "Failed to allocate memory for " + "control request"); + goto out; + } + req->bRequestType = MCS7830_WR_BMREQ; + req->bRequest = MCS7830_WR_BREQ; + req->wValue = 0; + req->wIndex = cpu_to_le16(index); + req->wLength = cpu_to_le16(size); + + usb_fill_control_urb(urb, dev->udev, + usb_sndctrlpipe(dev->udev, 0), + (void *)req, data, size, + mcs7830_async_cmd_callback, req); + + ret = usb_submit_urb(urb, GFP_ATOMIC); + if (ret < 0) { + dev_err(&dev->udev->dev, "Error submitting the control " + "message: ret=%d", ret); + goto out; + } + return; +out: + kfree(req); + usb_free_urb(urb); +} + +static int mcs7830_get_address(struct usbnet *dev) +{ + int ret; + ret = mcs7830_get_reg(dev, HIF_REG_ETHERNET_ADDR, ETH_ALEN, + dev->net->dev_addr); + if (ret < 0) + return ret; + return 0; +} + +static int mcs7830_read_phy(struct usbnet *dev, u8 index) +{ + int ret; + int i; + __le16 val; + + u8 cmd[2] = { + HIF_REG_PHY_CMD1_READ | HIF_REG_PHY_CMD1_PHYADDR, + HIF_REG_PHY_CMD2_PEND_FLAG_BIT | index, + }; + + /* write the MII command */ + ret = mcs7830_set_reg(dev, HIF_REG_PHY_CMD1, 2, cmd); + if (ret < 0) + goto out; + + /* wait for the data to become valid, should be within < 1ms */ + for (i = 0; i < 10; i++) { + ret = mcs7830_get_reg(dev, HIF_REG_PHY_CMD1, 2, cmd); + if ((ret < 0) || (cmd[1] & HIF_REG_PHY_CMD2_READY_FLAG_BIT)) + break; + ret = -EIO; + msleep(1); + } + if (ret < 0) + goto out; + + /* read actual register contents */ + ret = mcs7830_get_reg(dev, HIF_REG_PHY_DATA, 2, &val); + if (ret < 0) + goto out; + ret = le16_to_cpu(val); + dev_dbg(&dev->udev->dev, "read PHY reg %02x: %04x (%d tries)\n", + index, val, i); +out: + return ret; +} + +static int mcs7830_write_phy(struct usbnet *dev, u8 index, u16 val) +{ + int ret; + int i; + __le16 le_val; + + u8 cmd[2] = { + HIF_REG_PHY_CMD1_WRITE | HIF_REG_PHY_CMD1_PHYADDR, + HIF_REG_PHY_CMD2_PEND_FLAG_BIT | (index & 0x1F), + }; + + /* write the new register contents */ + le_val = cpu_to_le16(val); + ret = mcs7830_set_reg(dev, HIF_REG_PHY_DATA, 2, &le_val); + if (ret < 0) + goto out; + + /* write the MII command */ + ret = mcs7830_set_reg(dev, HIF_REG_PHY_CMD1, 2, cmd); + if (ret < 0) + goto out; + + /* wait for the command to be accepted by the PHY */ + for (i = 0; i < 10; i++) { + ret = mcs7830_get_reg(dev, HIF_REG_PHY_CMD1, 2, cmd); + if ((ret < 0) || (cmd[1] & HIF_REG_PHY_CMD2_READY_FLAG_BIT)) + break; + ret = -EIO; + msleep(1); + } + if (ret < 0) + goto out; + + ret = 0; + dev_dbg(&dev->udev->dev, "write PHY reg %02x: %04x (%d tries)\n", + index, val, i); +out: + return ret; +} + +/* + * This algorithm comes from the original mcs7830 version 1.4 driver, + * not sure if it is needed. + */ +static int mcs7830_set_autoneg(struct usbnet *dev, int ptrUserPhyMode) +{ + int ret; + /* Enable all media types */ + ret = mcs7830_write_phy(dev, MII_ADVERTISE, MCS7830_MII_ADVERTISE); + + /* First reset BMCR */ + if (!ret) + ret = mcs7830_write_phy(dev, MII_BMCR, 0x0000); + /* Enable Auto Neg */ + if (!ret) + ret = mcs7830_write_phy(dev, MII_BMCR, BMCR_ANENABLE); + /* Restart Auto Neg (Keep the Enable Auto Neg Bit Set) */ + if (!ret) + ret = mcs7830_write_phy(dev, MII_BMCR, + BMCR_ANENABLE | BMCR_ANRESTART ); + return ret < 0 ? : 0; +} + + +/* + * if we can read register 22, the chip revision is C or higher + */ +static int mcs7830_get_rev(struct usbnet *dev) +{ + u8 dummy[2]; + int ret; + ret = mcs7830_get_reg(dev, HIF_REG_22, 2, dummy); + if (ret > 0) + return 2; /* Rev C or later */ + return 1; /* earlier revision */ +} + +/* + * On rev. C we need to set the pause threshold + */ +static void mcs7830_rev_C_fixup(struct usbnet *dev) +{ + u8 pause_threshold = HIF_REG_PAUSE_THRESHOLD_DEFAULT; + int retry; + + for (retry = 0; retry < 2; retry++) { + if (mcs7830_get_rev(dev) == 2) { + dev_info(&dev->udev->dev, "applying rev.C fixup\n"); + mcs7830_set_reg(dev, HIF_REG_PAUSE_THRESHOLD, + 1, &pause_threshold); + } + msleep(1); + } +} + +static int mcs7830_init_dev(struct usbnet *dev) +{ + int ret; + int retry; + + /* Read MAC address from EEPROM */ + ret = -EINVAL; + for (retry = 0; retry < 5 && ret; retry++) + ret = mcs7830_get_address(dev); + if (ret) { + dev_warn(&dev->udev->dev, "Cannot read MAC address\n"); + goto out; + } + + /* Set up PHY */ + ret = mcs7830_set_autoneg(dev, 0); + if (ret) { + dev_info(&dev->udev->dev, "Cannot set autoneg\n"); + goto out; + } + + mcs7830_rev_C_fixup(dev); + ret = 0; +out: + return ret; +} + +static int mcs7830_mdio_read(struct net_device *netdev, int phy_id, + int location) +{ + struct usbnet *dev = netdev->priv; + return mcs7830_read_phy(dev, location); +} + +static void mcs7830_mdio_write(struct net_device *netdev, int phy_id, + int location, int val) +{ + struct usbnet *dev = netdev->priv; + mcs7830_write_phy(dev, location, val); +} + +static int mcs7830_ioctl(struct net_device *net, struct ifreq *rq, int cmd) +{ + struct usbnet *dev = netdev_priv(net); + return generic_mii_ioctl(&dev->mii, if_mii(rq), cmd, NULL); +} + +/* credits go to asix_set_multicast */ +static void mcs7830_set_multicast(struct net_device *net) +{ + struct usbnet *dev = netdev_priv(net); + struct mcs7830_data *data = (struct mcs7830_data *)&dev->data; + + data->config = HIF_REG_CONFIG_TXENABLE; + + /* this should not be needed, but it doesn't work otherwise */ + data->config |= HIF_REG_CONFIG_ALLMULTICAST; + + if (net->flags & IFF_PROMISC) { + data->config |= HIF_REG_CONFIG_PROMISCIOUS; + } else if (net->flags & IFF_ALLMULTI + || net->mc_count > MCS7830_MAX_MCAST) { + data->config |= HIF_REG_CONFIG_ALLMULTICAST; + } else if (net->mc_count == 0) { + /* just broadcast and directed */ + } else { + /* We use the 20 byte dev->data + * for our 8 byte filter buffer + * to avoid allocating memory that + * is tricky to free later */ + struct dev_mc_list *mc_list = net->mc_list; + u32 crc_bits; + int i; + + memset(data->multi_filter, 0, sizeof data->multi_filter); + + /* Build the multicast hash filter. */ + for (i = 0; i < net->mc_count; i++) { + crc_bits = ether_crc(ETH_ALEN, mc_list->dmi_addr) >> 26; + data->multi_filter[crc_bits >> 3] |= 1 << (crc_bits & 7); + mc_list = mc_list->next; + } + + mcs7830_set_reg_async(dev, HIF_REG_MULTICAST_HASH, + sizeof data->multi_filter, + data->multi_filter); + } + + mcs7830_set_reg_async(dev, HIF_REG_CONFIG, 1, &data->config); +} + +static int mcs7830_get_regs_len(struct net_device *net) +{ + struct usbnet *dev = netdev_priv(net); + + switch (mcs7830_get_rev(dev)) { + case 1: + return 21; + case 2: + return 32; + } + return 0; +} + +static void mcs7830_get_drvinfo(struct net_device *net, struct ethtool_drvinfo *drvinfo) +{ + usbnet_get_drvinfo(net, drvinfo); + drvinfo->regdump_len = mcs7830_get_regs_len(net); +} + +static void mcs7830_get_regs(struct net_device *net, struct ethtool_regs *regs, void *data) +{ + struct usbnet *dev = netdev_priv(net); + + regs->version = mcs7830_get_rev(dev); + mcs7830_get_reg(dev, 0, regs->len, data); +} + +static struct ethtool_ops mcs7830_ethtool_ops = { + .get_drvinfo = mcs7830_get_drvinfo, + .get_regs_len = mcs7830_get_regs_len, + .get_regs = mcs7830_get_regs, + + /* common usbnet calls */ + .get_msglevel = usbnet_get_msglevel, + .set_msglevel = usbnet_set_msglevel, +}; + +static int mcs7830_bind(struct usbnet *dev, struct usb_interface *udev) +{ + struct net_device *net = dev->net; + int ret; + + ret = mcs7830_init_dev(dev); + if (ret) + goto out; + + net->do_ioctl = mcs7830_ioctl; + net->ethtool_ops = &mcs7830_ethtool_ops; + net->set_multicast_list = mcs7830_set_multicast; + mcs7830_set_multicast(net); + + /* reserve space for the status byte on rx */ + dev->rx_urb_size = ETH_FRAME_LEN + 1; + + dev->mii.mdio_read = mcs7830_mdio_read; + dev->mii.mdio_write = mcs7830_mdio_write; + dev->mii.dev = net; + dev->mii.phy_id_mask = 0x3f; + dev->mii.reg_num_mask = 0x1f; + dev->mii.phy_id = *((u8 *) net->dev_addr + 1); + + ret = usbnet_get_endpoints(dev, udev); +out: + return ret; +} + +/* The chip always appends a status bytes that we need to strip */ +static int mcs7830_rx_fixup(struct usbnet *dev, struct sk_buff *skb) +{ + u8 status; + + if (skb->len == 0) { + dev_err(&dev->udev->dev, "unexpected empty rx frame\n"); + return 0; + } + + skb_trim(skb, skb->len - 1); + status = skb->data[skb->len]; + + if (status != 0x20) + dev_dbg(&dev->udev->dev, "rx fixup status %x\n", status); + + return skb->len > 0; +} + +static const struct driver_info moschip_info = { + .description = "MOSCHIP 7830 usb-NET adapter", + .bind = mcs7830_bind, + .rx_fixup = mcs7830_rx_fixup, + .flags = FLAG_ETHER, + .in = 1, + .out = 2, +}; + +static const struct usb_device_id products[] = { + { + USB_DEVICE(MCS7830_VENDOR_ID, MCS7830_PRODUCT_ID), + .driver_info = (unsigned long) &moschip_info, + }, + {}, +}; +MODULE_DEVICE_TABLE(usb, products); + +static struct usb_driver mcs7830_driver = { + .name = driver_name, + .id_table = products, + .probe = usbnet_probe, + .disconnect = usbnet_disconnect, + .suspend = usbnet_suspend, + .resume = usbnet_resume, +}; + +static int __init mcs7830_init(void) +{ + return usb_register(&mcs7830_driver); +} +module_init(mcs7830_init); + +static void __exit mcs7830_exit(void) +{ + usb_deregister(&mcs7830_driver); +} +module_exit(mcs7830_exit); + +MODULE_DESCRIPTION("USB to network adapter MCS7830)"); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From c41286fd42f3545513f8de9f61028120b6d38e89 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 9 Oct 2006 00:08:01 +0200 Subject: usbnet: improve generic ethtool support This adds generic support for the ethtool commands get_settings, set_settings, get_link and nway_reset to usbnet. These are now implemented using mii functions when a low-level driver supports mdio_read/mdio_write and does not override the usbnet ethtool commands with its own. Currently, this applies to the asix and the mcs7830 drivers. I have tested it on mcs7830. Signed-off-by: Arnd Bergmann Acked-by: David Hollis Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/asix.c b/drivers/usb/net/asix.c index c73dd22..5edd053 100644 --- a/drivers/usb/net/asix.c +++ b/drivers/usb/net/asix.c @@ -700,32 +700,6 @@ static void asix_get_drvinfo (struct net_device *net, info->eedump_len = data->eeprom_len; } -static int asix_get_settings(struct net_device *net, struct ethtool_cmd *cmd) -{ - struct usbnet *dev = netdev_priv(net); - - return mii_ethtool_gset(&dev->mii,cmd); -} - -static int asix_set_settings(struct net_device *net, struct ethtool_cmd *cmd) -{ - struct usbnet *dev = netdev_priv(net); - int res = mii_ethtool_sset(&dev->mii,cmd); - - /* link speed/duplex might have changed */ - if (dev->driver_info->link_reset) - dev->driver_info->link_reset(dev); - - return res; -} - -static int asix_nway_reset(struct net_device *net) -{ - struct usbnet *dev = netdev_priv(net); - - return mii_nway_restart(&dev->mii); -} - static u32 asix_get_link(struct net_device *net) { struct usbnet *dev = netdev_priv(net); @@ -746,15 +720,15 @@ static int asix_ioctl (struct net_device *net, struct ifreq *rq, int cmd) static struct ethtool_ops ax88172_ethtool_ops = { .get_drvinfo = asix_get_drvinfo, .get_link = asix_get_link, - .nway_reset = asix_nway_reset, .get_msglevel = usbnet_get_msglevel, .set_msglevel = usbnet_set_msglevel, .get_wol = asix_get_wol, .set_wol = asix_set_wol, .get_eeprom_len = asix_get_eeprom_len, .get_eeprom = asix_get_eeprom, - .get_settings = asix_get_settings, - .set_settings = asix_set_settings, + .get_settings = usbnet_get_settings, + .set_settings = usbnet_set_settings, + .nway_reset = usbnet_nway_reset, }; static void ax88172_set_multicast(struct net_device *net) @@ -885,15 +859,15 @@ out1: static struct ethtool_ops ax88772_ethtool_ops = { .get_drvinfo = asix_get_drvinfo, .get_link = asix_get_link, - .nway_reset = asix_nway_reset, .get_msglevel = usbnet_get_msglevel, .set_msglevel = usbnet_set_msglevel, .get_wol = asix_get_wol, .set_wol = asix_set_wol, .get_eeprom_len = asix_get_eeprom_len, .get_eeprom = asix_get_eeprom, - .get_settings = asix_get_settings, - .set_settings = asix_set_settings, + .get_settings = usbnet_get_settings, + .set_settings = usbnet_set_settings, + .nway_reset = usbnet_nway_reset, }; static int ax88772_link_reset(struct usbnet *dev) @@ -1046,15 +1020,15 @@ out1: static struct ethtool_ops ax88178_ethtool_ops = { .get_drvinfo = asix_get_drvinfo, .get_link = asix_get_link, - .nway_reset = asix_nway_reset, .get_msglevel = usbnet_get_msglevel, .set_msglevel = usbnet_set_msglevel, .get_wol = asix_get_wol, .set_wol = asix_set_wol, .get_eeprom_len = asix_get_eeprom_len, .get_eeprom = asix_get_eeprom, - .get_settings = asix_get_settings, - .set_settings = asix_set_settings, + .get_settings = usbnet_get_settings, + .set_settings = usbnet_set_settings, + .nway_reset = usbnet_nway_reset, }; static int marvell_phy_init(struct usbnet *dev) diff --git a/drivers/usb/net/mcs7830.c b/drivers/usb/net/mcs7830.c index 0266090..23a8066 100644 --- a/drivers/usb/net/mcs7830.c +++ b/drivers/usb/net/mcs7830.c @@ -430,8 +430,12 @@ static struct ethtool_ops mcs7830_ethtool_ops = { .get_regs = mcs7830_get_regs, /* common usbnet calls */ + .get_link = usbnet_get_link, .get_msglevel = usbnet_get_msglevel, .set_msglevel = usbnet_set_msglevel, + .get_settings = usbnet_get_settings, + .set_settings = usbnet_set_settings, + .nway_reset = usbnet_nway_reset, }; static int mcs7830_bind(struct usbnet *dev, struct usb_interface *udev) diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index af6d061..decc1b1 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -669,6 +669,37 @@ done: * they'll probably want to use this base set. */ +int usbnet_get_settings (struct net_device *net, struct ethtool_cmd *cmd) +{ + struct usbnet *dev = netdev_priv(net); + + if (!dev->mii.mdio_read) + return -EOPNOTSUPP; + + return mii_ethtool_gset(&dev->mii, cmd); +} +EXPORT_SYMBOL_GPL(usbnet_get_settings); + +int usbnet_set_settings (struct net_device *net, struct ethtool_cmd *cmd) +{ + struct usbnet *dev = netdev_priv(net); + int retval; + + if (!dev->mii.mdio_write) + return -EOPNOTSUPP; + + retval = mii_ethtool_sset(&dev->mii, cmd); + + /* link speed/duplex might have changed */ + if (dev->driver_info->link_reset) + dev->driver_info->link_reset(dev); + + return retval; + +} +EXPORT_SYMBOL_GPL(usbnet_set_settings); + + void usbnet_get_drvinfo (struct net_device *net, struct ethtool_drvinfo *info) { struct usbnet *dev = netdev_priv(net); @@ -682,7 +713,7 @@ void usbnet_get_drvinfo (struct net_device *net, struct ethtool_drvinfo *info) } EXPORT_SYMBOL_GPL(usbnet_get_drvinfo); -static u32 usbnet_get_link (struct net_device *net) +u32 usbnet_get_link (struct net_device *net) { struct usbnet *dev = netdev_priv(net); @@ -690,9 +721,14 @@ static u32 usbnet_get_link (struct net_device *net) if (dev->driver_info->check_connect) return dev->driver_info->check_connect (dev) == 0; + /* if the device has mii operations, use those */ + if (dev->mii.mdio_read) + return mii_link_ok(&dev->mii); + /* Otherwise, say we're up (to avoid breaking scripts) */ return 1; } +EXPORT_SYMBOL_GPL(usbnet_get_link); u32 usbnet_get_msglevel (struct net_device *net) { @@ -710,10 +746,24 @@ void usbnet_set_msglevel (struct net_device *net, u32 level) } EXPORT_SYMBOL_GPL(usbnet_set_msglevel); +int usbnet_nway_reset(struct net_device *net) +{ + struct usbnet *dev = netdev_priv(net); + + if (!dev->mii.mdio_write) + return -EOPNOTSUPP; + + return mii_nway_restart(&dev->mii); +} +EXPORT_SYMBOL_GPL(usbnet_nway_reset); + /* drivers may override default ethtool_ops in their bind() routine */ static struct ethtool_ops usbnet_ethtool_ops = { + .get_settings = usbnet_get_settings, + .set_settings = usbnet_set_settings, .get_drvinfo = usbnet_get_drvinfo, .get_link = usbnet_get_link, + .nway_reset = usbnet_nway_reset, .get_msglevel = usbnet_get_msglevel, .set_msglevel = usbnet_set_msglevel, }; diff --git a/drivers/usb/net/usbnet.h b/drivers/usb/net/usbnet.h index c0746f0..743947c 100644 --- a/drivers/usb/net/usbnet.h +++ b/drivers/usb/net/usbnet.h @@ -168,9 +168,13 @@ extern void usbnet_defer_kevent (struct usbnet *, int); extern void usbnet_skb_return (struct usbnet *, struct sk_buff *); extern void usbnet_unlink_rx_urbs(struct usbnet *); +extern int usbnet_get_settings (struct net_device *net, struct ethtool_cmd *cmd); +extern int usbnet_set_settings (struct net_device *net, struct ethtool_cmd *cmd); +extern u32 usbnet_get_link (struct net_device *net); extern u32 usbnet_get_msglevel (struct net_device *); extern void usbnet_set_msglevel (struct net_device *, u32); extern void usbnet_get_drvinfo (struct net_device *, struct ethtool_drvinfo *); +extern int usbnet_nway_reset(struct net_device *net); /* messaging support includes the interface name, so it must not be * used before it has one ... notably, in minidriver bind() calls. -- cgit v0.10.2 From a9fc6338bd51a3d5735839e756fe7b741c2e6fad Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 9 Oct 2006 00:08:02 +0200 Subject: usbnet: add a mutex around phy register access When working on the mcs7830, I noticed the need for a mutex in its mdio_read/mdio_write functions. A related problem seems to be present in the asix driver in the respective functions. This introduces a mutex in the common usbnet driver and uses it from the two hardware specific drivers. Acked-by: David Hollis Signed-off-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/asix.c b/drivers/usb/net/asix.c index 5edd053..881841e 100644 --- a/drivers/usb/net/asix.c +++ b/drivers/usb/net/asix.c @@ -569,10 +569,12 @@ static int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) struct usbnet *dev = netdev_priv(netdev); u16 res; + mutex_lock(&dev->phy_mutex); asix_set_sw_mii(dev); asix_read_cmd(dev, AX_CMD_READ_MII_REG, phy_id, (__u16)loc, 2, (u16 *)&res); asix_set_hw_mii(dev); + mutex_unlock(&dev->phy_mutex); devdbg(dev, "asix_mdio_read() phy_id=0x%02x, loc=0x%02x, returns=0x%04x", phy_id, loc, le16_to_cpu(res & 0xffff)); @@ -586,10 +588,12 @@ asix_mdio_write(struct net_device *netdev, int phy_id, int loc, int val) u16 res = cpu_to_le16(val); devdbg(dev, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x", phy_id, loc, val); + mutex_lock(&dev->phy_mutex); asix_set_sw_mii(dev); asix_write_cmd(dev, AX_CMD_WRITE_MII_REG, phy_id, (__u16)loc, 2, (u16 *)&res); asix_set_hw_mii(dev); + mutex_unlock(&dev->phy_mutex); } /* Get the PHY Identifier from the PHYSID1 & PHYSID2 MII registers */ diff --git a/drivers/usb/net/mcs7830.c b/drivers/usb/net/mcs7830.c index 23a8066..6240b97 100644 --- a/drivers/usb/net/mcs7830.c +++ b/drivers/usb/net/mcs7830.c @@ -184,6 +184,7 @@ static int mcs7830_read_phy(struct usbnet *dev, u8 index) HIF_REG_PHY_CMD2_PEND_FLAG_BIT | index, }; + mutex_lock(&dev->phy_mutex); /* write the MII command */ ret = mcs7830_set_reg(dev, HIF_REG_PHY_CMD1, 2, cmd); if (ret < 0) @@ -208,6 +209,7 @@ static int mcs7830_read_phy(struct usbnet *dev, u8 index) dev_dbg(&dev->udev->dev, "read PHY reg %02x: %04x (%d tries)\n", index, val, i); out: + mutex_unlock(&dev->phy_mutex); return ret; } @@ -222,6 +224,8 @@ static int mcs7830_write_phy(struct usbnet *dev, u8 index, u16 val) HIF_REG_PHY_CMD2_PEND_FLAG_BIT | (index & 0x1F), }; + mutex_lock(&dev->phy_mutex); + /* write the new register contents */ le_val = cpu_to_le16(val); ret = mcs7830_set_reg(dev, HIF_REG_PHY_DATA, 2, &le_val); @@ -248,6 +252,7 @@ static int mcs7830_write_phy(struct usbnet *dev, u8 index, u16 val) dev_dbg(&dev->udev->dev, "write PHY reg %02x: %04x (%d tries)\n", index, val, i); out: + mutex_unlock(&dev->phy_mutex); return ret; } diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index decc1b1..cf3d20e 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -1144,6 +1144,7 @@ usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod) dev->delay.function = usbnet_bh; dev->delay.data = (unsigned long) dev; init_timer (&dev->delay); + mutex_init (&dev->phy_mutex); SET_MODULE_OWNER (net); dev->net = net; diff --git a/drivers/usb/net/usbnet.h b/drivers/usb/net/usbnet.h index 743947c..07c70ab 100644 --- a/drivers/usb/net/usbnet.h +++ b/drivers/usb/net/usbnet.h @@ -30,6 +30,7 @@ struct usbnet { struct usb_device *udev; struct driver_info *driver_info; wait_queue_head_t *wait; + struct mutex phy_mutex; /* i/o info: pipes etc */ unsigned in, out; -- cgit v0.10.2 From 9fcde235270e6783600d1aee5bcda78c8282bcdd Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 10 Oct 2006 13:47:35 -0700 Subject: USB: move trancevibrator.c to the proper usb directory It's not a input driver, so it doesn't belong in the input directory. Cc: Sam Hocevar Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile index 97d57cf..825bf88 100644 --- a/drivers/usb/Makefile +++ b/drivers/usb/Makefile @@ -33,7 +33,6 @@ obj-$(CONFIG_USB_KBTAB) += input/ obj-$(CONFIG_USB_MOUSE) += input/ obj-$(CONFIG_USB_MTOUCH) += input/ obj-$(CONFIG_USB_POWERMATE) += input/ -obj-$(CONFIG_USB_TRANCEVIBRATOR)+= input/ obj-$(CONFIG_USB_WACOM) += input/ obj-$(CONFIG_USB_XPAD) += input/ @@ -66,6 +65,7 @@ obj-$(CONFIG_USB_PHIDGETSERVO) += misc/ obj-$(CONFIG_USB_RIO500) += misc/ obj-$(CONFIG_USB_SISUSBVGA) += misc/ obj-$(CONFIG_USB_TEST) += misc/ +obj-$(CONFIG_USB_TRANCEVIBRATOR)+= misc/ obj-$(CONFIG_USB_USS720) += misc/ obj-$(CONFIG_USB_ATM) += atm/ diff --git a/drivers/usb/input/Kconfig b/drivers/usb/input/Kconfig index 21cd226..20db364 100644 --- a/drivers/usb/input/Kconfig +++ b/drivers/usb/input/Kconfig @@ -348,13 +348,3 @@ config USB_APPLETOUCH To compile this driver as a module, choose M here: the module will be called appletouch. - -config USB_TRANCEVIBRATOR - tristate "PlayStation 2 Trance Vibrator driver support" - depends on USB - help - Say Y here if you want to connect a PlayStation 2 Trance Vibrator - device to your computer's USB port. - - To compile this driver as a module, choose M here: the - module will be called trancevibrator. diff --git a/drivers/usb/input/Makefile b/drivers/usb/input/Makefile index 71437db..d946d52 100644 --- a/drivers/usb/input/Makefile +++ b/drivers/usb/input/Makefile @@ -48,7 +48,6 @@ obj-$(CONFIG_USB_ACECAD) += acecad.o obj-$(CONFIG_USB_YEALINK) += yealink.o obj-$(CONFIG_USB_XPAD) += xpad.o obj-$(CONFIG_USB_APPLETOUCH) += appletouch.o -obj-$(CONFIG_USB_TRANCEVIBRATOR) += trancevibrator.o ifeq ($(CONFIG_USB_DEBUG),y) EXTRA_CFLAGS += -DDEBUG diff --git a/drivers/usb/input/trancevibrator.c b/drivers/usb/input/trancevibrator.c deleted file mode 100644 index 33cd91d..0000000 --- a/drivers/usb/input/trancevibrator.c +++ /dev/null @@ -1,159 +0,0 @@ -/* - * PlayStation 2 Trance Vibrator driver - * - * Copyright (C) 2006 Sam Hocevar - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - */ - -/* Standard include files */ -#include -#include -#include -#include -#include - -/* Version Information */ -#define DRIVER_VERSION "v1.1" -#define DRIVER_AUTHOR "Sam Hocevar, sam@zoy.org" -#define DRIVER_DESC "PlayStation 2 Trance Vibrator driver" - -#define TRANCEVIBRATOR_VENDOR_ID 0x0b49 /* ASCII Corporation */ -#define TRANCEVIBRATOR_PRODUCT_ID 0x064f /* Trance Vibrator */ - -static struct usb_device_id id_table [] = { - { USB_DEVICE(TRANCEVIBRATOR_VENDOR_ID, TRANCEVIBRATOR_PRODUCT_ID) }, - { }, -}; -MODULE_DEVICE_TABLE (usb, id_table); - -/* Driver-local specific stuff */ -struct trancevibrator { - struct usb_device *udev; - unsigned int speed; -}; - -static ssize_t show_speed(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct usb_interface *intf = to_usb_interface(dev); - struct trancevibrator *tv = usb_get_intfdata(intf); - - return sprintf(buf, "%d\n", tv->speed); -} - -static ssize_t set_speed(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct usb_interface *intf = to_usb_interface(dev); - struct trancevibrator *tv = usb_get_intfdata(intf); - int temp, retval; - - temp = simple_strtoul(buf, NULL, 10); - if (temp > 255) - temp = 255; - else if (temp < 0) - temp = 0; - tv->speed = temp; - - dev_dbg(&tv->udev->dev, "speed = %d\n", tv->speed); - - /* Set speed */ - retval = usb_control_msg(tv->udev, usb_sndctrlpipe(tv->udev, 0), - 0x01, /* vendor request: set speed */ - USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_OTHER, - tv->speed, /* speed value */ - 0, NULL, 0, USB_CTRL_GET_TIMEOUT); - if (retval) { - dev_dbg(&tv->udev->dev, "retval = %d\n", retval); - return retval; - } - return count; -} - -static DEVICE_ATTR(speed, S_IWUGO | S_IRUGO, show_speed, set_speed); - -static int tv_probe(struct usb_interface *interface, - const struct usb_device_id *id) -{ - struct usb_device *udev = interface_to_usbdev(interface); - struct trancevibrator *dev; - int retval; - - dev = kzalloc(sizeof(struct trancevibrator), GFP_KERNEL); - if (dev == NULL) { - dev_err(&interface->dev, "Out of memory\n"); - retval = -ENOMEM; - goto error; - } - - dev->udev = usb_get_dev(udev); - usb_set_intfdata(interface, dev); - retval = device_create_file(&interface->dev, &dev_attr_speed); - if (retval) - goto error_create_file; - - return 0; - -error_create_file: - usb_put_dev(udev); - usb_set_intfdata(interface, NULL); -error: - kfree(dev); - return retval; -} - -static void tv_disconnect(struct usb_interface *interface) -{ - struct trancevibrator *dev; - - dev = usb_get_intfdata (interface); - usb_set_intfdata(interface, NULL); - device_remove_file(&interface->dev, &dev_attr_speed); - usb_put_dev(dev->udev); - kfree(dev); -} - -/* USB subsystem object */ -static struct usb_driver tv_driver = { - .name = "trancevibrator", - .probe = tv_probe, - .disconnect = tv_disconnect, - .id_table = id_table, -}; - -static int __init tv_init(void) -{ - int retval = usb_register(&tv_driver); - if (retval) { - err("usb_register failed. Error number %d", retval); - return retval; - } - - info(DRIVER_VERSION ":" DRIVER_DESC); - return 0; -} - -static void __exit tv_exit(void) -{ - usb_deregister(&tv_driver); -} - -module_init (tv_init); -module_exit (tv_exit); - -MODULE_AUTHOR(DRIVER_AUTHOR); -MODULE_DESCRIPTION(DRIVER_DESC); -MODULE_LICENSE("GPL"); diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig index c29658f..a74bf86 100644 --- a/drivers/usb/misc/Kconfig +++ b/drivers/usb/misc/Kconfig @@ -223,6 +223,16 @@ config USB_LD To compile this driver as a module, choose M here: the module will be called ldusb. +config USB_TRANCEVIBRATOR + tristate "PlayStation 2 Trance Vibrator driver support" + depends on USB + help + Say Y here if you want to connect a PlayStation 2 Trance Vibrator + device to your computer's USB port. + + To compile this driver as a module, choose M here: the + module will be called trancevibrator. + config USB_TEST tristate "USB testing driver (DEVELOPMENT)" depends on USB && USB_DEVICEFS && EXPERIMENTAL diff --git a/drivers/usb/misc/Makefile b/drivers/usb/misc/Makefile index 2be70fa..11dc595 100644 --- a/drivers/usb/misc/Makefile +++ b/drivers/usb/misc/Makefile @@ -21,6 +21,7 @@ obj-$(CONFIG_USB_PHIDGETMOTORCONTROL) += phidgetmotorcontrol.o obj-$(CONFIG_USB_PHIDGETSERVO) += phidgetservo.o obj-$(CONFIG_USB_RIO500) += rio500.o obj-$(CONFIG_USB_TEST) += usbtest.o +obj-$(CONFIG_USB_TRANCEVIBRATOR) += trancevibrator.o obj-$(CONFIG_USB_USS720) += uss720.o obj-$(CONFIG_USB_SISUSBVGA) += sisusbvga/ diff --git a/drivers/usb/misc/trancevibrator.c b/drivers/usb/misc/trancevibrator.c new file mode 100644 index 0000000..33cd91d --- /dev/null +++ b/drivers/usb/misc/trancevibrator.c @@ -0,0 +1,159 @@ +/* + * PlayStation 2 Trance Vibrator driver + * + * Copyright (C) 2006 Sam Hocevar + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +/* Standard include files */ +#include +#include +#include +#include +#include + +/* Version Information */ +#define DRIVER_VERSION "v1.1" +#define DRIVER_AUTHOR "Sam Hocevar, sam@zoy.org" +#define DRIVER_DESC "PlayStation 2 Trance Vibrator driver" + +#define TRANCEVIBRATOR_VENDOR_ID 0x0b49 /* ASCII Corporation */ +#define TRANCEVIBRATOR_PRODUCT_ID 0x064f /* Trance Vibrator */ + +static struct usb_device_id id_table [] = { + { USB_DEVICE(TRANCEVIBRATOR_VENDOR_ID, TRANCEVIBRATOR_PRODUCT_ID) }, + { }, +}; +MODULE_DEVICE_TABLE (usb, id_table); + +/* Driver-local specific stuff */ +struct trancevibrator { + struct usb_device *udev; + unsigned int speed; +}; + +static ssize_t show_speed(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct usb_interface *intf = to_usb_interface(dev); + struct trancevibrator *tv = usb_get_intfdata(intf); + + return sprintf(buf, "%d\n", tv->speed); +} + +static ssize_t set_speed(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct usb_interface *intf = to_usb_interface(dev); + struct trancevibrator *tv = usb_get_intfdata(intf); + int temp, retval; + + temp = simple_strtoul(buf, NULL, 10); + if (temp > 255) + temp = 255; + else if (temp < 0) + temp = 0; + tv->speed = temp; + + dev_dbg(&tv->udev->dev, "speed = %d\n", tv->speed); + + /* Set speed */ + retval = usb_control_msg(tv->udev, usb_sndctrlpipe(tv->udev, 0), + 0x01, /* vendor request: set speed */ + USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_OTHER, + tv->speed, /* speed value */ + 0, NULL, 0, USB_CTRL_GET_TIMEOUT); + if (retval) { + dev_dbg(&tv->udev->dev, "retval = %d\n", retval); + return retval; + } + return count; +} + +static DEVICE_ATTR(speed, S_IWUGO | S_IRUGO, show_speed, set_speed); + +static int tv_probe(struct usb_interface *interface, + const struct usb_device_id *id) +{ + struct usb_device *udev = interface_to_usbdev(interface); + struct trancevibrator *dev; + int retval; + + dev = kzalloc(sizeof(struct trancevibrator), GFP_KERNEL); + if (dev == NULL) { + dev_err(&interface->dev, "Out of memory\n"); + retval = -ENOMEM; + goto error; + } + + dev->udev = usb_get_dev(udev); + usb_set_intfdata(interface, dev); + retval = device_create_file(&interface->dev, &dev_attr_speed); + if (retval) + goto error_create_file; + + return 0; + +error_create_file: + usb_put_dev(udev); + usb_set_intfdata(interface, NULL); +error: + kfree(dev); + return retval; +} + +static void tv_disconnect(struct usb_interface *interface) +{ + struct trancevibrator *dev; + + dev = usb_get_intfdata (interface); + usb_set_intfdata(interface, NULL); + device_remove_file(&interface->dev, &dev_attr_speed); + usb_put_dev(dev->udev); + kfree(dev); +} + +/* USB subsystem object */ +static struct usb_driver tv_driver = { + .name = "trancevibrator", + .probe = tv_probe, + .disconnect = tv_disconnect, + .id_table = id_table, +}; + +static int __init tv_init(void) +{ + int retval = usb_register(&tv_driver); + if (retval) { + err("usb_register failed. Error number %d", retval); + return retval; + } + + info(DRIVER_VERSION ":" DRIVER_DESC); + return 0; +} + +static void __exit tv_exit(void) +{ + usb_deregister(&tv_driver); +} + +module_init (tv_init); +module_exit (tv_exit); + +MODULE_AUTHOR(DRIVER_AUTHOR); +MODULE_DESCRIPTION(DRIVER_DESC); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 0f64478cbc7a008fe7b7e9ae79a73d8a6904ead8 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 9 Apr 2002 12:14:34 -0700 Subject: USB: add USB serial mos7720 driver Add support for Moschip 7720 USB dual port usb to serial device. This driver is originally based on the drivers/usb/io_edgeport.c driver. Cleaned up and forward ported by me. Cc: VijayaKumar Cc: AjayKumar Cc: Gurudeva Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig index 5076b9d..8ca6d3f 100644 --- a/drivers/usb/serial/Kconfig +++ b/drivers/usb/serial/Kconfig @@ -422,6 +422,16 @@ config USB_SERIAL_MCT_U232 To compile this driver as a module, choose M here: the module will be called mct_u232. +config USB_SERIAL_MOS7720 + tristate "USB Moschip 7720 Single Port Serial Driver" + depends on USB_SERIAL + ---help--- + Say Y here if you want to use a USB Serial single port adapter from + Moschip Semiconductor Tech. + + To compile this driver as a module, choose M here: the + module will be called mos7720. + config USB_SERIAL_MOS7840 tristate "USB Moschip 7840/7820 USB Serial Driver" depends on USB_SERIAL diff --git a/drivers/usb/serial/Makefile b/drivers/usb/serial/Makefile index 8dce833..a5047dc 100644 --- a/drivers/usb/serial/Makefile +++ b/drivers/usb/serial/Makefile @@ -34,6 +34,7 @@ obj-$(CONFIG_USB_SERIAL_KEYSPAN_PDA) += keyspan_pda.o obj-$(CONFIG_USB_SERIAL_KLSI) += kl5kusb105.o obj-$(CONFIG_USB_SERIAL_KOBIL_SCT) += kobil_sct.o obj-$(CONFIG_USB_SERIAL_MCT_U232) += mct_u232.o +obj-$(CONFIG_USB_SERIAL_MOS7720) += mos7720.o obj-$(CONFIG_USB_SERIAL_MOS7840) += mos7840.o obj-$(CONFIG_USB_SERIAL_NAVMAN) += navman.o obj-$(CONFIG_USB_SERIAL_OMNINET) += omninet.o diff --git a/drivers/usb/serial/mos7720.c b/drivers/usb/serial/mos7720.c new file mode 100644 index 0000000..82cd15b --- /dev/null +++ b/drivers/usb/serial/mos7720.c @@ -0,0 +1,1683 @@ +/* + * mos7720.c + * Controls the Moschip 7720 usb to dual port serial convertor + * + * Copyright 2006 Moschip Semiconductor Tech. Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, version 2 of the License. + * + * Developed by: + * VijayaKumar.G.N. + * AjayKumar + * Gurudeva.N. + * + * Cleaned up from the original by: + * Greg Kroah-Hartman + * + * Originally based on drivers/usb/serial/io_edgeport.c which is: + * Copyright (C) 2000 Inside Out Networks, All rights reserved. + * Copyright (C) 2001-2002 Greg Kroah-Hartman + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +/* + * Version Information + */ +#define DRIVER_VERSION "1.0.0.4F" +#define DRIVER_AUTHOR "Aspire Communications pvt Ltd." +#define DRIVER_DESC "Moschip USB Serial Driver" + +/* default urb timeout */ +#define MOS_WDR_TIMEOUT (HZ * 5) + +#define MOS_PORT1 0x0200 +#define MOS_PORT2 0x0300 +#define MOS_VENREG 0x0000 +#define MOS_MAX_PORT 0x02 +#define MOS_WRITE 0x0E +#define MOS_READ 0x0D + +/* Interrupt Rotinue Defines */ +#define SERIAL_IIR_RLS 0x06 +#define SERIAL_IIR_RDA 0x04 +#define SERIAL_IIR_CTI 0x0c +#define SERIAL_IIR_THR 0x02 +#define SERIAL_IIR_MS 0x00 + +#define NUM_URBS 16 /* URB Count */ +#define URB_TRANSFER_BUFFER_SIZE 32 /* URB Size */ + +/* This structure holds all of the local port information */ +struct moschip_port +{ + __u8 shadowLCR; /* last LCR value received */ + __u8 shadowMCR; /* last MCR value received */ + __u8 shadowMSR; /* last MSR value received */ + char open; + struct async_icount icount; + struct usb_serial_port *port; /* loop back to the owner */ + struct urb *write_urb_pool[NUM_URBS]; +}; + +/* This structure holds all of the individual serial device information */ +struct moschip_serial +{ + int interrupt_started; +}; + +static int debug; + +#define USB_VENDOR_ID_MOSCHIP 0x9710 +#define MOSCHIP_DEVICE_ID_7720 0x7720 +#define MOSCHIP_DEVICE_ID_7715 0x7715 + +static struct usb_device_id moschip_port_id_table [] = { + { USB_DEVICE(USB_VENDOR_ID_MOSCHIP,MOSCHIP_DEVICE_ID_7720) }, + { } /* terminating entry */ +}; +MODULE_DEVICE_TABLE(usb, moschip_port_id_table); + + +/* + * mos7720_interrupt_callback + * this is the callback function for when we have received data on the + * interrupt endpoint. + */ +static void mos7720_interrupt_callback(struct urb *urb) +{ + int result; + int length; + __u32 *data; + unsigned int status; + __u8 sp1; + __u8 sp2; + __u8 st; + + dbg("%s"," : Entering\n"); + + if (!urb) { + dbg("%s","Invalid Pointer !!!!:\n"); + return; + } + + switch (urb->status) { + case 0: + /* success */ + break; + case -ECONNRESET: + case -ENOENT: + case -ESHUTDOWN: + /* this urb is terminated, clean up */ + dbg("%s - urb shutting down with status: %d", __FUNCTION__, + urb->status); + return; + default: + dbg("%s - nonzero urb status received: %d", __FUNCTION__, + urb->status); + goto exit; + } + + length = urb->actual_length; + data = urb->transfer_buffer; + + /* Moschip get 4 bytes + * Byte 1 IIR Port 1 (port.number is 0) + * Byte 2 IIR Port 2 (port.number is 1) + * Byte 3 -------------- + * Byte 4 FIFO status for both */ + if (length && length > 4) { + dbg("Wrong data !!!"); + return; + } + + status = *data; + + sp1 = (status & 0xff000000)>>24; + sp2 = (status & 0x00ff0000)>>16; + st = status & 0x000000ff; + + if ((sp1 & 0x01) || (sp2 & 0x01)) { + /* No Interrupt Pending in both the ports */ + dbg("No Interrupt !!!"); + } else { + switch (sp1 & 0x0f) { + case SERIAL_IIR_RLS: + dbg("Serial Port 1: Receiver status error or address " + "bit detected in 9-bit mode\n"); + break; + case SERIAL_IIR_CTI: + dbg("Serial Port 1: Receiver time out"); + break; + case SERIAL_IIR_MS: + dbg("Serial Port 1: Modem status change"); + break; + } + + switch (sp2 & 0x0f) { + case SERIAL_IIR_RLS: + dbg("Serial Port 2: Receiver status error or address " + "bit detected in 9-bit mode"); + break; + case SERIAL_IIR_CTI: + dbg("Serial Port 2: Receiver time out"); + break; + case SERIAL_IIR_MS: + dbg("Serial Port 2: Modem status change"); + break; + } + } + +exit: + result = usb_submit_urb(urb, GFP_ATOMIC); + if (result) + dev_err(&urb->dev->dev, + "%s - Error %d submitting control urb\n", + __FUNCTION__, result); + return; +} + +/* + * mos7720_bulk_in_callback + * this is the callback function for when we have received data on the + * bulk in endpoint. + */ +static void mos7720_bulk_in_callback(struct urb *urb) +{ + int status; + unsigned char *data ; + struct usb_serial_port *port; + struct moschip_port *mos7720_port; + struct tty_struct *tty; + + if (urb->status) { + dbg("nonzero read bulk status received: %d",urb->status); + return; + } + + mos7720_port = urb->context; + if (!mos7720_port) { + dbg("%s","NULL mos7720_port pointer \n"); + return ; + } + + port = mos7720_port->port; + + dbg("Entering...%s", __FUNCTION__); + + data = urb->transfer_buffer; + + tty = port->tty; + if (tty && urb->actual_length) { + tty_buffer_request_room(tty, urb->actual_length); + tty_insert_flip_string(tty, data, urb->actual_length); + tty_flip_buffer_push(tty); + } + + if (!port->read_urb) { + dbg("URB KILLED !!!"); + return; + } + + if (port->read_urb->status != -EINPROGRESS) { + port->read_urb->dev = port->serial->dev; + + status = usb_submit_urb(port->read_urb, GFP_ATOMIC); + if (status) + dbg("usb_submit_urb(read bulk) failed, status = %d", + status); + } +} + +/* + * mos7720_bulk_out_data_callback + * this is the callback function for when we have finished sending serial + * data on the bulk out endpoint. + */ +static void mos7720_bulk_out_data_callback(struct urb *urb) +{ + struct moschip_port *mos7720_port; + struct tty_struct *tty; + + if (urb->status) { + dbg("nonzero write bulk status received:%d", urb->status); + return; + } + + mos7720_port = urb->context; + if (!mos7720_port) { + dbg("NULL mos7720_port pointer"); + return ; + } + + dbg("Entering ........."); + + tty = mos7720_port->port->tty; + + if (tty && mos7720_port->open) { + /* let the tty driver wakeup if it has a special * + * write_wakeup function */ + if ((tty->flags & (1 << TTY_DO_WRITE_WAKEUP)) && + tty->ldisc.write_wakeup) + (tty->ldisc.write_wakeup)(tty); + + /* tell the tty driver that something has changed */ + wake_up_interruptible(&tty->write_wait); + } + + /* schedule_work(&mos7720_port->port->work); */ +} + +/* + * send_mos_cmd + * this function will be used for sending command to device + */ +static int send_mos_cmd(struct usb_serial *serial, __u8 request, __u16 value, + __u16 index, void *data) +{ + int status; + unsigned int pipe; + u16 product = le16_to_cpu(serial->dev->descriptor.idProduct); + __u8 requesttype; + __u16 size = 0x0000; + + if (value < MOS_MAX_PORT) { + if (product == MOSCHIP_DEVICE_ID_7715) { + value = value*0x100+0x100; + } else { + value = value*0x100+0x200; + } + } else { + value = 0x0000; + if ((product == MOSCHIP_DEVICE_ID_7715) && + (index != 0x08)) { + dbg("serial->product== MOSCHIP_DEVICE_ID_7715"); + //index = 0x01 ; + } + } + + if (request == MOS_WRITE) { + request = (__u8)MOS_WRITE; + requesttype = (__u8)0x40; + value = value + (__u16)*((unsigned char *)data); + data = NULL; + pipe = usb_sndctrlpipe(serial->dev, 0); + } else { + request = (__u8)MOS_READ; + requesttype = (__u8)0xC0; + size = 0x01; + pipe = usb_rcvctrlpipe(serial->dev,0); + } + + status = usb_control_msg(serial->dev, pipe, request, requesttype, + value, index, data, size, MOS_WDR_TIMEOUT); + + if (status < 0) + dbg("Command Write failed Value %x index %x\n",value,index); + + return status; +} + +static int mos7720_open(struct usb_serial_port *port, struct file * filp) +{ + struct usb_serial *serial; + struct usb_serial_port *port0; + struct urb *urb; + struct moschip_serial *mos7720_serial; + struct moschip_port *mos7720_port; + int response; + int port_number; + char data; + int j; + + serial = port->serial; + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) + return -ENODEV; + + port0 = serial->port[0]; + + mos7720_serial = usb_get_serial_data(serial); + + if (mos7720_serial == NULL || port0 == NULL) + return -ENODEV; + + usb_clear_halt(serial->dev, port->write_urb->pipe); + usb_clear_halt(serial->dev, port->read_urb->pipe); + + /* Initialising the write urb pool */ + for (j = 0; j < NUM_URBS; ++j) { + urb = usb_alloc_urb(0,SLAB_ATOMIC); + mos7720_port->write_urb_pool[j] = urb; + + if (urb == NULL) { + err("No more urbs???"); + continue; + } + + urb->transfer_buffer = kmalloc(URB_TRANSFER_BUFFER_SIZE, + GFP_KERNEL); + if (!urb->transfer_buffer) { + err("%s-out of memory for urb buffers.", __FUNCTION__); + continue; + } + } + + /* Initialize MCS7720 -- Write Init values to corresponding Registers + * + * Register Index + * 1 : IER + * 2 : FCR + * 3 : LCR + * 4 : MCR + * + * 0x08 : SP1/2 Control Reg + */ + port_number = port->number - port->serial->minor; + send_mos_cmd(port->serial, MOS_READ, port_number, UART_LSR, &data); + dbg("SS::%p LSR:%x\n",mos7720_port, data); + + dbg("Check:Sending Command .........."); + + data = 0x02; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, 0x01, &data); + data = 0x02; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, 0x02, &data); + + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data); + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x02, &data); + + data = 0xCF; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x02, &data); + data = 0x03; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x03, &data); + data = 0x0b; + mos7720_port->shadowMCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + data = 0x0b; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + + data = 0x00; + send_mos_cmd(serial, MOS_READ, MOS_MAX_PORT, 0x08, &data); + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, 0x08, &data); + +/* data = 0x00; + send_mos_cmd(serial, MOS_READ, MOS_MAX_PORT, port_number + 1, &data); + data = 0x03; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, port_number + 1, &data); + data = 0x00; + send_mos_cmd(port->serial, MOS_WRITE, MOS_MAX_PORT, port_number + 1, &data); +*/ + data = 0x00; + send_mos_cmd(serial, MOS_READ, MOS_MAX_PORT, 0x08, &data); + + data = data | (port->number - port->serial->minor + 1); + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, 0x08, &data); + + data = 0x83; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x03, &data); + data = 0x0c; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x00, &data); + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data); + data = 0x03; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x03, &data); + data = 0x0c; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data); + data = 0x0c; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data); + +//Matrix + + /* force low_latency on so that our tty_push actually forces * + * the data through,otherwise it is scheduled, and with * + * high data rates (like with OHCI) data can get lost. */ + + if (port->tty) + port->tty->low_latency = 1; + + /* see if we've set up our endpoint info yet * + * (can't set it up in mos7720_startup as the * + * structures were not set up at that time.) */ + if (!mos7720_serial->interrupt_started) { + dbg("Interrupt buffer NULL !!!"); + + /* not set up yet, so do it now */ + mos7720_serial->interrupt_started = 1; + + dbg("To Submit URB !!!"); + + /* set up our interrupt urb */ + usb_fill_int_urb(port0->interrupt_in_urb, serial->dev, + usb_rcvintpipe(serial->dev, + port->interrupt_in_endpointAddress), + port0->interrupt_in_buffer, + port0->interrupt_in_urb->transfer_buffer_length, + mos7720_interrupt_callback, mos7720_port, + port0->interrupt_in_urb->interval); + + /* start interrupt read for this mos7720 this interrupt * + * will continue as long as the mos7720 is connected */ + dbg("Submit URB over !!!"); + response = usb_submit_urb(port0->interrupt_in_urb, GFP_KERNEL); + if (response) + dev_err(&port->dev, + "%s - Error %d submitting control urb", + __FUNCTION__, response); + } + + /* set up our bulk in urb */ + usb_fill_bulk_urb(port->read_urb, serial->dev, + usb_rcvbulkpipe(serial->dev, + port->bulk_in_endpointAddress), + port->bulk_in_buffer, + port->read_urb->transfer_buffer_length, + mos7720_bulk_in_callback, mos7720_port); + response = usb_submit_urb(port->read_urb, GFP_KERNEL); + if (response) + dev_err(&port->dev, + "%s - Error %d submitting read urb", __FUNCTION__, response); + + /* initialize our icount structure */ + memset(&(mos7720_port->icount), 0x00, sizeof(mos7720_port->icount)); + + /* initialize our port settings */ + mos7720_port->shadowMCR = UART_MCR_OUT2; /* Must set to enable ints! */ + + /* send a open port command */ + mos7720_port->open = 1; + + return 0; +} + +/* + * mos7720_chars_in_buffer + * this function is called by the tty driver when it wants to know how many + * bytes of data we currently have outstanding in the port (data that has + * been written, but hasn't made it out the port yet) + * If successful, we return the number of bytes left to be written in the + * system, + * Otherwise we return a negative error number. + */ +static int mos7720_chars_in_buffer(struct usb_serial_port *port) +{ + int i; + int chars = 0; + struct moschip_port *mos7720_port; + + dbg("%s:entering ...........", __FUNCTION__); + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) { + dbg("%s:leaving ...........", __FUNCTION__); + return -ENODEV; + } + + for (i = 0; i < NUM_URBS; ++i) { + if (mos7720_port->write_urb_pool[i]->status == -EINPROGRESS) + chars += URB_TRANSFER_BUFFER_SIZE; + } + dbg("%s - returns %d", __FUNCTION__, chars); + return chars; +} + +static void mos7720_close(struct usb_serial_port *port, struct file *filp) +{ + struct usb_serial *serial; + struct moschip_port *mos7720_port; + char data; + int j; + + dbg("mos7720_close:entering..."); + + serial = port->serial; + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) + return; + + for (j = 0; j < NUM_URBS; ++j) + usb_kill_urb(mos7720_port->write_urb_pool[j]); + + /* Freeing Write URBs */ + for (j = 0; j < NUM_URBS; ++j) { + if (mos7720_port->write_urb_pool[j]) { + kfree(mos7720_port->write_urb_pool[j]->transfer_buffer); + usb_free_urb(mos7720_port->write_urb_pool[j]); + } + } + + /* While closing port, shutdown all bulk read, write * + * and interrupt read if they exists */ + if (serial->dev) { + dbg("Shutdown bulk write"); + usb_kill_urb(port->write_urb); + dbg("Shutdown bulk read"); + usb_kill_urb(port->read_urb); + } + + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, port->number - port->serial->minor, + 0x04, &data); + + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, port->number - port->serial->minor, + 0x01, &data); + + mos7720_port->open = 0; + + dbg("Leaving %s", __FUNCTION__); +} + +static void mos7720_break(struct usb_serial_port *port, int break_state) +{ + unsigned char data; + struct usb_serial *serial; + struct moschip_port *mos7720_port; + + dbg("Entering %s", __FUNCTION__); + + serial = port->serial; + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) + return; + + if (break_state == -1) + data = mos7720_port->shadowLCR | UART_LCR_SBC; + else + data = mos7720_port->shadowLCR & ~UART_LCR_SBC; + + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, port->number - port->serial->minor, + 0x03, &data); + + return; +} + +/* + * mos7720_write_room + * this function is called by the tty driver when it wants to know how many + * bytes of data we can accept for a specific port. + * If successful, we return the amount of room that we have for this port + * Otherwise we return a negative error number. + */ +static int mos7720_write_room(struct usb_serial_port *port) +{ + struct moschip_port *mos7720_port; + int room = 0; + int i; + + dbg("%s:entering ...........", __FUNCTION__); + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) { + dbg("%s:leaving ...........", __FUNCTION__); + return -ENODEV; + } + + for (i = 0; i < NUM_URBS; ++i) { + if (mos7720_port->write_urb_pool[i]->status != -EINPROGRESS) + room += URB_TRANSFER_BUFFER_SIZE; + } + + dbg("%s - returns %d", __FUNCTION__, room); + return room; +} + +static int mos7720_write(struct usb_serial_port *port, + const unsigned char *data, int count) +{ + int status; + int i; + int bytes_sent = 0; + int transfer_size; + + struct moschip_port *mos7720_port; + struct usb_serial *serial; + struct urb *urb; + const unsigned char *current_position = data; + + dbg("%s:entering ...........", __FUNCTION__); + + serial = port->serial; + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) { + dbg("mos7720_port is NULL"); + return -ENODEV; + } + + /* try to find a free urb in the list */ + urb = NULL; + + for (i = 0; i < NUM_URBS; ++i) { + if (mos7720_port->write_urb_pool[i]->status != -EINPROGRESS) { + urb = mos7720_port->write_urb_pool[i]; + dbg("URB:%d",i); + break; + } + } + + if (urb == NULL) { + dbg("%s - no more free urbs", __FUNCTION__); + goto exit; + } + + if (urb->transfer_buffer == NULL) { + urb->transfer_buffer = kmalloc(URB_TRANSFER_BUFFER_SIZE, + GFP_KERNEL); + if (urb->transfer_buffer == NULL) { + err("%s no more kernel memory...", __FUNCTION__); + goto exit; + } + } + transfer_size = min (count, URB_TRANSFER_BUFFER_SIZE); + + memcpy(urb->transfer_buffer, current_position, transfer_size); + usb_serial_debug_data(debug, &port->dev, __FUNCTION__, transfer_size, + urb->transfer_buffer); + + /* fill urb with data and submit */ + usb_fill_bulk_urb(urb, serial->dev, + usb_sndbulkpipe(serial->dev, + port->bulk_out_endpointAddress), + urb->transfer_buffer, transfer_size, + mos7720_bulk_out_data_callback, mos7720_port); + + /* send it down the pipe */ + status = usb_submit_urb(urb,GFP_ATOMIC); + if (status) { + err("%s - usb_submit_urb(write bulk) failed with status = %d", + __FUNCTION__, status); + bytes_sent = status; + goto exit; + } + bytes_sent = transfer_size; + +exit: + return bytes_sent; +} + +static void mos7720_throttle(struct usb_serial_port *port) +{ + struct moschip_port *mos7720_port; + struct tty_struct *tty; + int status; + + dbg("%s- port %d\n", __FUNCTION__, port->number); + + mos7720_port = usb_get_serial_port_data(port); + + if (mos7720_port == NULL) + return; + + if (!mos7720_port->open) { + dbg("port not opened"); + return; + } + + dbg("%s: Entering ..........", __FUNCTION__); + + tty = port->tty; + if (!tty) { + dbg("%s - no tty available", __FUNCTION__); + return; + } + + /* if we are implementing XON/XOFF, send the stop character */ + if (I_IXOFF(tty)) { + unsigned char stop_char = STOP_CHAR(tty); + status = mos7720_write(port, &stop_char, 1); + if (status <= 0) + return; + } + + /* if we are implementing RTS/CTS, toggle that line */ + if (tty->termios->c_cflag & CRTSCTS) { + mos7720_port->shadowMCR &= ~UART_MCR_RTS; + status = send_mos_cmd(port->serial, MOS_WRITE, + port->number - port->serial->minor, + UART_MCR, &mos7720_port->shadowMCR); + if (status != 0) + return; + } +} + +static void mos7720_unthrottle(struct usb_serial_port *port) +{ + struct tty_struct *tty; + int status; + struct moschip_port *mos7720_port = usb_get_serial_port_data(port); + + if (mos7720_port == NULL) + return; + + if (!mos7720_port->open) { + dbg("%s - port not opened", __FUNCTION__); + return; + } + + dbg("%s: Entering ..........", __FUNCTION__); + + tty = port->tty; + if (!tty) { + dbg("%s - no tty available", __FUNCTION__); + return; + } + + /* if we are implementing XON/XOFF, send the start character */ + if (I_IXOFF(tty)) { + unsigned char start_char = START_CHAR(tty); + status = mos7720_write(port, &start_char, 1); + if (status <= 0) + return; + } + + /* if we are implementing RTS/CTS, toggle that line */ + if (tty->termios->c_cflag & CRTSCTS) { + mos7720_port->shadowMCR |= UART_MCR_RTS; + status = send_mos_cmd(port->serial, MOS_WRITE, + port->number - port->serial->minor, + UART_MCR, &mos7720_port->shadowMCR); + if (status != 0) + return; + } +} + +static int set_higher_rates(struct moschip_port *mos7720_port, + unsigned int baud) +{ + unsigned char data; + struct usb_serial_port *port; + struct usb_serial *serial; + int port_number; + + if (mos7720_port == NULL) + return -EINVAL; + + port = mos7720_port->port; + serial = port->serial; + + /*********************************************** + * Init Sequence for higher rates + ***********************************************/ + dbg("Sending Setting Commands .........."); + port_number = port->number - port->serial->minor; + + data = 0x000; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data); + data = 0x000; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x02, &data); + data = 0x0CF; + send_mos_cmd(serial, MOS_WRITE, port->number, 0x02, &data); + data = 0x00b; + mos7720_port->shadowMCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + data = 0x00b; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + + data = 0x000; + send_mos_cmd(serial, MOS_READ, MOS_MAX_PORT, 0x08, &data); + data = 0x000; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, 0x08, &data); + + + /*********************************************** + * Set for higher rates * + ***********************************************/ + + data = baud * 0x10; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, port_number + 1,&data); + + data = 0x003; + send_mos_cmd(serial, MOS_READ, MOS_MAX_PORT, 0x08, &data); + data = 0x003; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, 0x08, &data); + + data = 0x02b; + mos7720_port->shadowMCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + data = 0x02b; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + + /*********************************************** + * Set DLL/DLM + ***********************************************/ + + data = mos7720_port->shadowLCR | UART_LCR_DLAB; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x03, &data); + + data = 0x001; /* DLL */ + send_mos_cmd(serial, MOS_WRITE, port_number, 0x00, &data); + data = 0x000; /* DLM */ + send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data); + + data = mos7720_port->shadowLCR & ~UART_LCR_DLAB; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x03, &data); + + return 0; +} + +/* baud rate information */ +struct divisor_table_entry +{ + __u32 baudrate; + __u16 divisor; +}; + +/* Define table of divisors for moschip 7720 hardware * + * These assume a 3.6864MHz crystal, the standard /16, and * + * MCR.7 = 0. */ +static struct divisor_table_entry divisor_table[] = { + { 50, 2304}, + { 110, 1047}, /* 2094.545455 => 230450 => .0217 % over */ + { 134, 857}, /* 1713.011152 => 230398.5 => .00065% under */ + { 150, 768}, + { 300, 384}, + { 600, 192}, + { 1200, 96}, + { 1800, 64}, + { 2400, 48}, + { 4800, 24}, + { 7200, 16}, + { 9600, 12}, + { 19200, 6}, + { 38400, 3}, + { 57600, 2}, + { 115200, 1}, +}; + +/***************************************************************************** + * calc_baud_rate_divisor + * this function calculates the proper baud rate divisor for the specified + * baud rate. + *****************************************************************************/ +static int calc_baud_rate_divisor(int baudrate, int *divisor) +{ + int i; + __u16 custom; + __u16 round1; + __u16 round; + + + dbg("%s - %d", __FUNCTION__, baudrate); + + for (i = 0; i < ARRAY_SIZE(divisor_table); i++) { + if (divisor_table[i].baudrate == baudrate) { + *divisor = divisor_table[i].divisor; + return 0; + } + } + + /* After trying for all the standard baud rates * + * Try calculating the divisor for this baud rate */ + if (baudrate > 75 && baudrate < 230400) { + /* get the divisor */ + custom = (__u16)(230400L / baudrate); + + /* Check for round off */ + round1 = (__u16)(2304000L / baudrate); + round = (__u16)(round1 - (custom * 10)); + if (round > 4) + custom++; + *divisor = custom; + + dbg("Baud %d = %d",baudrate, custom); + return 0; + } + + dbg("Baud calculation Failed..."); + return -EINVAL; +} + +/* + * send_cmd_write_baud_rate + * this function sends the proper command to change the baud rate of the + * specified port. + */ +static int send_cmd_write_baud_rate(struct moschip_port *mos7720_port, + int baudrate) +{ + struct usb_serial_port *port; + struct usb_serial *serial; + int divisor; + int status; + unsigned char data; + unsigned char number; + + if (mos7720_port == NULL) + return -1; + + port = mos7720_port->port; + serial = port->serial; + + dbg("%s: Entering ..........", __FUNCTION__); + + number = port->number - port->serial->minor; + dbg("%s - port = %d, baud = %d", __FUNCTION__, port->number, baudrate); + + /* Calculate the Divisor */ + status = calc_baud_rate_divisor(baudrate, &divisor); + if (status) { + err("%s - bad baud rate", __FUNCTION__); + return status; + } + + /* Enable access to divisor latch */ + data = mos7720_port->shadowLCR | UART_LCR_DLAB; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, number, UART_LCR, &data); + + /* Write the divisor */ + data = ((unsigned char)(divisor & 0xff)); + send_mos_cmd(serial, MOS_WRITE, number, 0x00, &data); + + data = ((unsigned char)((divisor & 0xff00) >> 8)); + send_mos_cmd(serial, MOS_WRITE, number, 0x01, &data); + + /* Disable access to divisor latch */ + data = mos7720_port->shadowLCR & ~UART_LCR_DLAB; + mos7720_port->shadowLCR = data; + send_mos_cmd(serial, MOS_WRITE, number, 0x03, &data); + + return status; +} + +/* + * change_port_settings + * This routine is called to set the UART on the device to match + * the specified new settings. + */ +static void change_port_settings(struct moschip_port *mos7720_port, + struct termios *old_termios) +{ + struct usb_serial_port *port; + struct usb_serial *serial; + struct tty_struct *tty; + int baud; + unsigned cflag; + unsigned iflag; + __u8 mask = 0xff; + __u8 lData; + __u8 lParity; + __u8 lStop; + int status; + int port_number; + char data; + + if (mos7720_port == NULL) + return ; + + port = mos7720_port->port; + serial = port->serial; + port_number = port->number - port->serial->minor; + + dbg("%s - port %d", __FUNCTION__, port->number); + + if (!mos7720_port->open) { + dbg("%s - port not opened", __FUNCTION__); + return; + } + + tty = mos7720_port->port->tty; + + if ((!tty) || (!tty->termios)) { + dbg("%s - no tty structures", __FUNCTION__); + return; + } + + dbg("%s: Entering ..........", __FUNCTION__); + + lData = UART_LCR_WLEN8; + lStop = 0x00; /* 1 stop bit */ + lParity = 0x00; /* No parity */ + + cflag = tty->termios->c_cflag; + iflag = tty->termios->c_iflag; + + /* Change the number of bits */ + switch (cflag & CSIZE) { + case CS5: + lData = UART_LCR_WLEN5; + mask = 0x1f; + break; + + case CS6: + lData = UART_LCR_WLEN6; + mask = 0x3f; + break; + + case CS7: + lData = UART_LCR_WLEN7; + mask = 0x7f; + break; + default: + case CS8: + lData = UART_LCR_WLEN8; + break; + } + + /* Change the Parity bit */ + if (cflag & PARENB) { + if (cflag & PARODD) { + lParity = UART_LCR_PARITY; + dbg("%s - parity = odd", __FUNCTION__); + } else { + lParity = (UART_LCR_EPAR | UART_LCR_PARITY); + dbg("%s - parity = even", __FUNCTION__); + } + + } else { + dbg("%s - parity = none", __FUNCTION__); + } + + if (cflag & CMSPAR) + lParity = lParity | 0x20; + + /* Change the Stop bit */ + if (cflag & CSTOPB) { + lStop = UART_LCR_STOP; + dbg("%s - stop bits = 2", __FUNCTION__); + } else { + lStop = 0x00; + dbg("%s - stop bits = 1", __FUNCTION__); + } + +#define LCR_BITS_MASK 0x03 /* Mask for bits/char field */ +#define LCR_STOP_MASK 0x04 /* Mask for stop bits field */ +#define LCR_PAR_MASK 0x38 /* Mask for parity field */ + + /* Update the LCR with the correct value */ + mos7720_port->shadowLCR &= ~(LCR_BITS_MASK | LCR_STOP_MASK | LCR_PAR_MASK); + mos7720_port->shadowLCR |= (lData | lParity | lStop); + + + /* Disable Interrupts */ + data = 0x00; + send_mos_cmd(serial,MOS_WRITE,port->number - port->serial->minor, UART_IER, &data); + + data = 0x00; + send_mos_cmd(serial, MOS_WRITE, port_number, UART_FCR, &data); + + data = 0xcf; + send_mos_cmd(serial, MOS_WRITE, port_number, UART_FCR, &data); + + /* Send the updated LCR value to the mos7720 */ + data = mos7720_port->shadowLCR; + send_mos_cmd(serial, MOS_WRITE, port_number, UART_LCR, &data); + + data = 0x00b; + mos7720_port->shadowMCR = data; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + data = 0x00b; + send_mos_cmd(serial, MOS_WRITE, port_number, 0x04, &data); + + /* set up the MCR register and send it to the mos7720 */ + mos7720_port->shadowMCR = UART_MCR_OUT2; + if (cflag & CBAUD) + mos7720_port->shadowMCR |= (UART_MCR_DTR | UART_MCR_RTS); + + if (cflag & CRTSCTS) { + mos7720_port->shadowMCR |= (UART_MCR_XONANY); + + /* To set hardware flow control to the specified * + * serial port, in SP1/2_CONTROL_REG */ + if (port->number) { + data = 0x001; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, + 0x08, &data); + } else { + data = 0x002; + send_mos_cmd(serial, MOS_WRITE, MOS_MAX_PORT, + 0x08, &data); + } + } else { + mos7720_port->shadowMCR &= ~(UART_MCR_XONANY); + } + + data = mos7720_port->shadowMCR; + send_mos_cmd(serial, MOS_WRITE, port_number, UART_MCR, &data); + + /* Determine divisor based on baud rate */ + baud = tty_get_baud_rate(tty); + if (!baud) { + /* pick a default, any default... */ + dbg("Picked default baud..."); + baud = 9600; + } + + if (baud >= 230400) { + set_higher_rates(mos7720_port, baud); + /* Enable Interrupts */ + data = 0x0c; + send_mos_cmd(serial, MOS_WRITE, port_number, UART_IER, &data); + return; + } + + dbg("%s - baud rate = %d", __FUNCTION__, baud); + status = send_cmd_write_baud_rate(mos7720_port, baud); + + /* Enable Interrupts */ + data = 0x0c; + send_mos_cmd(serial, MOS_WRITE, port_number, UART_IER, &data); + + if (port->read_urb->status != -EINPROGRESS) { + port->read_urb->dev = serial->dev; + + status = usb_submit_urb(port->read_urb, GFP_ATOMIC); + if (status) + dbg("usb_submit_urb(read bulk) failed, status = %d", + status); + } + return; +} + +/* + * mos7720_set_termios + * this function is called by the tty driver when it wants to change the + * termios structure. + */ +static void mos7720_set_termios(struct usb_serial_port *port, + struct termios *old_termios) +{ + int status; + unsigned int cflag; + struct usb_serial *serial; + struct moschip_port *mos7720_port; + struct tty_struct *tty; + + serial = port->serial; + + mos7720_port = usb_get_serial_port_data(port); + + if (mos7720_port == NULL) + return; + + tty = port->tty; + + if (!port->tty || !port->tty->termios) { + dbg("%s - no tty or termios", __FUNCTION__); + return; + } + + if (!mos7720_port->open) { + dbg("%s - port not opened", __FUNCTION__); + return; + } + + dbg("%s\n","setting termios - ASPIRE"); + + cflag = tty->termios->c_cflag; + + if (!cflag) { + printk("%s %s\n",__FUNCTION__,"cflag is NULL"); + return; + } + + /* check that they really want us to change something */ + if (old_termios) { + if ((cflag == old_termios->c_cflag) && + (RELEVANT_IFLAG(tty->termios->c_iflag) == + RELEVANT_IFLAG(old_termios->c_iflag))) { + dbg("Nothing to change"); + return; + } + } + + dbg("%s - clfag %08x iflag %08x", __FUNCTION__, + tty->termios->c_cflag, + RELEVANT_IFLAG(tty->termios->c_iflag)); + + if (old_termios) + dbg("%s - old clfag %08x old iflag %08x", __FUNCTION__, + old_termios->c_cflag, + RELEVANT_IFLAG(old_termios->c_iflag)); + + dbg("%s - port %d", __FUNCTION__, port->number); + + /* change the port settings to the new ones specified */ + change_port_settings(mos7720_port, old_termios); + + if(!port->read_urb) { + dbg("%s","URB KILLED !!!!!\n"); + return; + } + + if(port->read_urb->status != -EINPROGRESS) { + port->read_urb->dev = serial->dev; + status = usb_submit_urb(port->read_urb, GFP_ATOMIC); + if (status) + dbg("usb_submit_urb(read bulk) failed, status = %d", + status); + } + return; +} + +/* + * get_lsr_info - get line status register info + * + * Purpose: Let user call ioctl() to get info when the UART physically + * is emptied. On bus types like RS485, the transmitter must + * release the bus after transmitting. This must be done when + * the transmit shift register is empty, not be done when the + * transmit holding register is empty. This functionality + * allows an RS485 driver to be written in user space. + */ +static int get_lsr_info(struct moschip_port *mos7720_port, + unsigned int __user *value) +{ + int count; + unsigned int result = 0; + + count = mos7720_chars_in_buffer(mos7720_port->port); + if (count == 0) { + dbg("%s -- Empty", __FUNCTION__); + result = TIOCSER_TEMT; + } + + if (copy_to_user(value, &result, sizeof(int))) + return -EFAULT; + return 0; +} + +/* + * get_number_bytes_avail - get number of bytes available + * + * Purpose: Let user call ioctl to get the count of number of bytes available. + */ +static int get_number_bytes_avail(struct moschip_port *mos7720_port, + unsigned int __user *value) +{ + unsigned int result = 0; + struct tty_struct *tty = mos7720_port->port->tty; + + if (!tty) + return -ENOIOCTLCMD; + + result = tty->read_cnt; + + dbg("%s(%d) = %d", __FUNCTION__, mos7720_port->port->number, result); + if (copy_to_user(value, &result, sizeof(int))) + return -EFAULT; + + return -ENOIOCTLCMD; +} + +static int set_modem_info(struct moschip_port *mos7720_port, unsigned int cmd, + unsigned int __user *value) +{ + unsigned int mcr ; + unsigned int arg; + unsigned char data; + + struct usb_serial_port *port; + + if (mos7720_port == NULL) + return -1; + + port = (struct usb_serial_port*)mos7720_port->port; + mcr = mos7720_port->shadowMCR; + + if (copy_from_user(&arg, value, sizeof(int))) + return -EFAULT; + + switch (cmd) { + case TIOCMBIS: + if (arg & TIOCM_RTS) + mcr |= UART_MCR_RTS; + if (arg & TIOCM_DTR) + mcr |= UART_MCR_RTS; + if (arg & TIOCM_LOOP) + mcr |= UART_MCR_LOOP; + break; + + case TIOCMBIC: + if (arg & TIOCM_RTS) + mcr &= ~UART_MCR_RTS; + if (arg & TIOCM_DTR) + mcr &= ~UART_MCR_RTS; + if (arg & TIOCM_LOOP) + mcr &= ~UART_MCR_LOOP; + break; + + case TIOCMSET: + /* turn off the RTS and DTR and LOOPBACK + * and then only turn on what was asked to */ + mcr &= ~(UART_MCR_RTS | UART_MCR_DTR | UART_MCR_LOOP); + mcr |= ((arg & TIOCM_RTS) ? UART_MCR_RTS : 0); + mcr |= ((arg & TIOCM_DTR) ? UART_MCR_DTR : 0); + mcr |= ((arg & TIOCM_LOOP) ? UART_MCR_LOOP : 0); + break; + } + + mos7720_port->shadowMCR = mcr; + + data = mos7720_port->shadowMCR; + send_mos_cmd(port->serial, MOS_WRITE, + port->number - port->serial->minor, UART_MCR, &data); + + return 0; +} + +static int get_modem_info(struct moschip_port *mos7720_port, + unsigned int __user *value) +{ + unsigned int result = 0; + unsigned int msr = mos7720_port->shadowMSR; + unsigned int mcr = mos7720_port->shadowMCR; + + result = ((mcr & UART_MCR_DTR) ? TIOCM_DTR: 0) /* 0x002 */ + | ((mcr & UART_MCR_RTS) ? TIOCM_RTS: 0) /* 0x004 */ + | ((msr & UART_MSR_CTS) ? TIOCM_CTS: 0) /* 0x020 */ + | ((msr & UART_MSR_DCD) ? TIOCM_CAR: 0) /* 0x040 */ + | ((msr & UART_MSR_RI) ? TIOCM_RI: 0) /* 0x080 */ + | ((msr & UART_MSR_DSR) ? TIOCM_DSR: 0); /* 0x100 */ + + + dbg("%s -- %x", __FUNCTION__, result); + + if (copy_to_user(value, &result, sizeof(int))) + return -EFAULT; + return 0; +} + +static int get_serial_info(struct moschip_port *mos7720_port, + struct serial_struct __user *retinfo) +{ + struct serial_struct tmp; + + if (!retinfo) + return -EFAULT; + + memset(&tmp, 0, sizeof(tmp)); + + tmp.type = PORT_16550A; + tmp.line = mos7720_port->port->serial->minor; + tmp.port = mos7720_port->port->number; + tmp.irq = 0; + tmp.flags = ASYNC_SKIP_TEST | ASYNC_AUTO_IRQ; + tmp.xmit_fifo_size = NUM_URBS * URB_TRANSFER_BUFFER_SIZE; + tmp.baud_base = 9600; + tmp.close_delay = 5*HZ; + tmp.closing_wait = 30*HZ; + + if (copy_to_user(retinfo, &tmp, sizeof(*retinfo))) + return -EFAULT; + return 0; +} + +static int mos7720_ioctl(struct usb_serial_port *port, struct file *file, + unsigned int cmd, unsigned long arg) +{ + struct moschip_port *mos7720_port; + struct async_icount cnow; + struct async_icount cprev; + struct serial_icounter_struct icount; + + mos7720_port = usb_get_serial_port_data(port); + if (mos7720_port == NULL) + return -ENODEV; + + dbg("%s - port %d, cmd = 0x%x", __FUNCTION__, port->number, cmd); + + switch (cmd) { + case TIOCINQ: + /* return number of bytes available */ + dbg("%s (%d) TIOCINQ", __FUNCTION__, port->number); + return get_number_bytes_avail(mos7720_port, + (unsigned int __user *)arg); + break; + + case TIOCSERGETLSR: + dbg("%s (%d) TIOCSERGETLSR", __FUNCTION__, port->number); + return get_lsr_info(mos7720_port, (unsigned int __user *)arg); + return 0; + + case TIOCMBIS: + case TIOCMBIC: + case TIOCMSET: + dbg("%s (%d) TIOCMSET/TIOCMBIC/TIOCMSET", __FUNCTION__, + port->number); + return set_modem_info(mos7720_port, cmd, + (unsigned int __user *)arg); + + case TIOCMGET: + dbg("%s (%d) TIOCMGET", __FUNCTION__, port->number); + return get_modem_info(mos7720_port, + (unsigned int __user *)arg); + + case TIOCGSERIAL: + dbg("%s (%d) TIOCGSERIAL", __FUNCTION__, port->number); + return get_serial_info(mos7720_port, + (struct serial_struct __user *)arg); + + case TIOCSSERIAL: + dbg("%s (%d) TIOCSSERIAL", __FUNCTION__, port->number); + break; + + case TIOCMIWAIT: + dbg("%s (%d) TIOCMIWAIT", __FUNCTION__, port->number); + cprev = mos7720_port->icount; + while (1) { + if (signal_pending(current)) + return -ERESTARTSYS; + cnow = mos7720_port->icount; + if (cnow.rng == cprev.rng && cnow.dsr == cprev.dsr && + cnow.dcd == cprev.dcd && cnow.cts == cprev.cts) + return -EIO; /* no change => error */ + if (((arg & TIOCM_RNG) && (cnow.rng != cprev.rng)) || + ((arg & TIOCM_DSR) && (cnow.dsr != cprev.dsr)) || + ((arg & TIOCM_CD) && (cnow.dcd != cprev.dcd)) || + ((arg & TIOCM_CTS) && (cnow.cts != cprev.cts)) ) { + return 0; + } + cprev = cnow; + } + /* NOTREACHED */ + break; + + case TIOCGICOUNT: + cnow = mos7720_port->icount; + icount.cts = cnow.cts; + icount.dsr = cnow.dsr; + icount.rng = cnow.rng; + icount.dcd = cnow.dcd; + icount.rx = cnow.rx; + icount.tx = cnow.tx; + icount.frame = cnow.frame; + icount.overrun = cnow.overrun; + icount.parity = cnow.parity; + icount.brk = cnow.brk; + icount.buf_overrun = cnow.buf_overrun; + + dbg("%s (%d) TIOCGICOUNT RX=%d, TX=%d", __FUNCTION__, + port->number, icount.rx, icount.tx ); + if (copy_to_user((void __user *)arg, &icount, sizeof(icount))) + return -EFAULT; + return 0; + } + + return -ENOIOCTLCMD; +} + +static int mos7720_startup(struct usb_serial *serial) +{ + struct moschip_serial *mos7720_serial; + struct moschip_port *mos7720_port; + struct usb_device *dev; + int i; + char data; + + dbg("%s: Entering ..........", __FUNCTION__); + + if (!serial) { + dbg("Invalid Handler"); + return -ENODEV; + } + + dev = serial->dev; + + /* create our private serial structure */ + mos7720_serial = kzalloc(sizeof(struct moschip_serial), GFP_KERNEL); + if (mos7720_serial == NULL) { + err("%s - Out of memory", __FUNCTION__); + return -ENOMEM; + } + + usb_set_serial_data(serial, mos7720_serial); + + /* we set up the pointers to the endpoints in the mos7720_open * + * function, as the structures aren't created yet. */ + + /* set up port private structures */ + for (i = 0; i < serial->num_ports; ++i) { + mos7720_port = kzalloc(sizeof(struct moschip_port), GFP_KERNEL); + if (mos7720_port == NULL) { + err("%s - Out of memory", __FUNCTION__); + usb_set_serial_data(serial, NULL); + kfree(mos7720_serial); + return -ENOMEM; + } + + /* Initialize all port interrupt end point to port 0 int + * endpoint. Our device has only one interrupt endpoint + * comman to all ports */ + serial->port[i]->interrupt_in_endpointAddress = serial->port[0]->interrupt_in_endpointAddress; + + mos7720_port->port = serial->port[i]; + usb_set_serial_port_data(serial->port[i], mos7720_port); + + dbg("port number is %d", serial->port[i]->number); + dbg("serial number is %d", serial->minor); + } + + + /* setting configuration feature to one */ + usb_control_msg(serial->dev, usb_sndctrlpipe(serial->dev, 0), + (__u8)0x03, 0x00,0x01,0x00, NULL, 0x00, 5*HZ); + + send_mos_cmd(serial,MOS_READ,0x00, UART_LSR, &data); // LSR For Port 1 + dbg("LSR:%x",data); + + send_mos_cmd(serial,MOS_READ,0x01, UART_LSR, &data); // LSR For Port 2 + dbg("LSR:%x",data); + + return 0; +} + +static void mos7720_shutdown(struct usb_serial *serial) +{ + int i; + + /* free private structure allocated for serial port */ + for (i=0; i < serial->num_ports; ++i) { + kfree(usb_get_serial_port_data(serial->port[i])); + usb_set_serial_port_data(serial->port[i], NULL); + } + + /* free private structure allocated for serial device */ + kfree(usb_get_serial_data(serial)); + usb_set_serial_data(serial, NULL); +} + +static struct usb_serial_driver moschip7720_2port_driver = { + .driver = { + .owner = THIS_MODULE, + .name = "moschip7720", + }, + .description = "Moschip 2 port adapter", + .id_table = moschip_port_id_table, + .num_interrupt_in = 1, + .num_bulk_in = 2, + .num_bulk_out = 2, + .num_ports = 2, + .open = mos7720_open, + .close = mos7720_close, + .throttle = mos7720_throttle, + .unthrottle = mos7720_unthrottle, + .attach = mos7720_startup, + .shutdown = mos7720_shutdown, + .ioctl = mos7720_ioctl, + .set_termios = mos7720_set_termios, + .write = mos7720_write, + .write_room = mos7720_write_room, + .chars_in_buffer = mos7720_chars_in_buffer, + .break_ctl = mos7720_break, + .read_bulk_callback = mos7720_bulk_in_callback, +}; + +static struct usb_driver usb_driver = { + .name = "moschip7720", + .probe = usb_serial_probe, + .disconnect = usb_serial_disconnect, + .id_table = moschip_port_id_table, +}; + +static int __init moschip7720_init(void) +{ + int retval; + + dbg("%s: Entering ..........", __FUNCTION__); + + /* Register with the usb serial */ + retval = usb_serial_register(&moschip7720_2port_driver); + if (retval) + goto failed_port_device_register; + + info(DRIVER_DESC " " DRIVER_VERSION); + + /* Register with the usb */ + retval = usb_register(&usb_driver); + if (retval) + goto failed_usb_register; + + return 0; + +failed_usb_register: + usb_serial_deregister(&moschip7720_2port_driver); + +failed_port_device_register: + return retval; +} + +static void __exit moschip7720_exit(void) +{ + usb_deregister(&usb_driver); + usb_serial_deregister(&moschip7720_2port_driver); +} + +module_init(moschip7720_init); +module_exit(moschip7720_exit); + +/* Module information */ +MODULE_AUTHOR( DRIVER_AUTHOR ); +MODULE_DESCRIPTION( DRIVER_DESC ); +MODULE_LICENSE("GPL"); + +module_param(debug, bool, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(debug, "Debug enabled or not"); -- cgit v0.10.2 From a65dc301c7448a9a8d24bf1cbfe292541d1fa390 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Fri, 6 Oct 2006 00:09:29 +0200 Subject: USB: fix dereference in drivers/usb/misc/adutux.c in two of the error cases, dev is still NULL, and we dereference it. Spotted by coverity (cid#1428, 1429) Signed-off-by: Eric Sesterhenn Cc: Randy Dunlap Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/misc/adutux.c b/drivers/usb/misc/adutux.c index aecd633..af2934e 100644 --- a/drivers/usb/misc/adutux.c +++ b/drivers/usb/misc/adutux.c @@ -370,7 +370,8 @@ static int adu_release(struct inode *inode, struct file *file) retval = adu_release_internal(dev); exit: - up(&dev->sem); + if (dev) + up(&dev->sem); dbg(2," %s : leave, return value %d", __FUNCTION__, retval); return retval; } -- cgit v0.10.2 From 1ff15e8efc1703eaae1eeec6fc09db6af1e4049f Mon Sep 17 00:00:00 2001 From: Tobias Lorenz Date: Sun, 8 Oct 2006 22:56:40 -0700 Subject: USB: Mitsumi USB FDD 061M: UNUSUAL_DEV multilun fix From: Tobias Lorenz Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 0a846e4..0093d9f 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -55,7 +55,8 @@ UNUSUAL_DEV( 0x03eb, 0x2002, 0x0100, 0x0100, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_IGNORE_RESIDUE), -UNUSUAL_DEV( 0x03ee, 0x6901, 0x0000, 0x0100, +/* modified by Tobias Lorenz */ +UNUSUAL_DEV( 0x03ee, 0x6901, 0x0000, 0x0200, "Mitsumi", "USB FDD", US_SC_DEVICE, US_PR_DEVICE, NULL, -- cgit v0.10.2 From c19ecd654209725444d1f47a4422e6f48846b53c Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 9 Oct 2006 01:16:24 +0200 Subject: USB: ftdi-elan.c: remove dead code The Coverity checker spotted this obviously dead code. Signed-off-by: Adrian Bunk Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/misc/ftdi-elan.c b/drivers/usb/misc/ftdi-elan.c index 37d1f4e..9b591b8 100644 --- a/drivers/usb/misc/ftdi-elan.c +++ b/drivers/usb/misc/ftdi-elan.c @@ -513,8 +513,6 @@ static void ftdi_elan_respond_work(void *data) ftdi->disconnected += 1; } else if (retval == -ENODEV) { ftdi->disconnected += 1; - } else if (retval == -ENODEV) { - ftdi->disconnected += 1; } else if (retval == -EILSEQ) { ftdi->disconnected += 1; } else { -- cgit v0.10.2 From ad18027f4909c8fc107056460c97dbedb6635128 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 9 Oct 2006 01:16:32 +0200 Subject: USB: mos7840.c: fix a check-after-dereference This patch fixes an obvious check-after-dereference spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c index 021be39..5b71962 100644 --- a/drivers/usb/serial/mos7840.c +++ b/drivers/usb/serial/mos7840.c @@ -2413,11 +2413,12 @@ static int mos7840_ioctl(struct usb_serial_port *port, struct file *file, } mos7840_port = mos7840_get_port_private(port); - tty = mos7840_port->port->tty; if (mos7840_port == NULL) return -1; + tty = mos7840_port->port->tty; + dbg("%s - port %d, cmd = 0x%x", __FUNCTION__, port->number, cmd); switch (cmd) { -- cgit v0.10.2 From 3ccf25ce185d4798e66a91812a7622f7fe6987df Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Fri, 13 Oct 2006 09:59:17 -0400 Subject: USB: unusual_devs entry for Nokia 6234 This patch (as803) adds an unusual_devs entry for the Nokia 6234 mobile phone. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 0093d9f..802f3a3 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -190,6 +190,13 @@ UNUSUAL_DEV( 0x0421, 0x047c, 0x0370, 0x0370, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_MAX_SECTORS_64 ), +/* Reported by Alex Corcoles */ +UNUSUAL_DEV( 0x0421, 0x0495, 0x0370, 0x0370, + "Nokia", + "6234", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_MAX_SECTORS_64 ), + /* Reported by Olaf Hering from novell bug #105878 */ UNUSUAL_DEV( 0x0424, 0x0fdc, 0x0210, 0x0210, "SMSC", -- cgit v0.10.2 From 521b600b58376b7c85a7c615ee32fae185c20b16 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 10 Oct 2006 14:42:46 -0700 Subject: USB: fix usbatm tiny race ia64: drivers/usb/atm/usbatm.c: In function `usbatm_do_heavy_init': drivers/usb/atm/usbatm.c:1004: warning: implicit declaration of function `get_current' drivers/usb/atm/usbatm.c:1004: error: invalid type argument of `->' Signed-off-by: Duncan Sands Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/usbatm.c b/drivers/usb/atm/usbatm.c index ab091fa..ec63b0e 100644 --- a/drivers/usb/atm/usbatm.c +++ b/drivers/usb/atm/usbatm.c @@ -1001,7 +1001,7 @@ static int usbatm_do_heavy_init(void *arg) daemonize(instance->driver->driver_name); allow_signal(SIGTERM); - instance->thread_pid = get_current()->pid; + instance->thread_pid = current->pid; complete(&instance->thread_started); -- cgit v0.10.2 From c0fc0ee06f6c9ab37f53afc62b0d94a700fa7a97 Mon Sep 17 00:00:00 2001 From: Jan Mate Date: Tue, 10 Oct 2006 14:42:47 -0700 Subject: USB Storage: unusual_devs.h entry for Sony Ericsson P990i USB Storage: this patch adds support for Sony Ericsson P990i Signed-off-by: Jan Mate Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 802f3a3..37ed8e0 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1306,6 +1306,13 @@ UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), +/* Reported by Jan Mate */ +UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, + "Sony Ericsson", + "P990i", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + /* Reported by Kevin Cernekee * Tested on hardware version 1.10. * Entry is needed only for the initializer function override. -- cgit v0.10.2 From 0e185b7922ac81516c5c4653dcf6aacbf6341e73 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Tue, 10 Oct 2006 14:42:50 -0700 Subject: USB: Memory leak in drivers/usb/serial/airprime.c the commit http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=5dda171202f94127e49c12daf780cdae1b4e668b added a memory leak. In case we cant allocate an urb, we dont free the buffer and leak it. Coverity id #1438 Signed-off-by: Eric Sesterhenn Acked-by: Andy Gay Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/airprime.c b/drivers/usb/serial/airprime.c index 392a512..ba93c72 100644 --- a/drivers/usb/serial/airprime.c +++ b/drivers/usb/serial/airprime.c @@ -134,6 +134,7 @@ static int airprime_open(struct usb_serial_port *port, struct file *filp) } urb = usb_alloc_urb(0, GFP_KERNEL); if (!urb) { + kfree(buffer); dev_err(&port->dev, "%s - no more urbs?\n", __FUNCTION__); result = -ENOMEM; -- cgit v0.10.2 From 4550718f6c75c9abe8b987fa4c625fd041aa95a2 Mon Sep 17 00:00:00 2001 From: Grant Grundler Date: Tue, 10 Oct 2006 14:42:51 -0700 Subject: USB: input: extract() and implement() are bit field manipulation routines extract() and implement() have brain damaged attempts to handle 32-bit wide "fields". The problem is the index math in the original code didn't clear all the relevant bits. (offset >> 5) only compensated for 32-bit index. We need (offset >> 6) if we want to use 64-bit loads. But it was also wrong in that it tried to use quasi-aligned loads. Ie "report" was only incremented in multiples of 4 bytes and then the offset was masked off for values greater than 4 bytes. The right way is to pretend "report" points at a byte array. And offset is then only minor adjustment for < 8 bits of offset. "n" (field width) can then be as big as 24 (assuming 32-bit loads) since "offset" will never be bigger than 7. If someone needs either function to handle more than 24-bits, please document why - point at a specification or specific USB hid device - in comments in the code. extract/implement() are also an eyesore to read. Please banish whoever wrote it to read CodingStyle 3 times in a row to a classroom full of 1st graders armed with rubberbands. Or just flame them. Whatever. Globbing all the code together on two lines does NOT make it faster and is Just Wrong. I've tested this patch on j6000 (dual 750Mhz PA-RISC, 32-bit 2.6.12-rc5). Kyle McMartin tested on c3000 (up 400Mhz PA-RISC, same kernel). "p2-mate" (Peter De Schrijver?) tested on sb1250 (dual core Mips, broadcom "swarm" eval board). Signed-off-by: Grant Grundler Signed-off-by: Matthew Wilcox Cc: Vojtech Pavlik Cc: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index a6738a8..feabda7 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -750,21 +750,31 @@ static __inline__ __u32 s32ton(__s32 value, unsigned n) } /* - * Extract/implement a data field from/to a report. + * Extract/implement a data field from/to a little endian report (bit array). */ static __inline__ __u32 extract(__u8 *report, unsigned offset, unsigned n) { - report += (offset >> 5) << 2; offset &= 31; - return (le64_to_cpu(get_unaligned((__le64*)report)) >> offset) & ((1ULL << n) - 1); + u32 x; + + report += offset >> 3; /* adjust byte index */ + offset &= 8 - 1; + x = get_unaligned((u32 *) report); + x = le32_to_cpu(x); + x = (x >> offset) & ((1 << n) - 1); + return x; } static __inline__ void implement(__u8 *report, unsigned offset, unsigned n, __u32 value) { - report += (offset >> 5) << 2; offset &= 31; - put_unaligned((get_unaligned((__le64*)report) - & cpu_to_le64(~((((__u64) 1 << n) - 1) << offset))) - | cpu_to_le64((__u64)value << offset), (__le64*)report); + u32 x; + + report += offset >> 3; + offset &= 8 - 1; + x = get_unaligned((u32 *)report); + x &= cpu_to_le32(~((((__u32) 1 << n) - 1) << offset)); + x |= cpu_to_le32(value << offset); + put_unaligned(x,(u32 *)report); } /* -- cgit v0.10.2 From deb8ee43a23d48116cb23eb8dd1de2348efb1e80 Mon Sep 17 00:00:00 2001 From: Dominic Cerquetti Date: Tue, 10 Oct 2006 14:42:48 -0700 Subject: USB: xpad: dance pad support Adds support for dance pads to the xpad driver. Dance pads require the d-pad to be mapped to four buttons instead of two axes, so that combinations of up/down and left/right can be hit simultaneously. Known dance pads are detected, and there is a module parameter added to default unknown xpad devices to map the d-pad to buttons if this is desired. (dpad_to_buttons). Minor modifications were made to port the changes in the original patch to a newer kernel version. This patch was originally from Dominic Cerquetti originally written for kernel 2.6.11.4, with minor modifications (API changes for USB, spelling fixes to the documentation added in the original patch) made to apply to the current kernel. I have modified Dominic's original patch per some suggestions from Dmitry Torokhov. (There was nothing in the patch format description about multiple From: lines, so I haven't added myself.) [akpm@osdl.org: cleanups] Signed-off-by: Adam Buchbinder Acked-by: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/input/xpad.txt b/Documentation/input/xpad.txt index b9111a7..5427bdf 100644 --- a/Documentation/input/xpad.txt +++ b/Documentation/input/xpad.txt @@ -3,20 +3,37 @@ xpad - Linux USB driver for X-Box gamepads This is the very first release of a driver for X-Box gamepads. Basically, this was hacked away in just a few hours, so don't expect miracles. + In particular, there is currently NO support for the rumble pack. You won't find many ff-aware linux applications anyway. -0. Status ---------- +0. Notes +-------- + +Driver updated for kernel 2.6.17.11. (Based on a patch for 2.6.11.4.) -For now, this driver has only been tested on just one Linux-Box. -This one is running a 2.4.18 kernel with usb-uhci on an amd athlon 600. +The number of buttons/axes reported varies based on 3 things: +- if you are using a known controller +- if you are using a known dance pad +- if using an unknown device (one not listed below), what you set in the + module configuration for "Map D-PAD to buttons rather than axes for unknown + pads" (module option dpad_to_buttons) -The jstest-program from joystick-1.2.15 (jstest-version 2.1.0) reports -8 axes and 10 buttons. +If you set dpad_to_buttons to 0 and you are using an unknown device (one +not listed below), the driver will map the directional pad to axes (X/Y), +if you said N it will map the d-pad to buttons, which is needed for dance +style games to function correctly. The default is Y. + +dpad_to_buttons has no effect for known pads. + +0.1 Normal Controllers +---------------------- +With a normal controller, the directional pad is mapped to its own X/Y axes. +The jstest-program from joystick-1.2.15 (jstest-version 2.1.0) will report 8 +axes and 10 buttons. -Alls 8 axes work, though they all have the same range (-32768..32767) +All 8 axes work, though they all have the same range (-32768..32767) and the zero-setting is not correct for the triggers (I don't know if that is some limitation of jstest, since the input device setup should be fine. I didn't have a look at jstest itself yet). @@ -30,16 +47,50 @@ in game functionality were OK. However, I find it rather difficult to play first person shooters with a pad. Your mileage may vary. +0.2 Xbox Dance Pads +------------------- +When using a known dance pad, jstest will report 6 axes and 14 buttons. + +For dance style pads (like the redoctane pad) several changes +have been made. The old driver would map the d-pad to axes, resulting +in the driver being unable to report when the user was pressing both +left+right or up+down, making DDR style games unplayable. + +Known dance pads automatically map the d-pad to buttons and will work +correctly out of the box. + +If your dance pad is recognized by the driver but is using axes instead +of buttons, see section 0.3 - Unknown Controllers + +I've tested this with Stepmania, and it works quite well. + + +0.3 Unkown Controllers +---------------------- +If you have an unkown xbox controller, it should work just fine with +the default settings. + +HOWEVER if you have an unknown dance pad not listed below, it will not +work UNLESS you set "dpad_to_buttons" to 1 in the module configuration. + +PLEASE if you have an unkown controller, email Dom with +a dump from /proc/bus/usb and a description of the pad (manufacturer, country, +whether it is a dance pad or normal controller) so that we can add your pad +to the list of supported devices, ensuring that it will work out of the +box in the future. + + 1. USB adapter -------------- Before you can actually use the driver, you need to get yourself an -adapter cable to connect the X-Box controller to your Linux-Box. +adapter cable to connect the X-Box controller to your Linux-Box. You +can buy these online fairly cheap, or build your own. -Such a cable is pretty easy to build. The Controller itself is a USB compound -device (a hub with three ports for two expansion slots and the controller -device) with the only difference in a nonstandard connector (5 pins vs. 4 on -standard USB connector). +Such a cable is pretty easy to build. The Controller itself is a USB +compound device (a hub with three ports for two expansion slots and +the controller device) with the only difference in a nonstandard connector +(5 pins vs. 4 on standard USB connector). You just need to solder a USB connector onto the cable and keep the yellow wire unconnected. The other pins have the same order on both @@ -51,36 +102,36 @@ original one. You can buy an extension cable and cut that instead. That way, you can still use the controller with your X-Box, if you have one ;) -2. driver installation +2. Driver Installation ---------------------- Once you have the adapter cable and the controller is connected, you need to load your USB subsystem and should cat /proc/bus/usb/devices. There should be an entry like the one at the end [4]. -Currently (as of version 0.0.4), the following three devices are included: +Currently (as of version 0.0.6), the following devices are included: original Microsoft XBOX controller (US), vendor=0x045e, product=0x0202 + smaller Microsoft XBOX controller (US), vendor=0x045e, product=0x0289 original Microsoft XBOX controller (Japan), vendor=0x045e, product=0x0285 InterAct PowerPad Pro (Germany), vendor=0x05fd, product=0x107a + RedOctane Xbox Dance Pad (US), vendor=0x0c12, product=0x8809 -If you have another controller that is not listed above and is not recognized -by the driver, please drop me a line with the appropriate info (that is, include -the name, vendor and product ID, as well as the country where you bought it; -sending the whole dump out of /proc/bus/usb/devices along would be even better). +The driver should work with xbox pads not listed above as well, however +you will need to do something extra for dance pads to work. -In theory, the driver should work with other controllers than mine -(InterAct PowerPad pro, bought in Germany) just fine, but I cannot test this -for I only have this one controller. +If you have a controller not listed above, see 0.3 - Unknown Controllers If you compiled and installed the driver, test the functionality: > modprobe xpad > modprobe joydev > jstest /dev/js0 -There should be a single line showing 18 inputs (8 axes, 10 buttons), and -it's values should change if you move the sticks and push the buttons. +If you're using a normal controller, there should be a single line showing +18 inputs (8 axes, 10 buttons), and its values should change if you move +the sticks and push the buttons. If you're using a dance pad, it should +show 20 inputs (6 axes, 14 buttons). -It works? Voila, your done ;) +It works? Voila, you're done ;) 3. Thanks @@ -111,6 +162,22 @@ I: If#= 0 Alt= 0 #EPs= 2 Cls=58(unk. ) Sub=42 Prot=00 Driver=(none) E: Ad=81(I) Atr=03(Int.) MxPS= 32 Ivl= 10ms E: Ad=02(O) Atr=03(Int.) MxPS= 32 Ivl= 10ms +5. /proc/bus/usb/devices - dump from Redoctane Xbox Dance Pad (US): + +T: Bus=01 Lev=02 Prnt=09 Port=00 Cnt=01 Dev#= 10 Spd=12 MxCh= 0 +D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1 +P: Vendor=0c12 ProdID=8809 Rev= 0.01 +S: Product=XBOX DDR +C:* #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=100mA +I: If#= 0 Alt= 0 #EPs= 2 Cls=58(unk. ) Sub=42 Prot=00 Driver=xpad +E: Ad=82(I) Atr=03(Int.) MxPS= 32 Ivl=4ms +E: Ad=02(O) Atr=03(Int.) MxPS= 32 Ivl=4ms + -- Marko Friedemann 2002-07-16 + - original doc + +Dominic Cerquetti +2005-03-19 + - added stuff for dance pads, new d-pad->axes mappings diff --git a/drivers/usb/input/xpad.c b/drivers/usb/input/xpad.c index cebb6c4..6a12a94 100644 --- a/drivers/usb/input/xpad.c +++ b/drivers/usb/input/xpad.c @@ -1,8 +1,9 @@ /* - * X-Box gamepad - v0.0.5 + * X-Box gamepad - v0.0.6 * * Copyright (c) 2002 Marko Friedemann - * + * 2005 Dominic Cerquetti + * 2006 Adam Buchbinder * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as @@ -30,9 +31,10 @@ * - Greg Kroah-Hartman - usb-skeleton driver * * TODO: - * - fine tune axes + * - fine tune axes (especially trigger axes) * - fix "analog" buttons (reported as digital now) * - get rumble working + * - need USB IDs for other dance pads * * History: * @@ -57,25 +59,40 @@ #include #include #include +#include #include +#include #include #include -#define DRIVER_VERSION "v0.0.5" +#define DRIVER_VERSION "v0.0.6" #define DRIVER_AUTHOR "Marko Friedemann " #define DRIVER_DESC "X-Box pad driver" #define XPAD_PKT_LEN 32 +/* xbox d-pads should map to buttons, as is required for DDR pads + but we map them to axes when possible to simplify things */ +#define MAP_DPAD_TO_BUTTONS 0 +#define MAP_DPAD_TO_AXES 1 +#define MAP_DPAD_UNKNOWN -1 + +static int dpad_to_buttons; +module_param(dpad_to_buttons, bool, S_IRUGO); +MODULE_PARM_DESC(dpad_to_buttons, "Map D-PAD to buttons rather than axes for unknown pads"); + static const struct xpad_device { u16 idVendor; u16 idProduct; char *name; + u8 dpad_mapping; } xpad_device[] = { - { 0x045e, 0x0202, "Microsoft X-Box pad (US)" }, - { 0x045e, 0x0285, "Microsoft X-Box pad (Japan)" }, - { 0x05fd, 0x107a, "InterAct 'PowerPad Pro' X-Box pad (Germany)" }, - { 0x0000, 0x0000, "X-Box pad" } + { 0x045e, 0x0202, "Microsoft X-Box pad v1 (US)", MAP_DPAD_TO_AXES }, + { 0x045e, 0x0289, "Microsoft X-Box pad v2 (US)", MAP_DPAD_TO_AXES }, + { 0x045e, 0x0285, "Microsoft X-Box pad (Japan)", MAP_DPAD_TO_AXES }, + { 0x05fd, 0x107a, "InterAct 'PowerPad Pro' X-Box pad (Germany)", MAP_DPAD_TO_AXES }, + { 0x0c12, 0x8809, "RedOctane Xbox Dance Pad", MAP_DPAD_TO_BUTTONS }, + { 0x0000, 0x0000, "Generic X-Box pad", MAP_DPAD_UNKNOWN } }; static const signed short xpad_btn[] = { @@ -84,11 +101,23 @@ static const signed short xpad_btn[] = { -1 /* terminating entry */ }; +/* only used if MAP_DPAD_TO_BUTTONS */ +static const signed short xpad_btn_pad[] = { + BTN_LEFT, BTN_RIGHT, /* d-pad left, right */ + BTN_0, BTN_1, /* d-pad up, down (XXX names??) */ + -1 /* terminating entry */ +}; + static const signed short xpad_abs[] = { ABS_X, ABS_Y, /* left stick */ ABS_RX, ABS_RY, /* right stick */ ABS_Z, ABS_RZ, /* triggers left/right */ - ABS_HAT0X, ABS_HAT0Y, /* digital pad */ + -1 /* terminating entry */ +}; + +/* only used if MAP_DPAD_TO_AXES */ +static const signed short xpad_abs_pad[] = { + ABS_HAT0X, ABS_HAT0Y, /* d-pad axes */ -1 /* terminating entry */ }; @@ -100,14 +129,16 @@ static struct usb_device_id xpad_table [] = { MODULE_DEVICE_TABLE (usb, xpad_table); struct usb_xpad { - struct input_dev *dev; /* input device interface */ - struct usb_device *udev; /* usb device */ + struct input_dev *dev; /* input device interface */ + struct usb_device *udev; /* usb device */ - struct urb *irq_in; /* urb for interrupt in report */ - unsigned char *idata; /* input data */ + struct urb *irq_in; /* urb for interrupt in report */ + unsigned char *idata; /* input data */ dma_addr_t idata_dma; - char phys[65]; /* physical device path */ + char phys[65]; /* physical device path */ + + int dpad_mapping; /* map d-pad to buttons or to axes */ }; /* @@ -137,14 +168,21 @@ static void xpad_process_packet(struct usb_xpad *xpad, u16 cmd, unsigned char *d input_report_abs(dev, ABS_RZ, data[11]); /* digital pad */ - input_report_abs(dev, ABS_HAT0X, !!(data[2] & 0x08) - !!(data[2] & 0x04)); - input_report_abs(dev, ABS_HAT0Y, !!(data[2] & 0x02) - !!(data[2] & 0x01)); + if (xpad->dpad_mapping == MAP_DPAD_TO_AXES) { + input_report_abs(dev, ABS_HAT0X, !!(data[2] & 0x08) - !!(data[2] & 0x04)); + input_report_abs(dev, ABS_HAT0Y, !!(data[2] & 0x02) - !!(data[2] & 0x01)); + } else /* xpad->dpad_mapping == MAP_DPAD_TO_BUTTONS */ { + input_report_key(dev, BTN_LEFT, data[2] & 0x04); + input_report_key(dev, BTN_RIGHT, data[2] & 0x08); + input_report_key(dev, BTN_0, data[2] & 0x01); // up + input_report_key(dev, BTN_1, data[2] & 0x02); // down + } /* start/back buttons and stick press left/right */ - input_report_key(dev, BTN_START, (data[2] & 0x10) >> 4); - input_report_key(dev, BTN_BACK, (data[2] & 0x20) >> 5); - input_report_key(dev, BTN_THUMBL, (data[2] & 0x40) >> 6); - input_report_key(dev, BTN_THUMBR, data[2] >> 7); + input_report_key(dev, BTN_START, data[2] & 0x10); + input_report_key(dev, BTN_BACK, data[2] & 0x20); + input_report_key(dev, BTN_THUMBL, data[2] & 0x40); + input_report_key(dev, BTN_THUMBR, data[2] & 0x80); /* "analog" buttons A, B, X, Y */ input_report_key(dev, BTN_A, data[4]); @@ -206,6 +244,28 @@ static void xpad_close (struct input_dev *dev) usb_kill_urb(xpad->irq_in); } +static void xpad_set_up_abs(struct input_dev *input_dev, signed short abs) +{ + set_bit(abs, input_dev->absbit); + + switch (abs) { + case ABS_X: + case ABS_Y: + case ABS_RX: + case ABS_RY: /* the two sticks */ + input_set_abs_params(input_dev, abs, -32768, 32767, 16, 128); + break; + case ABS_Z: + case ABS_RZ: /* the triggers */ + input_set_abs_params(input_dev, abs, 0, 255, 0, 0); + break; + case ABS_HAT0X: + case ABS_HAT0Y: /* the d-pad (only if MAP_DPAD_TO_AXES) */ + input_set_abs_params(input_dev, abs, -1, 1, 0, 0); + break; + } +} + static int xpad_probe(struct usb_interface *intf, const struct usb_device_id *id) { struct usb_device *udev = interface_to_usbdev (intf); @@ -235,6 +295,9 @@ static int xpad_probe(struct usb_interface *intf, const struct usb_device_id *id goto fail2; xpad->udev = udev; + xpad->dpad_mapping = xpad_device[i].dpad_mapping; + if (xpad->dpad_mapping == MAP_DPAD_UNKNOWN) + xpad->dpad_mapping = dpad_to_buttons; xpad->dev = input_dev; usb_make_path(udev, xpad->phys, sizeof(xpad->phys)); strlcat(xpad->phys, "/input0", sizeof(xpad->phys)); @@ -249,32 +312,19 @@ static int xpad_probe(struct usb_interface *intf, const struct usb_device_id *id input_dev->evbit[0] = BIT(EV_KEY) | BIT(EV_ABS); + /* set up buttons */ for (i = 0; xpad_btn[i] >= 0; i++) set_bit(xpad_btn[i], input_dev->keybit); + if (xpad->dpad_mapping == MAP_DPAD_TO_BUTTONS) + for (i = 0; xpad_btn_pad[i] >= 0; i++) + set_bit(xpad_btn_pad[i], input_dev->keybit); - for (i = 0; xpad_abs[i] >= 0; i++) { - - signed short t = xpad_abs[i]; - - set_bit(t, input_dev->absbit); - - switch (t) { - case ABS_X: - case ABS_Y: - case ABS_RX: - case ABS_RY: /* the two sticks */ - input_set_abs_params(input_dev, t, -32768, 32767, 16, 128); - break; - case ABS_Z: - case ABS_RZ: /* the triggers */ - input_set_abs_params(input_dev, t, 0, 255, 0, 0); - break; - case ABS_HAT0X: - case ABS_HAT0Y: /* the d-pad */ - input_set_abs_params(input_dev, t, -1, 1, 0, 0); - break; - } - } + /* set up axes */ + for (i = 0; xpad_abs[i] >= 0; i++) + xpad_set_up_abs(input_dev, xpad_abs[i]); + if (xpad->dpad_mapping == MAP_DPAD_TO_AXES) + for (i = 0; xpad_abs_pad[i] >= 0; i++) + xpad_set_up_abs(input_dev, xpad_abs_pad[i]); ep_irq_in = &intf->cur_altsetting->endpoint[0].desc; usb_fill_int_urb(xpad->irq_in, udev, @@ -305,7 +355,8 @@ static void xpad_disconnect(struct usb_interface *intf) usb_kill_urb(xpad->irq_in); input_unregister_device(xpad->dev); usb_free_urb(xpad->irq_in); - usb_buffer_free(interface_to_usbdev(intf), XPAD_PKT_LEN, xpad->idata, xpad->idata_dma); + usb_buffer_free(interface_to_usbdev(intf), XPAD_PKT_LEN, + xpad->idata, xpad->idata_dma); kfree(xpad); } } -- cgit v0.10.2 From 9ab99c8c513313c1c5931bdbd27dcc4bc7a3b7cd Mon Sep 17 00:00:00 2001 From: matthieu castet Date: Wed, 11 Oct 2006 14:20:56 -0700 Subject: UEAGLE: fix ueagle-atm Oops The array of attribute passed to sysfs_create_group() must be NULL-terminated. The sysfs entries are created before the start of the modem state machine to avoid to stop it in case of errors in sysfs creation. Also {destroy,create}_fs_entries are removed as they do nothing. Signed-off-by: Laurent Riffard Signed-off-by: Matthieu CASTET Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index 5705298..f6b9f7e 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -1648,16 +1648,12 @@ static struct attribute *attrs[] = { &dev_attr_stat_usunc.attr, &dev_attr_stat_dsunc.attr, &dev_attr_stat_firmid.attr, + NULL, }; static struct attribute_group attr_grp = { .attrs = attrs, }; -static int create_fs_entries(struct usb_interface *intf) -{ - return sysfs_create_group(&intf->dev.kobj, &attr_grp); -} - static int uea_bind(struct usbatm_data *usbatm, struct usb_interface *intf, const struct usb_device_id *id) { @@ -1717,31 +1713,25 @@ static int uea_bind(struct usbatm_data *usbatm, struct usb_interface *intf, } } + ret = sysfs_create_group(&intf->dev.kobj, &attr_grp); + if (ret < 0) + goto error; + ret = uea_boot(sc); - if (ret < 0) { - kfree(sc); - return ret; - } + if (ret < 0) + goto error; - ret = create_fs_entries(intf); - if (ret) { - uea_stop(sc); - kfree(sc); - return ret; - } return 0; -} - -static void destroy_fs_entries(struct usb_interface *intf) -{ - sysfs_remove_group(&intf->dev.kobj, &attr_grp); +error: + kfree(sc); + return ret; } static void uea_unbind(struct usbatm_data *usbatm, struct usb_interface *intf) { struct uea_softc *sc = usbatm->driver_data; - destroy_fs_entries(intf); + sysfs_remove_group(&intf->dev.kobj, &attr_grp); uea_stop(sc); kfree(sc); } -- cgit v0.10.2 From 5dfb5f1d060a6f7dfddb78dc59f9e4d299088cc1 Mon Sep 17 00:00:00 2001 From: Daniel Ritz Date: Wed, 11 Oct 2006 23:40:22 +0200 Subject: usbtouchscreen: fix data reading for ITM touchscreens ITM devices seem to report only garbage when not touched. update usbtouchscreen to do data reading like itmtouch. also fix wrong mask on pressure bits. Signed-off-by: Daniel Ritz Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/usbtouchscreen.c b/drivers/usb/input/usbtouchscreen.c index f26c1cd..2902742 100644 --- a/drivers/usb/input/usbtouchscreen.c +++ b/drivers/usb/input/usbtouchscreen.c @@ -256,10 +256,10 @@ static int itm_read_data(unsigned char *pkt, int *x, int *y, int *touch, int *pr { *x = ((pkt[0] & 0x1F) << 7) | (pkt[3] & 0x7F); *y = ((pkt[1] & 0x1F) << 7) | (pkt[4] & 0x7F); - *press = ((pkt[2] & 0x1F) << 7) | (pkt[5] & 0x7F); + *press = ((pkt[2] & 0x01) << 7) | (pkt[5] & 0x7F); *touch = ~pkt[7] & 0x20; - return 1; + return *touch; } #endif -- cgit v0.10.2 From b3899dacafb10347b1b7a9f589b6c70cf8f08a3e Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 11 Oct 2006 21:50:24 -0400 Subject: USB/gadget/net2280: handle sysfs errors Signed-off-by: Jeff Garzik Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c index d954daa..7cfe0e5 100644 --- a/drivers/usb/gadget/net2280.c +++ b/drivers/usb/gadget/net2280.c @@ -2044,8 +2044,10 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver) return retval; } - device_create_file (&dev->pdev->dev, &dev_attr_function); - device_create_file (&dev->pdev->dev, &dev_attr_queues); + retval = device_create_file (&dev->pdev->dev, &dev_attr_function); + if (retval) goto err_unbind; + retval = device_create_file (&dev->pdev->dev, &dev_attr_queues); + if (retval) goto err_func; /* ... then enable host detection and ep0; and we're ready * for set_configuration as well as eventual disconnect. @@ -2060,6 +2062,14 @@ int usb_gadget_register_driver (struct usb_gadget_driver *driver) /* pci writes may still be posted */ return 0; + +err_func: + device_remove_file (&dev->pdev->dev, &dev_attr_function); +err_unbind: + driver->unbind (&dev->gadget); + dev->gadget.dev.driver = NULL; + dev->driver = NULL; + return retval; } EXPORT_SYMBOL (usb_gadget_register_driver); @@ -2974,8 +2984,10 @@ static int net2280_probe (struct pci_dev *pdev, const struct pci_device_id *id) : "disabled"); the_controller = dev; - device_register (&dev->gadget.dev); - device_create_file (&pdev->dev, &dev_attr_registers); + retval = device_register (&dev->gadget.dev); + if (retval) goto done; + retval = device_create_file (&pdev->dev, &dev_attr_registers); + if (retval) goto done; return 0; -- cgit v0.10.2 From 61926b975d83aa6c0124b5b0ce40c08579e6cc98 Mon Sep 17 00:00:00 2001 From: Craig Shelley Date: Thu, 12 Oct 2006 22:09:56 +0100 Subject: USB-SERIAL:cp2101 Add new device ID This patch adds device ID 0xEA61. This is another factory default ID used by SILabs. Signed-off-by: Craig Shelley Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/cp2101.c b/drivers/usb/serial/cp2101.c index 486c741..bbf6532 100644 --- a/drivers/usb/serial/cp2101.c +++ b/drivers/usb/serial/cp2101.c @@ -65,6 +65,7 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x10C4, 0x813D) }, /* Burnside Telecom Deskmobile */ { USB_DEVICE(0x10C4, 0x815E) }, /* Helicomm IP-Link 1220-DVM */ { USB_DEVICE(0x10C4, 0xEA60) }, /* Silicon Labs factory default */ + { USB_DEVICE(0x10C4, 0xEA61) }, /* Silicon Labs factory default */ { USB_DEVICE(0x16D6, 0x0001) }, /* Jablotron serial interface */ { } /* Terminating Entry */ }; -- cgit v0.10.2 From 5c09d144ff94706c2a5df292329ad83a27380173 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Fri, 13 Oct 2006 15:57:58 -0700 Subject: USB: ftdi_sio whitespace fixes Whitespace fixups for drivers/usb/serial/ftdi_sio.c ... removing end-of-line whitespace, and space-before-tab. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index d3dc1a1..bd76b4c 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -1,16 +1,16 @@ /* * USB FTDI SIO driver * - * Copyright (C) 1999 - 2001 - * Greg Kroah-Hartman (greg@kroah.com) + * Copyright (C) 1999 - 2001 + * Greg Kroah-Hartman (greg@kroah.com) * Bill Ryder (bryder@sgi.com) * Copyright (C) 2002 * Kuba Ober (kuba@mareimbrium.org) * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. * * See Documentation/usb/usb-serial.txt for more information on using this driver * @@ -32,7 +32,7 @@ * Changed full name of USB-UIRT device to avoid "/" character. * Added FTDI's alternate PID (0x6006) for FT232/245 devices. * Added PID for "ELV USB Module UO100" from Stefan Frings. - * + * * (21/Oct/2003) Ian Abbott * Renamed some VID/PID macros for Matrix Orbital and Perle Systems * devices. Removed Matrix Orbital and Perle Systems devices from the @@ -69,7 +69,7 @@ * does not incure any measurable overhead. This also relies on the fact * that we have proper reference counting logic for urbs. I nicked this * from Greg KH's Visor driver. - * + * * (23/Jun/2003) Ian Abbott * Reduced flip buffer pushes and corrected a data length test in * ftdi_read_bulk_callback. @@ -77,7 +77,7 @@ * * (21/Jun/2003) Erik Nygren * Added support for Home Electronics Tira-1 IR transceiver using FT232BM chip. - * See . Only operates properly + * See . Only operates properly * at 100000 and RTS-CTS, so set custom divisor mode on startup. * Also force the Tira-1 and USB-UIRT to only use their custom baud rates. * @@ -137,17 +137,17 @@ * (17/Feb/2003) Bill Ryder * Added write urb buffer pool on a per device basis * Added more checking for open file on callbacks (fixed OOPS) - * Added CrystalFontz 632 and 634 PIDs + * Added CrystalFontz 632 and 634 PIDs * (thanx to CrystalFontz for the sample devices - they flushed out * some driver bugs) * Minor debugging message changes * Added throttle, unthrottle and chars_in_buffer functions * Fixed FTDI_SIO (the original device) bug * Fixed some shutdown handling - * - * - * - * + * + * + * + * * (07/Jun/2002) Kuba Ober * Changed FTDI_SIO_BASE_BAUD_TO_DIVISOR macro into ftdi_baud_to_divisor * function. It was getting too complex. @@ -158,7 +158,7 @@ * * (25/Jul/2002) Bill Ryder inserted Dmitri's TIOCMIWAIT patch * Not tested by me but it doesn't break anything I use. - * + * * (04/Jan/2002) Kuba Ober * Implemented 38400 baudrate kludge, where it can be substituted with other * values. That's the only way to set custom baudrates. @@ -179,7 +179,7 @@ * (the previous version caused panics) * Removed port iteration code since the device only has one I/O port and it * was wrong anyway. - * + * * (31/May/2001) gkh * Switched from using spinlock to a semaphore, which fixes lots of problems. * @@ -188,16 +188,16 @@ * Cleaned up comments for 8U232 * Added parity, framing and overrun error handling * Added receive break handling. - * + * * (04/08/2001) gb * Identify version on module load. - * + * * (18/March/2001) Bill Ryder * (Not released) * Added send break handling. (requires kernel patch too) * Fixed 8U232AM hardware RTS/CTS etc status reporting. * Added flipbuf fix copied from generic device - * + * * (12/3/2000) Bill Ryder * Added support for 8U232AM device. * Moved PID and VIDs into header file only. @@ -211,14 +211,14 @@ * Cleaned up comments. Removed multiple PID/VID definitions. * Factorised cts/dtr code * Made use of __FUNCTION__ in dbg's - * + * * (11/01/2000) Adam J. Richter * usb_device_id table support - * + * * (10/05/2000) gkh * Fixed bug with urb->dev not being set properly, now that the usb * core needs it. - * + * * (09/11/2000) gkh * Removed DEBUG #ifdefs with call to usb_serial_debug_data * @@ -226,11 +226,11 @@ * Added module_init and module_exit functions to handle the fact that this * driver is a loadable module now. * - * (04/04/2000) Bill Ryder + * (04/04/2000) Bill Ryder * Fixed bugs in TCGET/TCSET ioctls (by removing them - they are * handled elsewhere in the tty io driver chain). * - * (03/30/2000) Bill Ryder + * (03/30/2000) Bill Ryder * Implemented lots of ioctls * Fixed a race condition in write * Changed some dbg's to errs @@ -444,13 +444,13 @@ static struct usb_device_id id_table_combined [] = { /* { USB_DEVICE(FTDI_VID, FTDI_ELV_WS300PC_PID) }, */ /* { USB_DEVICE(FTDI_VID, FTDI_ELV_FHZ1300PC_PID) }, */ /* { USB_DEVICE(FTDI_VID, FTDI_ELV_WS500_PID) }, */ - { USB_DEVICE(FTDI_VID, LINX_SDMUSBQSS_PID) }, - { USB_DEVICE(FTDI_VID, LINX_MASTERDEVEL2_PID) }, - { USB_DEVICE(FTDI_VID, LINX_FUTURE_0_PID) }, - { USB_DEVICE(FTDI_VID, LINX_FUTURE_1_PID) }, - { USB_DEVICE(FTDI_VID, LINX_FUTURE_2_PID) }, - { USB_DEVICE(FTDI_VID, FTDI_CCSICDU20_0_PID) }, - { USB_DEVICE(FTDI_VID, FTDI_CCSICDU40_1_PID) }, + { USB_DEVICE(FTDI_VID, LINX_SDMUSBQSS_PID) }, + { USB_DEVICE(FTDI_VID, LINX_MASTERDEVEL2_PID) }, + { USB_DEVICE(FTDI_VID, LINX_FUTURE_0_PID) }, + { USB_DEVICE(FTDI_VID, LINX_FUTURE_1_PID) }, + { USB_DEVICE(FTDI_VID, LINX_FUTURE_2_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CCSICDU20_0_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CCSICDU40_1_PID) }, { USB_DEVICE(FTDI_VID, INSIDE_ACCESSO) }, { USB_DEVICE(INTREPID_VID, INTREPID_VALUECAN_PID) }, { USB_DEVICE(INTREPID_VID, INTREPID_NEOVI_PID) }, @@ -522,7 +522,7 @@ static struct usb_driver ftdi_driver = { .probe = usb_serial_probe, .disconnect = usb_serial_disconnect, .id_table = id_table_combined, - .no_dynamic_id = 1, + .no_dynamic_id = 1, }; static const char *ftdi_chip_name[] = { @@ -548,13 +548,13 @@ struct ftdi_private { int custom_divisor; /* custom_divisor kludge, this is for baud_base (different from what goes to the chip!) */ __u16 last_set_data_urb_value ; /* the last data state set - needed for doing a break */ - int write_offset; /* This is the offset in the usb data block to write the serial data - + int write_offset; /* This is the offset in the usb data block to write the serial data - * it is different between devices */ int flags; /* some ASYNC_xxxx flags are supported */ unsigned long last_dtr_rts; /* saved modem control outputs */ wait_queue_head_t delta_msr_wait; /* Used for TIOCMIWAIT */ - char prev_status, diff_status; /* Used for TIOCMIWAIT */ + char prev_status, diff_status; /* Used for TIOCMIWAIT */ __u8 rx_flags; /* receive state flags (throttling) */ spinlock_t rx_lock; /* spinlock for receive state */ struct work_struct rx_work; @@ -721,7 +721,7 @@ static int update_mctrl(struct usb_serial_port *port, unsigned int set, unsigned urb_value |= FTDI_SIO_SET_RTS_HIGH; rv = usb_control_msg(port->serial->dev, usb_sndctrlpipe(port->serial->dev, 0), - FTDI_SIO_SET_MODEM_CTRL_REQUEST, + FTDI_SIO_SET_MODEM_CTRL_REQUEST, FTDI_SIO_SET_MODEM_CTRL_REQUEST_TYPE, urb_value, priv->interface, buf, 0, WDR_TIMEOUT); @@ -768,7 +768,7 @@ static int change_speed(struct usb_serial_port *port) if (priv->interface) { /* FT2232C */ urb_index = (__u16)((urb_index << 8) | priv->interface); } - + rv = usb_control_msg(port->serial->dev, usb_sndctrlpipe(port->serial->dev, 0), FTDI_SIO_SET_BAUDRATE_REQUEST, @@ -827,7 +827,7 @@ static __u32 get_ftdi_divisor(struct usb_serial_port * port) /* 3. Convert baudrate to device-specific divisor */ - if (!baud) baud = 9600; + if (!baud) baud = 9600; switch(priv->chip_type) { case SIO: /* SIO chip */ switch(baud) { @@ -843,7 +843,7 @@ static __u32 get_ftdi_divisor(struct usb_serial_port * port) case 115200: div_value = ftdi_sio_b115200; break; } /* baud */ if (div_value == 0) { - dbg("%s - Baudrate (%d) requested is not supported", __FUNCTION__, baud); + dbg("%s - Baudrate (%d) requested is not supported", __FUNCTION__, baud); div_value = ftdi_sio_b9600; div_okay = 0; } @@ -925,7 +925,7 @@ static int set_serial_info(struct usb_serial_port * port, struct serial_struct _ /* Make the changes - these are privileged changes! */ priv->flags = ((priv->flags & ~ASYNC_FLAGS) | - (new_serial.flags & ASYNC_FLAGS)); + (new_serial.flags & ASYNC_FLAGS)); priv->custom_divisor = new_serial.custom_divisor; port->tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0; @@ -950,7 +950,7 @@ check_and_exit: (old_priv.custom_divisor != priv->custom_divisor))) { change_speed(port); } - + return (0); } /* set_serial_info */ @@ -1022,18 +1022,18 @@ static ssize_t show_latency_timer(struct device *dev, struct device_attribute *a struct usb_device *udev; unsigned short latency = 0; int rv = 0; - + udev = to_usb_device(dev); - + dbg("%s",__FUNCTION__); - + rv = usb_control_msg(udev, usb_rcvctrlpipe(udev, 0), FTDI_SIO_GET_LATENCY_TIMER_REQUEST, FTDI_SIO_GET_LATENCY_TIMER_REQUEST_TYPE, - 0, priv->interface, + 0, priv->interface, (char*) &latency, 1, WDR_TIMEOUT); - + if (rv < 0) { dev_err(dev, "Unable to read latency timer: %i", rv); return -EIO; @@ -1051,23 +1051,23 @@ static ssize_t store_latency_timer(struct device *dev, struct device_attribute * char buf[1]; int v = simple_strtoul(valbuf, NULL, 10); int rv = 0; - + udev = to_usb_device(dev); - + dbg("%s: setting latency timer = %i", __FUNCTION__, v); - + rv = usb_control_msg(udev, usb_sndctrlpipe(udev, 0), FTDI_SIO_SET_LATENCY_TIMER_REQUEST, FTDI_SIO_SET_LATENCY_TIMER_REQUEST_TYPE, - v, priv->interface, + v, priv->interface, buf, 0, WDR_TIMEOUT); - + if (rv < 0) { dev_err(dev, "Unable to write latency timer: %i", rv); return -EIO; } - + return count; } @@ -1082,23 +1082,23 @@ static ssize_t store_event_char(struct device *dev, struct device_attribute *att char buf[1]; int v = simple_strtoul(valbuf, NULL, 10); int rv = 0; - + udev = to_usb_device(dev); - + dbg("%s: setting event char = %i", __FUNCTION__, v); - + rv = usb_control_msg(udev, usb_sndctrlpipe(udev, 0), FTDI_SIO_SET_EVENT_CHAR_REQUEST, FTDI_SIO_SET_EVENT_CHAR_REQUEST_TYPE, - v, priv->interface, + v, priv->interface, buf, 0, WDR_TIMEOUT); - + if (rv < 0) { dbg("Unable to write event character: %i", rv); return -EIO; } - + return count; } @@ -1135,11 +1135,11 @@ static void remove_sysfs_attrs(struct usb_serial *serial) struct ftdi_private *priv; struct usb_device *udev; - dbg("%s",__FUNCTION__); + dbg("%s",__FUNCTION__); priv = usb_get_serial_port_data(serial->port[0]); udev = serial->dev; - + /* XXX see create_sysfs_attrs */ if (priv->chip_type != SIO) { device_remove_file(&udev->dev, &dev_attr_event_char); @@ -1147,7 +1147,7 @@ static void remove_sysfs_attrs(struct usb_serial *serial) device_remove_file(&udev->dev, &dev_attr_latency_timer); } } - + } /* @@ -1258,7 +1258,7 @@ static void ftdi_HE_TIRA1_setup (struct usb_serial *serial) } /* ftdi_HE_TIRA1_setup */ -/* ftdi_shutdown is called from usbserial:usb_serial_disconnect +/* ftdi_shutdown is called from usbserial:usb_serial_disconnect * it is called when the usb device is disconnected * * usbserial:usb_serial_disconnect @@ -1269,16 +1269,16 @@ static void ftdi_HE_TIRA1_setup (struct usb_serial *serial) static void ftdi_shutdown (struct usb_serial *serial) { /* ftdi_shutdown */ - + struct usb_serial_port *port = serial->port[0]; struct ftdi_private *priv = usb_get_serial_port_data(port); dbg("%s", __FUNCTION__); remove_sysfs_attrs(serial); - - /* all open ports are closed at this point - * (by usbserial.c:__serial_close, which calls ftdi_close) + + /* all open ports are closed at this point + * (by usbserial.c:__serial_close, which calls ftdi_close) */ if (priv) { @@ -1293,7 +1293,7 @@ static int ftdi_open (struct usb_serial_port *port, struct file *filp) struct usb_device *dev = port->serial->dev; struct ftdi_private *priv = usb_get_serial_port_data(port); unsigned long flags; - + int result = 0; char buf[1]; /* Needed for the usb_control_msg I think */ @@ -1312,8 +1312,8 @@ static int ftdi_open (struct usb_serial_port *port, struct file *filp) /* No error checking for this (will get errors later anyway) */ /* See ftdi_sio.h for description of what is reset */ usb_control_msg(dev, usb_sndctrlpipe(dev, 0), - FTDI_SIO_RESET_REQUEST, FTDI_SIO_RESET_REQUEST_TYPE, - FTDI_SIO_RESET_SIO, + FTDI_SIO_RESET_REQUEST, FTDI_SIO_RESET_REQUEST_TYPE, + FTDI_SIO_RESET_SIO, priv->interface, buf, 0, WDR_TIMEOUT); /* Termios defaults are set by usb_serial_init. We don't change @@ -1350,12 +1350,12 @@ static int ftdi_open (struct usb_serial_port *port, struct file *filp) -/* +/* * usbserial:__serial_close only calls ftdi_close if the point is open * * This only gets called when it is the last close - * - * + * + * */ static void ftdi_close (struct usb_serial_port *port, struct file *filp) @@ -1368,14 +1368,14 @@ static void ftdi_close (struct usb_serial_port *port, struct file *filp) if (c_cflag & HUPCL){ /* Disable flow control */ - if (usb_control_msg(port->serial->dev, + if (usb_control_msg(port->serial->dev, usb_sndctrlpipe(port->serial->dev, 0), FTDI_SIO_SET_FLOW_CTRL_REQUEST, FTDI_SIO_SET_FLOW_CTRL_REQUEST_TYPE, 0, priv->interface, buf, 0, WDR_TIMEOUT) < 0) { err("error from flowcontrol urb"); - } + } /* drop RTS and DTR */ clear_mctrl(port, TIOCM_DTR | TIOCM_RTS); @@ -1384,14 +1384,14 @@ static void ftdi_close (struct usb_serial_port *port, struct file *filp) /* cancel any scheduled reading */ cancel_delayed_work(&priv->rx_work); flush_scheduled_work(); - + /* shutdown our bulk read */ if (port->read_urb) usb_kill_urb(port->read_urb); } /* ftdi_close */ - + /* The SIO requires the first byte to have: * B0 1 * B1 0 @@ -1423,7 +1423,7 @@ static int ftdi_write (struct usb_serial_port *port, return 0; } spin_unlock_irqrestore(&priv->tx_lock, flags); - + data_offset = priv->write_offset; dbg("data_offset set to %d",data_offset); @@ -1462,7 +1462,7 @@ static int ftdi_write (struct usb_serial_port *port, user_pktsz = todo; } /* Write the control byte at the front of the packet*/ - *first_byte = 1 | ((user_pktsz) << 2); + *first_byte = 1 | ((user_pktsz) << 2); /* Copy data for packet */ memcpy (first_byte + data_offset, current_position, user_pktsz); @@ -1479,7 +1479,7 @@ static int ftdi_write (struct usb_serial_port *port, usb_serial_debug_data(debug, &port->dev, __FUNCTION__, transfer_size, buffer); /* fill the buffer and send it */ - usb_fill_bulk_urb(urb, port->serial->dev, + usb_fill_bulk_urb(urb, port->serial->dev, usb_sndbulkpipe(port->serial->dev, port->bulk_out_endpointAddress), buffer, transfer_size, ftdi_write_bulk_callback, port); @@ -1520,7 +1520,7 @@ static void ftdi_write_bulk_callback (struct urb *urb) kfree (urb->transfer_buffer); dbg("%s - port %d", __FUNCTION__, port->number); - + if (urb->status) { dbg("nonzero write bulk status received: %d", urb->status); return; @@ -1651,7 +1651,7 @@ static void ftdi_process_read (void *param) struct tty_struct *tty; struct ftdi_private *priv; char error_flag; - unsigned char *data; + unsigned char *data; int i; int result; @@ -1759,7 +1759,7 @@ static void ftdi_process_read (void *param) } if (length > 0) { for (i = 2; i < length+2; i++) { - /* Note that the error flag is duplicated for + /* Note that the error flag is duplicated for every character received since we don't know which character it applied to */ tty_insert_flip_char(tty, data[packet_offset+i], error_flag); @@ -1773,7 +1773,7 @@ static void ftdi_process_read (void *param) This doesn't work well since the application receives a never ending stream of bad data - even though new data hasn't been sent. Therefore I (bill) have taken this out. - However - this might make sense for framing errors and so on + However - this might make sense for framing errors and so on so I am leaving the code in for now. */ else { @@ -1827,7 +1827,7 @@ static void ftdi_process_read (void *param) /* if the port is closed stop trying to read */ if (port->open_count > 0){ /* Continue trying to always read */ - usb_fill_bulk_urb(port->read_urb, port->serial->dev, + usb_fill_bulk_urb(port->read_urb, port->serial->dev, usb_rcvbulkpipe(port->serial->dev, port->bulk_in_endpointAddress), port->read_urb->transfer_buffer, port->read_urb->transfer_buffer_length, ftdi_read_bulk_callback, port); @@ -1844,9 +1844,9 @@ static void ftdi_process_read (void *param) static void ftdi_break_ctl( struct usb_serial_port *port, int break_state ) { struct ftdi_private *priv = usb_get_serial_port_data(port); - __u16 urb_value = 0; + __u16 urb_value = 0; char buf[1]; - + /* break_state = -1 to turn on break, and 0 to turn off break */ /* see drivers/char/tty_io.c to see it used */ /* last_set_data_urb_value NEVER has the break bit set in it */ @@ -1854,20 +1854,20 @@ static void ftdi_break_ctl( struct usb_serial_port *port, int break_state ) if (break_state) { urb_value = priv->last_set_data_urb_value | FTDI_SIO_SET_BREAK; } else { - urb_value = priv->last_set_data_urb_value; + urb_value = priv->last_set_data_urb_value; } - + if (usb_control_msg(port->serial->dev, usb_sndctrlpipe(port->serial->dev, 0), - FTDI_SIO_SET_DATA_REQUEST, + FTDI_SIO_SET_DATA_REQUEST, FTDI_SIO_SET_DATA_REQUEST_TYPE, urb_value , priv->interface, buf, 0, WDR_TIMEOUT) < 0) { err("%s FAILED to enable/disable break state (state was %d)", __FUNCTION__,break_state); - } + } dbg("%s break state is %d - urb is %d", __FUNCTION__,break_state, urb_value); - + } @@ -1883,12 +1883,12 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct termios *old_ struct ftdi_private *priv = usb_get_serial_port_data(port); __u16 urb_value; /* will hold the new flags */ char buf[1]; /* Perhaps I should dynamically alloc this? */ - + // Added for xon/xoff support unsigned int iflag = port->tty->termios->c_iflag; unsigned char vstop; unsigned char vstart; - + dbg("%s", __FUNCTION__); /* Force baud rate if this device requires it, unless it is set to B0. */ @@ -1906,20 +1906,20 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct termios *old_ cflag = port->tty->termios->c_cflag; - /* FIXME -For this cut I don't care if the line is really changing or - not - so just do the change regardless - should be able to + /* FIXME -For this cut I don't care if the line is really changing or + not - so just do the change regardless - should be able to compare old_termios and tty->termios */ - /* NOTE These routines can get interrupted by - ftdi_sio_read_bulk_callback - need to examine what this + /* NOTE These routines can get interrupted by + ftdi_sio_read_bulk_callback - need to examine what this means - don't see any problems yet */ - + /* Set number of data bits, parity, stop bits */ - + urb_value = 0; urb_value |= (cflag & CSTOPB ? FTDI_SIO_SET_DATA_STOP_BITS_2 : FTDI_SIO_SET_DATA_STOP_BITS_1); - urb_value |= (cflag & PARENB ? - (cflag & PARODD ? FTDI_SIO_SET_DATA_PARITY_ODD : + urb_value |= (cflag & PARENB ? + (cflag & PARODD ? FTDI_SIO_SET_DATA_PARITY_ODD : FTDI_SIO_SET_DATA_PARITY_EVEN) : FTDI_SIO_SET_DATA_PARITY_NONE); if (cflag & CSIZE) { @@ -1936,25 +1936,25 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct termios *old_ /* This is needed by the break command since it uses the same command - but is * or'ed with this value */ priv->last_set_data_urb_value = urb_value; - + if (usb_control_msg(dev, usb_sndctrlpipe(dev, 0), - FTDI_SIO_SET_DATA_REQUEST, + FTDI_SIO_SET_DATA_REQUEST, FTDI_SIO_SET_DATA_REQUEST_TYPE, urb_value , priv->interface, buf, 0, WDR_SHORT_TIMEOUT) < 0) { err("%s FAILED to set databits/stopbits/parity", __FUNCTION__); - } + } /* Now do the baudrate */ if ((cflag & CBAUD) == B0 ) { /* Disable flow control */ if (usb_control_msg(dev, usb_sndctrlpipe(dev, 0), - FTDI_SIO_SET_FLOW_CTRL_REQUEST, + FTDI_SIO_SET_FLOW_CTRL_REQUEST, FTDI_SIO_SET_FLOW_CTRL_REQUEST_TYPE, - 0, priv->interface, + 0, priv->interface, buf, 0, WDR_TIMEOUT) < 0) { err("%s error from disable flowcontrol urb", __FUNCTION__); - } + } /* Drop RTS and DTR */ clear_mctrl(port, TIOCM_DTR | TIOCM_RTS); } else { @@ -1972,16 +1972,16 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct termios *old_ /* Note device also supports DTR/CD (ugh) and Xon/Xoff in hardware */ if (cflag & CRTSCTS) { dbg("%s Setting to CRTSCTS flow control", __FUNCTION__); - if (usb_control_msg(dev, + if (usb_control_msg(dev, usb_sndctrlpipe(dev, 0), - FTDI_SIO_SET_FLOW_CTRL_REQUEST, + FTDI_SIO_SET_FLOW_CTRL_REQUEST, FTDI_SIO_SET_FLOW_CTRL_REQUEST_TYPE, 0 , (FTDI_SIO_RTS_CTS_HS | priv->interface), buf, 0, WDR_TIMEOUT) < 0) { err("urb failed to set to rts/cts flow control"); - } - - } else { + } + + } else { /* * Xon/Xoff code * @@ -2011,16 +2011,16 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct termios *old_ /* else clause to only run if cfag ! CRTSCTS and iflag ! XOFF */ /* CHECKME Assuming XON/XOFF handled by tty stack - not by device */ dbg("%s Turning off hardware flow control", __FUNCTION__); - if (usb_control_msg(dev, + if (usb_control_msg(dev, usb_sndctrlpipe(dev, 0), - FTDI_SIO_SET_FLOW_CTRL_REQUEST, + FTDI_SIO_SET_FLOW_CTRL_REQUEST, FTDI_SIO_SET_FLOW_CTRL_REQUEST_TYPE, - 0, priv->interface, + 0, priv->interface, buf, 0, WDR_TIMEOUT) < 0) { err("urb failed to clear flow control"); - } + } } - + } return; } /* ftdi_termios */ @@ -2036,11 +2036,11 @@ static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file) switch (priv->chip_type) { case SIO: /* Request the status from the device */ - if ((ret = usb_control_msg(port->serial->dev, + if ((ret = usb_control_msg(port->serial->dev, usb_rcvctrlpipe(port->serial->dev, 0), - FTDI_SIO_GET_MODEM_STATUS_REQUEST, + FTDI_SIO_GET_MODEM_STATUS_REQUEST, FTDI_SIO_GET_MODEM_STATUS_REQUEST_TYPE, - 0, 0, + 0, 0, buf, 1, WDR_TIMEOUT)) < 0 ) { err("%s Could not get modem status of device - err: %d", __FUNCTION__, ret); @@ -2052,11 +2052,11 @@ static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file) case FT2232C: /* the 8U232AM returns a two byte value (the sio is a 1 byte value) - in the same format as the data returned from the in point */ - if ((ret = usb_control_msg(port->serial->dev, + if ((ret = usb_control_msg(port->serial->dev, usb_rcvctrlpipe(port->serial->dev, 0), - FTDI_SIO_GET_MODEM_STATUS_REQUEST, + FTDI_SIO_GET_MODEM_STATUS_REQUEST, FTDI_SIO_GET_MODEM_STATUS_REQUEST_TYPE, - 0, priv->interface, + 0, priv->interface, buf, 2, WDR_TIMEOUT)) < 0 ) { err("%s Could not get modem status of device - err: %d", __FUNCTION__, ret); @@ -2067,12 +2067,12 @@ static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file) return -EFAULT; break; } - + return (buf[0] & FTDI_SIO_DSR_MASK ? TIOCM_DSR : 0) | (buf[0] & FTDI_SIO_CTS_MASK ? TIOCM_CTS : 0) | (buf[0] & FTDI_SIO_RI_MASK ? TIOCM_RI : 0) | (buf[0] & FTDI_SIO_RLSD_MASK ? TIOCM_CD : 0) | - priv->last_dtr_rts; + priv->last_dtr_rts; } static int ftdi_tiocmset(struct usb_serial_port *port, struct file * file, unsigned int set, unsigned int clear) @@ -2138,11 +2138,11 @@ static int ftdi_ioctl (struct usb_serial_port *port, struct file * file, unsigne break; default: break; - + } - /* This is not necessarily an error - turns out the higher layers will do + /* This is not necessarily an error - turns out the higher layers will do * some ioctls itself (see comment above) */ dbg("%s arg not supported - it was 0x%04x - check /usr/include/asm/ioctls.h", __FUNCTION__, cmd); @@ -2199,7 +2199,7 @@ static int __init ftdi_init (void) if (retval) goto failed_sio_register; retval = usb_register(&ftdi_driver); - if (retval) + if (retval) goto failed_usb_register; info(DRIVER_VERSION ":" DRIVER_DESC); -- cgit v0.10.2 From 033a3fb980b041c5b1c865d3e9dce9217d1dc94b Mon Sep 17 00:00:00 2001 From: Kevin Lloyd Date: Fri, 13 Oct 2006 23:53:21 -0700 Subject: USB: Sierra Wireless driver update The largest feature in this patch is that it adds significant throughput increase to the Sierra driver and adds support for modem status line control (e.g. the DTR line). This patch also updates the current sierra.c driver so that it supports both 3-port Sierra devices and 1-port legacy devices and removes Sierra's references in other related files (Kconfig and airprime.c). Signed-off-by: Kevin Lloyd Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig index 8ca6d3f..9a6ec1b 100644 --- a/drivers/usb/serial/Kconfig +++ b/drivers/usb/serial/Kconfig @@ -537,8 +537,7 @@ config USB_SERIAL_OPTION The USB bus on these cards is not accessible externally. Supported devices include (some of?) those made by: - Option, Huawei, Audiovox, Sierra Wireless, Novatel Wireless, or - Anydata. + Option, Huawei, Audiovox, Novatel Wireless, or Anydata. To compile this driver as a module, choose M here: the module will be called option. diff --git a/drivers/usb/serial/airprime.c b/drivers/usb/serial/airprime.c index ba93c72..7f5d546 100644 --- a/drivers/usb/serial/airprime.c +++ b/drivers/usb/serial/airprime.c @@ -18,11 +18,6 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x0c88, 0x17da) }, /* Kyocera Wireless KPC650/Passport */ - { USB_DEVICE(0x0f3d, 0x0112) }, /* AirPrime CDMA Wireless PC Card */ - { USB_DEVICE(0x1199, 0x0017) }, /* Sierra Wireless EM5625 */ - { USB_DEVICE(0x1199, 0x0018) }, /* Sierra Wireless MC5720 */ - { USB_DEVICE(0x1199, 0x0112) }, /* Sierra Wireless Aircard 580 */ - { USB_DEVICE(0x1199, 0x0218) }, /* Sierra Wireless MC5720 */ { USB_DEVICE(0x1410, 0x1110) }, /* Novatel Wireless Merlin CDMA */ { USB_DEVICE(0x1410, 0x1100) }, /* ExpressCard34 Qualcomm 3G CDMA */ { }, diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c index d29638d..39799d2 100644 --- a/drivers/usb/serial/sierra.c +++ b/drivers/usb/serial/sierra.c @@ -1,21 +1,59 @@ /* - * Sierra Wireless CDMA Wireless Serial USB driver - * - * Current Copy modified by: Kevin Lloyd - * Original Copyright (C) 2005-2006 Greg Kroah-Hartman - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License version - * 2 as published by the Free Software Foundation. - */ + USB Driver for Sierra Wireless + + Copyright (C) 2006 Kevin Lloyd + + IMPORTANT DISCLAIMER: This driver is not commercially supported by + Sierra Wireless. Use at your own risk. + + This driver is free software; you can redistribute it and/or modify + it under the terms of Version 2 of the GNU General Public License as + published by the Free Software Foundation. + + Portions based on the option driver by Matthias Urlichs + Whom based his on the Keyspan driver by Hugh Blemings + + History: +*/ + +#define DRIVER_VERSION "v.1.0.5" +#define DRIVER_AUTHOR "Kevin Lloyd " +#define DRIVER_DESC "USB Driver for Sierra Wireless USB modems" #include -#include +#include +#include #include +#include #include #include #include +/* Function prototypes */ +static int sierra_open(struct usb_serial_port *port, struct file *filp); +static void sierra_close(struct usb_serial_port *port, struct file *filp); +static int sierra_startup(struct usb_serial *serial); +static void sierra_shutdown(struct usb_serial *serial); +static void sierra_rx_throttle(struct usb_serial_port *port); +static void sierra_rx_unthrottle(struct usb_serial_port *port); +static int sierra_write_room(struct usb_serial_port *port); + +static void sierra_instat_callback(struct urb *urb); + +static int sierra_write(struct usb_serial_port *port, + const unsigned char *buf, int count); + +static int sierra_chars_in_buffer(struct usb_serial_port *port); +static int sierra_ioctl(struct usb_serial_port *port, struct file *file, + unsigned int cmd, unsigned long arg); +static void sierra_set_termios(struct usb_serial_port *port, + struct termios *old); +static void sierra_break_ctl(struct usb_serial_port *port, int break_state); +static int sierra_tiocmget(struct usb_serial_port *port, struct file *file); +static int sierra_tiocmset(struct usb_serial_port *port, struct file *file, + unsigned int set, unsigned int clear); +static int sierra_send_setup(struct usb_serial_port *port); + static struct usb_device_id id_table [] = { { USB_DEVICE(0x1199, 0x0018) }, /* Sierra Wireless MC5720 */ { USB_DEVICE(0x1199, 0x0020) }, /* Sierra Wireless MC5725 */ @@ -25,51 +63,680 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x1199, 0x6803) }, /* Sierra Wireless MC8765 */ { USB_DEVICE(0x1199, 0x6812) }, /* Sierra Wireless MC8775 */ { USB_DEVICE(0x1199, 0x6820) }, /* Sierra Wireless AirCard 875 */ - /* Following devices are supported in the airprime.c driver */ - /* { USB_DEVICE(0x1199, 0x0112) }, */ /* Sierra Wireless AirCard 580 */ - /* { USB_DEVICE(0x0F3D, 0x0112) }, */ /* AirPrime/Sierra PC 5220 */ + + { USB_DEVICE(0x1199, 0x0112) }, /* Sierra Wireless AirCard 580 */ + { USB_DEVICE(0x0F3D, 0x0112) }, /* AirPrime/Sierra PC 5220 */ + { } +}; + +static struct usb_device_id id_table_1port [] = { + { USB_DEVICE(0x1199, 0x0112) }, /* Sierra Wireless AirCard 580 */ + { USB_DEVICE(0x0F3D, 0x0112) }, /* AirPrime/Sierra PC 5220 */ { } }; + +static struct usb_device_id id_table_3port [] = { + { USB_DEVICE(0x1199, 0x0018) }, /* Sierra Wireless MC5720 */ + { USB_DEVICE(0x1199, 0x0020) }, /* Sierra Wireless MC5725 */ + { USB_DEVICE(0x1199, 0x0017) }, /* Sierra Wireless EM5625 */ + { USB_DEVICE(0x1199, 0x0019) }, /* Sierra Wireless AirCard 595 */ + { USB_DEVICE(0x1199, 0x6802) }, /* Sierra Wireless MC8755 */ + { USB_DEVICE(0x1199, 0x6803) }, /* Sierra Wireless MC8765 */ + { USB_DEVICE(0x1199, 0x6812) }, /* Sierra Wireless MC8775 */ + { USB_DEVICE(0x1199, 0x6820) }, /* Sierra Wireless AirCard 875 */ + { } +}; + + MODULE_DEVICE_TABLE(usb, id_table); static struct usb_driver sierra_driver = { - .name = "sierra_wireless", - .probe = usb_serial_probe, - .disconnect = usb_serial_disconnect, - .id_table = id_table, + .name = "sierra", + .probe = usb_serial_probe, + .disconnect = usb_serial_disconnect, + .id_table = id_table, + .no_dynamic_id = 3, +}; + + +//static struct usb_serial_driver *sierra_device; + +static struct usb_serial_driver sierra_1port_device = { + .driver = { + .owner = THIS_MODULE, + .name = "sierra1", + }, + .description = "Sierra USB modem (1 port)", + .id_table = id_table_1port, + .num_interrupt_in = NUM_DONT_CARE, + .num_bulk_in = 1, + .num_bulk_out = 1, + .num_ports = 1, + .open = sierra_open, + .close = sierra_close, + .write = sierra_write, + .write_room = sierra_write_room, + .chars_in_buffer = sierra_chars_in_buffer, + .throttle = sierra_rx_throttle, + .unthrottle = sierra_rx_unthrottle, + .ioctl = sierra_ioctl, + .set_termios = sierra_set_termios, + .break_ctl = sierra_break_ctl, + .tiocmget = sierra_tiocmget, + .tiocmset = sierra_tiocmset, + .attach = sierra_startup, + .shutdown = sierra_shutdown, + .read_int_callback = sierra_instat_callback, }; -static struct usb_serial_driver sierra_device = { +static struct usb_serial_driver sierra_3port_device = { .driver = { - .owner = THIS_MODULE, - .name = "Sierra_Wireless", + .owner = THIS_MODULE, + .name = "sierra3", }, - .id_table = id_table, - .num_interrupt_in = NUM_DONT_CARE, - .num_bulk_in = NUM_DONT_CARE, - .num_bulk_out = NUM_DONT_CARE, - .num_ports = 3, + .description = "Sierra USB modem (3 port)", + .id_table = id_table_3port, + .num_interrupt_in = NUM_DONT_CARE, + .num_bulk_in = 3, + .num_bulk_out = 3, + .num_ports = 3, + .open = sierra_open, + .close = sierra_close, + .write = sierra_write, + .write_room = sierra_write_room, + .chars_in_buffer = sierra_chars_in_buffer, + .throttle = sierra_rx_throttle, + .unthrottle = sierra_rx_unthrottle, + .ioctl = sierra_ioctl, + .set_termios = sierra_set_termios, + .break_ctl = sierra_break_ctl, + .tiocmget = sierra_tiocmget, + .tiocmset = sierra_tiocmset, + .attach = sierra_startup, + .shutdown = sierra_shutdown, + .read_int_callback = sierra_instat_callback, }; +#ifdef CONFIG_USB_DEBUG +static int debug; +#else +#define debug 0 +#endif + +/* per port private data */ + +#define N_IN_URB 4 +#define N_OUT_URB 1 +#define IN_BUFLEN 4096 +#define OUT_BUFLEN 128 + +struct sierra_port_private { + /* Input endpoints and buffer for this port */ + struct urb *in_urbs[N_IN_URB]; + char in_buffer[N_IN_URB][IN_BUFLEN]; + /* Output endpoints and buffer for this port */ + struct urb *out_urbs[N_OUT_URB]; + char out_buffer[N_OUT_URB][OUT_BUFLEN]; + + /* Settings for the port */ + int rts_state; /* Handshaking pins (outputs) */ + int dtr_state; + int cts_state; /* Handshaking pins (inputs) */ + int dsr_state; + int dcd_state; + int ri_state; + + unsigned long tx_start_time[N_OUT_URB]; +}; + +/* Functions used by new usb-serial code. */ static int __init sierra_init(void) { int retval; - - retval = usb_serial_register(&sierra_device); + retval = usb_serial_register(&sierra_1port_device); + if (retval) + goto failed_1port_device_register; + retval = usb_serial_register(&sierra_3port_device); if (retval) - return retval; + goto failed_3port_device_register; + + retval = usb_register(&sierra_driver); if (retval) - usb_serial_deregister(&sierra_device); + goto failed_driver_register; + + info(DRIVER_DESC ": " DRIVER_VERSION); + + return 0; + +failed_driver_register: + usb_serial_deregister(&sierra_3port_device); +failed_3port_device_register: + usb_serial_deregister(&sierra_1port_device); +failed_1port_device_register: return retval; } static void __exit sierra_exit(void) { - usb_deregister(&sierra_driver); - usb_serial_deregister(&sierra_device); + usb_deregister (&sierra_driver); + usb_serial_deregister(&sierra_1port_device); + usb_serial_deregister(&sierra_3port_device); } module_init(sierra_init); module_exit(sierra_exit); + +static void sierra_rx_throttle(struct usb_serial_port *port) +{ + dbg("%s", __FUNCTION__); +} + +static void sierra_rx_unthrottle(struct usb_serial_port *port) +{ + dbg("%s", __FUNCTION__); +} + +static void sierra_break_ctl(struct usb_serial_port *port, int break_state) +{ + /* Unfortunately, I don't know how to send a break */ + dbg("%s", __FUNCTION__); +} + +static void sierra_set_termios(struct usb_serial_port *port, + struct termios *old_termios) +{ + dbg("%s", __FUNCTION__); + + sierra_send_setup(port); +} + +static int sierra_tiocmget(struct usb_serial_port *port, struct file *file) +{ + unsigned int value; + struct sierra_port_private *portdata; + + portdata = usb_get_serial_port_data(port); + + value = ((portdata->rts_state) ? TIOCM_RTS : 0) | + ((portdata->dtr_state) ? TIOCM_DTR : 0) | + ((portdata->cts_state) ? TIOCM_CTS : 0) | + ((portdata->dsr_state) ? TIOCM_DSR : 0) | + ((portdata->dcd_state) ? TIOCM_CAR : 0) | + ((portdata->ri_state) ? TIOCM_RNG : 0); + + return value; +} + +static int sierra_tiocmset(struct usb_serial_port *port, struct file *file, + unsigned int set, unsigned int clear) +{ + struct sierra_port_private *portdata; + + portdata = usb_get_serial_port_data(port); + + if (set & TIOCM_RTS) + portdata->rts_state = 1; + if (set & TIOCM_DTR) + portdata->dtr_state = 1; + + if (clear & TIOCM_RTS) + portdata->rts_state = 0; + if (clear & TIOCM_DTR) + portdata->dtr_state = 0; + return sierra_send_setup(port); +} + +static int sierra_ioctl(struct usb_serial_port *port, struct file *file, + unsigned int cmd, unsigned long arg) +{ + return -ENOIOCTLCMD; +} + +/* Write */ +static int sierra_write(struct usb_serial_port *port, + const unsigned char *buf, int count) +{ + struct sierra_port_private *portdata; + int i; + int left, todo; + struct urb *this_urb = NULL; /* spurious */ + int err; + + portdata = usb_get_serial_port_data(port); + + dbg("%s: write (%d chars)", __FUNCTION__, count); + + i = 0; + left = count; + for (i=0; left > 0 && i < N_OUT_URB; i++) { + todo = left; + if (todo > OUT_BUFLEN) + todo = OUT_BUFLEN; + + this_urb = portdata->out_urbs[i]; + if (this_urb->status == -EINPROGRESS) { + if (time_before(jiffies, + portdata->tx_start_time[i] + 10 * HZ)) + continue; + usb_unlink_urb(this_urb); + continue; + } + if (this_urb->status != 0) + dbg("usb_write %p failed (err=%d)", + this_urb, this_urb->status); + + dbg("%s: endpoint %d buf %d", __FUNCTION__, + usb_pipeendpoint(this_urb->pipe), i); + + /* send the data */ + memcpy (this_urb->transfer_buffer, buf, todo); + this_urb->transfer_buffer_length = todo; + + this_urb->dev = port->serial->dev; + err = usb_submit_urb(this_urb, GFP_ATOMIC); + if (err) { + dbg("usb_submit_urb %p (write bulk) failed " + "(%d, has %d)", this_urb, + err, this_urb->status); + continue; + } + portdata->tx_start_time[i] = jiffies; + buf += todo; + left -= todo; + } + + count -= left; + dbg("%s: wrote (did %d)", __FUNCTION__, count); + return count; +} + +static void sierra_indat_callback(struct urb *urb) +{ + int err; + int endpoint; + struct usb_serial_port *port; + struct tty_struct *tty; + unsigned char *data = urb->transfer_buffer; + + dbg("%s: %p", __FUNCTION__, urb); + + endpoint = usb_pipeendpoint(urb->pipe); + port = (struct usb_serial_port *) urb->context; + + if (urb->status) { + dbg("%s: nonzero status: %d on endpoint %02x.", + __FUNCTION__, urb->status, endpoint); + } else { + tty = port->tty; + if (urb->actual_length) { + tty_buffer_request_room(tty, urb->actual_length); + tty_insert_flip_string(tty, data, urb->actual_length); + tty_flip_buffer_push(tty); + } else { + dbg("%s: empty read urb received", __FUNCTION__); + } + + /* Resubmit urb so we continue receiving */ + if (port->open_count && urb->status != -ESHUTDOWN) { + err = usb_submit_urb(urb, GFP_ATOMIC); + if (err) + printk(KERN_ERR "%s: resubmit read urb failed. " + "(%d)", __FUNCTION__, err); + } + } + return; +} + +static void sierra_outdat_callback(struct urb *urb) +{ + struct usb_serial_port *port; + + dbg("%s", __FUNCTION__); + + port = (struct usb_serial_port *) urb->context; + + usb_serial_port_softint(port); +} + +static void sierra_instat_callback(struct urb *urb) +{ + int err; + struct usb_serial_port *port = (struct usb_serial_port *) urb->context; + struct sierra_port_private *portdata = usb_get_serial_port_data(port); + struct usb_serial *serial = port->serial; + + dbg("%s", __FUNCTION__); + dbg("%s: urb %p port %p has data %p", __FUNCTION__,urb,port,portdata); + + if (urb->status == 0) { + struct usb_ctrlrequest *req_pkt = + (struct usb_ctrlrequest *)urb->transfer_buffer; + + if (!req_pkt) { + dbg("%s: NULL req_pkt\n", __FUNCTION__); + return; + } + if ((req_pkt->bRequestType == 0xA1) && + (req_pkt->bRequest == 0x20)) { + int old_dcd_state; + unsigned char signals = *((unsigned char *) + urb->transfer_buffer + + sizeof(struct usb_ctrlrequest)); + + dbg("%s: signal x%x", __FUNCTION__, signals); + + old_dcd_state = portdata->dcd_state; + portdata->cts_state = 1; + portdata->dcd_state = ((signals & 0x01) ? 1 : 0); + portdata->dsr_state = ((signals & 0x02) ? 1 : 0); + portdata->ri_state = ((signals & 0x08) ? 1 : 0); + + if (port->tty && !C_CLOCAL(port->tty) && + old_dcd_state && !portdata->dcd_state) + tty_hangup(port->tty); + } else { + dbg("%s: type %x req %x", __FUNCTION__, + req_pkt->bRequestType,req_pkt->bRequest); + } + } else + dbg("%s: error %d", __FUNCTION__, urb->status); + + /* Resubmit urb so we continue receiving IRQ data */ + if (urb->status != -ESHUTDOWN) { + urb->dev = serial->dev; + err = usb_submit_urb(urb, GFP_ATOMIC); + if (err) + dbg("%s: resubmit intr urb failed. (%d)", + __FUNCTION__, err); + } +} + +static int sierra_write_room(struct usb_serial_port *port) +{ + struct sierra_port_private *portdata; + int i; + int data_len = 0; + struct urb *this_urb; + + portdata = usb_get_serial_port_data(port); + + for (i=0; i < N_OUT_URB; i++) { + this_urb = portdata->out_urbs[i]; + if (this_urb && this_urb->status != -EINPROGRESS) + data_len += OUT_BUFLEN; + } + + dbg("%s: %d", __FUNCTION__, data_len); + return data_len; +} + +static int sierra_chars_in_buffer(struct usb_serial_port *port) +{ + struct sierra_port_private *portdata; + int i; + int data_len = 0; + struct urb *this_urb; + + portdata = usb_get_serial_port_data(port); + + for (i=0; i < N_OUT_URB; i++) { + this_urb = portdata->out_urbs[i]; + if (this_urb && this_urb->status == -EINPROGRESS) + data_len += this_urb->transfer_buffer_length; + } + dbg("%s: %d", __FUNCTION__, data_len); + return data_len; +} + +static int sierra_open(struct usb_serial_port *port, struct file *filp) +{ + struct sierra_port_private *portdata; + struct usb_serial *serial = port->serial; + int i, err; + struct urb *urb; + + portdata = usb_get_serial_port_data(port); + + dbg("%s", __FUNCTION__); + + /* Set some sane defaults */ + portdata->rts_state = 1; + portdata->dtr_state = 1; + + /* Reset low level data toggle and start reading from endpoints */ + for (i = 0; i < N_IN_URB; i++) { + urb = portdata->in_urbs[i]; + if (! urb) + continue; + if (urb->dev != serial->dev) { + dbg("%s: dev %p != %p", __FUNCTION__, + urb->dev, serial->dev); + continue; + } + + /* + * make sure endpoint data toggle is synchronized with the + * device + */ + usb_clear_halt(urb->dev, urb->pipe); + + err = usb_submit_urb(urb, GFP_KERNEL); + if (err) { + dbg("%s: submit urb %d failed (%d) %d", + __FUNCTION__, i, err, + urb->transfer_buffer_length); + } + } + + /* Reset low level data toggle on out endpoints */ + for (i = 0; i < N_OUT_URB; i++) { + urb = portdata->out_urbs[i]; + if (! urb) + continue; + urb->dev = serial->dev; + /* usb_settoggle(urb->dev, usb_pipeendpoint(urb->pipe), + usb_pipeout(urb->pipe), 0); */ + } + + port->tty->low_latency = 1; + + sierra_send_setup(port); + + return (0); +} + +static inline void stop_urb(struct urb *urb) +{ + if (urb && urb->status == -EINPROGRESS) + usb_kill_urb(urb); +} + +static void sierra_close(struct usb_serial_port *port, struct file *filp) +{ + int i; + struct usb_serial *serial = port->serial; + struct sierra_port_private *portdata; + + dbg("%s", __FUNCTION__); + portdata = usb_get_serial_port_data(port); + + portdata->rts_state = 0; + portdata->dtr_state = 0; + + if (serial->dev) { + sierra_send_setup(port); + + /* Stop reading/writing urbs */ + for (i = 0; i < N_IN_URB; i++) + stop_urb(portdata->in_urbs[i]); + for (i = 0; i < N_OUT_URB; i++) + stop_urb(portdata->out_urbs[i]); + } + port->tty = NULL; +} + +/* Helper functions used by sierra_setup_urbs */ +static struct urb *sierra_setup_urb(struct usb_serial *serial, int endpoint, + int dir, void *ctx, char *buf, int len, + void (*callback)(struct urb *)) +{ + struct urb *urb; + + if (endpoint == -1) + return NULL; /* endpoint not needed */ + + urb = usb_alloc_urb(0, GFP_KERNEL); /* No ISO */ + if (urb == NULL) { + dbg("%s: alloc for endpoint %d failed.", __FUNCTION__, endpoint); + return NULL; + } + + /* Fill URB using supplied data. */ + usb_fill_bulk_urb(urb, serial->dev, + usb_sndbulkpipe(serial->dev, endpoint) | dir, + buf, len, callback, ctx); + + return urb; +} + +/* Setup urbs */ +static void sierra_setup_urbs(struct usb_serial *serial) +{ + int i,j; + struct usb_serial_port *port; + struct sierra_port_private *portdata; + + dbg("%s", __FUNCTION__); + + for (i = 0; i < serial->num_ports; i++) { + port = serial->port[i]; + portdata = usb_get_serial_port_data(port); + + /* Do indat endpoints first */ + for (j = 0; j < N_IN_URB; ++j) { + portdata->in_urbs[j] = sierra_setup_urb (serial, + port->bulk_in_endpointAddress, USB_DIR_IN, port, + portdata->in_buffer[j], IN_BUFLEN, sierra_indat_callback); + } + + /* outdat endpoints */ + for (j = 0; j < N_OUT_URB; ++j) { + portdata->out_urbs[j] = sierra_setup_urb (serial, + port->bulk_out_endpointAddress, USB_DIR_OUT, port, + portdata->out_buffer[j], OUT_BUFLEN, sierra_outdat_callback); + } + } +} + +static int sierra_send_setup(struct usb_serial_port *port) +{ + struct usb_serial *serial = port->serial; + struct sierra_port_private *portdata; + + dbg("%s", __FUNCTION__); + + portdata = usb_get_serial_port_data(port); + + if (port->tty) { + int val = 0; + if (portdata->dtr_state) + val |= 0x01; + if (portdata->rts_state) + val |= 0x02; + + return usb_control_msg(serial->dev, + usb_rcvctrlpipe(serial->dev, 0), + 0x22,0x21,val,0,NULL,0,USB_CTRL_SET_TIMEOUT); + } + + return 0; +} + +static int sierra_startup(struct usb_serial *serial) +{ + int i, err; + struct usb_serial_port *port; + struct sierra_port_private *portdata; + + dbg("%s", __FUNCTION__); + + /* Now setup per port private data */ + for (i = 0; i < serial->num_ports; i++) { + port = serial->port[i]; + portdata = kzalloc(sizeof(*portdata), GFP_KERNEL); + if (!portdata) { + dbg("%s: kmalloc for sierra_port_private (%d) failed!.", + __FUNCTION__, i); + return (1); + } + + usb_set_serial_port_data(port, portdata); + + if (! port->interrupt_in_urb) + continue; + err = usb_submit_urb(port->interrupt_in_urb, GFP_KERNEL); + if (err) + dbg("%s: submit irq_in urb failed %d", + __FUNCTION__, err); + } + + sierra_setup_urbs(serial); + + return (0); +} + +static void sierra_shutdown(struct usb_serial *serial) +{ + int i, j; + struct usb_serial_port *port; + struct sierra_port_private *portdata; + + dbg("%s", __FUNCTION__); + + /* Stop reading/writing urbs */ + for (i = 0; i < serial->num_ports; ++i) { + port = serial->port[i]; + portdata = usb_get_serial_port_data(port); + for (j = 0; j < N_IN_URB; j++) + stop_urb(portdata->in_urbs[j]); + for (j = 0; j < N_OUT_URB; j++) + stop_urb(portdata->out_urbs[j]); + } + + /* Now free them */ + for (i = 0; i < serial->num_ports; ++i) { + port = serial->port[i]; + portdata = usb_get_serial_port_data(port); + + for (j = 0; j < N_IN_URB; j++) { + if (portdata->in_urbs[j]) { + usb_free_urb(portdata->in_urbs[j]); + portdata->in_urbs[j] = NULL; + } + } + for (j = 0; j < N_OUT_URB; j++) { + if (portdata->out_urbs[j]) { + usb_free_urb(portdata->out_urbs[j]); + portdata->out_urbs[j] = NULL; + } + } + } + + /* Now free per port private data */ + for (i = 0; i < serial->num_ports; i++) { + port = serial->port[i]; + kfree(usb_get_serial_port_data(port)); + } +} + +MODULE_AUTHOR(DRIVER_AUTHOR); +MODULE_DESCRIPTION(DRIVER_DESC); +MODULE_VERSION(DRIVER_VERSION); MODULE_LICENSE("GPL"); + +#ifdef CONFIG_USB_DEBUG +module_param(debug, bool, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(debug, "Debug messages"); +#endif + -- cgit v0.10.2 From 964ee1deb3eac802902cd758ddb94b6a95c77987 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 17 Oct 2006 10:17:58 -0700 Subject: USB: cleanup sierra wireless driver a bit This saves over 30 lines and fixes a warning from sparse and allows debugging to work dynamically like all other usb-serial drivers. Cc: Kevin Lloyd Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c index 39799d2..6bdb117 100644 --- a/drivers/usb/serial/sierra.c +++ b/drivers/usb/serial/sierra.c @@ -29,30 +29,6 @@ #include #include -/* Function prototypes */ -static int sierra_open(struct usb_serial_port *port, struct file *filp); -static void sierra_close(struct usb_serial_port *port, struct file *filp); -static int sierra_startup(struct usb_serial *serial); -static void sierra_shutdown(struct usb_serial *serial); -static void sierra_rx_throttle(struct usb_serial_port *port); -static void sierra_rx_unthrottle(struct usb_serial_port *port); -static int sierra_write_room(struct usb_serial_port *port); - -static void sierra_instat_callback(struct urb *urb); - -static int sierra_write(struct usb_serial_port *port, - const unsigned char *buf, int count); - -static int sierra_chars_in_buffer(struct usb_serial_port *port); -static int sierra_ioctl(struct usb_serial_port *port, struct file *file, - unsigned int cmd, unsigned long arg); -static void sierra_set_termios(struct usb_serial_port *port, - struct termios *old); -static void sierra_break_ctl(struct usb_serial_port *port, int break_state); -static int sierra_tiocmget(struct usb_serial_port *port, struct file *file); -static int sierra_tiocmset(struct usb_serial_port *port, struct file *file, - unsigned int set, unsigned int clear); -static int sierra_send_setup(struct usb_serial_port *port); static struct usb_device_id id_table [] = { { USB_DEVICE(0x1199, 0x0018) }, /* Sierra Wireless MC5720 */ @@ -68,6 +44,7 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x0F3D, 0x0112) }, /* AirPrime/Sierra PC 5220 */ { } }; +MODULE_DEVICE_TABLE(usb, id_table); static struct usb_device_id id_table_1port [] = { { USB_DEVICE(0x1199, 0x0112) }, /* Sierra Wireless AirCard 580 */ @@ -87,88 +64,22 @@ static struct usb_device_id id_table_3port [] = { { } }; - -MODULE_DEVICE_TABLE(usb, id_table); - static struct usb_driver sierra_driver = { .name = "sierra", .probe = usb_serial_probe, .disconnect = usb_serial_disconnect, .id_table = id_table, - .no_dynamic_id = 3, + .no_dynamic_id = 1, }; -//static struct usb_serial_driver *sierra_device; - -static struct usb_serial_driver sierra_1port_device = { - .driver = { - .owner = THIS_MODULE, - .name = "sierra1", - }, - .description = "Sierra USB modem (1 port)", - .id_table = id_table_1port, - .num_interrupt_in = NUM_DONT_CARE, - .num_bulk_in = 1, - .num_bulk_out = 1, - .num_ports = 1, - .open = sierra_open, - .close = sierra_close, - .write = sierra_write, - .write_room = sierra_write_room, - .chars_in_buffer = sierra_chars_in_buffer, - .throttle = sierra_rx_throttle, - .unthrottle = sierra_rx_unthrottle, - .ioctl = sierra_ioctl, - .set_termios = sierra_set_termios, - .break_ctl = sierra_break_ctl, - .tiocmget = sierra_tiocmget, - .tiocmset = sierra_tiocmset, - .attach = sierra_startup, - .shutdown = sierra_shutdown, - .read_int_callback = sierra_instat_callback, -}; - -static struct usb_serial_driver sierra_3port_device = { - .driver = { - .owner = THIS_MODULE, - .name = "sierra3", - }, - .description = "Sierra USB modem (3 port)", - .id_table = id_table_3port, - .num_interrupt_in = NUM_DONT_CARE, - .num_bulk_in = 3, - .num_bulk_out = 3, - .num_ports = 3, - .open = sierra_open, - .close = sierra_close, - .write = sierra_write, - .write_room = sierra_write_room, - .chars_in_buffer = sierra_chars_in_buffer, - .throttle = sierra_rx_throttle, - .unthrottle = sierra_rx_unthrottle, - .ioctl = sierra_ioctl, - .set_termios = sierra_set_termios, - .break_ctl = sierra_break_ctl, - .tiocmget = sierra_tiocmget, - .tiocmset = sierra_tiocmset, - .attach = sierra_startup, - .shutdown = sierra_shutdown, - .read_int_callback = sierra_instat_callback, -}; - -#ifdef CONFIG_USB_DEBUG static int debug; -#else -#define debug 0 -#endif /* per port private data */ - -#define N_IN_URB 4 -#define N_OUT_URB 1 -#define IN_BUFLEN 4096 -#define OUT_BUFLEN 128 +#define N_IN_URB 4 +#define N_OUT_URB 1 +#define IN_BUFLEN 4096 +#define OUT_BUFLEN 128 struct sierra_port_private { /* Input endpoints and buffer for this port */ @@ -189,44 +100,30 @@ struct sierra_port_private { unsigned long tx_start_time[N_OUT_URB]; }; -/* Functions used by new usb-serial code. */ -static int __init sierra_init(void) +static int sierra_send_setup(struct usb_serial_port *port) { - int retval; - retval = usb_serial_register(&sierra_1port_device); - if (retval) - goto failed_1port_device_register; - retval = usb_serial_register(&sierra_3port_device); - if (retval) - goto failed_3port_device_register; - + struct usb_serial *serial = port->serial; + struct sierra_port_private *portdata; - retval = usb_register(&sierra_driver); - if (retval) - goto failed_driver_register; + dbg("%s", __FUNCTION__); - info(DRIVER_DESC ": " DRIVER_VERSION); + portdata = usb_get_serial_port_data(port); - return 0; + if (port->tty) { + int val = 0; + if (portdata->dtr_state) + val |= 0x01; + if (portdata->rts_state) + val |= 0x02; -failed_driver_register: - usb_serial_deregister(&sierra_3port_device); -failed_3port_device_register: - usb_serial_deregister(&sierra_1port_device); -failed_1port_device_register: - return retval; -} + return usb_control_msg(serial->dev, + usb_rcvctrlpipe(serial->dev, 0), + 0x22,0x21,val,0,NULL,0,USB_CTRL_SET_TIMEOUT); + } -static void __exit sierra_exit(void) -{ - usb_deregister (&sierra_driver); - usb_serial_deregister(&sierra_1port_device); - usb_serial_deregister(&sierra_3port_device); + return 0; } -module_init(sierra_init); -module_exit(sierra_exit); - static void sierra_rx_throttle(struct usb_serial_port *port) { dbg("%s", __FUNCTION__); @@ -578,8 +475,8 @@ static void sierra_close(struct usb_serial_port *port, struct file *filp) /* Helper functions used by sierra_setup_urbs */ static struct urb *sierra_setup_urb(struct usb_serial *serial, int endpoint, - int dir, void *ctx, char *buf, int len, - void (*callback)(struct urb *)) + int dir, void *ctx, char *buf, int len, + usb_complete_t callback) { struct urb *urb; @@ -629,30 +526,6 @@ static void sierra_setup_urbs(struct usb_serial *serial) } } -static int sierra_send_setup(struct usb_serial_port *port) -{ - struct usb_serial *serial = port->serial; - struct sierra_port_private *portdata; - - dbg("%s", __FUNCTION__); - - portdata = usb_get_serial_port_data(port); - - if (port->tty) { - int val = 0; - if (portdata->dtr_state) - val |= 0x01; - if (portdata->rts_state) - val |= 0x02; - - return usb_control_msg(serial->dev, - usb_rcvctrlpipe(serial->dev, 0), - 0x22,0x21,val,0,NULL,0,USB_CTRL_SET_TIMEOUT); - } - - return 0; -} - static int sierra_startup(struct usb_serial *serial) { int i, err; @@ -730,6 +603,100 @@ static void sierra_shutdown(struct usb_serial *serial) } } +static struct usb_serial_driver sierra_1port_device = { + .driver = { + .owner = THIS_MODULE, + .name = "sierra1", + }, + .description = "Sierra USB modem (1 port)", + .id_table = id_table_1port, + .num_interrupt_in = NUM_DONT_CARE, + .num_bulk_in = 1, + .num_bulk_out = 1, + .num_ports = 1, + .open = sierra_open, + .close = sierra_close, + .write = sierra_write, + .write_room = sierra_write_room, + .chars_in_buffer = sierra_chars_in_buffer, + .throttle = sierra_rx_throttle, + .unthrottle = sierra_rx_unthrottle, + .ioctl = sierra_ioctl, + .set_termios = sierra_set_termios, + .break_ctl = sierra_break_ctl, + .tiocmget = sierra_tiocmget, + .tiocmset = sierra_tiocmset, + .attach = sierra_startup, + .shutdown = sierra_shutdown, + .read_int_callback = sierra_instat_callback, +}; + +static struct usb_serial_driver sierra_3port_device = { + .driver = { + .owner = THIS_MODULE, + .name = "sierra3", + }, + .description = "Sierra USB modem (3 port)", + .id_table = id_table_3port, + .num_interrupt_in = NUM_DONT_CARE, + .num_bulk_in = 3, + .num_bulk_out = 3, + .num_ports = 3, + .open = sierra_open, + .close = sierra_close, + .write = sierra_write, + .write_room = sierra_write_room, + .chars_in_buffer = sierra_chars_in_buffer, + .throttle = sierra_rx_throttle, + .unthrottle = sierra_rx_unthrottle, + .ioctl = sierra_ioctl, + .set_termios = sierra_set_termios, + .break_ctl = sierra_break_ctl, + .tiocmget = sierra_tiocmget, + .tiocmset = sierra_tiocmset, + .attach = sierra_startup, + .shutdown = sierra_shutdown, + .read_int_callback = sierra_instat_callback, +}; + +/* Functions used by new usb-serial code. */ +static int __init sierra_init(void) +{ + int retval; + retval = usb_serial_register(&sierra_1port_device); + if (retval) + goto failed_1port_device_register; + retval = usb_serial_register(&sierra_3port_device); + if (retval) + goto failed_3port_device_register; + + + retval = usb_register(&sierra_driver); + if (retval) + goto failed_driver_register; + + info(DRIVER_DESC ": " DRIVER_VERSION); + + return 0; + +failed_driver_register: + usb_serial_deregister(&sierra_3port_device); +failed_3port_device_register: + usb_serial_deregister(&sierra_1port_device); +failed_1port_device_register: + return retval; +} + +static void __exit sierra_exit(void) +{ + usb_deregister (&sierra_driver); + usb_serial_deregister(&sierra_1port_device); + usb_serial_deregister(&sierra_3port_device); +} + +module_init(sierra_init); +module_exit(sierra_exit); + MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_VERSION(DRIVER_VERSION); -- cgit v0.10.2 From ab352c2687a4361aec06a184ddb20deb1e5091eb Mon Sep 17 00:00:00 2001 From: Jan Luebbe Date: Tue, 17 Oct 2006 00:09:00 +0200 Subject: USB: Add device id for Sierra Wireless MC8755 Adds the device id used by the UMTS cards in Lenovo X60s notebooks sold in Europe. Signed-off-by: Jan Luebbe Cc: Kevin Lloyd Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c index 6bdb117..ea16572 100644 --- a/drivers/usb/serial/sierra.c +++ b/drivers/usb/serial/sierra.c @@ -37,6 +37,7 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x1199, 0x0019) }, /* Sierra Wireless AirCard 595 */ { USB_DEVICE(0x1199, 0x6802) }, /* Sierra Wireless MC8755 */ { USB_DEVICE(0x1199, 0x6803) }, /* Sierra Wireless MC8765 */ + { USB_DEVICE(0x1199, 0x6804) }, /* Sierra Wireless MC8755 for Europe */ { USB_DEVICE(0x1199, 0x6812) }, /* Sierra Wireless MC8775 */ { USB_DEVICE(0x1199, 0x6820) }, /* Sierra Wireless AirCard 875 */ -- cgit v0.10.2 From ffc45571dfb4b70e7eda8d97f64a05f5e5a992ac Mon Sep 17 00:00:00 2001 From: Aron Griffis Date: Tue, 17 Oct 2006 00:28:15 -0400 Subject: [IA64] move ioremap/ioremap_nocache under __KERNEL__ I noticed these are declared extern outside of __KERNEL__, but surely they wouldn't be available to userland since they're defined in ioremap.c. Am I missing something here? If I'm right about this, then there's probably a good deal of other stuff in io.h that could move inside __KERNEL__, but at least this is a start. Signed-off-by: Aron Griffis Signed-off-by: Tony Luck diff --git a/include/asm-ia64/io.h b/include/asm-ia64/io.h index 43bfff6..855c30a 100644 --- a/include/asm-ia64/io.h +++ b/include/asm-ia64/io.h @@ -417,6 +417,8 @@ __writeq (unsigned long val, volatile void __iomem *addr) # define outl_p outl #endif +# ifdef __KERNEL__ + extern void __iomem * ioremap(unsigned long offset, unsigned long size); extern void __iomem * ioremap_nocache (unsigned long offset, unsigned long size); @@ -430,8 +432,6 @@ iounmap (volatile void __iomem *addr) #define dmi_iounmap(x,l) iounmap(x) #define dmi_alloc(l) kmalloc(l, GFP_ATOMIC) -# ifdef __KERNEL__ - /* * String version of IO memory access ops: */ -- cgit v0.10.2 From accaddb27a2d544e38e10ff2a2782b33bbbad913 Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Mon, 16 Oct 2006 12:56:54 -0500 Subject: [IA64] - Allow IPIs in timer loop Allow pending IPIs to interrupt a timer interrupt that is looping in the do_timer() "while" loop in timer_interrupt(). (Interrupts are allowed at only 1 spot in the code). Signed-off-by: Jack Steiner Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 41169a9..39e0cd3 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -84,6 +84,12 @@ timer_interrupt (int irq, void *dev_id) if (time_after(new_itm, ia64_get_itc())) break; + + /* + * Allow IPIs to interrupt the timer loop. + */ + local_irq_enable(); + local_irq_disable(); } do { -- cgit v0.10.2 From c12fb1885787dcc2e20c4b88149e1e607e1293b2 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 12 Oct 2006 16:20:59 -0600 Subject: [IA64] remove unused PAL_CALL_IC_OFF Linux maps PAL instructions with an ITR, but uses a DTC for PAL data. Section 11.10.2.1.3, "Making PAL Procedures Calls in Physical or Virtual Mode," of the SDM (rev 2.2), says we must therefore make all PAL calls with PSR.ic = 1 so that Linux can handle any TLB faults. PAL_CALL_IC_OFF is currently unused, and as long as we use the ITR + DTC strategy, we can't use it. So remove it. I also removed the code in ia64_pal_call_static() that conditionally cleared PSR.ic. Signed-off-by: Bjorn Helgaas Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/pal.S b/arch/ia64/kernel/pal.S index ebaf1e6..5f50e6b 100644 --- a/arch/ia64/kernel/pal.S +++ b/arch/ia64/kernel/pal.S @@ -50,12 +50,10 @@ END(ia64_pal_default_handler) * * in0 Index of PAL service * in1 - in3 Remaining PAL arguments - * in4 1 ==> clear psr.ic, 0 ==> don't clear psr.ic - * */ GLOBAL_ENTRY(ia64_pal_call_static) - .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(5) - alloc loc1 = ar.pfs,5,5,0,0 + .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(4) + alloc loc1 = ar.pfs,4,5,0,0 movl loc2 = pal_entry_point 1: { mov r28 = in0 @@ -64,7 +62,6 @@ GLOBAL_ENTRY(ia64_pal_call_static) } ;; ld8 loc2 = [loc2] // loc2 <- entry point - tbit.nz p6,p7 = in4, 0 adds r8 = 1f-1b,r8 mov loc4=ar.rsc // save RSE configuration ;; @@ -74,13 +71,11 @@ GLOBAL_ENTRY(ia64_pal_call_static) .body mov r30 = in2 -(p6) rsm psr.i | psr.ic mov r31 = in3 mov b7 = loc2 -(p7) rsm psr.i + rsm psr.i ;; -(p6) srlz.i mov rp = r8 br.cond.sptk.many b7 1: mov psr.l = loc3 diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h index 2c8fd92..4283ddc 100644 --- a/include/asm-ia64/pal.h +++ b/include/asm-ia64/pal.h @@ -764,7 +764,7 @@ struct ia64_pal_retval { * (generally 0) MUST be passed. Reserved parameters are not optional * parameters. */ -extern struct ia64_pal_retval ia64_pal_call_static (u64, u64, u64, u64, u64); +extern struct ia64_pal_retval ia64_pal_call_static (u64, u64, u64, u64); extern struct ia64_pal_retval ia64_pal_call_stacked (u64, u64, u64, u64); extern struct ia64_pal_retval ia64_pal_call_phys_static (u64, u64, u64, u64); extern struct ia64_pal_retval ia64_pal_call_phys_stacked (u64, u64, u64, u64); @@ -774,14 +774,7 @@ extern void ia64_load_scratch_fpregs (struct ia64_fpreg *); #define PAL_CALL(iprv,a0,a1,a2,a3) do { \ struct ia64_fpreg fr[6]; \ ia64_save_scratch_fpregs(fr); \ - iprv = ia64_pal_call_static(a0, a1, a2, a3, 0); \ - ia64_load_scratch_fpregs(fr); \ -} while (0) - -#define PAL_CALL_IC_OFF(iprv,a0,a1,a2,a3) do { \ - struct ia64_fpreg fr[6]; \ - ia64_save_scratch_fpregs(fr); \ - iprv = ia64_pal_call_static(a0, a1, a2, a3, 1); \ + iprv = ia64_pal_call_static(a0, a1, a2, a3); \ ia64_load_scratch_fpregs(fr); \ } while (0) -- cgit v0.10.2 From 4d5a31977cc69be8786e0033d5c148e5c62ae949 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 12 Oct 2006 16:21:17 -0600 Subject: [IA64] reformat pal.S to fit in 80 columns, fix typos Reformat to fit in 80 columns. Fix a couple typos. Remove a couple unused labels. Signed-off-by: Bjorn Helgaas Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/pal.S b/arch/ia64/kernel/pal.S index 5f50e6b..0b53344 100644 --- a/arch/ia64/kernel/pal.S +++ b/arch/ia64/kernel/pal.S @@ -21,11 +21,12 @@ pal_entry_point: .text /* - * Set the PAL entry point address. This could be written in C code, but we do it here - * to keep it all in one module (besides, it's so trivial that it's + * Set the PAL entry point address. This could be written in C code, but we + * do it here to keep it all in one module (besides, it's so trivial that it's * not a big deal). * - * in0 Address of the PAL entry point (text address, NOT a function descriptor). + * in0 Address of the PAL entry point (text address, NOT a function + * descriptor). */ GLOBAL_ENTRY(ia64_pal_handler_init) alloc r3=ar.pfs,1,0,0,0 @@ -36,9 +37,9 @@ GLOBAL_ENTRY(ia64_pal_handler_init) END(ia64_pal_handler_init) /* - * Default PAL call handler. This needs to be coded in assembly because it uses - * the static calling convention, i.e., the RSE may not be used and calls are - * done via "br.cond" (not "br.call"). + * Default PAL call handler. This needs to be coded in assembly because it + * uses the static calling convention, i.e., the RSE may not be used and + * calls are done via "br.cond" (not "br.call"). */ GLOBAL_ENTRY(ia64_pal_default_handler) mov r8=-1 @@ -91,8 +92,8 @@ END(ia64_pal_call_static) * Make a PAL call using the stacked registers calling convention. * * Inputs: - * in0 Index of PAL service - * in2 - in3 Remaning PAL arguments + * in0 Index of PAL service + * in2 - in3 Remaining PAL arguments */ GLOBAL_ENTRY(ia64_pal_call_stacked) .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(4) @@ -126,18 +127,18 @@ END(ia64_pal_call_stacked) * Make a physical mode PAL call using the static registers calling convention. * * Inputs: - * in0 Index of PAL service - * in2 - in3 Remaning PAL arguments + * in0 Index of PAL service + * in2 - in3 Remaining PAL arguments * * PSR_LP, PSR_TB, PSR_ID, PSR_DA are never set by the kernel. * So we don't need to clear them. */ -#define PAL_PSR_BITS_TO_CLEAR \ - (IA64_PSR_I | IA64_PSR_IT | IA64_PSR_DT | IA64_PSR_DB | IA64_PSR_RT | \ - IA64_PSR_DD | IA64_PSR_SS | IA64_PSR_RI | IA64_PSR_ED | \ +#define PAL_PSR_BITS_TO_CLEAR \ + (IA64_PSR_I | IA64_PSR_IT | IA64_PSR_DT | IA64_PSR_DB | IA64_PSR_RT |\ + IA64_PSR_DD | IA64_PSR_SS | IA64_PSR_RI | IA64_PSR_ED | \ IA64_PSR_DFL | IA64_PSR_DFH) -#define PAL_PSR_BITS_TO_SET \ +#define PAL_PSR_BITS_TO_SET \ (IA64_PSR_BN) @@ -173,7 +174,7 @@ GLOBAL_ENTRY(ia64_pal_call_phys_static) ;; andcm r16=loc3,r16 // removes bits to clear from psr br.call.sptk.many rp=ia64_switch_mode_phys -.ret1: mov rp = r8 // install return address (physical) + mov rp = r8 // install return address (physical) mov loc5 = r19 mov loc6 = r20 br.cond.sptk.many b7 @@ -183,7 +184,6 @@ GLOBAL_ENTRY(ia64_pal_call_phys_static) mov r19=loc5 mov r20=loc6 br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode -.ret2: mov psr.l = loc3 // restore init PSR mov ar.pfs = loc1 @@ -198,8 +198,8 @@ END(ia64_pal_call_phys_static) * Make a PAL call using the stacked registers in physical mode. * * Inputs: - * in0 Index of PAL service - * in2 - in3 Remaning PAL arguments + * in0 Index of PAL service + * in2 - in3 Remaining PAL arguments */ GLOBAL_ENTRY(ia64_pal_call_phys_stacked) .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(5) @@ -207,7 +207,7 @@ GLOBAL_ENTRY(ia64_pal_call_phys_stacked) movl loc2 = pal_entry_point 1: { mov r28 = in0 // copy procedure index - mov loc0 = rp // save rp + mov loc0 = rp // save rp } .body ;; @@ -240,7 +240,7 @@ GLOBAL_ENTRY(ia64_pal_call_phys_stacked) mov r16=loc3 // r16= original psr mov r19=loc5 mov r20=loc6 - br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode + br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode mov psr.l = loc3 // restore init PSR mov ar.pfs = loc1 @@ -252,10 +252,11 @@ GLOBAL_ENTRY(ia64_pal_call_phys_stacked) END(ia64_pal_call_phys_stacked) /* - * Save scratch fp scratch regs which aren't saved in pt_regs already (fp10-fp15). + * Save scratch fp scratch regs which aren't saved in pt_regs already + * (fp10-fp15). * - * NOTE: We need to do this since firmware (SAL and PAL) may use any of the scratch - * regs fp-low partition. + * NOTE: We need to do this since firmware (SAL and PAL) may use any of the + * scratch regs fp-low partition. * * Inputs: * in0 Address of stack storage for fp regs -- cgit v0.10.2 From 8611342f3dda0e13970e724129a8e2066de1362e Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 9 Oct 2006 05:24:57 -0400 Subject: [IA64] update sn2_defconfig The current sn2_defconfig is obsolete, in particular the recent ATA changes are a pain, so here's a patch to get things in sync ... at least for a day or two :) Update sn_defconfig to match current community kernel tree. Signed-off-by: Jes Sorensen Signed-off-by: Tony Luck diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig index 0f14a82..64e951d 100644 --- a/arch/ia64/configs/sn2_defconfig +++ b/arch/ia64/configs/sn2_defconfig @@ -1,8 +1,9 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.17-rc3 -# Thu Apr 27 11:48:23 2006 +# Linux kernel version: 2.6.19-rc1 +# Mon Oct 9 10:53:59 2006 # +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # # Code maturity level options @@ -18,16 +19,22 @@ CONFIG_LOCALVERSION="" # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SWAP=y CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set CONFIG_POSIX_MQUEUE=y # CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y +CONFIG_TASKSTATS=y +# CONFIG_TASK_DELAY_ACCT is not set +# CONFIG_UTS_NS is not set # CONFIG_AUDIT is not set # CONFIG_IKCONFIG is not set CONFIG_CPUSETS=y CONFIG_RELAY=y CONFIG_INITRAMFS_SOURCE="" CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_TASK_XACCT=y +CONFIG_SYSCTL=y # CONFIG_EMBEDDED is not set +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -40,6 +47,8 @@ CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y +CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -58,6 +67,7 @@ CONFIG_STOP_MACHINE=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_BLK_DEV_IO_TRACE is not set # @@ -89,7 +99,7 @@ CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y CONFIG_IA64_UNCACHED_ALLOCATOR=y -CONFIG_DMA_IS_DMA32=y +CONFIG_AUDIT_ARCH=y # CONFIG_IA64_GENERIC is not set # CONFIG_IA64_DIG is not set # CONFIG_IA64_HP_ZX1 is not set @@ -116,6 +126,7 @@ CONFIG_FORCE_MAX_ZONEORDER=17 CONFIG_SMP=y CONFIG_NR_CPUS=1024 # CONFIG_HOTPLUG_CPU is not set +CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y CONFIG_SCHED_SMT=y CONFIG_PREEMPT=y CONFIG_SELECT_MEMORY_MODEL=y @@ -128,6 +139,7 @@ CONFIG_NEED_MULTIPLE_NODES=y # CONFIG_SPARSEMEM_STATIC is not set CONFIG_SPLIT_PTLOCK_CPUS=4 CONFIG_MIGRATION=y +CONFIG_RESOURCES_64BIT=y CONFIG_ARCH_SELECT_MEMORY_MODEL=y CONFIG_ARCH_DISCONTIGMEM_ENABLE=y CONFIG_ARCH_FLATMEM_ENABLE=y @@ -135,15 +147,24 @@ CONFIG_ARCH_SPARSEMEM_ENABLE=y CONFIG_ARCH_DISCONTIGMEM_DEFAULT=y CONFIG_NUMA=y CONFIG_NODES_SHIFT=10 +CONFIG_ARCH_POPULATES_NODE_MAP=y CONFIG_VIRTUAL_MEM_MAP=y CONFIG_HOLES_IN_ZONE=y CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y +CONFIG_HAVE_ARCH_NODEDATA_EXTENSION=y CONFIG_IA32_SUPPORT=y CONFIG_COMPAT=y CONFIG_IA64_MCA_RECOVERY=y CONFIG_PERFMON=y CONFIG_IA64_PALINFO=y CONFIG_SGI_SN=y +# CONFIG_IA64_ESI is not set + +# +# SN Devices +# +CONFIG_SGI_IOC4=y +CONFIG_SGI_IOC3=y # # Firmware Drivers @@ -159,6 +180,7 @@ CONFIG_BINFMT_ELF=y CONFIG_PM=y # CONFIG_PM_LEGACY is not set # CONFIG_PM_DEBUG is not set +# CONFIG_PM_SYSFS_DEPRECATED is not set # # ACPI (Advanced Configuration and Power Interface) Support @@ -166,6 +188,7 @@ CONFIG_PM=y CONFIG_ACPI=y # CONFIG_ACPI_BUTTON is not set # CONFIG_ACPI_FAN is not set +# CONFIG_ACPI_DOCK is not set # CONFIG_ACPI_PROCESSOR is not set CONFIG_ACPI_NUMA=y CONFIG_ACPI_BLACKLIST_YEAR=0 @@ -185,7 +208,12 @@ CONFIG_ACPI_SYSTEM=y # CONFIG_PCI=y CONFIG_PCI_DOMAINS=y +CONFIG_PCIEPORTBUS=y +CONFIG_HOTPLUG_PCI_PCIE=y +# CONFIG_HOTPLUG_PCI_PCIE_POLL_EVENT_MODE is not set +CONFIG_PCIEAER=y # CONFIG_PCI_MSI is not set +# CONFIG_PCI_MULTITHREAD_PROBE is not set # CONFIG_PCI_DEBUG is not set # @@ -215,6 +243,9 @@ CONFIG_NET=y CONFIG_PACKET=y CONFIG_PACKET_MMAP=y CONFIG_UNIX=y +CONFIG_XFRM=y +# CONFIG_XFRM_USER is not set +# CONFIG_XFRM_SUB_POLICY is not set # CONFIG_NET_KEY is not set CONFIG_INET=y CONFIG_IP_MULTICAST=y @@ -231,19 +262,31 @@ CONFIG_SYN_COOKIES=y # CONFIG_INET_IPCOMP is not set # CONFIG_INET_XFRM_TUNNEL is not set # CONFIG_INET_TUNNEL is not set +CONFIG_INET_XFRM_MODE_TRANSPORT=y +CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=m CONFIG_INET_TCP_DIAG=m # CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" CONFIG_IPV6=m # CONFIG_IPV6_PRIVACY is not set # CONFIG_IPV6_ROUTER_PREF is not set # CONFIG_INET6_AH is not set # CONFIG_INET6_ESP is not set # CONFIG_INET6_IPCOMP is not set +# CONFIG_IPV6_MIP6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set +CONFIG_INET6_XFRM_MODE_TRANSPORT=m +CONFIG_INET6_XFRM_MODE_TUNNEL=m +CONFIG_INET6_XFRM_MODE_BEET=m +# CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set # CONFIG_IPV6_TUNNEL is not set +# CONFIG_IPV6_SUBTREES is not set +# CONFIG_IPV6_MULTIPLE_TABLES is not set +# CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set # @@ -269,7 +312,6 @@ CONFIG_IPV6=m # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -298,6 +340,7 @@ CONFIG_STANDALONE=y CONFIG_PREVENT_FIRMWARE_BUILD=y CONFIG_FW_LOADER=y # CONFIG_DEBUG_DRIVER is not set +# CONFIG_SYS_HYPERVISOR is not set # # Connector - unified userspace <-> kernelspace linker @@ -335,6 +378,7 @@ CONFIG_BLK_DEV_NBD=m CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set CONFIG_ATA_OVER_ETH=m @@ -381,6 +425,7 @@ CONFIG_IDEDMA_PCI_AUTO=y # CONFIG_BLK_DEV_CS5530 is not set # CONFIG_BLK_DEV_HPT34X is not set # CONFIG_BLK_DEV_HPT366 is not set +# CONFIG_BLK_DEV_JMICRON is not set # CONFIG_BLK_DEV_SC1200 is not set # CONFIG_BLK_DEV_PIIX is not set # CONFIG_BLK_DEV_IT821X is not set @@ -404,6 +449,7 @@ CONFIG_IDEDMA_AUTO=y # # CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y +CONFIG_SCSI_NETLINK=y CONFIG_SCSI_PROC_FS=y # @@ -425,12 +471,14 @@ CONFIG_SCSI_CONSTANTS=y # CONFIG_SCSI_LOGGING is not set # -# SCSI Transport Attributes +# SCSI Transports # CONFIG_SCSI_SPI_ATTRS=y CONFIG_SCSI_FC_ATTRS=y CONFIG_SCSI_ISCSI_ATTRS=m CONFIG_SCSI_SAS_ATTRS=y +CONFIG_SCSI_SAS_LIBSAS=y +# CONFIG_SCSI_SAS_LIBSAS_DEBUG is not set # # SCSI low-level drivers @@ -443,46 +491,82 @@ CONFIG_ISCSI_TCP=m # CONFIG_SCSI_AIC7XXX is not set # CONFIG_SCSI_AIC7XXX_OLD is not set # CONFIG_SCSI_AIC79XX is not set +# CONFIG_SCSI_AIC94XX is not set +# CONFIG_SCSI_ARCMSR is not set # CONFIG_MEGARAID_NEWGEN is not set # CONFIG_MEGARAID_LEGACY is not set # CONFIG_MEGARAID_SAS is not set -CONFIG_SCSI_SATA=y -# CONFIG_SCSI_SATA_AHCI is not set -# CONFIG_SCSI_SATA_SVW is not set -# CONFIG_SCSI_ATA_PIIX is not set -# CONFIG_SCSI_SATA_MV is not set -# CONFIG_SCSI_SATA_NV is not set -# CONFIG_SCSI_PDC_ADMA is not set -# CONFIG_SCSI_SATA_QSTOR is not set -# CONFIG_SCSI_SATA_PROMISE is not set -# CONFIG_SCSI_SATA_SX4 is not set -# CONFIG_SCSI_SATA_SIL is not set -# CONFIG_SCSI_SATA_SIL24 is not set -# CONFIG_SCSI_SATA_SIS is not set -# CONFIG_SCSI_SATA_ULI is not set -# CONFIG_SCSI_SATA_VIA is not set -CONFIG_SCSI_SATA_VITESSE=y +# CONFIG_SCSI_HPTIOP is not set # CONFIG_SCSI_DMX3191D is not set # CONFIG_SCSI_FUTURE_DOMAIN is not set # CONFIG_SCSI_IPS is not set # CONFIG_SCSI_INITIO is not set # CONFIG_SCSI_INIA100 is not set +# CONFIG_SCSI_STEX is not set # CONFIG_SCSI_SYM53C8XX_2 is not set # CONFIG_SCSI_IPR is not set CONFIG_SCSI_QLOGIC_1280=y CONFIG_SCSI_QLA_FC=y -CONFIG_SCSI_QLA2XXX_EMBEDDED_FIRMWARE=y -# CONFIG_SCSI_QLA21XX is not set -CONFIG_SCSI_QLA22XX=y -CONFIG_SCSI_QLA2300=y -CONFIG_SCSI_QLA2322=y -# CONFIG_SCSI_QLA24XX is not set +# CONFIG_SCSI_QLA_ISCSI is not set # CONFIG_SCSI_LPFC is not set # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set # CONFIG_SCSI_DEBUG is not set # +# Serial ATA (prod) and Parallel ATA (experimental) drivers +# +CONFIG_ATA=y +# CONFIG_SATA_AHCI is not set +# CONFIG_SATA_SVW is not set +# CONFIG_ATA_PIIX is not set +# CONFIG_SATA_MV is not set +# CONFIG_SATA_NV is not set +# CONFIG_PDC_ADMA is not set +# CONFIG_SATA_QSTOR is not set +# CONFIG_SATA_PROMISE is not set +# CONFIG_SATA_SX4 is not set +# CONFIG_SATA_SIL is not set +# CONFIG_SATA_SIL24 is not set +# CONFIG_SATA_SIS is not set +# CONFIG_SATA_ULI is not set +# CONFIG_SATA_VIA is not set +CONFIG_SATA_VITESSE=y +# CONFIG_PATA_ALI is not set +# CONFIG_PATA_AMD is not set +# CONFIG_PATA_ARTOP is not set +# CONFIG_PATA_ATIIXP is not set +# CONFIG_PATA_CMD64X is not set +# CONFIG_PATA_CS5520 is not set +# CONFIG_PATA_CS5530 is not set +# CONFIG_PATA_CYPRESS is not set +# CONFIG_PATA_EFAR is not set +# CONFIG_ATA_GENERIC is not set +# CONFIG_PATA_HPT366 is not set +# CONFIG_PATA_HPT37X is not set +# CONFIG_PATA_HPT3X2N is not set +# CONFIG_PATA_HPT3X3 is not set +# CONFIG_PATA_IT821X is not set +# CONFIG_PATA_JMICRON is not set +# CONFIG_PATA_TRIFLEX is not set +# CONFIG_PATA_MPIIX is not set +# CONFIG_PATA_OLDPIIX is not set +# CONFIG_PATA_NETCELL is not set +# CONFIG_PATA_NS87410 is not set +# CONFIG_PATA_OPTI is not set +# CONFIG_PATA_OPTIDMA is not set +# CONFIG_PATA_PDC_OLD is not set +# CONFIG_PATA_RADISYS is not set +# CONFIG_PATA_RZ1000 is not set +# CONFIG_PATA_SC1200 is not set +# CONFIG_PATA_SERVERWORKS is not set +# CONFIG_PATA_PDC2027X is not set +# CONFIG_PATA_SIL680 is not set +# CONFIG_PATA_SIS is not set +# CONFIG_PATA_VIA is not set +# CONFIG_PATA_WINBOND is not set + +# # Multi-device support (RAID and LVM) # CONFIG_MD=y @@ -491,12 +575,12 @@ CONFIG_MD_LINEAR=y CONFIG_MD_RAID0=y CONFIG_MD_RAID1=y # CONFIG_MD_RAID10 is not set -CONFIG_MD_RAID5=y +CONFIG_MD_RAID456=y # CONFIG_MD_RAID5_RESHAPE is not set -# CONFIG_MD_RAID6 is not set CONFIG_MD_MULTIPATH=y # CONFIG_MD_FAULTY is not set CONFIG_BLK_DEV_DM=y +# CONFIG_DM_DEBUG is not set CONFIG_DM_CRYPT=m CONFIG_DM_SNAPSHOT=m CONFIG_DM_MIRROR=m @@ -563,6 +647,7 @@ CONFIG_NETDEVICES=y # CONFIG_SK98LIN is not set CONFIG_TIGON3=y # CONFIG_BNX2 is not set +# CONFIG_QLA3XXX is not set # # Ethernet (10000 Mbit) @@ -571,6 +656,7 @@ CONFIG_CHELSIO_T1=m # CONFIG_IXGB is not set CONFIG_S2IO=m # CONFIG_S2IO_NAPI is not set +# CONFIG_MYRI10GE is not set # # Token Ring devices @@ -612,6 +698,7 @@ CONFIG_NET_POLL_CONTROLLER=y # Input device support # CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set # # Userland interfaces @@ -646,6 +733,7 @@ CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768 CONFIG_VT=y CONFIG_VT_CONSOLE=y CONFIG_HW_CONSOLE=y +# CONFIG_VT_HW_CONSOLE_BINDING is not set CONFIG_SERIAL_NONSTANDARD=y # CONFIG_COMPUTONE is not set # CONFIG_ROCKETPORT is not set @@ -659,10 +747,12 @@ CONFIG_SERIAL_NONSTANDARD=y # CONFIG_N_HDLC is not set # CONFIG_SPECIALIX is not set # CONFIG_SX is not set +# CONFIG_RIO is not set # CONFIG_STALDRV is not set CONFIG_SGI_SNSC=y CONFIG_SGI_TIOCX=y CONFIG_SGI_MBCS=m +CONFIG_MSPEC=y # # Serial drivers @@ -701,6 +791,7 @@ CONFIG_EFI_RTC=y # Ftape, the floppy tape device driver # CONFIG_AGP=y +# CONFIG_AGP_SIS is not set # CONFIG_AGP_VIA is not set CONFIG_AGP_SGI_TIOCA=y # CONFIG_DRM is not set @@ -730,7 +821,6 @@ CONFIG_MMTIMER=y # # Dallas's 1-wire bus # -# CONFIG_W1 is not set # # Hardware Monitoring support @@ -741,6 +831,7 @@ CONFIG_MMTIMER=y # # Misc devices # +# CONFIG_TIFM_CORE is not set # # Multimedia devices @@ -756,6 +847,7 @@ CONFIG_MMTIMER=y # # Graphics support # +CONFIG_FIRMWARE_EDID=y # CONFIG_FB is not set # @@ -764,6 +856,7 @@ CONFIG_MMTIMER=y CONFIG_VGA_CONSOLE=y # CONFIG_VGACON_SOFT_SCROLLBACK is not set CONFIG_DUMMY_CONSOLE=y +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set # # Sound @@ -794,6 +887,7 @@ CONFIG_USB=m CONFIG_USB_EHCI_HCD=m # CONFIG_USB_EHCI_SPLIT_ISO is not set # CONFIG_USB_EHCI_ROOT_HUB_TT is not set +# CONFIG_USB_EHCI_TT_NEWSCHED is not set # CONFIG_USB_ISP116X_HCD is not set CONFIG_USB_OHCI_HCD=m # CONFIG_USB_OHCI_BIG_ENDIAN is not set @@ -843,6 +937,7 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -874,15 +969,18 @@ CONFIG_USB_MON=y # # CONFIG_USB_EMI62 is not set # CONFIG_USB_EMI26 is not set +# CONFIG_USB_ADUTUX is not set # CONFIG_USB_AUERSWALD is not set # CONFIG_USB_RIO500 is not set # CONFIG_USB_LEGOTOWER is not set # CONFIG_USB_LCD is not set # CONFIG_USB_LED is not set +# CONFIG_USB_CYPRESS_CY7C63 is not set # CONFIG_USB_CYTHERM is not set -# CONFIG_USB_PHIDGETKIT is not set -# CONFIG_USB_PHIDGETSERVO is not set +# CONFIG_USB_PHIDGET is not set # CONFIG_USB_IDMOUSE is not set +# CONFIG_USB_FTDI_ELAN is not set +# CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set @@ -919,18 +1017,15 @@ CONFIG_USB_MON=y CONFIG_INFINIBAND=m # CONFIG_INFINIBAND_USER_MAD is not set CONFIG_INFINIBAND_USER_ACCESS=m +CONFIG_INFINIBAND_ADDR_TRANS=y CONFIG_INFINIBAND_MTHCA=m CONFIG_INFINIBAND_MTHCA_DEBUG=y +# CONFIG_INFINIBAND_AMSO1100 is not set CONFIG_INFINIBAND_IPOIB=m CONFIG_INFINIBAND_IPOIB_DEBUG=y # CONFIG_INFINIBAND_IPOIB_DEBUG_DATA is not set CONFIG_INFINIBAND_SRP=m - -# -# SN Devices -# -CONFIG_SGI_IOC4=y -CONFIG_SGI_IOC3=y +# CONFIG_INFINIBAND_ISER is not set # # EDAC - error detection and reporting (RAS) (EXPERIMENTAL) @@ -942,6 +1037,19 @@ CONFIG_SGI_IOC3=y # CONFIG_RTC_CLASS is not set # +# DMA Engine support +# +# CONFIG_DMA_ENGINE is not set + +# +# DMA Clients +# + +# +# DMA Devices +# + +# # File systems # CONFIG_EXT2_FS=y @@ -965,15 +1073,16 @@ CONFIG_REISERFS_FS_SECURITY=y # CONFIG_JFS_FS is not set CONFIG_FS_POSIX_ACL=y CONFIG_XFS_FS=y -CONFIG_XFS_EXPORT=y CONFIG_XFS_QUOTA=y # CONFIG_XFS_SECURITY is not set CONFIG_XFS_POSIX_ACL=y CONFIG_XFS_RT=y +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set CONFIG_INOTIFY=y +CONFIG_INOTIFY_USER=y CONFIG_QUOTA=y # CONFIG_QFMT_V1 is not set # CONFIG_QFMT_V2 is not set @@ -1007,8 +1116,10 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" # CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set CONFIG_HUGETLBFS=y CONFIG_HUGETLB_PAGE=y CONFIG_RAMFS=y @@ -1046,7 +1157,7 @@ CONFIG_NFSD_V4=y CONFIG_NFSD_TCP=y CONFIG_LOCKD=m CONFIG_LOCKD_V4=y -CONFIG_EXPORTFS=y +CONFIG_EXPORTFS=m CONFIG_NFS_COMMON=y CONFIG_SUNRPC=m CONFIG_SUNRPC_GSS=m @@ -1056,7 +1167,9 @@ CONFIG_SMB_FS=m # CONFIG_SMB_NLS_DEFAULT is not set CONFIG_CIFS=m # CONFIG_CIFS_STATS is not set +# CONFIG_CIFS_WEAK_PW_HASH is not set # CONFIG_CIFS_XATTR is not set +# CONFIG_CIFS_DEBUG2 is not set # CONFIG_CIFS_EXPERIMENTAL is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set @@ -1129,6 +1242,10 @@ CONFIG_NLS_ISO8859_1=y CONFIG_NLS_UTF8=y # +# Distributed Lock Manager +# + +# # Library routines # # CONFIG_CRC_CCITT is not set @@ -1138,9 +1255,11 @@ CONFIG_LIBCRC32C=m CONFIG_ZLIB_INFLATE=m CONFIG_ZLIB_DEFLATE=m CONFIG_GENERIC_ALLOCATOR=y +CONFIG_PLIST=y CONFIG_GENERIC_HARDIRQS=y CONFIG_GENERIC_IRQ_PROBE=y CONFIG_GENERIC_PENDING_IRQ=y +CONFIG_IRQ_PER_CPU=y # # Instrumentation Support @@ -1152,20 +1271,26 @@ CONFIG_GENERIC_PENDING_IRQ=y # Kernel hacking # # CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y CONFIG_MAGIC_SYSRQ=y +# CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=20 CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set -CONFIG_DEBUG_PREEMPT=y -# CONFIG_DEBUG_MUTEXES is not set +# CONFIG_DEBUG_RT_MUTEXES is not set +# CONFIG_RT_MUTEX_TESTER is not set # CONFIG_DEBUG_SPINLOCK is not set +# CONFIG_DEBUG_MUTEXES is not set +# CONFIG_DEBUG_RWSEMS is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set +# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set # CONFIG_DEBUG_KOBJECT is not set CONFIG_DEBUG_INFO=y # CONFIG_DEBUG_FS is not set # CONFIG_DEBUG_VM is not set +# CONFIG_DEBUG_LIST is not set CONFIG_FORCED_INLINING=y # CONFIG_RCU_TORTURE_TEST is not set CONFIG_IA64_GRANULE_16MB=y @@ -1186,6 +1311,10 @@ CONFIG_SYSVIPC_COMPAT=y # Cryptographic options # CONFIG_CRYPTO=y +CONFIG_CRYPTO_ALGAPI=y +CONFIG_CRYPTO_BLKCIPHER=m +CONFIG_CRYPTO_HASH=y +CONFIG_CRYPTO_MANAGER=m CONFIG_CRYPTO_HMAC=y # CONFIG_CRYPTO_NULL is not set # CONFIG_CRYPTO_MD4 is not set @@ -1195,6 +1324,8 @@ CONFIG_CRYPTO_SHA1=m # CONFIG_CRYPTO_SHA512 is not set # CONFIG_CRYPTO_WP512 is not set # CONFIG_CRYPTO_TGR192 is not set +CONFIG_CRYPTO_ECB=m +CONFIG_CRYPTO_CBC=m CONFIG_CRYPTO_DES=m # CONFIG_CRYPTO_BLOWFISH is not set # CONFIG_CRYPTO_TWOFISH is not set -- cgit v0.10.2 From 219902677351665bf0513115592a6dd665cf06e8 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Fri, 29 Sep 2006 12:58:31 -0600 Subject: [IA64] remove unused acpi_kbd_controller_present, acpi_legacy_devices Nobody uses either one anymore. Signed-off-by: Bjorn Helgaas Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 32c3abed..73ef4a85 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -64,9 +64,6 @@ EXPORT_SYMBOL(pm_idle); void (*pm_power_off) (void); EXPORT_SYMBOL(pm_power_off); -unsigned char acpi_kbd_controller_present = 1; -unsigned char acpi_legacy_devices; - unsigned int acpi_cpei_override; unsigned int acpi_cpei_phys_cpuid; @@ -628,12 +625,6 @@ static int __init acpi_parse_fadt(unsigned long phys_addr, unsigned long size) fadt = (struct fadt_descriptor *)fadt_header; - if (!(fadt->iapc_boot_arch & BAF_8042_KEYBOARD_CONTROLLER)) - acpi_kbd_controller_present = 0; - - if (fadt->iapc_boot_arch & BAF_LEGACY_DEVICES) - acpi_legacy_devices = 1; - acpi_register_gsi(fadt->sci_int, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW); return 0; } diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 5f2dcba..7a2d824 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -65,7 +65,6 @@ extern void sn_timer_init(void); extern unsigned long last_time_offset; extern void (*ia64_mark_idle) (int); extern void snidle(int); -extern unsigned char acpi_kbd_controller_present; extern unsigned long long (*ia64_printk_clock)(void); unsigned long sn_rtc_cycles_per_second; @@ -452,17 +451,6 @@ void __init sn_setup(char **cmdline_p) ia64_printk_clock = ia64_sn2_printk_clock; - /* - * Old PROMs do not provide an ACPI FADT. Disable legacy keyboard - * support here so we don't have to listen to failed keyboard probe - * messages. - */ - if (is_shub1() && version <= 0x0209 && acpi_kbd_controller_present) { - printk(KERN_INFO "Disabling legacy keyboard support as prom " - "is too old and doesn't provide FADT\n"); - acpi_kbd_controller_present = 0; - } - printk("SGI SAL version %x.%02x\n", version >> 8, version & 0x00FF); /* -- cgit v0.10.2 From 9b3377f9921766b6193861d0e3194854b1d765fe Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Mon, 16 Oct 2006 16:17:43 -0500 Subject: [IA64] Count resched interrupts Count the number of "resched" interrupts that each cpu receives. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index 68339dd0..9c6dafa 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -180,7 +180,9 @@ ia64_handle_irq (ia64_vector vector, struct pt_regs *regs) saved_tpr = ia64_getreg(_IA64_REG_CR_TPR); ia64_srlz_d(); while (vector != IA64_SPURIOUS_INT_VECTOR) { - if (!IS_RESCHEDULE(vector)) { + if (unlikely(IS_RESCHEDULE(vector))) + kstat_this_cpu.irqs[vector]++; + else { ia64_setreg(_IA64_REG_CR_TPR, vector); ia64_srlz_d(); @@ -225,7 +227,9 @@ void ia64_process_pending_intr(void) * Perform normal interrupt style processing */ while (vector != IA64_SPURIOUS_INT_VECTOR) { - if (!IS_RESCHEDULE(vector)) { + if (unlikely(IS_RESCHEDULE(vector))) + kstat_this_cpu.irqs[vector]++; + else { struct pt_regs *old_regs = set_irq_regs(NULL); ia64_setreg(_IA64_REG_CR_TPR, vector); @@ -258,11 +262,22 @@ void ia64_process_pending_intr(void) #ifdef CONFIG_SMP extern irqreturn_t handle_IPI (int irq, void *dev_id); +static irqreturn_t dummy_handler (int irq, void *dev_id) +{ + BUG(); +} + static struct irqaction ipi_irqaction = { .handler = handle_IPI, .flags = IRQF_DISABLED, .name = "IPI" }; + +static struct irqaction resched_irqaction = { + .handler = dummy_handler, + .flags = SA_INTERRUPT, + .name = "resched" +}; #endif void @@ -287,6 +302,7 @@ init_IRQ (void) register_percpu_irq(IA64_SPURIOUS_INT_VECTOR, NULL); #ifdef CONFIG_SMP register_percpu_irq(IA64_IPI_VECTOR, &ipi_irqaction); + register_percpu_irq(IA64_IPI_RESCHEDULE, &resched_irqaction); #endif #ifdef CONFIG_PERFMON pfm_init_percpu(); -- cgit v0.10.2 From 9950421c1e90d08d52ef47df1fcebe6078b04af3 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 17 Oct 2006 18:03:33 -0700 Subject: Fix USB gadget net2280.c compile It would fail the compile due to the newly added error checking testing a bad macro for a "return value" unless USB_GADGET_DEBUG_FILES was enabled. Pointed out by Stephen Hemminger. Acked-by: Stephen Hemminger Signed-off-by: Linus Torvalds diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c index 7cfe0e5..3acc896 100644 --- a/drivers/usb/gadget/net2280.c +++ b/drivers/usb/gadget/net2280.c @@ -1774,8 +1774,8 @@ static DEVICE_ATTR (queues, S_IRUGO, show_queues, NULL); #else -#define device_create_file(a,b) do {} while (0) -#define device_remove_file device_create_file +#define device_create_file(a,b) (0) +#define device_remove_file(a,b) do { } while (0) #endif -- cgit v0.10.2 From 7341df16a1f228be833e918cb42534a34381a37e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 17 Oct 2006 00:24:14 -0700 Subject: [SPARC]: Kill BOOTME_SINGLE. Unused, but still allow the '-s' boot option to be passed down to init. Based upon patches by Martin Habets. Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/setup.c b/arch/sparc/kernel/setup.c index f5ee1ac..383526a 100644 --- a/arch/sparc/kernel/setup.c +++ b/arch/sparc/kernel/setup.c @@ -103,7 +103,6 @@ void prom_sync_me(void) unsigned int boot_flags __initdata = 0; #define BOOTME_DEBUG 0x1 -#define BOOTME_SINGLE 0x2 /* Exported for mm/init.c:paging_init. */ unsigned long cmdline_memory_size __initdata = 0; @@ -132,7 +131,6 @@ static void __init process_switch(char c) boot_flags |= BOOTME_DEBUG; break; case 's': - boot_flags |= BOOTME_SINGLE; break; case 'h': prom_printf("boot_flags_init: Halt!\n"); diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c index cc8ad48..bf033b3 100644 --- a/arch/sparc64/kernel/setup.c +++ b/arch/sparc64/kernel/setup.c @@ -74,7 +74,6 @@ prom_console_write(struct console *con, const char *s, unsigned n) unsigned int boot_flags = 0; #define BOOTME_DEBUG 0x1 -#define BOOTME_SINGLE 0x2 /* Exported for mm/init.c:paging_init. */ unsigned long cmdline_memory_size = 0; @@ -102,7 +101,6 @@ static void __init process_switch(char c) boot_flags |= BOOTME_DEBUG; break; case 's': - boot_flags |= BOOTME_SINGLE; break; case 'h': prom_printf("boot_flags_init: Halt!\n"); -- cgit v0.10.2 From 5aee87c43e3a71a4aa4e72b0dc2180e4952c0848 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 17 Oct 2006 19:04:44 -0700 Subject: [SPARC64]: Fix PCI memory space root resource on Hummingbird. For Hummingbird PCI controllers, we should create the root PCI memory space resource as the full 4GB area, and then allocate the IOMMU DMA translation window out of there. The old code just assumed that the IOMMU DMA translation base to the top of the 4GB area was unusable. This is not true on many systems such as SB100 and SB150, where the IOMMU DMA translation window sits at 0xc0000000->0xdfffffff. So what would happen is that any device mapped by the firmware at the top section 0xe0000000->0xffffffff would get remapped by Linux somewhere else leading to all kinds of problems and boot failures. While we're here, report more cases of OBP resource assignment conflicts. The only truly valid ones are ROM resource conflicts. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/pci_common.c b/arch/sparc64/kernel/pci_common.c index 7a59cc7..827ae30 100644 --- a/arch/sparc64/kernel/pci_common.c +++ b/arch/sparc64/kernel/pci_common.c @@ -330,19 +330,6 @@ __init get_device_resource(struct linux_prom_pci_registers *ap, return res; } -static int __init pdev_resource_collisions_expected(struct pci_dev *pdev) -{ - if (pdev->vendor != PCI_VENDOR_ID_SUN) - return 0; - - if (pdev->device == PCI_DEVICE_ID_SUN_RIO_EBUS || - pdev->device == PCI_DEVICE_ID_SUN_RIO_1394 || - pdev->device == PCI_DEVICE_ID_SUN_RIO_USB) - return 1; - - return 0; -} - static void __init pdev_record_assignments(struct pci_pbm_info *pbm, struct pci_dev *pdev) { @@ -400,19 +387,23 @@ static void __init pdev_record_assignments(struct pci_pbm_info *pbm, pbm->parent->resource_adjust(pdev, res, root); if (request_resource(root, res) < 0) { + int rnum; + /* OK, there is some conflict. But this is fine * since we'll reassign it in the fixup pass. * - * We notify the user that OBP made an error if it - * is a case we don't expect. + * Do not print the warning for ROM resources + * as such a conflict is quite common and + * harmless as the ROM bar is disabled. */ - if (!pdev_resource_collisions_expected(pdev)) { - printk(KERN_ERR "PCI: Address space collision on region %ld " + rnum = (res - &pdev->resource[0]); + if (rnum != PCI_ROM_RESOURCE) + printk(KERN_ERR "PCI: Resource collision, " + "region %d " "[%016lx:%016lx] of device %s\n", - (res - &pdev->resource[0]), + rnum, res->start, res->end, pci_name(pdev)); - } } } } diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c index 6ec5698..de7f785 100644 --- a/arch/sparc64/kernel/pci_sabre.c +++ b/arch/sparc64/kernel/pci_sabre.c @@ -1196,7 +1196,7 @@ static void pbm_register_toplevel_resources(struct pci_controller_info *p, &pbm->mem_space); } -static void sabre_pbm_init(struct pci_controller_info *p, struct device_node *dp, u32 dma_begin) +static void sabre_pbm_init(struct pci_controller_info *p, struct device_node *dp, u32 dma_start, u32 dma_end) { struct pci_pbm_info *pbm; struct device_node *node; @@ -1261,6 +1261,8 @@ static void sabre_pbm_init(struct pci_controller_info *p, struct device_node *dp node = node->sibling; } if (simbas_found == 0) { + struct resource *rp; + /* No APBs underneath, probably this is a hummingbird * system. */ @@ -1302,8 +1304,10 @@ static void sabre_pbm_init(struct pci_controller_info *p, struct device_node *dp pbm->io_space.end = pbm->io_space.start + (1UL << 24) - 1UL; pbm->io_space.flags = IORESOURCE_IO; - pbm->mem_space.start = p->pbm_A.controller_regs + SABRE_MEMSPACE; - pbm->mem_space.end = pbm->mem_space.start + (unsigned long)dma_begin - 1UL; + pbm->mem_space.start = + (p->pbm_A.controller_regs + SABRE_MEMSPACE); + pbm->mem_space.end = + (pbm->mem_space.start + ((1UL << 32UL) - 1UL)); pbm->mem_space.flags = IORESOURCE_MEM; if (request_resource(&ioport_resource, &pbm->io_space) < 0) { @@ -1315,6 +1319,17 @@ static void sabre_pbm_init(struct pci_controller_info *p, struct device_node *dp prom_halt(); } + rp = kmalloc(sizeof(*rp), GFP_KERNEL); + if (!rp) { + prom_printf("Cannot allocate IOMMU resource.\n"); + prom_halt(); + } + rp->name = "IOMMU"; + rp->start = pbm->mem_space.start + (unsigned long) dma_start; + rp->end = pbm->mem_space.start + (unsigned long) dma_end - 1UL; + rp->flags = IORESOURCE_BUSY; + request_resource(&pbm->mem_space, rp); + pci_register_legacy_regions(&pbm->io_space, &pbm->mem_space); } @@ -1450,5 +1465,5 @@ void sabre_init(struct device_node *dp, char *model_name) /* * Look for APB underneath. */ - sabre_pbm_init(p, dp, vdma[0]); + sabre_pbm_init(p, dp, vdma[0], vdma[1]); } -- cgit v0.10.2 From b48194bf0dc0f8e2b617fab10df885513fbb3bad Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Tue, 17 Oct 2006 19:11:31 -0700 Subject: [SPARC]: Fix some section mismatch warnings in sparc drivers. Signed-off-by: Ben Collins Signed-off-by: David S. Miller diff --git a/drivers/net/myri_code.h b/drivers/net/myri_code.h index e21ec9b..ba7b865 100644 --- a/drivers/net/myri_code.h +++ b/drivers/net/myri_code.h @@ -1,8 +1,8 @@ /* This is the Myrinet MCP code for LANai4.x */ /* Generated by cat $MYRI_HOME/lib/lanai/mcp4.dat > myri_code4.h */ -static unsigned int lanai4_code_off = 0x0000; /* half-word offset */ -static unsigned char lanai4_code[76256] __initdata = { +static unsigned int __devinitdata lanai4_code_off = 0x0000; /* half-word offset */ +static unsigned char __devinitdata lanai4_code[76256] = { 0xF2,0x0E, 0xFE,0x00, 0xC2,0x90, 0x00,0x00, 0x07,0x88, 0x00,0x08, 0xE0,0x01, 0x01,0x4C, 0x97,0x93, 0xFF,0xFC, 0xE0,0x00, 0x00,0x14, 0x00,0x00, 0x00,0x01, 0x00,0x00, 0x00,0x00, 0x92,0x93, @@ -4774,8 +4774,8 @@ static unsigned char lanai4_code[76256] __initdata = { /* This is the LANai data */ -static unsigned int lanai4_data_off = 0x94F0; /* half-word offset */ -static unsigned char lanai4_data[20472] __initdata; +static unsigned int __devinitdata lanai4_data_off = 0x94F0; /* half-word offset */ +static unsigned char __devinitdata lanai4_data[20472]; #ifdef SYMBOL_DEFINES_COMPILED diff --git a/drivers/net/myri_sbus.c b/drivers/net/myri_sbus.c index 466b484..7747bfd 100644 --- a/drivers/net/myri_sbus.c +++ b/drivers/net/myri_sbus.c @@ -168,7 +168,7 @@ static int myri_do_handshake(struct myri_eth *mp) return 0; } -static int myri_load_lanai(struct myri_eth *mp) +static int __devinit myri_load_lanai(struct myri_eth *mp) { struct net_device *dev = mp->dev; struct myri_shmem __iomem *shmem = mp->shmem; @@ -891,7 +891,7 @@ static void dump_eeprom(struct myri_eth *mp) } #endif -static int __init myri_ether_init(struct sbus_dev *sdev) +static int __devinit myri_ether_init(struct sbus_dev *sdev) { static int num; static unsigned version_printed; diff --git a/drivers/net/sunbmac.c b/drivers/net/sunbmac.c index 6439b0c..18f8885 100644 --- a/drivers/net/sunbmac.c +++ b/drivers/net/sunbmac.c @@ -42,7 +42,7 @@ #define DRV_RELDATE "11/24/03" #define DRV_AUTHOR "David S. Miller (davem@redhat.com)" -static char version[] __initdata = +static char version[] = DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n"; MODULE_VERSION(DRV_VERSION); diff --git a/drivers/scsi/qlogicpti.c b/drivers/scsi/qlogicpti.c index ed58bb4..9b827ce 100644 --- a/drivers/scsi/qlogicpti.c +++ b/drivers/scsi/qlogicpti.c @@ -461,7 +461,7 @@ static int qlogicpti_reset_hardware(struct Scsi_Host *host) #define PTI_RESET_LIMIT 400 -static int __init qlogicpti_load_firmware(struct qlogicpti *qpti) +static int __devinit qlogicpti_load_firmware(struct qlogicpti *qpti) { struct Scsi_Host *host = qpti->qhost; unsigned short csum = 0; diff --git a/drivers/scsi/qlogicpti_asm.c b/drivers/scsi/qlogicpti_asm.c index 1545b30..19aa84f 100644 --- a/drivers/scsi/qlogicpti_asm.c +++ b/drivers/scsi/qlogicpti_asm.c @@ -1,5 +1,5 @@ /* Version 1.31.00 ISP1000 Initiator RISC firmware */ -unsigned short sbus_risc_code01[] __initdata = { +unsigned short sbus_risc_code01[] __devinitdata = { 0x0078, 0x1030, 0x0000, 0x2419, 0x0000, 0x12ff, 0x2043, 0x4f50, 0x5952, 0x4947, 0x4854, 0x2031, 0x3939, 0x312c, 0x3139, 0x3932, 0x2c31, 0x3939, 0x332c, 0x3139, 0x3934, 0x2051, 0x4c4f, 0x4749, @@ -1157,4 +1157,4 @@ unsigned short sbus_risc_code01[] __initdata = { 0x003c, 0x0040, 0x3415, 0x2019, 0x2626, 0x7b22, 0x7b26, 0x007c, 0x92a7 }; -unsigned short sbus_risc_code_length01 = 0x2419; +unsigned short __devinitdata sbus_risc_code_length01 = 0x2419; -- cgit v0.10.2 From 872ec6484720e7ddfebb8e15c232fa7ca158ef2e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 17 Oct 2006 19:19:08 -0700 Subject: [SPARC] {bbc_,}envctrl: Use call_usermodehelper(). We should not be calling kernel_execve() directly and this causes module build failures because kernel_execve() is not exported to modules. Signed-off-by: David S. Miller diff --git a/drivers/sbus/char/bbc_envctrl.c b/drivers/sbus/char/bbc_envctrl.c index 0d3660c..a54e414 100644 --- a/drivers/sbus/char/bbc_envctrl.c +++ b/drivers/sbus/char/bbc_envctrl.c @@ -5,8 +5,8 @@ */ #include -#include #include +#include #include #include @@ -195,7 +195,7 @@ static void do_envctrl_shutdown(struct bbc_cpu_temperature *tp) printk(KERN_CRIT "kenvctrld: Shutting down the system now.\n"); shutting_down = 1; - if (kernel_execve("/sbin/shutdown", argv, envp) < 0) + if (call_usermodehelper("/sbin/shutdown", argv, envp, 0) < 0) printk(KERN_CRIT "envctrl: shutdown execution failed\n"); } diff --git a/drivers/sbus/char/envctrl.c b/drivers/sbus/char/envctrl.c index 6b6a855..fff4660 100644 --- a/drivers/sbus/char/envctrl.c +++ b/drivers/sbus/char/envctrl.c @@ -25,7 +25,7 @@ #include #include #include -#include +#include #include #include @@ -976,7 +976,7 @@ static void envctrl_do_shutdown(void) inprog = 1; printk(KERN_CRIT "kenvctrld: WARNING: Shutting down the system now.\n"); - ret = kernel_execve("/sbin/shutdown", argv, envp); + ret = call_usermodehelper("/sbin/shutdown", argv, envp, 0); if (ret < 0) { printk(KERN_CRIT "kenvctrld: WARNING: system shutdown failed!\n"); inprog = 0; /* unlikely to succeed, but we could try again */ -- cgit v0.10.2 From 9550e59c4587f637d9aa34689e32eea460e6f50c Mon Sep 17 00:00:00 2001 From: Martin Habets Date: Tue, 17 Oct 2006 19:21:48 -0700 Subject: [SPARC]: Add sparc profiling support This patch adds profiling support to the sparc architecture. It is a copy of the sparc64 implementation. Signed-off-by: Martin Habets Signed-off-by: David S. Miller diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 9431e96..2f96610 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -289,6 +289,13 @@ endmenu source "fs/Kconfig" +menu "Instrumentation Support" + depends on EXPERIMENTAL + +source "arch/sparc/oprofile/Kconfig" + +endmenu + source "arch/sparc/Kconfig.debug" source "security/Kconfig" diff --git a/arch/sparc/Makefile b/arch/sparc/Makefile index 4cdbb2d..f33c381 100644 --- a/arch/sparc/Makefile +++ b/arch/sparc/Makefile @@ -30,6 +30,8 @@ HEAD_Y := $(head-y) core-y += arch/sparc/kernel/ arch/sparc/mm/ arch/sparc/math-emu/ libs-y += arch/sparc/prom/ arch/sparc/lib/ +drivers-$(CONFIG_OPROFILE) += arch/sparc/oprofile/ + # Export what is needed by arch/sparc/boot/Makefile # Renaming is done to avoid confusing pattern matching rules in 2.5.45 (multy-) INIT_Y := $(patsubst %/, %/built-in.o, $(init-y)) diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c index 7dcd1a1..6c7aa51 100644 --- a/arch/sparc/kernel/time.c +++ b/arch/sparc/kernel/time.c @@ -95,6 +95,8 @@ unsigned long profile_pc(struct pt_regs *regs) return pc; } +EXPORT_SYMBOL(profile_pc); + __volatile__ unsigned int *master_l10_counter; __volatile__ unsigned int *master_l10_limit; diff --git a/arch/sparc/oprofile/Kconfig b/arch/sparc/oprofile/Kconfig new file mode 100644 index 0000000..d8a8408 --- /dev/null +++ b/arch/sparc/oprofile/Kconfig @@ -0,0 +1,17 @@ +config PROFILING + bool "Profiling support (EXPERIMENTAL)" + help + Say Y here to enable the extended profiling support mechanisms used + by profilers such as OProfile. + + +config OPROFILE + tristate "OProfile system profiling (EXPERIMENTAL)" + depends on PROFILING + help + OProfile is a profiling system capable of profiling the + whole system, include the kernel, kernel modules, libraries, + and applications. + + If unsure, say N. + diff --git a/arch/sparc/oprofile/Makefile b/arch/sparc/oprofile/Makefile new file mode 100644 index 0000000..e9feca1 --- /dev/null +++ b/arch/sparc/oprofile/Makefile @@ -0,0 +1,9 @@ +obj-$(CONFIG_OPROFILE) += oprofile.o + +DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \ + oprof.o cpu_buffer.o buffer_sync.o \ + event_buffer.o oprofile_files.o \ + oprofilefs.o oprofile_stats.o \ + timer_int.o ) + +oprofile-y := $(DRIVER_OBJS) init.o diff --git a/arch/sparc/oprofile/init.c b/arch/sparc/oprofile/init.c new file mode 100644 index 0000000..9ab815b --- /dev/null +++ b/arch/sparc/oprofile/init.c @@ -0,0 +1,23 @@ +/** + * @file init.c + * + * @remark Copyright 2002 OProfile authors + * @remark Read the file COPYING + * + * @author John Levon + */ + +#include +#include +#include +#include + +int __init oprofile_arch_init(struct oprofile_operations * ops) +{ + return -ENODEV; +} + + +void oprofile_arch_exit(void) +{ +} -- cgit v0.10.2 From f6d7b8a7c9e21a58890a8532143e241409b78664 Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Tue, 17 Oct 2006 19:23:23 -0700 Subject: [SPARC]: Sparc compilation fix with floppy enabled This patch fixes a typo to make kernel compilable when floppy driver for Sparc is build. Signed-off-by: Krzysztof Helt Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/irq.c b/arch/sparc/kernel/irq.c index b3b6680..c8cb211 100644 --- a/arch/sparc/kernel/irq.c +++ b/arch/sparc/kernel/irq.c @@ -351,7 +351,7 @@ void handler_irq(int irq, struct pt_regs * regs) } #ifdef CONFIG_BLK_DEV_FD -extern void floppy_interrupt(int irq, void *dev_id) +extern void floppy_interrupt(int irq, void *dev_id); void sparc_floppy_irq(int irq, void *dev_id, struct pt_regs *regs) { -- cgit v0.10.2 From 73ed9a86cd77b6a3b46beec8d353ac3b0d4f50c1 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 17 Oct 2006 19:29:41 -0700 Subject: [SPARC64]: Update defconfig. Signed-off-by: David S. Miller diff --git a/arch/sparc64/defconfig b/arch/sparc64/defconfig index dcae559..2f4612f 100644 --- a/arch/sparc64/defconfig +++ b/arch/sparc64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc1 -# Thu Oct 5 02:08:41 2006 +# Linux kernel version: 2.6.19-rc2 +# Tue Oct 17 19:29:20 2006 # CONFIG_SPARC=y CONFIG_SPARC64=y @@ -217,6 +217,7 @@ CONFIG_INET6_XFRM_MODE_TRANSPORT=m CONFIG_INET6_XFRM_MODE_TUNNEL=m CONFIG_INET6_XFRM_MODE_BEET=m # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=m CONFIG_IPV6_TUNNEL=m # CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set @@ -334,6 +335,12 @@ CONFIG_CDROM_PKTCDVD_WCACHE=y CONFIG_ATA_OVER_ETH=m # +# Misc devices +# +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y @@ -728,7 +735,6 @@ CONFIG_RTC=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -842,11 +848,6 @@ CONFIG_HWMON=y # CONFIG_HWMON_DEBUG_CHIP is not set # -# Misc devices -# -# CONFIG_TIFM_CORE is not set - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set @@ -1104,7 +1105,6 @@ CONFIG_USB_HIDDEV=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set -# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -1150,6 +1150,7 @@ CONFIG_USB_HIDDEV=y # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1234,6 +1235,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y CONFIG_EXT3_FS_POSIX_ACL=y CONFIG_EXT3_FS_SECURITY=y +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1362,10 +1364,6 @@ CONFIG_NLS_DEFAULT="iso8859-1" # CONFIG_NLS_UTF8 is not set # -# Distributed Lock Manager -# - -# # Instrumentation Support # CONFIG_PROFILING=y @@ -1399,6 +1397,7 @@ CONFIG_DEBUG_FS=y # CONFIG_DEBUG_LIST is not set # CONFIG_UNWIND_INFO is not set CONFIG_FORCED_INLINING=y +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_LKDTM is not set # CONFIG_DEBUG_STACK_USAGE is not set -- cgit v0.10.2 From 5cacb9f8bca1ac67cc1f933a4e89b5130ffd1460 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:24:30 -0400 Subject: [alim7101] Add pci dev table for auto module loading. Also fixes comment for nowayout module param. Signed-off-by: Ben Collins diff --git a/drivers/char/watchdog/alim7101_wdt.c b/drivers/char/watchdog/alim7101_wdt.c index 5948863..bf25d0a 100644 --- a/drivers/char/watchdog/alim7101_wdt.c +++ b/drivers/char/watchdog/alim7101_wdt.c @@ -77,7 +77,8 @@ static struct pci_dev *alim7101_pmu; static int nowayout = WATCHDOG_NOWAYOUT; module_param(nowayout, int, 0); -MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CONFIG_WATCHDOG_NOWAYOUT)"); +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" + __stringify(CONFIG_WATCHDOG_NOWAYOUT) ")"); /* * Whack the dog @@ -415,6 +416,16 @@ err_out: module_init(alim7101_wdt_init); module_exit(alim7101_wdt_unload); +static struct pci_device_id alim7101_pci_tbl[] __devinitdata = { + { PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M7101, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { } +}; + +MODULE_DEVICE_TABLE(pci, alim7101_pci_tbl); + MODULE_AUTHOR("Steve Hill"); MODULE_DESCRIPTION("ALi M7101 PMU Computer Watchdog Timer driver"); MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 4596c75c23dde2623cbeec69357d5eb13d28387e Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:33:03 -0400 Subject: [mv643xx] Add pci device table for auto module loading. Signed-off-by: Ben Collins diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index 9997081..a4f861b 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -1557,6 +1557,12 @@ static void __exit mv643xx_cleanup_module(void) module_init(mv643xx_init_module); module_exit(mv643xx_cleanup_module); +static struct pci_device_id pci_marvell_mv64360[] = { + { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, PCI_DEVICE_ID_MARVELL_MV64360) }, + {} +}; +MODULE_DEVICE_TABLE(pci, pci_marvell_mv64360); + MODULE_LICENSE("GPL"); MODULE_AUTHOR( "Rabeeh Khoury, Assaf Hoffman, Matthew Dharm, Manish Lachwani" " and Dale Farnsworth"); -- cgit v0.10.2 From 745b5715fafccc8f0f992a7cccdd1eb2b1f5d23f Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:36:57 -0400 Subject: [BusLogic] Add pci dev table for auto module loading. Signed-off-by: Ben Collins diff --git a/drivers/scsi/BusLogic.c b/drivers/scsi/BusLogic.c index 7c59bba..cdd0337 100644 --- a/drivers/scsi/BusLogic.c +++ b/drivers/scsi/BusLogic.c @@ -3600,5 +3600,16 @@ static void __exit BusLogic_exit(void) __setup("BusLogic=", BusLogic_Setup); +static struct pci_device_id BusLogic_pci_tbl[] __devinitdata = { + { PCI_VENDOR_ID_BUSLOGIC, PCI_DEVICE_ID_BUSLOGIC_MULTIMASTER, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_BUSLOGIC, PCI_DEVICE_ID_BUSLOGIC_MULTIMASTER_NC, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_BUSLOGIC, PCI_DEVICE_ID_BUSLOGIC_FLASHPOINT, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { } +}; +MODULE_DEVICE_TABLE(pci, BusLogic_pci_tbl); + module_init(BusLogic_init); module_exit(BusLogic_exit); -- cgit v0.10.2 From 3985b977441f857404bb604a4b15911bbb0b9c5b Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:38:41 -0400 Subject: [fdomain] Add pci dev table for module auto loading. Signed-off-by: Ben Collins diff --git a/drivers/scsi/fdomain.c b/drivers/scsi/fdomain.c index 72794a7..65e6e7b 100644 --- a/drivers/scsi/fdomain.c +++ b/drivers/scsi/fdomain.c @@ -1736,6 +1736,15 @@ struct scsi_host_template fdomain_driver_template = { }; #ifndef PCMCIA + +static struct pci_device_id fdomain_pci_tbl[] __devinitdata = { + { PCI_VENDOR_ID_FD, PCI_DEVICE_ID_FD_36C70, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL }, + { } +}; +MODULE_DEVICE_TABLE(pci, fdomain_pci_tbl); + #define driver_template fdomain_driver_template #include "scsi_module.c" + #endif -- cgit v0.10.2 From 012887327875915f76a6208e81fe0d67a682ec15 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:40:57 -0400 Subject: [initio] Add pci dev table for module auto loading. Signed-off-by: Ben Collins diff --git a/drivers/scsi/initio.c b/drivers/scsi/initio.c index 911f2ff..afed293 100644 --- a/drivers/scsi/initio.c +++ b/drivers/scsi/initio.c @@ -142,8 +142,6 @@ #define i91u_MAXQUEUE 2 #define i91u_REVID "Initio INI-9X00U/UW SCSI device driver; Revision: 1.04a" -#define INI_VENDOR_ID 0x1101 /* Initio's PCI vendor ID */ -#define DMX_VENDOR_ID 0x134a /* Domex's PCI vendor ID */ #define I950_DEVICE_ID 0x9500 /* Initio's inic-950 product ID */ #define I940_DEVICE_ID 0x9400 /* Initio's inic-940 product ID */ #define I935_DEVICE_ID 0x9401 /* Initio's inic-935 product ID */ @@ -171,13 +169,16 @@ static int setup_debug = 0; static void i91uSCBPost(BYTE * pHcb, BYTE * pScb); -static const PCI_ID i91u_pci_devices[] = { - { INI_VENDOR_ID, I950_DEVICE_ID }, - { INI_VENDOR_ID, I940_DEVICE_ID }, - { INI_VENDOR_ID, I935_DEVICE_ID }, - { INI_VENDOR_ID, I920_DEVICE_ID }, - { DMX_VENDOR_ID, I920_DEVICE_ID }, +/* PCI Devices supported by this driver */ +static struct pci_device_id i91u_pci_devices[] __devinitdata = { + { PCI_VENDOR_ID_INIT, I950_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_INIT, I940_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_INIT, I935_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_INIT, I920_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_DOMEX, I920_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { } }; +MODULE_DEVICE_TABLE(pci, i91u_pci_devices); #define DEBUG_INTERRUPT 0 #define DEBUG_QUEUE 0 @@ -2771,7 +2772,7 @@ static int tul_NewReturnNumberOfAdapters(void) for (i = 0; i < ARRAY_SIZE(i91u_pci_devices); i++) { - while ((pDev = pci_find_device(i91u_pci_devices[i].vendor_id, i91u_pci_devices[i].device_id, pDev)) != NULL) { + while ((pDev = pci_find_device(i91u_pci_devices[i].vendor, i91u_pci_devices[i].device, pDev)) != NULL) { if (pci_enable_device(pDev)) continue; pci_read_config_dword(pDev, 0x44, (u32 *) & dRegValue); -- cgit v0.10.2 From d77f09e51f5793583ede9733a93bd31a324d051c Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:45:30 -0400 Subject: [ixj] Add pci dev table for module auto loading. Signed-off-by: Ben Collins diff --git a/drivers/telephony/ixj.c b/drivers/telephony/ixj.c index f6b2948..1b601b6 100644 --- a/drivers/telephony/ixj.c +++ b/drivers/telephony/ixj.c @@ -284,6 +284,14 @@ static int samplerate = 100; module_param(ixjdebug, int, 0); +static struct pci_device_id ixj_pci_tbl[] __devinitdata = { + { PCI_VENDOR_ID_QUICKNET, PCI_DEVICE_ID_QUICKNET_XJ, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { } +}; + +MODULE_DEVICE_TABLE(pci, ixj_pci_tbl); + /************************************************************************ * * ixjdebug meanings are now bit mapped instead of level based @@ -7683,7 +7691,8 @@ static int __init ixj_probe_pci(int *cnt) IXJ *j = NULL; for (i = 0; i < IXJMAX - *cnt; i++) { - pci = pci_find_device(0x15E2, 0x0500, pci); + pci = pci_find_device(PCI_VENDOR_ID_QUICKNET, + PCI_DEVICE_ID_QUICKNET_XJ, pci); if (!pci) break; -- cgit v0.10.2 From d57cdcffe1180cf9b8d1fce048f80d8c6b159827 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:47:37 -0400 Subject: [hid-core] TurboX Keyboard needs NOGET quirk. Signed-off-by: Ben Collins diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index feabda7..45f44fe 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1391,6 +1391,9 @@ void hid_close(struct hid_device *hid) #define USB_VENDOR_ID_PANJIT 0x134c +#define USB_VENDOR_ID_TURBOX 0x062a +#define USB_DEVICE_ID_TURBOX_KEYBOARD 0x0201 + /* * Initialize all reports */ @@ -1778,6 +1781,8 @@ static const struct hid_blacklist { { USB_VENDOR_ID_PANJIT, 0x0003, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_PANJIT, 0x0004, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_TURBOX, USB_DEVICE_ID_TURBOX_KEYBOARD, HID_QUIRK_NOGET }, + { 0, 0 } }; -- cgit v0.10.2 From 4938d3f4f8f1ffd744fa3626df8085118aeb1d79 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:49:31 -0400 Subject: [controlfb] Ifdef for when CONFIG_NVRAM isn't enabled. Signed-off-by: Ben Collins diff --git a/drivers/video/controlfb.c b/drivers/video/controlfb.c index 8cc6c0e..04c6d92 100644 --- a/drivers/video/controlfb.c +++ b/drivers/video/controlfb.c @@ -415,13 +415,15 @@ static int __init init_control(struct fb_info_control *p) full = p->total_vram == 0x400000; /* Try to pick a video mode out of NVRAM if we have one. */ +#ifdef CONFIG_NVRAM if (default_cmode == CMODE_NVRAM){ cmode = nvram_read_byte(NV_CMODE); if(cmode < CMODE_8 || cmode > CMODE_32) cmode = CMODE_8; } else +#endif cmode=default_cmode; - +#ifdef CONFIG_NVRAM if (default_vmode == VMODE_NVRAM) { vmode = nvram_read_byte(NV_VMODE); if (vmode < 1 || vmode > VMODE_MAX || @@ -432,7 +434,9 @@ static int __init init_control(struct fb_info_control *p) if (control_mac_modes[vmode - 1].m[full] < cmode) vmode = VMODE_640_480_60; } - } else { + } else +#endif + { vmode=default_vmode; if (control_mac_modes[vmode - 1].m[full] < cmode) { if (cmode > CMODE_8) -- cgit v0.10.2 From b023429ca30fc18b17a7b3e279b55bd652e9a989 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:50:49 -0400 Subject: [igafb] Add pci dev table for module auto loading. Signed-off-by: Ben Collins diff --git a/drivers/video/igafb.c b/drivers/video/igafb.c index 67f384f..e6df492c 100644 --- a/drivers/video/igafb.c +++ b/drivers/video/igafb.c @@ -573,3 +573,10 @@ int __init igafb_setup(char *options) module_init(igafb_init); MODULE_LICENSE("GPL"); +static struct pci_device_id igafb_pci_tbl[] __devinitdata = { + { PCI_VENDOR_ID_INTERG, PCI_DEVICE_ID_INTERG_1682, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { } +}; + +MODULE_DEVICE_TABLE(pci, igafb_pci_tbl); -- cgit v0.10.2 From f3f6f9aaced9c8453ef45956a23295e5c2328124 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:52:48 -0400 Subject: [platinumfb] Ifdef for when CONFIG_NVRAM isn't enabled. Signed-off-by: Ben Collins diff --git a/drivers/video/platinumfb.c b/drivers/video/platinumfb.c index 983be3e..fdb33cd 100644 --- a/drivers/video/platinumfb.c +++ b/drivers/video/platinumfb.c @@ -339,11 +339,12 @@ static int __devinit platinum_init_fb(struct fb_info *info) sense = read_platinum_sense(pinfo); printk(KERN_INFO "platinumfb: Monitor sense value = 0x%x, ", sense); - if (default_vmode == VMODE_NVRAM) { +#ifdef CONFIG_NVRAM default_vmode = nvram_read_byte(NV_VMODE); if (default_vmode <= 0 || default_vmode > VMODE_MAX || !platinum_reg_init[default_vmode-1]) +#endif default_vmode = VMODE_CHOOSE; } if (default_vmode == VMODE_CHOOSE) { @@ -351,8 +352,10 @@ static int __devinit platinum_init_fb(struct fb_info *info) } if (default_vmode <= 0 || default_vmode > VMODE_MAX) default_vmode = VMODE_640_480_60; +#ifdef CONFIG_NVRAM if (default_cmode == CMODE_NVRAM) default_cmode = nvram_read_byte(NV_CMODE); +#endif if (default_cmode < CMODE_8 || default_cmode > CMODE_32) default_cmode = CMODE_8; /* -- cgit v0.10.2 From 996f324767fd24b7cc682801277add2a4f276635 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:53:37 -0400 Subject: [valkyriefb] Ifdef for when CONFIG_NVRAM isn't enabled. Signed-off-by: Ben Collins diff --git a/drivers/video/valkyriefb.c b/drivers/video/valkyriefb.c index 47f2792..06fc19a 100644 --- a/drivers/video/valkyriefb.c +++ b/drivers/video/valkyriefb.c @@ -284,7 +284,7 @@ static void __init valkyrie_choose_mode(struct fb_info_valkyrie *p) printk(KERN_INFO "Monitor sense value = 0x%x\n", p->sense); /* Try to pick a video mode out of NVRAM if we have one. */ -#ifndef CONFIG_MAC +#if !defined(CONFIG_MAC) && defined(CONFIG_NVRAM) if (default_vmode == VMODE_NVRAM) { default_vmode = nvram_read_byte(NV_VMODE); if (default_vmode <= 0 @@ -297,7 +297,7 @@ static void __init valkyrie_choose_mode(struct fb_info_valkyrie *p) default_vmode = mac_map_monitor_sense(p->sense); if (!valkyrie_reg_init[default_vmode - 1]) default_vmode = VMODE_640_480_67; -#ifndef CONFIG_MAC +#if !defined(CONFIG_MAC) && defined(CONFIG_NVRAM) if (default_cmode == CMODE_NVRAM) default_cmode = nvram_read_byte(NV_CMODE); #endif -- cgit v0.10.2 From 74d919465a93b6c2b928b29a8ed3e5e41adbfa93 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:55:54 -0400 Subject: [pci_ids] Add Quicknet XJ vendor/device ID's. Signed-off-by: Ben Collins diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f069df2..f3a168f 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2351,3 +2351,5 @@ #define PCI_DEVICE_ID_RME_DIGI32_PRO 0x9897 #define PCI_DEVICE_ID_RME_DIGI32_8 0x9898 +#define PCI_VENDOR_ID_QUICKNET 0x15E2 +#define PCI_DEVICE_ID_QUICKNET_XJ 0x0500 -- cgit v0.10.2 From 1df23957ba2da6eb8c7a5c2ad7bb1d4732132873 Mon Sep 17 00:00:00 2001 From: Cedric Le Goater Date: Wed, 18 Oct 2006 18:30:41 +0200 Subject: [S390] fix vmlinux link when CONFIG_SYSIPC=n Fix the following compile error: CC init/version.o LD init/built-in.o LD .tmp_vmlinux1 arch/s390/kernel/built-in.o(.text+0xdba4): In function `sys32_ipc': : undefined reference to `compat_sys_semtimedop' arch/s390/kernel/built-in.o(.text+0xdbee): In function `sys32_ipc': : undefined reference to `compat_sys_semctl' arch/s390/kernel/built-in.o(.text+0xdc08): In function `sys32_ipc': : undefined reference to `compat_sys_msgsnd' arch/s390/kernel/built-in.o(.text+0xdc30): In function `sys32_ipc': : undefined reference to `compat_sys_msgrcv' arch/s390/kernel/built-in.o(.text+0xdc58): In function `sys32_ipc': : undefined reference to `compat_sys_msgctl' arch/s390/kernel/built-in.o(.text+0xdc76): In function `sys32_ipc': : undefined reference to `compat_sys_shmat' arch/s390/kernel/built-in.o(.text+0xdcb0): In function `sys32_ipc': : undefined reference to `compat_sys_shmctl' make: *** [.tmp_vmlinux1] Error 1 Signed-off-by: Cedric Le Goater Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index e15e148..2001767 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -295,6 +295,7 @@ static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i) * * This is really horribly ugly. */ +#ifdef CONFIG_SYSVIPC asmlinkage long sys32_ipc(u32 call, int first, int second, int third, u32 ptr) { if (call >> 16) /* hack for backward compatibility */ @@ -338,6 +339,7 @@ asmlinkage long sys32_ipc(u32 call, int first, int second, int third, u32 ptr) return -ENOSYS; } +#endif asmlinkage long sys32_truncate64(const char __user * path, unsigned long high, unsigned long low) { -- cgit v0.10.2 From 8b42f5c20251060cb1ab875459fe66ad7bb92cbd Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Wed, 18 Oct 2006 18:30:43 +0200 Subject: [S390] cio: invalid device operational notification Reset device operational notification flag when channel paths become unavailable during path verification. Signed-off-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index fcaf28d..de3d085 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -578,9 +578,13 @@ ccw_device_verify_done(struct ccw_device *cdev, int err) } break; case -ETIME: + /* Reset oper notify indication after verify error. */ + cdev->private->flags.donotify = 0; ccw_device_done(cdev, DEV_STATE_BOXED); break; default: + /* Reset oper notify indication after verify error. */ + cdev->private->flags.donotify = 0; PREPARE_WORK(&cdev->private->kick_work, ccw_device_nopath_notify, cdev); queue_work(ccw_device_notify_work, &cdev->private->kick_work); -- cgit v0.10.2 From f5956f84072804712cb7b663c5c64e9800180833 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 18 Oct 2006 18:30:45 +0200 Subject: [S390] Wire up epoll_pwait syscall. Signed-off-by: Heiko Carstens diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index e59baec..a4ceae3 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -320,3 +320,4 @@ SYSCALL(sys_tee,sys_tee,sys_tee_wrapper) SYSCALL(sys_vmsplice,sys_vmsplice,compat_sys_vmsplice_wrapper) NI_SYSCALL /* 310 sys_move_pages */ SYSCALL(sys_getcpu,sys_getcpu,sys_getcpu_wrapper) +SYSCALL(sys_epoll_pwait,sys_epoll_pwait,sys_ni_syscall) diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h index a19238c..71d3c21 100644 --- a/include/asm-s390/unistd.h +++ b/include/asm-s390/unistd.h @@ -249,8 +249,9 @@ #define __NR_vmsplice 309 /* Number 310 is reserved for new sys_move_pages */ #define __NR_getcpu 311 +#define __NR_epoll_pwait 312 -#define NR_syscalls 312 +#define NR_syscalls 313 /* * There are some system calls that are not present on 64 bit, some -- cgit v0.10.2 From 85a4aa64a85a8bd19b86887f70b13a27a363030d Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 18 Oct 2006 18:30:47 +0200 Subject: [S390] cio: sch_no -> schid.sch_no conversion. Overlooked one sch_no -> schid.sch_no conversion. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c index 0648ce5..476aa1d 100644 --- a/drivers/s390/cio/qdio.c +++ b/drivers/s390/cio/qdio.c @@ -3529,7 +3529,7 @@ do_QDIO(struct ccw_device *cdev,unsigned int callflags, #ifdef CONFIG_QDIO_DEBUG char dbf_text[20]; - sprintf(dbf_text,"doQD%04x",cdev->private->sch_no); + sprintf(dbf_text,"doQD%04x",cdev->private->schid.sch_no); QDIO_DBF_TEXT3(0,trace,dbf_text); #endif /* CONFIG_QDIO_DEBUG */ -- cgit v0.10.2 From 2c91971f84be168a35f937dd6c61e56e492b2185 Mon Sep 17 00:00:00 2001 From: Melissa Howland Date: Wed, 18 Oct 2006 18:30:49 +0200 Subject: [S390] monwriter find header logic. Fix logic for finding matching buffers. Signed-off-by: Melissa Howland Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/char/monwriter.c b/drivers/s390/char/monwriter.c index abd02ed..b9b0fc3 100644 --- a/drivers/s390/char/monwriter.c +++ b/drivers/s390/char/monwriter.c @@ -73,12 +73,15 @@ static inline struct mon_buf *monwrite_find_hdr(struct mon_private *monpriv, struct mon_buf *entry, *next; list_for_each_entry_safe(entry, next, &monpriv->list, list) - if (entry->hdr.applid == monhdr->applid && + if ((entry->hdr.mon_function == monhdr->mon_function || + monhdr->mon_function == MONWRITE_STOP_INTERVAL) && + entry->hdr.applid == monhdr->applid && entry->hdr.record_num == monhdr->record_num && entry->hdr.version == monhdr->version && entry->hdr.release == monhdr->release && entry->hdr.mod_level == monhdr->mod_level) return entry; + return NULL; } @@ -92,7 +95,9 @@ static int monwrite_new_hdr(struct mon_private *monpriv) monhdr->mon_function > MONWRITE_START_CONFIG || monhdr->hdrlen != sizeof(struct monwrite_hdr)) return -EINVAL; - monbuf = monwrite_find_hdr(monpriv, monhdr); + monbuf = NULL; + if (monhdr->mon_function != MONWRITE_GEN_EVENT) + monbuf = monwrite_find_hdr(monpriv, monhdr); if (monbuf) { if (monhdr->mon_function == MONWRITE_STOP_INTERVAL) { monhdr->datalen = monbuf->hdr.datalen; @@ -104,7 +109,7 @@ static int monwrite_new_hdr(struct mon_private *monpriv) kfree(monbuf); monbuf = NULL; } - } else { + } else if (monhdr->mon_function != MONWRITE_STOP_INTERVAL) { if (mon_buf_count >= mon_max_bufs) return -ENOSPC; monbuf = kzalloc(sizeof(struct mon_buf), GFP_KERNEL); @@ -118,7 +123,8 @@ static int monwrite_new_hdr(struct mon_private *monpriv) } monbuf->hdr = *monhdr; list_add_tail(&monbuf->list, &monpriv->list); - mon_buf_count++; + if (monhdr->mon_function != MONWRITE_GEN_EVENT) + mon_buf_count++; } monpriv->current_buf = monbuf; return 0; -- cgit v0.10.2 From 833774849d50a59f58e9bdfc3d9c88e682b3596d Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 18 Oct 2006 18:30:51 +0200 Subject: [S390] Fix pte type checking. handle_pte_fault uses pte_present, pte_none and pte_file to find out the type of a pte. That is done without holding the page table lock. This clashes with the way how ptep_clear_flush removes active page table entries from the system. First the ipte instruction is used to invalidate the pte and remove all plt entries for the page. The ipte sets the hardware invalid bit without changing any other bit. After the ipte finished the pte is cleared. A concurrent fault can observe the the previously valid pte with the invalid bit set. With the current encoding of the different pte types an invalidated read-only pte can be misinterpreted as a swap-pte. Signed-off-by: Martin Schwidefsky diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h index 519f0a5..36bb6da 100644 --- a/include/asm-s390/pgtable.h +++ b/include/asm-s390/pgtable.h @@ -200,18 +200,45 @@ extern char empty_zero_page[PAGE_SIZE]; */ /* Hardware bits in the page table entry */ -#define _PAGE_RO 0x200 /* HW read-only */ -#define _PAGE_INVALID 0x400 /* HW invalid */ +#define _PAGE_RO 0x200 /* HW read-only bit */ +#define _PAGE_INVALID 0x400 /* HW invalid bit */ +#define _PAGE_SWT 0x001 /* SW pte type bit t */ +#define _PAGE_SWX 0x002 /* SW pte type bit x */ -/* Mask and six different types of pages. */ -#define _PAGE_TYPE_MASK 0x601 +/* Six different types of pages. */ #define _PAGE_TYPE_EMPTY 0x400 #define _PAGE_TYPE_NONE 0x401 -#define _PAGE_TYPE_SWAP 0x600 -#define _PAGE_TYPE_FILE 0x601 +#define _PAGE_TYPE_SWAP 0x403 +#define _PAGE_TYPE_FILE 0x601 /* bit 0x002 is used for offset !! */ #define _PAGE_TYPE_RO 0x200 #define _PAGE_TYPE_RW 0x000 +/* + * PTE type bits are rather complicated. handle_pte_fault uses pte_present, + * pte_none and pte_file to find out the pte type WITHOUT holding the page + * table lock. ptep_clear_flush on the other hand uses ptep_clear_flush to + * invalidate a given pte. ipte sets the hw invalid bit and clears all tlbs + * for the page. The page table entry is set to _PAGE_TYPE_EMPTY afterwards. + * This change is done while holding the lock, but the intermediate step + * of a previously valid pte with the hw invalid bit set can be observed by + * handle_pte_fault. That makes it necessary that all valid pte types with + * the hw invalid bit set must be distinguishable from the four pte types + * empty, none, swap and file. + * + * irxt ipte irxt + * _PAGE_TYPE_EMPTY 1000 -> 1000 + * _PAGE_TYPE_NONE 1001 -> 1001 + * _PAGE_TYPE_SWAP 1011 -> 1011 + * _PAGE_TYPE_FILE 11?1 -> 11?1 + * _PAGE_TYPE_RO 0100 -> 1100 + * _PAGE_TYPE_RW 0000 -> 1000 + * + * pte_none is true for bits combinations 1000, 1100 + * pte_present is true for bits combinations 0000, 0010, 0100, 0110, 1001 + * pte_file is true for bits combinations 1101, 1111 + * swap pte is 1011 and 0001, 0011, 0101, 0111, 1010 and 1110 are invalid. + */ + #ifndef __s390x__ /* Bits in the segment table entry */ @@ -365,18 +392,21 @@ static inline int pmd_bad(pmd_t pmd) static inline int pte_none(pte_t pte) { - return (pte_val(pte) & _PAGE_TYPE_MASK) == _PAGE_TYPE_EMPTY; + return (pte_val(pte) & _PAGE_INVALID) && !(pte_val(pte) & _PAGE_SWT); } static inline int pte_present(pte_t pte) { - return !(pte_val(pte) & _PAGE_INVALID) || - (pte_val(pte) & _PAGE_TYPE_MASK) == _PAGE_TYPE_NONE; + unsigned long mask = _PAGE_RO | _PAGE_INVALID | _PAGE_SWT | _PAGE_SWX; + return (pte_val(pte) & mask) == _PAGE_TYPE_NONE || + (!(pte_val(pte) & _PAGE_INVALID) && + !(pte_val(pte) & _PAGE_SWT)); } static inline int pte_file(pte_t pte) { - return (pte_val(pte) & _PAGE_TYPE_MASK) == _PAGE_TYPE_FILE; + unsigned long mask = _PAGE_RO | _PAGE_INVALID | _PAGE_SWT; + return (pte_val(pte) & mask) == _PAGE_TYPE_FILE; } #define pte_same(a,b) (pte_val(a) == pte_val(b)) -- cgit v0.10.2 From 867dcd0f32b125f45a663f9374a36fb21193ab21 Mon Sep 17 00:00:00 2001 From: Stefan Weinhuber Date: Wed, 18 Oct 2006 18:30:53 +0200 Subject: [S390] dasd: clean up timer. Clean up dasd timer when when a dasd device is set offline. Signed-off-by: Stefan Weinhuber Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index d0647d1..79ffef6 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -203,6 +203,7 @@ dasd_state_basic_to_known(struct dasd_device * device) rc = dasd_flush_ccw_queue(device, 1); if (rc) return rc; + dasd_clear_timer(device); DBF_DEV_EVENT(DBF_EMERG, device, "%p debug area deleted", device); if (device->debug_area != NULL) { -- cgit v0.10.2 From 9b10fe5b70878fb4b7e3e1b300f1bff0c73d4e2e Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 18 Oct 2006 18:30:55 +0200 Subject: [S390] cio: update documentation. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/Documentation/s390/CommonIO b/Documentation/s390/CommonIO index 59d1166..d684a6a 100644 --- a/Documentation/s390/CommonIO +++ b/Documentation/s390/CommonIO @@ -66,7 +66,7 @@ Command line parameters When a device is un-ignored, device recognition and sensing is performed and the device driver will be notified if possible, so the device will become - available to the system. + available to the system. Note that un-ignoring is performed asynchronously. You can also add ranges of devices to be ignored by piping to /proc/cio_ignore; "add , , ..." will ignore the diff --git a/Documentation/s390/cds.txt b/Documentation/s390/cds.txt index d80e573..32a96cc 100644 --- a/Documentation/s390/cds.txt +++ b/Documentation/s390/cds.txt @@ -174,14 +174,10 @@ read_dev_chars() - Read Device Characteristics This routine returns the characteristics for the device specified. -The function is meant to be called with an irq handler in place; that is, +The function is meant to be called with the device already enabled; that is, at earliest during set_online() processing. -While the request is processed synchronously, the device interrupt -handler is called for final ending status. In case of error situations the -interrupt handler may recover appropriately. The device irq handler can -recognize the corresponding interrupts by the interruption parameter be -0x00524443. The ccw_device must not be locked prior to calling read_dev_chars(). +The ccw_device must not be locked prior to calling read_dev_chars(). The function may be called enabled or disabled. @@ -410,26 +406,7 @@ individual flag meanings. Usage Notes : -Prior to call ccw_device_start() the device driver must assure disabled state, -i.e. the I/O mask value in the PSW must be disabled. This can be accomplished -by calling local_save_flags( flags). The current PSW flags are preserved and -can be restored by local_irq_restore( flags) at a later time. - -If the device driver violates this rule while running in a uni-processor -environment an interrupt might be presented prior to the ccw_device_start() -routine returning to the device driver main path. In this case we will end in a -deadlock situation as the interrupt handler will try to obtain the irq -lock the device driver still owns (see below) ! - -The driver must assure to hold the device specific lock. This can be -accomplished by - -(i) spin_lock(get_ccwdev_lock(cdev)), or -(ii) spin_lock_irqsave(get_ccwdev_lock(cdev), flags) - -Option (i) should be used if the calling routine is running disabled for -I/O interrupts (see above) already. Option (ii) obtains the device gate und -puts the CPU into I/O disabled state by preserving the current PSW flags. +ccw_device_start() must be called disabled and with the ccw device lock held. The device driver is allowed to issue the next ccw_device_start() call from within its interrupt handler already. It is not required to schedule a @@ -488,7 +465,7 @@ int ccw_device_resume(struct ccw_device *cdev); cdev - ccw_device the resume operation is requested for -The resume_IO() function returns: +The ccw_device_resume() function returns: 0 - suspended channel program is resumed -EBUSY - status pending @@ -507,6 +484,8 @@ a long-running channel program or the device might require to initially issue a halt subchannel (HSCH) I/O command. For those purposes the ccw_device_halt() command is provided. +ccw_device_halt() must be called disabled and with the ccw device lock held. + int ccw_device_halt(struct ccw_device *cdev, unsigned long intparm); @@ -517,7 +496,7 @@ intparm : interruption parameter; value is only used if no I/O The ccw_device_halt() function returns : - 0 - successful completion or request successfully initiated + 0 - request successfully initiated -EBUSY - the device is currently busy, or status pending. -ENODEV - cdev invalid. -EINVAL - The device is not operational or the ccw device is not online. @@ -533,6 +512,23 @@ can then perform an appropriate action. Prior to interrupt of an outstanding read to a network device (with or without PCI flag) a ccw_device_halt() is required to end the pending operation. +ccw_device_clear() - Terminage I/O Request Processing + +In order to terminate all I/O processing at the subchannel, the clear subchannel +(CSCH) command is used. It can be issued via ccw_device_clear(). + +ccw_device_clear() must be called disabled and with the ccw device lock held. + +int ccw_device_clear(struct ccw_device *cdev, unsigned long intparm); + +cdev: ccw_device the clear operation is requested for +intparm: interruption parameter (see ccw_device_halt()) + +The ccw_device_clear() function returns: + + 0 - request successfully initiated +-ENODEV - cdev invalid +-EINVAL - The device is not operational or the ccw device is not online. Miscellaneous Support Routines diff --git a/Documentation/s390/driver-model.txt b/Documentation/s390/driver-model.txt index 62c0823..77bf450 100644 --- a/Documentation/s390/driver-model.txt +++ b/Documentation/s390/driver-model.txt @@ -239,6 +239,9 @@ status - Can be 'online' or 'offline'. type - The physical type of the channel path. +shared - Whether the channel path is shared. + +cmg - The channel measurement group. 3. System devices ----------------- -- cgit v0.10.2 From ce9b18f5cdd628196ca13bb783e8dcd5ad29ceaa Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 18 Oct 2006 18:30:59 +0200 Subject: [S390] update default configuration Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/defconfig b/arch/s390/defconfig index a325739..c313e9a 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18 -# Wed Oct 4 19:45:46 2006 +# Linux kernel version: 2.6.19-rc2 +# Wed Oct 18 17:11:10 2006 # CONFIG_MMU=y CONFIG_LOCKDEP_SUPPORT=y @@ -211,6 +211,7 @@ CONFIG_INET6_XFRM_MODE_TRANSPORT=y CONFIG_INET6_XFRM_MODE_TUNNEL=y CONFIG_INET6_XFRM_MODE_BEET=y # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=y # CONFIG_IPV6_TUNNEL is not set # CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set @@ -528,6 +529,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -646,10 +648,6 @@ CONFIG_MSDOS_PARTITION=y # CONFIG_NLS is not set # -# Distributed Lock Manager -# - -# # Instrumentation Support # @@ -669,7 +667,6 @@ CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=17 -# CONFIG_DETECT_SOFTLOCKUP is not set # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set CONFIG_DEBUG_PREEMPT=y @@ -690,6 +687,7 @@ CONFIG_DEBUG_FS=y # CONFIG_FRAME_POINTER is not set # CONFIG_UNWIND_INFO is not set CONFIG_FORCED_INLINING=y +CONFIG_HEADERS_CHECK=y # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_LKDTM is not set -- cgit v0.10.2 From b4a1efffcf8070dbc7734f27da10ce49fb9f2a34 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Fri, 22 Sep 2006 12:52:37 -0700 Subject: shpchp: fix shpchp_wait_cmd in poll This patch fixes the problem that issuing SHPC command in poll mode always fails with the following message. shpchp: Command not completed in 2000 msec Signed-off-by: Kenji Kaneshige Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c index 4d8aee1..446e9be 100644 --- a/drivers/pci/hotplug/shpchp_hpc.c +++ b/drivers/pci/hotplug/shpchp_hpc.c @@ -302,16 +302,43 @@ static void start_int_poll_timer(struct php_ctlr_state_s *php_ctlr, int sec) add_timer(&php_ctlr->int_poll_timer); } +/* + * Returns 1 if SHPC finishes executing a command within 1 sec, + * otherwise returns 0. + */ +static inline int shpc_poll_ctrl_busy(struct controller *ctrl) +{ + int i; + u16 cmd_status = shpc_readw(ctrl, CMD_STATUS); + + if (!(cmd_status & 0x1)) + return 1; + + /* Check every 0.1 sec for a total of 1 sec */ + for (i = 0; i < 10; i++) { + msleep(100); + cmd_status = shpc_readw(ctrl, CMD_STATUS); + if (!(cmd_status & 0x1)) + return 1; + } + + return 0; +} + static inline int shpc_wait_cmd(struct controller *ctrl) { int retval = 0; - unsigned int timeout_msec = shpchp_poll_mode ? 2000 : 1000; - unsigned long timeout = msecs_to_jiffies(timeout_msec); - int rc = wait_event_interruptible_timeout(ctrl->queue, - !ctrl->cmd_busy, timeout); + unsigned long timeout = msecs_to_jiffies(1000); + int rc; + + if (shpchp_poll_mode) + rc = shpc_poll_ctrl_busy(ctrl); + else + rc = wait_event_interruptible_timeout(ctrl->queue, + !ctrl->cmd_busy, timeout); if (!rc) { retval = -EIO; - err("Command not completed in %d msec\n", timeout_msec); + err("Command not completed in 1000 msec\n"); } else if (rc < 0) { retval = -EINTR; info("Command was interrupted by a signal\n"); @@ -327,26 +354,15 @@ static int shpc_write_cmd(struct slot *slot, u8 t_slot, u8 cmd) u16 cmd_status; int retval = 0; u16 temp_word; - int i; DBG_ENTER_ROUTINE mutex_lock(&slot->ctrl->cmd_lock); - for (i = 0; i < 10; i++) { - cmd_status = shpc_readw(ctrl, CMD_STATUS); - - if (!(cmd_status & 0x1)) - break; - /* Check every 0.1 sec for a total of 1 sec*/ - msleep(100); - } - - cmd_status = shpc_readw(ctrl, CMD_STATUS); - - if (cmd_status & 0x1) { + if (!shpc_poll_ctrl_busy(ctrl)) { /* After 1 sec and and the controller is still busy */ - err("%s : Controller is still busy after 1 sec.\n", __FUNCTION__); + err("%s : Controller is still busy after 1 sec.\n", + __FUNCTION__); retval = -EBUSY; goto out; } -- cgit v0.10.2 From 49ed2b4963cd00993eab518b820a6700f94f222d Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Fri, 22 Sep 2006 10:17:10 -0700 Subject: pciehp: fix improper info messages The slot number displayed in info messages would cause a confusion because those are displayed in several ways (decimal and hex). Furthermore, those slot number is not same as slot name (directory name). This patch fixes those improper info messages. Signed-off-by: Kenji Kaneshige Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index eaea9d3..b71f774 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -166,10 +166,10 @@ struct controller { * error Messages */ #define msg_initialization_err "Initialization failure, error=%d\n" -#define msg_button_on "PCI slot #%d - powering on due to button press.\n" -#define msg_button_off "PCI slot #%d - powering off due to button press.\n" -#define msg_button_cancel "PCI slot #%d - action canceled due to button press.\n" -#define msg_button_ignore "PCI slot #%d - button press ignored. (action in progress...)\n" +#define msg_button_on "PCI slot #%s - powering on due to button press.\n" +#define msg_button_off "PCI slot #%s - powering off due to button press.\n" +#define msg_button_cancel "PCI slot #%s - action canceled due to button press.\n" +#define msg_button_ignore "PCI slot #%s - button press ignored. (action in progress...)\n" /* controller functions */ extern int pciehp_event_start_thread (void); diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index 41290a1..f602b04 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -43,6 +43,11 @@ static int event_finished; static unsigned long pushbutton_pending; /* = 0 */ static unsigned long surprise_rm_pending; /* = 0 */ +static inline char *slot_name(struct slot *p_slot) +{ + return p_slot->hotplug_slot->name; +} + u8 pciehp_handle_attention_button(u8 hp_slot, void *inst_id) { struct controller *ctrl = (struct controller *) inst_id; @@ -68,7 +73,7 @@ u8 pciehp_handle_attention_button(u8 hp_slot, void *inst_id) /* * Button pressed - See if need to TAKE ACTION!!! */ - info("Button pressed on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Button pressed on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_BUTTON_PRESS; if ((p_slot->state == BLINKINGON_STATE) @@ -78,7 +83,7 @@ u8 pciehp_handle_attention_button(u8 hp_slot, void *inst_id) * or hot-remove */ taskInfo->event_type = INT_BUTTON_CANCEL; - info("Button cancel on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Button cancel on Slot(%s)\n", slot_name(p_slot)); } else if ((p_slot->state == POWERON_STATE) || (p_slot->state == POWEROFF_STATE)) { /* Ignore if the slot is on power-on or power-off state; this @@ -86,7 +91,7 @@ u8 pciehp_handle_attention_button(u8 hp_slot, void *inst_id) * hot-remove is undergoing */ taskInfo->event_type = INT_BUTTON_IGNORE; - info("Button ignore on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Button ignore on Slot(%s)\n", slot_name(p_slot)); } if (rc) @@ -122,13 +127,13 @@ u8 pciehp_handle_switch_change(u8 hp_slot, void *inst_id) /* * Switch opened */ - info("Latch open on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Latch open on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_SWITCH_OPEN; } else { /* * Switch closed */ - info("Latch close on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Latch close on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_SWITCH_CLOSE; } @@ -166,13 +171,13 @@ u8 pciehp_handle_presence_change(u8 hp_slot, void *inst_id) /* * Card Present */ - info("Card present on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Card present on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_PRESENCE_ON; } else { /* * Not Present */ - info("Card not present on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Card not present on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_PRESENCE_OFF; } @@ -206,13 +211,13 @@ u8 pciehp_handle_power_fault(u8 hp_slot, void *inst_id) /* * power fault Cleared */ - info("Power fault cleared on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Power fault cleared on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_POWER_FAULT_CLEAR; } else { /* * power fault */ - info("Power fault on Slot(%d)\n", ctrl->first_slot + hp_slot); + info("Power fault on Slot(%s)\n", slot_name(p_slot)); taskInfo->event_type = INT_POWER_FAULT; info("power fault bit %x set\n", hp_slot); } @@ -654,7 +659,7 @@ static void interrupt_event_handler(struct controller *ctrl) warn("Not a valid state\n"); return; } - info(msg_button_cancel, p_slot->number); + info(msg_button_cancel, slot_name(p_slot)); p_slot->state = STATIC_STATE; } /* ***********Button Pressed (No action on 1st press...) */ @@ -667,12 +672,12 @@ static void interrupt_event_handler(struct controller *ctrl) /* slot is on */ dbg("slot is on\n"); p_slot->state = BLINKINGOFF_STATE; - info(msg_button_off, p_slot->number); + info(msg_button_off, slot_name(p_slot)); } else { /* slot is off */ dbg("slot is off\n"); p_slot->state = BLINKINGON_STATE; - info(msg_button_on, p_slot->number); + info(msg_button_on, slot_name(p_slot)); } /* Wait for exclusive access to hardware */ @@ -760,14 +765,16 @@ int pciehp_enable_slot(struct slot *p_slot) rc = p_slot->hpc_ops->get_adapter_status(p_slot, &getstatus); if (rc || !getstatus) { - info("%s: no adapter on slot(%x)\n", __FUNCTION__, p_slot->number); + info("%s: no adapter on slot(%s)\n", __FUNCTION__, + slot_name(p_slot)); mutex_unlock(&p_slot->ctrl->crit_sect); return -ENODEV; } if (MRL_SENS(p_slot->ctrl->ctrlcap)) { rc = p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); if (rc || getstatus) { - info("%s: latch open on slot(%x)\n", __FUNCTION__, p_slot->number); + info("%s: latch open on slot(%s)\n", __FUNCTION__, + slot_name(p_slot)); mutex_unlock(&p_slot->ctrl->crit_sect); return -ENODEV; } @@ -776,7 +783,8 @@ int pciehp_enable_slot(struct slot *p_slot) if (POWER_CTRL(p_slot->ctrl->ctrlcap)) { rc = p_slot->hpc_ops->get_power_status(p_slot, &getstatus); if (rc || getstatus) { - info("%s: already enabled on slot(%x)\n", __FUNCTION__, p_slot->number); + info("%s: already enabled on slot(%s)\n", __FUNCTION__, + slot_name(p_slot)); mutex_unlock(&p_slot->ctrl->crit_sect); return -EINVAL; } @@ -811,7 +819,8 @@ int pciehp_disable_slot(struct slot *p_slot) if (!HP_SUPR_RM(p_slot->ctrl->ctrlcap)) { ret = p_slot->hpc_ops->get_adapter_status(p_slot, &getstatus); if (ret || !getstatus) { - info("%s: no adapter on slot(%x)\n", __FUNCTION__, p_slot->number); + info("%s: no adapter on slot(%s)\n", __FUNCTION__, + slot_name(p_slot)); mutex_unlock(&p_slot->ctrl->crit_sect); return -ENODEV; } @@ -820,7 +829,8 @@ int pciehp_disable_slot(struct slot *p_slot) if (MRL_SENS(p_slot->ctrl->ctrlcap)) { ret = p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); if (ret || getstatus) { - info("%s: latch open on slot(%x)\n", __FUNCTION__, p_slot->number); + info("%s: latch open on slot(%s)\n", __FUNCTION__, + slot_name(p_slot)); mutex_unlock(&p_slot->ctrl->crit_sect); return -ENODEV; } @@ -829,7 +839,8 @@ int pciehp_disable_slot(struct slot *p_slot) if (POWER_CTRL(p_slot->ctrl->ctrlcap)) { ret = p_slot->hpc_ops->get_power_status(p_slot, &getstatus); if (ret || !getstatus) { - info("%s: already disabled slot(%x)\n", __FUNCTION__, p_slot->number); + info("%s: already disabled slot(%s)\n", __FUNCTION__, + slot_name(p_slot)); mutex_unlock(&p_slot->ctrl->crit_sect); return -EINVAL; } -- cgit v0.10.2 From dd5619cb4407e830a8921a93c949be37c81105b5 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Fri, 22 Sep 2006 10:17:29 -0700 Subject: pciehp - add missing locking This patch fixes the problem that system will panic if multiple power on/off operations are issued to the same slot in parallel. This problem can be easily reproduced by commands below. # while true; do echo 1 > power; echo 0 > power; done & # while true; do echo 1 > power; echo 0 > power; done & The cause is lack of locking for enable/disable operations. This patch fixes this problem. Signed-off-by: Kenji Kaneshige Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index b71f774..30f021c 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -92,6 +92,7 @@ struct php_ctlr_state_s { struct controller { struct controller *next; struct mutex crit_sect; /* critical section mutex */ + struct mutex ctrl_lock; /* controller lock */ struct php_ctlr_state_s *hpc_ctlr_handle; /* HPC controller handle */ int num_slots; /* Number of slots on ctlr */ int slot_num_inc; /* 1 or -1 */ diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index c67b7c3..f93e81e 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -448,7 +448,7 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ } /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); t_slot->hpc_ops->get_adapter_status(t_slot, &value); /* Check if slot is occupied */ @@ -456,7 +456,7 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ rc = t_slot->hpc_ops->power_off_slot(t_slot); /* Power off slot if not occupied*/ if (rc) { /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); goto err_out_free_ctrl_slot; } else /* Wait for the command to complete */ @@ -464,7 +464,7 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); return 0; diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index f602b04..c206a3d 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -234,13 +234,13 @@ u8 pciehp_handle_power_fault(u8 hp_slot, void *inst_id) static void set_slot_off(struct controller *ctrl, struct slot * pslot) { /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); /* turn off slot, turn on Amber LED, turn off Green LED if supported*/ if (POWER_CTRL(ctrl->ctrlcap)) { if (pslot->hpc_ops->power_off_slot(pslot)) { err("%s: Issue of Slot Power Off command failed\n", __FUNCTION__); - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); return; } wait_for_ctrl_irq (ctrl); @@ -254,14 +254,14 @@ static void set_slot_off(struct controller *ctrl, struct slot * pslot) if (ATTN_LED(ctrl->ctrlcap)) { if (pslot->hpc_ops->set_attention_status(pslot, 1)) { err("%s: Issue of Set Attention Led command failed\n", __FUNCTION__); - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); return; } wait_for_ctrl_irq (ctrl); } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); } /** @@ -284,13 +284,13 @@ static int board_added(struct slot *p_slot) ctrl->slot_device_offset, hp_slot); /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); if (POWER_CTRL(ctrl->ctrlcap)) { /* Power on slot */ rc = p_slot->hpc_ops->power_on_slot(p_slot); if (rc) { - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); return -1; } @@ -306,7 +306,7 @@ static int board_added(struct slot *p_slot) } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); /* Wait for ~1 second */ wait_for_ctrl_irq (ctrl); @@ -340,7 +340,7 @@ static int board_added(struct slot *p_slot) pci_fixup_device(pci_fixup_final, ctrl->pci_dev); if (PWR_LED(ctrl->ctrlcap)) { /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); p_slot->hpc_ops->green_led_on(p_slot); @@ -348,7 +348,7 @@ static int board_added(struct slot *p_slot) wait_for_ctrl_irq (ctrl); /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); } return 0; @@ -380,14 +380,14 @@ static int remove_board(struct slot *p_slot) dbg("In %s, hp_slot = %d\n", __FUNCTION__, hp_slot); /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); if (POWER_CTRL(ctrl->ctrlcap)) { /* power off slot */ rc = p_slot->hpc_ops->power_off_slot(p_slot); if (rc) { err("%s: Issue of Slot Disable command failed\n", __FUNCTION__); - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); return rc; } /* Wait for the command to complete */ @@ -403,7 +403,7 @@ static int remove_board(struct slot *p_slot) } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); return 0; } @@ -450,7 +450,7 @@ static void pciehp_pushbutton_thread(unsigned long slot) if (pciehp_enable_slot(p_slot) && PWR_LED(p_slot->ctrl->ctrlcap)) { /* Wait for exclusive access to hardware */ - mutex_lock(&p_slot->ctrl->crit_sect); + mutex_lock(&p_slot->ctrl->ctrl_lock); p_slot->hpc_ops->green_led_off(p_slot); @@ -458,7 +458,7 @@ static void pciehp_pushbutton_thread(unsigned long slot) wait_for_ctrl_irq (p_slot->ctrl); /* Done with exclusive hardware access */ - mutex_unlock(&p_slot->ctrl->crit_sect); + mutex_unlock(&p_slot->ctrl->ctrl_lock); } p_slot->state = STATIC_STATE; } @@ -500,7 +500,7 @@ static void pciehp_surprise_rm_thread(unsigned long slot) if (pciehp_enable_slot(p_slot) && PWR_LED(p_slot->ctrl->ctrlcap)) { /* Wait for exclusive access to hardware */ - mutex_lock(&p_slot->ctrl->crit_sect); + mutex_lock(&p_slot->ctrl->ctrl_lock); p_slot->hpc_ops->green_led_off(p_slot); @@ -508,7 +508,7 @@ static void pciehp_surprise_rm_thread(unsigned long slot) wait_for_ctrl_irq (p_slot->ctrl); /* Done with exclusive hardware access */ - mutex_unlock(&p_slot->ctrl->crit_sect); + mutex_unlock(&p_slot->ctrl->ctrl_lock); } p_slot->state = STATIC_STATE; } @@ -621,7 +621,7 @@ static void interrupt_event_handler(struct controller *ctrl) switch (p_slot->state) { case BLINKINGOFF_STATE: /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); if (PWR_LED(ctrl->ctrlcap)) { p_slot->hpc_ops->green_led_on(p_slot); @@ -635,11 +635,11 @@ static void interrupt_event_handler(struct controller *ctrl) wait_for_ctrl_irq (ctrl); } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); break; case BLINKINGON_STATE: /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); if (PWR_LED(ctrl->ctrlcap)) { p_slot->hpc_ops->green_led_off(p_slot); @@ -652,7 +652,7 @@ static void interrupt_event_handler(struct controller *ctrl) wait_for_ctrl_irq (ctrl); } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); break; default: @@ -681,7 +681,7 @@ static void interrupt_event_handler(struct controller *ctrl) } /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); /* blink green LED and turn off amber */ if (PWR_LED(ctrl->ctrlcap)) { @@ -698,7 +698,7 @@ static void interrupt_event_handler(struct controller *ctrl) } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); init_timer(&p_slot->task_event); p_slot->task_event.expires = jiffies + 5 * HZ; /* 5 second delay */ @@ -713,7 +713,7 @@ static void interrupt_event_handler(struct controller *ctrl) if (POWER_CTRL(ctrl->ctrlcap)) { dbg("power fault\n"); /* Wait for exclusive access to hardware */ - mutex_lock(&ctrl->crit_sect); + mutex_lock(&ctrl->ctrl_lock); if (ATTN_LED(ctrl->ctrlcap)) { p_slot->hpc_ops->set_attention_status(p_slot, 1); @@ -726,7 +726,7 @@ static void interrupt_event_handler(struct controller *ctrl) } /* Done with exclusive hardware access */ - mutex_unlock(&ctrl->crit_sect); + mutex_unlock(&ctrl->ctrl_lock); } } /***********SURPRISE REMOVAL********************/ @@ -789,7 +789,6 @@ int pciehp_enable_slot(struct slot *p_slot) return -EINVAL; } } - mutex_unlock(&p_slot->ctrl->crit_sect); p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); @@ -801,6 +800,7 @@ int pciehp_enable_slot(struct slot *p_slot) if (p_slot) update_slot_info(p_slot); + mutex_unlock(&p_slot->ctrl->crit_sect); return rc; } @@ -846,10 +846,10 @@ int pciehp_disable_slot(struct slot *p_slot) } } - mutex_unlock(&p_slot->ctrl->crit_sect); - ret = remove_board(p_slot); update_slot_info(p_slot); + + mutex_unlock(&p_slot->ctrl->crit_sect); return ret; } diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 703a64a..1c551c6 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -1402,6 +1402,8 @@ int pcie_init(struct controller * ctrl, struct pcie_device *dev) pdev->subsystem_vendor, pdev->subsystem_device); mutex_init(&ctrl->crit_sect); + mutex_init(&ctrl->ctrl_lock); + /* setup wait queue */ init_waitqueue_head(&ctrl->queue); -- cgit v0.10.2 From 9ef9977cabc1b2c1718ef6eb883caec8dcb80b4c Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Mon, 25 Sep 2006 00:56:53 +0200 Subject: pciehp: Remove unnecessary check in pciehp_ctrl.c this was spotted by coverity (cid #819). We dereference p_slot earlier in the function, and i found no way it could become NULL anywhere. Signed-off-by: Eric Sesterhenn Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index c206a3d..372c63e 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -797,8 +797,7 @@ int pciehp_enable_slot(struct slot *p_slot) p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); } - if (p_slot) - update_slot_info(p_slot); + update_slot_info(p_slot); mutex_unlock(&p_slot->ctrl->crit_sect); return rc; -- cgit v0.10.2 From 09d6029f43ebbe7307854abdae204c25d711ff94 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Mon, 25 Sep 2006 16:52:19 -0700 Subject: PCI: VIA IRQ quirk behaviour change The most recent VIA IRQ quirk changes have broken various VIA devices for some users. We are not able to add these devices to the blacklist as they are also available in PCI-card form, and running the quirk on these devices brings us back to square one (running the VIA quirk on non-VIA boards where the quirk is not needed). This patch, based on suggestions from Sergey Vlasov, implements a scheme similar to but more restrictive than the scheme we had in 2.6.16 and earlier. It runs the quirk on all VIA hardware, but *only* if a VIA southbridge was detected on the system. To further reduce the amount of quirked devices, this patch includes a change suggested by Linus at http://lkml.org/lkml/2005/9/27/113 This ensures that devices bound to non-legacy IO-APIC interrupt lines are not quirked. We have made one change to Linus' suggestion: we do a comparison of ">15" rather than ">=15", as 15 is still in the legacy interrupt range. There is still a downside to this patch: if the user inserts a VIA PCI card into a VIA-based motherboard, in some circumstances the quirk will also run on the VIA PCI card. This corner case is hard to avoid. Signed-off-by: Daniel Drake Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 23b599d..e542507 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -648,11 +648,43 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4, quirk_vi * Some of the on-chip devices are actually '586 devices' so they are * listed here. */ + +static int via_irq_fixup_needed = -1; + +/* + * As some VIA hardware is available in PCI-card form, we need to restrict + * this quirk to VIA PCI hardware built onto VIA-based motherboards only. + * We try to locate a VIA southbridge before deciding whether the quirk + * should be applied. + */ +static const struct pci_device_id via_irq_fixup_tbl[] = { + { + .vendor = PCI_VENDOR_ID_VIA, + .device = PCI_ANY_ID, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .class = PCI_CLASS_BRIDGE_ISA << 8, + .class_mask = 0xffff00, + }, + { 0, }, +}; + static void quirk_via_irq(struct pci_dev *dev) { u8 irq, new_irq; - new_irq = dev->irq & 0xf; + if (via_irq_fixup_needed == -1) + via_irq_fixup_needed = pci_dev_present(via_irq_fixup_tbl); + + if (!via_irq_fixup_needed) + return; + + new_irq = dev->irq; + + /* Don't quirk interrupts outside the legacy IRQ range */ + if (!new_irq || new_irq > 15) + return; + pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq); if (new_irq != irq) { printk(KERN_INFO "PCI: VIA IRQ fixup for %s, from %d to %d\n", @@ -661,14 +693,7 @@ static void quirk_via_irq(struct pci_dev *dev) pci_write_config_byte(dev, PCI_INTERRUPT_LINE, new_irq); } } -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_0, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_1, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_2, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_3, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235_USB_2, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4, quirk_via_irq); -DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_5, quirk_via_irq); +DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_VIA, PCI_ANY_ID, quirk_via_irq); /* * VIA VT82C598 has its device ID settable and many BIOSes -- cgit v0.10.2 From 3ec6a8d02efd54a66640bd85afa8c162647b56c3 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 25 Sep 2006 16:52:20 -0700 Subject: PCI: pcie-check-and-return-bus_register-errors fix __must_check goes on the declaration, not the definition. Cc: "Randy.Dunlap" Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h index 67fcd17..3656e03 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -9,6 +9,8 @@ #ifndef _PORTDRV_H_ #define _PORTDRV_H_ +#include + #if !defined(PCI_CAP_ID_PME) #define PCI_CAP_ID_PME 1 #endif @@ -39,7 +41,7 @@ extern int pcie_port_device_suspend(struct pci_dev *dev, pm_message_t state); extern int pcie_port_device_resume(struct pci_dev *dev); #endif extern void pcie_port_device_remove(struct pci_dev *dev); -extern int pcie_port_bus_register(void); +extern int __must_check pcie_port_bus_register(void); extern void pcie_port_bus_unregister(void); #endif /* _PORTDRV_H_ */ diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index bd6615b..b20a9b8 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -6,7 +6,6 @@ * Copyright (C) Tom Long Nguyen (tom.l.nguyen@intel.com) */ -#include #include #include #include @@ -401,7 +400,7 @@ void pcie_port_device_remove(struct pci_dev *dev) pci_disable_msi(dev); } -int __must_check pcie_port_bus_register(void) +int pcie_port_bus_register(void) { return bus_register(&pcie_port_bus_type); } -- cgit v0.10.2 From bacedce32b171cd461a7da3160ad794e2240c67a Mon Sep 17 00:00:00 2001 From: Daniel Ritz Date: Mon, 25 Sep 2006 16:52:21 -0700 Subject: PCI: add ICH7/8 ACPI/GPIO io resource quirks Signed-off-by: Daniel Ritz Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index e542507..1d2ccda 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -453,6 +453,12 @@ static void __devinit quirk_ich6_lpc_acpi(struct pci_dev *dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_0, quirk_ich6_lpc_acpi ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1, quirk_ich6_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_0, quirk_ich6_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_1, quirk_ich6_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31, quirk_ich6_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_0, quirk_ich6_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_2, quirk_ich6_lpc_acpi ); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_3, quirk_ich6_lpc_acpi ); /* * VIA ACPI: One IO region pointed to by longword at -- cgit v0.10.2 From b5e4efe7e061ff52ac97b9fa45acca529d8daeea Mon Sep 17 00:00:00 2001 From: "eiichiro.oiwa.nm@hitachi.com" Date: Thu, 28 Sep 2006 13:55:47 +0900 Subject: PCI: Turn pci_fixup_video into generic for embedded VGA pci_fixup_video turns into generic code because there are many platforms need this fixup for embedded VGA as well as x86. The Video BIOS integrates into System BIOS on a machine has embedded VGA although embedded VGA generally don't have PCI ROM. As a result, embedded VGA need the way that the sysfs rom points to the Video BIOS of System RAM (0xC0000). PCI-to-PCI Bridge Architecture specification describes the condition whether or not PCI ROM forwards VGA compatible memory address. fixup_video suits this specification. Although the Video ROM generally implements in x86 code regardless of platform, some application such as X Window System can run this code by dosemu86. Therefore, pci_fixup_video should turn into generic code. Signed-off-by: Eiichiro Oiwa Acked-by: Alan Cox Acked-by: Jesse Barnes Signed-off-by: Greg Kroah-Hartman diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c index b60d7e8..908b410 100644 --- a/arch/i386/pci/fixup.c +++ b/arch/i386/pci/fixup.c @@ -343,51 +343,6 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC, pcie_ro DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC1, pcie_rootport_aspm_quirk ); /* - * Fixup to mark boot BIOS video selected by BIOS before it changes - * - * From information provided by "Jon Smirl" - * - * The standard boot ROM sequence for an x86 machine uses the BIOS - * to select an initial video card for boot display. This boot video - * card will have it's BIOS copied to C0000 in system RAM. - * IORESOURCE_ROM_SHADOW is used to associate the boot video - * card with this copy. On laptops this copy has to be used since - * the main ROM may be compressed or combined with another image. - * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW - * is marked here since the boot video device will be the only enabled - * video device at this point. - */ - -static void __devinit pci_fixup_video(struct pci_dev *pdev) -{ - struct pci_dev *bridge; - struct pci_bus *bus; - u16 config; - - if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) - return; - - /* Is VGA routed to us? */ - bus = pdev->bus; - while (bus) { - bridge = bus->self; - if (bridge) { - pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, - &config); - if (!(config & PCI_BRIDGE_CTL_VGA)) - return; - } - bus = bus->parent; - } - pci_read_config_word(pdev, PCI_COMMAND, &config); - if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { - pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; - printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_video); - -/* * Some Toshiba laptops need extra code to enable their TI TSB43AB22/A. * * We pretend to bring them out of full D3 state, and restore the proper diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 1d2ccda..371ab88 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1619,6 +1619,51 @@ static void __devinit fixup_rev1_53c810(struct pci_dev* dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1_53c810); +/* + * Fixup to mark boot BIOS video selected by BIOS before it changes + * + * From information provided by "Jon Smirl" + * + * The standard boot ROM sequence for an x86 machine uses the BIOS + * to select an initial video card for boot display. This boot video + * card will have it's BIOS copied to C0000 in system RAM. + * IORESOURCE_ROM_SHADOW is used to associate the boot video + * card with this copy. On laptops this copy has to be used since + * the main ROM may be compressed or combined with another image. + * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW + * is marked here since the boot video device will be the only enabled + * video device at this point. + */ + +static void __devinit fixup_video(struct pci_dev *pdev) +{ + struct pci_dev *bridge; + struct pci_bus *bus; + u16 config; + + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) + return; + + /* Is VGA routed to us? */ + bus = pdev->bus; + while (bus) { + bridge = bus->self; + if (bridge) { + pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, + &config); + if (!(config & PCI_BRIDGE_CTL_VGA)) + return; + } + bus = bus->parent; + } + pci_read_config_word(pdev, PCI_COMMAND, &config); + if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { + pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; + printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, fixup_video); + static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index f5ee7ce..43e4a49 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -71,7 +71,10 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) void __iomem *image; int last_image; - /* IORESOURCE_ROM_SHADOW only set on x86 */ + /* + * IORESOURCE_ROM_SHADOW set if the VGA enable bit of the Bridge Control + * register is set for embedded VGA. + */ if (res->flags & IORESOURCE_ROM_SHADOW) { /* primary video rom always starts here */ start = (loff_t)0xC0000; -- cgit v0.10.2 From ccc4c7bbd6a2d47bf5899c2c8cf2e0d176a4dc0f Mon Sep 17 00:00:00 2001 From: Vojtech Pavlik Date: Fri, 7 Apr 2006 20:00:27 +0200 Subject: Fix DMA resource allocation in ACPIPnP The ACPIPnP implementation had the understanding of Linux resource flags very wrong, resulting in a nonfunctional implementation of DMA resource allocation. This was usually not a problem, since almost no on-board PnP devices use ISA DMA, with the exception of ECP parallel ports. Even with that, parallel port DMA is preconfigured by the BIOS, so this routine isn't normally called. Except in the case where somebody does 'rmmod parport_pc; modprobe parport_pc', where the rmmod case disables the ECP parallel port resources, and they need to be enabled again to initialize the module. This didn't work, resulting in a non-printing printer. The application doing exactly the above to force reprobing of printers is the YaST printer module. Thus without this fix YaST wedged the printer when configuring it, and was not able to print a test page. Reported-by: Ralf Flaxa Reproduced-by: Jiri Dluhos Signed-off-by: Vojtech Pavlik Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pnp/pnpacpi/rsparser.c b/drivers/pnp/pnpacpi/rsparser.c index dc79b0a..379048f 100644 --- a/drivers/pnp/pnpacpi/rsparser.c +++ b/drivers/pnp/pnpacpi/rsparser.c @@ -776,21 +776,32 @@ static void pnpacpi_encode_dma(struct acpi_resource *resource, struct resource *p) { /* Note: pnp_assign_dma will copy pnp_dma->flags into p->flags */ - if (p->flags & IORESOURCE_DMA_COMPATIBLE) - resource->data.dma.type = ACPI_COMPATIBILITY; - else if (p->flags & IORESOURCE_DMA_TYPEA) - resource->data.dma.type = ACPI_TYPE_A; - else if (p->flags & IORESOURCE_DMA_TYPEB) - resource->data.dma.type = ACPI_TYPE_B; - else if (p->flags & IORESOURCE_DMA_TYPEF) - resource->data.dma.type = ACPI_TYPE_F; - if (p->flags & IORESOURCE_DMA_8BIT) - resource->data.dma.transfer = ACPI_TRANSFER_8; - else if (p->flags & IORESOURCE_DMA_8AND16BIT) - resource->data.dma.transfer = ACPI_TRANSFER_8_16; - else if (p->flags & IORESOURCE_DMA_16BIT) - resource->data.dma.transfer = ACPI_TRANSFER_16; - resource->data.dma.bus_master = p->flags & IORESOURCE_DMA_MASTER; + switch (p->flags & IORESOURCE_DMA_SPEED_MASK) { + case IORESOURCE_DMA_TYPEA: + resource->data.dma.type = ACPI_TYPE_A; + break; + case IORESOURCE_DMA_TYPEB: + resource->data.dma.type = ACPI_TYPE_B; + break; + case IORESOURCE_DMA_TYPEF: + resource->data.dma.type = ACPI_TYPE_F; + break; + default: + resource->data.dma.type = ACPI_COMPATIBILITY; + } + + switch (p->flags & IORESOURCE_DMA_TYPE_MASK) { + case IORESOURCE_DMA_8BIT: + resource->data.dma.transfer = ACPI_TRANSFER_8; + break; + case IORESOURCE_DMA_8AND16BIT: + resource->data.dma.transfer = ACPI_TRANSFER_8_16; + break; + default: + resource->data.dma.transfer = ACPI_TRANSFER_16; + } + + resource->data.dma.bus_master = !!(p->flags & IORESOURCE_DMA_MASTER); resource->data.dma.channel_count = 1; resource->data.dma.channels[0] = p->start; } -- cgit v0.10.2 From 0bed208efcb25bed4dc2026488a4417aa68e7c92 Mon Sep 17 00:00:00 2001 From: "Zhang, Yanmin" Date: Thu, 28 Sep 2006 14:35:59 +0800 Subject: PCI: fix pcie_portdrv_restore_config undefined without CONFIG_PM error On Thu, 2006-09-28 at 03:42, Olaf Hering wrote: > PCI-Express AER implemetation: pcie_portdrv error handler > > This patch breaks if CONFIG_PM is not enabled, > pcie_portdrv_restore_config() will be undefined. I move the definition of pcie_portdrv_restore_config out of CONFIG_PM. Below patch is against 2.6.18-mm1. Could you try it? Signed-off-by: Zhang Yanmin Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index 037690e..b4da795 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -37,7 +37,6 @@ static int pcie_portdrv_save_config(struct pci_dev *dev) return pci_save_state(dev); } -#ifdef CONFIG_PM static int pcie_portdrv_restore_config(struct pci_dev *dev) { int retval; @@ -50,6 +49,7 @@ static int pcie_portdrv_restore_config(struct pci_dev *dev) return 0; } +#ifdef CONFIG_PM static int pcie_portdrv_suspend(struct pci_dev *dev, pm_message_t state) { int ret = pcie_port_device_suspend(dev, state); -- cgit v0.10.2 From 094ed76e8988d46158b036ab150e0c22aff6db3a Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 29 Sep 2006 18:36:15 +0100 Subject: pci: Stamp out pci_find_* usage in fakephp pci_find is not hotplug safe, so it really doesn't want to be in an actual hotplug driver either. Signed-off-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c index 05a4f0f..aaeb112 100644 --- a/drivers/pci/hotplug/fakephp.c +++ b/drivers/pci/hotplug/fakephp.c @@ -181,7 +181,9 @@ static void pci_rescan_slot(struct pci_dev *temp) if (!pci_read_config_byte(temp, PCI_HEADER_TYPE, &hdr_type)) { temp->hdr_type = hdr_type & 0x7f; - if (!pci_find_slot(bus->number, temp->devfn)) { + if ((dev = pci_get_slot(bus, temp->devfn)) != NULL) + pci_dev_put(dev); + else { dev = pci_scan_single_device(bus, temp->devfn); if (dev) { dbg("New device on %s function %x:%x\n", @@ -205,7 +207,9 @@ static void pci_rescan_slot(struct pci_dev *temp) continue; temp->hdr_type = hdr_type & 0x7f; - if (!pci_find_slot(bus->number, temp->devfn)) { + if ((dev = pci_get_slot(bus, temp->devfn)) != NULL) + pci_dev_put(dev); + else { dev = pci_scan_single_device(bus, temp->devfn); if (dev) { dbg("New device on %s function %x:%x\n", @@ -305,7 +309,7 @@ static int disable_slot(struct hotplug_slot *slot) /* search for subfunctions and disable them first */ if (!(dslot->dev->devfn & 7)) { for (func = 1; func < 8; func++) { - dev = pci_find_slot(dslot->dev->bus->number, + dev = pci_get_slot(dslot->dev->bus, dslot->dev->devfn + func); if (dev) { hslot = get_slot_from_dev(dev); @@ -315,6 +319,7 @@ static int disable_slot(struct hotplug_slot *slot) err("Hotplug slot not found for subfunction of PCI device\n"); return -ENODEV; } + pci_dev_put(dev); } else dbg("No device in slot found\n"); } -- cgit v0.10.2 From d1729ccecd7ba9ceb6dca1c973dbfd87041d0637 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Thu, 28 Sep 2006 15:51:21 -0700 Subject: shpchp: fix command completion check This patch fixes the problem that shpchp driver could mis-detect command failures if the system was under heavy load. Signed-off-by: Kenji Kaneshige Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c index 446e9be..4826dd1 100644 --- a/drivers/pci/hotplug/shpchp_hpc.c +++ b/drivers/pci/hotplug/shpchp_hpc.c @@ -302,6 +302,12 @@ static void start_int_poll_timer(struct php_ctlr_state_s *php_ctlr, int sec) add_timer(&php_ctlr->int_poll_timer); } +static inline int is_ctrl_busy(struct controller *ctrl) +{ + u16 cmd_status = shpc_readw(ctrl, CMD_STATUS); + return cmd_status & 0x1; +} + /* * Returns 1 if SHPC finishes executing a command within 1 sec, * otherwise returns 0. @@ -309,16 +315,14 @@ static void start_int_poll_timer(struct php_ctlr_state_s *php_ctlr, int sec) static inline int shpc_poll_ctrl_busy(struct controller *ctrl) { int i; - u16 cmd_status = shpc_readw(ctrl, CMD_STATUS); - if (!(cmd_status & 0x1)) + if (!is_ctrl_busy(ctrl)) return 1; /* Check every 0.1 sec for a total of 1 sec */ for (i = 0; i < 10; i++) { msleep(100); - cmd_status = shpc_readw(ctrl, CMD_STATUS); - if (!(cmd_status & 0x1)) + if (!is_ctrl_busy(ctrl)) return 1; } @@ -336,7 +340,7 @@ static inline int shpc_wait_cmd(struct controller *ctrl) else rc = wait_event_interruptible_timeout(ctrl->queue, !ctrl->cmd_busy, timeout); - if (!rc) { + if (!rc && is_ctrl_busy(ctrl)) { retval = -EIO; err("Command not completed in 1000 msec\n"); } else if (rc < 0) { -- cgit v0.10.2 From 6aa562c248e05db993e4a5f405f899c0cfabb7f2 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Thu, 28 Sep 2006 15:51:36 -0700 Subject: shpchp: remove unnecessary cmd_busy member from struct controller This patch removes unnecessary cmd_busy member from struct controller. Read command status register instead of using cmd_busy. Signed-off-by: Kenji Kaneshige Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h index c7103ac..7e7d490 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -103,7 +103,6 @@ struct controller { u32 cap_offset; unsigned long mmio_base; unsigned long mmio_size; - volatile int cmd_busy; }; diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c index 4826dd1..bbe450f 100644 --- a/drivers/pci/hotplug/shpchp_hpc.c +++ b/drivers/pci/hotplug/shpchp_hpc.c @@ -339,7 +339,7 @@ static inline int shpc_wait_cmd(struct controller *ctrl) rc = shpc_poll_ctrl_busy(ctrl); else rc = wait_event_interruptible_timeout(ctrl->queue, - !ctrl->cmd_busy, timeout); + !is_ctrl_busy(ctrl), timeout); if (!rc && is_ctrl_busy(ctrl)) { retval = -EIO; err("Command not completed in 1000 msec\n"); @@ -347,7 +347,6 @@ static inline int shpc_wait_cmd(struct controller *ctrl) retval = -EINTR; info("Command was interrupted by a signal\n"); } - ctrl->cmd_busy = 0; return retval; } @@ -378,7 +377,6 @@ static int shpc_write_cmd(struct slot *slot, u8 t_slot, u8 cmd) /* To make sure the Controller Busy bit is 0 before we send out the * command. */ - slot->ctrl->cmd_busy = 1; shpc_writew(ctrl, CMD, temp_word); /* @@ -928,7 +926,6 @@ static irqreturn_t shpc_isr(int irq, void *dev_id) serr_int &= ~SERR_INTR_RSVDZ_MASK; shpc_writel(ctrl, SERR_INTR_ENABLE, serr_int); - ctrl->cmd_busy = 0; wake_up_interruptible(&ctrl->queue); } -- cgit v0.10.2 From 662a98fb8de5af4adb56e58f78753cdaa27b6459 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Thu, 5 Oct 2006 12:07:32 +0530 Subject: PCI hotplug: ioremap balanced with iounmap 1. ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. 2. Handle return value correctly Tested (compilation only) with: - allmodconfig Signed-off-by: Amol Lad Cc: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c index bbe450f..83a5226 100644 --- a/drivers/pci/hotplug/shpchp_hpc.c +++ b/drivers/pci/hotplug/shpchp_hpc.c @@ -1118,7 +1118,7 @@ int shpc_init(struct controller * ctrl, struct pci_dev * pdev) { struct php_ctlr_state_s *php_ctlr, *p; void *instance_id = ctrl; - int rc, num_slots = 0; + int rc = -1, num_slots = 0; u8 hp_slot; u32 shpc_base_offset; u32 tempdword, slot_reg, slot_config; @@ -1184,11 +1184,15 @@ int shpc_init(struct controller * ctrl, struct pci_dev * pdev) info("HPC vendor_id %x device_id %x ss_vid %x ss_did %x\n", pdev->vendor, pdev->device, pdev->subsystem_vendor, pdev->subsystem_device); - if (pci_enable_device(pdev)) + rc = pci_enable_device(pdev); + if (rc) { + err("%s: pci_enable_device failed\n", __FUNCTION__); goto abort_free_ctlr; + } if (!request_mem_region(ctrl->mmio_base, ctrl->mmio_size, MY_NAME)) { err("%s: cannot reserve MMIO region\n", __FUNCTION__); + rc = -1; goto abort_free_ctlr; } @@ -1197,6 +1201,7 @@ int shpc_init(struct controller * ctrl, struct pci_dev * pdev) err("%s: cannot remap MMIO region %lx @ %lx\n", __FUNCTION__, ctrl->mmio_size, ctrl->mmio_base); release_mem_region(ctrl->mmio_base, ctrl->mmio_size); + rc = -1; goto abort_free_ctlr; } dbg("%s: php_ctlr->creg %p\n", __FUNCTION__, php_ctlr->creg); @@ -1299,8 +1304,10 @@ int shpc_init(struct controller * ctrl, struct pci_dev * pdev) */ if (atomic_add_return(1, &shpchp_num_controllers) == 1) { shpchp_wq = create_singlethread_workqueue("shpchpd"); - if (!shpchp_wq) - return -ENOMEM; + if (!shpchp_wq) { + rc = -ENOMEM; + goto abort_free_ctlr; + } } /* @@ -1330,8 +1337,10 @@ int shpc_init(struct controller * ctrl, struct pci_dev * pdev) /* We end up here for the many possible ways to fail this API. */ abort_free_ctlr: + if (php_ctlr->creg) + iounmap(php_ctlr->creg); kfree(php_ctlr); abort: DBG_LEAVE_ROUTINE - return -1; + return rc; } -- cgit v0.10.2 From 0306ebfa3b45386401f80aa87cb4f7570bf3aadb Mon Sep 17 00:00:00 2001 From: Brice Goglin Date: Thu, 5 Oct 2006 10:24:31 +0200 Subject: PCI: Improve pci_msi_supported() comments Improve pci_msi_supported() comments. Signed-off-by: Brice Goglin Signed-off-by: Grant Grundler Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index f9fdc54..9fc9a34 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -627,22 +627,24 @@ static int msix_capability_init(struct pci_dev *dev, * pci_msi_supported - check whether MSI may be enabled on device * @dev: pointer to the pci_dev data structure of MSI device function * - * MSI must be globally enabled and supported by the device and its root - * bus. But, the root bus is not easy to find since some architectures - * have virtual busses on top of the PCI hierarchy (for instance the - * hypertransport bus), while the actual bus where MSI must be supported - * is below. So we test the MSI flag on all parent busses and assume - * that no quirk will ever set the NO_MSI flag on a non-root bus. + * Look at global flags, the device itself, and its parent busses + * to return 0 if MSI are supported for the device. **/ static int pci_msi_supported(struct pci_dev * dev) { struct pci_bus *bus; + /* MSI must be globally enabled and supported by the device */ if (!pci_msi_enable || !dev || dev->no_msi) return -EINVAL; - /* check MSI flags of all parent busses */ + /* Any bridge which does NOT route MSI transactions from it's + * secondary bus to it's primary bus must set NO_MSI flag on + * the secondary pci_bus. + * We expect only arch-specific PCI host bus controller driver + * or quirks for specific PCI bridges to be setting NO_MSI. + */ for (bus = dev->bus; bus; bus = bus->parent) if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI) return -EINVAL; -- cgit v0.10.2 From 0cc2b3763e06e84ae5a90b63e03cc1d585a109d0 Mon Sep 17 00:00:00 2001 From: Brice Goglin Date: Thu, 5 Oct 2006 10:24:42 +0200 Subject: PCI: Update MSI-HOWTO.txt according to pci_msi_supported() Update MSI-HOWTO.txt according to pci_msi_supported(). Signed-off-by: Brice Goglin Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/MSI-HOWTO.txt b/Documentation/MSI-HOWTO.txt index c70306a..5c34910 100644 --- a/Documentation/MSI-HOWTO.txt +++ b/Documentation/MSI-HOWTO.txt @@ -470,7 +470,68 @@ LOC: 324553 325068 ERR: 0 MIS: 0 -6. FAQ +6. MSI quirks + +Several PCI chipsets or devices are known to not support MSI. +The PCI stack provides 3 possible levels of MSI disabling: +* on a single device +* on all devices behind a specific bridge +* globally + +6.1. Disabling MSI on a single device + +Under some circumstances, it might be required to disable MSI on a +single device, It may be achived by either not calling pci_enable_msi() +or all, or setting the pci_dev->no_msi flag before (most of the time +in a quirk). + +6.2. Disabling MSI below a bridge + +The vast majority of MSI quirks are required by PCI bridges not +being able to route MSI between busses. In this case, MSI have to be +disabled on all devices behind this bridge. It is achieves by setting +the PCI_BUS_FLAGS_NO_MSI flag in the pci_bus->bus_flags of the bridge +subordinate bus. There is no need to set the same flag on bridges that +are below the broken brigde. When pci_enable_msi() is called to enable +MSI on a device, pci_msi_supported() takes care of checking the NO_MSI +flag in all parent busses of the device. + +Some bridges actually support dynamic MSI support enabling/disabling +by changing some bits in their PCI configuration space (especially +the Hypertransport chipsets such as the nVidia nForce and Serverworks +HT2000). It may then be required to update the NO_MSI flag on the +corresponding devices in the sysfs hierarchy. To enable MSI support +on device "0000:00:0e", do: + + echo 1 > /sys/bus/pci/devices/0000:00:0e/msi_bus + +To disable MSI support, echo 0 instead of 1. Note that it should be +used with caution since changing this value might break interrupts. + +6.3. Disabling MSI globally + +Some extreme cases may require to disable MSI globally on the system. +For now, the only known case is a Serverworks PCI-X chipsets (MSI are +not supported on several busses that are not all connected to the +chipset in the Linux PCI hierarchy). In the vast majority of other +cases, disabling only behind a specific bridge is enough. + +For debugging purpose, the user may also pass pci=nomsi on the kernel +command-line to explicitly disable MSI globally. But, once the appro- +priate quirks are added to the kernel, this option should not be +required anymore. + +6.4. Finding why MSI cannot be enabled on a device + +Assuming that MSI are not enabled on a device, you should look at +dmesg to find messages that quirks may output when disabling MSI +on some devices, some bridges or even globally. +Then, lspci -t gives the list of bridges above a device. Reading +/sys/bus/pci/devices/0000:00:0e/msi_bus will tell you whether MSI +are enabled (1) or disabled (0). In 0 is found in a single bridge +msi_bus file above the device, MSI cannot be enabled. + +7. FAQ Q1. Are there any limitations on using the MSI? -- cgit v0.10.2 From 11f242f04c6d886494cc83097cb6def044eabebb Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Tue, 10 Oct 2006 14:39:00 -0700 Subject: PCI: quirks: switch quirks code offender to use pci_get API Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 371ab88..e8a7f1b 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1840,7 +1840,7 @@ static void __devinit quirk_nvidia_ck804_msi_ht_cap(struct pci_dev *dev) /* check HT MSI cap on this chipset and the root one. * a single one having MSI is enough to be sure that MSI are supported. */ - pdev = pci_find_slot(dev->bus->number, 0); + pdev = pci_get_slot(dev->bus, 0); if (dev->subordinate && !msi_ht_cap_enabled(dev) && !msi_ht_cap_enabled(pdev)) { printk(KERN_WARNING "PCI: MSI quirk detected. " @@ -1848,6 +1848,7 @@ static void __devinit quirk_nvidia_ck804_msi_ht_cap(struct pci_dev *dev) pci_name(dev)); dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI; } + pci_dev_put(pdev); } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, quirk_nvidia_ck804_msi_ht_cap); -- cgit v0.10.2 From 29f3eb64634cf96903a3cdb56b1f9a80bebad17d Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:20:21 -0700 Subject: pci: Additional search functions In order to finish converting to pci_get_* interfaces we need to add a couple of bits of missing functionaility pci_get_bus_and_slot() provides the equivalent to pci_find_slot() (pci_get_slot is already taken as a name for something similar but not the same) pci_get_device_reverse() is the equivalent of pci_find_device_reverse but refcounting Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/search.c b/drivers/pci/search.c index d529462..2f13eba 100644 --- a/drivers/pci/search.c +++ b/drivers/pci/search.c @@ -140,6 +140,31 @@ struct pci_dev * pci_get_slot(struct pci_bus *bus, unsigned int devfn) } /** + * pci_get_bus_and_slot - locate PCI device from a given PCI slot + * @bus: number of PCI bus on which desired PCI device resides + * @devfn: encodes number of PCI slot in which the desired PCI + * device resides and the logical device number within that slot + * in case of multi-function devices. + * + * Given a PCI bus and slot/function number, the desired PCI device + * is located in system global list of PCI devices. If the device + * is found, a pointer to its data structure is returned. If no + * device is found, %NULL is returned. The returned device has its + * reference count bumped by one. + */ + +struct pci_dev * pci_get_bus_and_slot(unsigned int bus, unsigned int devfn) +{ + struct pci_dev *dev = NULL; + + while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { + if (dev->bus->number == bus && dev->devfn == devfn) + return dev; + } + return NULL; +} + +/** * pci_find_subsys - begin or continue searching for a PCI device by vendor/subvendor/device/subdevice id * @vendor: PCI vendor id to match, or %PCI_ANY_ID to match all vendor ids * @device: PCI device id to match, or %PCI_ANY_ID to match all device ids @@ -274,6 +299,45 @@ pci_get_device(unsigned int vendor, unsigned int device, struct pci_dev *from) return pci_get_subsys(vendor, device, PCI_ANY_ID, PCI_ANY_ID, from); } +/** + * pci_get_device_reverse - begin or continue searching for a PCI device by vendor/device id + * @vendor: PCI vendor id to match, or %PCI_ANY_ID to match all vendor ids + * @device: PCI device id to match, or %PCI_ANY_ID to match all device ids + * @from: Previous PCI device found in search, or %NULL for new search. + * + * Iterates through the list of known PCI devices in the reverse order of + * pci_get_device. + * If a PCI device is found with a matching @vendor and @device, the reference + * count to the device is incremented and a pointer to its device structure + * is returned Otherwise, %NULL is returned. A new search is initiated by + * passing %NULL as the @from argument. Otherwise if @from is not %NULL, + * searches continue from next device on the global list. The reference + * count for @from is always decremented if it is not %NULL. + */ +struct pci_dev * +pci_get_device_reverse(unsigned int vendor, unsigned int device, struct pci_dev *from) +{ + struct list_head *n; + struct pci_dev *dev; + + WARN_ON(in_interrupt()); + down_read(&pci_bus_sem); + n = from ? from->global_list.prev : pci_devices.prev; + + while (n && (n != &pci_devices)) { + dev = pci_dev_g(n); + if ((vendor == PCI_ANY_ID || dev->vendor == vendor) && + (device == PCI_ANY_ID || dev->device == device)) + goto exit; + n = n->prev; + } + dev = NULL; +exit: + dev = pci_dev_get(dev); + up_read(&pci_bus_sem); + pci_dev_put(from); + return dev; +} /** * pci_find_device_reverse - begin or continue searching for a PCI device by vendor/device id @@ -382,12 +446,16 @@ exit: } EXPORT_SYMBOL(pci_dev_present); -EXPORT_SYMBOL(pci_find_bus); -EXPORT_SYMBOL(pci_find_next_bus); EXPORT_SYMBOL(pci_find_device); EXPORT_SYMBOL(pci_find_device_reverse); EXPORT_SYMBOL(pci_find_slot); +/* For boot time work */ +EXPORT_SYMBOL(pci_find_bus); +EXPORT_SYMBOL(pci_find_next_bus); +/* For everyone */ EXPORT_SYMBOL(pci_get_device); +EXPORT_SYMBOL(pci_get_device_reverse); EXPORT_SYMBOL(pci_get_subsys); EXPORT_SYMBOL(pci_get_slot); +EXPORT_SYMBOL(pci_get_bus_and_slot); EXPORT_SYMBOL(pci_get_class); diff --git a/include/linux/pci.h b/include/linux/pci.h index 5c604f5..09bf88f 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -452,13 +452,14 @@ struct pci_dev *pci_find_slot (unsigned int bus, unsigned int devfn); int pci_find_capability (struct pci_dev *dev, int cap); int pci_find_next_capability (struct pci_dev *dev, u8 pos, int cap); int pci_find_ext_capability (struct pci_dev *dev, int cap); -struct pci_bus * pci_find_next_bus(const struct pci_bus *from); +struct pci_bus *pci_find_next_bus(const struct pci_bus *from); struct pci_dev *pci_get_device (unsigned int vendor, unsigned int device, struct pci_dev *from); struct pci_dev *pci_get_subsys (unsigned int vendor, unsigned int device, unsigned int ss_vendor, unsigned int ss_device, struct pci_dev *from); struct pci_dev *pci_get_slot (struct pci_bus *bus, unsigned int devfn); +struct pci_dev *pci_get_bus_and_slot (unsigned int bus, unsigned int devfn); struct pci_dev *pci_get_class (unsigned int class, struct pci_dev *from); int pci_dev_present(const struct pci_device_id *ids); -- cgit v0.10.2 From 49c61cca2b6591a28ffa4abb73c718091f569746 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sat, 14 Oct 2006 03:07:30 +0900 Subject: cpcihp_generic: prevent loading without "bridge" parameter cpcihp_generic module requires configured "bridge" module parameter. But it can be loaded successfully without that parameter. Because module init call ends up returning positive value. This patch prevents from loading without setting "bridge" module parameter. Signed-off-by: Akinbou Mita Signed-off-by: Scott Murray Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/cpcihp_generic.c b/drivers/pci/hotplug/cpcihp_generic.c index e847f0d..f3852a6 100644 --- a/drivers/pci/hotplug/cpcihp_generic.c +++ b/drivers/pci/hotplug/cpcihp_generic.c @@ -84,7 +84,7 @@ static int __init validate_parameters(void) if(!bridge) { info("not configured, disabling."); - return 1; + return -EINVAL; } str = bridge; if(!*str) @@ -147,7 +147,7 @@ static int __init cpcihp_generic_init(void) info(DRIVER_DESC " version: " DRIVER_VERSION); status = validate_parameters(); - if(status != 0) + if (status) return status; r = request_region(port, 1, "#ENUM hotswap signal register"); -- cgit v0.10.2 From 6b4b78fed47e7380dfe9280b154e8b9bfcd4c86c Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Fri, 29 Sep 2006 15:23:23 -0500 Subject: PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index ff571f9..dd00fd5 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1231,6 +1231,11 @@ and is between 256 and 4096 characters. It is defined in the file machine check when some devices' config space is read. But various workarounds are disabled and some IOMMU drivers will not work. + bfsort Sort PCI devices into breadth-first order. + This sorting is done to get a device + order compatible with older (<= 2.4) kernels. + nobfsort Don't sort PCI devices into breadth-first order. + pcmv= [HW,PCMCIA] BadgePAD 4 pd. [PARIDE] diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index 68bce194..6d5ace8 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -20,6 +20,7 @@ unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 | PCI_PROBE_MMCONF; +int pci_bf_sort; int pci_routeirq; int pcibios_last_bus = -1; unsigned long pirq_table_addr; @@ -118,6 +119,20 @@ void __devinit pcibios_fixup_bus(struct pci_bus *b) } /* + * Only use DMI information to set this if nothing was passed + * on the kernel command line (which was parsed earlier). + */ + +static int __devinit set_bf_sort(struct dmi_system_id *d) +{ + if (pci_bf_sort == pci_bf_sort_default) { + pci_bf_sort = pci_dmi_bf; + printk(KERN_INFO "PCI: %s detected, enabling pci=bfsort.\n", d->ident); + } + return 0; +} + +/* * Enable renumbering of PCI bus# ranges to reach all PCI busses (Cardbus) */ #ifdef __i386__ @@ -130,11 +145,11 @@ static int __devinit assign_all_busses(struct dmi_system_id *d) } #endif +static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { +#ifdef __i386__ /* * Laptops which need pci=assign-busses to see Cardbus cards */ -static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { -#ifdef __i386__ { .callback = assign_all_busses, .ident = "Samsung X20 Laptop", @@ -144,6 +159,38 @@ static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { }, }, #endif /* __i386__ */ + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 1950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1950"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 1955", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1955"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 2900", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2900"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 2950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2950"), + }, + }, {} }; @@ -189,6 +236,8 @@ static int __init pcibios_init(void) pcibios_resource_survey(); + if (pci_bf_sort >= pci_force_bf) + pci_sort_breadthfirst(); #ifdef CONFIG_PCI_BIOS if ((pci_probe & PCI_BIOS_SORT) && !(pci_probe & PCI_NO_SORT)) pcibios_sort(); @@ -203,6 +252,12 @@ char * __devinit pcibios_setup(char *str) if (!strcmp(str, "off")) { pci_probe = 0; return NULL; + } else if (!strcmp(str, "bfsort")) { + pci_bf_sort = pci_force_bf; + return NULL; + } else if (!strcmp(str, "nobfsort")) { + pci_bf_sort = pci_force_nobf; + return NULL; } #ifdef CONFIG_PCI_BIOS else if (!strcmp(str, "bios")) { diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h index 1814f74..ad065ce 100644 --- a/arch/i386/pci/pci.h +++ b/arch/i386/pci/pci.h @@ -30,6 +30,13 @@ extern unsigned int pci_probe; extern unsigned long pirq_table_addr; +enum pci_bf_sort_state { + pci_bf_sort_default, + pci_force_nobf, + pci_force_bf, + pci_dmi_bf, +}; + /* pci-i386.c */ extern unsigned int pcibios_max_latency; diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index a3b0a5e..e159d66 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1067,3 +1067,95 @@ EXPORT_SYMBOL(pci_scan_bridge); EXPORT_SYMBOL(pci_scan_single_device); EXPORT_SYMBOL_GPL(pci_scan_child_bus); #endif + +static int __init pci_sort_bf_cmp(const struct pci_dev *a, const struct pci_dev *b) +{ + if (pci_domain_nr(a->bus) < pci_domain_nr(b->bus)) return -1; + else if (pci_domain_nr(a->bus) > pci_domain_nr(b->bus)) return 1; + + if (a->bus->number < b->bus->number) return -1; + else if (a->bus->number > b->bus->number) return 1; + + if (a->devfn < b->devfn) return -1; + else if (a->devfn > b->devfn) return 1; + + return 0; +} + +/* + * Yes, this forcably breaks the klist abstraction temporarily. It + * just wants to sort the klist, not change reference counts and + * take/drop locks rapidly in the process. It does all this while + * holding the lock for the list, so objects can't otherwise be + * added/removed while we're swizzling. + */ +static void __init pci_insertion_sort_klist(struct pci_dev *a, struct list_head *list) +{ + struct list_head *pos; + struct klist_node *n; + struct device *dev; + struct pci_dev *b; + + list_for_each(pos, list) { + n = container_of(pos, struct klist_node, n_node); + dev = container_of(n, struct device, knode_bus); + b = to_pci_dev(dev); + if (pci_sort_bf_cmp(a, b) <= 0) { + list_move_tail(&a->dev.knode_bus.n_node, &b->dev.knode_bus.n_node); + return; + } + } + list_move_tail(&a->dev.knode_bus.n_node, list); +} + +static void __init pci_sort_breadthfirst_klist(void) +{ + LIST_HEAD(sorted_devices); + struct list_head *pos, *tmp; + struct klist_node *n; + struct device *dev; + struct pci_dev *pdev; + + spin_lock(&pci_bus_type.klist_devices.k_lock); + list_for_each_safe(pos, tmp, &pci_bus_type.klist_devices.k_list) { + n = container_of(pos, struct klist_node, n_node); + dev = container_of(n, struct device, knode_bus); + pdev = to_pci_dev(dev); + pci_insertion_sort_klist(pdev, &sorted_devices); + } + list_splice(&sorted_devices, &pci_bus_type.klist_devices.k_list); + spin_unlock(&pci_bus_type.klist_devices.k_lock); +} + +static void __init pci_insertion_sort_devices(struct pci_dev *a, struct list_head *list) +{ + struct pci_dev *b; + + list_for_each_entry(b, list, global_list) { + if (pci_sort_bf_cmp(a, b) <= 0) { + list_move_tail(&a->global_list, &b->global_list); + return; + } + } + list_move_tail(&a->global_list, list); +} + +static void __init pci_sort_breadthfirst_devices(void) +{ + LIST_HEAD(sorted_devices); + struct pci_dev *dev, *tmp; + + down_write(&pci_bus_sem); + list_for_each_entry_safe(dev, tmp, &pci_devices, global_list) { + pci_insertion_sort_devices(dev, &sorted_devices); + } + list_splice(&sorted_devices, &pci_devices); + up_write(&pci_bus_sem); +} + +void __init pci_sort_breadthfirst(void) +{ + pci_sort_breadthfirst_devices(); + pci_sort_breadthfirst_klist(); +} + diff --git a/include/linux/pci.h b/include/linux/pci.h index 09bf88f..4689e2a 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -443,6 +443,7 @@ extern void pci_remove_bus(struct pci_bus *b); extern void pci_remove_bus_device(struct pci_dev *dev); extern void pci_stop_bus_device(struct pci_dev *dev); void pci_setup_cardbus(struct pci_bus *bus); +extern void pci_sort_breadthfirst(void); /* Generic PCI functions exported to card drivers */ -- cgit v0.10.2 From fb5f4d7a74a140f8e033d1e6854989e88c36c6b8 Mon Sep 17 00:00:00 2001 From: Kristen Carlson Accardi Date: Fri, 29 Sep 2006 10:30:27 -0700 Subject: change pci hotplug subsystem maintainer to Kristen Here's a patch adding me to the maintainers file for the pci hotplug subsystem, as we discussed. Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/MAINTAINERS b/MAINTAINERS index 5305dd6..9b6b882 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2309,8 +2309,8 @@ T: quilt kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/ S: Supported PCI HOTPLUG CORE -P: Greg Kroah-Hartman -M: gregkh@suse.de +P: Kristen Carlson Accardi +M: kristen.c.accardi@intel.com S: Supported PCI HOTPLUG COMPAQ DRIVER diff --git a/drivers/pci/hotplug/pci_hotplug.h b/drivers/pci/hotplug/pci_hotplug.h index 772523d..a675a05 100644 --- a/drivers/pci/hotplug/pci_hotplug.h +++ b/drivers/pci/hotplug/pci_hotplug.h @@ -22,7 +22,7 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * Send feedback to + * Send feedback to * */ #ifndef _PCI_HOTPLUG_H diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c index e2823ea..fa666d0 100644 --- a/drivers/pci/hotplug/pci_hotplug_core.c +++ b/drivers/pci/hotplug/pci_hotplug_core.c @@ -21,9 +21,7 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * Send feedback to - * - * Filesystem portion based on work done by Pat Mochel on ddfs/driverfs + * Send feedback to * */ -- cgit v0.10.2 From 7a54f25cef6c763f16c9fd49ae382de162147873 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 13 Oct 2006 20:05:19 -0700 Subject: PCI Hotplug: move pci_hotplug.h to include/linux/ This makes it possible to build pci hotplug drivers outside of the main kernel tree, and Sam keeps telling me to move local header files to their proper places... Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index 51cb9f8..270a33c 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -29,10 +29,10 @@ #include #include #include +#include #include #include #include -#include "pci_hotplug.h" #define MY_NAME "acpi_pcihp" diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h index 7fff07e..59c5b24 100644 --- a/drivers/pci/hotplug/acpiphp.h +++ b/drivers/pci/hotplug/acpiphp.h @@ -38,7 +38,7 @@ #include #include /* for KOBJ_NAME_LEN */ #include -#include "pci_hotplug.h" +#include #define dbg(format, arg...) \ do { \ diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c index e2fef60..c57d9d5c 100644 --- a/drivers/pci/hotplug/acpiphp_core.c +++ b/drivers/pci/hotplug/acpiphp_core.c @@ -37,10 +37,10 @@ #include #include +#include #include #include #include -#include "pci_hotplug.h" #include "acpiphp.h" #define MY_NAME "acpiphp" diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 83e8e44..c44311a 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -45,11 +45,11 @@ #include #include +#include #include #include #include "../pci.h" -#include "pci_hotplug.h" #include "acpiphp.h" static LIST_HEAD(bridge_list); diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c index d0a07d9..bd40aee 100644 --- a/drivers/pci/hotplug/acpiphp_ibm.c +++ b/drivers/pci/hotplug/acpiphp_ibm.c @@ -35,7 +35,6 @@ #include #include "acpiphp.h" -#include "pci_hotplug.h" #define DRIVER_VERSION "1.0.1" #define DRIVER_AUTHOR "Irene Zubarev , Vernon Mauery " diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c index d06ab40..6845515 100644 --- a/drivers/pci/hotplug/cpci_hotplug_core.c +++ b/drivers/pci/hotplug/cpci_hotplug_core.c @@ -29,12 +29,12 @@ #include #include #include +#include #include #include #include #include #include -#include "pci_hotplug.h" #include "cpci_hotplug.h" #define DRIVER_AUTHOR "Scott Murray " diff --git a/drivers/pci/hotplug/cpci_hotplug_pci.c b/drivers/pci/hotplug/cpci_hotplug_pci.c index 4afcaff..7b1beaa 100644 --- a/drivers/pci/hotplug/cpci_hotplug_pci.c +++ b/drivers/pci/hotplug/cpci_hotplug_pci.c @@ -26,9 +26,9 @@ #include #include #include +#include #include #include "../pci.h" -#include "pci_hotplug.h" #include "cpci_hotplug.h" #define MY_NAME "cpci_hotplug" diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h index ea040c3..298ad7f 100644 --- a/drivers/pci/hotplug/cpqphp.h +++ b/drivers/pci/hotplug/cpqphp.h @@ -28,7 +28,6 @@ #ifndef _CPQPHP_H #define _CPQPHP_H -#include "pci_hotplug.h" #include #include /* for read? and write? functions */ #include /* for delays */ diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c index 1fc2599..5617cfd 100644 --- a/drivers/pci/hotplug/cpqphp_core.c +++ b/drivers/pci/hotplug/cpqphp_core.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c index 3ec2ad7..79ff6b4 100644 --- a/drivers/pci/hotplug/cpqphp_ctrl.c +++ b/drivers/pci/hotplug/cpqphp_ctrl.c @@ -36,6 +36,7 @@ #include #include #include +#include #include "cpqphp.h" static u32 configure_new_device(struct controller* ctrl, struct pci_func *func, diff --git a/drivers/pci/hotplug/cpqphp_nvram.c b/drivers/pci/hotplug/cpqphp_nvram.c index cf08789..298a6cf 100644 --- a/drivers/pci/hotplug/cpqphp_nvram.c +++ b/drivers/pci/hotplug/cpqphp_nvram.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include "cpqphp.h" diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c index 0d96889..fc7c74d 100644 --- a/drivers/pci/hotplug/cpqphp_pci.c +++ b/drivers/pci/hotplug/cpqphp_pci.c @@ -33,6 +33,7 @@ #include #include #include +#include #include "../pci.h" #include "cpqphp.h" #include "cpqphp_nvram.h" diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c index 5bab666..634f74d 100644 --- a/drivers/pci/hotplug/cpqphp_sysfs.c +++ b/drivers/pci/hotplug/cpqphp_sysfs.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include "cpqphp.h" diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c index aaeb112..e27907c 100644 --- a/drivers/pci/hotplug/fakephp.c +++ b/drivers/pci/hotplug/fakephp.c @@ -35,10 +35,10 @@ #include #include #include +#include #include #include #include -#include "pci_hotplug.h" #include "../pci.h" #if !defined(MODULE) diff --git a/drivers/pci/hotplug/ibmphp.h b/drivers/pci/hotplug/ibmphp.h index dba6d8c..612d963 100644 --- a/drivers/pci/hotplug/ibmphp.h +++ b/drivers/pci/hotplug/ibmphp.h @@ -30,7 +30,7 @@ * */ -#include "pci_hotplug.h" +#include extern int ibmphp_debug; diff --git a/drivers/pci/hotplug/pci_hotplug.h b/drivers/pci/hotplug/pci_hotplug.h deleted file mode 100644 index a675a05..0000000 --- a/drivers/pci/hotplug/pci_hotplug.h +++ /dev/null @@ -1,236 +0,0 @@ -/* - * PCI HotPlug Core Functions - * - * Copyright (C) 1995,2001 Compaq Computer Corporation - * Copyright (C) 2001 Greg Kroah-Hartman (greg@kroah.com) - * Copyright (C) 2001 IBM Corp. - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to - * - */ -#ifndef _PCI_HOTPLUG_H -#define _PCI_HOTPLUG_H - - -/* These values come from the PCI Hotplug Spec */ -enum pci_bus_speed { - PCI_SPEED_33MHz = 0x00, - PCI_SPEED_66MHz = 0x01, - PCI_SPEED_66MHz_PCIX = 0x02, - PCI_SPEED_100MHz_PCIX = 0x03, - PCI_SPEED_133MHz_PCIX = 0x04, - PCI_SPEED_66MHz_PCIX_ECC = 0x05, - PCI_SPEED_100MHz_PCIX_ECC = 0x06, - PCI_SPEED_133MHz_PCIX_ECC = 0x07, - PCI_SPEED_66MHz_PCIX_266 = 0x09, - PCI_SPEED_100MHz_PCIX_266 = 0x0a, - PCI_SPEED_133MHz_PCIX_266 = 0x0b, - PCI_SPEED_66MHz_PCIX_533 = 0x11, - PCI_SPEED_100MHz_PCIX_533 = 0x12, - PCI_SPEED_133MHz_PCIX_533 = 0x13, - PCI_SPEED_UNKNOWN = 0xff, -}; - -/* These values come from the PCI Express Spec */ -enum pcie_link_width { - PCIE_LNK_WIDTH_RESRV = 0x00, - PCIE_LNK_X1 = 0x01, - PCIE_LNK_X2 = 0x02, - PCIE_LNK_X4 = 0x04, - PCIE_LNK_X8 = 0x08, - PCIE_LNK_X12 = 0x0C, - PCIE_LNK_X16 = 0x10, - PCIE_LNK_X32 = 0x20, - PCIE_LNK_WIDTH_UNKNOWN = 0xFF, -}; - -enum pcie_link_speed { - PCIE_2PT5GB = 0x14, - PCIE_LNK_SPEED_UNKNOWN = 0xFF, -}; - -struct hotplug_slot; -struct hotplug_slot_attribute { - struct attribute attr; - ssize_t (*show)(struct hotplug_slot *, char *); - ssize_t (*store)(struct hotplug_slot *, const char *, size_t); -}; -#define to_hotplug_attr(n) container_of(n, struct hotplug_slot_attribute, attr); - -/** - * struct hotplug_slot_ops -the callbacks that the hotplug pci core can use - * @owner: The module owner of this structure - * @enable_slot: Called when the user wants to enable a specific pci slot - * @disable_slot: Called when the user wants to disable a specific pci slot - * @set_attention_status: Called to set the specific slot's attention LED to - * the specified value - * @hardware_test: Called to run a specified hardware test on the specified - * slot. - * @get_power_status: Called to get the current power status of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_attention_status: Called to get the current attention status of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_latch_status: Called to get the current latch status of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_adapter_status: Called to get see if an adapter is present in the slot or not. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_address: Called to get pci address of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_max_bus_speed: Called to get the max bus speed for a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_cur_bus_speed: Called to get the current bus speed for a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * - * The table of function pointers that is passed to the hotplug pci core by a - * hotplug pci driver. These functions are called by the hotplug pci core when - * the user wants to do something to a specific slot (query it for information, - * set an LED, enable / disable power, etc.) - */ -struct hotplug_slot_ops { - struct module *owner; - int (*enable_slot) (struct hotplug_slot *slot); - int (*disable_slot) (struct hotplug_slot *slot); - int (*set_attention_status) (struct hotplug_slot *slot, u8 value); - int (*hardware_test) (struct hotplug_slot *slot, u32 value); - int (*get_power_status) (struct hotplug_slot *slot, u8 *value); - int (*get_attention_status) (struct hotplug_slot *slot, u8 *value); - int (*get_latch_status) (struct hotplug_slot *slot, u8 *value); - int (*get_adapter_status) (struct hotplug_slot *slot, u8 *value); - int (*get_address) (struct hotplug_slot *slot, u32 *value); - int (*get_max_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); - int (*get_cur_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); -}; - -/** - * struct hotplug_slot_info - used to notify the hotplug pci core of the state of the slot - * @power: if power is enabled or not (1/0) - * @attention_status: if the attention light is enabled or not (1/0) - * @latch_status: if the latch (if any) is open or closed (1/0) - * @adapter_present: if there is a pci board present in the slot or not (1/0) - * @address: (domain << 16 | bus << 8 | dev) - * - * Used to notify the hotplug pci core of the status of a specific slot. - */ -struct hotplug_slot_info { - u8 power_status; - u8 attention_status; - u8 latch_status; - u8 adapter_status; - u32 address; - enum pci_bus_speed max_bus_speed; - enum pci_bus_speed cur_bus_speed; -}; - -/** - * struct hotplug_slot - used to register a physical slot with the hotplug pci core - * @name: the name of the slot being registered. This string must - * be unique amoung slots registered on this system. - * @ops: pointer to the &struct hotplug_slot_ops to be used for this slot - * @info: pointer to the &struct hotplug_slot_info for the initial values for - * this slot. - * @release: called during pci_hp_deregister to free memory allocated in a - * hotplug_slot structure. - * @private: used by the hotplug pci controller driver to store whatever it - * needs. - */ -struct hotplug_slot { - char *name; - struct hotplug_slot_ops *ops; - struct hotplug_slot_info *info; - void (*release) (struct hotplug_slot *slot); - void *private; - - /* Variables below this are for use only by the hotplug pci core. */ - struct list_head slot_list; - struct kobject kobj; -}; -#define to_hotplug_slot(n) container_of(n, struct hotplug_slot, kobj) - -extern int pci_hp_register (struct hotplug_slot *slot); -extern int pci_hp_deregister (struct hotplug_slot *slot); -extern int __must_check pci_hp_change_slot_info (struct hotplug_slot *slot, - struct hotplug_slot_info *info); -extern struct subsystem pci_hotplug_slots_subsys; - -/* PCI Setting Record (Type 0) */ -struct hpp_type0 { - u32 revision; - u8 cache_line_size; - u8 latency_timer; - u8 enable_serr; - u8 enable_perr; -}; - -/* PCI-X Setting Record (Type 1) */ -struct hpp_type1 { - u32 revision; - u8 max_mem_read; - u8 avg_max_split; - u16 tot_max_split; -}; - -/* PCI Express Setting Record (Type 2) */ -struct hpp_type2 { - u32 revision; - u32 unc_err_mask_and; - u32 unc_err_mask_or; - u32 unc_err_sever_and; - u32 unc_err_sever_or; - u32 cor_err_mask_and; - u32 cor_err_mask_or; - u32 adv_err_cap_and; - u32 adv_err_cap_or; - u16 pci_exp_devctl_and; - u16 pci_exp_devctl_or; - u16 pci_exp_lnkctl_and; - u16 pci_exp_lnkctl_or; - u32 sec_unc_err_sever_and; - u32 sec_unc_err_sever_or; - u32 sec_unc_err_mask_and; - u32 sec_unc_err_mask_or; -}; - -struct hotplug_params { - struct hpp_type0 *t0; /* Type0: NULL if not available */ - struct hpp_type1 *t1; /* Type1: NULL if not available */ - struct hpp_type2 *t2; /* Type2: NULL if not available */ - struct hpp_type0 type0_data; - struct hpp_type1 type1_data; - struct hpp_type2 type2_data; -}; - -#ifdef CONFIG_ACPI -#include -#include -#include -extern acpi_status acpi_run_oshp(acpi_handle handle); -extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, - struct hotplug_params *hpp); -int acpi_root_bridge(acpi_handle handle); -#endif -#endif - diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c index fa666d0..f5d632e 100644 --- a/drivers/pci/hotplug/pci_hotplug_core.c +++ b/drivers/pci/hotplug/pci_hotplug_core.c @@ -30,6 +30,8 @@ #include #include #include +#include +#include #include #include #include @@ -37,11 +39,8 @@ #include #include #include +#include #include -#include -#include -#include "pci_hotplug.h" - #define MY_NAME "pci_hotplug" diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 30f021c..4fb12fc 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -31,11 +31,11 @@ #include #include +#include #include #include /* signal_pending() */ #include #include -#include "pci_hotplug.h" #define MY_NAME "pciehp" diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c index 2b9e10e..50bcd3f 100644 --- a/drivers/pci/hotplug/pcihp_skeleton.c +++ b/drivers/pci/hotplug/pcihp_skeleton.c @@ -33,8 +33,8 @@ #include #include #include +#include #include -#include "pci_hotplug.h" #define SLOT_NAME_SIZE 10 struct slot { diff --git a/drivers/pci/hotplug/rpadlpar_sysfs.c b/drivers/pci/hotplug/rpadlpar_sysfs.c index db69be8..6c5be3f 100644 --- a/drivers/pci/hotplug/rpadlpar_sysfs.c +++ b/drivers/pci/hotplug/rpadlpar_sysfs.c @@ -14,7 +14,7 @@ */ #include #include -#include "pci_hotplug.h" +#include #include "rpadlpar.h" #define DLPAR_KOBJ_NAME "control" diff --git a/drivers/pci/hotplug/rpaphp_core.c b/drivers/pci/hotplug/rpaphp_core.c index 7288a3e..141486d 100644 --- a/drivers/pci/hotplug/rpaphp_core.c +++ b/drivers/pci/hotplug/rpaphp_core.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -36,7 +37,6 @@ #include "../pci.h" /* for pci_add_new_bus */ /* and pci_do_scan_bus */ #include "rpaphp.h" -#include "pci_hotplug.h" int debug; static struct semaphore rpaphp_sem; diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c index f31d83c..b62ad31 100644 --- a/drivers/pci/hotplug/sgi_hotplug.c +++ b/drivers/pci/hotplug/sgi_hotplug.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -29,7 +30,6 @@ #include #include "../pci.h" -#include "pci_hotplug.h" MODULE_LICENSE("GPL"); MODULE_AUTHOR("SGI (prarit@sgi.com, dickie@sgi.com, habeck@sgi.com)"); diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h index 7e7d490..ea2087c 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -31,12 +31,11 @@ #include #include +#include #include #include /* signal_pending(), struct timer_list */ #include -#include "pci_hotplug.h" - #if !defined(MODULE) #define MY_NAME "shpchp" #else diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h new file mode 100644 index 0000000..a675a05 --- /dev/null +++ b/include/linux/pci_hotplug.h @@ -0,0 +1,236 @@ +/* + * PCI HotPlug Core Functions + * + * Copyright (C) 1995,2001 Compaq Computer Corporation + * Copyright (C) 2001 Greg Kroah-Hartman (greg@kroah.com) + * Copyright (C) 2001 IBM Corp. + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Send feedback to + * + */ +#ifndef _PCI_HOTPLUG_H +#define _PCI_HOTPLUG_H + + +/* These values come from the PCI Hotplug Spec */ +enum pci_bus_speed { + PCI_SPEED_33MHz = 0x00, + PCI_SPEED_66MHz = 0x01, + PCI_SPEED_66MHz_PCIX = 0x02, + PCI_SPEED_100MHz_PCIX = 0x03, + PCI_SPEED_133MHz_PCIX = 0x04, + PCI_SPEED_66MHz_PCIX_ECC = 0x05, + PCI_SPEED_100MHz_PCIX_ECC = 0x06, + PCI_SPEED_133MHz_PCIX_ECC = 0x07, + PCI_SPEED_66MHz_PCIX_266 = 0x09, + PCI_SPEED_100MHz_PCIX_266 = 0x0a, + PCI_SPEED_133MHz_PCIX_266 = 0x0b, + PCI_SPEED_66MHz_PCIX_533 = 0x11, + PCI_SPEED_100MHz_PCIX_533 = 0x12, + PCI_SPEED_133MHz_PCIX_533 = 0x13, + PCI_SPEED_UNKNOWN = 0xff, +}; + +/* These values come from the PCI Express Spec */ +enum pcie_link_width { + PCIE_LNK_WIDTH_RESRV = 0x00, + PCIE_LNK_X1 = 0x01, + PCIE_LNK_X2 = 0x02, + PCIE_LNK_X4 = 0x04, + PCIE_LNK_X8 = 0x08, + PCIE_LNK_X12 = 0x0C, + PCIE_LNK_X16 = 0x10, + PCIE_LNK_X32 = 0x20, + PCIE_LNK_WIDTH_UNKNOWN = 0xFF, +}; + +enum pcie_link_speed { + PCIE_2PT5GB = 0x14, + PCIE_LNK_SPEED_UNKNOWN = 0xFF, +}; + +struct hotplug_slot; +struct hotplug_slot_attribute { + struct attribute attr; + ssize_t (*show)(struct hotplug_slot *, char *); + ssize_t (*store)(struct hotplug_slot *, const char *, size_t); +}; +#define to_hotplug_attr(n) container_of(n, struct hotplug_slot_attribute, attr); + +/** + * struct hotplug_slot_ops -the callbacks that the hotplug pci core can use + * @owner: The module owner of this structure + * @enable_slot: Called when the user wants to enable a specific pci slot + * @disable_slot: Called when the user wants to disable a specific pci slot + * @set_attention_status: Called to set the specific slot's attention LED to + * the specified value + * @hardware_test: Called to run a specified hardware test on the specified + * slot. + * @get_power_status: Called to get the current power status of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_attention_status: Called to get the current attention status of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_latch_status: Called to get the current latch status of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_adapter_status: Called to get see if an adapter is present in the slot or not. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_address: Called to get pci address of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_max_bus_speed: Called to get the max bus speed for a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_cur_bus_speed: Called to get the current bus speed for a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * + * The table of function pointers that is passed to the hotplug pci core by a + * hotplug pci driver. These functions are called by the hotplug pci core when + * the user wants to do something to a specific slot (query it for information, + * set an LED, enable / disable power, etc.) + */ +struct hotplug_slot_ops { + struct module *owner; + int (*enable_slot) (struct hotplug_slot *slot); + int (*disable_slot) (struct hotplug_slot *slot); + int (*set_attention_status) (struct hotplug_slot *slot, u8 value); + int (*hardware_test) (struct hotplug_slot *slot, u32 value); + int (*get_power_status) (struct hotplug_slot *slot, u8 *value); + int (*get_attention_status) (struct hotplug_slot *slot, u8 *value); + int (*get_latch_status) (struct hotplug_slot *slot, u8 *value); + int (*get_adapter_status) (struct hotplug_slot *slot, u8 *value); + int (*get_address) (struct hotplug_slot *slot, u32 *value); + int (*get_max_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); + int (*get_cur_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); +}; + +/** + * struct hotplug_slot_info - used to notify the hotplug pci core of the state of the slot + * @power: if power is enabled or not (1/0) + * @attention_status: if the attention light is enabled or not (1/0) + * @latch_status: if the latch (if any) is open or closed (1/0) + * @adapter_present: if there is a pci board present in the slot or not (1/0) + * @address: (domain << 16 | bus << 8 | dev) + * + * Used to notify the hotplug pci core of the status of a specific slot. + */ +struct hotplug_slot_info { + u8 power_status; + u8 attention_status; + u8 latch_status; + u8 adapter_status; + u32 address; + enum pci_bus_speed max_bus_speed; + enum pci_bus_speed cur_bus_speed; +}; + +/** + * struct hotplug_slot - used to register a physical slot with the hotplug pci core + * @name: the name of the slot being registered. This string must + * be unique amoung slots registered on this system. + * @ops: pointer to the &struct hotplug_slot_ops to be used for this slot + * @info: pointer to the &struct hotplug_slot_info for the initial values for + * this slot. + * @release: called during pci_hp_deregister to free memory allocated in a + * hotplug_slot structure. + * @private: used by the hotplug pci controller driver to store whatever it + * needs. + */ +struct hotplug_slot { + char *name; + struct hotplug_slot_ops *ops; + struct hotplug_slot_info *info; + void (*release) (struct hotplug_slot *slot); + void *private; + + /* Variables below this are for use only by the hotplug pci core. */ + struct list_head slot_list; + struct kobject kobj; +}; +#define to_hotplug_slot(n) container_of(n, struct hotplug_slot, kobj) + +extern int pci_hp_register (struct hotplug_slot *slot); +extern int pci_hp_deregister (struct hotplug_slot *slot); +extern int __must_check pci_hp_change_slot_info (struct hotplug_slot *slot, + struct hotplug_slot_info *info); +extern struct subsystem pci_hotplug_slots_subsys; + +/* PCI Setting Record (Type 0) */ +struct hpp_type0 { + u32 revision; + u8 cache_line_size; + u8 latency_timer; + u8 enable_serr; + u8 enable_perr; +}; + +/* PCI-X Setting Record (Type 1) */ +struct hpp_type1 { + u32 revision; + u8 max_mem_read; + u8 avg_max_split; + u16 tot_max_split; +}; + +/* PCI Express Setting Record (Type 2) */ +struct hpp_type2 { + u32 revision; + u32 unc_err_mask_and; + u32 unc_err_mask_or; + u32 unc_err_sever_and; + u32 unc_err_sever_or; + u32 cor_err_mask_and; + u32 cor_err_mask_or; + u32 adv_err_cap_and; + u32 adv_err_cap_or; + u16 pci_exp_devctl_and; + u16 pci_exp_devctl_or; + u16 pci_exp_lnkctl_and; + u16 pci_exp_lnkctl_or; + u32 sec_unc_err_sever_and; + u32 sec_unc_err_sever_or; + u32 sec_unc_err_mask_and; + u32 sec_unc_err_mask_or; +}; + +struct hotplug_params { + struct hpp_type0 *t0; /* Type0: NULL if not available */ + struct hpp_type1 *t1; /* Type1: NULL if not available */ + struct hpp_type2 *t2; /* Type2: NULL if not available */ + struct hpp_type0 type0_data; + struct hpp_type1 type1_data; + struct hpp_type2 type2_data; +}; + +#ifdef CONFIG_ACPI +#include +#include +#include +extern acpi_status acpi_run_oshp(acpi_handle handle); +extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, + struct hotplug_params *hpp); +int acpi_root_bridge(acpi_handle handle); +#endif +#endif + -- cgit v0.10.2 From acbd39fbc5d8757aa920c6045399374df7a6dd68 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Sat, 30 Sep 2006 22:41:43 -0400 Subject: Documentation: feature-removal-schedule typo Fix typo in newly added feature remove schedule item. Signed-off-by: Dominik Brodowski Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 24f3c63..1ac3c74 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -255,7 +255,7 @@ Who: Stephen Hemminger What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment -When: Oktober 2008 +When: October 2008 Why: The stacking of class devices makes these values misleading and inconsistent. Class devices should not carry any of these properties, and bus -- cgit v0.10.2 From 0fbf116d120a2dc5d808204c7d86ad35f7d7846f Mon Sep 17 00:00:00 2001 From: Duncan Sands Date: Wed, 27 Sep 2006 23:38:08 +0200 Subject: Driver core: plug device probe memory leak Make sure data is freed if the kthread fails to start. Signed-off-by: Duncan Sands Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/dd.c b/drivers/base/dd.c index b5f43c3..ef7db69 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -178,7 +178,7 @@ int driver_probe_device(struct device_driver * drv, struct device * dev) probe_task = kthread_run(really_probe, data, "probe-%s", dev->bus_id); if (IS_ERR(probe_task)) - ret = PTR_ERR(probe_task); + ret = really_probe(data); } else ret = really_probe(data); -- cgit v0.10.2 From 310a922d4307ed38b37982a6f93b11fdf3b8dcb1 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Sat, 23 Sep 2006 23:35:04 -0600 Subject: Fix dev_printk() is now GPL-only Make dev_printk usable from non-GPL modules again dev_printk now calls dev_driver_string. We want even proprietary modules to be calling dev_printk, so the export of dev_driver_string needs to be non-GPL-only. Signed-off-by: Matthew Wilcox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/core.c b/drivers/base/core.c index b224bb4..aee3743 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -44,7 +44,7 @@ const char *dev_driver_string(struct device *dev) return dev->driver ? dev->driver->name : (dev->bus ? dev->bus->name : ""); } -EXPORT_SYMBOL_GPL(dev_driver_string); +EXPORT_SYMBOL(dev_driver_string); #define to_dev(obj) container_of(obj, struct device, kobj) #define to_dev_attr(_attr) container_of(_attr, struct device_attribute, attr) -- cgit v0.10.2 From 722385f75efd82d9f480f0765a1e97a4d83cac0d Mon Sep 17 00:00:00 2001 From: Diego Calleja Date: Thu, 21 Sep 2006 22:37:10 +0200 Subject: HOWTO: bug report addition I suspect that not many people is subscribed to the bugzilla mailing list, not surprising since the URLs doesn't seem to be in the tree :) After fixing my english, I wonder if the following patch could be applied... Signed-off-by: Diego Calleja Acked-by: Randy Dunlap Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/HOWTO b/Documentation/HOWTO index d6f3dd1..8d51c14 100644 --- a/Documentation/HOWTO +++ b/Documentation/HOWTO @@ -395,6 +395,26 @@ bugme-janitor mailing list (every change in the bugzilla is mailed here) +Managing bug reports +-------------------- + +One of the best ways to put into practice your hacking skills is by fixing +bugs reported by other people. Not only you will help to make the kernel +more stable, you'll learn to fix real world problems and you will improve +your skills, and other developers will be aware of your presence. Fixing +bugs is one of the best ways to get merits among other developers, because +not many people like wasting time fixing other people's bugs. + +To work in the already reported bug reports, go to http://bugzilla.kernel.org. +If you want to be advised of the future bug reports, you can subscribe to the +bugme-new mailing list (only new bug reports are mailed here) or to the +bugme-janitor mailing list (every change in the bugzilla is mailed here) + + http://lists.osdl.org/mailman/listinfo/bugme-new + http://lists.osdl.org/mailman/listinfo/bugme-janitors + + + Mailing lists ------------- -- cgit v0.10.2 From e42344514c6e8ca7f5427da9b1407b56550dfa01 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Wed, 20 Sep 2006 16:38:00 +0900 Subject: sysfs: remove duplicated dput in sysfs_update_file Following function can drops d_count twice against one reference by lookup_one_len. /** * sysfs_update_file - update the modified timestamp on an object attribute. * @kobj: object we're acting for. * @attr: attribute descriptor. */ int sysfs_update_file(struct kobject * kobj, const struct attribute * attr) { struct dentry * dir = kobj->dentry; struct dentry * victim; int res = -ENOENT; mutex_lock(&dir->d_inode->i_mutex); victim = lookup_one_len(attr->name, dir, strlen(attr->name)); if (!IS_ERR(victim)) { /* make sure dentry is really there */ if (victim->d_inode && (victim->d_parent->d_inode == dir->d_inode)) { victim->d_inode->i_mtime = CURRENT_TIME; fsnotify_modify(victim); /** * Drop reference from initial sysfs_get_dentry(). */ dput(victim); res = 0; } else d_drop(victim); /** * Drop the reference acquired from sysfs_get_dentry() above. */ dput(victim); } mutex_unlock(&dir->d_inode->i_mutex); return res; } PCI-hotplug (drivers/pci/hotplug/pci_hotplug_core.c) is only user of this function. I confirmed that dentry of /sys/bus/pci/slots/XXX/* have negative d_count value. This patch removes unnecessary dput(). Signed-off-by: Hidetoshi Seto Acked-by: Maneesh Soni Signed-off-by: Greg Kroah-Hartman diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 146f1de..93218cc 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -483,11 +483,6 @@ int sysfs_update_file(struct kobject * kobj, const struct attribute * attr) (victim->d_parent->d_inode == dir->d_inode)) { victim->d_inode->i_mtime = CURRENT_TIME; fsnotify_modify(victim); - - /** - * Drop reference from initial sysfs_get_dentry(). - */ - dput(victim); res = 0; } else d_drop(victim); -- cgit v0.10.2 From 97a501849d60f3dbb8bfcd2300cf65dd5ebc0355 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Wed, 20 Sep 2006 16:49:02 +0900 Subject: sysfs: update obsolete comment in sysfs_update_file And the obsolete comment should be updated (or totally removed). Signed-off-by: Hidetoshi Seto Signed-off-by: Greg Kroah-Hartman diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 93218cc..298303b 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -488,7 +488,7 @@ int sysfs_update_file(struct kobject * kobj, const struct attribute * attr) d_drop(victim); /** - * Drop the reference acquired from sysfs_get_dentry() above. + * Drop the reference acquired from lookup_one_len() above. */ dput(victim); } -- cgit v0.10.2 From f0e1761ac528e9d28f3ba06bd268ec41fe872ac8 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 22 Sep 2006 11:37:00 +0200 Subject: driver core fixes: sysfs_create_link() retval check in class.c Check for return value of sysfs_create_link() in class_device_add(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/class.c b/drivers/base/class.c index b32b77f..0ff267a 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -562,7 +562,10 @@ int class_device_add(struct class_device *class_dev) goto out2; /* add the needed attributes to this device */ - sysfs_create_link(&class_dev->kobj, &parent_class->subsys.kset.kobj, "subsystem"); + error = sysfs_create_link(&class_dev->kobj, + &parent_class->subsys.kset.kobj, "subsystem"); + if (error) + goto out3; class_dev->uevent_attr.attr.name = "uevent"; class_dev->uevent_attr.attr.mode = S_IWUSR; class_dev->uevent_attr.attr.owner = parent_class->owner; -- cgit v0.10.2 From 1bb6881acae1c4f11a6e86f04df32ba45e95031d Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 22 Sep 2006 11:37:04 +0200 Subject: driver core fixes: bus_add_attrs() retval check Check return value of bus_add_attrs() in bus_register(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 12173d1..b90f6e6 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -732,11 +732,15 @@ int bus_register(struct bus_type * bus) klist_init(&bus->klist_devices, klist_devices_get, klist_devices_put); klist_init(&bus->klist_drivers, NULL, NULL); - bus_add_attrs(bus); + retval = bus_add_attrs(bus); + if (retval) + goto bus_attrs_fail; pr_debug("bus type '%s' registered\n", bus->name); return 0; +bus_attrs_fail: + kset_unregister(&bus->drivers); bus_drivers_fail: kset_unregister(&bus->devices); bus_devices_fail: -- cgit v0.10.2 From 513e7337adc32cdfbffecb99953e45a44e812c2d Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 22 Sep 2006 11:37:08 +0200 Subject: driver core fixes: bus_add_device() cleanup on error Correct cleanup in the error path of bus_add_device(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/bus.c b/drivers/base/bus.c index b90f6e6..d516f7d 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -372,19 +372,30 @@ int bus_add_device(struct device * dev) pr_debug("bus %s: add device %s\n", bus->name, dev->bus_id); error = device_add_attrs(bus, dev); if (error) - goto out; + goto out_put; error = sysfs_create_link(&bus->devices.kobj, &dev->kobj, dev->bus_id); if (error) - goto out; + goto out_id; error = sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "subsystem"); if (error) - goto out; + goto out_subsys; error = sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "bus"); + if (error) + goto out_deprecated; } -out: + return 0; + +out_deprecated: + sysfs_remove_link(&dev->kobj, "subsystem"); +out_subsys: + sysfs_remove_link(&bus->devices.kobj, dev->bus_id); +out_id: + device_remove_attrs(bus, dev); +out_put: + put_bus(dev->bus); return error; } -- cgit v0.10.2 From a306eea40952e6365301e8a2f7d5ffa9c6a1921b Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 22 Sep 2006 11:37:13 +0200 Subject: driver core fixes: device_add() cleanup on error Check for return code of device_create_file() and correct cleanup in the error case in device_add(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/core.c b/drivers/base/core.c index aee3743..365f709 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -433,14 +433,16 @@ int device_add(struct device *dev) if (dev->driver) dev->uevent_attr.attr.owner = dev->driver->owner; dev->uevent_attr.store = store_uevent; - device_create_file(dev, &dev->uevent_attr); + error = device_create_file(dev, &dev->uevent_attr); + if (error) + goto attrError; if (MAJOR(dev->devt)) { struct device_attribute *attr; attr = kzalloc(sizeof(*attr), GFP_KERNEL); if (!attr) { error = -ENOMEM; - goto PMError; + goto ueventattrError; } attr->attr.name = "dev"; attr->attr.mode = S_IRUGO; @@ -450,7 +452,7 @@ int device_add(struct device *dev) error = device_create_file(dev, attr); if (error) { kfree(attr); - goto attrError; + goto ueventattrError; } dev->devt_attr = attr; @@ -507,6 +509,8 @@ int device_add(struct device *dev) device_remove_file(dev, dev->devt_attr); kfree(dev->devt_attr); } + ueventattrError: + device_remove_file(dev, &dev->uevent_attr); attrError: kobject_uevent(&dev->kobj, KOBJ_REMOVE); kobject_del(&dev->kobj); -- cgit v0.10.2 From 141ecc5320147d363d060cfc8042d197a3cdd496 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 22 Sep 2006 11:37:27 +0200 Subject: driver core fixes: device_create_file() retval check in dmapool.c Check for device_create_file() return value in dma_pool_create(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/dmapool.c b/drivers/base/dmapool.c index 33c5cce..b2efbd4 100644 --- a/drivers/base/dmapool.c +++ b/drivers/base/dmapool.c @@ -141,11 +141,20 @@ dma_pool_create (const char *name, struct device *dev, init_waitqueue_head (&retval->waitq); if (dev) { + int ret; + down (&pools_lock); if (list_empty (&dev->dma_pools)) - device_create_file (dev, &dev_attr_pools); + ret = device_create_file (dev, &dev_attr_pools); + else + ret = 0; /* note: not currently insisting "name" be unique */ - list_add (&retval->pools, &dev->dma_pools); + if (!ret) + list_add (&retval->pools, &dev->dma_pools); + else { + kfree(retval); + retval = NULL; + } up (&pools_lock); } else INIT_LIST_HEAD (&retval->pools); -- cgit v0.10.2 From 221c324a336770a911b16bda02b9f4adad506a35 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 22 Sep 2006 11:37:32 +0200 Subject: driver core fixes: sysfs_create_group() retval in topology.c Return the return value of sysfs_create_group() in topology_add_dev(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 3ef9d51..28dccb7 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -97,8 +97,7 @@ static struct attribute_group topology_attr_group = { /* Add/Remove cpu_topology interface for CPU device */ static int __cpuinit topology_add_dev(struct sys_device * sys_dev) { - sysfs_create_group(&sys_dev->kobj, &topology_attr_group); - return 0; + return sysfs_create_group(&sys_dev->kobj, &topology_attr_group); } static int __cpuinit topology_remove_dev(struct sys_device * sys_dev) -- cgit v0.10.2 From 952ab431cd36c7ab573f685af8679c3677cbdc29 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 28 Sep 2006 23:56:01 +0200 Subject: Driver core: Don't leak 'old_class_name' in drivers/base/core.c::device_rename() If kmalloc() fails to allocate space for 'old_symlink_name' in drivers/base/core.c::device_rename(), then we'll leak 'old_class_name'. Spotted by the Coverity checker. Signed-off-by: Jesper Juhl Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/core.c b/drivers/base/core.c index 365f709..41f3dca 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -809,8 +809,10 @@ int device_rename(struct device *dev, char *new_name) if (dev->class) { old_symlink_name = kmalloc(BUS_ID_SIZE, GFP_KERNEL); - if (!old_symlink_name) - return -ENOMEM; + if (!old_symlink_name) { + error = -ENOMEM; + goto out_free_old_class; + } strlcpy(old_symlink_name, dev->bus_id, BUS_ID_SIZE); } @@ -834,9 +836,10 @@ int device_rename(struct device *dev, char *new_name) } put_device(dev); - kfree(old_class_name); kfree(new_class_name); kfree(old_symlink_name); + out_free_old_class: + kfree(old_class_name); return error; } -- cgit v0.10.2 From f70fa6296c2ec8f541f0a9b406ccc2d9d127d639 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 5 Oct 2006 17:03:24 -0400 Subject: Driver core: Don't ignore error returns from probing This patch (as797) fixes device_add() in the driver core. It needs to pay attention when the driver for a new device reports an error. At the same time, since bus_remove_device() undoes the effects of both bus_add_device() and bus_attach_device(), it needs to check whether the bus_attach_device step failed. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/bus.c b/drivers/base/bus.c index d516f7d..d7c5ea2 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -439,8 +439,10 @@ void bus_remove_device(struct device * dev) sysfs_remove_link(&dev->kobj, "bus"); sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id); device_remove_attrs(dev->bus, dev); - dev->is_registered = 0; - klist_del(&dev->knode_bus); + if (dev->is_registered) { + dev->is_registered = 0; + klist_del(&dev->knode_bus); + } pr_debug("bus %s: remove device %s\n", dev->bus->name, dev->bus_id); device_release_driver(dev); put_bus(dev->bus); diff --git a/drivers/base/core.c b/drivers/base/core.c index 41f3dca..68ad11a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -479,7 +479,8 @@ int device_add(struct device *dev) if ((error = bus_add_device(dev))) goto BusError; kobject_uevent(&dev->kobj, KOBJ_ADD); - bus_attach_device(dev); + if ((error = bus_attach_device(dev))) + goto AttachError; if (parent) klist_add_tail(&dev->knode_parent, &parent->klist_children); @@ -498,6 +499,8 @@ int device_add(struct device *dev) kfree(class_name); put_device(dev); return error; + AttachError: + bus_remove_device(dev); BusError: device_pm_remove(dev); PMError: -- cgit v0.10.2 From d9fd4d3b317a231e47f31d64d66c8cc7765d458f Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 4 Oct 2006 07:48:03 -0400 Subject: Driver core: bus: remove indentation level Before potentially fixing up these functions, this cosmetic change reduces the indentation level to make the code easier to read and maintain. No functional changes at all. Signed-off-by: Jeff Garzik Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/bus.c b/drivers/base/bus.c index d7c5ea2..7d8a7ce 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -518,34 +518,36 @@ int bus_add_driver(struct device_driver *drv) struct bus_type * bus = get_bus(drv->bus); int error = 0; - if (bus) { - pr_debug("bus %s: add driver %s\n", bus->name, drv->name); - error = kobject_set_name(&drv->kobj, "%s", drv->name); - if (error) - goto out_put_bus; - drv->kobj.kset = &bus->drivers; - if ((error = kobject_register(&drv->kobj))) - goto out_put_bus; - - error = driver_attach(drv); - if (error) - goto out_unregister; - klist_add_tail(&drv->knode_bus, &bus->klist_drivers); - module_add_driver(drv->owner, drv); - - error = driver_add_attrs(bus, drv); - if (error) { - /* How the hell do we get out of this pickle? Give up */ - printk(KERN_ERR "%s: driver_add_attrs(%s) failed\n", - __FUNCTION__, drv->name); - } - error = add_bind_files(drv); - if (error) { - /* Ditto */ - printk(KERN_ERR "%s: add_bind_files(%s) failed\n", - __FUNCTION__, drv->name); - } + if (!bus) + return 0; + + pr_debug("bus %s: add driver %s\n", bus->name, drv->name); + error = kobject_set_name(&drv->kobj, "%s", drv->name); + if (error) + goto out_put_bus; + drv->kobj.kset = &bus->drivers; + if ((error = kobject_register(&drv->kobj))) + goto out_put_bus; + + error = driver_attach(drv); + if (error) + goto out_unregister; + klist_add_tail(&drv->knode_bus, &bus->klist_drivers); + module_add_driver(drv->owner, drv); + + error = driver_add_attrs(bus, drv); + if (error) { + /* How the hell do we get out of this pickle? Give up */ + printk(KERN_ERR "%s: driver_add_attrs(%s) failed\n", + __FUNCTION__, drv->name); + } + error = add_bind_files(drv); + if (error) { + /* Ditto */ + printk(KERN_ERR "%s: add_bind_files(%s) failed\n", + __FUNCTION__, drv->name); } + return error; out_unregister: kobject_unregister(&drv->kobj); @@ -565,16 +567,17 @@ out_put_bus: void bus_remove_driver(struct device_driver * drv) { - if (drv->bus) { - remove_bind_files(drv); - driver_remove_attrs(drv->bus, drv); - klist_remove(&drv->knode_bus); - pr_debug("bus %s: remove driver %s\n", drv->bus->name, drv->name); - driver_detach(drv); - module_remove_driver(drv); - kobject_unregister(&drv->kobj); - put_bus(drv->bus); - } + if (!drv->bus) + return; + + remove_bind_files(drv); + driver_remove_attrs(drv->bus, drv); + klist_remove(&drv->knode_bus); + pr_debug("bus %s: remove driver %s\n", drv->bus->name, drv->name); + driver_detach(drv); + module_remove_driver(drv); + kobject_unregister(&drv->kobj); + put_bus(drv->bus); } -- cgit v0.10.2 From 4d664238207a82c4018757e2d87cf2a780462dcd Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 9 Oct 2006 18:04:30 +0900 Subject: driver core: kmalloc() failure check in driver_probe_device driver_probe_device() is missing kmalloc() failure check. Signed-off-by: Akinobu Mita Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/base/dd.c b/drivers/base/dd.c index ef7db69..db01b95 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -171,6 +171,8 @@ int driver_probe_device(struct device_driver * drv, struct device * dev) drv->bus->name, dev->bus_id, drv->name); data = kmalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; data->drv = drv; data->dev = dev; -- cgit v0.10.2 From 463c2c12dce78dd0cb77b65beba93f029a164ba3 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:34:41 -0400 Subject: aoe: eliminate isbusy message This message doesn't help users because the circumstance isn't problematic. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index ed4258a..c2bc3ed 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -20,11 +20,8 @@ aoedev_isbusy(struct aoedev *d) f = d->frames; e = f + d->nframes; do { - if (f->tag != FREETAG) { - printk(KERN_DEBUG "aoe: %ld.%ld isbusy\n", - d->aoemajor, d->aoeminor); + if (f->tag != FREETAG) return 1; - } } while (++f < e); return 0; -- cgit v0.10.2 From 2611464d7f36685fb1990275d3de1e72e6aff9d9 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:48 -0400 Subject: aoe: update copyright date Update the copyright year to 2006. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 6eebcb7..507c377 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ #define VERSION "22" #define AOE_MAJOR 152 #define DEVICE_NAME "aoe" diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index 393b86a..fa0e8ca 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ /* * aoeblk.c * block device routines diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index 1bc1cf9..8a7a081 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ /* * aoechr.c * AoE character device driver diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 39da28d..d1d8759 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ /* * aoecmd.c * Filesystem request handling methods diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index c2bc3ed..c7e05ed8 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ /* * aoedev.c * AoE device utility functions; maintains device list. diff --git a/drivers/block/aoe/aoemain.c b/drivers/block/aoe/aoemain.c index de08491..727c34d 100644 --- a/drivers/block/aoe/aoemain.c +++ b/drivers/block/aoe/aoemain.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ /* * aoemain.c * Module initialization routines, discover timer diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index c1434ed..1bba140 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -1,4 +1,4 @@ -/* Copyright (c) 2004 Coraid, Inc. See COPYING for GPL terms. */ +/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ /* * aoenet.c * Ethernet portion of AoE driver -- cgit v0.10.2 From 2fdc0ea75b26e3009cfdf72e79901e4e16bb99bd Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:48 -0400 Subject: aoe: remove unused NARGS enum The NARGS enum is left over from older code versions. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index 8a7a081..0c543d3 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -15,7 +15,6 @@ enum { MINOR_INTERFACES, MINOR_REVALIDATE, MSGSZ = 2048, - NARGS = 10, NMSG = 100, /* message backlog to retain */ }; -- cgit v0.10.2 From e407a7f6cd143b3ab4eb3d7e1cf882e96b710eb5 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:49 -0400 Subject: aoe: zero copy write 1 of 2 Avoid memory copy on writes. (This patch depends on fixes in patch 9 to follow.) Although skb->len should not be set when working with linear skbuffs, the skb->tail pointer maintained by skb_put/skb_trim is not relevant to what happens when the skb_fill_page_desc function is called. This issue was raised without comment in linux-kernel and netdev earlier this month: http://thread.gmane.org/gmane.linux.kernel/446474/ http://thread.gmane.org/gmane.linux.network/45444/ So until there is something analogous to skb_put that works for zero-copy write skbuffs, we will do what the other callers of skb_fill_page_desc are doing. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 507c377..fa2d804 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -107,11 +107,7 @@ struct frame { ulong waited; struct buf *buf; char *bufaddr; - int writedatalen; - int ndata; - - /* largest possible */ - unsigned char data[sizeof(struct aoe_hdr) + sizeof(struct aoe_atahdr)]; + struct sk_buff *skb; }; struct aoedev { @@ -157,6 +153,7 @@ void aoecmd_cfg(ushort aoemajor, unsigned char aoeminor); void aoecmd_ata_rsp(struct sk_buff *); void aoecmd_cfg_rsp(struct sk_buff *); void aoecmd_sleepwork(void *vp); +struct sk_buff *new_skb(ulong); int aoedev_init(void); void aoedev_exit(void); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index d1d8759..1aeb296 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -17,15 +17,14 @@ #define MAXTIMER (HZ << 1) #define MAXWAIT (60 * 3) /* After MAXWAIT seconds, give up and fail dev */ -static struct sk_buff * -new_skb(struct net_device *if_dev, ulong len) +struct sk_buff * +new_skb(ulong len) { struct sk_buff *skb; skb = alloc_skb(len, GFP_ATOMIC); if (skb) { skb->nh.raw = skb->mac.raw = skb->data; - skb->dev = if_dev; skb->protocol = __constant_htons(ETH_P_AOE); skb->priority = 0; skb_put(skb, len); @@ -40,29 +39,6 @@ new_skb(struct net_device *if_dev, ulong len) return skb; } -static struct sk_buff * -skb_prepare(struct aoedev *d, struct frame *f) -{ - struct sk_buff *skb; - char *p; - - skb = new_skb(d->ifp, f->ndata + f->writedatalen); - if (!skb) { - printk(KERN_INFO "aoe: skb_prepare: failure to allocate skb\n"); - return NULL; - } - - p = skb->mac.raw; - memcpy(p, f->data, f->ndata); - - if (f->writedatalen) { - p += sizeof(struct aoe_hdr) + sizeof(struct aoe_atahdr); - memcpy(p, f->bufaddr, f->writedatalen); - } - - return skb; -} - static struct frame * getframe(struct aoedev *d, int tag) { @@ -129,10 +105,11 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) bcnt = MAXATADATA; /* initialize the headers & frame */ - h = (struct aoe_hdr *) f->data; + skb = f->skb; + h = (struct aoe_hdr *) skb->mac.raw; ah = (struct aoe_atahdr *) (h+1); - f->ndata = sizeof *h + sizeof *ah; - memset(h, 0, f->ndata); + skb->len = sizeof *h + sizeof *ah; + memset(h, 0, skb->len); f->tag = aoehdr_atainit(d, h); f->waited = 0; f->buf = buf; @@ -155,11 +132,13 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) } if (bio_data_dir(buf->bio) == WRITE) { + skb_fill_page_desc(skb, 0, virt_to_page(f->bufaddr), + offset_in_page(f->bufaddr), bcnt); ah->aflags |= AOEAFL_WRITE; - f->writedatalen = bcnt; } else { + skb_shinfo(skb)->nr_frags = 0; + skb->len = ETH_ZLEN; writebit = 0; - f->writedatalen = 0; } ah->cmdstat = WIN_READ | writebit | extbit; @@ -179,15 +158,14 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) buf->bufaddr = page_address(buf->bv->bv_page) + buf->bv->bv_offset; } - skb = skb_prepare(d, f); - if (skb) { - skb->next = NULL; - if (d->sendq_hd) - d->sendq_tl->next = skb; - else - d->sendq_hd = skb; - d->sendq_tl = skb; - } + skb->dev = d->ifp; + skb_get(skb); + skb->next = NULL; + if (d->sendq_hd) + d->sendq_tl->next = skb; + else + d->sendq_hd = skb; + d->sendq_tl = skb; } /* some callers cannot sleep, and they can call this function, @@ -209,11 +187,12 @@ aoecmd_cfg_pkts(ushort aoemajor, unsigned char aoeminor, struct sk_buff **tail) if (!is_aoe_netif(ifp)) continue; - skb = new_skb(ifp, sizeof *h + sizeof *ch); + skb = new_skb(sizeof *h + sizeof *ch); if (skb == NULL) { printk(KERN_INFO "aoe: aoecmd_cfg: skb alloc failure\n"); continue; } + skb->dev = ifp; if (sl_tail == NULL) sl_tail = skb; h = (struct aoe_hdr *) skb->mac.raw; @@ -283,21 +262,21 @@ rexmit(struct aoedev *d, struct frame *f) d->aoemajor, d->aoeminor, f->tag, jiffies, n); aoechr_error(buf); - h = (struct aoe_hdr *) f->data; + skb = f->skb; + h = (struct aoe_hdr *) skb->mac.raw; f->tag = n; h->tag = cpu_to_be32(n); memcpy(h->dst, d->addr, sizeof h->dst); memcpy(h->src, d->ifp->dev_addr, sizeof h->src); - skb = skb_prepare(d, f); - if (skb) { - skb->next = NULL; - if (d->sendq_hd) - d->sendq_tl->next = skb; - else - d->sendq_hd = skb; - d->sendq_tl = skb; - } + skb->dev = d->ifp; + skb_get(skb); + skb->next = NULL; + if (d->sendq_hd) + d->sendq_tl->next = skb; + else + d->sendq_hd = skb; + d->sendq_tl = skb; } static int @@ -514,7 +493,7 @@ aoecmd_ata_rsp(struct sk_buff *skb) calc_rttavg(d, tsince(f->tag)); ahin = (struct aoe_atahdr *) (hin+1); - ahout = (struct aoe_atahdr *) (f->data + sizeof(struct aoe_hdr)); + ahout = (struct aoe_atahdr *) (f->skb->mac.raw + sizeof(struct aoe_hdr)); buf = f->buf; if (ahout->cmdstat == WIN_IDENTIFY) @@ -620,20 +599,21 @@ aoecmd_ata_id(struct aoedev *d) } /* initialize the headers & frame */ - h = (struct aoe_hdr *) f->data; + skb = f->skb; + h = (struct aoe_hdr *) skb->mac.raw; ah = (struct aoe_atahdr *) (h+1); - f->ndata = sizeof *h + sizeof *ah; - memset(h, 0, f->ndata); + skb->len = sizeof *h + sizeof *ah; + memset(h, 0, skb->len); f->tag = aoehdr_atainit(d, h); f->waited = 0; - f->writedatalen = 0; /* set up ata header */ ah->scnt = 1; ah->cmdstat = WIN_IDENTIFY; ah->lba3 = 0xa0; - skb = skb_prepare(d, f); + skb->dev = d->ifp; + skb_get(skb); d->rttavg = MAXTIMER; d->timer.function = rexmit_timer; diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index c7e05ed8..abf1d3c 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -63,22 +63,32 @@ aoedev_newdev(ulong nframes) struct frame *f, *e; d = kzalloc(sizeof *d, GFP_ATOMIC); - if (d == NULL) - return NULL; f = kcalloc(nframes, sizeof *f, GFP_ATOMIC); - if (f == NULL) { - kfree(d); + switch (!d || !f) { + case 0: + d->nframes = nframes; + d->frames = f; + e = f + nframes; + for (; ftag = FREETAG; + f->skb = new_skb(ETH_ZLEN); + if (!f->skb) + break; + } + if (f == e) + break; + while (f > d->frames) { + f--; + dev_kfree_skb(f->skb); + } + default: + if (f) + kfree(f); + if (d) + kfree(d); return NULL; } - INIT_WORK(&d->work, aoecmd_sleepwork, d); - - d->nframes = nframes; - d->frames = f; - e = f + nframes; - for (; ftag = FREETAG; - spin_lock_init(&d->lock); init_timer(&d->timer); d->timer.data = (ulong) d; @@ -160,11 +170,19 @@ aoedev_by_sysminor_m(ulong sysminor, ulong bufcnt) static void aoedev_freedev(struct aoedev *d) { + struct frame *f, *e; + if (d->gd) { aoedisk_rm_sysfs(d); del_gendisk(d->gd); put_disk(d->gd); } + f = d->frames; + e = f + d->nframes; + for (; fskb)->nr_frags = 0; + dev_kfree_skb(f->skb); + } kfree(d->frames); if (d->bufpool) mempool_destroy(d->bufpool); -- cgit v0.10.2 From 19bf26353c50bc2be375109ec73f2f0bbd616ed1 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:49 -0400 Subject: aoe: jumbo frame support 1 of 2 Add support for jumbo ethernet frames. (This patch depends on patch 7 to follow.) Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index fa2d804..1cec199 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -65,7 +65,7 @@ struct aoe_atahdr { struct aoe_cfghdr { __be16 bufcnt; __be16 fwver; - unsigned char res; + unsigned char scnt; unsigned char aoeccmd; unsigned char cslen[2]; }; @@ -78,12 +78,13 @@ enum { DEVFL_GDALLOC = (1<<4), /* need to alloc gendisk */ DEVFL_PAUSE = (1<<5), DEVFL_NEWSIZE = (1<<6), /* need to update dev size in block layer */ + DEVFL_MAXBCNT = (1<<7), /* d->maxbcnt is not changeable */ BUFFL_FAIL = 1, }; enum { - MAXATADATA = 1024, + DEFAULTBCNT = 2 * 512, /* 2 sectors */ NPERSHELF = 16, /* number of slots per shelf address */ FREETAG = -1, MIN_BUFS = 8, @@ -107,6 +108,8 @@ struct frame { ulong waited; struct buf *buf; char *bufaddr; + ulong bcnt; + sector_t lba; struct sk_buff *skb; }; @@ -120,6 +123,7 @@ struct aoedev { ulong nopen; /* (bd_openers isn't available without sleeping) */ ulong rttavg; /* round trip average of requests/responses */ u16 fw_ver; /* version of blade's firmware */ + u16 maxbcnt; struct work_struct work;/* disk create work struct */ struct gendisk *gd; request_queue_t blkq; @@ -134,7 +138,8 @@ struct aoedev { struct list_head bufq; /* queue of bios to work on */ struct buf *inprocess; /* the one we're currently working on */ ulong lasttag; /* last tag sent */ - ulong nframes; /* number of frames below */ + ushort lostjumbo; + ushort nframes; /* number of frames below */ struct frame *frames; }; diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index 0c543d3..2b5256c 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -89,6 +89,7 @@ revalidate(const char __user *str, size_t size) return -EINVAL; spin_lock_irqsave(&d->lock, flags); + d->flags &= ~DEVFL_MAXBCNT; d->flags |= DEVFL_PAUSE; spin_unlock_irqrestore(&d->lock, flags); aoecmd_cfg(major, minor); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 1aeb296..666797d 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -83,6 +83,17 @@ aoehdr_atainit(struct aoedev *d, struct aoe_hdr *h) return host_tag; } +static inline void +put_lba(struct aoe_atahdr *ah, sector_t lba) +{ + ah->lba0 = lba; + ah->lba1 = lba >>= 8; + ah->lba2 = lba >>= 8; + ah->lba3 = lba >>= 8; + ah->lba4 = lba >>= 8; + ah->lba5 = lba >>= 8; +} + static void aoecmd_ata_rw(struct aoedev *d, struct frame *f) { @@ -101,8 +112,8 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) sector = buf->sector; bcnt = buf->bv_resid; - if (bcnt > MAXATADATA) - bcnt = MAXATADATA; + if (bcnt > d->maxbcnt) + bcnt = d->maxbcnt; /* initialize the headers & frame */ skb = f->skb; @@ -114,17 +125,14 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) f->waited = 0; f->buf = buf; f->bufaddr = buf->bufaddr; + f->bcnt = bcnt; + f->lba = sector; /* set up ata header */ ah->scnt = bcnt >> 9; - ah->lba0 = sector; - ah->lba1 = sector >>= 8; - ah->lba2 = sector >>= 8; - ah->lba3 = sector >>= 8; + put_lba(ah, sector); if (d->flags & DEVFL_EXT) { ah->aflags |= AOEAFL_EXT; - ah->lba4 = sector >>= 8; - ah->lba5 = sector >>= 8; } else { extbit = 0; ah->lba3 &= 0x0f; @@ -251,6 +259,7 @@ rexmit(struct aoedev *d, struct frame *f) { struct sk_buff *skb; struct aoe_hdr *h; + struct aoe_atahdr *ah; char buf[128]; u32 n; @@ -264,11 +273,27 @@ rexmit(struct aoedev *d, struct frame *f) skb = f->skb; h = (struct aoe_hdr *) skb->mac.raw; + ah = (struct aoe_atahdr *) (h+1); f->tag = n; h->tag = cpu_to_be32(n); memcpy(h->dst, d->addr, sizeof h->dst); memcpy(h->src, d->ifp->dev_addr, sizeof h->src); + n = DEFAULTBCNT / 512; + if (ah->scnt > n) { + ah->scnt = n; + if (ah->aflags & AOEAFL_WRITE) + skb_fill_page_desc(skb, 0, virt_to_page(f->bufaddr), + offset_in_page(f->bufaddr), DEFAULTBCNT); + if (++d->lostjumbo > (d->nframes << 1)) + if (d->maxbcnt != DEFAULTBCNT) { + printk(KERN_INFO "aoe: rexmit: too many lost jumbo. " + "dropping back to 1KB frames.\n"); + d->maxbcnt = DEFAULTBCNT; + d->flags |= DEVFL_MAXBCNT; + } + } + skb->dev = d->ifp; skb_get(skb); skb->next = NULL; @@ -506,10 +531,10 @@ aoecmd_ata_rsp(struct sk_buff *skb) if (buf) buf->flags |= BUFFL_FAIL; } else { + n = ahout->scnt << 9; switch (ahout->cmdstat) { case WIN_READ: case WIN_READ_EXT: - n = ahout->scnt << 9; if (skb->len - sizeof *hin - sizeof *ahin < n) { printk(KERN_CRIT "aoe: aoecmd_ata_rsp: runt " "ata data size in read. skb->len=%d\n", @@ -521,6 +546,22 @@ aoecmd_ata_rsp(struct sk_buff *skb) memcpy(f->bufaddr, ahin+1, n); case WIN_WRITE: case WIN_WRITE_EXT: + if (f->bcnt -= n) { + f->bufaddr += n; + put_lba(ahout, f->lba += ahout->scnt); + n = f->bcnt > DEFAULTBCNT ? DEFAULTBCNT : f->bcnt; + ahout->scnt = n >> 9; + if (ahout->aflags & AOEAFL_WRITE) + skb_fill_page_desc(f->skb, 0, virt_to_page(f->bufaddr), + offset_in_page(f->bufaddr), n); + skb_get(f->skb); + f->skb->next = NULL; + spin_unlock_irqrestore(&d->lock, flags); + aoenet_xmit(f->skb); + return; + } + if (n > DEFAULTBCNT) + d->lostjumbo = 0; break; case WIN_IDENTIFY: if (skb->len - sizeof *hin - sizeof *ahin < 512) { @@ -628,9 +669,9 @@ aoecmd_cfg_rsp(struct sk_buff *skb) struct aoe_hdr *h; struct aoe_cfghdr *ch; ulong flags, sysminor, aoemajor; - u16 bufcnt; struct sk_buff *sl; enum { MAXFRAMES = 16 }; + u16 n; h = (struct aoe_hdr *) skb->mac.raw; ch = (struct aoe_cfghdr *) (h+1); @@ -654,11 +695,11 @@ aoecmd_cfg_rsp(struct sk_buff *skb) return; } - bufcnt = be16_to_cpu(ch->bufcnt); - if (bufcnt > MAXFRAMES) /* keep it reasonable */ - bufcnt = MAXFRAMES; + n = be16_to_cpu(ch->bufcnt); + if (n > MAXFRAMES) /* keep it reasonable */ + n = MAXFRAMES; - d = aoedev_by_sysminor_m(sysminor, bufcnt); + d = aoedev_by_sysminor_m(sysminor, n); if (d == NULL) { printk(KERN_INFO "aoe: aoecmd_cfg_rsp: device sysminor_m failure\n"); return; @@ -669,6 +710,14 @@ aoecmd_cfg_rsp(struct sk_buff *skb) /* permit device to migrate mac and network interface */ d->ifp = skb->dev; memcpy(d->addr, h->src, sizeof d->addr); + if (!(d->flags & DEVFL_MAXBCNT)) { + n = d->ifp->mtu; + n -= sizeof (struct aoe_hdr) + sizeof (struct aoe_atahdr); + n /= 512; + if (n > ch->scnt) + n = ch->scnt; + d->maxbcnt = n ? n * 512 : DEFAULTBCNT; + } /* don't change users' perspective */ if (d->nopen && !(d->flags & DEVFL_PAUSE)) { -- cgit v0.10.2 From 6bb6285fdb948cedee586c6bebc9ebc5e32a5c35 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:49 -0400 Subject: aoe: clean up printks via macros Use simple macros to clean up the printks. (This patch is reverted by the 14th patch to follow.) Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 1cec199..4d79f1e 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -10,6 +10,11 @@ #define AOE_PARTITIONS (16) #endif +#define xprintk(L, fmt, arg...) printk(L "aoe: " "%s: " fmt, __func__, ## arg) +#define iprintk(fmt, arg...) xprintk(KERN_INFO, fmt, ## arg) +#define eprintk(fmt, arg...) xprintk(KERN_ERR, fmt, ## arg) +#define dprintk(fmt, arg...) xprintk(KERN_DEBUG, fmt, ## arg) + #define SYSMINOR(aoemajor, aoeminor) ((aoemajor) * NPERSHELF + (aoeminor)) #define AOEMAJOR(sysminor) ((sysminor) / NPERSHELF) #define AOEMINOR(sysminor) ((sysminor) % NPERSHELF) diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index fa0e8ca..a7dbe6f 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -132,8 +132,7 @@ aoeblk_make_request(request_queue_t *q, struct bio *bio) d = bio->bi_bdev->bd_disk->private_data; buf = mempool_alloc(d->bufpool, GFP_NOIO); if (buf == NULL) { - printk(KERN_INFO "aoe: aoeblk_make_request: buf allocation " - "failure\n"); + iprintk("buf allocation failure\n"); bio_endio(bio, bio->bi_size, -ENOMEM); return 0; } @@ -150,8 +149,7 @@ aoeblk_make_request(request_queue_t *q, struct bio *bio) spin_lock_irqsave(&d->lock, flags); if ((d->flags & DEVFL_UP) == 0) { - printk(KERN_INFO "aoe: aoeblk_make_request: device %ld.%ld is not up\n", - d->aoemajor, d->aoeminor); + iprintk("device %ld.%ld is not up\n", d->aoemajor, d->aoeminor); spin_unlock_irqrestore(&d->lock, flags); mempool_free(buf, d->bufpool); bio_endio(bio, bio->bi_size, -ENXIO); @@ -176,7 +174,7 @@ aoeblk_getgeo(struct block_device *bdev, struct hd_geometry *geo) struct aoedev *d = bdev->bd_disk->private_data; if ((d->flags & DEVFL_UP) == 0) { - printk(KERN_ERR "aoe: aoeblk_ioctl: disk not up\n"); + eprintk("disk not up\n"); return -ENODEV; } @@ -203,8 +201,8 @@ aoeblk_gdalloc(void *vp) gd = alloc_disk(AOE_PARTITIONS); if (gd == NULL) { - printk(KERN_ERR "aoe: aoeblk_gdalloc: cannot allocate disk " - "structure for %ld.%ld\n", d->aoemajor, d->aoeminor); + eprintk("cannot allocate disk structure for %ld.%ld\n", + d->aoemajor, d->aoeminor); spin_lock_irqsave(&d->lock, flags); d->flags &= ~DEVFL_GDALLOC; spin_unlock_irqrestore(&d->lock, flags); @@ -213,8 +211,8 @@ aoeblk_gdalloc(void *vp) d->bufpool = mempool_create_slab_pool(MIN_BUFS, buf_pool_cache); if (d->bufpool == NULL) { - printk(KERN_ERR "aoe: aoeblk_gdalloc: cannot allocate bufpool " - "for %ld.%ld\n", d->aoemajor, d->aoeminor); + eprintk("cannot allocate bufpool for %ld.%ld\n", + d->aoemajor, d->aoeminor); put_disk(gd); spin_lock_irqsave(&d->lock, flags); d->flags &= ~DEVFL_GDALLOC; diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index 2b5256c..f5cab69 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -55,9 +55,7 @@ static int interfaces(const char __user *str, size_t size) { if (set_aoe_iflist(str, size)) { - printk(KERN_CRIT - "%s: could not set interface list: %s\n", - __FUNCTION__, "too many interfaces"); + eprintk("could not set interface list: too many interfaces\n"); return -EINVAL; } return 0; @@ -80,8 +78,7 @@ revalidate(const char __user *str, size_t size) /* should be e%d.%d format */ n = sscanf(buf, "e%d.%d", &major, &minor); if (n != 2) { - printk(KERN_ERR "aoe: %s: invalid device specification\n", - __FUNCTION__); + eprintk("invalid device specification\n"); return -EINVAL; } d = aoedev_by_aoeaddr(major, minor); @@ -116,7 +113,7 @@ bail: spin_unlock_irqrestore(&emsgs_lock, flags); mp = kmalloc(n, GFP_ATOMIC); if (mp == NULL) { - printk(KERN_CRIT "aoe: aoechr_error: allocation failure, len=%ld\n", n); + eprintk("allocation failure, len=%ld\n", n); goto bail; } @@ -141,7 +138,7 @@ aoechr_write(struct file *filp, const char __user *buf, size_t cnt, loff_t *offp switch ((unsigned long) filp->private_data) { default: - printk(KERN_INFO "aoe: aoechr_write: can't write to that file.\n"); + iprintk("can't write to that file.\n"); break; case MINOR_DISCOVER: ret = discover(); @@ -250,7 +247,7 @@ aoechr_init(void) n = register_chrdev(AOE_MAJOR, "aoechr", &aoe_fops); if (n < 0) { - printk(KERN_ERR "aoe: aoechr_init: can't register char device\n"); + eprintk("can't register char device\n"); return n; } sema_init(&emsgs_sema, 0); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 666797d..63c4560 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -155,7 +155,7 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) buf->nframesout += 1; buf->bufaddr += bcnt; buf->bv_resid -= bcnt; -/* printk(KERN_INFO "aoe: bv_resid=%ld\n", buf->bv_resid); */ +/* dprintk("bv_resid=%ld\n", buf->bv_resid); */ buf->resid -= bcnt; buf->sector += bcnt >> 9; if (buf->resid == 0) { @@ -197,7 +197,7 @@ aoecmd_cfg_pkts(ushort aoemajor, unsigned char aoeminor, struct sk_buff **tail) skb = new_skb(sizeof *h + sizeof *ch); if (skb == NULL) { - printk(KERN_INFO "aoe: aoecmd_cfg: skb alloc failure\n"); + iprintk("skb alloc failure\n"); continue; } skb->dev = ifp; @@ -247,7 +247,7 @@ loop: return; buf = container_of(d->bufq.next, struct buf, bufs); list_del(d->bufq.next); -/*printk(KERN_INFO "aoecmd_work: bi_size=%ld\n", buf->bio->bi_size); */ +/*dprintk("bi_size=%ld\n", buf->bio->bi_size); */ d->inprocess = buf; } aoecmd_ata_rw(d, f); @@ -287,8 +287,7 @@ rexmit(struct aoedev *d, struct frame *f) offset_in_page(f->bufaddr), DEFAULTBCNT); if (++d->lostjumbo > (d->nframes << 1)) if (d->maxbcnt != DEFAULTBCNT) { - printk(KERN_INFO "aoe: rexmit: too many lost jumbo. " - "dropping back to 1KB frames.\n"); + iprintk("too many lost jumbo - using 1KB frames.\n"); d->maxbcnt = DEFAULTBCNT; d->flags |= DEVFL_MAXBCNT; } @@ -435,8 +434,8 @@ ataid_complete(struct aoedev *d, unsigned char *id) } if (d->ssize != ssize) - printk(KERN_INFO "aoe: %012llx e%lu.%lu v%04x has %llu " - "sectors\n", (unsigned long long)mac_addr(d->addr), + iprintk("%012llx e%lu.%lu v%04x has %llu sectors\n", + (unsigned long long)mac_addr(d->addr), d->aoemajor, d->aoeminor, d->fw_ver, (long long)ssize); d->ssize = ssize; @@ -446,11 +445,9 @@ ataid_complete(struct aoedev *d, unsigned char *id) d->flags |= DEVFL_NEWSIZE; } else { if (d->flags & DEVFL_GDALLOC) { - printk(KERN_INFO "aoe: %s: %s e%lu.%lu, %s\n", - __FUNCTION__, - "can't schedule work for", + eprintk("can't schedule work for e%lu.%lu, %s\n", d->aoemajor, d->aoeminor, - "it's already on! (This really shouldn't happen).\n"); + "it's already on! This shouldn't happen.\n"); return; } d->flags |= DEVFL_GDALLOC; @@ -524,8 +521,7 @@ aoecmd_ata_rsp(struct sk_buff *skb) if (ahout->cmdstat == WIN_IDENTIFY) d->flags &= ~DEVFL_PAUSE; if (ahin->cmdstat & 0xa9) { /* these bits cleared on success */ - printk(KERN_CRIT "aoe: aoecmd_ata_rsp: ata error cmd=%2.2Xh " - "stat=%2.2Xh from e%ld.%ld\n", + eprintk("ata error cmd=%2.2Xh stat=%2.2Xh from e%ld.%ld\n", ahout->cmdstat, ahin->cmdstat, d->aoemajor, d->aoeminor); if (buf) @@ -536,8 +532,7 @@ aoecmd_ata_rsp(struct sk_buff *skb) case WIN_READ: case WIN_READ_EXT: if (skb->len - sizeof *hin - sizeof *ahin < n) { - printk(KERN_CRIT "aoe: aoecmd_ata_rsp: runt " - "ata data size in read. skb->len=%d\n", + eprintk("runt data size in read. skb->len=%d\n", skb->len); /* fail frame f? just returning will rexmit. */ spin_unlock_irqrestore(&d->lock, flags); @@ -549,10 +544,13 @@ aoecmd_ata_rsp(struct sk_buff *skb) if (f->bcnt -= n) { f->bufaddr += n; put_lba(ahout, f->lba += ahout->scnt); - n = f->bcnt > DEFAULTBCNT ? DEFAULTBCNT : f->bcnt; + n = f->bcnt; + if (n > DEFAULTBCNT) + n = DEFAULTBCNT; ahout->scnt = n >> 9; if (ahout->aflags & AOEAFL_WRITE) - skb_fill_page_desc(f->skb, 0, virt_to_page(f->bufaddr), + skb_fill_page_desc(f->skb, 0, + virt_to_page(f->bufaddr), offset_in_page(f->bufaddr), n); skb_get(f->skb); f->skb->next = NULL; @@ -565,19 +563,18 @@ aoecmd_ata_rsp(struct sk_buff *skb) break; case WIN_IDENTIFY: if (skb->len - sizeof *hin - sizeof *ahin < 512) { - printk(KERN_INFO "aoe: aoecmd_ata_rsp: runt data size " - "in ataid. skb->len=%d\n", skb->len); + iprintk("runt data size in ataid. skb->len=%d\n", + skb->len); spin_unlock_irqrestore(&d->lock, flags); return; } ataid_complete(d, (char *) (ahin+1)); break; default: - printk(KERN_INFO "aoe: aoecmd_ata_rsp: unrecognized " - "outbound ata command %2.2Xh for %d.%d\n", - ahout->cmdstat, - be16_to_cpu(hin->major), - hin->minor); + iprintk("unrecognized ata command %2.2Xh for %d.%d\n", + ahout->cmdstat, + be16_to_cpu(hin->major), + hin->minor); } } @@ -634,8 +631,7 @@ aoecmd_ata_id(struct aoedev *d) f = getframe(d, FREETAG); if (f == NULL) { - printk(KERN_CRIT "aoe: aoecmd_ata_id: can't get a frame. " - "This shouldn't happen.\n"); + eprintk("can't get a frame. This shouldn't happen.\n"); return NULL; } @@ -682,15 +678,14 @@ aoecmd_cfg_rsp(struct sk_buff *skb) */ aoemajor = be16_to_cpu(h->major); if (aoemajor == 0xfff) { - printk(KERN_CRIT "aoe: aoecmd_cfg_rsp: Warning: shelf " - "address is all ones. Check shelf dip switches\n"); + eprintk("Warning: shelf address is all ones. " + "Check shelf dip switches.\n"); return; } sysminor = SYSMINOR(aoemajor, h->minor); if (sysminor * AOE_PARTITIONS + AOE_PARTITIONS > MINORMASK) { - printk(KERN_INFO - "aoe: e%ld.%d: minor number too large\n", + iprintk("e%ld.%d: minor number too large\n", aoemajor, (int) h->minor); return; } @@ -701,7 +696,7 @@ aoecmd_cfg_rsp(struct sk_buff *skb) d = aoedev_by_sysminor_m(sysminor, n); if (d == NULL) { - printk(KERN_INFO "aoe: aoecmd_cfg_rsp: device sysminor_m failure\n"); + iprintk("device sysminor_m failure\n"); return; } diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index abf1d3c..f51d87b 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -155,7 +155,7 @@ aoedev_by_sysminor_m(ulong sysminor, ulong bufcnt) d = aoedev_newdev(bufcnt); if (d == NULL) { spin_unlock_irqrestore(&devlist_lock, flags); - printk(KERN_INFO "aoe: aoedev_set: aoedev_newdev failure.\n"); + iprintk("aoedev_newdev failure.\n"); return NULL; } d->sysminor = sysminor; diff --git a/drivers/block/aoe/aoemain.c b/drivers/block/aoe/aoemain.c index 727c34d..13e634db 100644 --- a/drivers/block/aoe/aoemain.c +++ b/drivers/block/aoe/aoemain.c @@ -84,13 +84,11 @@ aoe_init(void) goto net_fail; ret = register_blkdev(AOE_MAJOR, DEVICE_NAME); if (ret < 0) { - printk(KERN_ERR "aoe: aoeblk_init: can't register major\n"); + eprintk("can't register major\n"); goto blkreg_fail; } - printk(KERN_INFO - "aoe: aoe_init: AoE v%s initialised.\n", - VERSION); + iprintk("AoE v%s initialised.\n", VERSION); discover_timer(TINIT); return 0; @@ -103,7 +101,7 @@ aoe_init(void) chr_fail: aoedev_exit(); - printk(KERN_INFO "aoe: aoe_init: initialisation failure.\n"); + iprintk("initialisation failure.\n"); return ret; } diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index 1bba140..f1cf266 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -74,7 +74,7 @@ set_aoe_iflist(const char __user *user_str, size_t size) return -EINVAL; if (copy_from_user(aoe_iflist, user_str, size)) { - printk(KERN_INFO "aoe: %s: copy from user failed\n", __FUNCTION__); + iprintk("copy from user failed\n"); return -EFAULT; } aoe_iflist[size] = 0x00; @@ -132,8 +132,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, if (n > NECODES) n = 0; if (net_ratelimit()) - printk(KERN_ERR "aoe: aoenet_rcv: error packet from %d.%d; " - "ecode=%d '%s'\n", + eprintk("error packet from %d.%d; ecode=%d '%s'\n", be16_to_cpu(h->major), h->minor, h->err, aoe_errlist[n]); goto exit; @@ -147,7 +146,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, aoecmd_cfg_rsp(skb); break; default: - printk(KERN_INFO "aoe: aoenet_rcv: unknown cmd %d\n", h->cmd); + iprintk("unknown cmd %d\n", h->cmd); } exit: dev_kfree_skb(skb); -- cgit v0.10.2 From ddec63e86752b89776547e93aa68af01f1cbb10c Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:49 -0400 Subject: aoe: jumbo frame support 2 of 2 Add support for jumbo ethernet frames. (This patch follows patch 5.) Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 63c4560..621fdbb 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -475,7 +475,7 @@ void aoecmd_ata_rsp(struct sk_buff *skb) { struct aoedev *d; - struct aoe_hdr *hin; + struct aoe_hdr *hin, *hout; struct aoe_atahdr *ahin, *ahout; struct frame *f; struct buf *buf; @@ -515,7 +515,8 @@ aoecmd_ata_rsp(struct sk_buff *skb) calc_rttavg(d, tsince(f->tag)); ahin = (struct aoe_atahdr *) (hin+1); - ahout = (struct aoe_atahdr *) (f->skb->mac.raw + sizeof(struct aoe_hdr)); + hout = (struct aoe_hdr *) f->skb->mac.raw; + ahout = (struct aoe_atahdr *) (hout+1); buf = f->buf; if (ahout->cmdstat == WIN_IDENTIFY) @@ -552,6 +553,9 @@ aoecmd_ata_rsp(struct sk_buff *skb) skb_fill_page_desc(f->skb, 0, virt_to_page(f->bufaddr), offset_in_page(f->bufaddr), n); + f->tag = newtag(d); + hout->tag = cpu_to_be32(f->tag); + skb->dev = d->ifp; skb_get(f->skb); f->skb->next = NULL; spin_unlock_irqrestore(&d->lock, flags); -- cgit v0.10.2 From dced3a053dd5415a7321e1ae153c96dea644da4e Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:49 -0400 Subject: aoe: improve retransmission heuristics Add a dynamic minimum timer for better retransmission behavior. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 4d79f1e..7b11217 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -125,8 +125,10 @@ struct aoedev { ulong sysminor; ulong aoemajor; ulong aoeminor; - ulong nopen; /* (bd_openers isn't available without sleeping) */ - ulong rttavg; /* round trip average of requests/responses */ + u16 nopen; /* (bd_openers isn't available without sleeping) */ + u16 lasttag; /* last tag sent */ + u16 rttavg; /* round trip average of requests/responses */ + u16 mintimer; u16 fw_ver; /* version of blade's firmware */ u16 maxbcnt; struct work_struct work;/* disk create work struct */ @@ -142,7 +144,6 @@ struct aoedev { mempool_t *bufpool; /* for deadlock-free Buf allocation */ struct list_head bufq; /* queue of bios to work on */ struct buf *inprocess; /* the one we're currently working on */ - ulong lasttag; /* last tag sent */ ushort lostjumbo; ushort nframes; /* number of frames below */ struct frame *frames; diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 621fdbb..c0bdc1f 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -461,8 +461,15 @@ calc_rttavg(struct aoedev *d, int rtt) register long n; n = rtt; - if (n < MINTIMER) - n = MINTIMER; + if (n < 0) { + n = -rtt; + if (n < MINTIMER) + n = MINTIMER; + else if (n > MAXTIMER) + n = MAXTIMER; + d->mintimer += (n - d->mintimer) >> 1; + } else if (n < d->mintimer) + n = d->mintimer; else if (n > MAXTIMER) n = MAXTIMER; @@ -498,8 +505,10 @@ aoecmd_ata_rsp(struct sk_buff *skb) spin_lock_irqsave(&d->lock, flags); - f = getframe(d, be32_to_cpu(hin->tag)); + n = be32_to_cpu(hin->tag); + f = getframe(d, n); if (f == NULL) { + calc_rttavg(d, -tsince(n)); spin_unlock_irqrestore(&d->lock, flags); snprintf(ebuf, sizeof ebuf, "%15s e%d.%d tag=%08x@%08lx\n", @@ -724,6 +733,7 @@ aoecmd_cfg_rsp(struct sk_buff *skb) return; } d->flags |= DEVFL_PAUSE; /* force pause */ + d->mintimer = MINTIMER; d->fw_ver = be16_to_cpu(ch->fwver); /* check for already outstanding ataid */ -- cgit v0.10.2 From 4f51dc5e9ae195d2e8c22e5f574e004c2f6518a4 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:49 -0400 Subject: aoe: zero copy write 2 of 2 Avoid memory copy on writes. (This patch follows patch 4.) Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 7b11217..b41fdfe 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -84,6 +84,7 @@ enum { DEVFL_PAUSE = (1<<5), DEVFL_NEWSIZE = (1<<6), /* need to update dev size in block layer */ DEVFL_MAXBCNT = (1<<7), /* d->maxbcnt is not changeable */ + DEVFL_KICKME = (1<<8), BUFFL_FAIL = 1, }; diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index c0bdc1f..9ebc98a 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -120,7 +120,7 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) h = (struct aoe_hdr *) skb->mac.raw; ah = (struct aoe_atahdr *) (h+1); skb->len = sizeof *h + sizeof *ah; - memset(h, 0, skb->len); + memset(h, 0, ETH_ZLEN); f->tag = aoehdr_atainit(d, h); f->waited = 0; f->buf = buf; @@ -143,8 +143,9 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) skb_fill_page_desc(skb, 0, virt_to_page(f->bufaddr), offset_in_page(f->bufaddr), bcnt); ah->aflags |= AOEAFL_WRITE; + skb->len += bcnt; + skb->data_len = bcnt; } else { - skb_shinfo(skb)->nr_frags = 0; skb->len = ETH_ZLEN; writebit = 0; } @@ -167,8 +168,9 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) } skb->dev = d->ifp; - skb_get(skb); - skb->next = NULL; + skb = skb_clone(skb, GFP_ATOMIC); + if (skb == NULL) + return; if (d->sendq_hd) d->sendq_tl->next = skb; else @@ -224,6 +226,29 @@ aoecmd_cfg_pkts(ushort aoemajor, unsigned char aoeminor, struct sk_buff **tail) return sl; } +static struct frame * +freeframe(struct aoedev *d) +{ + struct frame *f, *e; + int n = 0; + + f = d->frames; + e = f + d->nframes; + for (; ftag != FREETAG) + continue; + if (atomic_read(&skb_shinfo(f->skb)->dataref) == 1) { + skb_shinfo(f->skb)->nr_frags = f->skb->data_len = 0; + return f; + } + n++; + } + if (n == d->nframes) /* wait for network layer */ + d->flags |= DEVFL_KICKME; + + return NULL; +} + /* enters with d->lock held */ void aoecmd_work(struct aoedev *d) @@ -239,7 +264,7 @@ aoecmd_work(struct aoedev *d) } loop: - f = getframe(d, FREETAG); + f = freeframe(d); if (f == NULL) return; if (d->inprocess == NULL) { @@ -282,20 +307,25 @@ rexmit(struct aoedev *d, struct frame *f) n = DEFAULTBCNT / 512; if (ah->scnt > n) { ah->scnt = n; - if (ah->aflags & AOEAFL_WRITE) + if (ah->aflags & AOEAFL_WRITE) { skb_fill_page_desc(skb, 0, virt_to_page(f->bufaddr), offset_in_page(f->bufaddr), DEFAULTBCNT); + skb->len = sizeof *h + sizeof *ah + DEFAULTBCNT; + skb->data_len = DEFAULTBCNT; + } if (++d->lostjumbo > (d->nframes << 1)) if (d->maxbcnt != DEFAULTBCNT) { - iprintk("too many lost jumbo - using 1KB frames.\n"); + iprintk("e%ld.%ld: too many lost jumbo on %s - using 1KB frames.\n", + d->aoemajor, d->aoeminor, d->ifp->name); d->maxbcnt = DEFAULTBCNT; d->flags |= DEVFL_MAXBCNT; } } skb->dev = d->ifp; - skb_get(skb); - skb->next = NULL; + skb = skb_clone(skb, GFP_ATOMIC); + if (skb == NULL) + return; if (d->sendq_hd) d->sendq_tl->next = skb; else @@ -350,6 +380,10 @@ rexmit_timer(ulong vp) rexmit(d, f); } } + if (d->flags & DEVFL_KICKME) { + d->flags &= ~DEVFL_KICKME; + aoecmd_work(d); + } sl = d->sendq_hd; d->sendq_hd = d->sendq_tl = NULL; @@ -552,23 +586,27 @@ aoecmd_ata_rsp(struct sk_buff *skb) case WIN_WRITE: case WIN_WRITE_EXT: if (f->bcnt -= n) { + skb = f->skb; f->bufaddr += n; put_lba(ahout, f->lba += ahout->scnt); n = f->bcnt; if (n > DEFAULTBCNT) n = DEFAULTBCNT; ahout->scnt = n >> 9; - if (ahout->aflags & AOEAFL_WRITE) - skb_fill_page_desc(f->skb, 0, + if (ahout->aflags & AOEAFL_WRITE) { + skb_fill_page_desc(skb, 0, virt_to_page(f->bufaddr), offset_in_page(f->bufaddr), n); + skb->len = sizeof *hout + sizeof *ahout + n; + skb->data_len = n; + } f->tag = newtag(d); hout->tag = cpu_to_be32(f->tag); skb->dev = d->ifp; - skb_get(f->skb); - f->skb->next = NULL; + skb = skb_clone(skb, GFP_ATOMIC); spin_unlock_irqrestore(&d->lock, flags); - aoenet_xmit(f->skb); + if (skb) + aoenet_xmit(skb); return; } if (n > DEFAULTBCNT) @@ -642,7 +680,7 @@ aoecmd_ata_id(struct aoedev *d) struct frame *f; struct sk_buff *skb; - f = getframe(d, FREETAG); + f = freeframe(d); if (f == NULL) { eprintk("can't get a frame. This shouldn't happen.\n"); return NULL; @@ -652,8 +690,8 @@ aoecmd_ata_id(struct aoedev *d) skb = f->skb; h = (struct aoe_hdr *) skb->mac.raw; ah = (struct aoe_atahdr *) (h+1); - skb->len = sizeof *h + sizeof *ah; - memset(h, 0, skb->len); + skb->len = ETH_ZLEN; + memset(h, 0, ETH_ZLEN); f->tag = aoehdr_atainit(d, h); f->waited = 0; @@ -663,12 +701,11 @@ aoecmd_ata_id(struct aoedev *d) ah->lba3 = 0xa0; skb->dev = d->ifp; - skb_get(skb); d->rttavg = MAXTIMER; d->timer.function = rexmit_timer; - return skb; + return skb_clone(skb, GFP_ATOMIC); } void @@ -724,7 +761,12 @@ aoecmd_cfg_rsp(struct sk_buff *skb) n /= 512; if (n > ch->scnt) n = ch->scnt; - d->maxbcnt = n ? n * 512 : DEFAULTBCNT; + n = n ? n * 512 : DEFAULTBCNT; + if (n != d->maxbcnt) { + iprintk("e%ld.%ld: setting %d byte data frames on %s\n", + d->aoemajor, d->aoeminor, n, d->ifp->name); + d->maxbcnt = n; + } } /* don't change users' perspective */ diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index f51d87b..7fd63d4 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -121,6 +121,7 @@ aoedev_downdev(struct aoedev *d) mempool_free(buf, d->bufpool); bio_endio(bio, bio->bi_size, -EIO); } + skb_shinfo(f->skb)->nr_frags = f->skb->data_len = 0; } d->inprocess = NULL; -- cgit v0.10.2 From b751e8b6590efdb76e1682c85bfcd5f3531ccae4 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:50 -0400 Subject: aoe: module parameter for device timeout The aoe_deadsecs module parameter sets the number of seconds that elapse before a nonresponsive AoE device is marked as dead. This is runtime settable in sysfs or settable with a module load or kernel boot parameter. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 9ebc98a..f2b8f55 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -15,7 +15,10 @@ #define TIMERTICK (HZ / 10) #define MINTIMER (2 * TIMERTICK) #define MAXTIMER (HZ << 1) -#define MAXWAIT (60 * 3) /* After MAXWAIT seconds, give up and fail dev */ + +static int aoe_deadsecs = 60 * 3; +module_param(aoe_deadsecs, int, 0644); +MODULE_PARM_DESC(aoe_deadsecs, "After aoe_deadsecs seconds, give up and fail dev."); struct sk_buff * new_skb(ulong len) @@ -373,7 +376,7 @@ rexmit_timer(ulong vp) if (f->tag != FREETAG && tsince(f->tag) >= timeout) { n = f->waited += timeout; n /= HZ; - if (n > MAXWAIT) { /* waited too long. device failure. */ + if (n > aoe_deadsecs) { /* waited too long for response */ aoedev_downdev(d); break; } -- cgit v0.10.2 From 392e4845f9728114f7ffa8d7612683397fd4d441 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:50 -0400 Subject: aoe: use bio->bi_idx Instead of starting with bio->bi_io_vec, use the offset in bio->bi_idx. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index a7dbe6f..196ae7a 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -142,7 +142,8 @@ aoeblk_make_request(request_queue_t *q, struct bio *bio) buf->bio = bio; buf->resid = bio->bi_size; buf->sector = bio->bi_sector; - buf->bv = buf->bio->bi_io_vec; + buf->bv = &bio->bi_io_vec[bio->bi_idx]; + WARN_ON(buf->bv->bv_len == 0); buf->bv_resid = buf->bv->bv_len; buf->bufaddr = page_address(buf->bv->bv_page) + buf->bv->bv_offset; diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index f2b8f55..2d0bcdd 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -166,6 +166,7 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) d->inprocess = NULL; } else if (buf->bv_resid == 0) { buf->bv++; + WARN_ON(buf->bv->bv_len == 0); buf->bv_resid = buf->bv->bv_len; buf->bufaddr = page_address(buf->bv->bv_page) + buf->bv->bv_offset; } -- cgit v0.10.2 From b849086d8f77f8a1269a01d5552fbf355311f7ac Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:51 -0400 Subject: aoe: remove sysfs comment Remove unecessary comment. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index 196ae7a..088acf4 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -14,7 +14,6 @@ static kmem_cache_t *buf_pool_cache; -/* add attributes for our block devices in sysfs */ static ssize_t aoedisk_show_state(struct gendisk * disk, char *page) { struct aoedev *d = disk->private_data; -- cgit v0.10.2 From 086216db1435f44a58c18454acfa59f013510c95 Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:51 -0400 Subject: aoe: update driver version Update aoe driver version number to 32. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index b41fdfe..188bf09 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -1,5 +1,5 @@ /* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */ -#define VERSION "22" +#define VERSION "32" #define AOE_MAJOR 152 #define DEVICE_NAME "aoe" -- cgit v0.10.2 From a12c93f08b8fc83b7fcdabaf92b1adcea7489f5e Mon Sep 17 00:00:00 2001 From: "Ed L. Cashin" Date: Wed, 20 Sep 2006 14:36:51 -0400 Subject: aoe: revert printk macros This patch addresses the concern that the aoe driver should not introduce unecessary conventions that must be learned by the reader. It reverts patch 6. Signed-off-by: "Ed L. Cashin" Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 188bf09..6d11122 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -10,11 +10,6 @@ #define AOE_PARTITIONS (16) #endif -#define xprintk(L, fmt, arg...) printk(L "aoe: " "%s: " fmt, __func__, ## arg) -#define iprintk(fmt, arg...) xprintk(KERN_INFO, fmt, ## arg) -#define eprintk(fmt, arg...) xprintk(KERN_ERR, fmt, ## arg) -#define dprintk(fmt, arg...) xprintk(KERN_DEBUG, fmt, ## arg) - #define SYSMINOR(aoemajor, aoeminor) ((aoemajor) * NPERSHELF + (aoeminor)) #define AOEMAJOR(sysminor) ((sysminor) / NPERSHELF) #define AOEMINOR(sysminor) ((sysminor) % NPERSHELF) diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index 088acf4..4259b52 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -131,7 +131,7 @@ aoeblk_make_request(request_queue_t *q, struct bio *bio) d = bio->bi_bdev->bd_disk->private_data; buf = mempool_alloc(d->bufpool, GFP_NOIO); if (buf == NULL) { - iprintk("buf allocation failure\n"); + printk(KERN_INFO "aoe: buf allocation failure\n"); bio_endio(bio, bio->bi_size, -ENOMEM); return 0; } @@ -149,7 +149,8 @@ aoeblk_make_request(request_queue_t *q, struct bio *bio) spin_lock_irqsave(&d->lock, flags); if ((d->flags & DEVFL_UP) == 0) { - iprintk("device %ld.%ld is not up\n", d->aoemajor, d->aoeminor); + printk(KERN_INFO "aoe: device %ld.%ld is not up\n", + d->aoemajor, d->aoeminor); spin_unlock_irqrestore(&d->lock, flags); mempool_free(buf, d->bufpool); bio_endio(bio, bio->bi_size, -ENXIO); @@ -174,7 +175,7 @@ aoeblk_getgeo(struct block_device *bdev, struct hd_geometry *geo) struct aoedev *d = bdev->bd_disk->private_data; if ((d->flags & DEVFL_UP) == 0) { - eprintk("disk not up\n"); + printk(KERN_ERR "aoe: disk not up\n"); return -ENODEV; } @@ -201,7 +202,7 @@ aoeblk_gdalloc(void *vp) gd = alloc_disk(AOE_PARTITIONS); if (gd == NULL) { - eprintk("cannot allocate disk structure for %ld.%ld\n", + printk(KERN_ERR "aoe: cannot allocate disk structure for %ld.%ld\n", d->aoemajor, d->aoeminor); spin_lock_irqsave(&d->lock, flags); d->flags &= ~DEVFL_GDALLOC; @@ -211,7 +212,7 @@ aoeblk_gdalloc(void *vp) d->bufpool = mempool_create_slab_pool(MIN_BUFS, buf_pool_cache); if (d->bufpool == NULL) { - eprintk("cannot allocate bufpool for %ld.%ld\n", + printk(KERN_ERR "aoe: cannot allocate bufpool for %ld.%ld\n", d->aoemajor, d->aoeminor); put_disk(gd); spin_lock_irqsave(&d->lock, flags); diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index f5cab69..e22b4c9 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -55,7 +55,8 @@ static int interfaces(const char __user *str, size_t size) { if (set_aoe_iflist(str, size)) { - eprintk("could not set interface list: too many interfaces\n"); + printk(KERN_ERR + "aoe: could not set interface list: too many interfaces\n"); return -EINVAL; } return 0; @@ -78,7 +79,7 @@ revalidate(const char __user *str, size_t size) /* should be e%d.%d format */ n = sscanf(buf, "e%d.%d", &major, &minor); if (n != 2) { - eprintk("invalid device specification\n"); + printk(KERN_ERR "aoe: invalid device specification\n"); return -EINVAL; } d = aoedev_by_aoeaddr(major, minor); @@ -113,7 +114,7 @@ bail: spin_unlock_irqrestore(&emsgs_lock, flags); mp = kmalloc(n, GFP_ATOMIC); if (mp == NULL) { - eprintk("allocation failure, len=%ld\n", n); + printk(KERN_ERR "aoe: allocation failure, len=%ld\n", n); goto bail; } @@ -138,7 +139,7 @@ aoechr_write(struct file *filp, const char __user *buf, size_t cnt, loff_t *offp switch ((unsigned long) filp->private_data) { default: - iprintk("can't write to that file.\n"); + printk(KERN_INFO "aoe: can't write to that file.\n"); break; case MINOR_DISCOVER: ret = discover(); @@ -247,7 +248,7 @@ aoechr_init(void) n = register_chrdev(AOE_MAJOR, "aoechr", &aoe_fops); if (n < 0) { - eprintk("can't register char device\n"); + printk(KERN_ERR "aoe: can't register char device\n"); return n; } sema_init(&emsgs_sema, 0); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 2d0bcdd..8a13b1a 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -159,7 +159,7 @@ aoecmd_ata_rw(struct aoedev *d, struct frame *f) buf->nframesout += 1; buf->bufaddr += bcnt; buf->bv_resid -= bcnt; -/* dprintk("bv_resid=%ld\n", buf->bv_resid); */ +/* printk(KERN_DEBUG "aoe: bv_resid=%ld\n", buf->bv_resid); */ buf->resid -= bcnt; buf->sector += bcnt >> 9; if (buf->resid == 0) { @@ -203,7 +203,7 @@ aoecmd_cfg_pkts(ushort aoemajor, unsigned char aoeminor, struct sk_buff **tail) skb = new_skb(sizeof *h + sizeof *ch); if (skb == NULL) { - iprintk("skb alloc failure\n"); + printk(KERN_INFO "aoe: skb alloc failure\n"); continue; } skb->dev = ifp; @@ -276,7 +276,7 @@ loop: return; buf = container_of(d->bufq.next, struct buf, bufs); list_del(d->bufq.next); -/*dprintk("bi_size=%ld\n", buf->bio->bi_size); */ +/*printk(KERN_DEBUG "aoe: bi_size=%ld\n", buf->bio->bi_size); */ d->inprocess = buf; } aoecmd_ata_rw(d, f); @@ -319,7 +319,7 @@ rexmit(struct aoedev *d, struct frame *f) } if (++d->lostjumbo > (d->nframes << 1)) if (d->maxbcnt != DEFAULTBCNT) { - iprintk("e%ld.%ld: too many lost jumbo on %s - using 1KB frames.\n", + printk(KERN_INFO "aoe: e%ld.%ld: too many lost jumbo on %s - using 1KB frames.\n", d->aoemajor, d->aoeminor, d->ifp->name); d->maxbcnt = DEFAULTBCNT; d->flags |= DEVFL_MAXBCNT; @@ -472,7 +472,7 @@ ataid_complete(struct aoedev *d, unsigned char *id) } if (d->ssize != ssize) - iprintk("%012llx e%lu.%lu v%04x has %llu sectors\n", + printk(KERN_INFO "aoe: %012llx e%lu.%lu v%04x has %llu sectors\n", (unsigned long long)mac_addr(d->addr), d->aoemajor, d->aoeminor, d->fw_ver, (long long)ssize); @@ -483,7 +483,7 @@ ataid_complete(struct aoedev *d, unsigned char *id) d->flags |= DEVFL_NEWSIZE; } else { if (d->flags & DEVFL_GDALLOC) { - eprintk("can't schedule work for e%lu.%lu, %s\n", + printk(KERN_ERR "aoe: can't schedule work for e%lu.%lu, %s\n", d->aoemajor, d->aoeminor, "it's already on! This shouldn't happen.\n"); return; @@ -569,7 +569,8 @@ aoecmd_ata_rsp(struct sk_buff *skb) if (ahout->cmdstat == WIN_IDENTIFY) d->flags &= ~DEVFL_PAUSE; if (ahin->cmdstat & 0xa9) { /* these bits cleared on success */ - eprintk("ata error cmd=%2.2Xh stat=%2.2Xh from e%ld.%ld\n", + printk(KERN_ERR + "aoe: ata error cmd=%2.2Xh stat=%2.2Xh from e%ld.%ld\n", ahout->cmdstat, ahin->cmdstat, d->aoemajor, d->aoeminor); if (buf) @@ -580,7 +581,8 @@ aoecmd_ata_rsp(struct sk_buff *skb) case WIN_READ: case WIN_READ_EXT: if (skb->len - sizeof *hin - sizeof *ahin < n) { - eprintk("runt data size in read. skb->len=%d\n", + printk(KERN_ERR + "aoe: runt data size in read. skb->len=%d\n", skb->len); /* fail frame f? just returning will rexmit. */ spin_unlock_irqrestore(&d->lock, flags); @@ -618,7 +620,8 @@ aoecmd_ata_rsp(struct sk_buff *skb) break; case WIN_IDENTIFY: if (skb->len - sizeof *hin - sizeof *ahin < 512) { - iprintk("runt data size in ataid. skb->len=%d\n", + printk(KERN_INFO + "aoe: runt data size in ataid. skb->len=%d\n", skb->len); spin_unlock_irqrestore(&d->lock, flags); return; @@ -626,7 +629,8 @@ aoecmd_ata_rsp(struct sk_buff *skb) ataid_complete(d, (char *) (ahin+1)); break; default: - iprintk("unrecognized ata command %2.2Xh for %d.%d\n", + printk(KERN_INFO + "aoe: unrecognized ata command %2.2Xh for %d.%d\n", ahout->cmdstat, be16_to_cpu(hin->major), hin->minor); @@ -686,7 +690,7 @@ aoecmd_ata_id(struct aoedev *d) f = freeframe(d); if (f == NULL) { - eprintk("can't get a frame. This shouldn't happen.\n"); + printk(KERN_ERR "aoe: can't get a frame. This shouldn't happen.\n"); return NULL; } @@ -732,14 +736,14 @@ aoecmd_cfg_rsp(struct sk_buff *skb) */ aoemajor = be16_to_cpu(h->major); if (aoemajor == 0xfff) { - eprintk("Warning: shelf address is all ones. " + printk(KERN_ERR "aoe: Warning: shelf address is all ones. " "Check shelf dip switches.\n"); return; } sysminor = SYSMINOR(aoemajor, h->minor); if (sysminor * AOE_PARTITIONS + AOE_PARTITIONS > MINORMASK) { - iprintk("e%ld.%d: minor number too large\n", + printk(KERN_INFO "aoe: e%ld.%d: minor number too large\n", aoemajor, (int) h->minor); return; } @@ -750,7 +754,7 @@ aoecmd_cfg_rsp(struct sk_buff *skb) d = aoedev_by_sysminor_m(sysminor, n); if (d == NULL) { - iprintk("device sysminor_m failure\n"); + printk(KERN_INFO "aoe: device sysminor_m failure\n"); return; } @@ -767,7 +771,8 @@ aoecmd_cfg_rsp(struct sk_buff *skb) n = ch->scnt; n = n ? n * 512 : DEFAULTBCNT; if (n != d->maxbcnt) { - iprintk("e%ld.%ld: setting %d byte data frames on %s\n", + printk(KERN_INFO + "aoe: e%ld.%ld: setting %d byte data frames on %s\n", d->aoemajor, d->aoeminor, n, d->ifp->name); d->maxbcnt = n; } diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index 7fd63d4..6125921 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -156,7 +156,7 @@ aoedev_by_sysminor_m(ulong sysminor, ulong bufcnt) d = aoedev_newdev(bufcnt); if (d == NULL) { spin_unlock_irqrestore(&devlist_lock, flags); - iprintk("aoedev_newdev failure.\n"); + printk(KERN_INFO "aoe: aoedev_newdev failure.\n"); return NULL; } d->sysminor = sysminor; diff --git a/drivers/block/aoe/aoemain.c b/drivers/block/aoe/aoemain.c index 13e634db..a04b7d6 100644 --- a/drivers/block/aoe/aoemain.c +++ b/drivers/block/aoe/aoemain.c @@ -84,11 +84,11 @@ aoe_init(void) goto net_fail; ret = register_blkdev(AOE_MAJOR, DEVICE_NAME); if (ret < 0) { - eprintk("can't register major\n"); + printk(KERN_ERR "aoe: can't register major\n"); goto blkreg_fail; } - iprintk("AoE v%s initialised.\n", VERSION); + printk(KERN_INFO "aoe: AoE v%s initialised.\n", VERSION); discover_timer(TINIT); return 0; @@ -101,7 +101,7 @@ aoe_init(void) chr_fail: aoedev_exit(); - iprintk("initialisation failure.\n"); + printk(KERN_INFO "aoe: initialisation failure.\n"); return ret; } diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index f1cf266..9626e0f 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -74,7 +74,7 @@ set_aoe_iflist(const char __user *user_str, size_t size) return -EINVAL; if (copy_from_user(aoe_iflist, user_str, size)) { - iprintk("copy from user failed\n"); + printk(KERN_INFO "aoe: copy from user failed\n"); return -EFAULT; } aoe_iflist[size] = 0x00; @@ -132,7 +132,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, if (n > NECODES) n = 0; if (net_ratelimit()) - eprintk("error packet from %d.%d; ecode=%d '%s'\n", + printk(KERN_ERR "aoe: error packet from %d.%d; ecode=%d '%s'\n", be16_to_cpu(h->major), h->minor, h->err, aoe_errlist[n]); goto exit; @@ -146,7 +146,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, aoecmd_cfg_rsp(skb); break; default: - iprintk("unknown cmd %d\n", h->cmd); + printk(KERN_INFO "aoe: unknown cmd %d\n", h->cmd); } exit: dev_kfree_skb(skb); -- cgit v0.10.2 From 4ca5224f3ea4779054d96e885ca9b3980801ce13 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 9 Apr 2002 12:14:34 -0700 Subject: aoe: fix sysfs_create_file warnings Moved the attributes into a group, making the compiler be quiet about ignoring the return value of the file create calls. This also also fixed a bug when removing the files, which were not symlinks. Cc: "Ed L. Cashin" Cc: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index 4259b52..d433f27 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -63,21 +63,26 @@ static struct disk_attribute disk_attr_fwver = { .show = aoedisk_show_fwver }; -static void +static struct attribute *aoe_attrs[] = { + &disk_attr_state.attr, + &disk_attr_mac.attr, + &disk_attr_netif.attr, + &disk_attr_fwver.attr, +}; + +static const struct attribute_group attr_group = { + .attrs = aoe_attrs, +}; + +static int aoedisk_add_sysfs(struct aoedev *d) { - sysfs_create_file(&d->gd->kobj, &disk_attr_state.attr); - sysfs_create_file(&d->gd->kobj, &disk_attr_mac.attr); - sysfs_create_file(&d->gd->kobj, &disk_attr_netif.attr); - sysfs_create_file(&d->gd->kobj, &disk_attr_fwver.attr); + return sysfs_create_group(&d->gd->kobj, &attr_group); } void aoedisk_rm_sysfs(struct aoedev *d) { - sysfs_remove_link(&d->gd->kobj, "state"); - sysfs_remove_link(&d->gd->kobj, "mac"); - sysfs_remove_link(&d->gd->kobj, "netif"); - sysfs_remove_link(&d->gd->kobj, "firmware-version"); + sysfs_remove_group(&d->gd->kobj, &attr_group); } static int -- cgit v0.10.2 From 34fc921a253f3ddfc4ad9de1dbc88683f84fbaaa Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Sun, 8 Oct 2006 21:56:29 +0200 Subject: w83791d: Fix unchecked return status Replace all unchecked calls to device_create_file with a single group declaration, and one call to sysfs_create_group, and check that one return status. Also remove the files on device detach. Signed-off-by: Jim Cromie Signed-off by: Charles Spirakis Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/hwmon/w83791d.c b/drivers/hwmon/w83791d.c index 371ed4f..9e5f885 100644 --- a/drivers/hwmon/w83791d.c +++ b/drivers/hwmon/w83791d.c @@ -746,6 +746,52 @@ static ssize_t store_vrm_reg(struct device *dev, static DEVICE_ATTR(vrm, S_IRUGO | S_IWUSR, show_vrm_reg, store_vrm_reg); +#define IN_UNIT_ATTRS(X) \ + &sda_in_input[X].dev_attr.attr, \ + &sda_in_min[X].dev_attr.attr, \ + &sda_in_max[X].dev_attr.attr + +#define FAN_UNIT_ATTRS(X) \ + &sda_fan_input[X].dev_attr.attr, \ + &sda_fan_min[X].dev_attr.attr, \ + &sda_fan_div[X].dev_attr.attr + +#define TEMP_UNIT_ATTRS(X) \ + &sda_temp_input[X].dev_attr.attr, \ + &sda_temp_max[X].dev_attr.attr, \ + &sda_temp_max_hyst[X].dev_attr.attr + +static struct attribute *w83791d_attributes[] = { + IN_UNIT_ATTRS(0), + IN_UNIT_ATTRS(1), + IN_UNIT_ATTRS(2), + IN_UNIT_ATTRS(3), + IN_UNIT_ATTRS(4), + IN_UNIT_ATTRS(5), + IN_UNIT_ATTRS(6), + IN_UNIT_ATTRS(7), + IN_UNIT_ATTRS(8), + IN_UNIT_ATTRS(9), + FAN_UNIT_ATTRS(0), + FAN_UNIT_ATTRS(1), + FAN_UNIT_ATTRS(2), + FAN_UNIT_ATTRS(3), + FAN_UNIT_ATTRS(4), + TEMP_UNIT_ATTRS(0), + TEMP_UNIT_ATTRS(1), + TEMP_UNIT_ATTRS(2), + &dev_attr_alarms.attr, + &sda_beep_ctrl[0].dev_attr.attr, + &sda_beep_ctrl[1].dev_attr.attr, + &dev_attr_cpu0_vid.attr, + &dev_attr_vrm.attr, + NULL +}; + +static const struct attribute_group w83791d_group = { + .attrs = w83791d_attributes, +}; + /* This function is called when: * w83791d_driver is inserted (when this module is loaded), for each available adapter @@ -967,41 +1013,20 @@ static int w83791d_detect(struct i2c_adapter *adapter, int address, int kind) } /* Register sysfs hooks */ + if ((err = sysfs_create_group(&client->dev.kobj, &w83791d_group))) + goto error3; + + /* Everything is ready, now register the working device */ data->class_dev = hwmon_device_register(dev); if (IS_ERR(data->class_dev)) { err = PTR_ERR(data->class_dev); - goto error3; + goto error4; } - for (i = 0; i < NUMBER_OF_VIN; i++) { - device_create_file(dev, &sda_in_input[i].dev_attr); - device_create_file(dev, &sda_in_min[i].dev_attr); - device_create_file(dev, &sda_in_max[i].dev_attr); - } - - for (i = 0; i < NUMBER_OF_FANIN; i++) { - device_create_file(dev, &sda_fan_input[i].dev_attr); - device_create_file(dev, &sda_fan_div[i].dev_attr); - device_create_file(dev, &sda_fan_min[i].dev_attr); - } - - for (i = 0; i < NUMBER_OF_TEMPIN; i++) { - device_create_file(dev, &sda_temp_input[i].dev_attr); - device_create_file(dev, &sda_temp_max[i].dev_attr); - device_create_file(dev, &sda_temp_max_hyst[i].dev_attr); - } - - device_create_file(dev, &dev_attr_alarms); - - for (i = 0; i < ARRAY_SIZE(sda_beep_ctrl); i++) { - device_create_file(dev, &sda_beep_ctrl[i].dev_attr); - } - - device_create_file(dev, &dev_attr_cpu0_vid); - device_create_file(dev, &dev_attr_vrm); - return 0; +error4: + sysfs_remove_group(&client->dev.kobj, &w83791d_group); error3: if (data->lm75[0] != NULL) { i2c_detach_client(data->lm75[0]); @@ -1025,8 +1050,10 @@ static int w83791d_detach_client(struct i2c_client *client) int err; /* main client */ - if (data) + if (data) { hwmon_device_unregister(data->class_dev); + sysfs_remove_group(&client->dev.kobj, &w83791d_group); + } if ((err = i2c_detach_client(client))) return err; -- cgit v0.10.2 From 2ca7b961c3c9f072d307293aad0f9705522e916a Mon Sep 17 00:00:00 2001 From: Grant Coady Date: Sun, 8 Oct 2006 21:57:41 +0200 Subject: adm9240: Update Grant Coady's email address Replace a bouncing email that I cannot recover from Mr Google. Signed-off-by: Grant Coady Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/hwmon/adm9240 b/Documentation/hwmon/adm9240 index 35f618f..2c6f1fe 100644 --- a/Documentation/hwmon/adm9240 +++ b/Documentation/hwmon/adm9240 @@ -24,7 +24,7 @@ Authors: Frodo Looijaard , Philip Edelbrock , Michiel Rook , - Grant Coady with guidance + Grant Coady with guidance from Jean Delvare Interface diff --git a/drivers/hwmon/adm9240.c b/drivers/hwmon/adm9240.c index 377961c..aad594a 100644 --- a/drivers/hwmon/adm9240.c +++ b/drivers/hwmon/adm9240.c @@ -5,7 +5,7 @@ * Copyright (C) 1999 Frodo Looijaard * Philip Edelbrock * Copyright (C) 2003 Michiel Rook - * Copyright (C) 2005 Grant Coady with valuable + * Copyright (C) 2005 Grant Coady with valuable * guidance from Jean Delvare * * Driver supports Analog Devices ADM9240 @@ -774,7 +774,7 @@ static void __exit sensors_adm9240_exit(void) } MODULE_AUTHOR("Michiel Rook , " - "Grant Coady and others"); + "Grant Coady and others"); MODULE_DESCRIPTION("ADM9240/DS1780/LM81 driver"); MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 15fe25ca67234514d7cf41af28096c1330f44950 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 8 Oct 2006 21:59:54 +0200 Subject: hwmon: Fix documentation typos Fix typos in hardware monitoring documentation. Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/hwmon/f71805f b/Documentation/hwmon/f71805f index 28c5b7d..2ca69df 100644 --- a/Documentation/hwmon/f71805f +++ b/Documentation/hwmon/f71805f @@ -17,7 +17,7 @@ Thanks to Kris Chen from Fintek for answering technical questions and providing additional documentation. Thanks to Chris Lin from Jetway for providing wiring schematics and -anwsering technical questions. +answering technical questions. Description diff --git a/Documentation/hwmon/w83627ehf b/Documentation/hwmon/w83627ehf index fae3b78..caa610a 100644 --- a/Documentation/hwmon/w83627ehf +++ b/Documentation/hwmon/w83627ehf @@ -26,7 +26,7 @@ fan control mode). Temperatures are measured in degrees Celsius and measurement resolution is 1 degC for temp1 and 0.5 degC for temp2 and temp3. An alarm is triggered when the temperature gets higher than high limit; it stays on until the temperature -falls below the Hysteresis value. +falls below the hysteresis value. Fan rotation speeds are reported in RPM (rotations per minute). An alarm is triggered if the rotation speed has dropped below a programmable limit. Fan @@ -67,9 +67,9 @@ Thermal Cruise mode If the temperature is in the range defined by: -pwm[1-4]_target - set target temperature, unit millidegree Celcius +pwm[1-4]_target - set target temperature, unit millidegree Celsius (range 0 - 127000) -pwm[1-4]_tolerance - tolerance, unit millidegree Celcius (range 0 - 15000) +pwm[1-4]_tolerance - tolerance, unit millidegree Celsius (range 0 - 15000) there are no changes to fan speed. Once the temperature leaves the interval, fan speed increases (temp is higher) or decreases if lower than desired. -- cgit v0.10.2 From 6091780eba5d195213747b515a62211ac97641f1 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 8 Oct 2006 22:00:44 +0200 Subject: smsc47m1: List the SMSC LPC47M112 as supported The SMSC LPC47M112 Super-I/O chip appears to be compatible with the LPC47M10x and LPC47M13x as far as hardware monitoring is concerned. The device ID is even the same, so it's really only a documentation update. Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/hwmon/smsc47m1 b/Documentation/hwmon/smsc47m1 index c15bbe6..04a1112 100644 --- a/Documentation/hwmon/smsc47m1 +++ b/Documentation/hwmon/smsc47m1 @@ -2,12 +2,14 @@ Kernel driver smsc47m1 ====================== Supported chips: - * SMSC LPC47B27x, LPC47M10x, LPC47M13x, LPC47M14x, LPC47M15x and LPC47M192 + * SMSC LPC47B27x, LPC47M112, LPC47M10x, LPC47M13x, LPC47M14x, + LPC47M15x and LPC47M192 Addresses scanned: none, address read from Super I/O config space Prefix: 'smsc47m1' Datasheets: http://www.smsc.com/main/datasheets/47b27x.pdf http://www.smsc.com/main/datasheets/47m10x.pdf + http://www.smsc.com/main/datasheets/47m112.pdf http://www.smsc.com/main/tools/discontinued/47m13x.pdf http://www.smsc.com/main/datasheets/47m14x.pdf http://www.smsc.com/main/tools/discontinued/47m15x.pdf diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig index 9b88b25..e0dede7 100644 --- a/drivers/hwmon/Kconfig +++ b/drivers/hwmon/Kconfig @@ -369,8 +369,8 @@ config SENSORS_SMSC47M1 help If you say yes here you get support for the integrated fan monitoring and control capabilities of the SMSC LPC47B27x, - LPC47M10x, LPC47M13x, LPC47M14x, LPC47M15x, LPC47M192 and - LPC47M997 chips. + LPC47M10x, LPC47M112, LPC47M13x, LPC47M14x, LPC47M15x, + LPC47M192 and LPC47M997 chips. The temperature and voltage sensor features of the LPC47M192 and LPC47M997 are supported by another driver, select also diff --git a/drivers/hwmon/smsc47m1.c b/drivers/hwmon/smsc47m1.c index 47132fd..beb881c 100644 --- a/drivers/hwmon/smsc47m1.c +++ b/drivers/hwmon/smsc47m1.c @@ -2,8 +2,8 @@ smsc47m1.c - Part of lm_sensors, Linux kernel modules for hardware monitoring - Supports the SMSC LPC47B27x, LPC47M10x, LPC47M13x, LPC47M14x, - LPC47M15x, LPC47M192 and LPC47M997 Super-I/O chips. + Supports the SMSC LPC47B27x, LPC47M10x, LPC47M112, LPC47M13x, + LPC47M14x, LPC47M15x, LPC47M192 and LPC47M997 Super-I/O chips. Copyright (C) 2002 Mark D. Studebaker Copyright (C) 2004 Jean Delvare @@ -380,8 +380,8 @@ static int __init smsc47m1_find(unsigned short *addr) val = superio_inb(SUPERIO_REG_DEVID); /* - * SMSC LPC47M10x/LPC47M13x (device id 0x59), LPC47M14x (device id - * 0x5F) and LPC47B27x (device id 0x51) have fan control. + * SMSC LPC47M10x/LPC47M112/LPC47M13x (device id 0x59), LPC47M14x + * (device id 0x5F) and LPC47B27x (device id 0x51) have fan control. * The LPC47M15x and LPC47M192 chips "with hardware monitoring block" * can do much more besides (device id 0x60). * The LPC47M997 is undocumented, but seems to be compatible with @@ -390,7 +390,8 @@ static int __init smsc47m1_find(unsigned short *addr) if (val == 0x51) printk(KERN_INFO "smsc47m1: Found SMSC LPC47B27x\n"); else if (val == 0x59) - printk(KERN_INFO "smsc47m1: Found SMSC LPC47M10x/LPC47M13x\n"); + printk(KERN_INFO "smsc47m1: Found SMSC " + "LPC47M10x/LPC47M112/LPC47M13x\n"); else if (val == 0x5F) printk(KERN_INFO "smsc47m1: Found SMSC LPC47M14x\n"); else if (val == 0x60) -- cgit v0.10.2 From 4660cb354a1dacbbc9c96f94eceedc38fe542fe2 Mon Sep 17 00:00:00 2001 From: Rudolf Marek Date: Sun, 8 Oct 2006 22:01:26 +0200 Subject: k8temp: Documentation update Update the documentation for the k8temp driver. Signed-off-by: Rudolf Marek Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/Documentation/hwmon/k8temp b/Documentation/hwmon/k8temp index bab445a..30d123b 100644 --- a/Documentation/hwmon/k8temp +++ b/Documentation/hwmon/k8temp @@ -2,7 +2,7 @@ Kernel driver k8temp ==================== Supported chips: - * AMD K8 CPU + * AMD Athlon64/FX or Opteron CPUs Prefix: 'k8temp' Addresses scanned: PCI space Datasheet: http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/32559.pdf @@ -13,10 +13,13 @@ Contact: Rudolf Marek Description ----------- -This driver permits reading temperature sensor(s) embedded inside AMD K8 CPUs. -Official documentation says that it works from revision F of K8 core, but -in fact it seems to be implemented for all revisions of K8 except the first -two revisions (SH-B0 and SH-B3). +This driver permits reading temperature sensor(s) embedded inside AMD K8 +family CPUs (Athlon64/FX, Opteron). Official documentation says that it works +from revision F of K8 core, but in fact it seems to be implemented for all +revisions of K8 except the first two revisions (SH-B0 and SH-B3). + +Please note that you will need at least lm-sensors 2.10.1 for proper userspace +support. There can be up to four temperature sensors inside single CPU. The driver will auto-detect the sensors and will display only temperatures from diff --git a/MAINTAINERS b/MAINTAINERS index 5305dd6..595b34c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1668,6 +1668,12 @@ M: sct@redhat.com, akpm@osdl.org L: ext2-devel@lists.sourceforge.net S: Maintained +K8TEMP HARDWARE MONITORING DRIVER +P: Rudolf Marek +M: r.marek@assembler.cz +L: lm-sensors@lm-sensors.org +S: Maintained + KCONFIG P: Roman Zippel M: zippel@linux-m68k.org diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig index e0dede7..e76d919 100644 --- a/drivers/hwmon/Kconfig +++ b/drivers/hwmon/Kconfig @@ -95,11 +95,13 @@ config SENSORS_ADM9240 will be called adm9240. config SENSORS_K8TEMP - tristate "AMD K8 processor sensor" + tristate "AMD Athlon64/FX or Opteron temperature sensor" depends on HWMON && X86 && PCI && EXPERIMENTAL help If you say yes here you get support for the temperature - sensor(s) inside your AMD K8 CPU. + sensor(s) inside your CPU. Supported is whole AMD K8 + microarchitecture. Please note that you will need at least + lm-sensors 2.10.1 for proper userspace support. This driver can also be built as a module. If so, the module will be called k8temp. -- cgit v0.10.2 From 14992c7eff937bb12c8ebf2d91dbaa8c2f0cfc87 Mon Sep 17 00:00:00 2001 From: Rudolf Marek Date: Sun, 8 Oct 2006 22:02:09 +0200 Subject: w83627ehf: Fix the detection of fan5 Fix the detection of fan5 and preserve the bit between the register writes, because the bit is write only. Signed-off-by: Rudolf Marek Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/hwmon/w83627ehf.c b/drivers/hwmon/w83627ehf.c index 833faa2..2257806 100644 --- a/drivers/hwmon/w83627ehf.c +++ b/drivers/hwmon/w83627ehf.c @@ -354,6 +354,8 @@ static void w83627ehf_write_fan_div(struct i2c_client *client, int nr) case 0: reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV1) & 0xcf) | ((data->fan_div[0] & 0x03) << 4); + /* fan5 input control bit is write only, compute the value */ + reg |= (data->has_fan & (1 << 4)) ? 1 : 0; w83627ehf_write_value(client, W83627EHF_REG_FANDIV1, reg); reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0xdf) | ((data->fan_div[0] & 0x04) << 3); @@ -362,6 +364,8 @@ static void w83627ehf_write_fan_div(struct i2c_client *client, int nr) case 1: reg = (w83627ehf_read_value(client, W83627EHF_REG_FANDIV1) & 0x3f) | ((data->fan_div[1] & 0x03) << 6); + /* fan5 input control bit is write only, compute the value */ + reg |= (data->has_fan & (1 << 4)) ? 1 : 0; w83627ehf_write_value(client, W83627EHF_REG_FANDIV1, reg); reg = (w83627ehf_read_value(client, W83627EHF_REG_VBAT) & 0xbf) | ((data->fan_div[1] & 0x04) << 4); @@ -1216,13 +1220,16 @@ static int w83627ehf_detect(struct i2c_adapter *adapter) superio_exit(); /* It looks like fan4 and fan5 pins can be alternatively used - as fan on/off switches */ + as fan on/off switches, but fan5 control is write only :/ + We assume that if the serial interface is disabled, designers + connected fan5 as input unless they are emitting log 1, which + is not the default. */ data->has_fan = 0x07; /* fan1, fan2 and fan3 */ i = w83627ehf_read_value(client, W83627EHF_REG_FANDIV1); if ((i & (1 << 2)) && (!fan4pin)) data->has_fan |= (1 << 3); - if ((i & (1 << 0)) && (!fan5pin)) + if (!(i & (1 << 1)) && (!fan5pin)) data->has_fan |= (1 << 4); /* Register sysfs hooks */ -- cgit v0.10.2 From e693810ce8495ce3e227dacaa83f501b3b8ab204 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Fri, 13 Oct 2006 16:56:28 +0200 Subject: hwmon: Let w83781d and lm78 load again Let the w83781d and lm78 hardware monitoring drivers load even when no chip was detected at the ISA address. There can still be supported chips connected to an I2C bus or SMBus. This fixes bug #7293. Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/hwmon/lm78.c b/drivers/hwmon/lm78.c index ac1b746..73bc2ff 100644 --- a/drivers/hwmon/lm78.c +++ b/drivers/hwmon/lm78.c @@ -815,18 +815,18 @@ static int __init sm_lm78_init(void) if (res) return res; - res = i2c_isa_add_driver(&lm78_isa_driver); - if (res) { - i2c_del_driver(&lm78_driver); - return res; - } + /* Don't exit if this one fails, we still want the I2C variants + to work! */ + if (i2c_isa_add_driver(&lm78_isa_driver)) + isa_address = 0; return 0; } static void __exit sm_lm78_exit(void) { - i2c_isa_del_driver(&lm78_isa_driver); + if (isa_address) + i2c_isa_del_driver(&lm78_isa_driver); i2c_del_driver(&lm78_driver); } diff --git a/drivers/hwmon/w83781d.c b/drivers/hwmon/w83781d.c index a4584ec..fea9206 100644 --- a/drivers/hwmon/w83781d.c +++ b/drivers/hwmon/w83781d.c @@ -1685,11 +1685,10 @@ sensors_w83781d_init(void) if (res) return res; - res = i2c_isa_add_driver(&w83781d_isa_driver); - if (res) { - i2c_del_driver(&w83781d_driver); - return res; - } + /* Don't exit if this one fails, we still want the I2C variants + to work! */ + if (i2c_isa_add_driver(&w83781d_isa_driver)) + isa_address = 0; return 0; } @@ -1697,7 +1696,8 @@ sensors_w83781d_init(void) static void __exit sensors_w83781d_exit(void) { - i2c_isa_del_driver(&w83781d_isa_driver); + if (isa_address) + i2c_isa_del_driver(&w83781d_isa_driver); i2c_del_driver(&w83781d_driver); } diff --git a/drivers/i2c/busses/i2c-isa.c b/drivers/i2c/busses/i2c-isa.c index 4380653..8ed59a2 100644 --- a/drivers/i2c/busses/i2c-isa.c +++ b/drivers/i2c/busses/i2c-isa.c @@ -91,7 +91,7 @@ int i2c_isa_add_driver(struct i2c_driver *driver) /* Now look for clients */ res = driver->attach_adapter(&isa_adapter); if (res) { - dev_err(&isa_adapter.dev, + dev_dbg(&isa_adapter.dev, "Driver %s failed to attach adapter, unregistering\n", driver->driver.name); driver_unregister(&driver->driver); -- cgit v0.10.2 From bd452e6f178a559408c54c2b4ca29191b812d47f Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Fri, 13 Oct 2006 17:03:42 +0200 Subject: hwmon: Fix debug messages in w83781d Fix debug messages in w83781d at detection time. We can't use dev_dbg() on an i2c client's device before calling i2c_attach_client() on that client. Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/hwmon/w83781d.c b/drivers/hwmon/w83781d.c index fea9206..1232171 100644 --- a/drivers/hwmon/w83781d.c +++ b/drivers/hwmon/w83781d.c @@ -1099,7 +1099,8 @@ w83781d_detect(struct i2c_adapter *adapter, int address, int kind) bank. */ if (kind < 0) { if (w83781d_read_value(client, W83781D_REG_CONFIG) & 0x80) { - dev_dbg(dev, "Detection failed at step 3\n"); + dev_dbg(&adapter->dev, "Detection of w83781d chip " + "failed at step 3\n"); err = -ENODEV; goto ERROR2; } @@ -1109,7 +1110,8 @@ w83781d_detect(struct i2c_adapter *adapter, int address, int kind) if ((!(val1 & 0x07)) && (((!(val1 & 0x80)) && (val2 != 0xa3) && (val2 != 0xc3)) || ((val1 & 0x80) && (val2 != 0x5c) && (val2 != 0x12)))) { - dev_dbg(dev, "Detection failed at step 4\n"); + dev_dbg(&adapter->dev, "Detection of w83781d chip " + "failed at step 4\n"); err = -ENODEV; goto ERROR2; } @@ -1119,7 +1121,8 @@ w83781d_detect(struct i2c_adapter *adapter, int address, int kind) ((val1 & 0x80) && (val2 == 0x5c)))) { if (w83781d_read_value (client, W83781D_REG_I2C_ADDR) != address) { - dev_dbg(dev, "Detection failed at step 5\n"); + dev_dbg(&adapter->dev, "Detection of w83781d " + "chip failed at step 5\n"); err = -ENODEV; goto ERROR2; } @@ -1141,8 +1144,8 @@ w83781d_detect(struct i2c_adapter *adapter, int address, int kind) else if (val2 == 0x12) vendid = asus; else { - dev_dbg(dev, "Chip was made by neither " - "Winbond nor Asus?\n"); + dev_dbg(&adapter->dev, "w83781d chip vendor is " + "neither Winbond nor Asus\n"); err = -ENODEV; goto ERROR2; } @@ -1161,10 +1164,9 @@ w83781d_detect(struct i2c_adapter *adapter, int address, int kind) kind = as99127f; else { if (kind == 0) - dev_warn(dev, "Ignoring 'force' " + dev_warn(&adapter->dev, "Ignoring 'force' " "parameter for unknown chip at " - "adapter %d, address 0x%02x\n", - i2c_adapter_id(adapter), address); + "address 0x%02x\n", address); err = -EINVAL; goto ERROR2; } -- cgit v0.10.2 From 065fd1772af2032bebdce006071df007c039734d Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:38:05 -0700 Subject: [TIPC]: Add missing unlock in port timeout code. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/port.c b/net/tipc/port.c index b9c8c6b..c1a1a76 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -505,8 +505,13 @@ static void port_timeout(unsigned long ref) struct port *p_ptr = tipc_port_lock(ref); struct sk_buff *buf = NULL; - if (!p_ptr || !p_ptr->publ.connected) + if (!p_ptr) + return; + + if (!p_ptr->publ.connected) { + tipc_port_unlock(p_ptr); return; + } /* Last probe answered ? */ if (p_ptr->probing_state == PROBING) { -- cgit v0.10.2 From 29ede244cc4cfb11432a0bffd158ba09e7b2c167 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:42:04 -0700 Subject: [TIPC]: Debug print buffer enhancements and fixes This change modifies TIPC's print buffer code as follows: 1) Now supports small print buffers (min. size reduced from 512 bytes to 64) 2) Now uses TIPC_NULL print buffer structure to indicate null device instead of NULL pointer (this simplified error handling) 3) Fixed misuse of console buffer structure by tipc_dump() 4) Added and corrected comments in various places Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/core.h b/net/tipc/core.h index 762aac2..47504ec 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -65,7 +65,7 @@ #define assert(i) BUG_ON(!(i)) struct tipc_msg; -extern struct print_buf *TIPC_CONS, *TIPC_LOG; +extern struct print_buf *TIPC_NULL, *TIPC_CONS, *TIPC_LOG; extern struct print_buf *TIPC_TEE(struct print_buf *, struct print_buf *); void tipc_msg_print(struct print_buf*,struct tipc_msg *,const char*); void tipc_printf(struct print_buf *, const char *fmt, ...); @@ -94,11 +94,11 @@ void tipc_dump(struct print_buf*,const char *fmt, ...); * here, or on a per .c file basis, by redefining these symbols. The following * print buffer options are available: * - * NULL : Output to null print buffer (i.e. print nowhere) - * TIPC_CONS : Output to system console - * TIPC_LOG : Output to TIPC log buffer - * &buf : Output to user-defined buffer (struct print_buf *) - * TIPC_TEE(&buf_a,&buf_b) : Output to two print buffers (eg. TIPC_TEE(TIPC_CONS,TIPC_LOG) ) + * TIPC_NULL : null buffer (i.e. print nowhere) + * TIPC_CONS : system console + * TIPC_LOG : TIPC log buffer + * &buf : user-defined buffer (struct print_buf *) + * TIPC_TEE(&buf_a,&buf_b) : list of buffers (eg. TIPC_TEE(TIPC_CONS,TIPC_LOG)) */ #ifndef TIPC_OUTPUT @@ -106,7 +106,7 @@ void tipc_dump(struct print_buf*,const char *fmt, ...); #endif #ifndef DBG_OUTPUT -#define DBG_OUTPUT NULL +#define DBG_OUTPUT TIPC_NULL #endif #else @@ -136,7 +136,7 @@ void tipc_dump(struct print_buf*,const char *fmt, ...); #define TIPC_OUTPUT TIPC_CONS #undef DBG_OUTPUT -#define DBG_OUTPUT NULL +#define DBG_OUTPUT TIPC_NULL #endif diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index 5513065..d8af4c2 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -1,8 +1,8 @@ /* - * net/tipc/dbg.c: TIPC print buffer routines for debuggign + * net/tipc/dbg.c: TIPC print buffer routines for debugging * * Copyright (c) 1996-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005-2006, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -38,11 +38,12 @@ #include "config.h" #include "dbg.h" -#define MAX_STRING 512 - -static char print_string[MAX_STRING]; +static char print_string[TIPC_PB_MAX_STR]; static DEFINE_SPINLOCK(print_lock); +static struct print_buf null_buf = { NULL, 0, NULL, NULL }; +struct print_buf *TIPC_NULL = &null_buf; + static struct print_buf cons_buf = { NULL, 0, NULL, NULL }; struct print_buf *TIPC_CONS = &cons_buf; @@ -62,68 +63,83 @@ struct print_buf *TIPC_LOG = &log_buf; /* * Locking policy when using print buffers. * - * 1) Routines of the form printbuf_XXX() rely on the caller to prevent - * simultaneous use of the print buffer(s) being manipulated. - * 2) tipc_printf() uses 'print_lock' to prevent simultaneous use of - * 'print_string' and to protect its print buffer(s). - * 3) TIPC_TEE() uses 'print_lock' to protect its print buffer(s). - * 4) Routines of the form log_XXX() uses 'print_lock' to protect TIPC_LOG. + * The following routines use 'print_lock' for protection: + * 1) tipc_printf() - to protect its print buffer(s) and 'print_string' + * 2) TIPC_TEE() - to protect its print buffer(s) + * 3) tipc_dump() - to protect its print buffer(s) and 'print_string' + * 4) tipc_log_XXX() - to protect TIPC_LOG + * + * All routines of the form tipc_printbuf_XXX() rely on the caller to prevent + * simultaneous use of the print buffer(s) being manipulated. */ /** * tipc_printbuf_init - initialize print buffer to empty + * @pb: pointer to print buffer structure + * @raw: pointer to character array used by print buffer + * @size: size of character array + * + * Makes the print buffer a null device that discards anything written to it + * if the character array is too small (or absent). */ -void tipc_printbuf_init(struct print_buf *pb, char *raw, u32 sz) +void tipc_printbuf_init(struct print_buf *pb, char *raw, u32 size) { - if (!pb || !raw || (sz < (MAX_STRING + 1))) - return; - - pb->crs = pb->buf = raw; - pb->size = sz; + pb->buf = raw; + pb->crs = raw; + pb->size = size; pb->next = NULL; - pb->buf[0] = 0; - pb->buf[sz-1] = ~0; + + if (size < TIPC_PB_MIN_SIZE) { + pb->buf = NULL; + } else if (raw) { + pb->buf[0] = 0; + pb->buf[size-1] = ~0; + } } /** * tipc_printbuf_reset - reinitialize print buffer to empty state + * @pb: pointer to print buffer structure */ void tipc_printbuf_reset(struct print_buf *pb) { - if (pb && pb->buf) - tipc_printbuf_init(pb, pb->buf, pb->size); + tipc_printbuf_init(pb, pb->buf, pb->size); } /** * tipc_printbuf_empty - test if print buffer is in empty state + * @pb: pointer to print buffer structure + * + * Returns non-zero if print buffer is empty. */ int tipc_printbuf_empty(struct print_buf *pb) { - return (!pb || !pb->buf || (pb->crs == pb->buf)); + return (!pb->buf || (pb->crs == pb->buf)); } /** * tipc_printbuf_validate - check for print buffer overflow + * @pb: pointer to print buffer structure * * Verifies that a print buffer has captured all data written to it. * If data has been lost, linearize buffer and prepend an error message * - * Returns length of print buffer data string (including trailing NULL) + * Returns length of print buffer data string (including trailing NUL) */ int tipc_printbuf_validate(struct print_buf *pb) { - char *err = " *** PRINT BUFFER WRAPPED AROUND ***\n"; + char *err = "\n\n*** PRINT BUFFER OVERFLOW ***\n\n"; char *cp_buf; struct print_buf cb; - if (!pb || !pb->buf) + if (!pb->buf) return 0; - if (pb->buf[pb->size - 1] == '\0') { + if (pb->buf[pb->size - 1] == 0) { cp_buf = kmalloc(pb->size, GFP_ATOMIC); if (cp_buf != NULL){ tipc_printbuf_init(&cb, cp_buf, pb->size); @@ -141,6 +157,8 @@ int tipc_printbuf_validate(struct print_buf *pb) /** * tipc_printbuf_move - move print buffer contents to another print buffer + * @pb_to: pointer to destination print buffer structure + * @pb_from: pointer to source print buffer structure * * Current contents of destination print buffer (if any) are discarded. * Source print buffer becomes empty if a successful move occurs. @@ -152,21 +170,22 @@ void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) /* Handle the cases where contents can't be moved */ - if (!pb_to || !pb_to->buf) + if (!pb_to->buf) return; - if (!pb_from || !pb_from->buf) { + if (!pb_from->buf) { tipc_printbuf_reset(pb_to); return; } if (pb_to->size < pb_from->size) { tipc_printbuf_reset(pb_to); - tipc_printf(pb_to, "*** PRINT BUFFER OVERFLOW ***"); + tipc_printf(pb_to, "*** PRINT BUFFER MOVE ERROR ***"); return; } /* Copy data from char after cursor to end (if used) */ + len = pb_from->buf + pb_from->size - pb_from->crs - 2; if ((pb_from->buf[pb_from->size-1] == 0) && (len > 0)) { strcpy(pb_to->buf, pb_from->crs + 1); @@ -175,6 +194,7 @@ void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) pb_to->crs = pb_to->buf; /* Copy data from start to cursor (always) */ + len = pb_from->crs - pb_from->buf; strcpy(pb_to->crs, pb_from->buf); pb_to->crs += len; @@ -184,6 +204,8 @@ void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) /** * tipc_printf - append formatted output to print buffer chain + * @pb: pointer to chain of print buffers (may be NULL) + * @fmt: formatted info to be printed */ void tipc_printf(struct print_buf *pb, const char *fmt, ...) @@ -195,8 +217,8 @@ void tipc_printf(struct print_buf *pb, const char *fmt, ...) spin_lock_bh(&print_lock); FORMAT(print_string, chars_to_add, fmt); - if (chars_to_add >= MAX_STRING) - strcpy(print_string, "*** STRING TOO LONG ***"); + if (chars_to_add >= TIPC_PB_MAX_STR) + strcpy(print_string, "*** PRINT BUFFER STRING TOO LONG ***"); while (pb) { if (pb == TIPC_CONS) @@ -206,6 +228,10 @@ void tipc_printf(struct print_buf *pb, const char *fmt, ...) if (chars_to_add <= chars_left) { strcpy(pb->crs, print_string); pb->crs += chars_to_add; + } else if (chars_to_add >= (pb->size - 1)) { + strcpy(pb->buf, print_string + chars_to_add + 1 + - pb->size); + pb->crs = pb->buf + pb->size - 1; } else { strcpy(pb->buf, print_string + chars_left); save_char = print_string[chars_left]; @@ -224,6 +250,10 @@ void tipc_printf(struct print_buf *pb, const char *fmt, ...) /** * TIPC_TEE - perform next output operation on both print buffers + * @b0: pointer to chain of print buffers (may be NULL) + * @b1: pointer to print buffer to add to chain + * + * Returns pointer to print buffer chain. */ struct print_buf *TIPC_TEE(struct print_buf *b0, struct print_buf *b1) @@ -232,8 +262,6 @@ struct print_buf *TIPC_TEE(struct print_buf *b0, struct print_buf *b1) if (!b0 || (b0 == b1)) return b1; - if (!b1) - return b0; spin_lock_bh(&print_lock); while (pb->next) { @@ -256,7 +284,7 @@ static void print_to_console(char *crs, int len) int rest = len; while (rest > 0) { - int sz = rest < MAX_STRING ? rest : MAX_STRING; + int sz = rest < TIPC_PB_MAX_STR ? rest : TIPC_PB_MAX_STR; char c = crs[sz]; crs[sz] = 0; @@ -275,36 +303,48 @@ static void printbuf_dump(struct print_buf *pb) { int len; + if (!pb->buf) { + printk("*** PRINT BUFFER NOT ALLOCATED ***"); + return; + } + /* Dump print buffer from char after cursor to end (if used) */ + len = pb->buf + pb->size - pb->crs - 2; if ((pb->buf[pb->size - 1] == 0) && (len > 0)) print_to_console(pb->crs + 1, len); /* Dump print buffer from start to cursor (always) */ + len = pb->crs - pb->buf; print_to_console(pb->buf, len); } /** * tipc_dump - dump non-console print buffer(s) to console + * @pb: pointer to chain of print buffers */ void tipc_dump(struct print_buf *pb, const char *fmt, ...) { + struct print_buf *pb_next; int len; spin_lock_bh(&print_lock); - FORMAT(TIPC_CONS->buf, len, fmt); - printk(TIPC_CONS->buf); + FORMAT(print_string, len, fmt); + printk(print_string); for (; pb; pb = pb->next) { - if (pb == TIPC_CONS) - continue; - printk("\n---- Start of dump,%s log ----\n\n", - (pb == TIPC_LOG) ? "global" : "local"); - printbuf_dump(pb); - tipc_printbuf_reset(pb); - printk("\n-------- End of dump --------\n"); + if (pb != TIPC_CONS) { + printk("\n---- Start of %s log dump ----\n\n", + (pb == TIPC_LOG) ? "global" : "local"); + printbuf_dump(pb); + tipc_printbuf_reset(pb); + printk("\n---- End of dump ----\n"); + } + pb_next = pb->next; + pb->next = NULL; + pb = pb_next; } spin_unlock_bh(&print_lock); } @@ -324,7 +364,8 @@ void tipc_log_stop(void) } /** - * tipc_log_reinit - set TIPC log print buffer to specified size + * tipc_log_reinit - (re)initialize TIPC log print buffer + * @log_size: print buffer size to use */ void tipc_log_reinit(int log_size) @@ -332,10 +373,11 @@ void tipc_log_reinit(int log_size) tipc_log_stop(); if (log_size) { - if (log_size <= MAX_STRING) - log_size = MAX_STRING + 1; + if (log_size < TIPC_PB_MIN_SIZE) + log_size = TIPC_PB_MIN_SIZE; spin_lock_bh(&print_lock); - tipc_printbuf_init(TIPC_LOG, kmalloc(log_size, GFP_ATOMIC), log_size); + tipc_printbuf_init(TIPC_LOG, kmalloc(log_size, GFP_ATOMIC), + log_size); spin_unlock_bh(&print_lock); } } diff --git a/net/tipc/dbg.h b/net/tipc/dbg.h index 227f050..467c0bc7 100644 --- a/net/tipc/dbg.h +++ b/net/tipc/dbg.h @@ -2,7 +2,7 @@ * net/tipc/dbg.h: Include file for TIPC print buffer routines * * Copyright (c) 1997-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005-2006, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -37,6 +37,14 @@ #ifndef _TIPC_DBG_H #define _TIPC_DBG_H +/** + * struct print_buf - TIPC print buffer structure + * @buf: pointer to character array containing print buffer contents + * @size: size of character array + * @crs: pointer to first unused space in character array (i.e. final NUL) + * @next: used to link print buffers when printing to more than one at a time + */ + struct print_buf { char *buf; u32 size; @@ -44,7 +52,10 @@ struct print_buf { struct print_buf *next; }; -void tipc_printbuf_init(struct print_buf *pb, char *buf, u32 sz); +#define TIPC_PB_MIN_SIZE 64 /* minimum size for a print buffer's array */ +#define TIPC_PB_MAX_STR 512 /* max printable string (with trailing NUL) */ + +void tipc_printbuf_init(struct print_buf *pb, char *buf, u32 size); void tipc_printbuf_reset(struct print_buf *pb); int tipc_printbuf_empty(struct print_buf *pb); int tipc_printbuf_validate(struct print_buf *pb); -- cgit v0.10.2 From eb5959c2bd290bf6c24ddf6d1f5ebcb496c54adb Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:43:54 -0700 Subject: [TIPC]: Stream socket can now send > 66000 bytes at a time The stream socket send code was not initializing some required fields of the temporary msghdr structure it was utilizing; this is now fixed. A check has also been added to detect if a user illegally specifies a destination address when sending on an established stream connection. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/socket.c b/net/tipc/socket.c index acfb852..ed900fe 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -2,7 +2,7 @@ * net/tipc/socket.c: TIPC socket API * * Copyright (c) 2001-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2006, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -629,6 +629,9 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, return -ENOTCONN; } + if (unlikely(m->msg_name)) + return -EISCONN; + /* * Send each iovec entry using one or more messages * @@ -641,6 +644,8 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, curr_iovlen = m->msg_iovlen; my_msg.msg_iov = &my_iov; my_msg.msg_iovlen = 1; + my_msg.msg_flags = m->msg_flags; + my_msg.msg_name = NULL; bytes_sent = 0; while (curr_iovlen--) { -- cgit v0.10.2 From e91ed0bcdfc4812c0342d64281ee985213df07c3 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:44:59 -0700 Subject: [TIPC]: Added duplicate node address detection capability TIPC now rejects and logs link setup requests from node if the receiving node already has a functional link to that node on the associated interface, or if the requestor is using the same as the receiver. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 75a5968..39744a3 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -2,7 +2,7 @@ * net/tipc/bearer.c: TIPC bearer code * * Copyright (c) 1996-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2006, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -191,14 +191,14 @@ void tipc_media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a) if ((i < media_count) && (m_ptr->addr2str != NULL)) { char addr_str[MAX_ADDR_STR]; - tipc_printf(pb, "%s(%s) ", m_ptr->name, + tipc_printf(pb, "%s(%s)", m_ptr->name, m_ptr->addr2str(a, addr_str, sizeof(addr_str))); } else { unchar *addr = (unchar *)&a->dev_addr; - tipc_printf(pb, "UNKNOWN(%u):", media_type); + tipc_printf(pb, "UNKNOWN(%u)", media_type); for (i = 0; i < (sizeof(*a) - sizeof(a->type)); i++) { - tipc_printf(pb, "%02x ", addr[i]); + tipc_printf(pb, "-%02x", addr[i]); } } } diff --git a/net/tipc/discover.c b/net/tipc/discover.c index ee94de9..cc748d4 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -132,6 +132,28 @@ static struct sk_buff *tipc_disc_init_msg(u32 type, } /** + * disc_dupl_alert - issue node address duplication alert + * @b_ptr: pointer to bearer detecting duplication + * @node_addr: duplicated node address + * @media_addr: media address advertised by duplicated node + */ + +static void disc_dupl_alert(struct bearer *b_ptr, u32 node_addr, + struct tipc_media_addr *media_addr) +{ + char node_addr_str[16]; + char media_addr_str[64]; + struct print_buf pb; + + addr_string_fill(node_addr_str, node_addr); + tipc_printbuf_init(&pb, media_addr_str, sizeof(media_addr_str)); + tipc_media_addr_printf(&pb, media_addr); + tipc_printbuf_validate(&pb); + warn("Duplicate %s using %s seen on <%s>\n", + node_addr_str, media_addr_str, b_ptr->publ.name); +} + +/** * tipc_disc_recv_msg - handle incoming link setup message (request or response) * @buf: buffer containing message */ @@ -157,8 +179,11 @@ void tipc_disc_recv_msg(struct sk_buff *buf) return; if (!tipc_addr_node_valid(orig)) return; - if (orig == tipc_own_addr) + if (orig == tipc_own_addr) { + if (memcmp(&media_addr, &b_ptr->publ.addr, sizeof(media_addr))) + disc_dupl_alert(b_ptr, tipc_own_addr, &media_addr); return; + } if (!in_scope(dest, tipc_own_addr)) return; if (is_slave(tipc_own_addr) && is_slave(orig)) @@ -190,6 +215,11 @@ void tipc_disc_recv_msg(struct sk_buff *buf) } addr = &link->media_addr; if (memcmp(addr, &media_addr, sizeof(*addr))) { + if (tipc_link_is_up(link) || (!link->started)) { + disc_dupl_alert(b_ptr, orig, &media_addr); + spin_unlock_bh(&n_ptr->lock); + return; + } warn("Resetting link <%s>, peer interface address changed\n", link->name); memcpy(addr, &media_addr, sizeof(*addr)); -- cgit v0.10.2 From cfb0c0890b200364c0886c0d9f0dc615f8114c43 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:47:18 -0700 Subject: [TIPC]: Optimize wakeup logic when socket has no waiting processes This patch adds a simple test so TIPC doesn't try waking up processes waiting on a socket if there are none waiting. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/socket.c b/net/tipc/socket.c index ed900fe..2a6a5a6 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1208,7 +1208,8 @@ static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) atomic_inc(&tipc_queue_size); skb_queue_tail(&sock->sk->sk_receive_queue, buf); - wake_up_interruptible(sock->sk->sk_sleep); + if (waitqueue_active(sock->sk->sk_sleep)) + wake_up_interruptible(sock->sk->sk_sleep); return TIPC_OK; } @@ -1223,7 +1224,8 @@ static void wakeupdispatch(struct tipc_port *tport) { struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; - wake_up_interruptible(tsock->sk.sk_sleep); + if (waitqueue_active(tsock->sk.sk_sleep)) + wake_up_interruptible(tsock->sk.sk_sleep); } /** -- cgit v0.10.2 From a3df92c73b92970dc4211189b87eb4cf874f5685 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:49:03 -0700 Subject: [TIPC]: Remove code bloat introduced by print buffer rework This patch allows the compiler to optimize out any code that tries to send debugging output to the null print buffer (TIPC_NULL), a capability that was unintentionally broken during the recent print buffer rework. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/core.h b/net/tipc/core.h index 47504ec..d1c3948 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -83,9 +83,9 @@ void tipc_dump(struct print_buf*,const char *fmt, ...); #define warn(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_WARNING "TIPC: " fmt, ## arg) #define info(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_NOTICE "TIPC: " fmt, ## arg) -#define dbg(fmt, arg...) do {if (DBG_OUTPUT) tipc_printf(DBG_OUTPUT, fmt, ## arg);} while(0) -#define msg_dbg(msg, txt) do {if (DBG_OUTPUT) tipc_msg_print(DBG_OUTPUT, msg, txt);} while(0) -#define dump(fmt, arg...) do {if (DBG_OUTPUT) tipc_dump(DBG_OUTPUT, fmt, ##arg);} while(0) +#define dbg(fmt, arg...) do {if (DBG_OUTPUT != TIPC_NULL) tipc_printf(DBG_OUTPUT, fmt, ## arg);} while(0) +#define msg_dbg(msg, txt) do {if (DBG_OUTPUT != TIPC_NULL) tipc_msg_print(DBG_OUTPUT, msg, txt);} while(0) +#define dump(fmt, arg...) do {if (DBG_OUTPUT != TIPC_NULL) tipc_dump(DBG_OUTPUT, fmt, ##arg);} while(0) /* diff --git a/net/tipc/link.c b/net/tipc/link.c index 53bc8cb..1bb983c 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -132,7 +132,7 @@ static void link_print(struct link *l_ptr, struct print_buf *buf, * allow the output from multiple links to be intermixed. For this reason * routines of the form "dbg_link_XXX()" have been created that will capture * debug info into a link's personal print buffer, which can then be dumped - * into the TIPC system log (LOG) upon request. + * into the TIPC system log (TIPC_LOG) upon request. * * To enable per-link debugging, use LINK_LOG_BUF_SIZE to specify the size * of the print buffer used by each link. If LINK_LOG_BUF_SIZE is set to 0, @@ -141,7 +141,7 @@ static void link_print(struct link *l_ptr, struct print_buf *buf, * when there is only a single link in the system being debugged. * * Notes: - * - When enabled, LINK_LOG_BUF_SIZE should be set to at least 1000 (bytes) + * - When enabled, LINK_LOG_BUF_SIZE should be set to at least TIPC_PB_MIN_SIZE * - "l_ptr" must be valid when using dbg_link_XXX() macros */ @@ -159,13 +159,13 @@ static void link_print(struct link *l_ptr, struct print_buf *buf, static void dbg_print_link(struct link *l_ptr, const char *str) { - if (DBG_OUTPUT) + if (DBG_OUTPUT != TIPC_NULL) link_print(l_ptr, DBG_OUTPUT, str); } static void dbg_print_buf_chain(struct sk_buff *root_buf) { - if (DBG_OUTPUT) { + if (DBG_OUTPUT != TIPC_NULL) { struct sk_buff *buf = root_buf; while (buf) { -- cgit v0.10.2 From 3a8d12142eab420ffcbbf3d1d2e637158e85aab8 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:50:20 -0700 Subject: [TIPC]: Add support for Ethernet VLANs This patch enhances TIPC's Ethernet support to include VLAN interfaces. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/core.c b/net/tipc/core.c index 0539a83..7f3f693 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -90,7 +90,7 @@ int tipc_random; atomic_t tipc_user_count = ATOMIC_INIT(0); const char tipc_alphabet[] = - "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_"; + "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_."; /* configurable TIPC parameters */ diff --git a/net/tipc/core.h b/net/tipc/core.h index d1c3948..4638947c 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -275,11 +275,15 @@ static inline void k_term_timer(struct timer_list *timer) /* * TIPC message buffer code * - * TIPC message buffer headroom leaves room for 14 byte Ethernet header, + * TIPC message buffer headroom reserves space for a link-level header + * (in case the message is sent off-node), * while ensuring TIPC header is word aligned for quicker access + * + * The largest header currently supported is 18 bytes, which is used when + * the standard 14 byte Ethernet header has 4 added bytes for VLAN info */ -#define BUF_HEADROOM 16u +#define BUF_HEADROOM 20u struct tipc_skb_cb { void *handle; -- cgit v0.10.2 From 08c31f7167b1bdc30cd0960b45d6f3076eb1f179 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:56:04 -0700 Subject: [TIPC]: Name publication events now delivered in chronological order This patch tivially re-orders the entries in TIPC's list of local publications so that applications will receive publication events in the order they were published. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index f0b063b..03bd659 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -122,7 +122,7 @@ void tipc_named_publish(struct publication *publ) struct sk_buff *buf; struct distr_item *item; - list_add(&publ->local_list, &publ_root); + list_add_tail(&publ->local_list, &publ_root); publ_cnt++; buf = named_prepare_buf(PUBLICATION, ITEM_SIZE, 0); -- cgit v0.10.2 From 2de07f6156fe664063207c010b3bd2500348884a Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:57:13 -0700 Subject: [TIPC]: Fixed slow link reactivation when link tolerance is large This patch corrects an issue wherein a previouly failed node could not reestablish a links to a non-failing node in the TIPC network until the latter node detected the link failure itself (which might be configured to take up to 30 seconds). The non-failing node now responds to link setup requests from a previously failed node in at most 1 second, allowing it to detect the link failure more quickly. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/discover.c b/net/tipc/discover.c index cc748d4..3b0cd12 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -195,7 +195,8 @@ void tipc_disc_recv_msg(struct sk_buff *buf) struct sk_buff *rbuf; struct tipc_media_addr *addr; struct node *n_ptr = tipc_node_find(orig); - int link_up; + int link_fully_up; + dbg(" in own cluster\n"); if (n_ptr == NULL) { n_ptr = tipc_node_create(orig); @@ -225,9 +226,9 @@ void tipc_disc_recv_msg(struct sk_buff *buf) memcpy(addr, &media_addr, sizeof(*addr)); tipc_link_reset(link); } - link_up = tipc_link_is_up(link); + link_fully_up = (link->state == WORKING_WORKING); spin_unlock_bh(&n_ptr->lock); - if ((type == DSC_RESP_MSG) || link_up) + if ((type == DSC_RESP_MSG) || link_fully_up) return; rbuf = tipc_disc_init_msg(DSC_RESP_MSG, 1, orig, b_ptr); if (rbuf != NULL) { -- cgit v0.10.2 From fc144deec6403c17e6d3f6a6574f701420f166ed Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 21:57:56 -0700 Subject: [TIPC]: Can now list multicast link on an isolated network node This patch fixes a minor bug that prevents "tipc-config -l" from displaying the multicast link if a TIPC node has never successfully established at least one unicast link. Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/node.c b/net/tipc/node.c index fc6d096..886bda5 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -648,7 +648,7 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); - if (!tipc_nodes) + if (tipc_mode != TIPC_NET_MODE) return tipc_cfg_reply_none(); /* Get space for all unicast links + multicast link */ -- cgit v0.10.2 From eb409460b1abec0e2a1f9c9d07019f4157a6d6bc Mon Sep 17 00:00:00 2001 From: Lijun Chen Date: Mon, 16 Oct 2006 21:59:42 -0700 Subject: [TIPC]: Added subscription cancellation capability This patch allows a TIPC application to cancel an existing topology service subscription by re-requesting the subscription with the TIPC_SUB_CANCEL filter bit set. (All other bits of the cancel request must match the original subscription request.) Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/include/linux/tipc.h b/include/linux/tipc.h index 243a15f..bea4694 100644 --- a/include/linux/tipc.h +++ b/include/linux/tipc.h @@ -129,6 +129,7 @@ static inline unsigned int tipc_node(__u32 addr) #define TIPC_SUB_PORTS 0x01 /* filter for port availability */ #define TIPC_SUB_SERVICE 0x02 /* filter for service availability */ +#define TIPC_SUB_CANCEL 0x04 /* cancel a subscription */ #if 0 /* The following filter options are not currently implemented */ #define TIPC_SUB_NO_BIND_EVTS 0x04 /* filter out "publish" events */ diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index c51600b..7a918f1 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -155,7 +155,7 @@ void tipc_subscr_report_overlap(struct subscription *sub, sub->seq.upper, found_lower, found_upper); if (!tipc_subscr_overlap(sub, found_lower, found_upper)) return; - if (!must && (sub->filter != TIPC_SUB_PORTS)) + if (!must && !(sub->filter & TIPC_SUB_PORTS)) return; subscr_send_event(sub, found_lower, found_upper, event, port_ref, node); } @@ -176,6 +176,13 @@ static void subscr_timeout(struct subscription *sub) if (subscriber == NULL) return; + /* Validate timeout (in case subscription is being cancelled) */ + + if (sub->timeout == TIPC_WAIT_FOREVER) { + tipc_ref_unlock(subscriber_ref); + return; + } + /* Unlink subscription from name table */ tipc_nametbl_unsubscribe(sub); @@ -199,6 +206,20 @@ static void subscr_timeout(struct subscription *sub) } /** + * subscr_del - delete a subscription within a subscription list + * + * Called with subscriber locked. + */ + +static void subscr_del(struct subscription *sub) +{ + tipc_nametbl_unsubscribe(sub); + list_del(&sub->subscription_list); + kfree(sub); + atomic_dec(&topsrv.subscription_count); +} + +/** * subscr_terminate - terminate communication with a subscriber * * Called with subscriber locked. Routine must temporarily release this lock @@ -227,12 +248,9 @@ static void subscr_terminate(struct subscriber *subscriber) k_cancel_timer(&sub->timer); k_term_timer(&sub->timer); } - tipc_nametbl_unsubscribe(sub); - list_del(&sub->subscription_list); - dbg("Term: Removed sub %u,%u,%u from subscriber %x list\n", + dbg("Term: Removing sub %u,%u,%u from subscriber %x list\n", sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); - kfree(sub); - atomic_dec(&topsrv.subscription_count); + subscr_del(sub); } /* Sever connection to subscriber */ @@ -253,6 +271,49 @@ static void subscr_terminate(struct subscriber *subscriber) } /** + * subscr_cancel - handle subscription cancellation request + * + * Called with subscriber locked. Routine must temporarily release this lock + * to enable the subscription timeout routine to finish without deadlocking; + * the lock is then reclaimed to allow caller to release it upon return. + * + * Note that fields of 's' use subscriber's endianness! + */ + +static void subscr_cancel(struct tipc_subscr *s, + struct subscriber *subscriber) +{ + struct subscription *sub; + struct subscription *sub_temp; + int found = 0; + + /* Find first matching subscription, exit if not found */ + + list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, + subscription_list) { + if (!memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) { + found = 1; + break; + } + } + if (!found) + return; + + /* Cancel subscription timer (if used), then delete subscription */ + + if (sub->timeout != TIPC_WAIT_FOREVER) { + sub->timeout = TIPC_WAIT_FOREVER; + spin_unlock_bh(subscriber->lock); + k_cancel_timer(&sub->timer); + k_term_timer(&sub->timer); + spin_lock_bh(subscriber->lock); + } + dbg("Cancel: removing sub %u,%u,%u from subscriber %x list\n", + sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); + subscr_del(sub); +} + +/** * subscr_subscribe - create subscription for subscriber * * Called with subscriber locked @@ -263,6 +324,21 @@ static void subscr_subscribe(struct tipc_subscr *s, { struct subscription *sub; + /* Determine/update subscriber's endianness */ + + if (s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE)) + subscriber->swap = 0; + else + subscriber->swap = 1; + + /* Detect & process a subscription cancellation request */ + + if (s->filter & htohl(TIPC_SUB_CANCEL, subscriber->swap)) { + s->filter &= ~htohl(TIPC_SUB_CANCEL, subscriber->swap); + subscr_cancel(s, subscriber); + return; + } + /* Refuse subscription if global limit exceeded */ if (atomic_read(&topsrv.subscription_count) >= tipc_max_subscriptions) { @@ -281,13 +357,6 @@ static void subscr_subscribe(struct tipc_subscr *s, return; } - /* Determine/update subscriber's endianness */ - - if ((s->filter == TIPC_SUB_PORTS) || (s->filter == TIPC_SUB_SERVICE)) - subscriber->swap = 0; - else - subscriber->swap = 1; - /* Initialize subscription object */ memset(sub, 0, sizeof(*sub)); @@ -296,8 +365,8 @@ static void subscr_subscribe(struct tipc_subscr *s, sub->seq.upper = htohl(s->seq.upper, subscriber->swap); sub->timeout = htohl(s->timeout, subscriber->swap); sub->filter = htohl(s->filter, subscriber->swap); - if ((((sub->filter != TIPC_SUB_PORTS) - && (sub->filter != TIPC_SUB_SERVICE))) + if ((!(sub->filter & TIPC_SUB_PORTS) + == !(sub->filter & TIPC_SUB_SERVICE)) || (sub->seq.lower > sub->seq.upper)) { warn("Subscription rejected, illegal request\n"); kfree(sub); -- cgit v0.10.2 From 53cfd1e102c759c958f907ee40a58bec3fc5911a Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 22:00:56 -0700 Subject: [TIPC]: Unrecognized configuration command now returns error message This patch causes TIPC to return an error message when it receives an unrecognized configuration command. (Previously, the sender received no feedback.) Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/config.c b/net/tipc/config.c index 285e1bc..ed1351e 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -2,7 +2,7 @@ * net/tipc/config.c: TIPC configuration management code * * Copyright (c) 2002-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2006, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -613,7 +613,8 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id); break; default: - rep_tlv_buf = NULL; + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (unknown command)"); break; } -- cgit v0.10.2 From 665d8669583e28c397d4333385d8f46ca5864048 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Mon, 16 Oct 2006 22:01:32 -0700 Subject: [TIPC]: Updated TIPC version number to 1.6.2 Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller diff --git a/net/tipc/core.c b/net/tipc/core.c index 7f3f693..6f5b7ee 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -57,7 +57,7 @@ void tipc_socket_stop(void); int tipc_netlink_start(void); void tipc_netlink_stop(void); -#define TIPC_MOD_VER "1.6.1" +#define TIPC_MOD_VER "1.6.2" #ifndef CONFIG_TIPC_ZONES #define CONFIG_TIPC_ZONES 3 -- cgit v0.10.2 From e320af1df4c47305e829e8e1a40e5fad0e5e9fba Mon Sep 17 00:00:00 2001 From: Ville Nuorvala Date: Mon, 16 Oct 2006 22:05:55 -0700 Subject: [IPV6]: Remove struct pol_chain. Struct pol_chain has existed since at least the 2.2 kernel, but isn't used anymore. As the IPv6 policy routing is implemented in a totally different way in the current kernel, just get rid of it. Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 6ca6b71..c14b70e 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -36,13 +36,6 @@ struct route_info { #define RT6_LOOKUP_F_REACHABLE 0x2 #define RT6_LOOKUP_F_HAS_SADDR 0x4 -struct pol_chain { - int type; - int priority; - struct fib6_node *rules; - struct pol_chain *next; -}; - extern struct rt6_info ip6_null_entry; #ifdef CONFIG_IPV6_MULTIPLE_TABLES -- cgit v0.10.2 From 23c435f7ff884caded4a1391ba2b308d465423c0 Mon Sep 17 00:00:00 2001 From: Ville Nuorvala Date: Mon, 16 Oct 2006 22:08:28 -0700 Subject: [SCTP]: Fix minor typo Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 9deec43..9f34dec 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -821,7 +821,7 @@ out: * addrs is a pointer to an array of one or more socket addresses. Each * address is contained in its appropriate structure (i.e. struct * sockaddr_in or struct sockaddr_in6) the family of the address type - * must be used to distengish the address length (note that this + * must be used to distinguish the address length (note that this * representation is termed a "packed array" of addresses). The caller * specifies the number of addresses in the array with addrcnt. * -- cgit v0.10.2 From 4251320fa2ef93207fbefeb2eda2d265b84fc116 Mon Sep 17 00:00:00 2001 From: Ville Nuorvala Date: Mon, 16 Oct 2006 22:10:05 -0700 Subject: [IPV6]: Make sure error handling is done when calling ip6_route_output(). As ip6_route_output() never returns NULL, error checking must be done by looking at dst->error in stead of comparing dst against NULL. Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 73cee2e..d400f8f 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -25,12 +25,14 @@ static struct dst_ops xfrm6_dst_ops; static struct xfrm_policy_afinfo xfrm6_policy_afinfo; -static int xfrm6_dst_lookup(struct xfrm_dst **dst, struct flowi *fl) +static int xfrm6_dst_lookup(struct xfrm_dst **xdst, struct flowi *fl) { - int err = 0; - *dst = (struct xfrm_dst*)ip6_route_output(NULL, fl); - if (!*dst) - err = -ENETUNREACH; + struct dst_entry *dst = ip6_route_output(NULL, fl); + int err = dst->error; + if (!err) + *xdst = (struct xfrm_dst *) dst; + else + dst_release(dst); return err; } diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 249e503..78071c6 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -215,17 +215,17 @@ static struct dst_entry *sctp_v6_get_dst(struct sctp_association *asoc, } dst = ip6_route_output(NULL, &fl); - if (dst) { + if (!dst->error) { struct rt6_info *rt; rt = (struct rt6_info *)dst; SCTP_DEBUG_PRINTK( "rt6_dst:" NIP6_FMT " rt6_src:" NIP6_FMT "\n", NIP6(rt->rt6i_dst.addr), NIP6(rt->rt6i_src.addr)); - } else { - SCTP_DEBUG_PRINTK("NO ROUTE\n"); + return dst; } - - return dst; + SCTP_DEBUG_PRINTK("NO ROUTE\n"); + dst_release(dst); + return NULL; } /* Returns the number of consecutive initial bits that match in the 2 ipv6 -- cgit v0.10.2 From e0eda7bbaae48ea56f1f1943ea90df72d459b9b0 Mon Sep 17 00:00:00 2001 From: Ville Nuorvala Date: Mon, 16 Oct 2006 22:11:11 -0700 Subject: [IPV6]: Clean up BACKTRACK(). The fn check is unnecessary as fn can never be NULL in BACKTRACK(). Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a1b0f07..263c057 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -484,7 +484,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, do { \ if (rt == &ip6_null_entry) { \ struct fib6_node *pn; \ - while (fn) { \ + while (1) { \ if (fn->fn_flags & RTN_TL_ROOT) \ goto out; \ pn = fn->parent; \ -- cgit v0.10.2 From 264e91b68aff1e39b558123498f28bf5aa68b4ee Mon Sep 17 00:00:00 2001 From: Ville Nuorvala Date: Mon, 16 Oct 2006 22:12:21 -0700 Subject: [IPV6]: Make IPV6_SUBTREES depend on IPV6_MULTIPLE_TABLES. As IPV6_SUBTREES can't work without IPV6_MULTIPLE_TABLES have IPV6_SUBTREES depend on it. Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index ef5eaad..6e48f52 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -175,9 +175,16 @@ config IPV6_TUNNEL If unsure, say N. +config IPV6_MULTIPLE_TABLES + bool "IPv6: Multiple Routing Tables" + depends on IPV6 && EXPERIMENTAL + select FIB_RULES + ---help--- + Support multiple routing tables. + config IPV6_SUBTREES bool "IPv6: source address based routing" - depends on IPV6 && EXPERIMENTAL + depends on IPV6_MULTIPLE_TABLES ---help--- Enable routing by source address or prefix. @@ -189,13 +196,6 @@ config IPV6_SUBTREES If unsure, say N. -config IPV6_MULTIPLE_TABLES - bool "IPv6: Multiple Routing Tables" - depends on IPV6 && EXPERIMENTAL - select FIB_RULES - ---help--- - Support multiple routing tables. - config IPV6_ROUTE_FWMARK bool "IPv6: use netfilter MARK value as routing key" depends on IPV6_MULTIPLE_TABLES && NETFILTER -- cgit v0.10.2 From 22e1e4d8dcb71762fcbe0c73d720aea8bb2013af Mon Sep 17 00:00:00 2001 From: Ville Nuorvala Date: Mon, 16 Oct 2006 22:14:26 -0700 Subject: [IPV6]: Always copy rt->u.dst.error when copying a rt6_info. Signed-off-by: Ville Nuorvala Signed-off-by: David S. Miller diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 263c057..aa96be8 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -618,8 +618,6 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d ipv6_addr_copy(&rt->rt6i_dst.addr, daddr); rt->rt6i_dst.plen = 128; rt->rt6i_flags |= RTF_CACHE; - if (rt->rt6i_flags & RTF_REJECT) - rt->u.dst.error = ort->u.dst.error; rt->u.dst.flags |= DST_HOST; rt->rt6i_nexthop = neigh_clone(ort->rt6i_nexthop); } @@ -1540,6 +1538,7 @@ static struct rt6_info * ip6_rt_copy(struct rt6_info *ort) rt->u.dst.output = ort->u.dst.output; memcpy(rt->u.dst.metrics, ort->u.dst.metrics, RTAX_MAX*sizeof(u32)); + rt->u.dst.error = ort->u.dst.error; rt->u.dst.dev = ort->u.dst.dev; if (rt->u.dst.dev) dev_hold(rt->u.dst.dev); -- cgit v0.10.2 From b52f070c9c3c09ed3b7f699280193aae7e25d816 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 18 Oct 2006 20:26:36 -0700 Subject: [IPv4] fib: Remove unused fib_config members Signed-off-by: Thomas Graf Signed-off-by: David S. Miller diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 8222914..949b932 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -21,17 +21,14 @@ #include struct fib_config { - u8 fc_family; u8 fc_dst_len; - u8 fc_src_len; u8 fc_tos; u8 fc_protocol; u8 fc_scope; u8 fc_type; - /* 1 byte unused */ + /* 3 bytes unused */ u32 fc_table; __be32 fc_dst; - __be32 fc_src; __be32 fc_gw; int fc_oif; u32 fc_flags; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 9c399a7..af0190d 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -482,9 +482,7 @@ static int rtm_to_fib_config(struct sk_buff *skb, struct nlmsghdr *nlh, memset(cfg, 0, sizeof(*cfg)); rtm = nlmsg_data(nlh); - cfg->fc_family = rtm->rtm_family; cfg->fc_dst_len = rtm->rtm_dst_len; - cfg->fc_src_len = rtm->rtm_src_len; cfg->fc_tos = rtm->rtm_tos; cfg->fc_table = rtm->rtm_table; cfg->fc_protocol = rtm->rtm_protocol; @@ -501,9 +499,6 @@ static int rtm_to_fib_config(struct sk_buff *skb, struct nlmsghdr *nlh, case RTA_DST: cfg->fc_dst = nla_get_be32(attr); break; - case RTA_SRC: - cfg->fc_src = nla_get_be32(attr); - break; case RTA_OIF: cfg->fc_oif = nla_get_u32(attr); break; -- cgit v0.10.2 From ae8064ac32d07f609114d73928cdef803be87134 Mon Sep 17 00:00:00 2001 From: John Heffner Date: Wed, 18 Oct 2006 20:36:48 -0700 Subject: [TCP]: Bound TSO defer time This patch limits the amount of time you will defer sending a TSO segment to less than two clock ticks, or the time between two acks, whichever is longer. On slow links, deferring causes significant bursts. See attached plots, which show RTT through a 1 Mbps link with a 100 ms RTT and ~100 ms queue for (a) non-TSO, (b) currnet TSO, and (c) patched TSO. This burstiness causes significant jitter, tends to overflow queues early (bad for short queues), and makes delay-based congestion control more difficult. Deferring by a couple clock ticks I believe will have a relatively small impact on performance. Signed-off-by: John Heffner Signed-off-by: David S. Miller diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 0e058a2..2d36f6d 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -342,6 +342,8 @@ struct tcp_sock { unsigned long last_synq_overflow; + __u32 tso_deferred; + /* Receiver side RTT estimation */ struct { __u32 rtt; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f22536e..ca40615 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1096,10 +1096,14 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ u32 send_win, cong_win, limit, in_flight; if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) - return 0; + goto send_now; if (icsk->icsk_ca_state != TCP_CA_Open) - return 0; + goto send_now; + + /* Defer for less than two clock ticks. */ + if (!tp->tso_deferred && ((jiffies<<1)>>1) - (tp->tso_deferred>>1) > 1) + goto send_now; in_flight = tcp_packets_in_flight(tp); @@ -1115,7 +1119,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ /* If a full-sized TSO skb can be sent, do it. */ if (limit >= 65536) - return 0; + goto send_now; if (sysctl_tcp_tso_win_divisor) { u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache); @@ -1125,7 +1129,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ */ chunk /= sysctl_tcp_tso_win_divisor; if (limit >= chunk) - return 0; + goto send_now; } else { /* Different approach, try not to defer past a single * ACK. Receiver should ACK every other full sized @@ -1133,11 +1137,17 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ * then send now. */ if (limit > tcp_max_burst(tp) * tp->mss_cache) - return 0; + goto send_now; } /* Ok, it looks like it is advisable to defer. */ + tp->tso_deferred = 1 | (jiffies<<1); + return 1; + +send_now: + tp->tso_deferred = 0; + return 0; } /* Create a new MTU probe if we are ready. -- cgit v0.10.2 From 3a31b9d2f2d98667832e87fa1853754ff062d6ef Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 18 Oct 2006 20:45:22 -0700 Subject: [DECNET]: Fix input routing bug This patch fixes a silly bug that has been in the input routing code for some time. It results in trying to send to a node directly when the origin of the packet is via the default router. Its been tested by Alan Kemmerer who reported the bug and its a fairly obvious fix for a typo. Signed-off-by: Steven Whitehouse Signed-off-by: Patrick Caulfield Signed-off-by: David S. Miller diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 491429c..23489f7 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1275,7 +1275,6 @@ static int dn_route_input_slow(struct sk_buff *skb) goto e_inval; res.type = RTN_LOCAL; - flags |= RTCF_DIRECTSRC; } else { __le16 src_map = fl.fld_src; free_res = 1; @@ -1346,7 +1345,7 @@ static int dn_route_input_slow(struct sk_buff *skb) goto make_route; /* Packet was intra-ethernet, so we know its on-link */ - if (cb->rt_flags | DN_RT_F_IE) { + if (cb->rt_flags & DN_RT_F_IE) { gateway = cb->src; flags |= RTCF_DIRECTSRC; goto make_route; -- cgit v0.10.2 From 9ce8ade015a3f82dbdf856df7a685878dd1cc0e1 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 18 Oct 2006 20:46:54 -0700 Subject: [IPv6] route: Fix prohibit and blackhole routing decision Lookups resolving to ip6_blk_hole_entry must result in silently discarding the packets whereas an ip6_pkt_prohibit_entry is supposed to cause an ICMPV6_ADM_PROHIBITED message to be sent. Thanks to Kim Nordlund for noticing this bug. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller diff --git a/net/ipv6/route.c b/net/ipv6/route.c index aa96be8..5c00ca4 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -94,6 +94,9 @@ static int ip6_dst_gc(void); static int ip6_pkt_discard(struct sk_buff *skb); static int ip6_pkt_discard_out(struct sk_buff *skb); +static int ip6_pkt_prohibit(struct sk_buff *skb); +static int ip6_pkt_prohibit_out(struct sk_buff *skb); +static int ip6_pkt_blk_hole(struct sk_buff *skb); static void ip6_link_failure(struct sk_buff *skb); static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu); @@ -150,8 +153,8 @@ struct rt6_info ip6_prohibit_entry = { .obsolete = -1, .error = -EACCES, .metrics = { [RTAX_HOPLIMIT - 1] = 255, }, - .input = ip6_pkt_discard, - .output = ip6_pkt_discard_out, + .input = ip6_pkt_prohibit, + .output = ip6_pkt_prohibit_out, .ops = &ip6_dst_ops, .path = (struct dst_entry*)&ip6_prohibit_entry, } @@ -170,8 +173,8 @@ struct rt6_info ip6_blk_hole_entry = { .obsolete = -1, .error = -EINVAL, .metrics = { [RTAX_HOPLIMIT - 1] = 255, }, - .input = ip6_pkt_discard, - .output = ip6_pkt_discard_out, + .input = ip6_pkt_blk_hole, + .output = ip6_pkt_blk_hole, .ops = &ip6_dst_ops, .path = (struct dst_entry*)&ip6_blk_hole_entry, } @@ -1742,24 +1745,46 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg) * Drop the packet on the floor */ -static int ip6_pkt_discard(struct sk_buff *skb) +static inline int ip6_pkt_drop(struct sk_buff *skb, int code) { int type = ipv6_addr_type(&skb->nh.ipv6h->daddr); if (type == IPV6_ADDR_ANY || type == IPV6_ADDR_RESERVED) IP6_INC_STATS(IPSTATS_MIB_INADDRERRORS); IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_NOROUTE, 0, skb->dev); + icmpv6_send(skb, ICMPV6_DEST_UNREACH, code, 0, skb->dev); kfree_skb(skb); return 0; } +static int ip6_pkt_discard(struct sk_buff *skb) +{ + return ip6_pkt_drop(skb, ICMPV6_NOROUTE); +} + static int ip6_pkt_discard_out(struct sk_buff *skb) { skb->dev = skb->dst->dev; return ip6_pkt_discard(skb); } +static int ip6_pkt_prohibit(struct sk_buff *skb) +{ + return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED); +} + +static int ip6_pkt_prohibit_out(struct sk_buff *skb) +{ + skb->dev = skb->dst->dev; + return ip6_pkt_prohibit(skb); +} + +static int ip6_pkt_blk_hole(struct sk_buff *skb) +{ + kfree_skb(skb); + return 0; +} + /* * Allocate a dst for local (unicast / anycast) address. */ -- cgit v0.10.2 From 5175c3786c244f8b689854db24c9e79b1c6a084f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 18 Oct 2006 20:51:57 -0700 Subject: [NET]: reduce per cpu ram used for loopback stats We dont need a full struct net_device_stats (currently 23 long : 184 bytes on x86_64) per possible CPU, but only two counters : bytes and packets We save few CPU cycles too in loopback_xmit() not updating 4 fields, but 2. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index 4178b4b1..93fbea1 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -58,7 +58,11 @@ #include #include -static DEFINE_PER_CPU(struct net_device_stats, loopback_stats); +struct pcpu_lstats { + unsigned long packets; + unsigned long bytes; +}; +static DEFINE_PER_CPU(struct pcpu_lstats, pcpu_lstats); #define LOOPBACK_OVERHEAD (128 + MAX_HEADER + 16 + 16) @@ -128,7 +132,7 @@ static void emulate_large_send_offload(struct sk_buff *skb) */ static int loopback_xmit(struct sk_buff *skb, struct net_device *dev) { - struct net_device_stats *lb_stats; + struct pcpu_lstats *lb_stats; skb_orphan(skb); @@ -149,11 +153,9 @@ static int loopback_xmit(struct sk_buff *skb, struct net_device *dev) #endif dev->last_rx = jiffies; - lb_stats = &per_cpu(loopback_stats, get_cpu()); - lb_stats->rx_bytes += skb->len; - lb_stats->tx_bytes = lb_stats->rx_bytes; - lb_stats->rx_packets++; - lb_stats->tx_packets = lb_stats->rx_packets; + lb_stats = &per_cpu(pcpu_lstats, get_cpu()); + lb_stats->bytes += skb->len; + lb_stats->packets++; put_cpu(); netif_rx(skb); @@ -166,20 +168,21 @@ static struct net_device_stats loopback_stats; static struct net_device_stats *get_stats(struct net_device *dev) { struct net_device_stats *stats = &loopback_stats; + unsigned long bytes = 0; + unsigned long packets = 0; int i; - memset(stats, 0, sizeof(struct net_device_stats)); - for_each_possible_cpu(i) { - struct net_device_stats *lb_stats; + const struct pcpu_lstats *lb_stats; - lb_stats = &per_cpu(loopback_stats, i); - stats->rx_bytes += lb_stats->rx_bytes; - stats->tx_bytes += lb_stats->tx_bytes; - stats->rx_packets += lb_stats->rx_packets; - stats->tx_packets += lb_stats->tx_packets; + lb_stats = &per_cpu(pcpu_lstats, i); + bytes += lb_stats->bytes; + packets += lb_stats->packets; } - + stats->rx_packets = packets; + stats->tx_packets = packets; + stats->rx_bytes = bytes; + stats->tx_bytes = bytes; return stats; } -- cgit v0.10.2 From 42952231c6a8623117ee3cc89c82d382dc69ca30 Mon Sep 17 00:00:00 2001 From: Ranjit Manomohan Date: Wed, 18 Oct 2006 20:54:26 -0700 Subject: [TG3]: Fix set ring params tx ring size implementation Fixes the implementation of the ethtool set ring parameters for the tg3 transmit ring. The size of tx_pending is taken into account before doing a netif_wake_queue. This prevents the interface from locking up when smaller transmit ring sizes are used. Signed-off-by: Ranjit Manomohan Acked-by: Michael Chan Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 327836b..39e4833 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -129,7 +129,7 @@ #define RX_JUMBO_PKT_BUF_SZ (9046 + tp->rx_offset + 64) /* minimum number of free TX descriptors required to wake up TX process */ -#define TG3_TX_WAKEUP_THRESH (TG3_TX_RING_SIZE / 4) +#define TG3_TX_WAKEUP_THRESH(tp) ((tp)->tx_pending / 4) /* number of ETHTOOL_GSTATS u64's */ #define TG3_NUM_STATS (sizeof(struct tg3_ethtool_stats)/sizeof(u64)) @@ -3075,10 +3075,10 @@ static void tg3_tx(struct tg3 *tp) smp_mb(); if (unlikely(netif_queue_stopped(tp->dev) && - (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH))) { + (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH(tp)))) { netif_tx_lock(tp->dev); if (netif_queue_stopped(tp->dev) && - (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH)) + (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH(tp))) netif_wake_queue(tp->dev); netif_tx_unlock(tp->dev); } @@ -3928,7 +3928,7 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev) tp->tx_prod = entry; if (unlikely(tg3_tx_avail(tp) <= (MAX_SKB_FRAGS + 1))) { netif_stop_queue(dev); - if (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH) + if (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH(tp)) netif_wake_queue(tp->dev); } @@ -4143,7 +4143,7 @@ static int tg3_start_xmit_dma_bug(struct sk_buff *skb, struct net_device *dev) tp->tx_prod = entry; if (unlikely(tg3_tx_avail(tp) <= (MAX_SKB_FRAGS + 1))) { netif_stop_queue(dev); - if (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH) + if (tg3_tx_avail(tp) > TG3_TX_WAKEUP_THRESH(tp)) netif_wake_queue(tp->dev); } -- cgit v0.10.2 From bc3a9254a7925b6278f6d882746fe6a0bdf6f610 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 18 Oct 2006 20:55:18 -0700 Subject: [TG3]: Add lower bound checks for tx ring size. The minimum tx ring size must be greater than MAX_SKB_FRAGS or 3 times that on some chips with TSO bugs. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 39e4833..7ba9dd2 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -8106,7 +8106,10 @@ static int tg3_set_ringparam(struct net_device *dev, struct ethtool_ringparam *e if ((ering->rx_pending > TG3_RX_RING_SIZE - 1) || (ering->rx_jumbo_pending > TG3_RX_JUMBO_RING_SIZE - 1) || - (ering->tx_pending > TG3_TX_RING_SIZE - 1)) + (ering->tx_pending > TG3_TX_RING_SIZE - 1) || + (ering->tx_pending <= MAX_SKB_FRAGS) || + ((tp->tg3_flags2 & TG3_FLG2_HW_TSO_1_BUG) && + (ering->tx_pending <= (MAX_SKB_FRAGS * 3)))) return -EINVAL; if (netif_running(dev)) { -- cgit v0.10.2 From 6ba7511b7c8b71d5148f7516584d0b677acc384a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Oct 2006 20:56:06 -0700 Subject: [TG3]: Bump driver version and release date. Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 7ba9dd2..8e39849 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -68,8 +68,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.66" -#define DRV_MODULE_RELDATE "September 23, 2006" +#define DRV_MODULE_VERSION "3.67" +#define DRV_MODULE_RELDATE "October 18, 2006" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 -- cgit v0.10.2 From 6723ab549df777742801716d7aeea397e6e97f2c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Oct 2006 21:20:57 -0700 Subject: [IPV6]: Fix route.c warnings when multiple tables are disabled. Signed-off-by: David S. Miller diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 5c00ca4..c953466 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -94,9 +94,6 @@ static int ip6_dst_gc(void); static int ip6_pkt_discard(struct sk_buff *skb); static int ip6_pkt_discard_out(struct sk_buff *skb); -static int ip6_pkt_prohibit(struct sk_buff *skb); -static int ip6_pkt_prohibit_out(struct sk_buff *skb); -static int ip6_pkt_blk_hole(struct sk_buff *skb); static void ip6_link_failure(struct sk_buff *skb); static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu); @@ -144,6 +141,10 @@ struct rt6_info ip6_null_entry = { #ifdef CONFIG_IPV6_MULTIPLE_TABLES +static int ip6_pkt_prohibit(struct sk_buff *skb); +static int ip6_pkt_prohibit_out(struct sk_buff *skb); +static int ip6_pkt_blk_hole(struct sk_buff *skb); + struct rt6_info ip6_prohibit_entry = { .u = { .dst = { @@ -1768,6 +1769,8 @@ static int ip6_pkt_discard_out(struct sk_buff *skb) return ip6_pkt_discard(skb); } +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + static int ip6_pkt_prohibit(struct sk_buff *skb) { return ip6_pkt_drop(skb, ICMPV6_ADM_PROHIBITED); @@ -1785,6 +1788,8 @@ static int ip6_pkt_blk_hole(struct sk_buff *skb) return 0; } +#endif + /* * Allocate a dst for local (unicast / anycast) address. */ -- cgit v0.10.2 From 6b0022305f80cf249de69e746f6f5ccf7ffc5b7c Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 12 Oct 2006 17:07:45 +0900 Subject: sh: Proper show_stack/show_trace() implementation. This splits out some of the previous show_stack() implementation which was mostly doing the show_trace() work without actually dumping any of the stack contents. This now gets split in to two sections, where we do the fetching of the stack pointer and subsequent stack dumping in show_stack(), while moving the call trace in to show_trace(). Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c index 91516dc..a52b13a 100644 --- a/arch/sh/kernel/process.c +++ b/arch/sh/kernel/process.c @@ -105,7 +105,7 @@ void show_regs(struct pt_regs * regs) { printk("\n"); printk("Pid : %d, Comm: %20s\n", current->pid, current->comm); - print_symbol("PC is at %s\n", regs->pc); + print_symbol("PC is at %s\n", instruction_pointer(regs)); printk("PC : %08lx SP : %08lx SR : %08lx ", regs->pc, regs->regs[15], regs->sr); #ifdef CONFIG_MMU @@ -130,15 +130,7 @@ void show_regs(struct pt_regs * regs) printk("MACH: %08lx MACL: %08lx GBR : %08lx PR : %08lx\n", regs->mach, regs->macl, regs->gbr, regs->pr); - /* - * If we're in kernel mode, dump the stack too.. - */ - if (!user_mode(regs)) { - extern void show_task(unsigned long *sp); - unsigned long sp = regs->regs[15]; - - show_task((unsigned long *)sp); - } + show_trace(NULL, (unsigned long *)regs->regs[15], regs); } /* diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c index c2c597e..ffe127f 100644 --- a/arch/sh/kernel/traps.c +++ b/arch/sh/kernel/traps.c @@ -1,16 +1,15 @@ -/* $Id: traps.c,v 1.17 2004/05/02 01:46:30 sugioka Exp $ - * - * linux/arch/sh/traps.c +/* + * 'traps.c' handles hardware traps and faults after we have saved some + * state in 'entry.S'. * * SuperH version: Copyright (C) 1999 Niibe Yutaka * Copyright (C) 2000 Philipp Rumpf * Copyright (C) 2000 David Howells - * Copyright (C) 2002, 2003 Paul Mundt - */ - -/* - * 'Traps.c' handles hardware traps and faults after we have saved some - * state in 'entry.S'. + * Copyright (C) 2002 - 2006 Paul Mundt + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. */ #include #include @@ -53,13 +52,32 @@ #define TRAP_ILLEGAL_SLOT_INST 13 #endif -/* - * These constants are for searching for possible module text - * segments. VMALLOC_OFFSET comes from mm/vmalloc.c; MODULE_RANGE is - * a guess of how much space is likely to be vmalloced. - */ -#define VMALLOC_OFFSET (8*1024*1024) -#define MODULE_RANGE (8*1024*1024) +static void dump_mem(const char *str, unsigned long bottom, unsigned long top) +{ + unsigned long p; + int i; + + printk("%s(0x%08lx to 0x%08lx)\n", str, bottom, top); + + for (p = bottom & ~31; p < top; ) { + printk("%04lx: ", p & 0xffff); + + for (i = 0; i < 8; i++, p += 4) { + unsigned int val; + + if (p < bottom || p >= top) + printk(" "); + else { + if (__get_user(val, (unsigned int __user *)p)) { + printk("\n"); + return; + } + printk("%08x ", val); + } + } + printk("\n"); + } +} DEFINE_SPINLOCK(die_lock); @@ -69,14 +87,28 @@ void die(const char * str, struct pt_regs * regs, long err) console_verbose(); spin_lock_irq(&die_lock); + bust_spinlocks(1); + printk("%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter); + CHK_REMOTE_DEBUG(regs); + print_modules(); show_regs(regs); + + printk("Process: %s (pid: %d, stack limit = %p)\n", + current->comm, current->pid, task_stack_page(current) + 1); + + if (!user_mode(regs) || in_interrupt()) + dump_mem("Stack: ", regs->regs[15], THREAD_SIZE + + (unsigned long)task_stack_page(current)); + + bust_spinlocks(0); spin_unlock_irq(&die_lock); do_exit(SIGSEGV); } -static inline void die_if_kernel(const char * str, struct pt_regs * regs, long err) +static inline void die_if_kernel(const char *str, struct pt_regs *regs, + long err) { if (!user_mode(regs)) die(str, regs, err); @@ -93,8 +125,7 @@ static int handle_unaligned_notify_count = 10; */ static int die_if_no_fixup(const char * str, struct pt_regs * regs, long err) { - if (!user_mode(regs)) - { + if (!user_mode(regs)) { const struct exception_table_entry *fixup; fixup = search_exception_tables(regs->pc); if (fixup) { @@ -734,52 +765,43 @@ void __init trap_init(void) per_cpu_trap_init(); } -void show_stack(struct task_struct *tsk, unsigned long *sp) +void show_trace(struct task_struct *tsk, unsigned long *sp, + struct pt_regs *regs) { - unsigned long *stack, addr; - unsigned long module_start = VMALLOC_START; - unsigned long module_end = VMALLOC_END; - int i = 1; + unsigned long addr; - if (!tsk) - tsk = current; - if (tsk == current) - sp = (unsigned long *)current_stack_pointer; - else - sp = (unsigned long *)tsk->thread.sp; - - stack = sp; + if (regs && user_mode(regs)) + return; printk("\nCall trace: "); #ifdef CONFIG_KALLSYMS printk("\n"); #endif - while (!kstack_end(stack)) { - addr = *stack++; - if (((addr >= (unsigned long)_text) && - (addr <= (unsigned long)_etext)) || - ((addr >= module_start) && (addr <= module_end))) { - /* - * For 80-columns display, 6 entry is maximum. - * NOTE: '[<8c00abcd>] ' consumes 13 columns . - */ -#ifndef CONFIG_KALLSYMS - if (i && ((i % 6) == 0)) - printk("\n "); -#endif - printk("[<%08lx>] ", addr); - print_symbol("%s\n", addr); - i++; - } + while (!kstack_end(sp)) { + addr = *sp++; + if (kernel_text_address(addr)) + print_ip_sym(addr); } printk("\n"); } -void show_task(unsigned long *sp) +void show_stack(struct task_struct *tsk, unsigned long *sp) { - show_stack(NULL, sp); + unsigned long stack; + + if (!tsk) + tsk = current; + if (tsk == current) + sp = (unsigned long *)current_stack_pointer; + else + sp = (unsigned long *)tsk->thread.sp; + + stack = (unsigned long)sp; + dump_mem("Stack: ", stack, THREAD_SIZE + + (unsigned long)task_stack_page(tsk)); + show_trace(tsk, sp, NULL); } void dump_stack(void) diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h index 4747738..45bb74e 100644 --- a/include/asm-sh/processor.h +++ b/include/asm-sh/processor.h @@ -255,6 +255,8 @@ extern void save_fpu(struct task_struct *__tsk, struct pt_regs *regs); */ #define thread_saved_pc(tsk) (tsk->thread.pc) +void show_trace(struct task_struct *tsk, unsigned long *sp, + struct pt_regs *regs); extern unsigned long get_wchan(struct task_struct *p); #define KSTK_EIP(tsk) ((tsk)->thread.pc) -- cgit v0.10.2 From 4a58eaca7ca68abea37d6d2a4ea7deb394906183 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 19 Oct 2006 16:15:13 +0900 Subject: sh: Remove board-specific ide.h headers. The driver that these were using never made it in to drivers/ide, so kill off the rest of the cruft. These will have to be reworked for board-specific platform devices through libata when they're added back through the setup code. Signed-off-by: Paul Mundt diff --git a/include/asm-sh/hp6xx/ide.h b/include/asm-sh/hp6xx/ide.h deleted file mode 100644 index 570395a..0000000 --- a/include/asm-sh/hp6xx/ide.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef __ASM_SH_HP6XX_IDE_H -#define __ASM_SH_HP6XX_IDE_H - -#define IRQ_CFCARD 93 -#define IRQ_PCMCIA 94 - -#endif /* __ASM_SH_HP6XX_IDE_H */ - diff --git a/include/asm-sh/hs7751rvoip/ide.h b/include/asm-sh/hs7751rvoip/ide.h deleted file mode 100644 index 65ad1d0..0000000 --- a/include/asm-sh/hs7751rvoip/ide.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef __ASM_SH_HS7751RVOIP_IDE_H -#define __ASM_SH_HS7751RVOIP_IDE_H - -/* Nothing to see here.. */ -#include - -#endif /* __ASM_SH_HS7751RVOIP_IDE_H */ - diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h index 28996f9..1837bdb 100644 --- a/include/asm-sh/irq.h +++ b/include/asm-sh/irq.h @@ -14,16 +14,6 @@ #include #include /* for pt_regs */ -#if defined(CONFIG_SH_HP6XX) || \ - defined(CONFIG_SH_RTS7751R2D) || \ - defined(CONFIG_SH_HS7751RVOIP) || \ - defined(CONFIG_SH_HS7751RVOIP) || \ - defined(CONFIG_SH_SH03) || \ - defined(CONFIG_SH_R7780RP) || \ - defined(CONFIG_SH_LANDISK) -#include -#endif - #ifndef CONFIG_CPU_SUBTYPE_SH7780 #define INTC_DMAC0_MSK 0 @@ -38,15 +28,6 @@ #define INTC_IPRD 0xffd00010UL #endif -#ifdef CONFIG_IDE -# ifndef IRQ_CFCARD -# define IRQ_CFCARD 14 -# endif -# ifndef IRQ_PCMCIA -# define IRQ_PCMCIA 15 -# endif -#endif - #define TIMER_IRQ 16 #define TIMER_IPR_ADDR INTC_IPRA #define TIMER_IPR_POS 3 diff --git a/include/asm-sh/landisk/ide.h b/include/asm-sh/landisk/ide.h deleted file mode 100644 index 6490e28..0000000 --- a/include/asm-sh/landisk/ide.h +++ /dev/null @@ -1,14 +0,0 @@ -/* - * modifed by kogiidena - * 2005.03.03 - */ - -#ifndef __ASM_SH_LANDISK_IDE_H -#define __ASM_SH_LANDISK_IDE_H - -/* Nothing to see here.. */ -#include -#define IRQ_CFCARD IRQ_FATA /* CF Card IRQ */ -#define IRQ_PCMCIA IRQ_ATA /* PCMCIA IRQ */ - -#endif /* __ASM_SH_LANDISK_IDE_H */ diff --git a/include/asm-sh/r7780rp/ide.h b/include/asm-sh/r7780rp/ide.h deleted file mode 100644 index a1ed78e..0000000 --- a/include/asm-sh/r7780rp/ide.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef __ASM_SH_R7780RP_IDE_H -#define __ASM_SH_R7780RP_IDE_H - -/* Nothing to see here.. */ -#include - -#endif /* __ASM_SH_R7780RP_IDE_H */ - diff --git a/include/asm-sh/rts7751r2d/ide.h b/include/asm-sh/rts7751r2d/ide.h deleted file mode 100644 index 416f96b..0000000 --- a/include/asm-sh/rts7751r2d/ide.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef __ASM_SH_RTS7751R2D_IDE_H -#define __ASM_SH_RTS7751R2D_IDE_H - -/* Nothing to see here.. */ -#include - -#endif /* __ASM_SH_RTS7751R2D_IDE_H */ - diff --git a/include/asm-sh/sh03/ide.h b/include/asm-sh/sh03/ide.h deleted file mode 100644 index 73ee92e..0000000 --- a/include/asm-sh/sh03/ide.h +++ /dev/null @@ -1,7 +0,0 @@ -#ifndef __ASM_SH_SH03_IDE_H -#define __ASM_SH_SH03_IDE_H - -#define IRQ_CFCARD 8 -#define IRQ_PCMCIA 8 - -#endif /* __ASM_SH_SH03_IDE_H */ -- cgit v0.10.2 From 082c44d20eb4c6c4aa60ae7429ea184854cb0610 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 19 Oct 2006 16:16:18 +0900 Subject: sh: Cleanup board header directories. Now with the ide.h mess sorted out, most of these boards don't need their own directory. Move the headers out, and update the driver paths. Signed-off-by: Paul Mundt diff --git a/arch/sh/boards/hp6xx/hp6xx_apm.c b/arch/sh/boards/hp6xx/hp6xx_apm.c index 2191791..d146cda 100644 --- a/arch/sh/boards/hp6xx/hp6xx_apm.c +++ b/arch/sh/boards/hp6xx/hp6xx_apm.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #define SH7709_PGDR 0xa400012c diff --git a/arch/sh/boards/hp6xx/pm.c b/arch/sh/boards/hp6xx/pm.c index 83d3272..d194773 100644 --- a/arch/sh/boards/hp6xx/pm.c +++ b/arch/sh/boards/hp6xx/pm.c @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c index 2d3a5b4..b5a9664 100644 --- a/arch/sh/boards/hp6xx/setup.c +++ b/arch/sh/boards/hp6xx/setup.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include #define SCPCR 0xa4000116 diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c index 51f3f65..bb9aa0d 100644 --- a/arch/sh/boards/renesas/hs7751rvoip/io.c +++ b/arch/sh/boards/renesas/hs7751rvoip/io.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include #include extern void *area6_io8_base; /* Area 6 8bit I/O Base address */ diff --git a/arch/sh/boards/renesas/hs7751rvoip/irq.c b/arch/sh/boards/renesas/hs7751rvoip/irq.c index c617b18..943f93a 100644 --- a/arch/sh/boards/renesas/hs7751rvoip/irq.c +++ b/arch/sh/boards/renesas/hs7751rvoip/irq.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include static int mask_pos[] = {8, 9, 10, 11, 12, 13, 0, 1, 2, 3, 4, 5, 6, 7}; diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c index 0414c15..1d997ff 100644 --- a/arch/sh/boards/renesas/hs7751rvoip/setup.c +++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c @@ -10,15 +10,10 @@ #include #include #include -#include -#include -#include #include +#include #include -#include #include -#include -#include static void __init hs7751rvoip_init_irq(void) { diff --git a/arch/sh/boards/renesas/r7780rp/io.c b/arch/sh/boards/renesas/r7780rp/io.c index db92d6e..311cccc 100644 --- a/arch/sh/boards/renesas/r7780rp/io.c +++ b/arch/sh/boards/renesas/r7780rp/io.c @@ -11,7 +11,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c index b544772..cbb4ea2 100644 --- a/arch/sh/boards/renesas/r7780rp/irq.c +++ b/arch/sh/boards/renesas/r7780rp/irq.c @@ -10,7 +10,8 @@ */ #include #include -#include +#include +#include #ifdef CONFIG_SH_R7780MP static int mask_pos[] = {12, 11, 9, 14, 15, 8, 13, 6, 5, 4, 3, 2, 0, 0, 1, 0}; diff --git a/arch/sh/boards/renesas/r7780rp/setup.c b/arch/sh/boards/renesas/r7780rp/setup.c index b941aa0..c331cae 100644 --- a/arch/sh/boards/renesas/r7780rp/setup.c +++ b/arch/sh/boards/renesas/r7780rp/setup.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/sh/boards/renesas/rts7751r2d/io.c b/arch/sh/boards/renesas/rts7751r2d/io.c index 135aa0b..f2507a8 100644 --- a/arch/sh/boards/renesas/rts7751r2d/io.c +++ b/arch/sh/boards/renesas/rts7751r2d/io.c @@ -11,8 +11,8 @@ #include #include #include -#include -#include +#include +#include #include /* diff --git a/arch/sh/boards/renesas/rts7751r2d/irq.c b/arch/sh/boards/renesas/rts7751r2d/irq.c index c915e7a..cb0eb20 100644 --- a/arch/sh/boards/renesas/rts7751r2d/irq.c +++ b/arch/sh/boards/renesas/rts7751r2d/irq.c @@ -8,12 +8,10 @@ * Modified for RTS7751R2D by * Atom Create Engineering Co., Ltd. 2002. */ - #include #include -#include -#include -#include +#include +#include #if defined(CONFIG_RTS7751R2D_REV11) static int mask_pos[] = {11, 9, 8, 12, 10, 6, 5, 4, 7, 14, 13, 0, 0, 0, 0}; diff --git a/arch/sh/boards/renesas/rts7751r2d/led.c b/arch/sh/boards/renesas/rts7751r2d/led.c index a7ce66c..509f548 100644 --- a/arch/sh/boards/renesas/rts7751r2d/led.c +++ b/arch/sh/boards/renesas/rts7751r2d/led.c @@ -8,13 +8,9 @@ * * This file contains Renesas Technology Sales RTS7751R2D specific LED code. */ - -#include -#include - -#ifdef CONFIG_HEARTBEAT - +#include #include +#include /* Cycle the LED's in the clasic Knightriger/Sun pattern */ void heartbeat_rts7751r2d(void) @@ -46,10 +42,3 @@ void heartbeat_rts7751r2d(void) else bit--; } -#endif /* CONFIG_HEARTBEAT */ - -void rts7751r2d_led(unsigned short value) -{ - ctrl_outw(value, PA_OUTPORT); -} - diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c index 20597a6..5c042d3 100644 --- a/arch/sh/boards/renesas/rts7751r2d/setup.c +++ b/arch/sh/boards/renesas/rts7751r2d/setup.c @@ -12,9 +12,9 @@ #include #include #include -#include #include #include +#include #include extern void heartbeat_rts7751r2d(void); diff --git a/arch/sh/boards/shmin/setup.c b/arch/sh/boards/shmin/setup.c index 2f0c197..a31a1d1 100644 --- a/arch/sh/boards/shmin/setup.c +++ b/arch/sh/boards/shmin/setup.c @@ -7,7 +7,7 @@ */ #include #include -#include +#include #include #include #include diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c index bf1b28f..f7ea700 100644 --- a/arch/sh/cchips/voyagergx/irq.c +++ b/arch/sh/cchips/voyagergx/irq.c @@ -17,29 +17,18 @@ Copyright 2003 (c) Lineo uSolutions,Inc. */ -/* -------------------------------------------------------------------- */ - -#undef DEBUG - -#include -#include -#include -#include -#include #include #include -#include - -#include -#include +#include #include +#include static void disable_voyagergx_irq(unsigned int irq) { unsigned long val; unsigned long mask = 1 << (irq - VOYAGER_IRQ_BASE); - pr_debug("disable_voyagergx_irq(%d): mask=%x\n", irq, mask); + pr_debug("disable_voyagergx_irq(%d): mask=%lx\n", irq, mask); val = inl(VOYAGER_INT_MASK); val &= ~mask; outl(val, VOYAGER_INT_MASK); @@ -50,7 +39,7 @@ static void enable_voyagergx_irq(unsigned int irq) unsigned long val; unsigned long mask = 1 << (irq - VOYAGER_IRQ_BASE); - pr_debug("disable_voyagergx_irq(%d): mask=%x\n", irq, mask); + pr_debug("disable_voyagergx_irq(%d): mask=%lx\n", irq, mask); val = inl(VOYAGER_INT_MASK); val |= mask; outl(val, VOYAGER_INT_MASK); @@ -137,7 +126,7 @@ int voyagergx_irq_demux(int irq) } else { printk("Unexpected IRQ irq = %d status = 0x%08lx\n", irq, val); } - pr_debug("voyagergx_irq_demux %d \n", i); + pr_debug("voyagergx_irq_demux %ld\n", i); #else for (bit = 1, i = 0 ; i < VOYAGER_IRQ_NUM ; bit <<= 1, i++) if (val & bit) @@ -185,4 +174,3 @@ void __init setup_voyagergx_irq(void) setup_irq(IRQ_VOYAGER, &irq0); } - diff --git a/arch/sh/drivers/pci/ops-r7780rp.c b/arch/sh/drivers/pci/ops-r7780rp.c index 6e3ba9c6..eeea157 100644 --- a/arch/sh/drivers/pci/ops-r7780rp.c +++ b/arch/sh/drivers/pci/ops-r7780rp.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include #include "pci-sh4.h" diff --git a/arch/sh/drivers/pci/ops-rts7751r2d.c b/arch/sh/drivers/pci/ops-rts7751r2d.c index b68824c..4a518d9 100644 --- a/arch/sh/drivers/pci/ops-rts7751r2d.c +++ b/arch/sh/drivers/pci/ops-rts7751r2d.c @@ -10,28 +10,24 @@ * * PCI initialization for the Renesas SH7751R RTS7751R2D board */ - #include #include #include -#include #include -#include -#include -#include +#include +#include #include "pci-sh4.h" +static u8 rts7751r2d_irq_tab[] __initdata = { + IRQ_PCISLOT1, + IRQ_PCISLOT2, + IRQ_PCMCIA, + IRQ_PCIETH, +}; + int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin) { - switch (slot) { - case 0: return IRQ_PCISLOT1; /* PCI Extend slot #1 */ - case 1: return IRQ_PCISLOT2; /* PCI Extend slot #2 */ - case 2: return IRQ_PCMCIA; /* PCI Cardbus Bridge */ - case 3: return IRQ_PCIETH; /* Realtek Ethernet controller */ - default: - printk("PCI: Bad IRQ mapping request for slot %d\n", slot); - return -1; - } + return rts7751r2d_irq_tab[slot]; } static struct resource sh7751_io_resource = { diff --git a/drivers/input/touchscreen/hp680_ts_input.c b/drivers/input/touchscreen/hp680_ts_input.c index e31c6c5..58fca31 100644 --- a/drivers/input/touchscreen/hp680_ts_input.c +++ b/drivers/input/touchscreen/hp680_ts_input.c @@ -6,7 +6,7 @@ #include #include #include -#include +#include #define MODNAME "hp680_ts_input" diff --git a/drivers/video/hitfb.c b/drivers/video/hitfb.c index 3afb472..3dc4942 100644 --- a/drivers/video/hitfb.c +++ b/drivers/video/hitfb.c @@ -29,7 +29,6 @@ #include #include #include -#include #define WIDTH 640 diff --git a/include/asm-sh/edosk7705/io.h b/include/asm-sh/edosk7705/io.h deleted file mode 100644 index a1089a6..0000000 --- a/include/asm-sh/edosk7705/io.h +++ /dev/null @@ -1,30 +0,0 @@ -/* - * include/asm-sh/edosk7705/io.h - * - * Modified version of io_se.h for the EDOSK7705 specific functions. - * - * May be copied or modified under the terms of the GNU General Public - * License. See linux/COPYING for more information. - * - * IO functions for an Hitachi EDOSK7705 development board - */ - -#ifndef __ASM_SH_EDOSK7705_IO_H -#define __ASM_SH_EDOSK7705_IO_H - -#include - -extern unsigned char sh_edosk7705_inb(unsigned long port); -extern unsigned int sh_edosk7705_inl(unsigned long port); - -extern void sh_edosk7705_outb(unsigned char value, unsigned long port); -extern void sh_edosk7705_outl(unsigned int value, unsigned long port); - -extern void sh_edosk7705_insb(unsigned long port, void *addr, unsigned long count); -extern void sh_edosk7705_insl(unsigned long port, void *addr, unsigned long count); -extern void sh_edosk7705_outsb(unsigned long port, const void *addr, unsigned long count); -extern void sh_edosk7705_outsl(unsigned long port, const void *addr, unsigned long count); - -extern unsigned long sh_edosk7705_isa_port2addr(unsigned long offset); - -#endif /* __ASM_SH_EDOSK7705_IO_H */ diff --git a/include/asm-sh/hp6xx/hp6xx.h b/include/asm-sh/hp6xx/hp6xx.h deleted file mode 100644 index f35134c..0000000 --- a/include/asm-sh/hp6xx/hp6xx.h +++ /dev/null @@ -1,80 +0,0 @@ -#ifndef __ASM_SH_HP6XX_H -#define __ASM_SH_HP6XX_H - -/* - * Copyright (C) 2003, 2004, 2005 Andriy Skulysh - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - */ - -#define HP680_BTN_IRQ IRQ0_IRQ -#define HP680_TS_IRQ IRQ3_IRQ -#define HP680_HD64461_IRQ IRQ4_IRQ - -#define DAC_LCD_BRIGHTNESS 0 -#define DAC_SPEAKER_VOLUME 1 - -#define PGDR_OPENED 0x01 -#define PGDR_MAIN_BATTERY_OUT 0x04 -#define PGDR_PLAY_BUTTON 0x08 -#define PGDR_REWIND_BUTTON 0x10 -#define PGDR_RECORD_BUTTON 0x20 - -#define PHDR_TS_PEN_DOWN 0x08 - -#define PJDR_LED_BLINK 0x02 - -#define PKDR_LED_GREEN 0x10 - -#define SCPDR_TS_SCAN_ENABLE 0x20 -#define SCPDR_TS_SCAN_Y 0x02 -#define SCPDR_TS_SCAN_X 0x01 - -#define SCPCR_TS_ENABLE 0x405 -#define SCPCR_TS_MASK 0xc0f - -#define ADC_CHANNEL_TS_Y 1 -#define ADC_CHANNEL_TS_X 2 -#define ADC_CHANNEL_BATTERY 3 -#define ADC_CHANNEL_BACKUP 4 -#define ADC_CHANNEL_CHARGE 5 - -#define HD64461_GPADR_SPEAKER 0x01 -#define HD64461_GPADR_PCMCIA0 (0x02|0x08) - -#define HD64461_GPBDR_LCDOFF 0x01 -#define HD64461_GPBDR_LCD_CONTRAST_MASK 0x78 -#define HD64461_GPBDR_LED_RED 0x80 - -#include -#include - -#define PJDR 0xa4000130 -#define PKDR 0xa4000132 - -static inline void hp6xx_led_red(int on) -{ - u16 v16; - v16 = ctrl_inw(CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000); - if (on) - ctrl_outw(v16 & (~HD64461_GPBDR_LED_RED), CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000); - else - ctrl_outw(v16 | HD64461_GPBDR_LED_RED, CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000); -} - -static inline void hp6xx_led_green(int on) -{ - u8 v8; - - v8 = ctrl_inb(PKDR); - if (on) - ctrl_outb(v8 & (~PKDR_LED_GREEN), PKDR); - else - ctrl_outb(v8 | PKDR_LED_GREEN, PKDR); -} - - -#endif /* __ASM_SH_HP6XX_H */ diff --git a/include/asm-sh/hp6xx/io.h b/include/asm-sh/hp6xx/io.h deleted file mode 100644 index 2044476..0000000 --- a/include/asm-sh/hp6xx/io.h +++ /dev/null @@ -1,10 +0,0 @@ -#ifndef __ASM_SH_HP6XX_IO_H -#define __ASM_SH_HP6XX_IO_H - -/* - * Nothing special here.. just use the generic cchip io routines. - */ -#include - -#endif /* __ASM_SH_HP6XX_IO_H */ - diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h deleted file mode 100644 index c4cff9d..0000000 --- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h +++ /dev/null @@ -1,54 +0,0 @@ -#ifndef __ASM_SH_RENESAS_HS7751RVOIP_H -#define __ASM_SH_RENESAS_HS7751RVOIP_H - -/* - * linux/include/asm-sh/hs7751rvoip/hs7751rvoip.h - * - * Copyright (C) 2000 Atom Create Engineering Co., Ltd. - * - * Renesas Technology Sales HS7751RVoIP support - */ - -/* Box specific addresses. */ - -#define PA_BCR 0xa4000000 /* FPGA */ -#define PA_SLICCNTR1 0xa4000006 /* SLIC PIO Control 1 */ -#define PA_SLICCNTR2 0xa4000008 /* SLIC PIO Control 2 */ -#define PA_DMACNTR 0xa400000a /* USB DMA Control */ -#define PA_INPORTR 0xa400000c /* Input Port Register */ -#define PA_OUTPORTR 0xa400000e /* Output Port Reguster */ -#define PA_VERREG 0xa4000014 /* FPGA Version Register */ - -#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ - -#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ -#define IRLCNTR2 (PA_BCR + 2) /* Interrupt Control Register2 */ -#define IRLCNTR3 (PA_BCR + 4) /* Interrupt Control Register3 */ -#define IRLCNTR4 (PA_BCR + 16) /* Interrupt Control Register4 */ -#define IRLCNTR5 (PA_BCR + 18) /* Interrupt Control Register5 */ - -#define IRQ_PCIETH 6 /* PCI Ethernet IRQ */ -#define IRQ_PCIHUB 7 /* PCI Ethernet Hub IRQ */ -#define IRQ_USBCOM 8 /* USB Comunication IRQ */ -#define IRQ_USBCON 9 /* USB Connect IRQ */ -#define IRQ_USBDMA 10 /* USB DMA IRQ */ -#define IRQ_CFCARD 11 /* CF Card IRQ */ -#define IRQ_PCMCIA 12 /* PCMCIA IRQ */ -#define IRQ_PCISLOT 13 /* PCI Slot #1 IRQ */ -#define IRQ_ONHOOK1 0 /* ON HOOK1 IRQ */ -#define IRQ_OFFHOOK1 1 /* OFF HOOK1 IRQ */ -#define IRQ_ONHOOK2 2 /* ON HOOK2 IRQ */ -#define IRQ_OFFHOOK2 3 /* OFF HOOK2 IRQ */ -#define IRQ_RINGING 4 /* Ringing IRQ */ -#define IRQ_CODEC 5 /* CODEC IRQ */ - -#define __IO_PREFIX hs7751rvoip -#include - -/* arch/sh/boards/renesas/hs7751rvoip/irq.c */ -void init_hs7751rvoip_IRQ(void); - -/* arch/sh/boards/renesas/hs7751rvoip/io.c */ -void *hs7751rvoip_ioremap(unsigned long, unsigned long); - -#endif /* __ASM_SH_RENESAS_HS7751RVOIP */ diff --git a/include/asm-sh/r7780rp/r7780rp.h b/include/asm-sh/r7780rp/r7780rp.h deleted file mode 100644 index f95d9db..0000000 --- a/include/asm-sh/r7780rp/r7780rp.h +++ /dev/null @@ -1,177 +0,0 @@ -#ifndef __ASM_SH_RENESAS_R7780RP_H -#define __ASM_SH_RENESAS_R7780RP_H - -/* - * linux/include/asm-sh/r7780rp.h - * - * Copyright (C) 2000 Atom Create Engineering Co., Ltd. - * - * Renesas Solutions Highlander R7780RP support - */ - -/* Box specific addresses. */ -#if defined(CONFIG_SH_R7780MP) -#define PA_BCR 0xa4000000 /* FPGA */ -#define PA_IRLMSK (PA_BCR+0x0000) /* Interrupt Mask control */ -#define PA_IRLMON (PA_BCR+0x0002) /* Interrupt Status control */ -#define PA_IRLPRI1 (PA_BCR+0x0004) /* Interrupt Priorty 1 */ -#define PA_IRLPRI2 (PA_BCR+0x0006) /* Interrupt Priorty 2 */ -#define PA_IRLPRI3 (PA_BCR+0x0008) /* Interrupt Priorty 3 */ -#define PA_IRLPRI4 (PA_BCR+0x000a) /* Interrupt Priorty 4 */ -#define PA_RSTCTL (PA_BCR+0x000c) /* Reset Control */ -#define PA_PCIBD (PA_BCR+0x000e) /* PCI Board detect control */ -#define PA_PCICD (PA_BCR+0x0010) /* PCI Conector detect control */ -#define PA_EXTGIO (PA_BCR+0x0016) /* Extension GPIO Control */ -#define PA_IVDRMON (PA_BCR+0x0018) /* iVDR Moniter control */ -#define PA_IVDRCTL (PA_BCR+0x001a) /* iVDR control */ -#define PA_OBLED (PA_BCR+0x001c) /* On Board LED control */ -#define PA_OBSW (PA_BCR+0x001e) /* On Board Switch control */ -#define PA_AUDIOSEL (PA_BCR+0x0020) /* Sound Interface Select control */ -#define PA_EXTPLR (PA_BCR+0x001e) /* Extention Pin Polarity control */ -#define PA_TPCTL (PA_BCR+0x0100) /* Touch Panel Access control */ -#define PA_TPDCKCTL (PA_BCR+0x0102) /* Touch Panel Access data control */ -#define PA_TPCTLCLR (PA_BCR+0x0104) /* Touch Panel Access control */ -#define PA_TPXPOS (PA_BCR+0x0106) /* Touch Panel X position control */ -#define PA_TPYPOS (PA_BCR+0x0108) /* Touch Panel Y position control */ -#define PA_DBSW (PA_BCR+0x0200) /* Debug Board Switch control */ -#define PA_CFCTL (PA_BCR+0x0300) /* CF Timing control */ -#define PA_CFPOW (PA_BCR+0x0302) /* CF Power control */ -#define PA_CFCDINTCLR (PA_BCR+0x0304) /* CF Insert Interrupt clear */ -#define PA_SCSMR0 (PA_BCR+0x0400) /* SCIF0 Serial mode control */ -#define PA_SCBRR0 (PA_BCR+0x0404) /* SCIF0 Bit rate control */ -#define PA_SCSCR0 (PA_BCR+0x0408) /* SCIF0 Serial control */ -#define PA_SCFTDR0 (PA_BCR+0x040c) /* SCIF0 Send FIFO control */ -#define PA_SCFSR0 (PA_BCR+0x0410) /* SCIF0 Serial status control */ -#define PA_SCFRDR0 (PA_BCR+0x0414) /* SCIF0 Receive FIFO control */ -#define PA_SCFCR0 (PA_BCR+0x0418) /* SCIF0 FIFO control */ -#define PA_SCTFDR0 (PA_BCR+0x041c) /* SCIF0 Send FIFO data control */ -#define PA_SCRFDR0 (PA_BCR+0x0420) /* SCIF0 Receive FIFO data control */ -#define PA_SCSPTR0 (PA_BCR+0x0424) /* SCIF0 Serial Port control */ -#define PA_SCLSR0 (PA_BCR+0x0428) /* SCIF0 Line Status control */ -#define PA_SCRER0 (PA_BCR+0x042c) /* SCIF0 Serial Error control */ -#define PA_SCSMR1 (PA_BCR+0x0500) /* SCIF1 Serial mode control */ -#define PA_SCBRR1 (PA_BCR+0x0504) /* SCIF1 Bit rate control */ -#define PA_SCSCR1 (PA_BCR+0x0508) /* SCIF1 Serial control */ -#define PA_SCFTDR1 (PA_BCR+0x050c) /* SCIF1 Send FIFO control */ -#define PA_SCFSR1 (PA_BCR+0x0510) /* SCIF1 Serial status control */ -#define PA_SCFRDR1 (PA_BCR+0x0514) /* SCIF1 Receive FIFO control */ -#define PA_SCFCR1 (PA_BCR+0x0518) /* SCIF1 FIFO control */ -#define PA_SCTFDR1 (PA_BCR+0x051c) /* SCIF1 Send FIFO data control */ -#define PA_SCRFDR1 (PA_BCR+0x0520) /* SCIF1 Receive FIFO data control */ -#define PA_SCSPTR1 (PA_BCR+0x0524) /* SCIF1 Serial Port control */ -#define PA_SCLSR1 (PA_BCR+0x0528) /* SCIF1 Line Status control */ -#define PA_SCRER1 (PA_BCR+0x052c) /* SCIF1 Serial Error control */ -#define PA_ICCR (PA_BCR+0x0600) /* Serial control */ -#define PA_SAR (PA_BCR+0x0602) /* Serial Slave control */ -#define PA_MDR (PA_BCR+0x0604) /* Serial Mode control */ -#define PA_ADR1 (PA_BCR+0x0606) /* Serial Address1 control */ -#define PA_DAR1 (PA_BCR+0x0646) /* Serial Data1 control */ -#define PA_VERREG (PA_BCR+0x0700) /* FPGA Version Register */ -#define PA_POFF (PA_BCR+0x0800) /* System Power Off control */ -#define PA_PMR (PA_BCR+0x0900) /* */ - -#define PA_AX88796L 0xa4100400 /* AX88796L Area */ -#define PA_SC1602BSLB 0xa6000000 /* SC1602BSLB Area */ -#define PA_AREA5_IO 0xb4000000 /* Area 5 IO Memory */ -#define PA_AREA6_IO 0xb8000000 /* Area 6 IO Memory */ -#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ -#define AX88796L_IO_BASE 0x1000 /* AX88796L IO Base Address */ - -#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ - -#define IRQ_PCISLOT1 65 /* PCI Slot #1 IRQ */ -#define IRQ_PCISLOT2 66 /* PCI Slot #2 IRQ */ -#define IRQ_PCISLOT3 67 /* PCI Slot #3 IRQ */ -#define IRQ_PCISLOT4 68 /* PCI Slot #4 IRQ */ -#define IRQ_CFCARD 1 /* CF Card IRQ */ -// #define IRQ_CFINST 0 /* CF Card Insert IRQ */ -#define IRQ_TP 2 /* Touch Panel IRQ */ -#define IRQ_SCI1 3 /* SCI1 IRQ */ -#define IRQ_SCI0 4 /* SCI0 IRQ */ -#define IRQ_2SERIAL 5 /* Serial IRQ */ -#define IRQ_RTC 6 /* RTC A / B IRQ */ -#define IRQ_EXTENTION6 7 /* EXT6n IRQ */ -#define IRQ_EXTENTION5 8 /* EXT5n IRQ */ -#define IRQ_EXTENTION4 9 /* EXT4n IRQ */ -#define IRQ_EXTENTION2 10 /* EXT2n IRQ */ -#define IRQ_EXTENTION1 11 /* EXT1n IRQ */ -#define IRQ_ONETH 13 /* On board Ethernet IRQ */ -#define IRQ_PSW 14 /* Push Switch IRQ */ - -#else /* R7780RP */ - -#define PA_BCR 0xa5000000 /* FPGA */ -#define PA_IRLMSK (PA_BCR+0x0000) /* Interrupt Mask control */ -#define PA_IRLMON (PA_BCR+0x0002) /* Interrupt Status control */ -#define PA_SDPOW (PA_BCR+0x0004) /* SD Power control */ -#define PA_RSTCTL (PA_BCR+0x0006) /* Device Reset control */ -#define PA_PCIBD (PA_BCR+0x0008) /* PCI Board detect control */ -#define PA_PCICD (PA_BCR+0x000a) /* PCI Conector detect control */ -#define PA_ZIGIO1 (PA_BCR+0x000c) /* Zigbee IO control 1 */ -#define PA_ZIGIO2 (PA_BCR+0x000e) /* Zigbee IO control 2 */ -#define PA_ZIGIO3 (PA_BCR+0x0010) /* Zigbee IO control 3 */ -#define PA_ZIGIO4 (PA_BCR+0x0012) /* Zigbee IO control 4 */ -#define PA_IVDRMON (PA_BCR+0x0014) /* iVDR Moniter control */ -#define PA_IVDRCTL (PA_BCR+0x0016) /* iVDR control */ -#define PA_OBLED (PA_BCR+0x0018) /* On Board LED control */ -#define PA_OBSW (PA_BCR+0x001a) /* On Board Switch control */ -#define PA_AUDIOSEL (PA_BCR+0x001c) /* Sound Interface Select control */ -#define PA_EXTPLR (PA_BCR+0x001e) /* Extention Pin Polarity control */ -#define PA_TPCTL (PA_BCR+0x0100) /* Touch Panel Access control */ -#define PA_TPDCKCTL (PA_BCR+0x0102) /* Touch Panel Access data control */ -#define PA_TPCTLCLR (PA_BCR+0x0104) /* Touch Panel Access control */ -#define PA_TPXPOS (PA_BCR+0x0106) /* Touch Panel X position control */ -#define PA_TPYPOS (PA_BCR+0x0108) /* Touch Panel Y position control */ -#define PA_DBDET (PA_BCR+0x0200) /* Debug Board detect control */ -#define PA_DBDISPCTL (PA_BCR+0x0202) /* Debug Board Dot timing control */ -#define PA_DBSW (PA_BCR+0x0204) /* Debug Board Switch control */ -#define PA_CFCTL (PA_BCR+0x0300) /* CF Timing control */ -#define PA_CFPOW (PA_BCR+0x0302) /* CF Power control */ -#define PA_CFCDINTCLR (PA_BCR+0x0304) /* CF Insert Interrupt clear */ -#define PA_SCSMR (PA_BCR+0x0400) /* SCIF Serial mode control */ -#define PA_SCBRR (PA_BCR+0x0402) /* SCIF Bit rate control */ -#define PA_SCSCR (PA_BCR+0x0404) /* SCIF Serial control */ -#define PA_SCFDTR (PA_BCR+0x0406) /* SCIF Send FIFO control */ -#define PA_SCFSR (PA_BCR+0x0408) /* SCIF Serial status control */ -#define PA_SCFRDR (PA_BCR+0x040a) /* SCIF Receive FIFO control */ -#define PA_SCFCR (PA_BCR+0x040c) /* SCIF FIFO control */ -#define PA_SCFDR (PA_BCR+0x040e) /* SCIF FIFO data control */ -#define PA_SCLSR (PA_BCR+0x0412) /* SCIF Line Status control */ -#define PA_ICCR (PA_BCR+0x0500) /* Serial control */ -#define PA_SAR (PA_BCR+0x0502) /* Serial Slave control */ -#define PA_MDR (PA_BCR+0x0504) /* Serial Mode control */ -#define PA_ADR1 (PA_BCR+0x0506) /* Serial Address1 control */ -#define PA_DAR1 (PA_BCR+0x0546) /* Serial Data1 control */ -#define PA_VERREG (PA_BCR+0x0600) /* FPGA Version Register */ - -#define PA_AX88796L 0xa5800400 /* AX88796L Area */ -#define PA_SC1602BSLB 0xa6000000 /* SC1602BSLB Area */ -#define PA_AREA5_IO 0xb4000000 /* Area 5 IO Memory */ -#define PA_AREA6_IO 0xb8000000 /* Area 6 IO Memory */ -#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ -#define AX88796L_IO_BASE 0x1000 /* AX88796L IO Base Address */ - -#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ - -#define IRQ_PCISLOT1 0 /* PCI Slot #1 IRQ */ -#define IRQ_PCISLOT2 1 /* PCI Slot #2 IRQ */ -#define IRQ_PCISLOT3 2 /* PCI Slot #3 IRQ */ -#define IRQ_PCISLOT4 3 /* PCI Slot #4 IRQ */ -#define IRQ_CFCARD 4 /* CF Card IRQ */ -#define IRQ_CFINST 5 /* CF Card Insert IRQ */ -#define IRQ_M66596 6 /* M66596 IRQ */ -#define IRQ_SDCARD 7 /* SD Card IRQ */ -#define IRQ_TUCHPANEL 8 /* Touch Panel IRQ */ -#define IRQ_SCI 9 /* SCI IRQ */ -#define IRQ_2SERIAL 10 /* Serial IRQ */ -#define IRQ_EXTENTION 11 /* EXTn IRQ */ -#define IRQ_ONETH 12 /* On board Ethernet IRQ */ -#define IRQ_PSW 13 /* Push Switch IRQ */ -#define IRQ_ZIGBEE 14 /* Ziggbee IO IRQ */ - -#endif /* CONFIG_SH_R7780MP */ - -#define __IO_PREFIX r7780rp -#include - -#endif /* __ASM_SH_RENESAS_R7780RP */ diff --git a/include/asm-sh/rts7751r2d/rts7751r2d.h b/include/asm-sh/rts7751r2d/rts7751r2d.h deleted file mode 100644 index 796b8fc..0000000 --- a/include/asm-sh/rts7751r2d/rts7751r2d.h +++ /dev/null @@ -1,74 +0,0 @@ -#ifndef __ASM_SH_RENESAS_RTS7751R2D_H -#define __ASM_SH_RENESAS_RTS7751R2D_H - -/* - * linux/include/asm-sh/renesas_rts7751r2d.h - * - * Copyright (C) 2000 Atom Create Engineering Co., Ltd. - * - * Renesas Technology Sales RTS7751R2D support - */ - -/* Box specific addresses. */ - -#define PA_BCR 0xa4000000 /* FPGA */ -#define PA_IRLMON 0xa4000002 /* Interrupt Status control */ -#define PA_CFCTL 0xa4000004 /* CF Timing control */ -#define PA_CFPOW 0xa4000006 /* CF Power control */ -#define PA_DISPCTL 0xa4000008 /* Display Timing control */ -#define PA_SDMPOW 0xa400000a /* SD Power control */ -#define PA_RTCCE 0xa400000c /* RTC(9701) Enable control */ -#define PA_PCICD 0xa400000e /* PCI Extention detect control */ -#define PA_VOYAGERRTS 0xa4000020 /* VOYAGER Reset control */ -#if defined(CONFIG_RTS7751R2D_REV11) -#define PA_AXRST 0xa4000022 /* AX_LAN Reset control */ -#define PA_CFRST 0xa4000024 /* CF Reset control */ -#define PA_ADMRTS 0xa4000026 /* SD Reset control */ -#define PA_EXTRST 0xa4000028 /* Extention Reset control */ -#define PA_CFCDINTCLR 0xa400002a /* CF Insert Interrupt clear */ -#else -#define PA_CFRST 0xa4000022 /* CF Reset control */ -#define PA_ADMRTS 0xa4000024 /* SD Reset control */ -#define PA_EXTRST 0xa4000026 /* Extention Reset control */ -#define PA_CFCDINTCLR 0xa4000028 /* CF Insert Interrupt clear */ -#define PA_KEYCTLCLR 0xa400002a /* Key Interrupt clear */ -#endif -#define PA_POWOFF 0xa4000030 /* Board Power OFF control */ -#define PA_VERREG 0xa4000032 /* FPGA Version Register */ -#define PA_INPORT 0xa4000034 /* KEY Input Port control */ -#define PA_OUTPORT 0xa4000036 /* LED control */ -#define PA_DMPORT 0xa4000038 /* DM270 Output Port control */ - -#define PA_AX88796L 0xaa000400 /* AX88796L Area */ -#define PA_VOYAGER 0xab000000 /* VOYAGER GX Area */ -#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ -#define AX88796L_IO_BASE 0x1000 /* AX88796L IO Base Address */ - -#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ - -#if defined(CONFIG_RTS7751R2D_REV11) -#define IRQ_PCIETH 0 /* PCI Ethernet IRQ */ -#define IRQ_CFCARD 1 /* CF Card IRQ */ -#define IRQ_CFINST 2 /* CF Card Insert IRQ */ -#define IRQ_PCMCIA 3 /* PCMCIA IRQ */ -#define IRQ_VOYAGER 4 /* VOYAGER IRQ */ -#define IRQ_ONETH 5 /* On board Ethernet IRQ */ -#else -#define IRQ_KEYIN 0 /* Key Input IRQ */ -#define IRQ_PCIETH 1 /* PCI Ethernet IRQ */ -#define IRQ_CFCARD 2 /* CF Card IRQ */ -#define IRQ_CFINST 3 /* CF Card Insert IRQ */ -#define IRQ_PCMCIA 4 /* PCMCIA IRQ */ -#define IRQ_VOYAGER 5 /* VOYAGER IRQ */ -#endif -#define IRQ_RTCALM 6 /* RTC Alarm IRQ */ -#define IRQ_RTCTIME 7 /* RTC Timer IRQ */ -#define IRQ_SDCARD 8 /* SD Card IRQ */ -#define IRQ_PCISLOT1 9 /* PCI Slot #1 IRQ */ -#define IRQ_PCISLOT2 10 /* PCI Slot #2 IRQ */ -#define IRQ_EXTENTION 11 /* EXTn IRQ */ - -#define __IO_PREFIX rts7751r2d -#include - -#endif /* __ASM_SH_RENESAS_RTS7751R2D */ diff --git a/include/asm-sh/shmin/shmin.h b/include/asm-sh/shmin/shmin.h deleted file mode 100644 index 36ba138..0000000 --- a/include/asm-sh/shmin/shmin.h +++ /dev/null @@ -1,9 +0,0 @@ -#ifndef __ASM_SH_SHMIN_H -#define __ASM_SH_SHMIN_H - -#define SHMIN_IO_BASE 0xb0000000UL - -#define SHMIN_NE_IRQ IRQ2_IRQ -#define SHMIN_NE_BASE 0x300 - -#endif diff --git a/sound/oss/sh_dac_audio.c b/sound/oss/sh_dac_audio.c index 3b3b4da..51f5541 100644 --- a/sound/oss/sh_dac_audio.c +++ b/sound/oss/sh_dac_audio.c @@ -26,7 +26,7 @@ #include #include #include -#include +#include #include #define MODNAME "sh_dac_audio" -- cgit v0.10.2 From 1f666587dbf6bc660b23d8dd8abb6c572ce3eae5 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 19 Oct 2006 16:20:25 +0900 Subject: sh: Fix exception_handling_table alignment. With the recent change ripping out interrupt_table, explicit padding of the table was missing, causing bad things to happen when manually inserting handlers in to the table. This problem particularly showed up in relation to do_fpu_state_restore() which was inserted quite deeply in to the table and ended up scribbling over a slab object. Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/cpu/sh3/ex.S b/arch/sh/kernel/cpu/sh3/ex.S index 6be46f0..ba3082d 100644 --- a/arch/sh/kernel/cpu/sh3/ex.S +++ b/arch/sh/kernel/cpu/sh3/ex.S @@ -4,7 +4,7 @@ * The SH-3 exception vector table. * Copyright (C) 1999, 2000, 2002 Niibe Yutaka - * Copyright (C) 2003 Paul Mundt + * Copyright (C) 2003 - 2006 Paul Mundt * * This file is subject to the terms and conditions of the GNU General Public * License. See the file "COPYING" in the main directory of this archive @@ -49,3 +49,10 @@ ENTRY(nmi_slot) #endif ENTRY(user_break_point_trap) .long break_point_trap /* 1E0 */ + + /* + * Pad the remainder of the table out, exceptions residing in far + * away offsets can be manually inserted in to their appropriate + * location via set_exception_table_{evt,vec}(). + */ + .balign 4096,0,4096 diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S index 3f4cd04..ac8ab57 100644 --- a/arch/sh/kernel/cpu/sh4/ex.S +++ b/arch/sh/kernel/cpu/sh4/ex.S @@ -4,7 +4,7 @@ * The SH-4 exception vector table. * Copyright (C) 1999, 2000, 2002 Niibe Yutaka - * Copyright (C) 2003 Paul Mundt + * Copyright (C) 2003 - 2006 Paul Mundt * * This file is subject to the terms and conditions of the GNU General Public * License. See the file "COPYING" in the main directory of this archive @@ -53,3 +53,10 @@ ENTRY(nmi_slot) #endif ENTRY(user_break_point_trap) .long break_point_trap /* 1E0 */ + + /* + * Pad the remainder of the table out, exceptions residing in far + * away offsets can be manually inserted in to their appropriate + * location via set_exception_table_{evt,vec}(). + */ + .balign 4096,0,4096 diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c index ffe127f..53dfa55 100644 --- a/arch/sh/kernel/traps.c +++ b/arch/sh/kernel/traps.c @@ -11,27 +11,15 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. */ -#include #include -#include -#include #include -#include -#include -#include -#include #include -#include #include #include #include - +#include #include #include -#include -#include -#include -#include #ifdef CONFIG_SH_KGDB #include @@ -581,7 +569,10 @@ int is_dsp_inst(struct pt_regs *regs) #define is_dsp_inst(regs) (0) #endif /* CONFIG_SH_DSP */ -extern int do_fpu_inst(unsigned short, struct pt_regs*); +/* arch/sh/kernel/cpu/sh4/fpu.c */ +extern int do_fpu_inst(unsigned short, struct pt_regs *); +extern asmlinkage void do_fpu_state_restore(unsigned long r4, unsigned long r5, + unsigned long r6, unsigned long r7, struct pt_regs regs); asmlinkage void do_reserved_inst(unsigned long r4, unsigned long r5, unsigned long r6, unsigned long r7, @@ -740,14 +731,20 @@ void __init per_cpu_trap_init(void) : "memory"); } -void __init trap_init(void) +void *set_exception_table_vec(unsigned int vec, void *handler) { extern void *exception_handling_table[]; + void *old_handler; + + old_handler = exception_handling_table[vec]; + exception_handling_table[vec] = handler; + return old_handler; +} - exception_handling_table[TRAP_RESERVED_INST] - = (void *)do_reserved_inst; - exception_handling_table[TRAP_ILLEGAL_SLOT_INST] - = (void *)do_illegal_slot_inst; +void __init trap_init(void) +{ + set_exception_table_vec(TRAP_RESERVED_INST, do_reserved_inst); + set_exception_table_vec(TRAP_ILLEGAL_SLOT_INST, do_illegal_slot_inst); #if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SH_FPU) || \ defined(CONFIG_SH_FPU_EMU) @@ -756,9 +753,11 @@ void __init trap_init(void) * reserved. They'll be handled in the math-emu case, or faulted on * otherwise. */ - /* entry 64 corresponds to EXPEVT=0x800 */ - exception_handling_table[64] = (void *)do_reserved_inst; - exception_handling_table[65] = (void *)do_illegal_slot_inst; + set_exception_table_evt(0x800, do_reserved_inst); + set_exception_table_evt(0x820, do_illegal_slot_inst); +#elif defined(CONFIG_SH_FPU) + set_exception_table_evt(0x800, do_fpu_state_restore); + set_exception_table_evt(0x820, do_fpu_state_restore); #endif /* Setup VBR for boot cpu */ diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h index 6c1f8fd..3340126 100644 --- a/include/asm-sh/system.h +++ b/include/asm-sh/system.h @@ -353,6 +353,13 @@ static inline unsigned long __cmpxchg(volatile void * ptr, unsigned long old, (unsigned long)_n_, sizeof(*(ptr))); \ }) +extern void *set_exception_table_vec(unsigned int vec, void *handler); + +static inline void *set_exception_table_evt(unsigned int evt, void *handler) +{ + return set_exception_table_vec(evt >> 5, handler); +} + /* XXX * disable hlt during certain critical i/o operations */ -- cgit v0.10.2 From c2a560f5334c55da1e8bfa17586cc1d4e7f8ed85 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 19 Oct 2006 17:31:22 +0900 Subject: sh: Add some missing board headers. Some of these were dropped in the header directory rework, add the few missing ones back in. Signed-off-by: Paul Mundt diff --git a/include/asm-sh/edosk7705.h b/include/asm-sh/edosk7705.h new file mode 100644 index 0000000..a1089a6 --- /dev/null +++ b/include/asm-sh/edosk7705.h @@ -0,0 +1,30 @@ +/* + * include/asm-sh/edosk7705/io.h + * + * Modified version of io_se.h for the EDOSK7705 specific functions. + * + * May be copied or modified under the terms of the GNU General Public + * License. See linux/COPYING for more information. + * + * IO functions for an Hitachi EDOSK7705 development board + */ + +#ifndef __ASM_SH_EDOSK7705_IO_H +#define __ASM_SH_EDOSK7705_IO_H + +#include + +extern unsigned char sh_edosk7705_inb(unsigned long port); +extern unsigned int sh_edosk7705_inl(unsigned long port); + +extern void sh_edosk7705_outb(unsigned char value, unsigned long port); +extern void sh_edosk7705_outl(unsigned int value, unsigned long port); + +extern void sh_edosk7705_insb(unsigned long port, void *addr, unsigned long count); +extern void sh_edosk7705_insl(unsigned long port, void *addr, unsigned long count); +extern void sh_edosk7705_outsb(unsigned long port, const void *addr, unsigned long count); +extern void sh_edosk7705_outsl(unsigned long port, const void *addr, unsigned long count); + +extern unsigned long sh_edosk7705_isa_port2addr(unsigned long offset); + +#endif /* __ASM_SH_EDOSK7705_IO_H */ diff --git a/include/asm-sh/hp6xx.h b/include/asm-sh/hp6xx.h new file mode 100644 index 0000000..f35134c --- /dev/null +++ b/include/asm-sh/hp6xx.h @@ -0,0 +1,80 @@ +#ifndef __ASM_SH_HP6XX_H +#define __ASM_SH_HP6XX_H + +/* + * Copyright (C) 2003, 2004, 2005 Andriy Skulysh + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + */ + +#define HP680_BTN_IRQ IRQ0_IRQ +#define HP680_TS_IRQ IRQ3_IRQ +#define HP680_HD64461_IRQ IRQ4_IRQ + +#define DAC_LCD_BRIGHTNESS 0 +#define DAC_SPEAKER_VOLUME 1 + +#define PGDR_OPENED 0x01 +#define PGDR_MAIN_BATTERY_OUT 0x04 +#define PGDR_PLAY_BUTTON 0x08 +#define PGDR_REWIND_BUTTON 0x10 +#define PGDR_RECORD_BUTTON 0x20 + +#define PHDR_TS_PEN_DOWN 0x08 + +#define PJDR_LED_BLINK 0x02 + +#define PKDR_LED_GREEN 0x10 + +#define SCPDR_TS_SCAN_ENABLE 0x20 +#define SCPDR_TS_SCAN_Y 0x02 +#define SCPDR_TS_SCAN_X 0x01 + +#define SCPCR_TS_ENABLE 0x405 +#define SCPCR_TS_MASK 0xc0f + +#define ADC_CHANNEL_TS_Y 1 +#define ADC_CHANNEL_TS_X 2 +#define ADC_CHANNEL_BATTERY 3 +#define ADC_CHANNEL_BACKUP 4 +#define ADC_CHANNEL_CHARGE 5 + +#define HD64461_GPADR_SPEAKER 0x01 +#define HD64461_GPADR_PCMCIA0 (0x02|0x08) + +#define HD64461_GPBDR_LCDOFF 0x01 +#define HD64461_GPBDR_LCD_CONTRAST_MASK 0x78 +#define HD64461_GPBDR_LED_RED 0x80 + +#include +#include + +#define PJDR 0xa4000130 +#define PKDR 0xa4000132 + +static inline void hp6xx_led_red(int on) +{ + u16 v16; + v16 = ctrl_inw(CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000); + if (on) + ctrl_outw(v16 & (~HD64461_GPBDR_LED_RED), CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000); + else + ctrl_outw(v16 | HD64461_GPBDR_LED_RED, CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000); +} + +static inline void hp6xx_led_green(int on) +{ + u8 v8; + + v8 = ctrl_inb(PKDR); + if (on) + ctrl_outb(v8 & (~PKDR_LED_GREEN), PKDR); + else + ctrl_outb(v8 | PKDR_LED_GREEN, PKDR); +} + + +#endif /* __ASM_SH_HP6XX_H */ diff --git a/include/asm-sh/hs7751rvoip.h b/include/asm-sh/hs7751rvoip.h new file mode 100644 index 0000000..c4cff9d --- /dev/null +++ b/include/asm-sh/hs7751rvoip.h @@ -0,0 +1,54 @@ +#ifndef __ASM_SH_RENESAS_HS7751RVOIP_H +#define __ASM_SH_RENESAS_HS7751RVOIP_H + +/* + * linux/include/asm-sh/hs7751rvoip/hs7751rvoip.h + * + * Copyright (C) 2000 Atom Create Engineering Co., Ltd. + * + * Renesas Technology Sales HS7751RVoIP support + */ + +/* Box specific addresses. */ + +#define PA_BCR 0xa4000000 /* FPGA */ +#define PA_SLICCNTR1 0xa4000006 /* SLIC PIO Control 1 */ +#define PA_SLICCNTR2 0xa4000008 /* SLIC PIO Control 2 */ +#define PA_DMACNTR 0xa400000a /* USB DMA Control */ +#define PA_INPORTR 0xa400000c /* Input Port Register */ +#define PA_OUTPORTR 0xa400000e /* Output Port Reguster */ +#define PA_VERREG 0xa4000014 /* FPGA Version Register */ + +#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ + +#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ +#define IRLCNTR2 (PA_BCR + 2) /* Interrupt Control Register2 */ +#define IRLCNTR3 (PA_BCR + 4) /* Interrupt Control Register3 */ +#define IRLCNTR4 (PA_BCR + 16) /* Interrupt Control Register4 */ +#define IRLCNTR5 (PA_BCR + 18) /* Interrupt Control Register5 */ + +#define IRQ_PCIETH 6 /* PCI Ethernet IRQ */ +#define IRQ_PCIHUB 7 /* PCI Ethernet Hub IRQ */ +#define IRQ_USBCOM 8 /* USB Comunication IRQ */ +#define IRQ_USBCON 9 /* USB Connect IRQ */ +#define IRQ_USBDMA 10 /* USB DMA IRQ */ +#define IRQ_CFCARD 11 /* CF Card IRQ */ +#define IRQ_PCMCIA 12 /* PCMCIA IRQ */ +#define IRQ_PCISLOT 13 /* PCI Slot #1 IRQ */ +#define IRQ_ONHOOK1 0 /* ON HOOK1 IRQ */ +#define IRQ_OFFHOOK1 1 /* OFF HOOK1 IRQ */ +#define IRQ_ONHOOK2 2 /* ON HOOK2 IRQ */ +#define IRQ_OFFHOOK2 3 /* OFF HOOK2 IRQ */ +#define IRQ_RINGING 4 /* Ringing IRQ */ +#define IRQ_CODEC 5 /* CODEC IRQ */ + +#define __IO_PREFIX hs7751rvoip +#include + +/* arch/sh/boards/renesas/hs7751rvoip/irq.c */ +void init_hs7751rvoip_IRQ(void); + +/* arch/sh/boards/renesas/hs7751rvoip/io.c */ +void *hs7751rvoip_ioremap(unsigned long, unsigned long); + +#endif /* __ASM_SH_RENESAS_HS7751RVOIP */ diff --git a/include/asm-sh/r7780rp.h b/include/asm-sh/r7780rp.h new file mode 100644 index 0000000..ddd67b6 --- /dev/null +++ b/include/asm-sh/r7780rp.h @@ -0,0 +1,173 @@ +#ifndef __ASM_SH_RENESAS_R7780RP_H +#define __ASM_SH_RENESAS_R7780RP_H + +/* + * linux/include/asm-sh/r7780rp.h + * + * Copyright (C) 2000 Atom Create Engineering Co., Ltd. + * + * Renesas Solutions Highlander R7780RP support + */ + +/* Box specific addresses. */ +#if defined(CONFIG_SH_R7780MP) +#define PA_BCR 0xa4000000 /* FPGA */ +#define PA_IRLMSK (PA_BCR+0x0000) /* Interrupt Mask control */ +#define PA_IRLMON (PA_BCR+0x0002) /* Interrupt Status control */ +#define PA_IRLPRI1 (PA_BCR+0x0004) /* Interrupt Priorty 1 */ +#define PA_IRLPRI2 (PA_BCR+0x0006) /* Interrupt Priorty 2 */ +#define PA_IRLPRI3 (PA_BCR+0x0008) /* Interrupt Priorty 3 */ +#define PA_IRLPRI4 (PA_BCR+0x000a) /* Interrupt Priorty 4 */ +#define PA_RSTCTL (PA_BCR+0x000c) /* Reset Control */ +#define PA_PCIBD (PA_BCR+0x000e) /* PCI Board detect control */ +#define PA_PCICD (PA_BCR+0x0010) /* PCI Conector detect control */ +#define PA_EXTGIO (PA_BCR+0x0016) /* Extension GPIO Control */ +#define PA_IVDRMON (PA_BCR+0x0018) /* iVDR Moniter control */ +#define PA_IVDRCTL (PA_BCR+0x001a) /* iVDR control */ +#define PA_OBLED (PA_BCR+0x001c) /* On Board LED control */ +#define PA_OBSW (PA_BCR+0x001e) /* On Board Switch control */ +#define PA_AUDIOSEL (PA_BCR+0x0020) /* Sound Interface Select control */ +#define PA_EXTPLR (PA_BCR+0x001e) /* Extention Pin Polarity control */ +#define PA_TPCTL (PA_BCR+0x0100) /* Touch Panel Access control */ +#define PA_TPDCKCTL (PA_BCR+0x0102) /* Touch Panel Access data control */ +#define PA_TPCTLCLR (PA_BCR+0x0104) /* Touch Panel Access control */ +#define PA_TPXPOS (PA_BCR+0x0106) /* Touch Panel X position control */ +#define PA_TPYPOS (PA_BCR+0x0108) /* Touch Panel Y position control */ +#define PA_DBSW (PA_BCR+0x0200) /* Debug Board Switch control */ +#define PA_CFCTL (PA_BCR+0x0300) /* CF Timing control */ +#define PA_CFPOW (PA_BCR+0x0302) /* CF Power control */ +#define PA_CFCDINTCLR (PA_BCR+0x0304) /* CF Insert Interrupt clear */ +#define PA_SCSMR0 (PA_BCR+0x0400) /* SCIF0 Serial mode control */ +#define PA_SCBRR0 (PA_BCR+0x0404) /* SCIF0 Bit rate control */ +#define PA_SCSCR0 (PA_BCR+0x0408) /* SCIF0 Serial control */ +#define PA_SCFTDR0 (PA_BCR+0x040c) /* SCIF0 Send FIFO control */ +#define PA_SCFSR0 (PA_BCR+0x0410) /* SCIF0 Serial status control */ +#define PA_SCFRDR0 (PA_BCR+0x0414) /* SCIF0 Receive FIFO control */ +#define PA_SCFCR0 (PA_BCR+0x0418) /* SCIF0 FIFO control */ +#define PA_SCTFDR0 (PA_BCR+0x041c) /* SCIF0 Send FIFO data control */ +#define PA_SCRFDR0 (PA_BCR+0x0420) /* SCIF0 Receive FIFO data control */ +#define PA_SCSPTR0 (PA_BCR+0x0424) /* SCIF0 Serial Port control */ +#define PA_SCLSR0 (PA_BCR+0x0428) /* SCIF0 Line Status control */ +#define PA_SCRER0 (PA_BCR+0x042c) /* SCIF0 Serial Error control */ +#define PA_SCSMR1 (PA_BCR+0x0500) /* SCIF1 Serial mode control */ +#define PA_SCBRR1 (PA_BCR+0x0504) /* SCIF1 Bit rate control */ +#define PA_SCSCR1 (PA_BCR+0x0508) /* SCIF1 Serial control */ +#define PA_SCFTDR1 (PA_BCR+0x050c) /* SCIF1 Send FIFO control */ +#define PA_SCFSR1 (PA_BCR+0x0510) /* SCIF1 Serial status control */ +#define PA_SCFRDR1 (PA_BCR+0x0514) /* SCIF1 Receive FIFO control */ +#define PA_SCFCR1 (PA_BCR+0x0518) /* SCIF1 FIFO control */ +#define PA_SCTFDR1 (PA_BCR+0x051c) /* SCIF1 Send FIFO data control */ +#define PA_SCRFDR1 (PA_BCR+0x0520) /* SCIF1 Receive FIFO data control */ +#define PA_SCSPTR1 (PA_BCR+0x0524) /* SCIF1 Serial Port control */ +#define PA_SCLSR1 (PA_BCR+0x0528) /* SCIF1 Line Status control */ +#define PA_SCRER1 (PA_BCR+0x052c) /* SCIF1 Serial Error control */ +#define PA_ICCR (PA_BCR+0x0600) /* Serial control */ +#define PA_SAR (PA_BCR+0x0602) /* Serial Slave control */ +#define PA_MDR (PA_BCR+0x0604) /* Serial Mode control */ +#define PA_ADR1 (PA_BCR+0x0606) /* Serial Address1 control */ +#define PA_DAR1 (PA_BCR+0x0646) /* Serial Data1 control */ +#define PA_VERREG (PA_BCR+0x0700) /* FPGA Version Register */ +#define PA_POFF (PA_BCR+0x0800) /* System Power Off control */ +#define PA_PMR (PA_BCR+0x0900) /* */ + +#define PA_AX88796L 0xa4100400 /* AX88796L Area */ +#define PA_SC1602BSLB 0xa6000000 /* SC1602BSLB Area */ +#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ +#define AX88796L_IO_BASE 0x1000 /* AX88796L IO Base Address */ + +#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ + +#define IRQ_PCISLOT1 65 /* PCI Slot #1 IRQ */ +#define IRQ_PCISLOT2 66 /* PCI Slot #2 IRQ */ +#define IRQ_PCISLOT3 67 /* PCI Slot #3 IRQ */ +#define IRQ_PCISLOT4 68 /* PCI Slot #4 IRQ */ +#define IRQ_CFCARD 1 /* CF Card IRQ */ +// #define IRQ_CFINST 0 /* CF Card Insert IRQ */ +#define IRQ_TP 2 /* Touch Panel IRQ */ +#define IRQ_SCI1 3 /* SCI1 IRQ */ +#define IRQ_SCI0 4 /* SCI0 IRQ */ +#define IRQ_2SERIAL 5 /* Serial IRQ */ +#define IRQ_RTC 6 /* RTC A / B IRQ */ +#define IRQ_EXTENTION6 7 /* EXT6n IRQ */ +#define IRQ_EXTENTION5 8 /* EXT5n IRQ */ +#define IRQ_EXTENTION4 9 /* EXT4n IRQ */ +#define IRQ_EXTENTION2 10 /* EXT2n IRQ */ +#define IRQ_EXTENTION1 11 /* EXT1n IRQ */ +#define IRQ_ONETH 13 /* On board Ethernet IRQ */ +#define IRQ_PSW 14 /* Push Switch IRQ */ + +#else /* R7780RP */ + +#define PA_BCR 0xa5000000 /* FPGA */ +#define PA_IRLMSK (PA_BCR+0x0000) /* Interrupt Mask control */ +#define PA_IRLMON (PA_BCR+0x0002) /* Interrupt Status control */ +#define PA_SDPOW (PA_BCR+0x0004) /* SD Power control */ +#define PA_RSTCTL (PA_BCR+0x0006) /* Device Reset control */ +#define PA_PCIBD (PA_BCR+0x0008) /* PCI Board detect control */ +#define PA_PCICD (PA_BCR+0x000a) /* PCI Conector detect control */ +#define PA_ZIGIO1 (PA_BCR+0x000c) /* Zigbee IO control 1 */ +#define PA_ZIGIO2 (PA_BCR+0x000e) /* Zigbee IO control 2 */ +#define PA_ZIGIO3 (PA_BCR+0x0010) /* Zigbee IO control 3 */ +#define PA_ZIGIO4 (PA_BCR+0x0012) /* Zigbee IO control 4 */ +#define PA_IVDRMON (PA_BCR+0x0014) /* iVDR Moniter control */ +#define PA_IVDRCTL (PA_BCR+0x0016) /* iVDR control */ +#define PA_OBLED (PA_BCR+0x0018) /* On Board LED control */ +#define PA_OBSW (PA_BCR+0x001a) /* On Board Switch control */ +#define PA_AUDIOSEL (PA_BCR+0x001c) /* Sound Interface Select control */ +#define PA_EXTPLR (PA_BCR+0x001e) /* Extention Pin Polarity control */ +#define PA_TPCTL (PA_BCR+0x0100) /* Touch Panel Access control */ +#define PA_TPDCKCTL (PA_BCR+0x0102) /* Touch Panel Access data control */ +#define PA_TPCTLCLR (PA_BCR+0x0104) /* Touch Panel Access control */ +#define PA_TPXPOS (PA_BCR+0x0106) /* Touch Panel X position control */ +#define PA_TPYPOS (PA_BCR+0x0108) /* Touch Panel Y position control */ +#define PA_DBDET (PA_BCR+0x0200) /* Debug Board detect control */ +#define PA_DBDISPCTL (PA_BCR+0x0202) /* Debug Board Dot timing control */ +#define PA_DBSW (PA_BCR+0x0204) /* Debug Board Switch control */ +#define PA_CFCTL (PA_BCR+0x0300) /* CF Timing control */ +#define PA_CFPOW (PA_BCR+0x0302) /* CF Power control */ +#define PA_CFCDINTCLR (PA_BCR+0x0304) /* CF Insert Interrupt clear */ +#define PA_SCSMR (PA_BCR+0x0400) /* SCIF Serial mode control */ +#define PA_SCBRR (PA_BCR+0x0402) /* SCIF Bit rate control */ +#define PA_SCSCR (PA_BCR+0x0404) /* SCIF Serial control */ +#define PA_SCFDTR (PA_BCR+0x0406) /* SCIF Send FIFO control */ +#define PA_SCFSR (PA_BCR+0x0408) /* SCIF Serial status control */ +#define PA_SCFRDR (PA_BCR+0x040a) /* SCIF Receive FIFO control */ +#define PA_SCFCR (PA_BCR+0x040c) /* SCIF FIFO control */ +#define PA_SCFDR (PA_BCR+0x040e) /* SCIF FIFO data control */ +#define PA_SCLSR (PA_BCR+0x0412) /* SCIF Line Status control */ +#define PA_ICCR (PA_BCR+0x0500) /* Serial control */ +#define PA_SAR (PA_BCR+0x0502) /* Serial Slave control */ +#define PA_MDR (PA_BCR+0x0504) /* Serial Mode control */ +#define PA_ADR1 (PA_BCR+0x0506) /* Serial Address1 control */ +#define PA_DAR1 (PA_BCR+0x0546) /* Serial Data1 control */ +#define PA_VERREG (PA_BCR+0x0600) /* FPGA Version Register */ + +#define PA_AX88796L 0xa5800400 /* AX88796L Area */ +#define PA_SC1602BSLB 0xa6000000 /* SC1602BSLB Area */ +#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ +#define AX88796L_IO_BASE 0x1000 /* AX88796L IO Base Address */ + +#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ + +#define IRQ_PCISLOT1 0 /* PCI Slot #1 IRQ */ +#define IRQ_PCISLOT2 1 /* PCI Slot #2 IRQ */ +#define IRQ_PCISLOT3 2 /* PCI Slot #3 IRQ */ +#define IRQ_PCISLOT4 3 /* PCI Slot #4 IRQ */ +#define IRQ_CFCARD 4 /* CF Card IRQ */ +#define IRQ_CFINST 5 /* CF Card Insert IRQ */ +#define IRQ_M66596 6 /* M66596 IRQ */ +#define IRQ_SDCARD 7 /* SD Card IRQ */ +#define IRQ_TUCHPANEL 8 /* Touch Panel IRQ */ +#define IRQ_SCI 9 /* SCI IRQ */ +#define IRQ_2SERIAL 10 /* Serial IRQ */ +#define IRQ_EXTENTION 11 /* EXTn IRQ */ +#define IRQ_ONETH 12 /* On board Ethernet IRQ */ +#define IRQ_PSW 13 /* Push Switch IRQ */ +#define IRQ_ZIGBEE 14 /* Ziggbee IO IRQ */ + +#endif /* CONFIG_SH_R7780MP */ + +#define __IO_PREFIX r7780rp +#include + +#endif /* __ASM_SH_RENESAS_R7780RP */ diff --git a/include/asm-sh/rts7751r2d.h b/include/asm-sh/rts7751r2d.h new file mode 100644 index 0000000..796b8fc --- /dev/null +++ b/include/asm-sh/rts7751r2d.h @@ -0,0 +1,74 @@ +#ifndef __ASM_SH_RENESAS_RTS7751R2D_H +#define __ASM_SH_RENESAS_RTS7751R2D_H + +/* + * linux/include/asm-sh/renesas_rts7751r2d.h + * + * Copyright (C) 2000 Atom Create Engineering Co., Ltd. + * + * Renesas Technology Sales RTS7751R2D support + */ + +/* Box specific addresses. */ + +#define PA_BCR 0xa4000000 /* FPGA */ +#define PA_IRLMON 0xa4000002 /* Interrupt Status control */ +#define PA_CFCTL 0xa4000004 /* CF Timing control */ +#define PA_CFPOW 0xa4000006 /* CF Power control */ +#define PA_DISPCTL 0xa4000008 /* Display Timing control */ +#define PA_SDMPOW 0xa400000a /* SD Power control */ +#define PA_RTCCE 0xa400000c /* RTC(9701) Enable control */ +#define PA_PCICD 0xa400000e /* PCI Extention detect control */ +#define PA_VOYAGERRTS 0xa4000020 /* VOYAGER Reset control */ +#if defined(CONFIG_RTS7751R2D_REV11) +#define PA_AXRST 0xa4000022 /* AX_LAN Reset control */ +#define PA_CFRST 0xa4000024 /* CF Reset control */ +#define PA_ADMRTS 0xa4000026 /* SD Reset control */ +#define PA_EXTRST 0xa4000028 /* Extention Reset control */ +#define PA_CFCDINTCLR 0xa400002a /* CF Insert Interrupt clear */ +#else +#define PA_CFRST 0xa4000022 /* CF Reset control */ +#define PA_ADMRTS 0xa4000024 /* SD Reset control */ +#define PA_EXTRST 0xa4000026 /* Extention Reset control */ +#define PA_CFCDINTCLR 0xa4000028 /* CF Insert Interrupt clear */ +#define PA_KEYCTLCLR 0xa400002a /* Key Interrupt clear */ +#endif +#define PA_POWOFF 0xa4000030 /* Board Power OFF control */ +#define PA_VERREG 0xa4000032 /* FPGA Version Register */ +#define PA_INPORT 0xa4000034 /* KEY Input Port control */ +#define PA_OUTPORT 0xa4000036 /* LED control */ +#define PA_DMPORT 0xa4000038 /* DM270 Output Port control */ + +#define PA_AX88796L 0xaa000400 /* AX88796L Area */ +#define PA_VOYAGER 0xab000000 /* VOYAGER GX Area */ +#define PA_IDE_OFFSET 0x1f0 /* CF IDE Offset */ +#define AX88796L_IO_BASE 0x1000 /* AX88796L IO Base Address */ + +#define IRLCNTR1 (PA_BCR + 0) /* Interrupt Control Register1 */ + +#if defined(CONFIG_RTS7751R2D_REV11) +#define IRQ_PCIETH 0 /* PCI Ethernet IRQ */ +#define IRQ_CFCARD 1 /* CF Card IRQ */ +#define IRQ_CFINST 2 /* CF Card Insert IRQ */ +#define IRQ_PCMCIA 3 /* PCMCIA IRQ */ +#define IRQ_VOYAGER 4 /* VOYAGER IRQ */ +#define IRQ_ONETH 5 /* On board Ethernet IRQ */ +#else +#define IRQ_KEYIN 0 /* Key Input IRQ */ +#define IRQ_PCIETH 1 /* PCI Ethernet IRQ */ +#define IRQ_CFCARD 2 /* CF Card IRQ */ +#define IRQ_CFINST 3 /* CF Card Insert IRQ */ +#define IRQ_PCMCIA 4 /* PCMCIA IRQ */ +#define IRQ_VOYAGER 5 /* VOYAGER IRQ */ +#endif +#define IRQ_RTCALM 6 /* RTC Alarm IRQ */ +#define IRQ_RTCTIME 7 /* RTC Timer IRQ */ +#define IRQ_SDCARD 8 /* SD Card IRQ */ +#define IRQ_PCISLOT1 9 /* PCI Slot #1 IRQ */ +#define IRQ_PCISLOT2 10 /* PCI Slot #2 IRQ */ +#define IRQ_EXTENTION 11 /* EXTn IRQ */ + +#define __IO_PREFIX rts7751r2d +#include + +#endif /* __ASM_SH_RENESAS_RTS7751R2D */ diff --git a/include/asm-sh/shmin.h b/include/asm-sh/shmin.h new file mode 100644 index 0000000..36ba138 --- /dev/null +++ b/include/asm-sh/shmin.h @@ -0,0 +1,9 @@ +#ifndef __ASM_SH_SHMIN_H +#define __ASM_SH_SHMIN_H + +#define SHMIN_IO_BASE 0xb0000000UL + +#define SHMIN_NE_IRQ IRQ2_IRQ +#define SHMIN_NE_BASE 0x300 + +#endif -- cgit v0.10.2 From 709bc44c31db4eeeec7dcf7d3f3fefd057adf7fb Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 19 Oct 2006 17:32:56 +0900 Subject: sh: Updates for irq-flow-type naming changes. handle_irq_name() and set_irq_chip_and_handler() disappeared, update for desc->name and set_irq_chip_and_handler_name() use. Signed-off-by: Paul Mundt diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c index cbb4ea2..aa15ec5 100644 --- a/arch/sh/boards/renesas/r7780rp/irq.c +++ b/arch/sh/boards/renesas/r7780rp/irq.c @@ -33,7 +33,7 @@ static void disable_r7780rp_irq(unsigned int irq) } static struct irq_chip r7780rp_irq_chip __read_mostly = { - .name = "r7780rp", + .name = "R7780RP", .mask = disable_r7780rp_irq, .unmask = enable_r7780rp_irq, .mask_ack = disable_r7780rp_irq, @@ -48,8 +48,8 @@ void __init init_r7780rp_IRQ(void) for (i = 0; i < 15; i++) { disable_irq_nosync(i); - set_irq_chip_and_handler(i, &r7780rp_irq_chip, - handle_level_irq); + set_irq_chip_and_handler_name(i, &r7780rp_irq_chip, + handle_level_irq, "level"); enable_r7780rp_irq(i); } } diff --git a/arch/sh/kernel/cpu/irq/intc2.c b/arch/sh/kernel/cpu/irq/intc2.c index d4b2bb7..212884a 100644 --- a/arch/sh/kernel/cpu/irq/intc2.c +++ b/arch/sh/kernel/cpu/irq/intc2.c @@ -31,7 +31,7 @@ static void enable_intc2_irq(unsigned int irq) } static struct irq_chip intc2_irq_chip = { - .typename = "intc2", + .name = "INTC2", .mask = disable_intc2_irq, .unmask = enable_intc2_irq, .mask_ack = disable_intc2_irq, @@ -64,7 +64,8 @@ void make_intc2_irq(struct intc2_data *p) local_irq_restore(flags); - set_irq_chip_and_handler(p->irq, &intc2_irq_chip, handle_level_irq); + set_irq_chip_and_handler_name(p->irq, &intc2_irq_chip, + handle_level_irq, "level"); set_irq_chip_data(p->irq, p); enable_intc2_irq(p->irq); diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c index 8944abd..f799731 100644 --- a/arch/sh/kernel/cpu/irq/ipr.c +++ b/arch/sh/kernel/cpu/irq/ipr.c @@ -44,7 +44,7 @@ static void enable_ipr_irq(unsigned int irq) } static struct irq_chip ipr_irq_chip = { - .name = "ipr", + .name = "IPR", .mask = disable_ipr_irq, .unmask = enable_ipr_irq, .mask_ack = disable_ipr_irq, @@ -60,7 +60,8 @@ void make_ipr_irq(unsigned int irq, unsigned int addr, int pos, int priority) ipr_data.shift = pos*4; /* POSition (0-3) x 4 means shift */ ipr_data.priority = priority; - set_irq_chip_and_handler(irq, &ipr_irq_chip, handle_level_irq); + set_irq_chip_and_handler_name(irq, &ipr_irq_chip, + handle_level_irq, "level"); set_irq_chip_data(irq, &ipr_data); enable_ipr_irq(irq); diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c index acf2602..944128c 100644 --- a/arch/sh/kernel/irq.c +++ b/arch/sh/kernel/irq.c @@ -54,7 +54,7 @@ int show_interrupts(struct seq_file *p, void *v) for_each_online_cpu(j) seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); seq_printf(p, " %14s", irq_desc[i].chip->name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) -- cgit v0.10.2 From 55b7428303d390c53d3a1bc587de8592ce65900e Mon Sep 17 00:00:00 2001 From: Franck Bui-Huu Date: Fri, 13 Oct 2006 13:37:35 +0200 Subject: [MIPS] Use kallsyms_lookup_size_offset() instead of kallsyms_lookup() This new routine doesn't lookup for symbol names. So we needn't to pass any char buffers or pointer since we don't care about names. Signed-off-by: Franck Bui-Huu Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c index 9f307eb..ec8209f 100644 --- a/arch/mips/kernel/process.c +++ b/arch/mips/kernel/process.c @@ -358,10 +358,8 @@ static int __init frame_info_init(void) unsigned long size = 0; #ifdef CONFIG_KALLSYMS unsigned long ofs; - char *modname; - char namebuf[KSYM_NAME_LEN + 1]; - kallsyms_lookup((unsigned long)schedule, &size, &ofs, &modname, namebuf); + kallsyms_lookup_size_offset((unsigned long)schedule, &size, &ofs); #endif schedule_mfi.func = schedule; schedule_mfi.func_size = size; @@ -403,8 +401,6 @@ unsigned long unwind_stack(struct task_struct *task, unsigned long *sp, { unsigned long stack_page; struct mips_frame_info info; - char *modname; - char namebuf[KSYM_NAME_LEN + 1]; unsigned long size, ofs; int leaf; extern void ret_from_irq(void); @@ -433,7 +429,7 @@ unsigned long unwind_stack(struct task_struct *task, unsigned long *sp, } return 0; } - if (!kallsyms_lookup(pc, &size, &ofs, &modname, namebuf)) + if (!kallsyms_lookup_size_offset(pc, &size, &ofs)) return 0; /* * Return ra if an exception occured at the first instruction -- cgit v0.10.2 From 22c56c3a03b377d21d8363f737aa2a855d892458 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 15 Oct 2006 09:15:19 +0100 Subject: [MIPS] Delete unneeded pt_regs forward declaration. Signed-off-by: Ralf Baechle diff --git a/arch/mips/oprofile/op_impl.h b/arch/mips/oprofile/op_impl.h index 354e544..fa6b4aa 100644 --- a/arch/mips/oprofile/op_impl.h +++ b/arch/mips/oprofile/op_impl.h @@ -10,8 +10,6 @@ #ifndef OP_IMPL_H #define OP_IMPL_H 1 -struct pt_regs; - extern int null_perf_irq(void); extern int (*perf_irq)(void); -- cgit v0.10.2 From ba38cdf94285a1fc6a01fd493bdc5304bc83f61e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 15 Oct 2006 09:17:43 +0100 Subject: [MIPS] Malta: Fix uninitialized regs pointer. Signed-off-by: Ralf Baechle diff --git a/arch/mips/mips-boards/malta/malta_int.c b/arch/mips/mips-boards/malta/malta_int.c index 6244d0e..90ad5bf 100644 --- a/arch/mips/mips-boards/malta/malta_int.c +++ b/arch/mips/mips-boards/malta/malta_int.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -131,7 +132,7 @@ static void corehi_irqdispatch(void) unsigned int intedge, intsteer, pcicmd, pcibadaddr; unsigned int pcimstat, intisr, inten, intpol; unsigned int intrcause,datalo,datahi; - struct pt_regs *regs; + struct pt_regs *regs = get_irq_regs(); printk("CoreHI interrupt, shouldn't happen, so we die here!!!\n"); printk("epc : %08lx\nStatus: %08lx\n" -- cgit v0.10.2 From 36d98e79b798fb27e38b9be4f36c5a96025f9281 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 15 Oct 2006 09:19:58 +0100 Subject: [MIPS] A few more pt_regs fixups. Signed-off-by: Ralf Baechle diff --git a/arch/mips/au1000/common/dbdma.c b/arch/mips/au1000/common/dbdma.c index c4fae8f..626de44 100644 --- a/arch/mips/au1000/common/dbdma.c +++ b/arch/mips/au1000/common/dbdma.c @@ -849,7 +849,7 @@ au1xxx_dbdma_chan_free(u32 chanid) EXPORT_SYMBOL(au1xxx_dbdma_chan_free); static irqreturn_t -dbdma_interrupt(int irq, void *dev_id, struct pt_regs *regs) +dbdma_interrupt(int irq, void *dev_id) { u32 intstat; u32 chan_index; diff --git a/arch/mips/oprofile/op_model_rm9000.c b/arch/mips/oprofile/op_model_rm9000.c index b7063fe..7dc9bf6 100644 --- a/arch/mips/oprofile/op_model_rm9000.c +++ b/arch/mips/oprofile/op_model_rm9000.c @@ -80,8 +80,7 @@ static void rm9000_cpu_stop(void *args) write_c0_perfcontrol(0); } -static irqreturn_t rm9000_perfcount_handler(int irq, void * dev_id, - struct pt_regs *regs) +static irqreturn_t rm9000_perfcount_handler(int irq, void * dev_id) { unsigned int control = read_c0_perfcontrol(); uint32_t counter1, counter2; diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c index fd0932b..db80844 100644 --- a/arch/mips/sgi-ip32/ip32-reset.c +++ b/arch/mips/sgi-ip32/ip32-reset.c @@ -135,7 +135,7 @@ static inline void ip32_power_button(void) add_timer(&power_timer); } -static irqreturn_t ip32_rtc_int(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t ip32_rtc_int(int irq, void *dev_id) { volatile unsigned char reg_c; diff --git a/arch/mips/sibyte/sb1250/bcm1250_tbprof.c b/arch/mips/sibyte/sb1250/bcm1250_tbprof.c index 992e0d8..d1a906e 100644 --- a/arch/mips/sibyte/sb1250/bcm1250_tbprof.c +++ b/arch/mips/sibyte/sb1250/bcm1250_tbprof.c @@ -88,7 +88,7 @@ static void arm_tb(void) sbp.tb_armed = 1; } -static irqreturn_t sbprof_tb_intr(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t sbprof_tb_intr(int irq, void *dev_id) { int i; DBG(printk(DEVNAME ": tb_intr\n")); @@ -138,7 +138,7 @@ static irqreturn_t sbprof_tb_intr(int irq, void *dev_id, struct pt_regs *regs) return IRQ_HANDLED; } -static irqreturn_t sbprof_pc_intr(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t sbprof_pc_intr(int irq, void *dev_id) { printk(DEVNAME ": unexpected pc_intr"); return IRQ_NONE; diff --git a/arch/mips/sibyte/sb1250/bus_watcher.c b/arch/mips/sibyte/sb1250/bus_watcher.c index bb90649..45274bd 100644 --- a/arch/mips/sibyte/sb1250/bus_watcher.c +++ b/arch/mips/sibyte/sb1250/bus_watcher.c @@ -171,7 +171,7 @@ static void create_proc_decoder(struct bw_stats_struct *stats) * notes: possible re-entry due to multiple sources * should check/indicate saturation */ -static irqreturn_t sibyte_bw_int(int irq, void *data, struct pt_regs *regs) +static irqreturn_t sibyte_bw_int(int irq, void *data) { struct bw_stats_struct *stats = data; unsigned long cntr; -- cgit v0.10.2 From 53571ce47010562f5e67782ea00206f379a5cd65 Mon Sep 17 00:00:00 2001 From: Thiemo Seufer Date: Sun, 13 Aug 2006 00:53:29 +0100 Subject: [MIPS] Fix O32 personality(2) call with 0xffffffff argument. A sign extension bug did result in sys_personality being invoked with a 0xffffffffffffffffUL argument, so querying the current personality didn't work. Signed-off-by: Thiemo Seufer Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c index 53f4171..7a3ebbe 100644 --- a/arch/mips/kernel/linux32.c +++ b/arch/mips/kernel/linux32.c @@ -1055,7 +1055,9 @@ asmlinkage long sys32_newuname(struct new_utsname __user * name) asmlinkage int sys32_personality(unsigned long personality) { int ret; - if (current->personality == PER_LINUX32 && personality == PER_LINUX) + personality &= 0xffffffff; + if (personality(current->personality) == PER_LINUX32 && + personality == PER_LINUX) personality = PER_LINUX32; ret = sys_personality(personality); if (ret == PER_LINUX32) -- cgit v0.10.2 From 089c7e7f2da7c3245de47377252683bd9edae738 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 16 Oct 2006 16:49:37 +0100 Subject: [MIPS] Use compat_sys_mount. This fixes mount problems with smbfs, ncpfs and NFSv4. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 6d9f187..8c453f8 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -280,7 +280,7 @@ EXPORT(sysn32_call_table) PTR sys_sync PTR sys_acct PTR sys32_settimeofday - PTR sys_mount /* 6160 */ + PTR compat_sys_mount /* 6160 */ PTR sys_umount PTR sys_swapon PTR sys_swapoff diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 2e6d067..d105917 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -226,7 +226,7 @@ sys_call_table: PTR sys_ni_syscall /* was sys_stat */ PTR sys_lseek PTR sys_getpid /* 4020 */ - PTR sys_mount + PTR compat_sys_mount PTR sys_oldumount PTR sys_setuid PTR sys_getuid -- cgit v0.10.2 From eea32d4c6e272b6c324c8c22df4c28274fcb5a21 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Mon, 16 Oct 2006 22:48:49 +0900 Subject: [MIPS] save_context_stack fix CONFIG_KALLSYMS=n case is obviously wrong, though it is harmless since CONFIG_KALLSYMS is always enabled with CONFIG_STACKTRACE for now. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/stacktrace.c b/arch/mips/kernel/stacktrace.c index 4aabe52..a586aba 100644 --- a/arch/mips/kernel/stacktrace.c +++ b/arch/mips/kernel/stacktrace.c @@ -57,7 +57,7 @@ static void save_context_stack(struct stack_trace *trace, pc = unwind_stack(task, &sp, pc, &ra); } while (pc); #else - save_raw_context_stack(sp); + save_raw_context_stack(trace, sp); #endif } -- cgit v0.10.2 From 94399ea62fc1047eded76b45b972e7850a800a1b Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 18 Oct 2006 23:27:29 +0900 Subject: [MIPS] More vr41xx pt_regs fixups Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/include/asm-mips/vr41xx/vr41xx.h b/include/asm-mips/vr41xx/vr41xx.h index dd3eb3d..88b492f 100644 --- a/include/asm-mips/vr41xx/vr41xx.h +++ b/include/asm-mips/vr41xx/vr41xx.h @@ -75,7 +75,7 @@ extern void vr41xx_mask_clock(vr41xx_clock_t clock); * Interrupt Control Unit */ extern int vr41xx_set_intassign(unsigned int irq, unsigned char intassign); -extern int cascade_irq(unsigned int irq, int (*get_irq)(unsigned int, struct pt_regs *)); +extern int cascade_irq(unsigned int irq, int (*get_irq)(unsigned int)); #define PIUINT_COMMAND 0x0040 #define PIUINT_DATA 0x0020 -- cgit v0.10.2 From 00ddf2be159e3b96ba8872b407240e1bb449b75e Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 18 Oct 2006 23:36:15 +0900 Subject: [MIPS] Update pnx8500-jbs_defconfig In the current pnx8550-jbs_defconfig CONFIG_SGI_IP22 has been selected. Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/configs/pnx8550-jbs_defconfig b/arch/mips/configs/pnx8550-jbs_defconfig index 26b0b98..280a800 100644 --- a/arch/mips/configs/pnx8550-jbs_defconfig +++ b/arch/mips/configs/pnx8550-jbs_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc1 -# Thu Jul 6 10:04:18 2006 +# Linux kernel version: 2.6.19-rc2 +# Sat Oct 14 23:01:16 2006 # CONFIG_MIPS=y @@ -25,8 +25,6 @@ CONFIG_MIPS=y # CONFIG_MIPS_COBALT is not set # CONFIG_MACH_DECSTATION is not set # CONFIG_MIPS_EV64120 is not set -# CONFIG_MIPS_IVR is not set -# CONFIG_MIPS_ITE8172 is not set # CONFIG_MACH_JAZZ is not set # CONFIG_LASAT is not set # CONFIG_MIPS_ATLAS is not set @@ -41,13 +39,13 @@ CONFIG_MIPS=y # CONFIG_MOMENCO_OCELOT_G is not set # CONFIG_MIPS_XXS1500 is not set # CONFIG_PNX8550_V2PCI is not set -# CONFIG_PNX8550_JBS is not set +CONFIG_PNX8550_JBS=y # CONFIG_DDB5477 is not set # CONFIG_MACH_VR41XX is not set # CONFIG_PMC_YOSEMITE is not set # CONFIG_QEMU is not set # CONFIG_MARKEINS is not set -CONFIG_SGI_IP22=y +# CONFIG_SGI_IP22 is not set # CONFIG_SGI_IP27 is not set # CONFIG_SGI_IP32 is not set # CONFIG_SIBYTE_BIGSUR is not set @@ -67,25 +65,21 @@ CONFIG_RWSEM_GENERIC_SPINLOCK=y CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y -CONFIG_ARC=y CONFIG_DMA_NONCOHERENT=y CONFIG_DMA_NEED_PCI_MAP_STATE=y -CONFIG_CPU_BIG_ENDIAN=y -# CONFIG_CPU_LITTLE_ENDIAN is not set -CONFIG_SYS_SUPPORTS_BIG_ENDIAN=y -CONFIG_IRQ_CPU=y -CONFIG_SWAP_IO_SPACE=y -CONFIG_ARC32=y -CONFIG_BOOT_ELF32=y +# CONFIG_CPU_BIG_ENDIAN is not set +CONFIG_CPU_LITTLE_ENDIAN=y +CONFIG_SYS_SUPPORTS_LITTLE_ENDIAN=y +CONFIG_PNX8550=y +CONFIG_SOC_PNX8550=y CONFIG_MIPS_L1_CACHE_SHIFT=5 -# CONFIG_ARC_CONSOLE is not set -CONFIG_ARC_PROMLIB=y # # CPU selection # -# CONFIG_CPU_MIPS32_R1 is not set +CONFIG_CPU_MIPS32_R1=y # CONFIG_CPU_MIPS32_R2 is not set # CONFIG_CPU_MIPS64_R1 is not set # CONFIG_CPU_MIPS64_R2 is not set @@ -93,7 +87,7 @@ CONFIG_ARC_PROMLIB=y # CONFIG_CPU_TX39XX is not set # CONFIG_CPU_VR41XX is not set # CONFIG_CPU_R4300 is not set -CONFIG_CPU_R4X00=y +# CONFIG_CPU_R4X00 is not set # CONFIG_CPU_TX49XX is not set # CONFIG_CPU_R5000 is not set # CONFIG_CPU_R5432 is not set @@ -104,12 +98,11 @@ CONFIG_CPU_R4X00=y # CONFIG_CPU_RM7000 is not set # CONFIG_CPU_RM9000 is not set # CONFIG_CPU_SB1 is not set -CONFIG_SYS_HAS_CPU_R4X00=y -CONFIG_SYS_HAS_CPU_R5000=y +CONFIG_SYS_HAS_CPU_MIPS32_R1=y +CONFIG_CPU_MIPS32=y +CONFIG_CPU_MIPSR1=y CONFIG_SYS_SUPPORTS_32BIT_KERNEL=y -CONFIG_SYS_SUPPORTS_64BIT_KERNEL=y CONFIG_CPU_SUPPORTS_32BIT_KERNEL=y -CONFIG_CPU_SUPPORTS_64BIT_KERNEL=y # # Kernel type @@ -120,17 +113,17 @@ CONFIG_PAGE_SIZE_4KB=y # CONFIG_PAGE_SIZE_8KB is not set # CONFIG_PAGE_SIZE_16KB is not set # CONFIG_PAGE_SIZE_64KB is not set -CONFIG_BOARD_SCACHE=y -CONFIG_IP22_CPU_SCACHE=y +CONFIG_CPU_HAS_PREFETCH=y CONFIG_MIPS_MT_DISABLED=y -# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_MT_SMP is not set +# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_VPE_LOADER is not set # CONFIG_64BIT_PHYS_ADDR is not set CONFIG_CPU_HAS_LLSC=y CONFIG_CPU_HAS_SYNC=y CONFIG_GENERIC_HARDIRQS=y CONFIG_GENERIC_IRQ_PROBE=y +CONFIG_CPU_SUPPORTS_HIGHMEM=y CONFIG_ARCH_FLATMEM_ENABLE=y CONFIG_SELECT_MEMORY_MODEL=y CONFIG_FLATMEM_MANUAL=y @@ -144,12 +137,12 @@ CONFIG_SPLIT_PTLOCK_CPUS=4 # CONFIG_HZ_48 is not set # CONFIG_HZ_100 is not set # CONFIG_HZ_128 is not set -# CONFIG_HZ_250 is not set +CONFIG_HZ_250=y # CONFIG_HZ_256 is not set -CONFIG_HZ_1000=y +# CONFIG_HZ_1000 is not set # CONFIG_HZ_1024 is not set CONFIG_SYS_SUPPORTS_ARBIT_HZ=y -CONFIG_HZ=1000 +CONFIG_HZ=250 CONFIG_PREEMPT_NONE=y # CONFIG_PREEMPT_VOLUNTARY is not set # CONFIG_PREEMPT is not set @@ -171,16 +164,20 @@ CONFIG_LOCALVERSION="" CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set # CONFIG_POSIX_MQUEUE is not set # CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y +# CONFIG_TASKSTATS is not set +# CONFIG_UTS_NS is not set # CONFIG_AUDIT is not set CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y CONFIG_EMBEDDED=y +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -189,12 +186,12 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y CONFIG_BASE_FULL=y -CONFIG_RT_MUTEXES=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -211,6 +208,7 @@ CONFIG_KMOD=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_LBD is not set # CONFIG_BLK_DEV_IO_TRACE is not set # CONFIG_LSF is not set @@ -231,8 +229,10 @@ CONFIG_DEFAULT_IOSCHED="anticipatory" # # Bus options (PCI, PCMCIA, EISA, ISA, TC) # -CONFIG_HW_HAS_EISA=y -# CONFIG_EISA is not set +CONFIG_HW_HAS_PCI=y +CONFIG_PCI=y +# CONFIG_PCI_MULTITHREAD_PROBE is not set +# CONFIG_PCI_DEBUG is not set CONFIG_MMU=y # @@ -243,6 +243,7 @@ CONFIG_MMU=y # # PCI Hotplug Support # +# CONFIG_HOTPLUG_PCI is not set # # Executable file formats @@ -265,6 +266,7 @@ CONFIG_PACKET=y CONFIG_UNIX=y CONFIG_XFRM=y # CONFIG_XFRM_USER is not set +# CONFIG_XFRM_SUB_POLICY is not set # CONFIG_NET_KEY is not set CONFIG_INET=y # CONFIG_IP_MULTICAST is not set @@ -283,16 +285,18 @@ CONFIG_IP_PNP_BOOTP=y # CONFIG_INET_IPCOMP is not set # CONFIG_INET_XFRM_TUNNEL is not set # CONFIG_INET_TUNNEL is not set -CONFIG_INET_XFRM_MODE_TRANSPORT=m -CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_TRANSPORT=y +CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set -CONFIG_NETWORK_SECMARK=y +# CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set # @@ -318,7 +322,6 @@ CONFIG_NETWORK_SECMARK=y # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -371,13 +374,20 @@ CONFIG_FW_LOADER=y # # Block devices # +# CONFIG_BLK_CPQ_DA is not set +# CONFIG_BLK_CPQ_CISS_DA is not set +# CONFIG_BLK_DEV_DAC960 is not set +# CONFIG_BLK_DEV_UMEM is not set # CONFIG_BLK_DEV_COW_COMMON is not set CONFIG_BLK_DEV_LOOP=y # CONFIG_BLK_DEV_CRYPTOLOOP is not set # CONFIG_BLK_DEV_NBD is not set +# CONFIG_BLK_DEV_SX8 is not set +# CONFIG_BLK_DEV_UB is not set CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=8192 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set @@ -386,6 +396,7 @@ CONFIG_BLK_DEV_INITRD=y # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y +CONFIG_IDE_MAX_HWIFS=4 CONFIG_BLK_DEV_IDE=y # @@ -404,8 +415,39 @@ CONFIG_BLK_DEV_IDESCSI=y # IDE chipset support/bugfixes # CONFIG_IDE_GENERIC=y +CONFIG_BLK_DEV_IDEPCI=y +CONFIG_IDEPCI_SHARE_IRQ=y +CONFIG_BLK_DEV_OFFBOARD=y +CONFIG_BLK_DEV_GENERIC=y +# CONFIG_BLK_DEV_OPTI621 is not set +CONFIG_BLK_DEV_IDEDMA_PCI=y +# CONFIG_BLK_DEV_IDEDMA_FORCED is not set +# CONFIG_IDEDMA_PCI_AUTO is not set +# CONFIG_BLK_DEV_AEC62XX is not set +# CONFIG_BLK_DEV_ALI15X3 is not set +# CONFIG_BLK_DEV_AMD74XX is not set +# CONFIG_BLK_DEV_CMD64X is not set +# CONFIG_BLK_DEV_TRIFLEX is not set +# CONFIG_BLK_DEV_CY82C693 is not set +# CONFIG_BLK_DEV_CS5520 is not set +# CONFIG_BLK_DEV_CS5530 is not set +# CONFIG_BLK_DEV_HPT34X is not set +CONFIG_BLK_DEV_HPT366=y +# CONFIG_BLK_DEV_JMICRON is not set +# CONFIG_BLK_DEV_SC1200 is not set +# CONFIG_BLK_DEV_PIIX is not set +# CONFIG_BLK_DEV_IT821X is not set +# CONFIG_BLK_DEV_NS87415 is not set +# CONFIG_BLK_DEV_PDC202XX_OLD is not set +# CONFIG_BLK_DEV_PDC202XX_NEW is not set +# CONFIG_BLK_DEV_SVWKS is not set +# CONFIG_BLK_DEV_SIIMAGE is not set +# CONFIG_BLK_DEV_SLC90E66 is not set +# CONFIG_BLK_DEV_TRM290 is not set +# CONFIG_BLK_DEV_VIA82CXXX is not set # CONFIG_IDE_ARM is not set -# CONFIG_BLK_DEV_IDEDMA is not set +CONFIG_BLK_DEV_IDEDMA=y +# CONFIG_IDEDMA_IVB is not set # CONFIG_IDEDMA_AUTO is not set # CONFIG_BLK_DEV_HD is not set @@ -414,6 +456,7 @@ CONFIG_IDE_GENERIC=y # # CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y +CONFIG_SCSI_NETLINK=y CONFIG_SCSI_PROC_FS=y # @@ -434,22 +477,54 @@ CONFIG_SCSI_CONSTANTS=y # CONFIG_SCSI_LOGGING is not set # -# SCSI Transport Attributes +# SCSI Transports # # CONFIG_SCSI_SPI_ATTRS is not set CONFIG_SCSI_FC_ATTRS=y CONFIG_SCSI_ISCSI_ATTRS=m # CONFIG_SCSI_SAS_ATTRS is not set +# CONFIG_SCSI_SAS_LIBSAS is not set # # SCSI low-level drivers # CONFIG_ISCSI_TCP=m -# CONFIG_SGIWD93_SCSI is not set -# CONFIG_SCSI_SATA is not set +# CONFIG_BLK_DEV_3W_XXXX_RAID is not set +# CONFIG_SCSI_3W_9XXX is not set +# CONFIG_SCSI_ACARD is not set +# CONFIG_SCSI_AACRAID is not set +# CONFIG_SCSI_AIC7XXX is not set +# CONFIG_SCSI_AIC7XXX_OLD is not set +# CONFIG_SCSI_AIC79XX is not set +# CONFIG_SCSI_AIC94XX is not set +# CONFIG_SCSI_DPT_I2O is not set +# CONFIG_SCSI_ARCMSR is not set +# CONFIG_MEGARAID_NEWGEN is not set +# CONFIG_MEGARAID_LEGACY is not set +# CONFIG_MEGARAID_SAS is not set +# CONFIG_SCSI_HPTIOP is not set +# CONFIG_SCSI_DMX3191D is not set +# CONFIG_SCSI_FUTURE_DOMAIN is not set +# CONFIG_SCSI_IPS is not set +# CONFIG_SCSI_INITIO is not set +# CONFIG_SCSI_INIA100 is not set +# CONFIG_SCSI_STEX is not set +# CONFIG_SCSI_SYM53C8XX_2 is not set +# CONFIG_SCSI_QLOGIC_1280 is not set +# CONFIG_SCSI_QLA_FC is not set +# CONFIG_SCSI_QLA_ISCSI is not set +# CONFIG_SCSI_LPFC is not set +# CONFIG_SCSI_DC395x is not set +# CONFIG_SCSI_DC390T is not set +# CONFIG_SCSI_NSP32 is not set # CONFIG_SCSI_DEBUG is not set # +# Serial ATA (prod) and Parallel ATA (experimental) drivers +# +# CONFIG_ATA is not set + +# # Multi-device support (RAID and LVM) # # CONFIG_MD is not set @@ -458,14 +533,19 @@ CONFIG_ISCSI_TCP=m # Fusion MPT device support # # CONFIG_FUSION is not set +# CONFIG_FUSION_SPI is not set +# CONFIG_FUSION_FC is not set +# CONFIG_FUSION_SAS is not set # # IEEE 1394 (FireWire) support # +# CONFIG_IEEE1394 is not set # # I2O device support # +# CONFIG_I2O is not set # # Network device support @@ -477,6 +557,11 @@ CONFIG_NETDEVICES=y # CONFIG_TUN is not set # +# ARCnet devices +# +# CONFIG_ARCNET is not set + +# # PHY device support # # CONFIG_PHYLIB is not set @@ -486,20 +571,73 @@ CONFIG_NETDEVICES=y # CONFIG_NET_ETHERNET=y CONFIG_MII=y +# CONFIG_HAPPYMEAL is not set +# CONFIG_SUNGEM is not set +# CONFIG_CASSINI is not set +# CONFIG_NET_VENDOR_3COM is not set # CONFIG_DM9000 is not set -# CONFIG_SGISEEQ is not set + +# +# Tulip family network device support +# +# CONFIG_NET_TULIP is not set +# CONFIG_HP100 is not set +CONFIG_NET_PCI=y +# CONFIG_PCNET32 is not set +# CONFIG_AMD8111_ETH is not set +# CONFIG_ADAPTEC_STARFIRE is not set +# CONFIG_B44 is not set +# CONFIG_FORCEDETH is not set +# CONFIG_DGRS is not set +# CONFIG_EEPRO100 is not set +# CONFIG_E100 is not set +# CONFIG_FEALNX is not set +# CONFIG_NATSEMI is not set +# CONFIG_NE2K_PCI is not set +# CONFIG_8139CP is not set +CONFIG_8139TOO=y +# CONFIG_8139TOO_PIO is not set +CONFIG_8139TOO_TUNE_TWISTER=y +CONFIG_8139TOO_8129=y +# CONFIG_8139_OLD_RX_RESET is not set +# CONFIG_SIS900 is not set +# CONFIG_EPIC100 is not set +# CONFIG_SUNDANCE is not set +# CONFIG_TLAN is not set +# CONFIG_VIA_RHINE is not set +# CONFIG_LAN_SAA9730 is not set # # Ethernet (1000 Mbit) # +# CONFIG_ACENIC is not set +# CONFIG_DL2K is not set +# CONFIG_E1000 is not set +# CONFIG_NS83820 is not set +# CONFIG_HAMACHI is not set +# CONFIG_YELLOWFIN is not set +# CONFIG_R8169 is not set +# CONFIG_SIS190 is not set +# CONFIG_SKGE is not set +# CONFIG_SKY2 is not set +# CONFIG_SK98LIN is not set +# CONFIG_VIA_VELOCITY is not set +# CONFIG_TIGON3 is not set +# CONFIG_BNX2 is not set +# CONFIG_QLA3XXX is not set # # Ethernet (10000 Mbit) # +# CONFIG_CHELSIO_T1 is not set +# CONFIG_IXGB is not set +# CONFIG_S2IO is not set +# CONFIG_MYRI10GE is not set # # Token Ring devices # +# CONFIG_TR is not set # # Wireless LAN (non-hamradio) @@ -510,8 +648,11 @@ CONFIG_MII=y # Wan interfaces # # CONFIG_WAN is not set +# CONFIG_FDDI is not set +# CONFIG_HIPPI is not set # CONFIG_PPP is not set # CONFIG_SLIP is not set +# CONFIG_NET_FC is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set # CONFIG_NETPOLL is not set @@ -531,6 +672,7 @@ CONFIG_MII=y # Input device support # CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set # # Userland interfaces @@ -556,6 +698,7 @@ CONFIG_INPUT=y CONFIG_SERIO=y # CONFIG_SERIO_I8042 is not set # CONFIG_SERIO_SERPORT is not set +# CONFIG_SERIO_PCIPS2 is not set CONFIG_SERIO_LIBPS2=y # CONFIG_SERIO_RAW is not set # CONFIG_GAMEPORT is not set @@ -566,7 +709,7 @@ CONFIG_SERIO_LIBPS2=y CONFIG_VT=y CONFIG_VT_CONSOLE=y CONFIG_HW_CONSOLE=y -CONFIG_VT_HW_CONSOLE_BINDING=y +# CONFIG_VT_HW_CONSOLE_BINDING is not set # CONFIG_SERIAL_NONSTANDARD is not set # @@ -577,7 +720,8 @@ CONFIG_VT_HW_CONSOLE_BINDING=y # # Non-8250 serial port support # -# CONFIG_SERIAL_IP22_ZILOG is not set +# CONFIG_SERIAL_IP3106 is not set +# CONFIG_SERIAL_JSM is not set CONFIG_UNIX98_PTYS=y CONFIG_LEGACY_PTYS=y CONFIG_LEGACY_PTY_COUNT=256 @@ -591,16 +735,17 @@ CONFIG_LEGACY_PTY_COUNT=256 # Watchdog Cards # # CONFIG_WATCHDOG is not set -# CONFIG_HW_RANDOM is not set +CONFIG_HW_RANDOM=y # CONFIG_RTC is not set -# CONFIG_SGI_DS1286 is not set # CONFIG_GEN_RTC is not set # CONFIG_DTLK is not set # CONFIG_R3964 is not set +# CONFIG_APPLICOM is not set # # Ftape, the floppy tape device driver # +# CONFIG_DRM is not set # CONFIG_RAW_DRIVER is not set # @@ -631,35 +776,37 @@ CONFIG_HWMON=y # CONFIG_HWMON_VID is not set # CONFIG_SENSORS_ABITUGURU is not set # CONFIG_SENSORS_F71805F is not set +# CONFIG_SENSORS_VT1211 is not set # CONFIG_HWMON_DEBUG_CHIP is not set # # Misc devices # +# CONFIG_TIFM_CORE is not set # # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices # # CONFIG_DVB is not set +# CONFIG_USB_DABUSB is not set # # Graphics support # -# CONFIG_FIRMWARE_EDID is not set +CONFIG_FIRMWARE_EDID=y # CONFIG_FB is not set # # Console display driver support # # CONFIG_VGA_CONSOLE is not set -# CONFIG_SGI_NEWPORT_CONSOLE is not set CONFIG_DUMMY_CONSOLE=y +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set # # Sound @@ -669,15 +816,131 @@ CONFIG_DUMMY_CONSOLE=y # # USB support # -# CONFIG_USB_ARCH_HAS_HCD is not set -# CONFIG_USB_ARCH_HAS_OHCI is not set -# CONFIG_USB_ARCH_HAS_EHCI is not set +CONFIG_USB_ARCH_HAS_HCD=y +CONFIG_USB_ARCH_HAS_OHCI=y +CONFIG_USB_ARCH_HAS_EHCI=y +CONFIG_USB=y +# CONFIG_USB_DEBUG is not set + +# +# Miscellaneous USB options +# +# CONFIG_USB_DEVICEFS is not set +# CONFIG_USB_BANDWIDTH is not set +# CONFIG_USB_DYNAMIC_MINORS is not set +# CONFIG_USB_OTG is not set + +# +# USB Host Controller Drivers +# +# CONFIG_USB_EHCI_HCD is not set +# CONFIG_USB_ISP116X_HCD is not set +CONFIG_USB_OHCI_HCD=y +# CONFIG_USB_OHCI_BIG_ENDIAN is not set +CONFIG_USB_OHCI_LITTLE_ENDIAN=y +# CONFIG_USB_UHCI_HCD is not set +# CONFIG_USB_SL811_HCD is not set + +# +# USB Device Class drivers +# +# CONFIG_USB_ACM is not set +# CONFIG_USB_PRINTER is not set # # NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' # # +# may also be needed; see USB_STORAGE Help for more information +# +CONFIG_USB_STORAGE=y +# CONFIG_USB_STORAGE_DEBUG is not set +CONFIG_USB_STORAGE_DATAFAB=y +CONFIG_USB_STORAGE_FREECOM=y +CONFIG_USB_STORAGE_ISD200=y +CONFIG_USB_STORAGE_DPCM=y +CONFIG_USB_STORAGE_USBAT=y +CONFIG_USB_STORAGE_SDDR09=y +CONFIG_USB_STORAGE_SDDR55=y +CONFIG_USB_STORAGE_JUMPSHOT=y +# CONFIG_USB_STORAGE_ALAUDA is not set +# CONFIG_USB_STORAGE_KARMA is not set +# CONFIG_USB_LIBUSUAL is not set + +# +# USB Input Devices +# +# CONFIG_USB_HID is not set + +# +# USB HID Boot Protocol drivers +# +# CONFIG_USB_KBD is not set +# CONFIG_USB_MOUSE is not set +# CONFIG_USB_AIPTEK is not set +# CONFIG_USB_WACOM is not set +# CONFIG_USB_ACECAD is not set +# CONFIG_USB_KBTAB is not set +# CONFIG_USB_POWERMATE is not set +# CONFIG_USB_TOUCHSCREEN is not set +# CONFIG_USB_YEALINK is not set +# CONFIG_USB_XPAD is not set +# CONFIG_USB_ATI_REMOTE is not set +# CONFIG_USB_ATI_REMOTE2 is not set +# CONFIG_USB_KEYSPAN_REMOTE is not set +# CONFIG_USB_APPLETOUCH is not set +# CONFIG_USB_TRANCEVIBRATOR is not set + +# +# USB Imaging devices +# +# CONFIG_USB_MDC800 is not set +# CONFIG_USB_MICROTEK is not set + +# +# USB Network Adapters +# +# CONFIG_USB_CATC is not set +# CONFIG_USB_KAWETH is not set +# CONFIG_USB_PEGASUS is not set +# CONFIG_USB_RTL8150 is not set +# CONFIG_USB_USBNET is not set +CONFIG_USB_MON=y + +# +# USB port drivers +# + +# +# USB Serial Converter support +# +# CONFIG_USB_SERIAL is not set + +# +# USB Miscellaneous drivers +# +# CONFIG_USB_EMI62 is not set +# CONFIG_USB_EMI26 is not set +# CONFIG_USB_ADUTUX is not set +# CONFIG_USB_AUERSWALD is not set +# CONFIG_USB_RIO500 is not set +# CONFIG_USB_LEGOTOWER is not set +# CONFIG_USB_LCD is not set +# CONFIG_USB_LED is not set +# CONFIG_USB_CYPRESS_CY7C63 is not set +# CONFIG_USB_CYTHERM is not set +# CONFIG_USB_PHIDGET is not set +# CONFIG_USB_IDMOUSE is not set +# CONFIG_USB_FTDI_ELAN is not set +# CONFIG_USB_APPLEDISPLAY is not set +# CONFIG_USB_LD is not set + +# +# USB DSL modem support +# + +# # USB Gadget Support # # CONFIG_USB_GADGET is not set @@ -703,6 +966,7 @@ CONFIG_DUMMY_CONSOLE=y # # InfiniBand support # +# CONFIG_INFINIBAND is not set # # EDAC - error detection and reporting (RAS) (EXPERIMENTAL) @@ -733,10 +997,12 @@ CONFIG_EXT2_FS=y # CONFIG_EXT2_FS_XATTR is not set # CONFIG_EXT2_FS_XIP is not set # CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set # CONFIG_REISERFS_FS is not set # CONFIG_JFS_FS is not set # CONFIG_FS_POSIX_ACL is not set # CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set @@ -769,8 +1035,10 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" # CONFIG_PROC_FS=y # CONFIG_PROC_KCORE is not set +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y # CONFIG_CONFIGFS_FS is not set @@ -813,7 +1081,6 @@ CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set # CONFIG_CIFS is not set -# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -824,7 +1091,6 @@ CONFIG_SUNRPC=y # # CONFIG_PARTITION_ADVANCED is not set CONFIG_MSDOS_PARTITION=y -CONFIG_SGI_PARTITION=y # # Native Language Support @@ -880,6 +1146,7 @@ CONFIG_NLS_DEFAULT="iso8859-1" # CONFIG_TRACE_IRQFLAGS_SUPPORT=y # CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y @@ -893,13 +1160,17 @@ CONFIG_DEBUG_SLAB=y # CONFIG_DEBUG_SPINLOCK is not set CONFIG_DEBUG_MUTEXES=y # CONFIG_DEBUG_RWSEMS is not set +# CONFIG_DEBUG_LOCK_ALLOC is not set +# CONFIG_PROVE_LOCKING is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set # CONFIG_DEBUG_KOBJECT is not set # CONFIG_DEBUG_INFO is not set # CONFIG_DEBUG_FS is not set # CONFIG_DEBUG_VM is not set +# CONFIG_DEBUG_LIST is not set CONFIG_FORCED_INLINING=y +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set CONFIG_CROSSCOMPILE=y CONFIG_CMDLINE="console=ttyS1,38400n8 kgdb=ttyS0 root=/dev/nfs ip=bootp" @@ -918,6 +1189,9 @@ CONFIG_CMDLINE="console=ttyS1,38400n8 kgdb=ttyS0 root=/dev/nfs ip=bootp" # Cryptographic options # CONFIG_CRYPTO=y +CONFIG_CRYPTO_ALGAPI=m +CONFIG_CRYPTO_BLKCIPHER=m +CONFIG_CRYPTO_MANAGER=m # CONFIG_CRYPTO_HMAC is not set # CONFIG_CRYPTO_NULL is not set # CONFIG_CRYPTO_MD4 is not set @@ -927,6 +1201,8 @@ CONFIG_CRYPTO_MD5=m # CONFIG_CRYPTO_SHA512 is not set # CONFIG_CRYPTO_WP512 is not set # CONFIG_CRYPTO_TGR192 is not set +CONFIG_CRYPTO_ECB=m +CONFIG_CRYPTO_CBC=m # CONFIG_CRYPTO_DES is not set # CONFIG_CRYPTO_BLOWFISH is not set # CONFIG_CRYPTO_TWOFISH is not set -- cgit v0.10.2 From 0895b19ebdbb44d6d2eee0d7a138f07b2ddb04ec Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 18 Oct 2006 23:38:12 +0900 Subject: [MIPS] Update pnx8550-v2pci_defconfig In the current pnx8550-v2pci_defconfig CONFIG_SGI_IP22 has been selected. Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/configs/pnx8550-v2pci_defconfig b/arch/mips/configs/pnx8550-v2pci_defconfig index e93266b..64b9fbf 100644 --- a/arch/mips/configs/pnx8550-v2pci_defconfig +++ b/arch/mips/configs/pnx8550-v2pci_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc1 -# Thu Jul 6 10:04:18 2006 +# Linux kernel version: 2.6.19-rc2 +# Sat Oct 14 23:12:15 2006 # CONFIG_MIPS=y @@ -25,8 +25,6 @@ CONFIG_MIPS=y # CONFIG_MIPS_COBALT is not set # CONFIG_MACH_DECSTATION is not set # CONFIG_MIPS_EV64120 is not set -# CONFIG_MIPS_IVR is not set -# CONFIG_MIPS_ITE8172 is not set # CONFIG_MACH_JAZZ is not set # CONFIG_LASAT is not set # CONFIG_MIPS_ATLAS is not set @@ -40,14 +38,14 @@ CONFIG_MIPS=y # CONFIG_MOMENCO_OCELOT_C is not set # CONFIG_MOMENCO_OCELOT_G is not set # CONFIG_MIPS_XXS1500 is not set -# CONFIG_PNX8550_V2PCI is not set +CONFIG_PNX8550_V2PCI=y # CONFIG_PNX8550_JBS is not set # CONFIG_DDB5477 is not set # CONFIG_MACH_VR41XX is not set # CONFIG_PMC_YOSEMITE is not set # CONFIG_QEMU is not set # CONFIG_MARKEINS is not set -CONFIG_SGI_IP22=y +# CONFIG_SGI_IP22 is not set # CONFIG_SGI_IP27 is not set # CONFIG_SGI_IP32 is not set # CONFIG_SIBYTE_BIGSUR is not set @@ -67,25 +65,21 @@ CONFIG_RWSEM_GENERIC_SPINLOCK=y CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y -CONFIG_ARC=y CONFIG_DMA_NONCOHERENT=y CONFIG_DMA_NEED_PCI_MAP_STATE=y -CONFIG_CPU_BIG_ENDIAN=y -# CONFIG_CPU_LITTLE_ENDIAN is not set -CONFIG_SYS_SUPPORTS_BIG_ENDIAN=y -CONFIG_IRQ_CPU=y -CONFIG_SWAP_IO_SPACE=y -CONFIG_ARC32=y -CONFIG_BOOT_ELF32=y +# CONFIG_CPU_BIG_ENDIAN is not set +CONFIG_CPU_LITTLE_ENDIAN=y +CONFIG_SYS_SUPPORTS_LITTLE_ENDIAN=y +CONFIG_PNX8550=y +CONFIG_SOC_PNX8550=y CONFIG_MIPS_L1_CACHE_SHIFT=5 -# CONFIG_ARC_CONSOLE is not set -CONFIG_ARC_PROMLIB=y # # CPU selection # -# CONFIG_CPU_MIPS32_R1 is not set +CONFIG_CPU_MIPS32_R1=y # CONFIG_CPU_MIPS32_R2 is not set # CONFIG_CPU_MIPS64_R1 is not set # CONFIG_CPU_MIPS64_R2 is not set @@ -93,7 +87,7 @@ CONFIG_ARC_PROMLIB=y # CONFIG_CPU_TX39XX is not set # CONFIG_CPU_VR41XX is not set # CONFIG_CPU_R4300 is not set -CONFIG_CPU_R4X00=y +# CONFIG_CPU_R4X00 is not set # CONFIG_CPU_TX49XX is not set # CONFIG_CPU_R5000 is not set # CONFIG_CPU_R5432 is not set @@ -104,12 +98,11 @@ CONFIG_CPU_R4X00=y # CONFIG_CPU_RM7000 is not set # CONFIG_CPU_RM9000 is not set # CONFIG_CPU_SB1 is not set -CONFIG_SYS_HAS_CPU_R4X00=y -CONFIG_SYS_HAS_CPU_R5000=y +CONFIG_SYS_HAS_CPU_MIPS32_R1=y +CONFIG_CPU_MIPS32=y +CONFIG_CPU_MIPSR1=y CONFIG_SYS_SUPPORTS_32BIT_KERNEL=y -CONFIG_SYS_SUPPORTS_64BIT_KERNEL=y CONFIG_CPU_SUPPORTS_32BIT_KERNEL=y -CONFIG_CPU_SUPPORTS_64BIT_KERNEL=y # # Kernel type @@ -120,17 +113,17 @@ CONFIG_PAGE_SIZE_4KB=y # CONFIG_PAGE_SIZE_8KB is not set # CONFIG_PAGE_SIZE_16KB is not set # CONFIG_PAGE_SIZE_64KB is not set -CONFIG_BOARD_SCACHE=y -CONFIG_IP22_CPU_SCACHE=y +CONFIG_CPU_HAS_PREFETCH=y CONFIG_MIPS_MT_DISABLED=y -# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_MT_SMP is not set +# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_VPE_LOADER is not set # CONFIG_64BIT_PHYS_ADDR is not set CONFIG_CPU_HAS_LLSC=y CONFIG_CPU_HAS_SYNC=y CONFIG_GENERIC_HARDIRQS=y CONFIG_GENERIC_IRQ_PROBE=y +CONFIG_CPU_SUPPORTS_HIGHMEM=y CONFIG_ARCH_FLATMEM_ENABLE=y CONFIG_SELECT_MEMORY_MODEL=y CONFIG_FLATMEM_MANUAL=y @@ -144,12 +137,12 @@ CONFIG_SPLIT_PTLOCK_CPUS=4 # CONFIG_HZ_48 is not set # CONFIG_HZ_100 is not set # CONFIG_HZ_128 is not set -# CONFIG_HZ_250 is not set +CONFIG_HZ_250=y # CONFIG_HZ_256 is not set -CONFIG_HZ_1000=y +# CONFIG_HZ_1000 is not set # CONFIG_HZ_1024 is not set CONFIG_SYS_SUPPORTS_ARBIT_HZ=y -CONFIG_HZ=1000 +CONFIG_HZ=250 CONFIG_PREEMPT_NONE=y # CONFIG_PREEMPT_VOLUNTARY is not set # CONFIG_PREEMPT is not set @@ -171,16 +164,20 @@ CONFIG_LOCALVERSION="" CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set # CONFIG_POSIX_MQUEUE is not set # CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y +# CONFIG_TASKSTATS is not set +# CONFIG_UTS_NS is not set # CONFIG_AUDIT is not set CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y CONFIG_EMBEDDED=y +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_EXTRA_PASS is not set CONFIG_HOTPLUG=y @@ -188,12 +185,12 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y CONFIG_BASE_FULL=y -CONFIG_RT_MUTEXES=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -210,6 +207,7 @@ CONFIG_KMOD=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_LBD is not set # CONFIG_BLK_DEV_IO_TRACE is not set # CONFIG_LSF is not set @@ -230,8 +228,9 @@ CONFIG_DEFAULT_IOSCHED="anticipatory" # # Bus options (PCI, PCMCIA, EISA, ISA, TC) # -CONFIG_HW_HAS_EISA=y -# CONFIG_EISA is not set +CONFIG_HW_HAS_PCI=y +CONFIG_PCI=y +# CONFIG_PCI_MULTITHREAD_PROBE is not set CONFIG_MMU=y # @@ -242,6 +241,7 @@ CONFIG_MMU=y # # PCI Hotplug Support # +# CONFIG_HOTPLUG_PCI is not set # # Executable file formats @@ -264,6 +264,7 @@ CONFIG_PACKET=y CONFIG_UNIX=y CONFIG_XFRM=y # CONFIG_XFRM_USER is not set +# CONFIG_XFRM_SUB_POLICY is not set # CONFIG_NET_KEY is not set CONFIG_INET=y # CONFIG_IP_MULTICAST is not set @@ -282,12 +283,14 @@ CONFIG_IP_PNP=y # CONFIG_INET_IPCOMP is not set # CONFIG_INET_XFRM_TUNNEL is not set # CONFIG_INET_TUNNEL is not set -CONFIG_INET_XFRM_MODE_TRANSPORT=m -CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_TRANSPORT=y +CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" # # IP: Virtual Server Configuration @@ -300,12 +303,18 @@ CONFIG_IPV6_ROUTE_INFO=y # CONFIG_INET6_AH is not set # CONFIG_INET6_ESP is not set # CONFIG_INET6_IPCOMP is not set +# CONFIG_IPV6_MIP6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set CONFIG_INET6_XFRM_MODE_TRANSPORT=m CONFIG_INET6_XFRM_MODE_TUNNEL=m +CONFIG_INET6_XFRM_MODE_BEET=m +# CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=m # CONFIG_IPV6_TUNNEL is not set -CONFIG_NETWORK_SECMARK=y +# CONFIG_IPV6_SUBTREES is not set +# CONFIG_IPV6_MULTIPLE_TABLES is not set +# CONFIG_NETWORK_SECMARK is not set CONFIG_NETFILTER=y # CONFIG_NETFILTER_DEBUG is not set @@ -318,9 +327,9 @@ CONFIG_NETFILTER_XTABLES=m CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m CONFIG_NETFILTER_XT_TARGET_MARK=m CONFIG_NETFILTER_XT_TARGET_NFQUEUE=m -CONFIG_NETFILTER_XT_TARGET_SECMARK=m CONFIG_NETFILTER_XT_MATCH_COMMENT=m CONFIG_NETFILTER_XT_MATCH_DCCP=m +# CONFIG_NETFILTER_XT_MATCH_DSCP is not set CONFIG_NETFILTER_XT_MATCH_ESP=m CONFIG_NETFILTER_XT_MATCH_LENGTH=m CONFIG_NETFILTER_XT_MATCH_LIMIT=m @@ -329,10 +338,10 @@ CONFIG_NETFILTER_XT_MATCH_MARK=m # CONFIG_NETFILTER_XT_MATCH_POLICY is not set CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m CONFIG_NETFILTER_XT_MATCH_PKTTYPE=m -CONFIG_NETFILTER_XT_MATCH_QUOTA=m +# CONFIG_NETFILTER_XT_MATCH_QUOTA is not set CONFIG_NETFILTER_XT_MATCH_REALM=m CONFIG_NETFILTER_XT_MATCH_SCTP=m -CONFIG_NETFILTER_XT_MATCH_STATISTIC=m +# CONFIG_NETFILTER_XT_MATCH_STATISTIC is not set CONFIG_NETFILTER_XT_MATCH_STRING=m CONFIG_NETFILTER_XT_MATCH_TCPMSS=m @@ -373,7 +382,6 @@ CONFIG_NETFILTER_XT_MATCH_TCPMSS=m # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -426,13 +434,20 @@ CONFIG_FW_LOADER=y # # Block devices # +# CONFIG_BLK_CPQ_DA is not set +# CONFIG_BLK_CPQ_CISS_DA is not set +# CONFIG_BLK_DEV_DAC960 is not set +# CONFIG_BLK_DEV_UMEM is not set # CONFIG_BLK_DEV_COW_COMMON is not set CONFIG_BLK_DEV_LOOP=y # CONFIG_BLK_DEV_CRYPTOLOOP is not set # CONFIG_BLK_DEV_NBD is not set +# CONFIG_BLK_DEV_SX8 is not set +# CONFIG_BLK_DEV_UB is not set CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=8192 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set @@ -441,6 +456,7 @@ CONFIG_BLK_DEV_INITRD=y # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y +CONFIG_IDE_MAX_HWIFS=4 CONFIG_BLK_DEV_IDE=y # @@ -459,9 +475,41 @@ CONFIG_IDEDISK_MULTI_MODE=y # IDE chipset support/bugfixes # CONFIG_IDE_GENERIC=y +CONFIG_BLK_DEV_IDEPCI=y +CONFIG_IDEPCI_SHARE_IRQ=y +# CONFIG_BLK_DEV_OFFBOARD is not set +# CONFIG_BLK_DEV_GENERIC is not set +# CONFIG_BLK_DEV_OPTI621 is not set +CONFIG_BLK_DEV_IDEDMA_PCI=y +# CONFIG_BLK_DEV_IDEDMA_FORCED is not set +CONFIG_IDEDMA_PCI_AUTO=y +# CONFIG_IDEDMA_ONLYDISK is not set +# CONFIG_BLK_DEV_AEC62XX is not set +# CONFIG_BLK_DEV_ALI15X3 is not set +# CONFIG_BLK_DEV_AMD74XX is not set +CONFIG_BLK_DEV_CMD64X=y +# CONFIG_BLK_DEV_TRIFLEX is not set +# CONFIG_BLK_DEV_CY82C693 is not set +# CONFIG_BLK_DEV_CS5520 is not set +# CONFIG_BLK_DEV_CS5530 is not set +# CONFIG_BLK_DEV_HPT34X is not set +# CONFIG_BLK_DEV_HPT366 is not set +# CONFIG_BLK_DEV_JMICRON is not set +# CONFIG_BLK_DEV_SC1200 is not set +# CONFIG_BLK_DEV_PIIX is not set +# CONFIG_BLK_DEV_IT821X is not set +# CONFIG_BLK_DEV_NS87415 is not set +# CONFIG_BLK_DEV_PDC202XX_OLD is not set +# CONFIG_BLK_DEV_PDC202XX_NEW is not set +# CONFIG_BLK_DEV_SVWKS is not set +# CONFIG_BLK_DEV_SIIMAGE is not set +# CONFIG_BLK_DEV_SLC90E66 is not set +# CONFIG_BLK_DEV_TRM290 is not set +# CONFIG_BLK_DEV_VIA82CXXX is not set # CONFIG_IDE_ARM is not set -# CONFIG_BLK_DEV_IDEDMA is not set -# CONFIG_IDEDMA_AUTO is not set +CONFIG_BLK_DEV_IDEDMA=y +# CONFIG_IDEDMA_IVB is not set +CONFIG_IDEDMA_AUTO=y # CONFIG_BLK_DEV_HD is not set # @@ -469,6 +517,7 @@ CONFIG_IDE_GENERIC=y # # CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y +CONFIG_SCSI_NETLINK=y CONFIG_SCSI_PROC_FS=y # @@ -489,22 +538,59 @@ CONFIG_BLK_DEV_SD=y # CONFIG_SCSI_LOGGING is not set # -# SCSI Transport Attributes +# SCSI Transports # CONFIG_SCSI_SPI_ATTRS=m CONFIG_SCSI_FC_ATTRS=y CONFIG_SCSI_ISCSI_ATTRS=m # CONFIG_SCSI_SAS_ATTRS is not set +# CONFIG_SCSI_SAS_LIBSAS is not set # # SCSI low-level drivers # CONFIG_ISCSI_TCP=m -# CONFIG_SGIWD93_SCSI is not set -# CONFIG_SCSI_SATA is not set +# CONFIG_BLK_DEV_3W_XXXX_RAID is not set +# CONFIG_SCSI_3W_9XXX is not set +# CONFIG_SCSI_ACARD is not set +# CONFIG_SCSI_AACRAID is not set +CONFIG_SCSI_AIC7XXX=m +CONFIG_AIC7XXX_CMDS_PER_DEVICE=32 +CONFIG_AIC7XXX_RESET_DELAY_MS=15000 +# CONFIG_AIC7XXX_DEBUG_ENABLE is not set +CONFIG_AIC7XXX_DEBUG_MASK=0 +# CONFIG_AIC7XXX_REG_PRETTY_PRINT is not set +# CONFIG_SCSI_AIC7XXX_OLD is not set +# CONFIG_SCSI_AIC79XX is not set +# CONFIG_SCSI_AIC94XX is not set +# CONFIG_SCSI_DPT_I2O is not set +# CONFIG_SCSI_ARCMSR is not set +# CONFIG_MEGARAID_NEWGEN is not set +# CONFIG_MEGARAID_LEGACY is not set +# CONFIG_MEGARAID_SAS is not set +# CONFIG_SCSI_HPTIOP is not set +# CONFIG_SCSI_DMX3191D is not set +# CONFIG_SCSI_FUTURE_DOMAIN is not set +# CONFIG_SCSI_IPS is not set +# CONFIG_SCSI_INITIO is not set +# CONFIG_SCSI_INIA100 is not set +# CONFIG_SCSI_STEX is not set +# CONFIG_SCSI_SYM53C8XX_2 is not set +# CONFIG_SCSI_QLOGIC_1280 is not set +# CONFIG_SCSI_QLA_FC is not set +# CONFIG_SCSI_QLA_ISCSI is not set +# CONFIG_SCSI_LPFC is not set +# CONFIG_SCSI_DC395x is not set +# CONFIG_SCSI_DC390T is not set +# CONFIG_SCSI_NSP32 is not set # CONFIG_SCSI_DEBUG is not set # +# Serial ATA (prod) and Parallel ATA (experimental) drivers +# +# CONFIG_ATA is not set + +# # Multi-device support (RAID and LVM) # # CONFIG_MD is not set @@ -513,14 +599,19 @@ CONFIG_ISCSI_TCP=m # Fusion MPT device support # # CONFIG_FUSION is not set +# CONFIG_FUSION_SPI is not set +# CONFIG_FUSION_FC is not set +# CONFIG_FUSION_SAS is not set # # IEEE 1394 (FireWire) support # +# CONFIG_IEEE1394 is not set # # I2O device support # +# CONFIG_I2O is not set # # Network device support @@ -532,6 +623,11 @@ CONFIG_NETDEVICES=y CONFIG_TUN=m # +# ARCnet devices +# +# CONFIG_ARCNET is not set + +# # PHY device support # # CONFIG_PHYLIB is not set @@ -541,20 +637,73 @@ CONFIG_TUN=m # CONFIG_NET_ETHERNET=y CONFIG_MII=y +# CONFIG_HAPPYMEAL is not set +# CONFIG_SUNGEM is not set +# CONFIG_CASSINI is not set +# CONFIG_NET_VENDOR_3COM is not set # CONFIG_DM9000 is not set -# CONFIG_SGISEEQ is not set + +# +# Tulip family network device support +# +# CONFIG_NET_TULIP is not set +# CONFIG_HP100 is not set +CONFIG_NET_PCI=y +# CONFIG_PCNET32 is not set +# CONFIG_AMD8111_ETH is not set +# CONFIG_ADAPTEC_STARFIRE is not set +# CONFIG_B44 is not set +# CONFIG_FORCEDETH is not set +# CONFIG_DGRS is not set +# CONFIG_EEPRO100 is not set +# CONFIG_E100 is not set +# CONFIG_FEALNX is not set +CONFIG_NATSEMI=y +# CONFIG_NE2K_PCI is not set +# CONFIG_8139CP is not set +CONFIG_8139TOO=y +# CONFIG_8139TOO_PIO is not set +# CONFIG_8139TOO_TUNE_TWISTER is not set +# CONFIG_8139TOO_8129 is not set +# CONFIG_8139_OLD_RX_RESET is not set +# CONFIG_SIS900 is not set +# CONFIG_EPIC100 is not set +# CONFIG_SUNDANCE is not set +# CONFIG_TLAN is not set +# CONFIG_VIA_RHINE is not set +# CONFIG_LAN_SAA9730 is not set # # Ethernet (1000 Mbit) # +# CONFIG_ACENIC is not set +# CONFIG_DL2K is not set +# CONFIG_E1000 is not set +# CONFIG_NS83820 is not set +# CONFIG_HAMACHI is not set +# CONFIG_YELLOWFIN is not set +# CONFIG_R8169 is not set +# CONFIG_SIS190 is not set +# CONFIG_SKGE is not set +# CONFIG_SKY2 is not set +# CONFIG_SK98LIN is not set +# CONFIG_VIA_VELOCITY is not set +# CONFIG_TIGON3 is not set +# CONFIG_BNX2 is not set +# CONFIG_QLA3XXX is not set # # Ethernet (10000 Mbit) # +# CONFIG_CHELSIO_T1 is not set +# CONFIG_IXGB is not set +# CONFIG_S2IO is not set +# CONFIG_MYRI10GE is not set # # Token Ring devices # +# CONFIG_TR is not set # # Wireless LAN (non-hamradio) @@ -565,6 +714,8 @@ CONFIG_MII=y # Wan interfaces # # CONFIG_WAN is not set +# CONFIG_FDDI is not set +# CONFIG_HIPPI is not set CONFIG_PPP=m # CONFIG_PPP_MULTILINK is not set # CONFIG_PPP_FILTER is not set @@ -575,6 +726,8 @@ CONFIG_PPP_DEFLATE=m CONFIG_PPP_MPPE=m # CONFIG_PPPOE is not set # CONFIG_SLIP is not set +CONFIG_SLHC=m +# CONFIG_NET_FC is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set # CONFIG_NETPOLL is not set @@ -594,6 +747,7 @@ CONFIG_PPP_MPPE=m # Input device support # CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set # # Userland interfaces @@ -616,6 +770,7 @@ CONFIG_KEYBOARD_ATKBD=y # CONFIG_KEYBOARD_LKKBD is not set # CONFIG_KEYBOARD_XTKBD is not set # CONFIG_KEYBOARD_NEWTON is not set +# CONFIG_KEYBOARD_STOWAWAY is not set CONFIG_INPUT_MOUSE=y CONFIG_MOUSE_PS2=y # CONFIG_MOUSE_SERIAL is not set @@ -630,6 +785,7 @@ CONFIG_MOUSE_PS2=y CONFIG_SERIO=y CONFIG_SERIO_I8042=y CONFIG_SERIO_SERPORT=y +# CONFIG_SERIO_PCIPS2 is not set CONFIG_SERIO_LIBPS2=y # CONFIG_SERIO_RAW is not set # CONFIG_GAMEPORT is not set @@ -640,7 +796,7 @@ CONFIG_SERIO_LIBPS2=y CONFIG_VT=y # CONFIG_VT_CONSOLE is not set CONFIG_HW_CONSOLE=y -CONFIG_VT_HW_CONSOLE_BINDING=y +# CONFIG_VT_HW_CONSOLE_BINDING is not set CONFIG_SERIAL_NONSTANDARD=y # CONFIG_COMPUTONE is not set # CONFIG_ROCKETPORT is not set @@ -650,6 +806,7 @@ CONFIG_SERIAL_NONSTANDARD=y # CONFIG_MOXA_SMARTIO is not set # CONFIG_ISI is not set # CONFIG_SYNCLINKMP is not set +# CONFIG_SYNCLINK_GT is not set # CONFIG_N_HDLC is not set # CONFIG_RISCOM8 is not set # CONFIG_SPECIALIX is not set @@ -665,7 +822,8 @@ CONFIG_SERIAL_NONSTANDARD=y # # Non-8250 serial port support # -# CONFIG_SERIAL_IP22_ZILOG is not set +# CONFIG_SERIAL_IP3106 is not set +# CONFIG_SERIAL_JSM is not set CONFIG_UNIX98_PTYS=y CONFIG_LEGACY_PTYS=y CONFIG_LEGACY_PTY_COUNT=256 @@ -679,16 +837,17 @@ CONFIG_LEGACY_PTY_COUNT=256 # Watchdog Cards # # CONFIG_WATCHDOG is not set -# CONFIG_HW_RANDOM is not set +CONFIG_HW_RANDOM=y # CONFIG_RTC is not set -# CONFIG_SGI_DS1286 is not set # CONFIG_GEN_RTC is not set # CONFIG_DTLK is not set # CONFIG_R3964 is not set +# CONFIG_APPLICOM is not set # # Ftape, the floppy tape device driver # +# CONFIG_DRM is not set # CONFIG_RAW_DRIVER is not set # @@ -709,14 +868,30 @@ CONFIG_I2C_CHARDEV=m CONFIG_I2C_ALGOBIT=m # CONFIG_I2C_ALGOPCF is not set # CONFIG_I2C_ALGOPCA is not set -# CONFIG_I2C_ALGO_SGI is not set # # I2C Hardware Bus support # +# CONFIG_I2C_ALI1535 is not set +# CONFIG_I2C_ALI1563 is not set +# CONFIG_I2C_ALI15X3 is not set +# CONFIG_I2C_AMD756 is not set +# CONFIG_I2C_AMD8111 is not set +# CONFIG_I2C_I801 is not set +# CONFIG_I2C_I810 is not set +# CONFIG_I2C_PIIX4 is not set +# CONFIG_I2C_NFORCE2 is not set # CONFIG_I2C_OCORES is not set # CONFIG_I2C_PARPORT_LIGHT is not set +# CONFIG_I2C_PROSAVAGE is not set +# CONFIG_I2C_SAVAGE4 is not set +# CONFIG_I2C_SIS5595 is not set +# CONFIG_I2C_SIS630 is not set +# CONFIG_I2C_SIS96X is not set # CONFIG_I2C_STUB is not set +# CONFIG_I2C_VIA is not set +# CONFIG_I2C_VIAPRO is not set +# CONFIG_I2C_VOODOO3 is not set # CONFIG_I2C_PCA_ISA is not set # @@ -776,9 +951,13 @@ CONFIG_HWMON=y # CONFIG_SENSORS_LM92 is not set # CONFIG_SENSORS_MAX1619 is not set # CONFIG_SENSORS_PC87360 is not set +# CONFIG_SENSORS_SIS5595 is not set # CONFIG_SENSORS_SMSC47M1 is not set # CONFIG_SENSORS_SMSC47M192 is not set # CONFIG_SENSORS_SMSC47B397 is not set +# CONFIG_SENSORS_VIA686A is not set +# CONFIG_SENSORS_VT1211 is not set +# CONFIG_SENSORS_VT8231 is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83791D is not set # CONFIG_SENSORS_W83792D is not set @@ -790,23 +969,25 @@ CONFIG_HWMON=y # # Misc devices # +# CONFIG_TIFM_CORE is not set # # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices # # CONFIG_DVB is not set +# CONFIG_USB_DABUSB is not set # # Graphics support # -# CONFIG_FIRMWARE_EDID is not set +CONFIG_FIRMWARE_EDID=y CONFIG_FB=y +# CONFIG_FB_DDC is not set # CONFIG_FB_CFB_FILLRECT is not set # CONFIG_FB_CFB_COPYAREA is not set # CONFIG_FB_CFB_IMAGEBLIT is not set @@ -814,14 +995,32 @@ CONFIG_FB=y # CONFIG_FB_BACKLIGHT is not set # CONFIG_FB_MODE_HELPERS is not set # CONFIG_FB_TILEBLITTING is not set +# CONFIG_FB_CIRRUS is not set +# CONFIG_FB_PM2 is not set +# CONFIG_FB_CYBER2000 is not set +# CONFIG_FB_ASILIANT is not set +# CONFIG_FB_IMSTT is not set # CONFIG_FB_S1D13XXX is not set +# CONFIG_FB_NVIDIA is not set +# CONFIG_FB_RIVA is not set +# CONFIG_FB_MATROX is not set +# CONFIG_FB_RADEON is not set +# CONFIG_FB_ATY128 is not set +# CONFIG_FB_ATY is not set +# CONFIG_FB_SAVAGE is not set +# CONFIG_FB_SIS is not set +# CONFIG_FB_NEOMAGIC is not set +# CONFIG_FB_KYRO is not set +# CONFIG_FB_3DFX is not set +# CONFIG_FB_VOODOO1 is not set +# CONFIG_FB_SMIVGX is not set +# CONFIG_FB_TRIDENT is not set # CONFIG_FB_VIRTUAL is not set # # Console display driver support # # CONFIG_VGA_CONSOLE is not set -# CONFIG_SGI_NEWPORT_CONSOLE is not set CONFIG_DUMMY_CONSOLE=y # CONFIG_FRAMEBUFFER_CONSOLE is not set @@ -839,15 +1038,129 @@ CONFIG_DUMMY_CONSOLE=y # # USB support # -# CONFIG_USB_ARCH_HAS_HCD is not set -# CONFIG_USB_ARCH_HAS_OHCI is not set -# CONFIG_USB_ARCH_HAS_EHCI is not set +CONFIG_USB_ARCH_HAS_HCD=y +CONFIG_USB_ARCH_HAS_OHCI=y +CONFIG_USB_ARCH_HAS_EHCI=y +CONFIG_USB=y +# CONFIG_USB_DEBUG is not set + +# +# Miscellaneous USB options +# +CONFIG_USB_DEVICEFS=y +# CONFIG_USB_BANDWIDTH is not set +# CONFIG_USB_DYNAMIC_MINORS is not set +# CONFIG_USB_OTG is not set + +# +# USB Host Controller Drivers +# +# CONFIG_USB_EHCI_HCD is not set +# CONFIG_USB_ISP116X_HCD is not set +# CONFIG_USB_OHCI_HCD is not set +# CONFIG_USB_UHCI_HCD is not set +# CONFIG_USB_SL811_HCD is not set + +# +# USB Device Class drivers +# +# CONFIG_USB_ACM is not set +# CONFIG_USB_PRINTER is not set # # NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' # # +# may also be needed; see USB_STORAGE Help for more information +# +CONFIG_USB_STORAGE=y +# CONFIG_USB_STORAGE_DEBUG is not set +# CONFIG_USB_STORAGE_DATAFAB is not set +# CONFIG_USB_STORAGE_FREECOM is not set +# CONFIG_USB_STORAGE_ISD200 is not set +# CONFIG_USB_STORAGE_DPCM is not set +# CONFIG_USB_STORAGE_USBAT is not set +# CONFIG_USB_STORAGE_SDDR09 is not set +# CONFIG_USB_STORAGE_SDDR55 is not set +# CONFIG_USB_STORAGE_JUMPSHOT is not set +# CONFIG_USB_STORAGE_ALAUDA is not set +# CONFIG_USB_STORAGE_ONETOUCH is not set +# CONFIG_USB_STORAGE_KARMA is not set +# CONFIG_USB_LIBUSUAL is not set + +# +# USB Input Devices +# +CONFIG_USB_HID=y +CONFIG_USB_HIDINPUT=y +# CONFIG_USB_HIDINPUT_POWERBOOK is not set +# CONFIG_HID_FF is not set +CONFIG_USB_HIDDEV=y +# CONFIG_USB_AIPTEK is not set +# CONFIG_USB_WACOM is not set +# CONFIG_USB_ACECAD is not set +# CONFIG_USB_KBTAB is not set +# CONFIG_USB_POWERMATE is not set +# CONFIG_USB_TOUCHSCREEN is not set +# CONFIG_USB_YEALINK is not set +# CONFIG_USB_XPAD is not set +# CONFIG_USB_ATI_REMOTE is not set +# CONFIG_USB_ATI_REMOTE2 is not set +# CONFIG_USB_KEYSPAN_REMOTE is not set +# CONFIG_USB_APPLETOUCH is not set +# CONFIG_USB_TRANCEVIBRATOR is not set + +# +# USB Imaging devices +# +# CONFIG_USB_MDC800 is not set +# CONFIG_USB_MICROTEK is not set + +# +# USB Network Adapters +# +# CONFIG_USB_CATC is not set +# CONFIG_USB_KAWETH is not set +# CONFIG_USB_PEGASUS is not set +# CONFIG_USB_RTL8150 is not set +# CONFIG_USB_USBNET is not set +CONFIG_USB_MON=y + +# +# USB port drivers +# + +# +# USB Serial Converter support +# +# CONFIG_USB_SERIAL is not set + +# +# USB Miscellaneous drivers +# +# CONFIG_USB_EMI62 is not set +# CONFIG_USB_EMI26 is not set +# CONFIG_USB_ADUTUX is not set +# CONFIG_USB_AUERSWALD is not set +# CONFIG_USB_RIO500 is not set +# CONFIG_USB_LEGOTOWER is not set +# CONFIG_USB_LCD is not set +# CONFIG_USB_LED is not set +# CONFIG_USB_CYPRESS_CY7C63 is not set +# CONFIG_USB_CYTHERM is not set +# CONFIG_USB_PHIDGET is not set +# CONFIG_USB_IDMOUSE is not set +# CONFIG_USB_FTDI_ELAN is not set +# CONFIG_USB_APPLEDISPLAY is not set +# CONFIG_USB_LD is not set +# CONFIG_USB_TEST is not set + +# +# USB DSL modem support +# + +# # USB Gadget Support # # CONFIG_USB_GADGET is not set @@ -873,6 +1186,7 @@ CONFIG_DUMMY_CONSOLE=y # # InfiniBand support # +# CONFIG_INFINIBAND is not set # # EDAC - error detection and reporting (RAS) (EXPERIMENTAL) @@ -906,6 +1220,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -917,6 +1232,7 @@ CONFIG_XFS_FS=m # CONFIG_XFS_SECURITY is not set # CONFIG_XFS_POSIX_ACL is not set # CONFIG_XFS_RT is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set @@ -949,8 +1265,10 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" # CONFIG_PROC_FS=y # CONFIG_PROC_KCORE is not set +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y # CONFIG_CONFIGFS_FS is not set @@ -994,7 +1312,6 @@ CONFIG_SUNRPC=y CONFIG_SMB_FS=m # CONFIG_SMB_NLS_DEFAULT is not set # CONFIG_CIFS is not set -# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -1005,7 +1322,6 @@ CONFIG_SMB_FS=m # # CONFIG_PARTITION_ADVANCED is not set CONFIG_MSDOS_PARTITION=y -CONFIG_SGI_PARTITION=y # # Native Language Support @@ -1061,11 +1377,13 @@ CONFIG_NLS_DEFAULT="iso8859-1" # CONFIG_TRACE_IRQFLAGS_SUPPORT=y # CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y # CONFIG_MAGIC_SYSRQ is not set # CONFIG_UNUSED_SYMBOLS is not set # CONFIG_DEBUG_KERNEL is not set CONFIG_LOG_BUF_SHIFT=14 # CONFIG_DEBUG_FS is not set +# CONFIG_HEADERS_CHECK is not set CONFIG_CROSSCOMPILE=y CONFIG_CMDLINE="" @@ -1079,6 +1397,9 @@ CONFIG_CMDLINE="" # Cryptographic options # CONFIG_CRYPTO=y +CONFIG_CRYPTO_ALGAPI=m +CONFIG_CRYPTO_BLKCIPHER=m +CONFIG_CRYPTO_MANAGER=m # CONFIG_CRYPTO_HMAC is not set # CONFIG_CRYPTO_NULL is not set # CONFIG_CRYPTO_MD4 is not set @@ -1088,6 +1409,8 @@ CONFIG_CRYPTO_SHA1=m # CONFIG_CRYPTO_SHA512 is not set # CONFIG_CRYPTO_WP512 is not set # CONFIG_CRYPTO_TGR192 is not set +CONFIG_CRYPTO_ECB=m +CONFIG_CRYPTO_CBC=m # CONFIG_CRYPTO_DES is not set # CONFIG_CRYPTO_BLOWFISH is not set # CONFIG_CRYPTO_TWOFISH is not set -- cgit v0.10.2 From e7a6f9c1f4e46f7fd776c6ee38bdb8f8ae131a82 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 18 Oct 2006 23:48:31 +0900 Subject: [MIPS] Update tb0287_defconfig Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/configs/tb0287_defconfig b/arch/mips/configs/tb0287_defconfig index ad7271b..f7e8194 100644 --- a/arch/mips/configs/tb0287_defconfig +++ b/arch/mips/configs/tb0287_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc1 -# Thu Jul 6 10:04:21 2006 +# Linux kernel version: 2.6.19-rc2 +# Wed Oct 18 12:57:11 2006 # CONFIG_MIPS=y @@ -25,8 +25,6 @@ CONFIG_MIPS=y # CONFIG_MIPS_COBALT is not set # CONFIG_MACH_DECSTATION is not set # CONFIG_MIPS_EV64120 is not set -# CONFIG_MIPS_IVR is not set -# CONFIG_MIPS_ITE8172 is not set # CONFIG_MACH_JAZZ is not set # CONFIG_LASAT is not set # CONFIG_MIPS_ATLAS is not set @@ -72,11 +70,11 @@ CONFIG_TANBAC_TB0287=y # CONFIG_VICTOR_MPC30X is not set # CONFIG_ZAO_CAPCELLA is not set CONFIG_PCI_VR41XX=y -# CONFIG_VRC4173 is not set CONFIG_RWSEM_GENERIC_SPINLOCK=y CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y CONFIG_DMA_NONCOHERENT=y CONFIG_DMA_NEED_PCI_MAP_STATE=y @@ -123,8 +121,8 @@ CONFIG_PAGE_SIZE_4KB=y # CONFIG_PAGE_SIZE_16KB is not set # CONFIG_PAGE_SIZE_64KB is not set CONFIG_MIPS_MT_DISABLED=y -# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_MT_SMP is not set +# CONFIG_MIPS_MT_SMTC is not set # CONFIG_MIPS_VPE_LOADER is not set CONFIG_CPU_HAS_SYNC=y CONFIG_GENERIC_HARDIRQS=y @@ -169,15 +167,19 @@ CONFIG_LOCALVERSION="" CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set # CONFIG_POSIX_MQUEUE is not set # CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y +# CONFIG_TASKSTATS is not set +# CONFIG_UTS_NS is not set # CONFIG_AUDIT is not set # CONFIG_IKCONFIG is not set # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y CONFIG_EMBEDDED=y +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_EXTRA_PASS is not set # CONFIG_HOTPLUG is not set @@ -185,12 +187,12 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y CONFIG_BASE_FULL=y -CONFIG_RT_MUTEXES=y CONFIG_FUTEX=y CONFIG_EPOLL=y CONFIG_SHMEM=y CONFIG_SLAB=y CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 # CONFIG_SLOB is not set @@ -208,6 +210,7 @@ CONFIG_KMOD=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_LBD is not set # CONFIG_BLK_DEV_IO_TRACE is not set # CONFIG_LSF is not set @@ -230,17 +233,16 @@ CONFIG_DEFAULT_IOSCHED="anticipatory" # CONFIG_HW_HAS_PCI=y CONFIG_PCI=y +# CONFIG_PCI_MULTITHREAD_PROBE is not set CONFIG_MMU=y # # PCCARD (PCMCIA/CardBus) support # -# CONFIG_PCCARD is not set # # PCI Hotplug Support # -# CONFIG_HOTPLUG_PCI is not set # # Executable file formats @@ -263,6 +265,7 @@ CONFIG_PACKET=y CONFIG_UNIX=y CONFIG_XFRM=y # CONFIG_XFRM_USER is not set +# CONFIG_XFRM_SUB_POLICY is not set # CONFIG_NET_KEY is not set CONFIG_INET=y CONFIG_IP_MULTICAST=y @@ -291,13 +294,10 @@ CONFIG_SYN_COOKIES=y CONFIG_INET_TUNNEL=m CONFIG_INET_XFRM_MODE_TRANSPORT=m CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y CONFIG_TCP_CONG_ADVANCED=y - -# -# TCP congestion control -# CONFIG_TCP_CONG_BIC=y CONFIG_TCP_CONG_CUBIC=m CONFIG_TCP_CONG_WESTWOOD=m @@ -308,7 +308,13 @@ CONFIG_TCP_CONG_HTCP=m # CONFIG_TCP_CONG_SCALABLE is not set # CONFIG_TCP_CONG_LP is not set # CONFIG_TCP_CONG_VENO is not set -# CONFIG_TCP_CONG_COMPOUND is not set +CONFIG_DEFAULT_BIC=y +# CONFIG_DEFAULT_CUBIC is not set +# CONFIG_DEFAULT_HTCP is not set +# CONFIG_DEFAULT_VEGAS is not set +# CONFIG_DEFAULT_WESTWOOD is not set +# CONFIG_DEFAULT_RENO is not set +CONFIG_DEFAULT_TCP_CONG="bic" # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set @@ -338,7 +344,6 @@ CONFIG_NETWORK_SECMARK=y # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -355,6 +360,7 @@ CONFIG_NETWORK_SECMARK=y # CONFIG_IRDA is not set # CONFIG_BT is not set # CONFIG_IEEE80211 is not set +CONFIG_FIB_RULES=y # # Device Drivers @@ -365,7 +371,6 @@ CONFIG_NETWORK_SECMARK=y # CONFIG_STANDALONE=y CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set # CONFIG_SYS_HYPERVISOR is not set # @@ -403,6 +408,7 @@ CONFIG_BLK_DEV_NBD=m CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 # CONFIG_BLK_DEV_INITRD is not set # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set @@ -410,65 +416,14 @@ CONFIG_BLK_DEV_RAM_SIZE=4096 # # ATA/ATAPI/MFM/RLL support # -CONFIG_IDE=y -CONFIG_BLK_DEV_IDE=y - -# -# Please see Documentation/ide.txt for help/info on IDE drives -# -# CONFIG_BLK_DEV_IDE_SATA is not set -CONFIG_BLK_DEV_IDEDISK=y -# CONFIG_IDEDISK_MULTI_MODE is not set -# CONFIG_BLK_DEV_IDECD is not set -# CONFIG_BLK_DEV_IDETAPE is not set -# CONFIG_BLK_DEV_IDEFLOPPY is not set -# CONFIG_BLK_DEV_IDESCSI is not set -# CONFIG_IDE_TASK_IOCTL is not set - -# -# IDE chipset support/bugfixes -# -CONFIG_IDE_GENERIC=y -CONFIG_BLK_DEV_IDEPCI=y -# CONFIG_IDEPCI_SHARE_IRQ is not set -# CONFIG_BLK_DEV_OFFBOARD is not set -# CONFIG_BLK_DEV_GENERIC is not set -# CONFIG_BLK_DEV_OPTI621 is not set -CONFIG_BLK_DEV_IDEDMA_PCI=y -# CONFIG_BLK_DEV_IDEDMA_FORCED is not set -# CONFIG_IDEDMA_PCI_AUTO is not set -# CONFIG_BLK_DEV_AEC62XX is not set -# CONFIG_BLK_DEV_ALI15X3 is not set -# CONFIG_BLK_DEV_AMD74XX is not set -# CONFIG_BLK_DEV_CMD64X is not set -# CONFIG_BLK_DEV_TRIFLEX is not set -# CONFIG_BLK_DEV_CY82C693 is not set -# CONFIG_BLK_DEV_CS5520 is not set -# CONFIG_BLK_DEV_CS5530 is not set -# CONFIG_BLK_DEV_HPT34X is not set -# CONFIG_BLK_DEV_HPT366 is not set -# CONFIG_BLK_DEV_SC1200 is not set -# CONFIG_BLK_DEV_PIIX is not set -# CONFIG_BLK_DEV_IT821X is not set -# CONFIG_BLK_DEV_NS87415 is not set -# CONFIG_BLK_DEV_PDC202XX_OLD is not set -# CONFIG_BLK_DEV_PDC202XX_NEW is not set -# CONFIG_BLK_DEV_SVWKS is not set -CONFIG_BLK_DEV_SIIMAGE=y -# CONFIG_BLK_DEV_SLC90E66 is not set -# CONFIG_BLK_DEV_TRM290 is not set -# CONFIG_BLK_DEV_VIA82CXXX is not set -# CONFIG_IDE_ARM is not set -CONFIG_BLK_DEV_IDEDMA=y -# CONFIG_IDEDMA_IVB is not set -# CONFIG_IDEDMA_AUTO is not set -# CONFIG_BLK_DEV_HD is not set +# CONFIG_IDE is not set # # SCSI device support # # CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y +# CONFIG_SCSI_NETLINK is not set CONFIG_SCSI_PROC_FS=y # @@ -489,12 +444,13 @@ CONFIG_BLK_DEV_SD=y # CONFIG_SCSI_LOGGING is not set # -# SCSI Transport Attributes +# SCSI Transports # # CONFIG_SCSI_SPI_ATTRS is not set # CONFIG_SCSI_FC_ATTRS is not set # CONFIG_SCSI_ISCSI_ATTRS is not set # CONFIG_SCSI_SAS_ATTRS is not set +# CONFIG_SCSI_SAS_LIBSAS is not set # # SCSI low-level drivers @@ -507,21 +463,24 @@ CONFIG_BLK_DEV_SD=y # CONFIG_SCSI_AIC7XXX is not set # CONFIG_SCSI_AIC7XXX_OLD is not set # CONFIG_SCSI_AIC79XX is not set +# CONFIG_SCSI_AIC94XX is not set # CONFIG_SCSI_DPT_I2O is not set +# CONFIG_SCSI_ARCMSR is not set # CONFIG_MEGARAID_NEWGEN is not set # CONFIG_MEGARAID_LEGACY is not set # CONFIG_MEGARAID_SAS is not set -# CONFIG_SCSI_SATA is not set # CONFIG_SCSI_HPTIOP is not set # CONFIG_SCSI_DMX3191D is not set # CONFIG_SCSI_FUTURE_DOMAIN is not set # CONFIG_SCSI_IPS is not set # CONFIG_SCSI_INITIO is not set # CONFIG_SCSI_INIA100 is not set +# CONFIG_SCSI_STEX is not set # CONFIG_SCSI_SYM53C8XX_2 is not set # CONFIG_SCSI_IPR is not set # CONFIG_SCSI_QLOGIC_1280 is not set # CONFIG_SCSI_QLA_FC is not set +# CONFIG_SCSI_QLA_ISCSI is not set # CONFIG_SCSI_LPFC is not set # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set @@ -529,6 +488,59 @@ CONFIG_BLK_DEV_SD=y # CONFIG_SCSI_DEBUG is not set # +# Serial ATA (prod) and Parallel ATA (experimental) drivers +# +CONFIG_ATA=y +# CONFIG_SATA_AHCI is not set +# CONFIG_SATA_SVW is not set +# CONFIG_ATA_PIIX is not set +# CONFIG_SATA_MV is not set +# CONFIG_SATA_NV is not set +# CONFIG_PDC_ADMA is not set +# CONFIG_SATA_QSTOR is not set +# CONFIG_SATA_PROMISE is not set +# CONFIG_SATA_SX4 is not set +# CONFIG_SATA_SIL is not set +# CONFIG_SATA_SIL24 is not set +# CONFIG_SATA_SIS is not set +# CONFIG_SATA_ULI is not set +# CONFIG_SATA_VIA is not set +# CONFIG_SATA_VITESSE is not set +# CONFIG_PATA_ALI is not set +# CONFIG_PATA_AMD is not set +# CONFIG_PATA_ARTOP is not set +# CONFIG_PATA_ATIIXP is not set +# CONFIG_PATA_CMD64X is not set +# CONFIG_PATA_CS5520 is not set +# CONFIG_PATA_CS5530 is not set +# CONFIG_PATA_CYPRESS is not set +# CONFIG_PATA_EFAR is not set +# CONFIG_ATA_GENERIC is not set +# CONFIG_PATA_HPT366 is not set +# CONFIG_PATA_HPT37X is not set +# CONFIG_PATA_HPT3X2N is not set +# CONFIG_PATA_HPT3X3 is not set +# CONFIG_PATA_IT821X is not set +# CONFIG_PATA_JMICRON is not set +# CONFIG_PATA_TRIFLEX is not set +# CONFIG_PATA_MPIIX is not set +# CONFIG_PATA_OLDPIIX is not set +# CONFIG_PATA_NETCELL is not set +# CONFIG_PATA_NS87410 is not set +# CONFIG_PATA_OPTI is not set +# CONFIG_PATA_OPTIDMA is not set +# CONFIG_PATA_PDC_OLD is not set +# CONFIG_PATA_RADISYS is not set +# CONFIG_PATA_RZ1000 is not set +# CONFIG_PATA_SC1200 is not set +# CONFIG_PATA_SERVERWORKS is not set +# CONFIG_PATA_PDC2027X is not set +CONFIG_PATA_SIL680=y +# CONFIG_PATA_SIS is not set +# CONFIG_PATA_VIA is not set +# CONFIG_PATA_WINBOND is not set + +# # Multi-device support (RAID and LVM) # # CONFIG_MD is not set @@ -632,6 +644,7 @@ CONFIG_R8169=y # CONFIG_SK98LIN is not set # CONFIG_TIGON3 is not set # CONFIG_BNX2 is not set +# CONFIG_QLA3XXX is not set # # Ethernet (10000 Mbit) @@ -679,6 +692,7 @@ CONFIG_R8169=y # Input device support # CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set # # Userland interfaces @@ -758,7 +772,6 @@ CONFIG_GPIO_VR41XX=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -784,12 +797,12 @@ CONFIG_GPIO_VR41XX=y # # Misc devices # +# CONFIG_TIFM_CORE is not set # # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -897,13 +910,13 @@ CONFIG_USB_STORAGE=m # CONFIG_USB_STORAGE_DEBUG is not set # CONFIG_USB_STORAGE_DATAFAB is not set # CONFIG_USB_STORAGE_FREECOM is not set -# CONFIG_USB_STORAGE_ISD200 is not set # CONFIG_USB_STORAGE_DPCM is not set # CONFIG_USB_STORAGE_USBAT is not set # CONFIG_USB_STORAGE_SDDR09 is not set # CONFIG_USB_STORAGE_SDDR55 is not set # CONFIG_USB_STORAGE_JUMPSHOT is not set # CONFIG_USB_STORAGE_ALAUDA is not set +# CONFIG_USB_STORAGE_KARMA is not set # CONFIG_USB_LIBUSUAL is not set # @@ -932,6 +945,7 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -963,16 +977,17 @@ CONFIG_USB_MON=y # # CONFIG_USB_EMI62 is not set # CONFIG_USB_EMI26 is not set +# CONFIG_USB_ADUTUX is not set # CONFIG_USB_AUERSWALD is not set # CONFIG_USB_RIO500 is not set # CONFIG_USB_LEGOTOWER is not set # CONFIG_USB_LCD is not set # CONFIG_USB_LED is not set -# CONFIG_USB_CY7C63 is not set +# CONFIG_USB_CYPRESS_CY7C63 is not set # CONFIG_USB_CYTHERM is not set -# CONFIG_USB_PHIDGETKIT is not set -# CONFIG_USB_PHIDGETSERVO is not set +# CONFIG_USB_PHIDGET is not set # CONFIG_USB_IDMOUSE is not set +# CONFIG_USB_FTDI_ELAN is not set # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set @@ -1041,6 +1056,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1052,6 +1068,7 @@ CONFIG_XFS_QUOTA=y # CONFIG_XFS_SECURITY is not set CONFIG_XFS_POSIX_ACL=y # CONFIG_XFS_RT is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set CONFIG_ROMFS_FS=m @@ -1082,8 +1099,10 @@ CONFIG_AUTOFS4_FS=y # CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y # CONFIG_CONFIGFS_FS is not set @@ -1123,7 +1142,6 @@ CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set # CONFIG_CIFS is not set -# CONFIG_CIFS_DEBUG2 is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -1150,11 +1168,13 @@ CONFIG_MSDOS_PARTITION=y # CONFIG_TRACE_IRQFLAGS_SUPPORT=y # CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y # CONFIG_MAGIC_SYSRQ is not set # CONFIG_UNUSED_SYMBOLS is not set # CONFIG_DEBUG_KERNEL is not set CONFIG_LOG_BUF_SHIFT=14 # CONFIG_DEBUG_FS is not set +# CONFIG_HEADERS_CHECK is not set CONFIG_CROSSCOMPILE=y CONFIG_CMDLINE="mem=64M console=ttyVR0,115200 ip=any root=/dev/nfs" @@ -1170,10 +1190,6 @@ CONFIG_CMDLINE="mem=64M console=ttyVR0,115200 ip=any root=/dev/nfs" # CONFIG_CRYPTO is not set # -# Hardware crypto devices -# - -# # Library routines # # CONFIG_CRC_CCITT is not set -- cgit v0.10.2 From d2bcf87d0fcdc10d1be65b03fd032bec05efe49f Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 18 Oct 2006 23:52:17 +0100 Subject: [MIPS] Reserve syscall numbers for kexec_load. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index 61362e6..720fac3 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -652,7 +652,8 @@ einval: li v0, -EINVAL sys sys_vmsplice 4 sys sys_move_pages 6 sys sys_set_robust_list 2 - sys sys_get_robust_list 3 + sys sys_get_robust_list 3 /* 4310 */ + sys sys_ni_syscall 0 .endm /* We pre-compute the number of _instruction_ bytes needed to diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 6c7b5ed..3a34f62 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -468,3 +468,4 @@ sys_call_table: PTR sys_move_pages PTR sys_set_robust_list PTR sys_get_robust_list + PTR sys_ni_syscall /* 5270 */ diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 8c453f8..67b92a1 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -394,3 +394,4 @@ EXPORT(sysn32_call_table) PTR sys_move_pages PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list + PTR sys_ni_syscall diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index d105917..2875c4a 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -516,4 +516,5 @@ sys_call_table: PTR compat_sys_move_pages PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list /* 4310 */ + PTR sys_ni_syscall .size sys_call_table,.-sys_call_table diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h index 685c914..30240a4 100644 --- a/include/asm-mips/unistd.h +++ b/include/asm-mips/unistd.h @@ -331,16 +331,17 @@ #define __NR_move_pages (__NR_Linux + 308) #define __NR_set_robust_list (__NR_Linux + 309) #define __NR_get_robust_list (__NR_Linux + 310) +#define __NR_kexec_load (__NR_Linux + 311) /* * Offset of the last Linux o32 flavoured syscall */ -#define __NR_Linux_syscalls 310 +#define __NR_Linux_syscalls 311 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */ #define __NR_O32_Linux 4000 -#define __NR_O32_Linux_syscalls 310 +#define __NR_O32_Linux_syscalls 311 #if _MIPS_SIM == _MIPS_SIM_ABI64 @@ -618,16 +619,17 @@ #define __NR_move_pages (__NR_Linux + 267) #define __NR_set_robust_list (__NR_Linux + 268) #define __NR_get_robust_list (__NR_Linux + 269) +#define __NR_kexec_load (__NR_Linux + 270) /* * Offset of the last Linux 64-bit flavoured syscall */ -#define __NR_Linux_syscalls 269 +#define __NR_Linux_syscalls 270 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */ #define __NR_64_Linux 5000 -#define __NR_64_Linux_syscalls 269 +#define __NR_64_Linux_syscalls 270 #if _MIPS_SIM == _MIPS_SIM_NABI32 @@ -909,16 +911,17 @@ #define __NR_move_pages (__NR_Linux + 271) #define __NR_set_robust_list (__NR_Linux + 272) #define __NR_get_robust_list (__NR_Linux + 273) +#define __NR_kexec_load (__NR_Linux + 274) /* * Offset of the last N32 flavoured syscall */ -#define __NR_Linux_syscalls 273 +#define __NR_Linux_syscalls 274 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */ #define __NR_N32_Linux 6000 -#define __NR_N32_Linux_syscalls 273 +#define __NR_N32_Linux_syscalls 274 #ifdef __KERNEL__ -- cgit v0.10.2 From d89e36d8df547fde2beaea82211954868da2282d Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 19 Oct 2006 14:21:47 +0100 Subject: [MIPS] Fix iounmap argument to const volatile. With the existing prototype the following code: const void __iomem *io = ioremap(); x = readb(io); iounmap(io); did result in a warning. Signed-off-by: Ralf Baechle diff --git a/arch/mips/mm/ioremap.c b/arch/mips/mm/ioremap.c index 3101d1d..cea7d0e 100644 --- a/arch/mips/mm/ioremap.c +++ b/arch/mips/mm/ioremap.c @@ -176,7 +176,7 @@ void __iomem * __ioremap(phys_t phys_addr, phys_t size, unsigned long flags) #define IS_KSEG1(addr) (((unsigned long)(addr) & ~0x1fffffffUL) == CKSEG1) -void __iounmap(volatile void __iomem *addr) +void __iounmap(const volatile void __iomem *addr) { struct vm_struct *p; diff --git a/include/asm-mips/io.h b/include/asm-mips/io.h index c2d124b..bc5f3c5 100644 --- a/include/asm-mips/io.h +++ b/include/asm-mips/io.h @@ -172,7 +172,7 @@ extern unsigned long isa_slot_offset; #define page_to_phys(page) ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT) extern void __iomem * __ioremap(phys_t offset, phys_t size, unsigned long flags); -extern void __iounmap(volatile void __iomem *addr); +extern void __iounmap(const volatile void __iomem *addr); static inline void __iomem * __ioremap_mode(phys_t offset, unsigned long size, unsigned long flags) @@ -279,7 +279,7 @@ static inline void __iomem * __ioremap_mode(phys_t offset, unsigned long size, #define ioremap_uncached_accelerated(offset, size) \ __ioremap_mode((offset), (size), _CACHE_UNCACHED_ACCELERATED) -static inline void iounmap(volatile void __iomem *addr) +static inline void iounmap(const volatile void __iomem *addr) { #define __IS_KSEG1(addr) (((unsigned long)(addr) & ~0x1fffffffUL) == CKSEG1) -- cgit v0.10.2 From 62752ee198dca9209b7dee504763e51b11e9e0ca Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 17 Oct 2006 10:31:38 +0200 Subject: [PATCH] Take i_mutex in splice_from_pipe() The splice_actor may be calling ->prepare_write() and ->commit_write(). We want i_mutex on the inode being written to before calling those so that we don't race i_size changes. The double locking behavior is done elsewhere in splice.c, and if we eventually want _nolock variants of generic_file_splice_write(), fs modules might have to replicate the nasty locking code. We introduce inode_double_lock() and inode_double_unlock() to consolidate the locking rules into one set of functions. Signed-off-by: Mark Fasheh Signed-off-by: Jens Axboe diff --git a/fs/inode.c b/fs/inode.c index d9a21d1..26cdb11 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1306,6 +1306,42 @@ void wake_up_inode(struct inode *inode) wake_up_bit(&inode->i_state, __I_LOCK); } +/* + * We rarely want to lock two inodes that do not have a parent/child + * relationship (such as directory, child inode) simultaneously. The + * vast majority of file systems should be able to get along fine + * without this. Do not use these functions except as a last resort. + */ +void inode_double_lock(struct inode *inode1, struct inode *inode2) +{ + if (inode1 == NULL || inode2 == NULL || inode1 == inode2) { + if (inode1) + mutex_lock(&inode1->i_mutex); + else if (inode2) + mutex_lock(&inode2->i_mutex); + return; + } + + if (inode1 < inode2) { + mutex_lock_nested(&inode1->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode2->i_mutex, I_MUTEX_CHILD); + } else { + mutex_lock_nested(&inode2->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode1->i_mutex, I_MUTEX_CHILD); + } +} +EXPORT_SYMBOL(inode_double_lock); + +void inode_double_unlock(struct inode *inode1, struct inode *inode2) +{ + if (inode1) + mutex_unlock(&inode1->i_mutex); + + if (inode2 && inode2 != inode1) + mutex_unlock(&inode2->i_mutex); +} +EXPORT_SYMBOL(inode_double_unlock); + static __initdata unsigned long ihash_entries; static int __init set_ihash_entries(char *str) { diff --git a/fs/splice.c b/fs/splice.c index a567010..c1072b6 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -713,6 +713,7 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, { int ret, do_wakeup, err; struct splice_desc sd; + struct inode *inode = out->f_mapping->host; ret = 0; do_wakeup = 0; @@ -722,8 +723,13 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, sd.file = out; sd.pos = *ppos; - if (pipe->inode) - mutex_lock(&pipe->inode->i_mutex); + /* + * The actor worker might be calling ->prepare_write and + * ->commit_write. Most of the time, these expect i_mutex to + * be held. Since this may result in an ABBA deadlock with + * pipe->inode, we have to order lock acquiry here. + */ + inode_double_lock(inode, pipe->inode); for (;;) { if (pipe->nrbufs) { @@ -797,8 +803,7 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, pipe_wait(pipe); } - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); + inode_double_unlock(inode, pipe->inode); if (do_wakeup) { smp_mb(); @@ -1400,13 +1405,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, * grabbing by inode address. Otherwise two different processes * could deadlock (one doing tee from A -> B, the other from B -> A). */ - if (ipipe->inode < opipe->inode) { - mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_CHILD); - } else { - mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_CHILD); - } + inode_double_lock(ipipe->inode, opipe->inode); do { if (!opipe->readers) { @@ -1450,8 +1449,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, i++; } while (len); - mutex_unlock(&ipipe->inode->i_mutex); - mutex_unlock(&opipe->inode->i_mutex); + inode_double_unlock(ipipe->inode, opipe->inode); /* * If we put data in the output pipe, wakeup any potential readers. diff --git a/include/linux/fs.h b/include/linux/fs.h index 661c7c5..853a02f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -623,6 +623,9 @@ enum inode_i_mutex_lock_class I_MUTEX_QUOTA }; +extern void inode_double_lock(struct inode *inode1, struct inode *inode2); +extern void inode_double_unlock(struct inode *inode1, struct inode *inode2); + /* * NOTE: in a 32bit arch with a preemptable kernel and * an UP compile the i_size_read/write must be atomic -- cgit v0.10.2 From 6da61809822c22634a3de2dcb3c60283b836a88a Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 17 Oct 2006 18:43:07 +0200 Subject: [PATCH] Introduce generic_file_splice_write_nolock() This allows file systems to manage their own i_mutex locking while still re-using the generic_file_splice_write() logic. OCFS2 in particular wants this so that it can order cluster locks within i_mutex. Signed-off-by: Mark Fasheh Signed-off-by: Jens Axboe diff --git a/fs/splice.c b/fs/splice.c index c1072b6..68e20e6 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -707,13 +707,12 @@ out_ret: * key here is the 'actor' worker passed in that actually moves the data * to the wanted destination. See pipe_to_file/pipe_to_sendpage above. */ -ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, - loff_t *ppos, size_t len, unsigned int flags, - splice_actor *actor) +static ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, + struct file *out, loff_t *ppos, size_t len, + unsigned int flags, splice_actor *actor) { int ret, do_wakeup, err; struct splice_desc sd; - struct inode *inode = out->f_mapping->host; ret = 0; do_wakeup = 0; @@ -723,14 +722,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, sd.file = out; sd.pos = *ppos; - /* - * The actor worker might be calling ->prepare_write and - * ->commit_write. Most of the time, these expect i_mutex to - * be held. Since this may result in an ABBA deadlock with - * pipe->inode, we have to order lock acquiry here. - */ - inode_double_lock(inode, pipe->inode); - for (;;) { if (pipe->nrbufs) { struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; @@ -803,8 +794,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, pipe_wait(pipe); } - inode_double_unlock(inode, pipe->inode); - if (do_wakeup) { smp_mb(); if (waitqueue_active(&pipe->wait)) @@ -815,6 +804,69 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, return ret; } +ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags, + splice_actor *actor) +{ + ssize_t ret; + struct inode *inode = out->f_mapping->host; + + /* + * The actor worker might be calling ->prepare_write and + * ->commit_write. Most of the time, these expect i_mutex to + * be held. Since this may result in an ABBA deadlock with + * pipe->inode, we have to order lock acquiry here. + */ + inode_double_lock(inode, pipe->inode); + ret = __splice_from_pipe(pipe, out, ppos, len, flags, actor); + inode_double_unlock(inode, pipe->inode); + + return ret; +} + +/** + * generic_file_splice_write_nolock - generic_file_splice_write without mutexes + * @pipe: pipe info + * @out: file to write to + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Will either move or copy pages (determined by @flags options) from + * the given pipe inode to the given file. The caller is responsible + * for acquiring i_mutex on both inodes. + * + */ +ssize_t +generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; + ssize_t ret; + int err; + + ret = __splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); + if (ret > 0) { + *ppos += ret; + + /* + * If file or inode is SYNC and we actually wrote some data, + * sync it. + */ + if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { + err = generic_osync_inode(inode, mapping, + OSYNC_METADATA|OSYNC_DATA); + + if (err) + ret = err; + } + } + + return ret; +} + +EXPORT_SYMBOL(generic_file_splice_write_nolock); + /** * generic_file_splice_write - splice data from a pipe to a file * @pipe: pipe info diff --git a/include/linux/fs.h b/include/linux/fs.h index 853a02f..d695ba2 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1758,6 +1758,8 @@ extern ssize_t generic_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); extern ssize_t generic_file_splice_write(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); +extern ssize_t generic_file_splice_write_nolock(struct pipe_inode_info *, + struct file *, loff_t *, size_t, unsigned int); extern ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe, struct file *out, loff_t *, size_t len, unsigned int flags); extern long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, -- cgit v0.10.2 From 01de85e057328ecbef36e108673b1e81059d54c1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 17 Oct 2006 19:50:36 +0200 Subject: [PATCH] Add lockless helpers for remove_suid() Right now users have to grab i_mutex before calling remove_suid(), in the unlikely event that a call to ->setattr() may be needed. Split up the function in two parts: - One to check if we need to remove suid - One to actually remove it The first we can call lockless. Signed-off-by: Jens Axboe diff --git a/include/linux/fs.h b/include/linux/fs.h index d695ba2..2fe6e3f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1712,6 +1712,8 @@ extern void __iget(struct inode * inode); extern void clear_inode(struct inode *); extern void destroy_inode(struct inode *); extern struct inode *new_inode(struct super_block *); +extern int __remove_suid(struct dentry *, int); +extern int should_remove_suid(struct dentry *); extern int remove_suid(struct dentry *); extern void remove_dquot_ref(struct super_block *, int, struct list_head *); diff --git a/mm/filemap.c b/mm/filemap.c index 3464b68..7c7addb 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1884,11 +1884,10 @@ repeat: * if suid or (sgid and xgrp) * remove privs */ -int remove_suid(struct dentry *dentry) +int should_remove_suid(struct dentry *dentry) { mode_t mode = dentry->d_inode->i_mode; int kill = 0; - int result = 0; /* suid always must be killed */ if (unlikely(mode & S_ISUID)) @@ -1901,13 +1900,28 @@ int remove_suid(struct dentry *dentry) if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) kill |= ATTR_KILL_SGID; - if (unlikely(kill && !capable(CAP_FSETID))) { - struct iattr newattrs; + if (unlikely(kill && !capable(CAP_FSETID))) + return kill; - newattrs.ia_valid = ATTR_FORCE | kill; - result = notify_change(dentry, &newattrs); - } - return result; + return 0; +} + +int __remove_suid(struct dentry *dentry, int kill) +{ + struct iattr newattrs; + + newattrs.ia_valid = ATTR_FORCE | kill; + return notify_change(dentry, &newattrs); +} + +int remove_suid(struct dentry *dentry) +{ + int kill = should_remove_suid(dentry); + + if (unlikely(kill)) + return __remove_suid(dentry, kill); + + return 0; } EXPORT_SYMBOL(remove_suid); -- cgit v0.10.2 From 8c34e2d63231d4bf4852bac8521883944d770fe3 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 17 Oct 2006 19:43:22 +0200 Subject: [PATCH] Remove SUID when splicing into an inode Originally from Mark Fasheh generic_file_splice_write() does not remove S_ISUID or S_ISGID. This is inconsistent with the way we generally write to files. Signed-off-by: Mark Fasheh Signed-off-by: Jens Axboe diff --git a/fs/splice.c b/fs/splice.c index 68e20e6..49fb9f1 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -845,6 +845,10 @@ generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; int err; + err = remove_suid(out->f_dentry); + if (unlikely(err)) + return err; + ret = __splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); if (ret > 0) { *ppos += ret; @@ -883,12 +887,21 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, loff_t *ppos, size_t len, unsigned int flags) { struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; ssize_t ret; + int err; + + err = should_remove_suid(out->f_dentry); + if (unlikely(err)) { + mutex_lock(&inode->i_mutex); + err = __remove_suid(out->f_dentry, err); + mutex_unlock(&inode->i_mutex); + if (err) + return err; + } ret = splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); if (ret > 0) { - struct inode *inode = mapping->host; - *ppos += ret; /* @@ -896,8 +909,6 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, * sync it. */ if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { - int err; - mutex_lock(&inode->i_mutex); err = generic_osync_inode(inode, mapping, OSYNC_METADATA|OSYNC_DATA); -- cgit v0.10.2 From 98978edb6278f0fb30c8d636c6a79a28f9164cb9 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Tue, 17 Oct 2006 10:54:36 -0400 Subject: [PATCH] wireless: WE-20 compatibility for ESSID and NICKN ioctls WE-21 changed the ABI for the SIOC[SG]IW{ESSID,NICKN} ioctls by dropping NULL termination. This patch adds compatibility code so that WE-21 can work properly with WE-20 (and older) tools. Signed-off-by: John W. Linville diff --git a/net/core/wireless.c b/net/core/wireless.c index ffff0da..cb1b872 100644 --- a/net/core/wireless.c +++ b/net/core/wireless.c @@ -748,11 +748,39 @@ static int ioctl_standard_call(struct net_device * dev, int extra_size; int user_length = 0; int err; + int essid_compat = 0; /* Calculate space needed by arguments. Always allocate * for max space. Easier, and won't last long... */ extra_size = descr->max_tokens * descr->token_size; + /* Check need for ESSID compatibility for WE < 21 */ + switch (cmd) { + case SIOCSIWESSID: + case SIOCGIWESSID: + case SIOCSIWNICKN: + case SIOCGIWNICKN: + if (iwr->u.data.length == descr->max_tokens + 1) + essid_compat = 1; + else if (IW_IS_SET(cmd) && (iwr->u.data.length != 0)) { + char essid[IW_ESSID_MAX_SIZE + 1]; + + err = copy_from_user(essid, iwr->u.data.pointer, + iwr->u.data.length * + descr->token_size); + if (err) + return -EFAULT; + + if (essid[iwr->u.data.length - 1] == '\0') + essid_compat = 1; + } + break; + default: + break; + } + + iwr->u.data.length -= essid_compat; + /* Check what user space is giving us */ if(IW_IS_SET(cmd)) { /* Check NULL pointer */ @@ -795,7 +823,8 @@ static int ioctl_standard_call(struct net_device * dev, #endif /* WE_IOCTL_DEBUG */ /* Create the kernel buffer */ - extra = kmalloc(extra_size, GFP_KERNEL); + /* kzalloc ensures NULL-termination for essid_compat */ + extra = kzalloc(extra_size, GFP_KERNEL); if (extra == NULL) { return -ENOMEM; } @@ -819,6 +848,8 @@ static int ioctl_standard_call(struct net_device * dev, /* Call the handler */ ret = handler(dev, &info, &(iwr->u), extra); + iwr->u.data.length += essid_compat; + /* If we have something to return to the user */ if (!ret && IW_IS_GET(cmd)) { /* Check if there is enough buffer up there */ -- cgit v0.10.2 From e83f214e45a80459d2931ae1259ab5fae82c8a8e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Oct 2006 12:48:35 -0700 Subject: [SPARC64]: Compute dma_end argument to sabre_pbm_init() correctly. virtual-dma property layout is [start, size] not [start, end]. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c index de7f785..94bb681 100644 --- a/arch/sparc64/kernel/pci_sabre.c +++ b/arch/sparc64/kernel/pci_sabre.c @@ -1465,5 +1465,5 @@ void sabre_init(struct device_node *dp, char *model_name) /* * Look for APB underneath. */ - sabre_pbm_init(p, dp, vdma[0], vdma[1]); + sabre_pbm_init(p, dp, vdma[0], vdma[0] + vdma[1]); } -- cgit v0.10.2 From 6bda57365a5fda4743d83a5987c6aab66e90771c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 18 Oct 2006 23:00:35 -0700 Subject: [SPARC64]: Fix of_ioremap(). Use request_mem_region() if IORESOURCE_MEM. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c index d822c7c..d3dfb2a 100644 --- a/arch/sparc64/kernel/of_device.c +++ b/arch/sparc64/kernel/of_device.c @@ -131,8 +131,13 @@ static int of_device_resume(struct device * dev) void __iomem *of_ioremap(struct resource *res, unsigned long offset, unsigned long size, char *name) { unsigned long ret = res->start + offset; + struct resource *r; - if (!request_region(ret, size, name)) + if (res->flags & IORESOURCE_MEM) + r = request_mem_region(ret, size, name); + else + r = request_region(ret, size, name); + if (!r) ret = 0; return (void __iomem *) ret; -- cgit v0.10.2 From 66a740572d7bcb18469e71cb014bfed3ff75a773 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 20 Oct 2006 15:30:55 +0900 Subject: sh: Convert INTC2 to IRQ table registration. Currently the INTC2 code contains a fixed IRQ table that it iterates through to set the handler type, we move this in to the CPU subtype setup code instead and allow for submitting the table that way. This drops the ST40 tables, as nothing has been happening with those processors, while converting the only existing users to use the new table directly (SH7760 and SH7780). Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/cpu/irq/intc2.c b/arch/sh/kernel/cpu/irq/intc2.c index 212884a..74ca576 100644 --- a/arch/sh/kernel/cpu/irq/intc2.c +++ b/arch/sh/kernel/cpu/irq/intc2.c @@ -11,10 +11,9 @@ * Hitachi 7751, the STM ST40 STB1, SH7760, and SH7780. */ #include -#include #include +#include #include -#include static void disable_intc2_irq(unsigned int irq) { @@ -45,151 +44,36 @@ static struct irq_chip intc2_irq_chip = { * PIO1 which is INTPRI00[19,16] and INTMSK00[13] * would be: ^ ^ ^ ^ * | | | | - * make_intc2_irq(84, 0, 16, 0, 13); + * { 84, 0, 16, 0, 13 }, + * + * in the intc2_data table. */ -void make_intc2_irq(struct intc2_data *p) +void make_intc2_irq(struct intc2_data *table, unsigned int nr_irqs) { - unsigned int flags; - unsigned long ipr; - - disable_irq_nosync(p->irq); - - /* Set the priority level */ - local_irq_save(flags); - - ipr = ctrl_inl(INTC2_BASE + INTC2_INTPRI_OFFSET + p->ipr_offset); - ipr &= ~(0xf << p->ipr_shift); - ipr |= p->priority << p->ipr_shift; - ctrl_outl(ipr, INTC2_BASE + INTC2_INTPRI_OFFSET + p->ipr_offset); - - local_irq_restore(flags); + int i; - set_irq_chip_and_handler_name(p->irq, &intc2_irq_chip, - handle_level_irq, "level"); - set_irq_chip_data(p->irq, p); + for (i = 0; i < nr_irqs; i++) { + unsigned long ipr, flags; + struct intc2_data *p = table + i; - enable_intc2_irq(p->irq); -} + disable_irq_nosync(p->irq); -static struct intc2_data intc2_irq_table[] = { -#if defined(CONFIG_CPU_SUBTYPE_ST40) - {64, 0, 0, 0, 0, 13}, /* PCI serr */ - {65, 0, 4, 0, 1, 13}, /* PCI err */ - {66, 0, 4, 0, 2, 13}, /* PCI ad */ - {67, 0, 4, 0, 3, 13}, /* PCI pwd down */ - {72, 0, 8, 0, 5, 13}, /* DMAC INT0 */ - {73, 0, 8, 0, 6, 13}, /* DMAC INT1 */ - {74, 0, 8, 0, 7, 13}, /* DMAC INT2 */ - {75, 0, 8, 0, 8, 13}, /* DMAC INT3 */ - {76, 0, 8, 0, 9, 13}, /* DMAC INT4 */ - {78, 0, 8, 0, 11, 13}, /* DMAC ERR */ - {80, 0, 12, 0, 12, 13}, /* PIO0 */ - {84, 0, 16, 0, 13, 13}, /* PIO1 */ - {88, 0, 20, 0, 14, 13}, /* PIO2 */ - {112, 4, 0, 4, 0, 13}, /* Mailbox */ - #ifdef CONFIG_CPU_SUBTYPE_ST40GX1 - {116, 4, 4, 4, 4, 13}, /* SSC0 */ - {120, 4, 8, 4, 8, 13}, /* IR Blaster */ - {124, 4, 12, 4, 12, 13}, /* USB host */ - {128, 4, 16, 4, 16, 13}, /* Video processor BLITTER */ - {132, 4, 20, 4, 20, 13}, /* UART0 */ - {134, 4, 20, 4, 22, 13}, /* UART2 */ - {136, 4, 24, 4, 24, 13}, /* IO_PIO0 */ - {140, 4, 28, 4, 28, 13}, /* EMPI */ - {144, 8, 0, 8, 0, 13}, /* MAFE */ - {148, 8, 4, 8, 4, 13}, /* PWM */ - {152, 8, 8, 8, 8, 13}, /* SSC1 */ - {156, 8, 12, 8, 12, 13}, /* IO_PIO1 */ - {160, 8, 16, 8, 16, 13}, /* USB target */ - {164, 8, 20, 8, 20, 13}, /* UART1 */ - {168, 8, 24, 8, 24, 13}, /* Teletext */ - {172, 8, 28, 8, 28, 13}, /* VideoSync VTG */ - {173, 8, 28, 8, 29, 13}, /* VideoSync DVP0 */ - {174, 8, 28, 8, 30, 13}, /* VideoSync DVP1 */ -#endif -#elif defined(CONFIG_CPU_SUBTYPE_SH7760) -/* - * SH7760 INTC2-Style interrupts, vectors IRQ48-111 INTEVT 0x800-0xFE0 - */ - /* INTPRIO0 | INTMSK0 */ - {48, 0, 28, 0, 31, 3}, /* IRQ 4 */ - {49, 0, 24, 0, 30, 3}, /* IRQ 3 */ - {50, 0, 20, 0, 29, 3}, /* IRQ 2 */ - {51, 0, 16, 0, 28, 3}, /* IRQ 1 */ - /* 52-55 (INTEVT 0x880-0x8E0) unused/reserved */ - /* INTPRIO4 | INTMSK0 */ - {56, 4, 28, 0, 25, 3}, /* HCAN2_CHAN0 */ - {57, 4, 24, 0, 24, 3}, /* HCAN2_CHAN1 */ - {58, 4, 20, 0, 23, 3}, /* I2S_CHAN0 */ - {59, 4, 16, 0, 22, 3}, /* I2S_CHAN1 */ - {60, 4, 12, 0, 21, 3}, /* AC97_CHAN0 */ - {61, 4, 8, 0, 20, 3}, /* AC97_CHAN1 */ - {62, 4, 4, 0, 19, 3}, /* I2C_CHAN0 */ - {63, 4, 0, 0, 18, 3}, /* I2C_CHAN1 */ - /* INTPRIO8 | INTMSK0 */ - {52, 8, 16, 0, 11, 3}, /* SCIF0_ERI_IRQ */ - {53, 8, 16, 0, 10, 3}, /* SCIF0_RXI_IRQ */ - {54, 8, 16, 0, 9, 3}, /* SCIF0_BRI_IRQ */ - {55, 8, 16, 0, 8, 3}, /* SCIF0_TXI_IRQ */ - {64, 8, 28, 0, 17, 3}, /* USBHI_IRQ */ - {65, 8, 24, 0, 16, 3}, /* LCDC */ - /* 66, 67 unused */ - {68, 8, 20, 0, 14, 13}, /* DMABRGI0_IRQ */ - {69, 8, 20, 0, 13, 13}, /* DMABRGI1_IRQ */ - {70, 8, 20, 0, 12, 13}, /* DMABRGI2_IRQ */ - /* 71 unused */ - {72, 8, 12, 0, 7, 3}, /* SCIF1_ERI_IRQ */ - {73, 8, 12, 0, 6, 3}, /* SCIF1_RXI_IRQ */ - {74, 8, 12, 0, 5, 3}, /* SCIF1_BRI_IRQ */ - {75, 8, 12, 0, 4, 3}, /* SCIF1_TXI_IRQ */ - {76, 8, 8, 0, 3, 3}, /* SCIF2_ERI_IRQ */ - {77, 8, 8, 0, 2, 3}, /* SCIF2_RXI_IRQ */ - {78, 8, 8, 0, 1, 3}, /* SCIF2_BRI_IRQ */ - {79, 8, 8, 0, 0, 3}, /* SCIF2_TXI_IRQ */ - /* | INTMSK4 */ - {80, 8, 4, 4, 23, 3}, /* SIM_ERI */ - {81, 8, 4, 4, 22, 3}, /* SIM_RXI */ - {82, 8, 4, 4, 21, 3}, /* SIM_TXI */ - {83, 8, 4, 4, 20, 3}, /* SIM_TEI */ - {84, 8, 0, 4, 19, 3}, /* HSPII */ - /* INTPRIOC | INTMSK4 */ - /* 85-87 unused/reserved */ - {88, 12, 20, 4, 18, 3}, /* MMCI0 */ - {89, 12, 20, 4, 17, 3}, /* MMCI1 */ - {90, 12, 20, 4, 16, 3}, /* MMCI2 */ - {91, 12, 20, 4, 15, 3}, /* MMCI3 */ - {92, 12, 12, 4, 6, 3}, /* MFI (unsure, bug? in my 7760 manual*/ - /* 93-107 reserved/undocumented */ - {108,12, 4, 4, 1, 3}, /* ADC */ - {109,12, 0, 4, 0, 3}, /* CMTI */ - /* 110-111 reserved/unused */ -#elif defined(CONFIG_CPU_SUBTYPE_SH7780) - { TIMER_IRQ, 0, 24, 0, INTC_TMU0_MSK, 2}, - { 21, 1, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY }, - { 22, 1, 1, 0, INTC_RTC_MSK, TIMER_PRIORITY }, - { 23, 1, 2, 0, INTC_RTC_MSK, TIMER_PRIORITY }, - { SCIF0_ERI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, - { SCIF0_RXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, - { SCIF0_BRI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, - { SCIF0_TXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, + /* Set the priority level */ + local_irq_save(flags); - { SCIF1_ERI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, - { SCIF1_RXI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, - { SCIF1_BRI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, - { SCIF1_TXI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, + ipr = ctrl_inl(INTC2_BASE + INTC2_INTPRI_OFFSET + + p->ipr_offset); + ipr &= ~(0xf << p->ipr_shift); + ipr |= p->priority << p->ipr_shift; + ctrl_outl(ipr, INTC2_BASE + INTC2_INTPRI_OFFSET + + p->ipr_offset); - { PCIC0_IRQ, 0x10, 8, 0, INTC_PCIC0_MSK, PCIC0_PRIORITY }, - { PCIC1_IRQ, 0x10, 0, 0, INTC_PCIC1_MSK, PCIC1_PRIORITY }, - { PCIC2_IRQ, 0x14, 24, 0, INTC_PCIC2_MSK, PCIC2_PRIORITY }, - { PCIC3_IRQ, 0x14, 16, 0, INTC_PCIC3_MSK, PCIC3_PRIORITY }, - { PCIC4_IRQ, 0x14, 8, 0, INTC_PCIC4_MSK, PCIC4_PRIORITY }, -#endif -}; + local_irq_restore(flags); -void __init init_IRQ_intc2(void) -{ - int i; + set_irq_chip_and_handler_name(p->irq, &intc2_irq_chip, + handle_level_irq, "level"); + set_irq_chip_data(p->irq, p); - for (i = 0; i < ARRAY_SIZE(intc2_irq_table); i++) - make_intc2_irq(intc2_irq_table + i); + enable_intc2_irq(p->irq); + } } diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7760.c b/arch/sh/kernel/cpu/sh4/setup-sh7760.c index 97f1c9a..07e5377 100644 --- a/arch/sh/kernel/cpu/sh4/setup-sh7760.c +++ b/arch/sh/kernel/cpu/sh4/setup-sh7760.c @@ -51,3 +51,66 @@ static int __init sh7760_devices_setup(void) ARRAY_SIZE(sh7760_devices)); } __initcall(sh7760_devices_setup); + +/* + * SH7760 INTC2-Style interrupts, vectors IRQ48-111 INTEVT 0x800-0xFE0 + */ +static struct intc2_data intc2_irq_table[] = { + /* INTPRIO0 | INTMSK0 */ + {48, 0, 28, 0, 31, 3}, /* IRQ 4 */ + {49, 0, 24, 0, 30, 3}, /* IRQ 3 */ + {50, 0, 20, 0, 29, 3}, /* IRQ 2 */ + {51, 0, 16, 0, 28, 3}, /* IRQ 1 */ + /* 52-55 (INTEVT 0x880-0x8E0) unused/reserved */ + /* INTPRIO4 | INTMSK0 */ + {56, 4, 28, 0, 25, 3}, /* HCAN2_CHAN0 */ + {57, 4, 24, 0, 24, 3}, /* HCAN2_CHAN1 */ + {58, 4, 20, 0, 23, 3}, /* I2S_CHAN0 */ + {59, 4, 16, 0, 22, 3}, /* I2S_CHAN1 */ + {60, 4, 12, 0, 21, 3}, /* AC97_CHAN0 */ + {61, 4, 8, 0, 20, 3}, /* AC97_CHAN1 */ + {62, 4, 4, 0, 19, 3}, /* I2C_CHAN0 */ + {63, 4, 0, 0, 18, 3}, /* I2C_CHAN1 */ + /* INTPRIO8 | INTMSK0 */ + {52, 8, 16, 0, 11, 3}, /* SCIF0_ERI_IRQ */ + {53, 8, 16, 0, 10, 3}, /* SCIF0_RXI_IRQ */ + {54, 8, 16, 0, 9, 3}, /* SCIF0_BRI_IRQ */ + {55, 8, 16, 0, 8, 3}, /* SCIF0_TXI_IRQ */ + {64, 8, 28, 0, 17, 3}, /* USBHI_IRQ */ + {65, 8, 24, 0, 16, 3}, /* LCDC */ + /* 66, 67 unused */ + {68, 8, 20, 0, 14, 13}, /* DMABRGI0_IRQ */ + {69, 8, 20, 0, 13, 13}, /* DMABRGI1_IRQ */ + {70, 8, 20, 0, 12, 13}, /* DMABRGI2_IRQ */ + /* 71 unused */ + {72, 8, 12, 0, 7, 3}, /* SCIF1_ERI_IRQ */ + {73, 8, 12, 0, 6, 3}, /* SCIF1_RXI_IRQ */ + {74, 8, 12, 0, 5, 3}, /* SCIF1_BRI_IRQ */ + {75, 8, 12, 0, 4, 3}, /* SCIF1_TXI_IRQ */ + {76, 8, 8, 0, 3, 3}, /* SCIF2_ERI_IRQ */ + {77, 8, 8, 0, 2, 3}, /* SCIF2_RXI_IRQ */ + {78, 8, 8, 0, 1, 3}, /* SCIF2_BRI_IRQ */ + {79, 8, 8, 0, 0, 3}, /* SCIF2_TXI_IRQ */ + /* | INTMSK4 */ + {80, 8, 4, 4, 23, 3}, /* SIM_ERI */ + {81, 8, 4, 4, 22, 3}, /* SIM_RXI */ + {82, 8, 4, 4, 21, 3}, /* SIM_TXI */ + {83, 8, 4, 4, 20, 3}, /* SIM_TEI */ + {84, 8, 0, 4, 19, 3}, /* HSPII */ + /* INTPRIOC | INTMSK4 */ + /* 85-87 unused/reserved */ + {88, 12, 20, 4, 18, 3}, /* MMCI0 */ + {89, 12, 20, 4, 17, 3}, /* MMCI1 */ + {90, 12, 20, 4, 16, 3}, /* MMCI2 */ + {91, 12, 20, 4, 15, 3}, /* MMCI3 */ + {92, 12, 12, 4, 6, 3}, /* MFI (unsure, bug? in my 7760 manual*/ + /* 93-107 reserved/undocumented */ + {108,12, 4, 4, 1, 3}, /* ADC */ + {109,12, 0, 4, 0, 3}, /* CMTI */ + /* 110-111 reserved/unused */ +}; + +void __init init_IRQ_intc2(void) +{ + make_intc2_irq(intc2_irq_table, ARRAY_SIZE(intc2_irq_table)); +} diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7780.c b/arch/sh/kernel/cpu/sh4/setup-sh7780.c index 72493f2..814ddb2 100644 --- a/arch/sh/kernel/cpu/sh4/setup-sh7780.c +++ b/arch/sh/kernel/cpu/sh4/setup-sh7780.c @@ -77,3 +77,30 @@ static int __init sh7780_devices_setup(void) ARRAY_SIZE(sh7780_devices)); } __initcall(sh7780_devices_setup); + +static struct intc2_data intc2_irq_table[] = { + { TIMER_IRQ, 0, 24, 0, INTC_TMU0_MSK, 2 }, + { 21, 1, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY }, + { 22, 1, 1, 0, INTC_RTC_MSK, TIMER_PRIORITY }, + { 23, 1, 2, 0, INTC_RTC_MSK, TIMER_PRIORITY }, + { SCIF0_ERI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, + { SCIF0_RXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, + { SCIF0_BRI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, + { SCIF0_TXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY }, + + { SCIF1_ERI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, + { SCIF1_RXI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, + { SCIF1_BRI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, + { SCIF1_TXI_IRQ, 8, 16, 0, INTC_SCIF1_MSK, SCIF1_PRIORITY }, + + { PCIC0_IRQ, 0x10, 8, 0, INTC_PCIC0_MSK, PCIC0_PRIORITY }, + { PCIC1_IRQ, 0x10, 0, 0, INTC_PCIC1_MSK, PCIC1_PRIORITY }, + { PCIC2_IRQ, 0x14, 24, 0, INTC_PCIC2_MSK, PCIC2_PRIORITY }, + { PCIC3_IRQ, 0x14, 16, 0, INTC_PCIC3_MSK, PCIC3_PRIORITY }, + { PCIC4_IRQ, 0x14, 8, 0, INTC_PCIC4_MSK, PCIC4_PRIORITY }, +}; + +void __init init_IRQ_intc2(void) +{ + make_intc2_irq(intc2_irq_table, ARRAY_SIZE(intc2_irq_table)); +} diff --git a/include/asm-sh/irq-sh7780.h b/include/asm-sh/irq-sh7780.h index 895c578..19912ae 100644 --- a/include/asm-sh/irq-sh7780.h +++ b/include/asm-sh/irq-sh7780.h @@ -6,16 +6,6 @@ * * Copyright (C) 2004 Takashi SHUDO */ - -#ifdef CONFIG_IDE -# ifndef IRQ_CFCARD -# define IRQ_CFCARD 14 -# endif -# ifndef IRQ_PCMCIA -# define IRQ_PCMCIA 15 -# endif -#endif - #define INTC_BASE 0xffd00000 #define INTC_ICR0 (INTC_BASE+0x0) #define INTC_ICR1 (INTC_BASE+0x1c) diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h index 1837bdb..7596ab8 100644 --- a/include/asm-sh/irq.h +++ b/include/asm-sh/irq.h @@ -685,7 +685,7 @@ struct intc2_data { unsigned char priority; }; -void make_intc2_irq(struct intc2_data *); +void make_intc2_irq(struct intc2_data *, unsigned int nr_irqs); void init_IRQ_intc2(void); #endif diff --git a/include/asm-sh/r7780rp.h b/include/asm-sh/r7780rp.h index ddd67b6..c18f648 100644 --- a/include/asm-sh/r7780rp.h +++ b/include/asm-sh/r7780rp.h @@ -81,7 +81,6 @@ #define IRQ_PCISLOT2 66 /* PCI Slot #2 IRQ */ #define IRQ_PCISLOT3 67 /* PCI Slot #3 IRQ */ #define IRQ_PCISLOT4 68 /* PCI Slot #4 IRQ */ -#define IRQ_CFCARD 1 /* CF Card IRQ */ // #define IRQ_CFINST 0 /* CF Card Insert IRQ */ #define IRQ_TP 2 /* Touch Panel IRQ */ #define IRQ_SCI1 3 /* SCI1 IRQ */ @@ -153,7 +152,6 @@ #define IRQ_PCISLOT2 1 /* PCI Slot #2 IRQ */ #define IRQ_PCISLOT3 2 /* PCI Slot #3 IRQ */ #define IRQ_PCISLOT4 3 /* PCI Slot #4 IRQ */ -#define IRQ_CFCARD 4 /* CF Card IRQ */ #define IRQ_CFINST 5 /* CF Card Insert IRQ */ #define IRQ_M66596 6 /* M66596 IRQ */ #define IRQ_SDCARD 7 /* SD Card IRQ */ -- cgit v0.10.2 From 206daaf77f68ce0f103164e6406336068c87a4a5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 19 Oct 2006 23:58:23 -0700 Subject: [NETPOLL]: initialize skb for UDP Need to fully initialize skb to keep lower layers and queueing happy. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller diff --git a/net/core/netpoll.c b/net/core/netpoll.c index ead5920..9308af0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -335,13 +335,13 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) memcpy(skb->data, msg, len); skb->len += len; - udph = (struct udphdr *) skb_push(skb, sizeof(*udph)); + skb->h.uh = udph = (struct udphdr *) skb_push(skb, sizeof(*udph)); udph->source = htons(np->local_port); udph->dest = htons(np->remote_port); udph->len = htons(udp_len); udph->check = 0; - iph = (struct iphdr *)skb_push(skb, sizeof(*iph)); + skb->nh.iph = iph = (struct iphdr *)skb_push(skb, sizeof(*iph)); /* iph->version = 4; iph->ihl = 5; */ put_unaligned(0x45, (unsigned char *)iph); @@ -357,8 +357,8 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); eth = (struct ethhdr *) skb_push(skb, ETH_HLEN); - - eth->h_proto = htons(ETH_P_IP); + skb->mac.raw = skb->data; + skb->protocol = eth->h_proto = htons(ETH_P_IP); memcpy(eth->h_source, np->local_mac, 6); memcpy(eth->h_dest, np->remote_mac, 6); -- cgit v0.10.2 From 82fac0542e11c0d3316cc8fdafd2a990d2aab692 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B6rn=20Steinbrink?= Date: Fri, 20 Oct 2006 00:21:10 -0700 Subject: [NETFILTER]: Missing check for CAP_NET_ADMIN in iptables compat layer MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The 32bit compatibility layer has no CAP_NET_ADMIN check in compat_do_ipt_get_ctl, which for example allows to list the current iptables rules even without having that capability (the non-compat version requires it). Other capabilities might be required to exploit the bug (eg. CAP_NET_RAW to get the nfnetlink socket?), so a plain user can't exploit it, but a setup actually using the posix capability system might very well hit such a constellation of granted capabilities. Signed-off-by: Björn Steinbrink Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 78a44b0..4b90927 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1932,6 +1932,9 @@ compat_do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) { int ret; + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + switch (cmd) { case IPT_SO_GET_INFO: ret = get_info(user, len, 1); -- cgit v0.10.2 From 06ca719faddaf5ea46c6356b12847663c3ed8806 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 20 Oct 2006 00:22:25 -0700 Subject: [TCP]: One NET_INC_STATS() could be NET_INC_STATS_BH in tcp_v4_err() I believe this NET_INC_STATS() call can be replaced by NET_INC_STATS_BH(), a little bit cheaper. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 6bbd985..22ef8bd 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -373,7 +373,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) seq = ntohl(th->seq); if (sk->sk_state != TCP_LISTEN && !between(seq, tp->snd_una, tp->snd_nxt)) { - NET_INC_STATS(LINUX_MIB_OUTOFWINDOWICMPS); + NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } -- cgit v0.10.2 From 78d79423179c0efc7ec34b55d287e7be4ca07da6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 20 Oct 2006 00:28:35 -0700 Subject: [IPV4] inet_peer: Group together avl_left, avl_right, v4daddr to speedup lookups on some CPUS Lot of routers/embedded devices still use CPUS with 16/32 bytes cache lines. (486, Pentium, ... PIII) It makes sense to group together fields used at lookup time so they fit in one cache line. This reduce cache footprint and speedup lookups. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h index f13cc0c..aa10a81 100644 --- a/include/net/inetpeer.h +++ b/include/net/inetpeer.h @@ -17,14 +17,15 @@ struct inet_peer { + /* group together avl_left,avl_right,v4daddr to speedup lookups */ struct inet_peer *avl_left, *avl_right; + __be32 v4daddr; /* peer's address */ + __u16 avl_height; + __u16 ip_id_count; /* IP ID for the next packet */ struct inet_peer *unused_next, **unused_prevp; __u32 dtime; /* the time of last use of not * referenced entries */ atomic_t refcnt; - __be32 v4daddr; /* peer's address */ - __u16 avl_height; - __u16 ip_id_count; /* IP ID for the next packet */ atomic_t rid; /* Frag reception counter */ __u32 tcp_ts; unsigned long tcp_ts_stamp; -- cgit v0.10.2 From 6cf431d77c3e917399a847e3a7ec239d5163056b Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 20 Oct 2006 00:29:33 -0700 Subject: [SPARC]: Clean up asm-sparc/elf.h pollution in userspace. We don't need to export sparc_elf_hwcap() to userspace, and it doesn't build there. Remove it by moving it inside #ifdef __KERNEL__, along with some other things which don't need to be exported. Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/include/asm-sparc/elf.h b/include/asm-sparc/elf.h index 83a3dd1..aaf6ef4 100644 --- a/include/asm-sparc/elf.h +++ b/include/asm-sparc/elf.h @@ -8,11 +8,6 @@ #include -#ifdef __KERNEL__ -#include -#include -#endif - /* * Sparc section types */ @@ -77,6 +72,23 @@ typedef unsigned long elf_greg_t; #define ELF_NGREG 38 typedef elf_greg_t elf_gregset_t[ELF_NGREG]; +typedef struct { + union { + unsigned long pr_regs[32]; + double pr_dregs[16]; + } pr_fr; + unsigned long __unused; + unsigned long pr_fsr; + unsigned char pr_qcnt; + unsigned char pr_q_entrysize; + unsigned char pr_en; + unsigned int pr_q[64]; +} elf_fpregset_t; + +#ifdef __KERNEL__ +#include +#include + /* Format is: * G0 --> G7 * O0 --> O7 @@ -99,20 +111,7 @@ do { unsigned long *dest = &(__elf_regs[0]); \ dest[34] = src->npc; \ dest[35] = src->y; \ dest[36] = dest[37] = 0; /* XXX */ \ -} while(0); /* Janitors: Don't touch this colon. */ - -typedef struct { - union { - unsigned long pr_regs[32]; - double pr_dregs[16]; - } pr_fr; - unsigned long __unused; - unsigned long pr_fsr; - unsigned char pr_qcnt; - unsigned char pr_q_entrysize; - unsigned char pr_en; - unsigned int pr_q[64]; -} elf_fpregset_t; +} while(0); /* Janitors: Don't touch this semicolon. */ #define ELF_CORE_COPY_TASK_REGS(__tsk, __elf_regs) \ ({ ELF_CORE_COPY_REGS((*(__elf_regs)), (__tsk)->thread.kregs); 1; }) @@ -165,8 +164,8 @@ typedef struct { #define ELF_PLATFORM (NULL) -#ifdef __KERNEL__ #define SET_PERSONALITY(ex, ibcs2) set_personality((ibcs2)?PER_SVR4:PER_LINUX) -#endif + +#endif /* __KERNEL__ */ #endif /* !(__ASMSPARC_ELF_H) */ -- cgit v0.10.2 From 58f539740b1ccfc5ef4e509ec2efe82621b546e3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 20 Oct 2006 00:32:41 -0700 Subject: [NET]: Can use __get_cpu_var() instead of per_cpu() in loopback driver. As BHs are off in loopback_xmit(), preemption cannot occurs, so we can use __get_cpu_var() instead of per_cpu() (and avoid a preempt_enable()/preempt_disable() pair) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index 93fbea1..82c10de 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -153,14 +153,14 @@ static int loopback_xmit(struct sk_buff *skb, struct net_device *dev) #endif dev->last_rx = jiffies; - lb_stats = &per_cpu(pcpu_lstats, get_cpu()); + /* it's OK to use __get_cpu_var() because BHs are off */ + lb_stats = &__get_cpu_var(pcpu_lstats); lb_stats->bytes += skb->len; lb_stats->packets++; - put_cpu(); netif_rx(skb); - return(0); + return 0; } static struct net_device_stats loopback_stats; -- cgit v0.10.2 From 0eab934f4b9668669cffebfa8a9542fedf9082af Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Fri, 20 Oct 2006 08:55:29 +0200 Subject: [Bluetooth] Add support for newer ANYCOM USB dongles This patch adds the vendor and product id of the ANYCOM Bluetooth USB-200 and USB-250 dongles and sets a flag to send HCI_Reset as the first command. Signed-off-by: Marcel Holtmann Date: Fri, 20 Oct 2006 08:55:34 +0200 Subject: [Bluetooth] Add missing entry for Nokia DTL-4 PCMCIA card The device id for the Nokia DTL-4 PCMCIA card was missing. This patch adds it back to the list of supported devices. Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/dtl1_cs.c b/drivers/bluetooth/dtl1_cs.c index e7c800f..07eafbc 100644 --- a/drivers/bluetooth/dtl1_cs.c +++ b/drivers/bluetooth/dtl1_cs.c @@ -711,6 +711,7 @@ static void dtl1_release(struct pcmcia_device *link) static struct pcmcia_device_id dtl1_ids[] = { PCMCIA_DEVICE_PROD_ID12("Nokia Mobile Phones", "DTL-1", 0xe1bfdd64, 0xe168480d), + PCMCIA_DEVICE_PROD_ID12("Nokia Mobile Phones", "DTL-4", 0xe1bfdd64, 0x9102bc82), PCMCIA_DEVICE_PROD_ID12("Socket", "CF", 0xb38bcc2e, 0x44ebf863), PCMCIA_DEVICE_PROD_ID12("Socket", "CF+ Personal Network Card", 0xb38bcc2e, 0xe732bae3), PCMCIA_DEVICE_NULL -- cgit v0.10.2 From 7b19ffc40b0247fcfe083644fdb621fdb3c05ef6 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Fri, 20 Oct 2006 08:55:48 +0200 Subject: [Bluetooth] Fix HID disconnect NULL pointer dereference The latest HID disconnect sequence change introduced a NULL pointer dereference. For the quirk to handle buggy remote HID implementations, it is enough to wait for a potential control channel disconnect from the remote side and it is also enough to wait only 500 msecs. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 9a562cf..6678201 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -507,12 +507,10 @@ static int hidp_session(void *arg) hidp_del_timer(session); - if (intr_sk->sk_state != BT_CONNECTED) - wait_event_timeout(*(ctrl_sk->sk_sleep), (ctrl_sk->sk_state == BT_CLOSED), HZ); - fput(session->intr_sock->file); - wait_event_timeout(*(intr_sk->sk_sleep), (intr_sk->sk_state == BT_CLOSED), HZ); + wait_event_timeout(*(ctrl_sk->sk_sleep), + (ctrl_sk->sk_state == BT_CLOSED), msecs_to_jiffies(500)); fput(session->ctrl_sock->file); -- cgit v0.10.2 From 42fb00838a644d03f9a2a5fbbe0b668a5ff5df4d Mon Sep 17 00:00:00 2001 From: Patrick Caulfield Date: Fri, 13 Oct 2006 17:12:05 +0100 Subject: [DLM] fix iovec length in recvmsg I didn't spot that the msg_iovlen was set to 2 if there were two elements in the iovec but left at zero if not :( I think this might be why bob was still seeing trouble. Signed-Off-By: Patrick Caulfield Signed-off-by: Steven Whitehouse diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 867f93d..6da6b14 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -519,6 +519,7 @@ static int receive_from_sock(void) msg.msg_flags = 0; msg.msg_control = incmsg; msg.msg_controllen = sizeof(incmsg); + msg.msg_iovlen = 1; /* I don't see why this circular buffer stuff is necessary for SCTP * which is a packet-based protocol, but the whole thing breaks under -- cgit v0.10.2 From 23591256d61354e20f12e98d7a496ad5c23de74c Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Fri, 13 Oct 2006 17:25:45 -0400 Subject: [GFS2] Fix bmap to map extents properly This fix means that bmap will map extents of the length requested by the VFS rather than guessing at it, or just mapping one block at a time. The other callers of gfs2_block_map are audited to ensure they send the correct max extent lengths (i.e. set bh->b_size correctly). Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index cc57f2e..06e9a8c 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -434,8 +434,7 @@ static int lookup_block(struct gfs2_inode *ip, struct buffer_head *bh, */ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh_map, struct metapath *mp, - unsigned int maxlen) + struct buffer_head *bh_map, struct metapath *mp) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); @@ -448,6 +447,7 @@ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, int new = 0; u64 dblock = 0; int boundary; + unsigned int maxlen = bh_map->b_size >> inode->i_blkbits; BUG_ON(maxlen == 0); @@ -541,13 +541,13 @@ static inline void bmap_unlock(struct inode *inode, int create) } int gfs2_block_map(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh, unsigned int maxlen) + struct buffer_head *bh) { struct metapath mp; int ret; bmap_lock(inode, create); - ret = gfs2_block_pointers(inode, lblock, create, bh, &mp, maxlen); + ret = gfs2_block_pointers(inode, lblock, create, bh, &mp); bmap_unlock(inode, create); return ret; } @@ -555,7 +555,7 @@ int gfs2_block_map(struct inode *inode, u64 lblock, int create, int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen) { struct metapath mp; - struct buffer_head bh = { .b_state = 0, .b_blocknr = 0, .b_size = 0 }; + struct buffer_head bh = { .b_state = 0, .b_blocknr = 0 }; int ret; int create = *new; @@ -563,8 +563,9 @@ int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsi BUG_ON(!dblock); BUG_ON(!new); + bh.b_size = 1 << (inode->i_blkbits + 5); bmap_lock(inode, create); - ret = gfs2_block_pointers(inode, lblock, create, &bh, &mp, 32); + ret = gfs2_block_pointers(inode, lblock, create, &bh, &mp); bmap_unlock(inode, create); *extlen = bh.b_size >> inode->i_blkbits; *dblock = bh.b_blocknr; diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h index 0fd379b..ac2fd04 100644 --- a/fs/gfs2/bmap.h +++ b/fs/gfs2/bmap.h @@ -15,7 +15,7 @@ struct gfs2_inode; struct page; int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page); -int gfs2_block_map(struct inode *inode, u64 lblock, int create, struct buffer_head *bh, unsigned int maxlen); +int gfs2_block_map(struct inode *inode, u64 lblock, int create, struct buffer_head *bh); int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen); int gfs2_truncatei(struct gfs2_inode *ip, u64 size); diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 72eec65..0cace3d 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -312,10 +312,12 @@ void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks) static u64 log_bmap(struct gfs2_sbd *sdp, unsigned int lbn) { + struct inode *inode = sdp->sd_jdesc->jd_inode; int error; - struct buffer_head bh_map; + struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; - error = gfs2_block_map(sdp->sd_jdesc->jd_inode, lbn, 0, &bh_map, 1); + bh_map.b_size = 1 << inode->i_blkbits; + error = gfs2_block_map(inode, lbn, 0, &bh_map); if (error || !bh_map.b_blocknr) printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error, bh_map.b_blocknr, lbn); gfs2_assert_withdraw(sdp, !error && bh_map.b_blocknr); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index e0599fe..8d5963c 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -65,7 +65,7 @@ static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page, int gfs2_get_block(struct inode *inode, sector_t lblock, struct buffer_head *bh_result, int create) { - return gfs2_block_map(inode, lblock, create, bh_result, 32); + return gfs2_block_map(inode, lblock, create, bh_result); } /** @@ -83,7 +83,7 @@ static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, { int error; - error = gfs2_block_map(inode, lblock, 0, bh_result, 1); + error = gfs2_block_map(inode, lblock, 0, bh_result); if (error) return error; if (bh_result->b_blocknr == 0) @@ -94,7 +94,7 @@ static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, static int gfs2_get_block_direct(struct inode *inode, sector_t lblock, struct buffer_head *bh_result, int create) { - return gfs2_block_map(inode, lblock, 0, bh_result, 32); + return gfs2_block_map(inode, lblock, 0, bh_result); } /** diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index c69b94a..a3deae7 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -251,7 +251,7 @@ static int bh_get(struct gfs2_quota_data *qd) unsigned int block, offset; struct buffer_head *bh; int error; - struct buffer_head bh_map; + struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; mutex_lock(&sdp->sd_quota_mutex); @@ -263,7 +263,8 @@ static int bh_get(struct gfs2_quota_data *qd) block = qd->qd_slot / sdp->sd_qc_per_block; offset = qd->qd_slot % sdp->sd_qc_per_block;; - error = gfs2_block_map(&ip->i_inode, block, 0, &bh_map, 1); + bh_map.b_size = 1 << ip->i_inode.i_blkbits; + error = gfs2_block_map(&ip->i_inode, block, 0, &bh_map); if (error) goto fail; error = gfs2_meta_read(ip->i_gl, bh_map.b_blocknr, DIO_WAIT, &bh); diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index 0a8a4b8..62cd223 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -372,11 +372,12 @@ static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header *head) u32 hash; struct buffer_head *bh; int error; - struct buffer_head bh_map; + struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; lblock = head->lh_blkno; gfs2_replay_incr_blk(sdp, &lblock); - error = gfs2_block_map(&ip->i_inode, lblock, 0, &bh_map, 1); + bh_map.b_size = 1 << ip->i_inode.i_blkbits; + error = gfs2_block_map(&ip->i_inode, lblock, 0, &bh_map); if (error) return error; if (!bh_map.b_blocknr) { -- cgit v0.10.2 From a2d7d021d78dbc00d24d9c809c64a7f3e61fa773 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 14 Oct 2006 16:49:30 +0100 Subject: [GFS2] gfs2 endianness bug: be16 assigned to be32 field Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 459498c..d43caf0 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -815,7 +815,7 @@ static struct gfs2_leaf *new_leaf(struct inode *inode, struct buffer_head **pbh, leaf = (struct gfs2_leaf *)bh->b_data; leaf->lf_depth = cpu_to_be16(depth); leaf->lf_entries = 0; - leaf->lf_dirent_format = cpu_to_be16(GFS2_FORMAT_DE); + leaf->lf_dirent_format = cpu_to_be32(GFS2_FORMAT_DE); leaf->lf_next = 0; memset(leaf->lf_reserved, 0, sizeof(leaf->lf_reserved)); dent = (struct gfs2_dirent *)(leaf+1); -- cgit v0.10.2 From abbdbd2065e74411dc2c401501c2c85a82f60e06 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 19 Oct 2006 15:12:24 +0200 Subject: [GFS2] fs/gfs2/dir.c:gfs2_dir_write_data(): remove dead code The Coverity checker spotted this obviously dead code. Signed-off-by: Adrian Bunk Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index d43caf0..ce52bd9 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -212,8 +212,6 @@ static int gfs2_dir_write_data(struct gfs2_inode *ip, const char *buf, gfs2_trans_add_bh(ip->i_gl, bh, 1); memcpy(bh->b_data + o, buf, amount); brelse(bh); - if (error) - goto fail; buf += amount; copied += amount; -- cgit v0.10.2 From b0cb66955f4bf7a72b544096ceef48a829361a3c Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 19 Oct 2006 15:13:26 +0200 Subject: [GFS2] fs/gfs2/ops_fstype.c:gfs2_get_sb_meta(): remove unused variable The Coverity checker spotted this unused variable. Signed-off-by: Adrian Bunk Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 178b339..e99444d 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -854,7 +854,6 @@ static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, int error = 0; struct super_block *sb = NULL, *new; struct gfs2_sbd *sdp; - char *gfs2mnt = NULL; sb = get_gfs2_sb(dev_name); if (!sb) { @@ -892,8 +891,6 @@ static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, atomic_inc(&sdp->sd_gfs2mnt->mnt_count); return simple_set_mnt(mnt, new); error: - if (gfs2mnt) - kfree(gfs2mnt); return error; } -- cgit v0.10.2 From 348acd48f050f5ba7fa917b1421ae34443be97dd Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 19 Oct 2006 15:20:04 +0200 Subject: [GFS2] fs/gfs2/dir.c:gfs2_dir_write_data(): don't use an uninitialized variable In the "if (extlen)" case, "new" might be used uninitialized. Looking at the code, it should be initialized to 0. Spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index ce52bd9..ead7df0 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -184,7 +184,7 @@ static int gfs2_dir_write_data(struct gfs2_inode *ip, const char *buf, while (copied < size) { unsigned int amount; struct buffer_head *bh; - int new; + int new = 0; amount = size - copied; if (amount > sdp->sd_sb.sb_bsize - o) -- cgit v0.10.2 From bbbe4512735eb0f15f09ffd14876091a8e91bc69 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 19 Oct 2006 15:27:00 +0200 Subject: [GFS2] fs/gfs2/ops_fstype.c:fill_super_meta(): fix NULL dereference Don't dereference new->s_root when we do know it's NULL. Spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index e99444d..882873a 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -794,8 +794,8 @@ static int fill_super_meta(struct super_block *sb, struct super_block *new, fs_err(sdp, "can't get root dentry\n"); error = -ENOMEM; iput(inode); - } - new->s_root->d_op = &gfs2_dops; + } else + new->s_root->d_op = &gfs2_dops; return error; } -- cgit v0.10.2 From b7d8ac3e1779c30ddef0a8f38042076c5007a23d Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 19 Oct 2006 16:02:07 +0200 Subject: [GFS2] gfs2_dir_read_data(): fix uninitialized variable usage In the "if (extlen)" case, "bh" was used uninitialized. This patch changes the code to what seems to have been intended. Spotted by the Coverity checker. This patch also removes a pointless "bh = NULL" asignment (the variable is never accessed again after this point). Signed-off-by: Adrian Bunk Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index ead7df0..e24af28b1 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -315,8 +315,7 @@ static int gfs2_dir_read_data(struct gfs2_inode *ip, char *buf, u64 offset, if (!ra) extlen = 1; bh = gfs2_meta_ra(ip->i_gl, dblock, extlen); - } - if (!bh) { + } else { error = gfs2_meta_read(ip->i_gl, dblock, DIO_WAIT, &bh); if (error) goto fail; @@ -330,7 +329,6 @@ static int gfs2_dir_read_data(struct gfs2_inode *ip, char *buf, u64 offset, extlen--; memcpy(buf, bh->b_data + o, amount); brelse(bh); - bh = NULL; buf += amount; copied += amount; lblock++; -- cgit v0.10.2 From e8e82b76e0312827f5ae04b573a05b02854a447e Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Thu, 19 Oct 2006 23:28:12 -0700 Subject: [PATCH] e100: fix reboot -f with netconsole enabled When rebooting with netconsole over e100, the driver shutdown code would deadlock with netpoll. Reduce shutdown code to a bare minimum while retaining WoL and suspend functionality. Signed-off-by: Auke Kok Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 27d5d2f..a3a08a5d 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -2039,7 +2039,6 @@ static int e100_change_mtu(struct net_device *netdev, int new_mtu) return 0; } -#ifdef CONFIG_PM static int e100_asf(struct nic *nic) { /* ASF can be enabled from eeprom */ @@ -2048,7 +2047,6 @@ static int e100_asf(struct nic *nic) !(nic->eeprom[eeprom_config_asf] & eeprom_gcl) && ((nic->eeprom[eeprom_smbus_addr] & 0xFF) != 0xFE)); } -#endif static int e100_up(struct nic *nic) { @@ -2715,34 +2713,32 @@ static void __devexit e100_remove(struct pci_dev *pdev) } } +#ifdef CONFIG_PM static int e100_suspend(struct pci_dev *pdev, pm_message_t state) { struct net_device *netdev = pci_get_drvdata(pdev); struct nic *nic = netdev_priv(netdev); - if (netif_running(netdev)) - e100_down(nic); - e100_hw_reset(nic); - netif_device_detach(netdev); + netif_poll_disable(nic->netdev); + del_timer_sync(&nic->watchdog); + netif_carrier_off(nic->netdev); -#ifdef CONFIG_PM pci_save_state(pdev); - if (nic->flags & (wol_magic | e100_asf(nic))) -#else - if (nic->flags & (wol_magic)) -#endif - pci_enable_wake(pdev, pci_choose_state(pdev, state), 1); - else - /* disable PME */ - pci_enable_wake(pdev, 0, 0); + + if ((nic->flags & wol_magic) | e100_asf(nic)) { + pci_enable_wake(pdev, PCI_D3hot, 1); + pci_enable_wake(pdev, PCI_D3cold, 1); + } else { + pci_enable_wake(pdev, PCI_D3hot, 0); + pci_enable_wake(pdev, PCI_D3cold, 0); + } pci_disable_device(pdev); - pci_set_power_state(pdev, pci_choose_state(pdev, state)); + pci_set_power_state(pdev, PCI_D3hot); return 0; } -#ifdef CONFIG_PM static int e100_resume(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); @@ -2764,7 +2760,23 @@ static int e100_resume(struct pci_dev *pdev) static void e100_shutdown(struct pci_dev *pdev) { - e100_suspend(pdev, PMSG_SUSPEND); + struct net_device *netdev = pci_get_drvdata(pdev); + struct nic *nic = netdev_priv(netdev); + + netif_poll_disable(nic->netdev); + del_timer_sync(&nic->watchdog); + netif_carrier_off(nic->netdev); + + if ((nic->flags & wol_magic) | e100_asf(nic)) { + pci_enable_wake(pdev, PCI_D3hot, 1); + pci_enable_wake(pdev, PCI_D3cold, 1); + } else { + pci_enable_wake(pdev, PCI_D3hot, 0); + pci_enable_wake(pdev, PCI_D3cold, 0); + } + + pci_disable_device(pdev); + pci_set_power_state(pdev, PCI_D3hot); } /* ------------------ PCI Error Recovery infrastructure -------------- */ @@ -2848,9 +2860,9 @@ static struct pci_driver e100_driver = { .id_table = e100_id_table, .probe = e100_probe, .remove = __devexit_p(e100_remove), +#ifdef CONFIG_PM /* Power Management hooks */ .suspend = e100_suspend, -#ifdef CONFIG_PM .resume = e100_resume, #endif .shutdown = e100_shutdown, -- cgit v0.10.2 From fb5527e68d495650a7658fec9a7246bf922db212 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Thu, 19 Oct 2006 23:28:13 -0700 Subject: [PATCH] direct-io: sync and invalidate file region when falling back to buffered write When direct-io falls back to buffered write, it will just leave the dirty data floating about in pagecache, pending regular writeback. But normal direct-io semantics are that IO is synchronous, and that it leaves no pagecache behind. So change the fallback-to-buffered-write code to sync the file region and to then strip away the pagecache, just as a regular direct-io write would do. Acked-by: Jeff Moyer Cc: Zach Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/filemap.c b/mm/filemap.c index 3464b68..57faa8d 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2222,7 +2222,7 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t *ppos) { struct file *file = iocb->ki_filp; - const struct address_space * mapping = file->f_mapping; + struct address_space * mapping = file->f_mapping; size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; @@ -2275,8 +2275,11 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, /* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { - written = generic_file_direct_write(iocb, iov, - &nr_segs, pos, ppos, count, ocount); + loff_t endbyte; + ssize_t written_buffered; + + written = generic_file_direct_write(iocb, iov, &nr_segs, pos, + ppos, count, ocount); if (written < 0 || written == count) goto out; /* @@ -2285,10 +2288,46 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, */ pos += written; count -= written; - } + written_buffered = generic_file_buffered_write(iocb, iov, + nr_segs, pos, ppos, count, + written); + /* + * If generic_file_buffered_write() retuned a synchronous error + * then we want to return the number of bytes which were + * direct-written, or the error code if that was zero. Note + * that this differs from normal direct-io semantics, which + * will return -EFOO even if some bytes were written. + */ + if (written_buffered < 0) { + err = written_buffered; + goto out; + } - written = generic_file_buffered_write(iocb, iov, nr_segs, - pos, ppos, count, written); + /* + * We need to ensure that the page cache pages are written to + * disk and invalidated to preserve the expected O_DIRECT + * semantics. + */ + endbyte = pos + written_buffered - written - 1; + err = do_sync_file_range(file, pos, endbyte, + SYNC_FILE_RANGE_WAIT_BEFORE| + SYNC_FILE_RANGE_WRITE| + SYNC_FILE_RANGE_WAIT_AFTER); + if (err == 0) { + written = written_buffered; + invalidate_mapping_pages(mapping, + pos >> PAGE_CACHE_SHIFT, + endbyte >> PAGE_CACHE_SHIFT); + } else { + /* + * We don't know how much we wrote, so just return + * the number of bytes which were direct-written + */ + } + } else { + written = generic_file_buffered_write(iocb, iov, nr_segs, + pos, ppos, count, written); + } out: current->backing_dev_info = NULL; return written ? written : err; -- cgit v0.10.2 From 26da82058e62ea173559a26881b16d10089645ba Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Thu, 19 Oct 2006 23:28:14 -0700 Subject: [PATCH] ecryptfs: use special_file() Use the special_file() macro to check whether an inode is special instead of open-coding it. Acked-by: Mike Halcrow Cc: Phillip Hellewell Signed-off-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 7a11b8a..5938a23 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -104,10 +104,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, inode->i_op = &ecryptfs_dir_iops; if (S_ISDIR(lower_inode->i_mode)) inode->i_fop = &ecryptfs_dir_fops; - /* TODO: Is there a better way to identify if the inode is - * special? */ - if (S_ISBLK(lower_inode->i_mode) || S_ISCHR(lower_inode->i_mode) || - S_ISFIFO(lower_inode->i_mode) || S_ISSOCK(lower_inode->i_mode)) + if (special_file(lower_inode->i_mode)) init_special_inode(inode, lower_inode->i_mode, lower_inode->i_rdev); dentry->d_op = &ecryptfs_dops; -- cgit v0.10.2 From 79e2de4bc53d7ca2a8eedee49e4a92479b4b530e Mon Sep 17 00:00:00 2001 From: Thomas Maier Date: Thu, 19 Oct 2006 23:28:15 -0700 Subject: [PATCH] export clear_queue_congested and set_queue_congested Export the clear_queue_congested() and set_queue_congested() functions located in ll_rw_blk.c The functions are renamed to blk_clear_queue_congested() and blk_set_queue_congested(). (needed in the pktcdvd driver's bio write congestion control) Signed-off-by: Thomas Maier Cc: Peter Osterlund Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index c847e17..132a858 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -117,7 +117,7 @@ static void blk_queue_congestion_threshold(struct request_queue *q) * congested queues, and wake up anyone who was waiting for requests to be * put back. */ -static void clear_queue_congested(request_queue_t *q, int rw) +void blk_clear_queue_congested(request_queue_t *q, int rw) { enum bdi_state bit; wait_queue_head_t *wqh = &congestion_wqh[rw]; @@ -128,18 +128,20 @@ static void clear_queue_congested(request_queue_t *q, int rw) if (waitqueue_active(wqh)) wake_up(wqh); } +EXPORT_SYMBOL(blk_clear_queue_congested); /* * A queue has just entered congestion. Flag that in the queue's VM-visible * state flags and increment the global gounter of congested queues. */ -static void set_queue_congested(request_queue_t *q, int rw) +void blk_set_queue_congested(request_queue_t *q, int rw) { enum bdi_state bit; bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; set_bit(bit, &q->backing_dev_info.state); } +EXPORT_SYMBOL(blk_set_queue_congested); /** * blk_get_backing_dev_info - get the address of a queue's backing_dev_info @@ -159,7 +161,6 @@ struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev) ret = &q->backing_dev_info; return ret; } - EXPORT_SYMBOL(blk_get_backing_dev_info); void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data) @@ -167,7 +168,6 @@ void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data) q->activity_fn = fn; q->activity_data = data; } - EXPORT_SYMBOL(blk_queue_activity_fn); /** @@ -2067,7 +2067,7 @@ static void __freed_request(request_queue_t *q, int rw) struct request_list *rl = &q->rq; if (rl->count[rw] < queue_congestion_off_threshold(q)) - clear_queue_congested(q, rw); + blk_clear_queue_congested(q, rw); if (rl->count[rw] + 1 <= q->nr_requests) { if (waitqueue_active(&rl->wait[rw])) @@ -2137,7 +2137,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio, } } } - set_queue_congested(q, rw); + blk_set_queue_congested(q, rw); } /* @@ -3765,14 +3765,14 @@ queue_requests_store(struct request_queue *q, const char *page, size_t count) blk_queue_congestion_threshold(q); if (rl->count[READ] >= queue_congestion_on_threshold(q)) - set_queue_congested(q, READ); + blk_set_queue_congested(q, READ); else if (rl->count[READ] < queue_congestion_off_threshold(q)) - clear_queue_congested(q, READ); + blk_clear_queue_congested(q, READ); if (rl->count[WRITE] >= queue_congestion_on_threshold(q)) - set_queue_congested(q, WRITE); + blk_set_queue_congested(q, WRITE); else if (rl->count[WRITE] < queue_congestion_off_threshold(q)) - clear_queue_congested(q, WRITE); + blk_clear_queue_congested(q, WRITE); if (rl->count[READ] >= q->nr_requests) { blk_set_queue_full(q, READ); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d370d2c..9575e3a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -651,6 +651,8 @@ extern void blk_recount_segments(request_queue_t *, struct bio *); extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *); extern int sg_scsi_ioctl(struct file *, struct request_queue *, struct gendisk *, struct scsi_ioctl_command __user *); +extern void blk_clear_queue_congested(request_queue_t *q, int rw); +extern void blk_set_queue_congested(request_queue_t *q, int rw); extern void blk_start_queue(request_queue_t *q); extern void blk_stop_queue(request_queue_t *q); extern void blk_sync_queue(struct request_queue *q); -- cgit v0.10.2 From 3fcfab16c5b86eaa3db3a9a31adba550c5b67141 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 19 Oct 2006 23:28:16 -0700 Subject: [PATCH] separate bdi congestion functions from queue congestion functions Separate out the concept of "queue congestion" from "backing-dev congestion". Congestion is a backing-dev concept, not a queue concept. The blk_* congestion functions are retained, as wrappers around the core backing-dev congestion functions. This proper layering is needed so that NFS can cleanly use the congestion functions, and so that CONFIG_BLOCK=n actually links. Cc: "Thomas Maier" Cc: "Jens Axboe" Cc: Trond Myklebust Cc: David Howells Cc: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/lib/usercopy.c b/arch/i386/lib/usercopy.c index 258df6b..d22cfc9 100644 --- a/arch/i386/lib/usercopy.c +++ b/arch/i386/lib/usercopy.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -741,7 +742,7 @@ survive: if (retval == -ENOMEM && is_init(current)) { up_read(¤t->mm->mmap_sem); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto survive; } diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 132a858..1360665 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -56,11 +56,6 @@ static kmem_cache_t *requestq_cachep; */ static kmem_cache_t *iocontext_cachep; -static wait_queue_head_t congestion_wqh[2] = { - __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), - __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) - }; - /* * Controlling structure to kblockd */ @@ -112,37 +107,6 @@ static void blk_queue_congestion_threshold(struct request_queue *q) q->nr_congestion_off = nr; } -/* - * A queue has just exitted congestion. Note this in the global counter of - * congested queues, and wake up anyone who was waiting for requests to be - * put back. - */ -void blk_clear_queue_congested(request_queue_t *q, int rw) -{ - enum bdi_state bit; - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; - clear_bit(bit, &q->backing_dev_info.state); - smp_mb__after_clear_bit(); - if (waitqueue_active(wqh)) - wake_up(wqh); -} -EXPORT_SYMBOL(blk_clear_queue_congested); - -/* - * A queue has just entered congestion. Flag that in the queue's VM-visible - * state flags and increment the global gounter of congested queues. - */ -void blk_set_queue_congested(request_queue_t *q, int rw) -{ - enum bdi_state bit; - - bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; - set_bit(bit, &q->backing_dev_info.state); -} -EXPORT_SYMBOL(blk_set_queue_congested); - /** * blk_get_backing_dev_info - get the address of a queue's backing_dev_info * @bdev: device @@ -2755,41 +2719,6 @@ void blk_end_sync_rq(struct request *rq, int error) } EXPORT_SYMBOL(blk_end_sync_rq); -/** - * blk_congestion_wait - wait for a queue to become uncongested - * @rw: READ or WRITE - * @timeout: timeout in jiffies - * - * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion. - * If no queues are congested then just wait for the next request to be - * returned. - */ -long blk_congestion_wait(int rw, long timeout) -{ - long ret; - DEFINE_WAIT(wait); - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); - ret = io_schedule_timeout(timeout); - finish_wait(wqh, &wait); - return ret; -} - -EXPORT_SYMBOL(blk_congestion_wait); - -/** - * blk_congestion_end - wake up sleepers on a congestion queue - * @rw: READ or WRITE - */ -void blk_congestion_end(int rw) -{ - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - if (waitqueue_active(wqh)) - wake_up(wqh); -} - /* * Has to be called with the request spinlock acquired */ diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 655d816..a625576 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -602,7 +603,7 @@ static void process_write(struct crypt_io *io) /* out of memory -> run queues */ if (remaining) - blk_congestion_wait(bio_data_dir(clone), HZ/100); + congestion_wait(bio_data_dir(clone), HZ/100); } } diff --git a/fs/fat/file.c b/fs/fat/file.c index f4b8f8b..8337451 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -13,6 +13,7 @@ #include #include #include +#include #include int fat_generic_ioctl(struct inode *inode, struct file *filp, @@ -118,7 +119,7 @@ static int fat_file_release(struct inode *inode, struct file *filp) if ((filp->f_mode & FMODE_WRITE) && MSDOS_SB(inode->i_sb)->options.flush) { fat_flush_inodes(inode->i_sb, inode, NULL); - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } return 0; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f6675d2..ca92ac3 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -57,6 +57,8 @@ #include #include #include +#include + #include #include @@ -395,7 +397,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) out: clear_bit(BDI_write_congested, &bdi->state); wake_up_all(&nfs_write_congestion); - writeback_congestion_end(); + congestion_end(WRITE); return err; } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index ad8cbc4..85ce232 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -53,6 +53,7 @@ #include #include #include +#include /* gets a struct reiserfs_journal_list * from a list head */ #define JOURNAL_LIST_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \ @@ -970,7 +971,7 @@ int reiserfs_async_progress_wait(struct super_block *s) DEFINE_WAIT(wait); struct reiserfs_journal *j = SB_JOURNAL(s); if (atomic_read(&j->j_async_throttle)) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); return 0; } diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index d597375..004baf6 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "time.h" #include "kmem.h" @@ -53,7 +54,7 @@ kmem_alloc(size_t size, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } @@ -131,7 +132,7 @@ kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 9bbadaf..db5f5a3 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "xfs_linux.h" STATIC kmem_zone_t *xfs_buf_zone; @@ -395,7 +396,7 @@ _xfs_buf_lookup_pages( XFS_STATS_INC(xb_page_retries); xfsbufd_wakeup(0, gfp_mask); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto retry; } diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index f7a1390..7011d625 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -10,6 +10,8 @@ #include +struct page; + /* * Bits in backing_dev_info.state */ @@ -88,6 +90,11 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi) (1 << BDI_write_congested)); } +void clear_bdi_congested(struct backing_dev_info *bdi, int rw); +void set_bdi_congested(struct backing_dev_info *bdi, int rw); +long congestion_wait(int rw, long timeout); +void congestion_end(int rw); + #define bdi_cap_writeback_dirty(bdi) \ (!((bdi)->capabilities & BDI_CAP_NO_WRITEBACK)) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 9575e3a..7bfcde2 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -651,8 +651,26 @@ extern void blk_recount_segments(request_queue_t *, struct bio *); extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *); extern int sg_scsi_ioctl(struct file *, struct request_queue *, struct gendisk *, struct scsi_ioctl_command __user *); -extern void blk_clear_queue_congested(request_queue_t *q, int rw); -extern void blk_set_queue_congested(request_queue_t *q, int rw); + +/* + * A queue has just exitted congestion. Note this in the global counter of + * congested queues, and wake up anyone who was waiting for requests to be + * put back. + */ +static inline void blk_clear_queue_congested(request_queue_t *q, int rw) +{ + clear_bdi_congested(&q->backing_dev_info, rw); +} + +/* + * A queue has just entered congestion. Flag that in the queue's VM-visible + * state flags and increment the global gounter of congested queues. + */ +static inline void blk_set_queue_congested(request_queue_t *q, int rw) +{ + set_bdi_congested(&q->backing_dev_info, rw); +} + extern void blk_start_queue(request_queue_t *q); extern void blk_stop_queue(request_queue_t *q); extern void blk_sync_queue(struct request_queue *q); @@ -767,10 +785,8 @@ extern int blk_queue_init_tags(request_queue_t *, int, struct blk_queue_tag *); extern void blk_queue_free_tags(request_queue_t *); extern int blk_queue_resize_tags(request_queue_t *, int); extern void blk_queue_invalidate_tags(request_queue_t *); -extern long blk_congestion_wait(int rw, long timeout); extern struct blk_queue_tag *blk_init_tags(int); extern void blk_free_tags(struct blk_queue_tag *); -extern void blk_congestion_end(int rw); static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt, int tag) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index a341c80..fc35e6b 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -85,7 +85,6 @@ int wakeup_pdflush(long nr_pages); void laptop_io_completion(void); void laptop_sync_completion(void); void throttle_vm_writeout(void); -void writeback_congestion_end(void); /* These are exported to sysctl. */ extern int dirty_background_ratio; diff --git a/mm/Makefile b/mm/Makefile index 12b3a4e..f3c077e 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -10,7 +10,8 @@ mmu-$(CONFIG_MMU) := fremap.o highmem.o madvise.o memory.o mincore.o \ obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \ page_alloc.o page-writeback.o pdflush.o \ readahead.o swap.o truncate.o vmscan.o \ - prio_tree.o util.o mmzone.o vmstat.o $(mmu-y) + prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \ + $(mmu-y) ifeq ($(CONFIG_MMU)$(CONFIG_BLOCK),yy) obj-y += bounce.o diff --git a/mm/backing-dev.c b/mm/backing-dev.c new file mode 100644 index 0000000..f50a281 --- /dev/null +++ b/mm/backing-dev.c @@ -0,0 +1,69 @@ + +#include +#include +#include +#include +#include + +static wait_queue_head_t congestion_wqh[2] = { + __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), + __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) + }; + + +void clear_bdi_congested(struct backing_dev_info *bdi, int rw) +{ + enum bdi_state bit; + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; + clear_bit(bit, &bdi->state); + smp_mb__after_clear_bit(); + if (waitqueue_active(wqh)) + wake_up(wqh); +} +EXPORT_SYMBOL(clear_bdi_congested); + +void set_bdi_congested(struct backing_dev_info *bdi, int rw) +{ + enum bdi_state bit; + + bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; + set_bit(bit, &bdi->state); +} +EXPORT_SYMBOL(set_bdi_congested); + +/** + * congestion_wait - wait for a backing_dev to become uncongested + * @rw: READ or WRITE + * @timeout: timeout in jiffies + * + * Waits for up to @timeout jiffies for a backing_dev (any backing_dev) to exit + * write congestion. If no backing_devs are congested then just wait for the + * next write to be completed. + */ +long congestion_wait(int rw, long timeout) +{ + long ret; + DEFINE_WAIT(wait); + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); + ret = io_schedule_timeout(timeout); + finish_wait(wqh, &wait); + return ret; +} +EXPORT_SYMBOL(congestion_wait); + +/** + * congestion_end - wake up sleepers on a congested backing_dev_info + * @rw: READ or WRITE + */ +void congestion_end(int rw) +{ + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + if (waitqueue_active(wqh)) + wake_up(wqh); +} +EXPORT_SYMBOL(congestion_end); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a0f3390..8d9b19f 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -222,7 +222,7 @@ static void balance_dirty_pages(struct address_space *mapping) if (pages_written >= write_chunk) break; /* We've done our duty */ } - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } if (nr_reclaimable + global_page_state(NR_WRITEBACK) @@ -314,7 +314,7 @@ void throttle_vm_writeout(void) if (global_page_state(NR_UNSTABLE_NFS) + global_page_state(NR_WRITEBACK) <= dirty_thresh) break; - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } } @@ -351,7 +351,7 @@ static void background_writeout(unsigned long _min_pages) min_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write; if (wbc.nr_to_write > 0 || wbc.pages_skipped > 0) { /* Wrote less than expected */ - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); if (!wbc.encountered_congestion) break; } @@ -422,7 +422,7 @@ static void wb_kupdate(unsigned long arg) writeback_inodes(&wbc); if (wbc.nr_to_write > 0) { if (wbc.encountered_congestion) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); else break; /* All the old data is written */ } @@ -956,15 +956,6 @@ int test_set_page_writeback(struct page *page) EXPORT_SYMBOL(test_set_page_writeback); /* - * Wakes up tasks that are being throttled due to writeback congestion - */ -void writeback_congestion_end(void) -{ - blk_congestion_end(WRITE); -} -EXPORT_SYMBOL(writeback_congestion_end); - -/* * Return true if any of the pages in the mapping are marged with the * passed tag. */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 40db96a..afee38f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include @@ -1050,7 +1051,7 @@ nofail_alloc: if (page) goto got_pg; if (gfp_mask & __GFP_NOFAIL) { - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto nofail_alloc; } } @@ -1113,7 +1114,7 @@ rebalance: do_retry = 1; } if (do_retry) { - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto rebalance; } diff --git a/mm/shmem.c b/mm/shmem.c index b378f66..4959535 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -48,6 +48,7 @@ #include #include #include +#include #include #include @@ -1131,7 +1132,7 @@ repeat: page_cache_release(swappage); if (error == -ENOMEM) { /* let kswapd refresh zone for GFP_ATOMICs */ - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } goto repeat; } diff --git a/mm/vmscan.c b/mm/vmscan.c index af73c14..f05527b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1059,7 +1059,7 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) /* Take a nap, wait for some writeback to complete */ if (sc.nr_scanned && priority < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } /* top priority shrink_caches still had more to do? don't OOM, then */ if (!sc.all_unreclaimable) @@ -1214,7 +1214,7 @@ scan: * another pass across the zones. */ if (total_scanned && priority < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); /* * We do this so kswapd doesn't build up large priorities for @@ -1458,7 +1458,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages) goto out; if (sc.nr_scanned && prio < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); } lru_pages = 0; -- cgit v0.10.2 From 34e856e6a522a8fc0feba7497f5b05aeaa13d473 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 19 Oct 2006 23:28:17 -0700 Subject: [PATCH] Make userspace proof contains the constants for personality(2) but also some defintions that are useless or even harmful in userspace such as the personality() macro. Signed-off-by: Ralf Baechle Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/personality.h b/include/linux/personality.h index 80d780e..bf4cf20 100644 --- a/include/linux/personality.h +++ b/include/linux/personality.h @@ -1,6 +1,8 @@ #ifndef _LINUX_PERSONALITY_H #define _LINUX_PERSONALITY_H +#ifdef __KERNEL__ + /* * Handling of different ABIs (personalities). */ @@ -12,6 +14,8 @@ extern int register_exec_domain(struct exec_domain *); extern int unregister_exec_domain(struct exec_domain *); extern int __set_personality(unsigned long); +#endif /* __KERNEL__ */ + /* * Flags for bug emulation. * @@ -71,6 +75,7 @@ enum { PER_MASK = 0x00ff, }; +#ifdef __KERNEL__ /* * Description of an execution domain. @@ -111,4 +116,6 @@ struct exec_domain { #define set_personality(pers) \ ((current->personality == pers) ? 0 : __set_personality(pers)) +#endif /* __KERNEL__ */ + #endif /* _LINUX_PERSONALITY_H */ -- cgit v0.10.2 From 1a3e9ad163b77a3c7504f58f2780b5a62238c0ac Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Thu, 19 Oct 2006 23:28:18 -0700 Subject: [PATCH] uml: MODE_TT is bust arch/um/sys-x86_64/ptrace.c:20:1: warning: "SC_SS" redefined In file included from arch/um/include/sysdep/ptrace.h:18, from include/asm/ptrace-generic.h:12, from include/asm/ptrace.h:15, from arch/um/sys-x86_64/ptrace.c:8: arch/um/include/sysdep/sc.h:38:1: warning: this is the location of the previous definition arch/um/sys-x86_64/ptrace.c: In function 'putreg': arch/um/sys-x86_64/ptrace.c:63: warning: implicit declaration of function 'SC_FS_BASE' arch/um/sys-x86_64/ptrace.c:63: error: invalid lvalue in unary '&' arch/um/sys-x86_64/ptrace.c:63: warning: implicit declaration of function 'SC_GS_BASE' arch/um/sys-x86_64/ptrace.c:63: error: invalid lvalue in unary '&' arch/um/sys-x86_64/ptrace.c: In function 'getreg': arch/um/sys-x86_64/ptrace.c:101: error: invalid lvalue in unary '&' arch/um/sys-x86_64/ptrace.c:101: error: invalid lvalue in unary '&' I'd have to say that the fix for this, for now, is this: Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 78fb619..50fb89e 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -52,6 +52,7 @@ menu "UML-specific options" config MODE_TT bool "Tracing thread support (DEPRECATED)" default n + depends on BROKEN help This option controls whether tracing thread support is compiled into UML. This option is largely obsolete, given that skas0 provides -- cgit v0.10.2 From 3fda982c501c6a8baa3fa79aaea1bfa7bb2a5def Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:19 -0700 Subject: [PATCH] fix typo in memory barrier docs Fix cut'n'paste typo - &a and &b are used in other examples, in this one the doc uses &u and &v. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt index 994355b..7f790f6 100644 --- a/Documentation/memory-barriers.txt +++ b/Documentation/memory-barriers.txt @@ -1898,7 +1898,7 @@ queue before processing any further requests: smp_wmb(); - p = &b; q = p; + p = &v; q = p; -- cgit v0.10.2 From 02a5323d8060d7259277e9e2936fd02129dc0984 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:20 -0700 Subject: [PATCH] uml: remove some leftover PPC code I happened to notice that this code is a leftover and it should be removed - since there are sporadical efforts to revive the PPC port doing such cleanups is not useless. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-um/archparam-ppc.h b/include/asm-um/archparam-ppc.h index 172cd6f..4269d8a 100644 --- a/include/asm-um/archparam-ppc.h +++ b/include/asm-um/archparam-ppc.h @@ -1,15 +1,6 @@ #ifndef __UM_ARCHPARAM_PPC_H #define __UM_ARCHPARAM_PPC_H -/********* Bits for asm-um/hw_irq.h **********/ - -struct hw_interrupt_type; - -/********* Bits for asm-um/hardirq.h **********/ - -#define irq_enter(cpu, irq) hardirq_enter(cpu) -#define irq_exit(cpu, irq) hardirq_exit(cpu) - /********* Bits for asm-um/string.h **********/ #define __HAVE_ARCH_STRRCHR -- cgit v0.10.2 From c13e569073b89eb75216a2551e89ae93ad1f9951 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:20 -0700 Subject: [PATCH] uml: split memory allocation prototypes out of user.h user.h is too generic a header name. I've split out allocation routines from it. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/cow_sys.h b/arch/um/drivers/cow_sys.h index 7a5b4af..c6a3084 100644 --- a/arch/um/drivers/cow_sys.h +++ b/arch/um/drivers/cow_sys.h @@ -5,6 +5,7 @@ #include "user_util.h" #include "os.h" #include "user.h" +#include "um_malloc.h" static inline void *cow_malloc(int size) { diff --git a/arch/um/drivers/daemon_user.c b/arch/um/drivers/daemon_user.c index 77954ea..310af0f 100644 --- a/arch/um/drivers/daemon_user.c +++ b/arch/um/drivers/daemon_user.c @@ -17,6 +17,7 @@ #include "user_util.h" #include "user.h" #include "os.h" +#include "um_malloc.h" #define MAX_PACKET (ETH_MAX_PACKET + ETH_HEADER_OTHER) diff --git a/arch/um/drivers/fd.c b/arch/um/drivers/fd.c index 108b7da..218aa0e 100644 --- a/arch/um/drivers/fd.c +++ b/arch/um/drivers/fd.c @@ -12,6 +12,7 @@ #include "user_util.h" #include "chan_user.h" #include "os.h" +#include "um_malloc.h" struct fd_chan { int fd; diff --git a/arch/um/drivers/mcast_user.c b/arch/um/drivers/mcast_user.c index 4d2bd39..8138f5e 100644 --- a/arch/um/drivers/mcast_user.c +++ b/arch/um/drivers/mcast_user.c @@ -23,6 +23,7 @@ #include "user_util.h" #include "user.h" #include "os.h" +#include "um_malloc.h" #define MAX_PACKET (ETH_MAX_PACKET + ETH_HEADER_OTHER) diff --git a/arch/um/drivers/net_user.c b/arch/um/drivers/net_user.c index f3a3f8a..0ffd7ac 100644 --- a/arch/um/drivers/net_user.c +++ b/arch/um/drivers/net_user.c @@ -18,6 +18,7 @@ #include "kern_util.h" #include "net_user.h" #include "os.h" +#include "um_malloc.h" int tap_open_common(void *dev, char *gate_addr) { diff --git a/arch/um/drivers/pcap_user.c b/arch/um/drivers/pcap_user.c index 2ef641d..11921a7 100644 --- a/arch/um/drivers/pcap_user.c +++ b/arch/um/drivers/pcap_user.c @@ -12,6 +12,7 @@ #include "net_user.h" #include "pcap_user.h" #include "user.h" +#include "um_malloc.h" #define MAX_PACKET (ETH_MAX_PACKET + ETH_HEADER_OTHER) diff --git a/arch/um/drivers/port_user.c b/arch/um/drivers/port_user.c index f2e8fc4..bc6afaf 100644 --- a/arch/um/drivers/port_user.c +++ b/arch/um/drivers/port_user.c @@ -19,6 +19,7 @@ #include "chan_user.h" #include "port.h" #include "os.h" +#include "um_malloc.h" struct port_chan { int raw; diff --git a/arch/um/drivers/pty.c b/arch/um/drivers/pty.c index abec620..829a5ec 100644 --- a/arch/um/drivers/pty.c +++ b/arch/um/drivers/pty.c @@ -13,6 +13,7 @@ #include "user_util.h" #include "kern_util.h" #include "os.h" +#include "um_malloc.h" struct pty_chan { void (*announce)(char *dev_name, int dev); diff --git a/arch/um/drivers/slip_user.c b/arch/um/drivers/slip_user.c index 8460285..7eddacc 100644 --- a/arch/um/drivers/slip_user.c +++ b/arch/um/drivers/slip_user.c @@ -15,6 +15,7 @@ #include "slip.h" #include "slip_common.h" #include "os.h" +#include "um_malloc.h" void slip_user_init(void *data, void *dev) { diff --git a/arch/um/drivers/tty.c b/arch/um/drivers/tty.c index 11de3ac..d95d643 100644 --- a/arch/um/drivers/tty.c +++ b/arch/um/drivers/tty.c @@ -11,6 +11,7 @@ #include "user_util.h" #include "user.h" #include "os.h" +#include "um_malloc.h" struct tty_chan { char *dev; diff --git a/arch/um/include/um_malloc.h b/arch/um/include/um_malloc.h new file mode 100644 index 0000000..0363a9b --- /dev/null +++ b/arch/um/include/um_malloc.h @@ -0,0 +1,17 @@ +/* + * Copyright (C) 2005 Paolo 'Blaisorblade' Giarrusso + * Licensed under the GPL + */ + +#ifndef __UM_MALLOC_H__ +#define __UM_MALLOC_H__ + +extern void *um_kmalloc(int size); +extern void *um_kmalloc_atomic(int size); +extern void kfree(const void *ptr); + +extern void *um_vmalloc(int size); +extern void *um_vmalloc_atomic(int size); +extern void vfree(void *ptr); + +#endif /* __UM_MALLOC_H__ */ diff --git a/arch/um/include/user.h b/arch/um/include/user.h index 39f8c88..acadce3 100644 --- a/arch/um/include/user.h +++ b/arch/um/include/user.h @@ -11,17 +11,11 @@ extern void panic(const char *fmt, ...) extern int printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 2))); extern void schedule(void); -extern void *um_kmalloc(int size); -extern void *um_kmalloc_atomic(int size); -extern void kfree(void *ptr); extern int in_aton(char *str); extern int open_gdb_chan(void); /* These use size_t, however unsigned long is correct on both i386 and x86_64. */ extern unsigned long strlcpy(char *, const char *, unsigned long); extern unsigned long strlcat(char *, const char *, unsigned long); -extern void *um_vmalloc(int size); -extern void *um_vmalloc_atomic(int size); -extern void vfree(void *ptr); #endif diff --git a/arch/um/include/user_util.h b/arch/um/include/user_util.h index 802d784..06625fe 100644 --- a/arch/um/include/user_util.h +++ b/arch/um/include/user_util.h @@ -52,7 +52,6 @@ extern int linux_main(int argc, char **argv); extern void set_cmdline(char *cmd); extern void input_cb(void (*proc)(void *), void *arg, int arg_len); extern int get_pty(void); -extern void *um_kmalloc(int size); extern int switcheroo(int fd, int prot, void *from, void *to, int size); extern void do_exec(int old_pid, int new_pid); extern void tracer_panic(char *msg, ...) diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index ef25956..5c1e611 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -31,6 +31,7 @@ #include "irq_kern.h" #include "os.h" #include "sigio.h" +#include "um_malloc.h" #include "misc_constants.h" /* diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index fe6c64a..348b272 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -46,6 +46,7 @@ #include "mode.h" #include "mode_kern.h" #include "choose-mode.h" +#include "um_malloc.h" /* This is a per-cpu array. A processor only modifies its entry and it only * cares about its entry, so it's OK if another processor is modifying its diff --git a/arch/um/os-Linux/drivers/ethertap_user.c b/arch/um/os-Linux/drivers/ethertap_user.c index f559bdf7..863981b 100644 --- a/arch/um/os-Linux/drivers/ethertap_user.c +++ b/arch/um/os-Linux/drivers/ethertap_user.c @@ -20,6 +20,7 @@ #include "net_user.h" #include "etap.h" #include "os.h" +#include "um_malloc.h" #define MAX_PACKET ETH_MAX_PACKET diff --git a/arch/um/os-Linux/irq.c b/arch/um/os-Linux/irq.c index a97206d..d46b818 100644 --- a/arch/um/os-Linux/irq.c +++ b/arch/um/os-Linux/irq.c @@ -18,6 +18,7 @@ #include "sigio.h" #include "irq_user.h" #include "os.h" +#include "um_malloc.h" static struct pollfd *pollfds = NULL; static int pollfds_num = 0; diff --git a/arch/um/os-Linux/main.c b/arch/um/os-Linux/main.c index d1c5670..685feaa 100644 --- a/arch/um/os-Linux/main.c +++ b/arch/um/os-Linux/main.c @@ -23,6 +23,7 @@ #include "choose-mode.h" #include "uml-config.h" #include "os.h" +#include "um_malloc.h" /* Set in set_stklim, which is called from main and __wrap_malloc. * __wrap_malloc only calls it if main hasn't started. diff --git a/arch/um/os-Linux/sigio.c b/arch/um/os-Linux/sigio.c index f645776..925a652 100644 --- a/arch/um/os-Linux/sigio.c +++ b/arch/um/os-Linux/sigio.c @@ -19,6 +19,7 @@ #include "user_util.h" #include "sigio.h" #include "os.h" +#include "um_malloc.h" /* Protected by sigio_lock(), also used by sigio_cleanup, which is an * exitcall. -- cgit v0.10.2 From 8b028bcd0e746ae0f2f218b911032232a32dedd5 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:21 -0700 Subject: [PATCH] uml: code convention cleanup of a file Fix coding conventions violations is arch/um/os-Linux/helper.c. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/os-Linux/helper.c b/arch/um/os-Linux/helper.c index cd15b9d..8a78bf0 100644 --- a/arch/um/os-Linux/helper.c +++ b/arch/um/os-Linux/helper.c @@ -35,18 +35,18 @@ static int helper_child(void *arg) char **argv = data->argv; int errval; - if(helper_pause){ + if (helper_pause){ signal(SIGHUP, helper_hup); pause(); } - if(data->pre_exec != NULL) + if (data->pre_exec != NULL) (*data->pre_exec)(data->pre_data); execvp(argv[0], argv); errval = -errno; printk("helper_child - execve of '%s' failed - errno = %d\n", argv[0], errno); os_write_file(data->fd, &errval, sizeof(errval)); kill(os_getpid(), SIGKILL); - return(0); + return 0; } /* Returns either the pid of the child process we run or -E* on failure. @@ -58,20 +58,21 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, unsigned long stack, sp; int pid, fds[2], ret, n; - if((stack_out != NULL) && (*stack_out != 0)) + if ((stack_out != NULL) && (*stack_out != 0)) stack = *stack_out; - else stack = alloc_stack(0, __cant_sleep()); - if(stack == 0) + else + stack = alloc_stack(0, __cant_sleep()); + if (stack == 0) return -ENOMEM; ret = os_pipe(fds, 1, 0); - if(ret < 0){ + if (ret < 0) { printk("run_helper : pipe failed, ret = %d\n", -ret); goto out_free; } ret = os_set_exec_close(fds[1], 1); - if(ret < 0){ + if (ret < 0) { printk("run_helper : setting FD_CLOEXEC failed, ret = %d\n", -ret); goto out_close; @@ -83,7 +84,7 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, data.argv = argv; data.fd = fds[1]; pid = clone(helper_child, (void *) sp, CLONE_VM | SIGCHLD, &data); - if(pid < 0){ + if (pid < 0) { ret = -errno; printk("run_helper : clone failed, errno = %d\n", errno); goto out_close; @@ -95,10 +96,10 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, /* Read the errno value from the child, if the exec failed, or get 0 if * the exec succeeded because the pipe fd was set as close-on-exec. */ n = os_read_file(fds[0], &ret, sizeof(ret)); - if(n == 0) + if (n == 0) { ret = pid; - else { - if(n < 0){ + } else { + if (n < 0) { printk("run_helper : read on pipe failed, ret = %d\n", -n); ret = n; @@ -112,10 +113,11 @@ out_close: close(fds[1]); close(fds[0]); out_free: - if(stack_out == NULL) + if (stack_out == NULL) free_stack(stack, 0); - else *stack_out = stack; - return(ret); + else + *stack_out = stack; + return ret; } int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, @@ -125,31 +127,32 @@ int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, int pid, status, err; stack = alloc_stack(stack_order, __cant_sleep()); - if(stack == 0) return(-ENOMEM); + if (stack == 0) + return -ENOMEM; sp = stack + (page_size() << stack_order) - sizeof(void *); pid = clone(proc, (void *) sp, flags | SIGCHLD, arg); - if(pid < 0){ + if (pid < 0) { err = -errno; printk("run_helper_thread : clone failed, errno = %d\n", errno); return err; } - if(stack_out == NULL){ + if (stack_out == NULL) { CATCH_EINTR(pid = waitpid(pid, &status, 0)); - if(pid < 0){ + if (pid < 0) { err = -errno; printk("run_helper_thread - wait failed, errno = %d\n", errno); pid = err; } - if(!WIFEXITED(status) || (WEXITSTATUS(status) != 0)) + if (!WIFEXITED(status) || (WEXITSTATUS(status) != 0)) printk("run_helper_thread - thread returned status " "0x%x\n", status); free_stack(stack, stack_order); - } - else *stack_out = stack; - return(pid); + } else + *stack_out = stack; + return pid; } int helper_wait(int pid) @@ -157,9 +160,9 @@ int helper_wait(int pid) int ret; CATCH_EINTR(ret = waitpid(pid, NULL, WNOHANG)); - if(ret < 0){ + if (ret < 0) { ret = -errno; printk("helper_wait : waitpid failed, errno = %d\n", errno); } - return(ret); + return ret; } -- cgit v0.10.2 From 493e3758be1d5628b4d799fe21d68969edbe32aa Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:22 -0700 Subject: [PATCH] uml: reenable compilation of enable_timer, disabled by mistake CONFIG_MODE_TT does not work there, the UML_ prefixed version must be used - this causes a link-time failure when CONFIG_MODE_TT is enabled (i.e. always here, never by Jeff). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/os-Linux/time.c b/arch/um/os-Linux/time.c index 38be096..2115b8b 100644 --- a/arch/um/os-Linux/time.c +++ b/arch/um/os-Linux/time.c @@ -16,6 +16,7 @@ #include "process.h" #include "kern_constants.h" #include "os.h" +#include "uml-config.h" int set_interval(int is_virtual) { @@ -30,7 +31,7 @@ int set_interval(int is_virtual) return 0; } -#ifdef CONFIG_MODE_TT +#ifdef UML_CONFIG_MODE_TT void enable_timer(void) { set_interval(1); -- cgit v0.10.2 From b2670eacfb013169b8bf151a5078a9ef8ef86466 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:23 -0700 Subject: [PATCH] uml: use DEFCONFIG_LIST to avoid reading host's config This should make sure that, for UML, host's configuration files are not considered, which avoids various pains to the user. Our dependency are such that the obtained Kconfig will be valid and will lead to successful compilation - however they cannot prevent an user from disabling any boot device, and if an option is not set in the read .config (say /boot/config-XXX), with make menuconfig ARCH=um, it is not set. This always disables UBD and all console I/O channels, which leads to non-working UML kernels, so this bothers users - especially now, since it will happen on almost every machine (/boot/config-`uname -r` exists almost on every machine). It can be workarounded with make defconfig ARCH=um, but it is non-obvious and can be avoided, so please _do_ merge this patch. Given the existence of options, it could be interesting to implement (additionally) "option required" - with it, Kconfig will refuse reading a .config file (from wherever it comes) if the given option is not set. With this, one could mark with it the option characteristic of the given architecture (it was an old proposal of Roman Zippel, when I pointed out our problem): config UML option required default y However this should be further discussed: *) for x86, it must support constructs like: ==arch/i386/Kconfig== config 64BIT option required default n where Kconfig must require that CONFIG_64BIT is disabled or not present in the read .config. *) do we want to do such checks only for the starting defconfig or also for .config? Which leads to: *) I may want to port a x86_64 .config to x86 and viceversa, or even among more different archs. Should that be allowed, and in which measure (the user may force skipping the check for a .config or it is only given a warning by default)? Cc: Roman Zippel Cc: Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 50fb89e..5ac1f29 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -1,3 +1,8 @@ +config DEFCONFIG_LIST + string + option defconfig_list + default "arch/$ARCH/defconfig" + # UML uses the generic IRQ sugsystem config GENERIC_HARDIRQS bool diff --git a/init/Kconfig b/init/Kconfig index 1038293..c8b2624 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1,5 +1,6 @@ config DEFCONFIG_LIST string + depends on !UML option defconfig_list default "/lib/modules/$UNAME_RELEASE/.config" default "/etc/kernel-config" -- cgit v0.10.2 From d9d645f06a8f50659bbae2be64ed8367ba068fc0 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:24 -0700 Subject: [PATCH] uml: cleanup run_helper() API to fix a leak Freeing the stack is left uselessly to the caller of run_helper in some cases - this is taken from run_helper_thread, but here it is useless, so no caller needs it and the only place where this happens has a potential leak - in case of error neither run_helper() nor xterm_open() call free_stack(). At this point passing a pointer is not needed - the stack pointer should be passed directly, but this change is not done here. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/xterm.c b/arch/um/drivers/xterm.c index 386f8b9..850221d 100644 --- a/arch/um/drivers/xterm.c +++ b/arch/um/drivers/xterm.c @@ -136,8 +136,6 @@ int xterm_open(int input, int output, int primary, void *d, return(pid); } - if(data->stack == 0) free_stack(stack, 0); - if (data->direct_rcv) { new = os_rcv_fd(fd, &data->helper_pid); } else { diff --git a/arch/um/os-Linux/helper.c b/arch/um/os-Linux/helper.c index 8a78bf0..d13299c 100644 --- a/arch/um/os-Linux/helper.c +++ b/arch/um/os-Linux/helper.c @@ -50,7 +50,8 @@ static int helper_child(void *arg) } /* Returns either the pid of the child process we run or -E* on failure. - * XXX The alloc_stack here breaks if this is called in the tracing thread */ + * XXX The alloc_stack here breaks if this is called in the tracing thread, so + * we need to receive a preallocated stack (a local buffer is ok). */ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, unsigned long *stack_out) { @@ -113,10 +114,8 @@ out_close: close(fds[1]); close(fds[0]); out_free: - if (stack_out == NULL) + if ((stack_out == NULL) || (*stack_out == 0)) free_stack(stack, 0); - else - *stack_out = stack; return ret; } -- cgit v0.10.2 From f43e6a5a72566675da3581dc0d7076042e6701f7 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:25 -0700 Subject: [PATCH] uml: kconfig - silence warning Silence useless warning about undefined symbol in Kconfig. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig.char b/arch/um/Kconfig.char index 62d87b7..e03e40c 100644 --- a/arch/um/Kconfig.char +++ b/arch/um/Kconfig.char @@ -190,6 +190,11 @@ config HOSTAUDIO tristate default UML_SOUND +#It is selected elsewhere, so kconfig would warn without this. +config HW_RANDOM + tristate + default n + config UML_RANDOM tristate "Hardware random number generator" help -- cgit v0.10.2 From 1ba0ce6f9c19a8573832e14b418dc835dc6f3bbd Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 19 Oct 2006 23:28:26 -0700 Subject: [PATCH] uml: mmapper - remove just added but wrong "const" attribute When enabling the mmapper driver I got warnings because this "const" miscdevice structure is passed to function as non-const pointer; unlike struct tty_operations, however, I verified that misc_{de,}register _do_ modify their parameter, so this const attribute must be removed. Since the purpose of the change was to guarantee that no lock was needed, add a comment to prove this differently. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/mmapper_kern.c b/arch/um/drivers/mmapper_kern.c index 9a3b5da..df3516e 100644 --- a/arch/um/drivers/mmapper_kern.c +++ b/arch/um/drivers/mmapper_kern.c @@ -95,7 +95,8 @@ static const struct file_operations mmapper_fops = { .release = mmapper_release, }; -static const struct miscdevice mmapper_dev = { +/* No locking needed - only used (and modified) by below initcall and exitcall. */ +static struct miscdevice mmapper_dev = { .minor = MISC_DYNAMIC_MINOR, .name = "mmapper", .fops = &mmapper_fops -- cgit v0.10.2 From 74e8b51d254865b8abe4a94b5eb82b1940ec820c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 19 Oct 2006 23:28:26 -0700 Subject: [PATCH] Kconfig serial typos Fix typo (repeated) in serial Kconfig. Signed-off-by: Randy Dunlap Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig index b0d5026..0b71e7d 100644 --- a/drivers/serial/Kconfig +++ b/drivers/serial/Kconfig @@ -767,37 +767,37 @@ config SERIAL_CPM_SCC1 bool "Support for SCC1 serial port" depends on SERIAL_CPM=y help - Select the is option to use SCC1 as a serial port + Select this option to use SCC1 as a serial port config SERIAL_CPM_SCC2 bool "Support for SCC2 serial port" depends on SERIAL_CPM=y help - Select the is option to use SCC2 as a serial port + Select this option to use SCC2 as a serial port config SERIAL_CPM_SCC3 bool "Support for SCC3 serial port" depends on SERIAL_CPM=y help - Select the is option to use SCC3 as a serial port + Select this option to use SCC3 as a serial port config SERIAL_CPM_SCC4 bool "Support for SCC4 serial port" depends on SERIAL_CPM=y help - Select the is option to use SCC4 as a serial port + Select this option to use SCC4 as a serial port config SERIAL_CPM_SMC1 bool "Support for SMC1 serial port" depends on SERIAL_CPM=y help - Select the is option to use SMC1 as a serial port + Select this option to use SMC1 as a serial port config SERIAL_CPM_SMC2 bool "Support for SMC2 serial port" depends on SERIAL_CPM=y help - Select the is option to use SMC2 as a serial port + Select this option to use SMC2 as a serial port config SERIAL_SGI_L1_CONSOLE bool "SGI Altix L1 serial console support" -- cgit v0.10.2 From c5a114f1fb2d3c54be62779a705e088471063b47 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Thu, 19 Oct 2006 23:28:28 -0700 Subject: [PATCH] fix "ACPI: Processor native C-states using MWAIT" This patch breaks C-state discovery on my IBM IntelliStation Z30 because the return value of acpi_processor_get_power_info_fadt is not assigned to "result" in the case that acpi_processor_get_power_info_cst returns -ENODEV. Thus, if ACPI provides C-state data via the FADT and not _CST (as is the case on this machine), we incorrectly exit the function with -ENODEV after reading the FADT. The attached patch sets the value of result so that we don't exit early. Signed-off-by: Darrick J. Wong Acked-by: "Pallipadi, Venkatesh" Acked-by: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index e67144c..65b3f05 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -962,7 +962,7 @@ static int acpi_processor_get_power_info(struct acpi_processor *pr) result = acpi_processor_get_power_info_cst(pr); if (result == -ENODEV) - acpi_processor_get_power_info_fadt(pr); + result = acpi_processor_get_power_info_fadt(pr); if (result) return result; -- cgit v0.10.2 From 145fc655a1ceabda76cf2ad74f7cf96863c65b65 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 19 Oct 2006 23:28:28 -0700 Subject: [PATCH] genirq: clean up irq-flow-type naming, fix Re-add the set_irq_chip_and_handler() prototype, it's still widely used. Signed-off-by: Ingo Molnar Cc: Olaf Hering Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/irq.h b/include/linux/irq.h index 775f5a7..52fc405 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -322,6 +322,9 @@ extern struct irq_chip no_irq_chip; extern struct irq_chip dummy_irq_chip; extern void +set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle); +extern void set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, irq_flow_handler_t handle, const char *name); -- cgit v0.10.2 From 91fcdd4e0314145d7d4fa52dba2f9c2da25346fd Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Thu, 19 Oct 2006 23:28:29 -0700 Subject: [PATCH] readjust comments of task_timeslice for kernel doc Signed-off-by: Borislav Petkov Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/sched.c b/kernel/sched.c index 094b568..3399701 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -160,15 +160,6 @@ #define TASK_PREEMPTS_CURR(p, rq) \ ((p)->prio < (rq)->curr->prio) -/* - * task_timeslice() scales user-nice values [ -20 ... 0 ... 19 ] - * to time slice values: [800ms ... 100ms ... 5ms] - * - * The higher a thread's priority, the bigger timeslices - * it gets during one round of execution. But even the lowest - * priority thread gets MIN_TIMESLICE worth of execution time. - */ - #define SCALE_PRIO(x, prio) \ max(x * (MAX_PRIO - prio) / (MAX_USER_PRIO / 2), MIN_TIMESLICE) @@ -180,6 +171,15 @@ static unsigned int static_prio_timeslice(int static_prio) return SCALE_PRIO(DEF_TIMESLICE, static_prio); } +/* + * task_timeslice() scales user-nice values [ -20 ... 0 ... 19 ] + * to time slice values: [800ms ... 100ms ... 5ms] + * + * The higher a thread's priority, the bigger timeslices + * it gets during one round of execution. But even the lowest + * priority thread gets MIN_TIMESLICE worth of execution time. + */ + static inline unsigned int task_timeslice(struct task_struct *p) { return static_prio_timeslice(p->static_prio); -- cgit v0.10.2 From 6cbe44cd8d48a92856295f445183f52bf42a544d Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Thu, 19 Oct 2006 23:28:30 -0700 Subject: [PATCH] Change log level of a message of acpi_memhotplug to KERN_DEBUG I suppose this message seems quite useless except debugging. It just shows "Hotplug Mem Device". System admin can't know anything by this message. So, I would like to change it to KERN_DEBUG. Signed-off-by: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c index 98099de..9090b5a 100644 --- a/drivers/acpi/acpi_memhotplug.c +++ b/drivers/acpi/acpi_memhotplug.c @@ -414,7 +414,7 @@ static int acpi_memory_device_add(struct acpi_device *device) /* Set the device state */ mem_device->state = MEMORY_POWER_ON_STATE; - printk(KERN_INFO "%s \n", acpi_device_name(device)); + printk(KERN_DEBUG "%s \n", acpi_device_name(device)); return result; } -- cgit v0.10.2 From 887b95931b4072e60e3bf4253ff7bffe372bca46 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Thu, 19 Oct 2006 23:28:31 -0700 Subject: [PATCH] acpi memory hotplug: remove strange add_memory fail message I wrote a patch to avoid redundant memory hot-add call at boot time. This was cause of strange fail message of memory hotplug like "ACPI: add_memory failed". Memory is recognized by early boot code with EFI/E820. But, if DSDT describes memory devices for them, then hot-add code is called for already recognized memory, and it shows fail messages with -EEXIST. So, sys admin will misunderstand this message as something wrong by it. This patch avoids them by preventing redundant hot-add call until completion of driver initialization. [akpm@osdl.org: cleanups] Signed-off-by: Yasunori Goto Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c index 9090b5a..6bcd9e8 100644 --- a/drivers/acpi/acpi_memhotplug.c +++ b/drivers/acpi/acpi_memhotplug.c @@ -85,6 +85,8 @@ struct acpi_memory_device { struct list_head res_list; }; +static int acpi_hotmem_initialized; + static acpi_status acpi_memory_get_resource(struct acpi_resource *resource, void *context) { @@ -438,6 +440,15 @@ static int acpi_memory_device_start (struct acpi_device *device) struct acpi_memory_device *mem_device; int result = 0; + /* + * Early boot code has recognized memory area by EFI/E820. + * If DSDT shows these memory devices on boot, hotplug is not necessary + * for them. So, it just returns until completion of this driver's + * start up. + */ + if (!acpi_hotmem_initialized) + return 0; + mem_device = acpi_driver_data(device); if (!acpi_memory_check_device(mem_device)) { @@ -537,6 +548,7 @@ static int __init acpi_memory_device_init(void) return -ENODEV; } + acpi_hotmem_initialized = 1; return 0; } -- cgit v0.10.2 From 8ac773b4f73afa6fd66695131103944b975d5d5c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 19 Oct 2006 23:28:32 -0700 Subject: [PATCH] OOM killer meets userspace headers Despite mm.h is not being exported header, it does contain one thing which is part of userspace ABI -- value disabling OOM killer for given process. So, a) create and export include/linux/oom.h b) move OOM_DISABLE define there. c) turn bounding values of /proc/$PID/oom_adj into defines and export them too. Note: mass __KERNEL__ removal will be done later. Signed-off-by: Alexey Dobriyan Cc: Nick Piggin Cc: David Woodhouse Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/base.c b/fs/proc/base.c index 26a8f841..8df27401 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -72,6 +72,7 @@ #include #include #include +#include #include "internal.h" /* NOTE: @@ -689,7 +690,8 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, if (copy_from_user(buffer, buf, count)) return -EFAULT; oom_adjust = simple_strtol(buffer, &end, 0); - if ((oom_adjust < -16 || oom_adjust > 15) && oom_adjust != OOM_DISABLE) + if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) && + oom_adjust != OOM_DISABLE) return -EINVAL; if (*end == '\n') end++; diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 5114ff1..a1155a2 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -120,6 +120,7 @@ header-y += netrom.h header-y += nfs2.h header-y += nfs4_mount.h header-y += nfs_mount.h +header-y += oom.h header-y += param.h header-y += pci_ids.h header-y += pci_regs.h diff --git a/include/linux/mm.h b/include/linux/mm.h index 5a6068f..d538de9 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1115,9 +1115,6 @@ int in_gate_area_no_task(unsigned long addr); #define in_gate_area(task, addr) ({(void)task; in_gate_area_no_task(addr);}) #endif /* __HAVE_ARCH_GATE_AREA */ -/* /proc//oom_adj set to -17 protects from the oom-killer */ -#define OOM_DISABLE -17 - int drop_caches_sysctl_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask, diff --git a/include/linux/oom.h b/include/linux/oom.h new file mode 100644 index 0000000..ad76463 --- /dev/null +++ b/include/linux/oom.h @@ -0,0 +1,10 @@ +#ifndef __INCLUDE_LINUX_OOM_H +#define __INCLUDE_LINUX_OOM_H + +/* /proc//oom_adj set to -17 protects from the oom-killer */ +#define OOM_DISABLE (-17) +/* inclusive */ +#define OOM_ADJUST_MIN (-16) +#define OOM_ADJUST_MAX 15 + +#endif diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 20f41b0..2e3ce3a 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -15,6 +15,7 @@ * kernel subsystems and hints as to where to find out what things do. */ +#include #include #include #include -- cgit v0.10.2 From a31baca58cc16fe0584685f54c6d17494a231c92 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 19 Oct 2006 23:28:33 -0700 Subject: [PATCH] irq updates: make eata_pio compile Signed-off-by: Alan Cox Cc: James Bottomley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/scsi/eata_pio.c b/drivers/scsi/eata_pio.c index 811d884..2dbb66d 100644 --- a/drivers/scsi/eata_pio.c +++ b/drivers/scsi/eata_pio.c @@ -203,7 +203,7 @@ static irqreturn_t do_eata_pio_int_handler(int irq, void *dev_id) irqreturn_t ret; spin_lock_irqsave(dev->host_lock, flags); - ret = eata_pio_int_handler(irq, dev_id, regs); + ret = eata_pio_int_handler(irq, dev_id); spin_unlock_irqrestore(dev->host_lock, flags); return ret; } -- cgit v0.10.2 From 8c7c7c9bf39470c9689ad43cae3142cf948f4cfb Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 19 Oct 2006 23:28:34 -0700 Subject: [PATCH] Fix warnings for WARN_ON if CONFIG_BUG is disabled MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In most cases the return value of WARN_ON() is ignored. If the generic definition for the !CONFIG_BUG case is used this will result in a warning: CC kernel/sched.o In file included from include/linux/bio.h:25, from include/linux/blkdev.h:14, from kernel/sched.c:39: include/linux/ioprio.h: In function ‘task_ioprio’: include/linux/ioprio.h:50: warning: statement with no effect kernel/sched.c: In function ‘context_switch’: kernel/sched.c:1834: warning: statement with no effect Signed-off-by: Ralf Baechle Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index 1d9573c..c92ae0f 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -37,7 +37,10 @@ #endif #ifndef HAVE_ARCH_WARN_ON -#define WARN_ON(condition) unlikely((condition)) +#define WARN_ON(condition) ({ \ + typeof(condition) __ret_warn_on = (condition); \ + unlikely(__ret_warn_on); \ +}) #endif #endif -- cgit v0.10.2 From d6f8ff7381501887233666b508b9eac70143303d Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 19 Oct 2006 23:28:34 -0700 Subject: [PATCH] cad_pid sysctl with PROC_FS=n If CONFIG_PROC_FS=n: kernel/sysctl.c:148: warning: 'proc_do_cad_pid' used but never defined kernel/built-in.o:(.data+0x1228): undefined reference to `proc_do_cad_pid' make: *** [.tmp_vmlinux1] Error 1 Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 8020fb2..8bff2c1 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -136,8 +136,10 @@ static int parse_table(int __user *, int, void __user *, size_t __user *, static int proc_do_uts_string(ctl_table *table, int write, struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos); +#ifdef CONFIG_PROC_SYSCTL static int proc_do_cad_pid(ctl_table *table, int write, struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos); +#endif static ctl_table root_table[]; static struct ctl_table_header root_table_header = @@ -542,6 +544,7 @@ static ctl_table kern_table[] = { .proc_handler = &proc_dointvec, }, #endif +#ifdef CONFIG_PROC_SYSCTL { .ctl_name = KERN_CADPID, .procname = "cad_pid", @@ -550,6 +553,7 @@ static ctl_table kern_table[] = { .mode = 0600, .proc_handler = &proc_do_cad_pid, }, +#endif { .ctl_name = KERN_MAX_THREADS, .procname = "threads-max", -- cgit v0.10.2 From f2fbc6c2dad7bbcbf226c094749534f1e84d3be2 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 19 Oct 2006 23:28:35 -0700 Subject: [PATCH] fs/Kconfig: move GENERIC_ACL, fix acl() call errors GENERIC_ACL shouldn't be under Network File Systems (which made it depend on NET) as far as I can tell. Having it there and having many (FS) config symbols disabled gives this (which the patch fixes): mm/built-in.o: In function `shmem_check_acl': shmem_acl.c:(.text.shmem_check_acl+0x33): undefined reference to `posix_acl_permission' fs/built-in.o: In function `generic_acl_get': (.text.generic_acl_get+0x30): undefined reference to `posix_acl_to_xattr' fs/built-in.o: In function `generic_acl_set': (.text.generic_acl_set+0x75): undefined reference to `posix_acl_from_xattr' fs/built-in.o: In function `generic_acl_set': (.text.generic_acl_set+0x94): undefined reference to `posix_acl_valid' fs/built-in.o: In function `generic_acl_set': (.text.generic_acl_set+0xc1): undefined reference to `posix_acl_equiv_mode' fs/built-in.o: In function `generic_acl_init': (.text.generic_acl_init+0x7a): undefined reference to `posix_acl_clone' fs/built-in.o: In function `generic_acl_init': (.text.generic_acl_init+0xb4): undefined reference to `posix_acl_clone' fs/built-in.o: In function `generic_acl_init': (.text.generic_acl_init+0xc8): undefined reference to `posix_acl_create_masq' fs/built-in.o: In function `generic_acl_chmod': (.text.generic_acl_chmod+0x49): undefined reference to `posix_acl_clone' fs/built-in.o: In function `generic_acl_chmod': (.text.generic_acl_chmod+0x76): undefined reference to `posix_acl_chmod_masq' Signed-off-by: Randy Dunlap Acked-by: Andreas Gruenbacher Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/Kconfig b/fs/Kconfig index 6a3df05..fee318e 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -634,6 +634,10 @@ config FUSE_FS If you want to develop a userspace FS, or if you want to use a filesystem based on FUSE, answer Y or M. +config GENERIC_ACL + bool + select FS_POSIX_ACL + if BLOCK menu "CD-ROM/DVD Filesystems" @@ -2080,10 +2084,6 @@ config 9P_FS If unsure, say N. -config GENERIC_ACL - bool - select FS_POSIX_ACL - endmenu if BLOCK -- cgit v0.10.2 From 0e7d73824e6b0024100701da246fec769dd8f087 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 19 Oct 2006 23:28:36 -0700 Subject: [PATCH] autofs3: Make sure all dentries refs are released before calling kill_anon_super() Make sure all dentries refs are released before calling kill_anon_super() so that the assumption that generic_shutdown_super() can completely destroy the dentry tree for there will be no external references holds true. What was being done in the put_super() superblock op, is now done in the kill_sb() filesystem op instead, prior to calling kill_anon_super(). The call to shrink_dcache_sb() is removed as it is redundant since shrink_dcache_for_umount() will now be called after the cleanup routine. Signed-off-by: David Howells Acked-by: Ian Kent Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/autofs/autofs_i.h b/fs/autofs/autofs_i.h index c7700d9..906ba5c 100644 --- a/fs/autofs/autofs_i.h +++ b/fs/autofs/autofs_i.h @@ -149,6 +149,7 @@ extern const struct file_operations autofs_root_operations; /* Initializing function */ int autofs_fill_super(struct super_block *, void *, int); +void autofs_kill_sb(struct super_block *sb); /* Queue management functions */ diff --git a/fs/autofs/dirhash.c b/fs/autofs/dirhash.c index 3fded38..bf8c8af 100644 --- a/fs/autofs/dirhash.c +++ b/fs/autofs/dirhash.c @@ -246,5 +246,4 @@ void autofs_hash_nuke(struct autofs_sb_info *sbi) kfree(ent); } } - shrink_dcache_sb(sbi->sb); } diff --git a/fs/autofs/init.c b/fs/autofs/init.c index aca1237..cea5219 100644 --- a/fs/autofs/init.c +++ b/fs/autofs/init.c @@ -24,7 +24,7 @@ static struct file_system_type autofs_fs_type = { .owner = THIS_MODULE, .name = "autofs", .get_sb = autofs_get_sb, - .kill_sb = kill_anon_super, + .kill_sb = autofs_kill_sb, }; static int __init init_autofs_fs(void) diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index 2c9759ba..54c518c 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -20,7 +20,7 @@ #include "autofs_i.h" #include -static void autofs_put_super(struct super_block *sb) +void autofs_kill_sb(struct super_block *sb) { struct autofs_sb_info *sbi = autofs_sbi(sb); unsigned int n; @@ -37,13 +37,13 @@ static void autofs_put_super(struct super_block *sb) kfree(sb->s_fs_info); DPRINTK(("autofs: shutting down\n")); + kill_anon_super(sb); } static void autofs_read_inode(struct inode *inode); static struct super_operations autofs_sops = { .read_inode = autofs_read_inode, - .put_super = autofs_put_super, .statfs = simple_statfs, }; -- cgit v0.10.2 From 575b5c7870c940326a11614e0279b74356c1d44f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Oct 2006 23:28:37 -0700 Subject: [PATCH] NFSv4: Fix thinko in fs/nfs/super.c Duh. addr.sin_port should be in network byte order. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 28659a9..28108c8 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -834,7 +834,7 @@ static int nfs4_get_sb(struct file_system_type *fs_type, } /* RFC3530: The default port for NFS is 2049 */ if (addr.sin_port == 0) - addr.sin_port = NFS_PORT; + addr.sin_port = htons(NFS_PORT); /* Grab the authentication type */ authflavour = RPC_AUTH_UNIX; -- cgit v0.10.2 From b6dff26a08189932eeb0fa4261e09e733b0fc540 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Oct 2006 23:28:38 -0700 Subject: [PATCH] NFS: Fix oops in nfs_cancel_commit_list Fix two bugs: - nfs_inode_remove_request will call nfs_clear_request, so we cannot reference req->wb_page after it. Move the call to dec_zone_page_state so that it occurs while req->wb_page is still valid. - Calling nfs_clear_page_writeback is unnecessary since the radix tree tags will have been cleared by the call to nfs_inode_remove_request. Replace with a simple call to nfs_unlock_request. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/write.c b/fs/nfs/write.c index ca92ac3..883dd4a 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -590,10 +590,10 @@ static void nfs_cancel_commit_list(struct list_head *head) while(!list_empty(head)) { req = nfs_list_entry(head->next); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); nfs_list_remove_request(req); nfs_inode_remove_request(req); - dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); - nfs_clear_page_writeback(req); + nfs_unlock_request(req); } } -- cgit v0.10.2 From eda3cef8dd2b83875affe82595db9d0c278879b2 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Oct 2006 23:28:38 -0700 Subject: [PATCH] NFS: Fix error handling in nfs_direct_write_result() If the RPC call tanked, we should not be checking the return value of data->res.verf->committed, since it is unlikely to even be initialised. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 9f7f8b9..1e873fc 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -532,10 +532,12 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata) spin_lock(&dreq->lock); - if (likely(status >= 0)) - dreq->count += data->res.count; - else - dreq->error = task->tk_status; + if (unlikely(status < 0)) { + dreq->error = status; + goto out_unlock; + } + + dreq->count += data->res.count; if (data->res.verf->committed != NFS_FILE_SYNC) { switch (dreq->flags) { @@ -550,7 +552,7 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata) } } } - +out_unlock: spin_unlock(&dreq->lock); } -- cgit v0.10.2 From 7d9ac06f26fe8d477c813405f1a8c7c90eecef2d Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 19 Oct 2006 23:28:39 -0700 Subject: [PATCH] nfs4: initialize cl_ipaddr David forgot to do this. I'm not sure if this is the right place to put it.... Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 34c3996..8b123f6 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -849,6 +849,7 @@ error: */ static int nfs4_init_client(struct nfs_client *clp, int proto, int timeo, int retrans, + const char *ip_addr, rpc_authflavor_t authflavour) { int error; @@ -865,6 +866,7 @@ static int nfs4_init_client(struct nfs_client *clp, error = nfs_create_rpc_client(clp, proto, timeo, retrans, authflavour); if (error < 0) goto error; + memcpy(clp->cl_ipaddr, ip_addr, sizeof(clp->cl_ipaddr)); error = nfs_idmap_new(clp); if (error < 0) { @@ -888,6 +890,7 @@ error: */ static int nfs4_set_client(struct nfs_server *server, const char *hostname, const struct sockaddr_in *addr, + const char *ip_addr, rpc_authflavor_t authflavour, int proto, int timeo, int retrans) { @@ -902,7 +905,7 @@ static int nfs4_set_client(struct nfs_server *server, error = PTR_ERR(clp); goto error; } - error = nfs4_init_client(clp, proto, timeo, retrans, authflavour); + error = nfs4_init_client(clp, proto, timeo, retrans, ip_addr, authflavour); if (error < 0) goto error_put; @@ -971,7 +974,7 @@ struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data, return ERR_PTR(-ENOMEM); /* Get a client record */ - error = nfs4_set_client(server, hostname, addr, authflavour, + error = nfs4_set_client(server, hostname, addr, ip_addr, authflavour, data->proto, data->timeo, data->retrans); if (error < 0) goto error; @@ -1041,6 +1044,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, /* Get a client representation. * Note: NFSv4 always uses TCP, */ error = nfs4_set_client(server, data->hostname, data->addr, + parent_client->cl_ipaddr, data->authflavor, parent_server->client->cl_xprt->prot, parent_client->retrans_timeo, -- cgit v0.10.2 From 13bbc06af8a5f65df0f888b442e557c617cadba7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Oct 2006 23:28:40 -0700 Subject: [PATCH] NFS: Fix NFSv4 callback regression The change in semantics for nfs_find_client() introduced by David breaks the NFSv4 callback channel. Also, replace another completely broken BUG_ON() in nfs_find_client(). In initialised clients, clp->cl_cons_state == 0, and callers of that function should in any case never want to see clients that are uninitialised. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 8b123f6..5fea638 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -232,11 +232,15 @@ void nfs_put_client(struct nfs_client *clp) * Find a client by address * - caller must hold nfs_client_lock */ -static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion) +static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion, int match_port) { struct nfs_client *clp; list_for_each_entry(clp, &nfs_client_list, cl_share_link) { + /* Don't match clients that failed to initialise properly */ + if (clp->cl_cons_state < 0) + continue; + /* Different NFS versions cannot share the same nfs_client */ if (clp->cl_nfsversion != nfsversion) continue; @@ -245,7 +249,7 @@ static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int sizeof(clp->cl_addr.sin_addr)) != 0) continue; - if (clp->cl_addr.sin_port == addr->sin_port) + if (!match_port || clp->cl_addr.sin_port == addr->sin_port) goto found; } @@ -265,11 +269,12 @@ struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversio struct nfs_client *clp; spin_lock(&nfs_client_lock); - clp = __nfs_find_client(addr, nfsversion); + clp = __nfs_find_client(addr, nfsversion, 0); spin_unlock(&nfs_client_lock); - - BUG_ON(clp && clp->cl_cons_state == 0); - + if (clp != NULL && clp->cl_cons_state != NFS_CS_READY) { + nfs_put_client(clp); + clp = NULL; + } return clp; } @@ -292,7 +297,7 @@ static struct nfs_client *nfs_get_client(const char *hostname, do { spin_lock(&nfs_client_lock); - clp = __nfs_find_client(addr, nfsversion); + clp = __nfs_find_client(addr, nfsversion, 1); if (clp) goto found_client; if (new) -- cgit v0.10.2 From cd9ae2b6a75bb1fa0d370929c2d7a7da1ed719d9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Oct 2006 23:28:40 -0700 Subject: [PATCH] NFS: Deal with failure of invalidate_inode_pages2() If invalidate_inode_pages2() fails, then it should in principle just be because the current process was signalled. In that case, we just want to ensure that the inode's page cache remains marked as invalid. Also add a helper to allow the O_DIRECT code to simply mark the page cache as invalid once it is finished writing, instead of calling invalidate_inode_pages2() itself. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 481f889..58d4405 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -203,8 +203,10 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) * Note: assumes we have exclusive access to this mapping either * through inode->i_mutex or some other mechanism. */ - if (page->index == 0) - invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1); + if (page->index == 0 && invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1) < 0) { + /* Should never happen */ + nfs_zap_mapping(inode, inode->i_mapping); + } unlock_page(page); return 0; error: diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 1e873fc..bdfabf8 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -497,6 +497,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode if (dreq->commit_data != NULL) nfs_commit_free(dreq->commit_data); nfs_direct_free_writedata(dreq); + nfs_zap_mapping(inode, inode->i_mapping); nfs_direct_complete(dreq); } } @@ -517,6 +518,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode { nfs_end_data_update(inode); nfs_direct_free_writedata(dreq); + nfs_zap_mapping(inode, inode->i_mapping); nfs_direct_complete(dreq); } #endif @@ -830,17 +832,6 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, const struct iovec *iov, retval = nfs_direct_write(iocb, (unsigned long) buf, count, pos); - /* - * XXX: nfs_end_data_update() already ensures this file's - * cached data is subsequently invalidated. Do we really - * need to call invalidate_inode_pages2() again here? - * - * For aio writes, this invalidation will almost certainly - * occur before the writes complete. Kind of racey. - */ - if (mapping->nrpages) - invalidate_inode_pages2(mapping); - if (retval > 0) iocb->ki_pos = pos + retval; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index bc9376c..9979ad1 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -131,6 +131,15 @@ void nfs_zap_caches(struct inode *inode) spin_unlock(&inode->i_lock); } +void nfs_zap_mapping(struct inode *inode, struct address_space *mapping) +{ + if (mapping->nrpages != 0) { + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; + spin_unlock(&inode->i_lock); + } +} + static void nfs_zap_acl_cache(struct inode *inode) { void (*clear_acl_cache)(struct inode *); @@ -671,13 +680,20 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode)) ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (ret < 0) + goto out; if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { - nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); - if (S_ISREG(inode->i_mode)) - nfs_sync_mapping(mapping); - invalidate_inode_pages2(mapping); - + if (mapping->nrpages != 0) { + if (S_ISREG(inode->i_mode)) { + ret = nfs_sync_mapping(mapping); + if (ret < 0) + goto out; + } + ret = invalidate_inode_pages2(mapping); + if (ret < 0) + goto out; + } spin_lock(&inode->i_lock); nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; if (S_ISDIR(inode->i_mode)) { @@ -687,10 +703,12 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) } spin_unlock(&inode->i_lock); + nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); } +out: return ret; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 76ff548..6b2de1b 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -290,6 +290,7 @@ static inline int nfs_verify_change_attribute(struct inode *inode, unsigned long * linux/fs/nfs/inode.c */ extern int nfs_sync_mapping(struct address_space *mapping); +extern void nfs_zap_mapping(struct inode *inode, struct address_space *mapping); extern void nfs_zap_caches(struct inode *); extern struct inode *nfs_fhget(struct super_block *, struct nfs_fh *, struct nfs_fattr *); -- cgit v0.10.2 From 39cf8a1374dc51fea169190674d5e4996a7d7ea2 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Oct 2006 23:28:41 -0700 Subject: [PATCH] NFS: fix minor bug in new NFS symlink code The original code confused a zero return code from pagevec_add() as success. Test plan: None. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 58d4405..c86a1ea 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1519,8 +1519,8 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym pagevec_init(&lru_pvec, 0); if (!add_to_page_cache(page, dentry->d_inode->i_mapping, 0, GFP_KERNEL)) { - if (!pagevec_add(&lru_pvec, page)) - __pagevec_lru_add(&lru_pvec); + pagevec_add(&lru_pvec, page); + pagevec_lru_add(&lru_pvec); SetPageUptodate(page); unlock_page(page); } else -- cgit v0.10.2 From 85233a7a436a48a0b98e7976a66797e5da79c9d6 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Oct 2006 23:28:42 -0700 Subject: [PATCH] NFS: __nfs_revalidate_inode() can use "inode" before checking it is non-NULL The "!inode" check in __nfs_revalidate_inode() occurs well after the first time it is dereferenced, so get rid of it. Coverity: #cid 1372, 1373 Test plan: Code review; recheck with Coverity. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 9979ad1..08cc4c5 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -583,7 +583,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); lock_kernel(); - if (!inode || is_bad_inode(inode)) + if (is_bad_inode(inode)) goto out_nowait; if (NFS_STALE(inode)) goto out_nowait; -- cgit v0.10.2 From b87c0adfeaaf8d8310c4f790d76072a5961b3518 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Oct 2006 23:28:42 -0700 Subject: [PATCH] NFS: remove unused check in nfs4_open_revalidate Coverity spotted a superfluous error check in nfs4_open_revalidate(). Remove it. Coverity: #cid 847 Test plan: Code inspection; another pass through Coverity. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 47c7e6e..7421bcb 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1314,11 +1314,9 @@ nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, st case -EROFS: lookup_instantiate_filp(nd, (struct dentry *)state, NULL); return 1; - case -ENOENT: - if (dentry->d_inode == NULL) - return 1; + default: + goto out_drop; } - goto out_drop; } if (state->inode == dentry->d_inode) { nfs4_intent_set_file(nd, dentry, state); -- cgit v0.10.2 From 71bdcf8056f910dc57ea3d0def80a9329e7dc52d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Oct 2006 23:28:43 -0700 Subject: [PATCH] SUNRPC: fix race in in-kernel RPC portmapper client When submitting a request to a fast portmapper (such as the local rpcbind daemon), the request can complete before the parent task is even queued up on xprt->binding. Fix this by queuing before submitting the rpcbind request. Test plan: Connectathon locking test with UDP. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c index 919d5ba..e52afab 100644 --- a/net/sunrpc/pmap_clnt.c +++ b/net/sunrpc/pmap_clnt.c @@ -101,11 +101,13 @@ void rpc_getport(struct rpc_task *task) /* Autobind on cloned rpc clients is discouraged */ BUG_ON(clnt->cl_parent != clnt); - if (xprt_test_and_set_binding(xprt)) { - task->tk_status = -EACCES; /* tell caller to check again */ - rpc_sleep_on(&xprt->binding, task, NULL, NULL); - return; - } + /* Put self on queue before sending rpcbind request, in case + * pmap_getport_done completes before we return from rpc_run_task */ + rpc_sleep_on(&xprt->binding, task, NULL, NULL); + + status = -EACCES; /* tell caller to check again */ + if (xprt_test_and_set_binding(xprt)) + goto bailout_nofree; /* Someone else may have bound if we slept */ status = 0; @@ -134,8 +136,6 @@ void rpc_getport(struct rpc_task *task) goto bailout; rpc_release_task(child); - rpc_sleep_on(&xprt->binding, task, NULL, NULL); - task->tk_xprt->stat.bind_count++; return; -- cgit v0.10.2 From b7766da7f7395b74dec9e52005b7dac0d09391a4 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 19 Oct 2006 23:28:44 -0700 Subject: [PATCH] SUNRPC: fix a typo Yes, this actually passed tests the way it was. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 28100e0..757fc91 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1366,7 +1366,7 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to) if (xprt->slot == NULL) return -ENOMEM; - if (ntohs(addr->sin_port != 0)) + if (ntohs(addr->sin_port) != 0) xprt_set_bound(xprt); xprt->port = xs_get_random_port(); -- cgit v0.10.2 From cc45f0175088e000ac7493e5e3f05579b6f7d240 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:44 -0700 Subject: [PATCH] bug: nfsd/nfs4xdr.c misuse of ERR_PTR() a) ERR_PTR(nfserr_something) is a bad idea; IS_ERR() will be false for it. b) mixing nfserr_.... with -EOPNOTSUPP is even worse idea. nfsd4_path() does both; caller expects to get NFS protocol error out it if anything goes wrong, but if it does we either do not notice (see (a)) or get host-endian negative (see (b)). IOW, that's a case when we can't use ERR_PTR() to return error, even though we return a pointer in case of success. Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 41fc241..77be0c4 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1292,16 +1292,15 @@ static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, * Returned string is safe to use as long as the caller holds a reference * to @exp. */ -static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp) +static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *stat) { struct svc_fh tmp_fh; char *path, *rootpath; - int stat; fh_init(&tmp_fh, NFS4_FHSIZE); - stat = exp_pseudoroot(rqstp->rq_client, &tmp_fh, &rqstp->rq_chandle); - if (stat) - return ERR_PTR(stat); + *stat = exp_pseudoroot(rqstp->rq_client, &tmp_fh, &rqstp->rq_chandle); + if (*stat) + return NULL; rootpath = tmp_fh.fh_export->ex_path; path = exp->ex_path; @@ -1309,7 +1308,8 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp) if (strncmp(path, rootpath, strlen(rootpath))) { printk("nfsd: fs_locations failed;" "%s is not contained in %s\n", path, rootpath); - return ERR_PTR(-EOPNOTSUPP); + *stat = nfserr_notsupp; + return NULL; } return path + strlen(rootpath); @@ -1322,13 +1322,14 @@ static int nfsd4_encode_fs_locations(struct svc_rqst *rqstp, struct svc_export *exp, u32 **pp, int *buflen) { - int status, i; + u32 status; + int i; u32 *p = *pp; struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs; - char *root = nfsd4_path(rqstp, exp); + char *root = nfsd4_path(rqstp, exp, &status); - if (IS_ERR(root)) - return PTR_ERR(root); + if (status) + return status; status = nfsd4_encode_components('/', root, &p, buflen); if (status) return status; -- cgit v0.10.2 From 7111c66e4e70588c9602035a4996c9cdc2087d2d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:45 -0700 Subject: [PATCH] fix svc_procfunc declaration svc_procfunc instances return __be32, not int Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 399ad11..4e71986 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -68,7 +68,7 @@ no_locks: /* * NULL: Test for presence of service */ -static int +static __be32 nlm4svc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { dprintk("lockd: NULL called\n"); @@ -78,7 +78,7 @@ nlm4svc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * TEST: Check for conflicting lock */ -static int +static __be32 nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -107,7 +107,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -150,7 +150,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -183,7 +183,7 @@ nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNLOCK: release a lock */ -static int +static __be32 nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -217,7 +217,7 @@ nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, * GRANTED: A server calls us to tell that a process' lock request * was granted */ -static int +static __be32 nlm4svc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -253,12 +253,12 @@ static const struct rpc_call_ops nlm4svc_callback_ops = { * because we send the callback before the reply proper. I hope this * doesn't break any clients. */ -static int nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, - int (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) +static __be32 nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, + __be32 (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) { struct nlm_host *host; struct nlm_rqst *call; - int stat; + __be32 stat; host = nlmsvc_lookup_host(rqstp, argp->lock.caller, @@ -282,35 +282,35 @@ static int nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *a return rpc_success; } -static int nlm4svc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: TEST_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_TEST_RES, argp, nlm4svc_proc_test); } -static int nlm4svc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: LOCK_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_LOCK_RES, argp, nlm4svc_proc_lock); } -static int nlm4svc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: CANCEL_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_CANCEL_RES, argp, nlm4svc_proc_cancel); } -static int nlm4svc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: UNLOCK_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_UNLOCK_RES, argp, nlm4svc_proc_unlock); } -static int nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: GRANTED_MSG called\n"); @@ -320,7 +320,7 @@ static int nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *arg /* * SHARE: create a DOS share or alter existing share. */ -static int +static __be32 nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -353,7 +353,7 @@ nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNSHARE: Release a DOS share. */ -static int +static __be32 nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -386,7 +386,7 @@ nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* * NM_LOCK: Create an unmonitored lock */ -static int +static __be32 nlm4svc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -399,7 +399,7 @@ nlm4svc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* * FREE_ALL: Release all locks and shares held by client */ -static int +static __be32 nlm4svc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -417,7 +417,7 @@ nlm4svc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SM_NOTIFY: private callback from statd (not part of official NLM proto) */ -static int +static __be32 nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { @@ -446,7 +446,7 @@ nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, /* * client sent a GRANTED_RES, let's remove the associated block */ -static int +static __be32 nlm4svc_proc_granted_res(struct svc_rqst *rqstp, struct nlm_res *argp, void *resp) { diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 6a931f4..db8d85c 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -96,7 +96,7 @@ no_locks: /* * NULL: Test for presence of service */ -static int +static __be32 nlmsvc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { dprintk("lockd: NULL called\n"); @@ -106,7 +106,7 @@ nlmsvc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * TEST: Check for conflicting lock */ -static int +static __be32 nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -136,7 +136,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -179,7 +179,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -212,7 +212,7 @@ nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNLOCK: release a lock */ -static int +static __be32 nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -246,7 +246,7 @@ nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, * GRANTED: A server calls us to tell that a process' lock request * was granted */ -static int +static __be32 nlmsvc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -282,12 +282,12 @@ static const struct rpc_call_ops nlmsvc_callback_ops = { * because we send the callback before the reply proper. I hope this * doesn't break any clients. */ -static int nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, - int (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) +static __be32 nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, + __be32 (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) { struct nlm_host *host; struct nlm_rqst *call; - int stat; + __be32 stat; host = nlmsvc_lookup_host(rqstp, argp->lock.caller, @@ -311,28 +311,28 @@ static int nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *ar return rpc_success; } -static int nlmsvc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: TEST_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_TEST_RES, argp, nlmsvc_proc_test); } -static int nlmsvc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: LOCK_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_LOCK_RES, argp, nlmsvc_proc_lock); } -static int nlmsvc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: CANCEL_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_CANCEL_RES, argp, nlmsvc_proc_cancel); } -static int +static __be32 nlmsvc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -340,7 +340,7 @@ nlmsvc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, return nlmsvc_callback(rqstp, NLMPROC_UNLOCK_RES, argp, nlmsvc_proc_unlock); } -static int +static __be32 nlmsvc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -351,7 +351,7 @@ nlmsvc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SHARE: create a DOS share or alter existing share. */ -static int +static __be32 nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -384,7 +384,7 @@ nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNSHARE: Release a DOS share. */ -static int +static __be32 nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -417,7 +417,7 @@ nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* * NM_LOCK: Create an unmonitored lock */ -static int +static __be32 nlmsvc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -430,7 +430,7 @@ nlmsvc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* * FREE_ALL: Release all locks and shares held by client */ -static int +static __be32 nlmsvc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -448,7 +448,7 @@ nlmsvc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SM_NOTIFY: private callback from statd (not part of official NLM proto) */ -static int +static __be32 nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { @@ -477,7 +477,7 @@ nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, /* * client sent a GRANTED_RES, let's remove the associated block */ -static int +static __be32 nlmsvc_proc_granted_res(struct svc_rqst *rqstp, struct nlm_res *argp, void *resp) { diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 29f9321..5998d0c 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -36,7 +36,7 @@ struct callback_op { static struct callback_op callback_ops[]; -static int nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) +static __be32 nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) { return htonl(NFS4_OK); } @@ -399,7 +399,7 @@ static unsigned process_op(struct svc_rqst *rqstp, /* * Decode, process and encode a COMPOUND */ -static int nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *resp) +static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *resp) { struct cb_compound_hdr_arg hdr_arg; struct cb_compound_hdr_res hdr_res; diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 9187755..8d48616 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -21,7 +21,7 @@ /* * NULL call. */ -static int +static __be32 nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -30,7 +30,7 @@ nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get the Access and/or Default ACL of a file. */ -static int nfsacld_proc_getacl(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_getacl(struct svc_rqst * rqstp, struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) { svc_fh *fh; @@ -97,7 +97,7 @@ fail: /* * Set the Access and/or Default ACL of a file. */ -static int nfsacld_proc_setacl(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_setaclargs *argp, struct nfsd_attrstat *resp) { @@ -128,7 +128,7 @@ static int nfsacld_proc_setacl(struct svc_rqst * rqstp, /* * Check file attributes */ -static int nfsacld_proc_getattr(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_getattr(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); @@ -140,7 +140,7 @@ static int nfsacld_proc_getattr(struct svc_rqst * rqstp, /* * Check file access */ -static int nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, +static __be32 nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { int nfserr; diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index d4bdc00..ed6e2c2 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -19,7 +19,7 @@ /* * NULL call. */ -static int +static __be32 nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -28,7 +28,7 @@ nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get the Access and/or Default ACL of a file. */ -static int nfsd3_proc_getacl(struct svc_rqst * rqstp, +static __be32 nfsd3_proc_getacl(struct svc_rqst * rqstp, struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) { svc_fh *fh; @@ -93,7 +93,7 @@ fail: /* * Set the Access and/or Default ACL of a file. */ -static int nfsd3_proc_setacl(struct svc_rqst * rqstp, +static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_setaclargs *argp, struct nfsd3_attrstat *resp) { diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index a5ebc7d..a12663f 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -43,7 +43,7 @@ static int nfs3_ftypes[] = { /* * NULL call. */ -static int +static __be32 nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -52,7 +52,7 @@ nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get a file's attributes */ -static int +static __be32 nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd3_attrstat *resp) { @@ -76,7 +76,7 @@ nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, /* * Set a file's attributes */ -static int +static __be32 nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, struct nfsd3_attrstat *resp) { @@ -94,7 +94,7 @@ nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, /* * Look up a path name component */ -static int +static __be32 nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_diropres *resp) { @@ -118,7 +118,7 @@ nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, /* * Check file access */ -static int +static __be32 nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { @@ -137,7 +137,7 @@ nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, /* * Read a symlink. */ -static int +static __be32 nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, struct nfsd3_readlinkres *resp) { @@ -155,7 +155,7 @@ nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, /* * Read a portion of a file. */ -static int +static __be32 nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, struct nfsd3_readres *resp) { @@ -195,7 +195,7 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, /* * Write data to a file */ -static int +static __be32 nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, struct nfsd3_writeres *resp) { @@ -223,7 +223,7 @@ nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, * At least in theory; we'll see how it fares in practice when the * first reports about SunOS compatibility problems start to pour in... */ -static int +static __be32 nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { @@ -265,7 +265,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, /* * Make directory. This operation is not idempotent. */ -static int +static __be32 nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { @@ -285,7 +285,7 @@ nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, struct nfsd3_diropres *resp) { @@ -307,7 +307,7 @@ nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, /* * Make socket/fifo/device. */ -static int +static __be32 nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, struct nfsd3_diropres *resp) { @@ -343,7 +343,7 @@ nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, /* * Remove file/fifo/socket etc. */ -static int +static __be32 nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { @@ -363,7 +363,7 @@ nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, /* * Remove a directory */ -static int +static __be32 nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { @@ -379,7 +379,7 @@ nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, struct nfsd3_renameres *resp) { @@ -401,7 +401,7 @@ nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, struct nfsd3_linkres *resp) { @@ -424,7 +424,7 @@ nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, /* * Read a portion of a directory. */ -static int +static __be32 nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { @@ -459,7 +459,7 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, * Read a portion of a directory, including file handles and attrs. * For now, we choose to ignore the dircount parameter. */ -static int +static __be32 nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { @@ -517,7 +517,7 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, /* * Get file system stats */ -static int +static __be32 nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsstatres *resp) { @@ -534,7 +534,7 @@ nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Get file system info */ -static int +static __be32 nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsinfores *resp) { @@ -576,7 +576,7 @@ nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Get pathconf info for the specified file */ -static int +static __be32 nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_pathconfres *resp) { @@ -619,7 +619,7 @@ nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Commit a file (range) to stable storage. */ -static int +static __be32 nfsd3_proc_commit(struct svc_rqst * rqstp, struct nfsd3_commitargs *argp, struct nfsd3_commitres *resp) { diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index d1fac68..795ad6c5 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -715,7 +715,7 @@ out_kfree: /* * NULL call. */ -static int +static __be32 nfsd4_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -731,7 +731,7 @@ static inline void nfsd4_increment_op_stats(u32 opnum) /* * COMPOUND call. */ -static int +static __be32 nfsd4_proc_compound(struct svc_rqst *rqstp, struct nfsd4_compoundargs *args, struct nfsd4_compoundres *resp) diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 9ee1dab..09030af 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -30,7 +30,7 @@ typedef struct svc_buf svc_buf; #define NFSDDBG_FACILITY NFSDDBG_PROC -static int +static __be32 nfsd_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -56,7 +56,7 @@ nfsd_return_dirop(int err, struct nfsd_diropres *resp) * Get a file's attributes * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { @@ -72,7 +72,7 @@ nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, * Set a file's attributes * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, struct nfsd_attrstat *resp) { @@ -92,7 +92,7 @@ nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, * doesn't exist yet. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, struct nfsd_diropres *resp) { @@ -112,7 +112,7 @@ nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, /* * Read a symlink. */ -static int +static __be32 nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, struct nfsd_readlinkres *resp) { @@ -132,7 +132,7 @@ nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, * Read a portion of a file. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, struct nfsd_readres *resp) { @@ -172,7 +172,7 @@ nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, * Write data to a file * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, struct nfsd_attrstat *resp) { @@ -197,7 +197,7 @@ nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, * and the actual create() call in compliance with VFS protocols. * N.B. After this call _both_ argp->fh and resp->fh need an fh_put */ -static int +static __be32 nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { @@ -348,7 +348,7 @@ done: return nfsd_return_dirop(nfserr, resp); } -static int +static __be32 nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { @@ -363,7 +363,7 @@ nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, void *resp) { @@ -381,7 +381,7 @@ nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, void *resp) { @@ -401,7 +401,7 @@ nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, void *resp) { @@ -430,7 +430,7 @@ nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, * Make directory. This operation is not idempotent. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { @@ -454,7 +454,7 @@ nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, /* * Remove a directory */ -static int +static __be32 nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { @@ -470,7 +470,7 @@ nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, /* * Read a portion of a directory. */ -static int +static __be32 nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, struct nfsd_readdirres *resp) { @@ -509,7 +509,7 @@ nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, /* * Get file system info */ -static int +static __be32 nfsd_proc_statfs(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_statfsres *resp) { diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 9c9a8ad..965d6c2 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -335,7 +335,7 @@ struct svc_version { /* * RPC procedure info */ -typedef int (*svc_procfunc)(struct svc_rqst *, void *argp, void *resp); +typedef __be32 (*svc_procfunc)(struct svc_rqst *, void *argp, void *resp); struct svc_procedure { svc_procfunc pc_func; /* process the request */ kxdrproc_t pc_decode; /* XDR decode args */ -- cgit v0.10.2 From 52921e02a4f4163a7b1f4b5dde71e1debc71de4a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:46 -0700 Subject: [PATCH] lockd endianness annotations Signed-off-by: Alexey Dobriyan Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index e8c7765..b85a0ad 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -100,12 +100,12 @@ int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout) /* * The server lockd has called us back to tell us the lock was granted */ -u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) +__be32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) { const struct file_lock *fl = &lock->fl; const struct nfs_fh *fh = &lock->fh; struct nlm_wait *block; - u32 res = nlm_lck_denied; + __be32 res = nlm_lck_denied; /* * Look up blocked request based on arguments. diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index e0179f8..eb243ed 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -148,8 +148,8 @@ nsm_create(void) * XDR functions for NSM. */ -static u32 * -xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +static __be32 * +xdr_encode_common(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { char buffer[20], *name; @@ -176,7 +176,7 @@ xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +xdr_encode_mon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { p = xdr_encode_common(rqstp, p, argp); if (IS_ERR(p)) @@ -192,7 +192,7 @@ xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +xdr_encode_unmon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { p = xdr_encode_common(rqstp, p, argp); if (IS_ERR(p)) @@ -202,7 +202,7 @@ xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +xdr_decode_stat_res(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) { resp->status = ntohl(*p++); resp->state = ntohl(*p++); @@ -212,7 +212,7 @@ xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) } static int -xdr_decode_stat(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +xdr_decode_stat(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) { resp->state = ntohl(*p++); return 0; diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 4e71986..0ce5c81 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -24,14 +24,14 @@ /* * Obtain client and file from arguments */ -static u32 +static __be32 nlm4svc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host **hostp, struct nlm_file **filp) { struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error = 0; + __be32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index 814c606..7e219b9 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -334,13 +334,13 @@ static void nlmsvc_freegrantargs(struct nlm_rqst *call) * Attempt to establish a lock, and if it can't be granted, block it * if required. */ -u32 +__be32 nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, struct nlm_lock *lock, int wait, struct nlm_cookie *cookie) { struct nlm_block *block, *newblock = NULL; int error; - u32 ret; + __be32 ret; dprintk("lockd: nlmsvc_lock(%s/%ld, ty=%d, pi=%d, %Ld-%Ld, bl=%d)\n", file->f_file->f_dentry->d_inode->i_sb->s_id, @@ -415,7 +415,7 @@ out: /* * Test for presence of a conflicting lock. */ -u32 +__be32 nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, struct nlm_lock *conflock) { @@ -448,7 +448,7 @@ nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, * afterwards. In this case the block will still be there, and hence * must be removed. */ -u32 +__be32 nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) { int error; @@ -476,7 +476,7 @@ nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) * be in progress. * The calling procedure must check whether the file can be closed. */ -u32 +__be32 nlmsvc_cancel_blocked(struct nlm_file *file, struct nlm_lock *lock) { struct nlm_block *block; diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index db8d85c..32e99a6 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -22,8 +22,8 @@ #define NLMDBG_FACILITY NLMDBG_CLIENT #ifdef CONFIG_LOCKD_V4 -static u32 -cast_to_nlm(u32 status, u32 vers) +static __be32 +cast_to_nlm(__be32 status, u32 vers) { /* Note: status is assumed to be in network byte order !!! */ if (vers != 4){ @@ -52,14 +52,14 @@ cast_to_nlm(u32 status, u32 vers) /* * Obtain client and file from arguments */ -static u32 +static __be32 nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host **hostp, struct nlm_file **filp) { struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error = 0; + __be32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) diff --git a/fs/lockd/svcshare.c b/fs/lockd/svcshare.c index b9926ce..6220dc2a 100644 --- a/fs/lockd/svcshare.c +++ b/fs/lockd/svcshare.c @@ -23,7 +23,7 @@ nlm_cmp_owner(struct nlm_share *share, struct xdr_netobj *oh) && !memcmp(share->s_owner.data, oh->data, oh->len); } -u32 +__be32 nlmsvc_share_file(struct nlm_host *host, struct nlm_file *file, struct nlm_args *argp) { @@ -64,7 +64,7 @@ update: /* * Delete a share. */ -u32 +__be32 nlmsvc_unshare_file(struct nlm_host *host, struct nlm_file *file, struct nlm_args *argp) { diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 7dac96e..e83024e 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -78,14 +78,14 @@ static inline unsigned int file_hash(struct nfs_fh *f) * This is not quite right, but for now, we assume the client performs * the proper R/W checking. */ -u32 +__be32 nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, struct nfs_fh *f) { struct hlist_node *pos; struct nlm_file *file; unsigned int hash; - u32 nfserr; + __be32 nfserr; nlm_debug_print_fh("nlm_file_lookup", f); diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c index 61c46fa..b7c9492 100644 --- a/fs/lockd/xdr.c +++ b/fs/lockd/xdr.c @@ -43,7 +43,7 @@ loff_t_to_s32(loff_t offset) /* * XDR functions for basic NLM types */ -static u32 *nlm_decode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 *nlm_decode_cookie(__be32 *p, struct nlm_cookie *c) { unsigned int len; @@ -69,8 +69,8 @@ static u32 *nlm_decode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static inline u32 * -nlm_encode_cookie(u32 *p, struct nlm_cookie *c) +static inline __be32 * +nlm_encode_cookie(__be32 *p, struct nlm_cookie *c) { *p++ = htonl(c->len); memcpy(p, c->data, c->len); @@ -78,8 +78,8 @@ nlm_encode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm_decode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm_decode_fh(__be32 *p, struct nfs_fh *f) { unsigned int len; @@ -95,8 +95,8 @@ nlm_decode_fh(u32 *p, struct nfs_fh *f) return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32 * -nlm_encode_fh(u32 *p, struct nfs_fh *f) +static inline __be32 * +nlm_encode_fh(__be32 *p, struct nfs_fh *f) { *p++ = htonl(NFS2_FHSIZE); memcpy(p, f->data, NFS2_FHSIZE); @@ -106,20 +106,20 @@ nlm_encode_fh(u32 *p, struct nfs_fh *f) /* * Encode and decode owner handle */ -static inline u32 * -nlm_decode_oh(u32 *p, struct xdr_netobj *oh) +static inline __be32 * +nlm_decode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_decode_netobj(p, oh); } -static inline u32 * -nlm_encode_oh(u32 *p, struct xdr_netobj *oh) +static inline __be32 * +nlm_encode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_encode_netobj(p, oh); } -static u32 * -nlm_decode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm_decode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; s32 start, len, end; @@ -153,8 +153,8 @@ nlm_decode_lock(u32 *p, struct nlm_lock *lock) /* * Encode a lock as part of an NLM call */ -static u32 * -nlm_encode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm_encode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s32 start, len; @@ -184,8 +184,8 @@ nlm_encode_lock(u32 *p, struct nlm_lock *lock) /* * Encode result of a TEST/TEST_MSG call */ -static u32 * -nlm_encode_testres(u32 *p, struct nlm_res *resp) +static __be32 * +nlm_encode_testres(__be32 *p, struct nlm_res *resp) { s32 start, len; @@ -221,7 +221,7 @@ nlm_encode_testres(u32 *p, struct nlm_res *resp) * First, the server side XDR functions */ int -nlmsvc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_testargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -238,7 +238,7 @@ nlmsvc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_testres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_testres(p, resp))) return 0; @@ -246,7 +246,7 @@ nlmsvc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_lockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -266,7 +266,7 @@ nlmsvc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_cancargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -282,7 +282,7 @@ nlmsvc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { if (!(p = nlm_decode_cookie(p, &argp->cookie)) || !(p = nlm_decode_lock(p, &argp->lock))) @@ -292,7 +292,7 @@ nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_shareargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -313,7 +313,7 @@ nlmsvc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_shareres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return 0; @@ -323,7 +323,7 @@ nlmsvc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return 0; @@ -332,7 +332,7 @@ nlmsvc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) +nlmsvc_decode_notify(struct svc_rqst *rqstp, __be32 *p, struct nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -344,7 +344,7 @@ nlmsvc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) } int -nlmsvc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) +nlmsvc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) { if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; @@ -357,7 +357,7 @@ nlmsvc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) } int -nlmsvc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_decode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return 0; @@ -366,13 +366,13 @@ nlmsvc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlmsvc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nlmsvc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlmsvc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -389,7 +389,7 @@ nlmclt_decode_void(struct rpc_rqst *req, u32 *p, void *ptr) #endif static int -nlmclt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_testargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -403,7 +403,7 @@ nlmclt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_decode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return -EIO; @@ -438,7 +438,7 @@ nlmclt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) static int -nlmclt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_lockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -455,7 +455,7 @@ nlmclt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_cancargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -470,7 +470,7 @@ nlmclt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_unlockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -483,7 +483,7 @@ nlmclt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_encode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return -EIO; @@ -493,7 +493,7 @@ nlmclt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlmclt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_encode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_testres(p, resp))) return -EIO; @@ -502,7 +502,7 @@ nlmclt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlmclt_decode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_decode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return -EIO; diff --git a/fs/lockd/xdr4.c b/fs/lockd/xdr4.c index 36eb175..f4c0b2b 100644 --- a/fs/lockd/xdr4.c +++ b/fs/lockd/xdr4.c @@ -44,8 +44,8 @@ loff_t_to_s64(loff_t offset) /* * XDR functions for basic NLM types */ -static u32 * -nlm4_decode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 * +nlm4_decode_cookie(__be32 *p, struct nlm_cookie *c) { unsigned int len; @@ -71,8 +71,8 @@ nlm4_decode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm4_encode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 * +nlm4_encode_cookie(__be32 *p, struct nlm_cookie *c) { *p++ = htonl(c->len); memcpy(p, c->data, c->len); @@ -80,8 +80,8 @@ nlm4_encode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm4_decode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm4_decode_fh(__be32 *p, struct nfs_fh *f) { memset(f->data, 0, sizeof(f->data)); f->size = ntohl(*p++); @@ -95,8 +95,8 @@ nlm4_decode_fh(u32 *p, struct nfs_fh *f) return p + XDR_QUADLEN(f->size); } -static u32 * -nlm4_encode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm4_encode_fh(__be32 *p, struct nfs_fh *f) { *p++ = htonl(f->size); if (f->size) p[XDR_QUADLEN(f->size)-1] = 0; /* don't leak anything */ @@ -107,20 +107,20 @@ nlm4_encode_fh(u32 *p, struct nfs_fh *f) /* * Encode and decode owner handle */ -static u32 * -nlm4_decode_oh(u32 *p, struct xdr_netobj *oh) +static __be32 * +nlm4_decode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_decode_netobj(p, oh); } -static u32 * -nlm4_encode_oh(u32 *p, struct xdr_netobj *oh) +static __be32 * +nlm4_encode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_encode_netobj(p, oh); } -static u32 * -nlm4_decode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm4_decode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s64 len, start, end; @@ -153,8 +153,8 @@ nlm4_decode_lock(u32 *p, struct nlm_lock *lock) /* * Encode a lock as part of an NLM call */ -static u32 * -nlm4_encode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm4_encode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s64 start, len; @@ -185,8 +185,8 @@ nlm4_encode_lock(u32 *p, struct nlm_lock *lock) /* * Encode result of a TEST/TEST_MSG call */ -static u32 * -nlm4_encode_testres(u32 *p, struct nlm_res *resp) +static __be32 * +nlm4_encode_testres(__be32 *p, struct nlm_res *resp) { s64 start, len; @@ -227,7 +227,7 @@ nlm4_encode_testres(u32 *p, struct nlm_res *resp) * First, the server side XDR functions */ int -nlm4svc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_testargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -244,7 +244,7 @@ nlm4svc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_testres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_testres(p, resp))) return 0; @@ -252,7 +252,7 @@ nlm4svc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_lockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -272,7 +272,7 @@ nlm4svc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_cancargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -288,7 +288,7 @@ nlm4svc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { if (!(p = nlm4_decode_cookie(p, &argp->cookie)) || !(p = nlm4_decode_lock(p, &argp->lock))) @@ -298,7 +298,7 @@ nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_shareargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -319,7 +319,7 @@ nlm4svc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_shareres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return 0; @@ -329,7 +329,7 @@ nlm4svc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return 0; @@ -338,7 +338,7 @@ nlm4svc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) +nlm4svc_decode_notify(struct svc_rqst *rqstp, __be32 *p, struct nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -350,7 +350,7 @@ nlm4svc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) } int -nlm4svc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) +nlm4svc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) { if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; @@ -363,7 +363,7 @@ nlm4svc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) } int -nlm4svc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_decode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return 0; @@ -372,13 +372,13 @@ nlm4svc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlm4svc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nlm4svc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlm4svc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -388,14 +388,14 @@ nlm4svc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) */ #ifdef NLMCLNT_SUPPORT_SHARES static int -nlm4clt_decode_void(struct rpc_rqst *req, u32 *p, void *ptr) +nlm4clt_decode_void(struct rpc_rqst *req, __be32 *p, void *ptr) { return 0; } #endif static int -nlm4clt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_testargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -409,7 +409,7 @@ nlm4clt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_decode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return -EIO; @@ -444,7 +444,7 @@ nlm4clt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) static int -nlm4clt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_lockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -461,7 +461,7 @@ nlm4clt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_cancargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -476,7 +476,7 @@ nlm4clt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_unlockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -489,7 +489,7 @@ nlm4clt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_encode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return -EIO; @@ -499,7 +499,7 @@ nlm4clt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlm4clt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_encode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_testres(p, resp))) return -EIO; @@ -508,7 +508,7 @@ nlm4clt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlm4clt_decode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_decode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return -EIO; diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 2909619..862d973 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -154,7 +154,7 @@ int nlm_async_reply(struct nlm_rqst *, u32, const struct rpc_call_ops *); struct nlm_wait * nlmclnt_prepare_block(struct nlm_host *host, struct file_lock *fl); void nlmclnt_finish_block(struct nlm_wait *block); int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout); -u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *); +__be32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *); void nlmclnt_recovery(struct nlm_host *); int nlmclnt_reclaim(struct nlm_host *, struct file_lock *); void nlmclnt_next_cookie(struct nlm_cookie *); @@ -184,12 +184,12 @@ typedef int (*nlm_host_match_fn_t)(struct nlm_host *cur, struct nlm_host *ref) /* * Server-side lock handling */ -u32 nlmsvc_lock(struct svc_rqst *, struct nlm_file *, +__be32 nlmsvc_lock(struct svc_rqst *, struct nlm_file *, struct nlm_lock *, int, struct nlm_cookie *); -u32 nlmsvc_unlock(struct nlm_file *, struct nlm_lock *); -u32 nlmsvc_testlock(struct nlm_file *, struct nlm_lock *, +__be32 nlmsvc_unlock(struct nlm_file *, struct nlm_lock *); +__be32 nlmsvc_testlock(struct nlm_file *, struct nlm_lock *, struct nlm_lock *); -u32 nlmsvc_cancel_blocked(struct nlm_file *, struct nlm_lock *); +__be32 nlmsvc_cancel_blocked(struct nlm_file *, struct nlm_lock *); unsigned long nlmsvc_retry_blocked(void); void nlmsvc_traverse_blocks(struct nlm_host *, struct nlm_file *, nlm_host_match_fn_t match); @@ -198,7 +198,7 @@ void nlmsvc_grant_reply(struct nlm_cookie *, u32); /* * File handling for the server personality */ -u32 nlm_lookup_file(struct svc_rqst *, struct nlm_file **, +__be32 nlm_lookup_file(struct svc_rqst *, struct nlm_file **, struct nfs_fh *); void nlm_release_file(struct nlm_file *); void nlmsvc_mark_resources(void); diff --git a/include/linux/lockd/share.h b/include/linux/lockd/share.h index cd7816e..630c5bf 100644 --- a/include/linux/lockd/share.h +++ b/include/linux/lockd/share.h @@ -21,9 +21,9 @@ struct nlm_share { u32 s_mode; /* deny mode */ }; -u32 nlmsvc_share_file(struct nlm_host *, struct nlm_file *, +__be32 nlmsvc_share_file(struct nlm_host *, struct nlm_file *, struct nlm_args *); -u32 nlmsvc_unshare_file(struct nlm_host *, struct nlm_file *, +__be32 nlmsvc_unshare_file(struct nlm_host *, struct nlm_file *, struct nlm_args *); void nlmsvc_traverse_shares(struct nlm_host *, struct nlm_file *, nlm_host_match_fn_t); diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h index 66fdae3..29e7d9f 100644 --- a/include/linux/lockd/xdr.h +++ b/include/linux/lockd/xdr.h @@ -90,19 +90,19 @@ struct nlm_reboot { */ #define NLMSVC_XDRSIZE sizeof(struct nlm_args) -int nlmsvc_decode_testargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_encode_testres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_decode_lockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_decode_cancargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_decode_unlockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_encode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_decode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_encode_void(struct svc_rqst *, u32 *, void *); -int nlmsvc_decode_void(struct svc_rqst *, u32 *, void *); -int nlmsvc_decode_shareargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_encode_shareres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_decode_notify(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_decode_reboot(struct svc_rqst *, u32 *, struct nlm_reboot *); +int nlmsvc_decode_testargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_encode_testres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_decode_lockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_decode_cancargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_decode_unlockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_encode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_decode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_encode_void(struct svc_rqst *, __be32 *, void *); +int nlmsvc_decode_void(struct svc_rqst *, __be32 *, void *); +int nlmsvc_decode_shareargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_encode_shareres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_decode_notify(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_decode_reboot(struct svc_rqst *, __be32 *, struct nlm_reboot *); /* int nlmclt_encode_testargs(struct rpc_rqst *, u32 *, struct nlm_args *); int nlmclt_encode_lockargs(struct rpc_rqst *, u32 *, struct nlm_args *); diff --git a/include/linux/lockd/xdr4.h b/include/linux/lockd/xdr4.h index 3cc1ae2..dd12b4c 100644 --- a/include/linux/lockd/xdr4.h +++ b/include/linux/lockd/xdr4.h @@ -23,19 +23,19 @@ -int nlm4svc_decode_testargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_encode_testres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_decode_lockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_decode_cancargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_decode_unlockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_encode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_decode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_encode_void(struct svc_rqst *, u32 *, void *); -int nlm4svc_decode_void(struct svc_rqst *, u32 *, void *); -int nlm4svc_decode_shareargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_encode_shareres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_decode_notify(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_decode_reboot(struct svc_rqst *, u32 *, struct nlm_reboot *); +int nlm4svc_decode_testargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_encode_testres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_decode_lockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_decode_cancargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_decode_unlockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_encode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_decode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_encode_void(struct svc_rqst *, __be32 *, void *); +int nlm4svc_decode_void(struct svc_rqst *, __be32 *, void *); +int nlm4svc_decode_shareargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_encode_shareres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_decode_notify(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_decode_reboot(struct svc_rqst *, __be32 *, struct nlm_reboot *); /* int nlmclt_encode_testargs(struct rpc_rqst *, u32 *, struct nlm_args *); int nlmclt_encode_lockargs(struct rpc_rqst *, u32 *, struct nlm_args *); -- cgit v0.10.2 From 9d787a75a00679c3ebcb88236a7af7b38a0b5932 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:47 -0700 Subject: [PATCH] xdr annotations: NFSv2 on-the-wire data is big-endian [in large part pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index b49501f..1d801e3 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -66,15 +66,15 @@ /* * Common NFS XDR functions as inlines */ -static inline u32 * -xdr_encode_fhandle(u32 *p, struct nfs_fh *fhandle) +static inline __be32 * +xdr_encode_fhandle(__be32 *p, struct nfs_fh *fhandle) { memcpy(p, fhandle->data, NFS2_FHSIZE); return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32 * -xdr_decode_fhandle(u32 *p, struct nfs_fh *fhandle) +static inline __be32 * +xdr_decode_fhandle(__be32 *p, struct nfs_fh *fhandle) { /* NFSv2 handles have a fixed length */ fhandle->size = NFS2_FHSIZE; @@ -82,8 +82,8 @@ xdr_decode_fhandle(u32 *p, struct nfs_fh *fhandle) return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32* -xdr_encode_time(u32 *p, struct timespec *timep) +static inline __be32* +xdr_encode_time(__be32 *p, struct timespec *timep) { *p++ = htonl(timep->tv_sec); /* Convert nanoseconds into microseconds */ @@ -91,8 +91,8 @@ xdr_encode_time(u32 *p, struct timespec *timep) return p; } -static inline u32* -xdr_encode_current_server_time(u32 *p, struct timespec *timep) +static inline __be32* +xdr_encode_current_server_time(__be32 *p, struct timespec *timep) { /* * Passing the invalid value useconds=1000000 is a @@ -108,8 +108,8 @@ xdr_encode_current_server_time(u32 *p, struct timespec *timep) return p; } -static inline u32* -xdr_decode_time(u32 *p, struct timespec *timep) +static inline __be32* +xdr_decode_time(__be32 *p, struct timespec *timep) { timep->tv_sec = ntohl(*p++); /* Convert microseconds into nanoseconds */ @@ -117,8 +117,8 @@ xdr_decode_time(u32 *p, struct timespec *timep) return p; } -static u32 * -xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) +static __be32 * +xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) { u32 rdev; fattr->type = (enum nfs_ftype) ntohl(*p++); @@ -146,10 +146,10 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) return p; } -static inline u32 * -xdr_encode_sattr(u32 *p, struct iattr *attr) +static inline __be32 * +xdr_encode_sattr(__be32 *p, struct iattr *attr) { - const u32 not_set = __constant_htonl(0xFFFFFFFF); + const __be32 not_set = __constant_htonl(0xFFFFFFFF); *p++ = (attr->ia_valid & ATTR_MODE) ? htonl(attr->ia_mode) : not_set; *p++ = (attr->ia_valid & ATTR_UID) ? htonl(attr->ia_uid) : not_set; @@ -184,7 +184,7 @@ xdr_encode_sattr(u32 *p, struct iattr *attr) * GETATTR, READLINK, STATFS */ static int -nfs_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) +nfs_xdr_fhandle(struct rpc_rqst *req, __be32 *p, struct nfs_fh *fh) { p = xdr_encode_fhandle(p, fh); req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); @@ -195,7 +195,7 @@ nfs_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) * Encode SETATTR arguments */ static int -nfs_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs_sattrargs *args) +nfs_xdr_sattrargs(struct rpc_rqst *req, __be32 *p, struct nfs_sattrargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_sattr(p, args->sattr); @@ -208,7 +208,7 @@ nfs_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs_sattrargs *args) * LOOKUP, REMOVE, RMDIR */ static int -nfs_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs_diropargs *args) +nfs_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs_diropargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -222,7 +222,7 @@ nfs_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs_diropargs *args) * exactly to the page we want to fetch. */ static int -nfs_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) +nfs_xdr_readargs(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -246,7 +246,7 @@ nfs_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) * Decode READ reply */ static int -nfs_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) +nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; int status, count, recvd, hdrlen; @@ -286,7 +286,7 @@ nfs_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) * Write arguments. Splice the buffer to be written into the iovec. */ static int -nfs_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) +nfs_xdr_writeargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_buf *sndbuf = &req->rq_snd_buf; u32 offset = (u32)args->offset; @@ -309,7 +309,7 @@ nfs_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) * CREATE, MKDIR */ static int -nfs_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs_createargs *args) +nfs_xdr_createargs(struct rpc_rqst *req, __be32 *p, struct nfs_createargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -322,7 +322,7 @@ nfs_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs_createargs *args) * Encode RENAME arguments */ static int -nfs_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs_renameargs *args) +nfs_xdr_renameargs(struct rpc_rqst *req, __be32 *p, struct nfs_renameargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); @@ -336,7 +336,7 @@ nfs_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs_renameargs *args) * Encode LINK arguments */ static int -nfs_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs_linkargs *args) +nfs_xdr_linkargs(struct rpc_rqst *req, __be32 *p, struct nfs_linkargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_fhandle(p, args->tofh); @@ -349,7 +349,7 @@ nfs_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs_linkargs *args) * Encode SYMLINK arguments */ static int -nfs_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_symlinkargs *args) +nfs_xdr_symlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs_symlinkargs *args) { struct xdr_buf *sndbuf = &req->rq_snd_buf; size_t pad; @@ -378,7 +378,7 @@ nfs_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_symlinkargs *args) * Encode arguments to readdir call */ static int -nfs_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs_readdirargs *args) +nfs_xdr_readdirargs(struct rpc_rqst *req, __be32 *p, struct nfs_readdirargs *args) { struct rpc_task *task = req->rq_task; struct rpc_auth *auth = task->tk_auth; @@ -404,7 +404,7 @@ nfs_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs_readdirargs *args) * from nfs_readdir for each entry. */ static int -nfs_xdr_readdirres(struct rpc_rqst *req, u32 *p, void *dummy) +nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -412,7 +412,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, u32 *p, void *dummy) int hdrlen, recvd; int status, nr; unsigned int len, pglen; - u32 *end, *entry, *kaddr; + __be32 *end, *entry, *kaddr; if ((status = ntohl(*p++))) return -nfs_stat_to_errno(status); @@ -432,8 +432,8 @@ nfs_xdr_readdirres(struct rpc_rqst *req, u32 *p, void *dummy) if (pglen > recvd) pglen = recvd; page = rcvbuf->pages; - kaddr = p = (u32 *)kmap_atomic(*page, KM_USER0); - end = (u32 *)((char *)p + pglen); + kaddr = p = kmap_atomic(*page, KM_USER0); + end = (__be32 *)((char *)p + pglen); entry = p; for (nr = 0; *p++; nr++) { if (p + 2 > end) @@ -496,7 +496,7 @@ nfs_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) * Decode simple status reply */ static int -nfs_xdr_stat(struct rpc_rqst *req, u32 *p, void *dummy) +nfs_xdr_stat(struct rpc_rqst *req, __be32 *p, void *dummy) { int status; @@ -510,7 +510,7 @@ nfs_xdr_stat(struct rpc_rqst *req, u32 *p, void *dummy) * GETATTR, SETATTR, WRITE */ static int -nfs_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs_xdr_attrstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status; @@ -525,7 +525,7 @@ nfs_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * LOOKUP, CREATE, MKDIR */ static int -nfs_xdr_diropres(struct rpc_rqst *req, u32 *p, struct nfs_diropok *res) +nfs_xdr_diropres(struct rpc_rqst *req, __be32 *p, struct nfs_diropok *res) { int status; @@ -540,7 +540,7 @@ nfs_xdr_diropres(struct rpc_rqst *req, u32 *p, struct nfs_diropok *res) * Encode READLINK args */ static int -nfs_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_readlinkargs *args) +nfs_xdr_readlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs_readlinkargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -558,7 +558,7 @@ nfs_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_readlinkargs *args * Decode READLINK reply */ static int -nfs_xdr_readlinkres(struct rpc_rqst *req, u32 *p, void *dummy) +nfs_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, void *dummy) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -601,7 +601,7 @@ nfs_xdr_readlinkres(struct rpc_rqst *req, u32 *p, void *dummy) * Decode WRITE reply */ static int -nfs_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) +nfs_xdr_writeres(struct rpc_rqst *req, __be32 *p, struct nfs_writeres *res) { res->verf->committed = NFS_FILE_SYNC; return nfs_xdr_attrstat(req, p, res->fattr); @@ -611,7 +611,7 @@ nfs_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) * Decode STATFS reply */ static int -nfs_xdr_statfsres(struct rpc_rqst *req, u32 *p, struct nfs2_fsstat *res) +nfs_xdr_statfsres(struct rpc_rqst *req, __be32 *p, struct nfs2_fsstat *res) { int status; -- cgit v0.10.2 From d61005a6855160091dca44b718db93fe7aa9876f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:48 -0700 Subject: [PATCH] xdr annotations: NFSv3 on-the-wire data is big-endian [in large part pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 16556fa..b4e740e 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -105,14 +105,14 @@ static struct { /* * Common NFS XDR functions as inlines */ -static inline u32 * -xdr_encode_fhandle(u32 *p, struct nfs_fh *fh) +static inline __be32 * +xdr_encode_fhandle(__be32 *p, struct nfs_fh *fh) { return xdr_encode_array(p, fh->data, fh->size); } -static inline u32 * -xdr_decode_fhandle(u32 *p, struct nfs_fh *fh) +static inline __be32 * +xdr_decode_fhandle(__be32 *p, struct nfs_fh *fh) { if ((fh->size = ntohl(*p++)) <= NFS3_FHSIZE) { memcpy(fh->data, p, fh->size); @@ -124,24 +124,24 @@ xdr_decode_fhandle(u32 *p, struct nfs_fh *fh) /* * Encode/decode time. */ -static inline u32 * -xdr_encode_time3(u32 *p, struct timespec *timep) +static inline __be32 * +xdr_encode_time3(__be32 *p, struct timespec *timep) { *p++ = htonl(timep->tv_sec); *p++ = htonl(timep->tv_nsec); return p; } -static inline u32 * -xdr_decode_time3(u32 *p, struct timespec *timep) +static inline __be32 * +xdr_decode_time3(__be32 *p, struct timespec *timep) { timep->tv_sec = ntohl(*p++); timep->tv_nsec = ntohl(*p++); return p; } -static u32 * -xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) +static __be32 * +xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) { unsigned int type, major, minor; int fmode; @@ -177,8 +177,8 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) return p; } -static inline u32 * -xdr_encode_sattr(u32 *p, struct iattr *attr) +static inline __be32 * +xdr_encode_sattr(__be32 *p, struct iattr *attr) { if (attr->ia_valid & ATTR_MODE) { *p++ = xdr_one; @@ -223,8 +223,8 @@ xdr_encode_sattr(u32 *p, struct iattr *attr) return p; } -static inline u32 * -xdr_decode_wcc_attr(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_wcc_attr(__be32 *p, struct nfs_fattr *fattr) { p = xdr_decode_hyper(p, &fattr->pre_size); p = xdr_decode_time3(p, &fattr->pre_mtime); @@ -233,16 +233,16 @@ xdr_decode_wcc_attr(u32 *p, struct nfs_fattr *fattr) return p; } -static inline u32 * -xdr_decode_post_op_attr(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_post_op_attr(__be32 *p, struct nfs_fattr *fattr) { if (*p++) p = xdr_decode_fattr(p, fattr); return p; } -static inline u32 * -xdr_decode_pre_op_attr(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_pre_op_attr(__be32 *p, struct nfs_fattr *fattr) { if (*p++) return xdr_decode_wcc_attr(p, fattr); @@ -250,8 +250,8 @@ xdr_decode_pre_op_attr(u32 *p, struct nfs_fattr *fattr) } -static inline u32 * -xdr_decode_wcc_data(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_wcc_data(__be32 *p, struct nfs_fattr *fattr) { p = xdr_decode_pre_op_attr(p, fattr); return xdr_decode_post_op_attr(p, fattr); @@ -265,7 +265,7 @@ xdr_decode_wcc_data(u32 *p, struct nfs_fattr *fattr) * Encode file handle argument */ static int -nfs3_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) +nfs3_xdr_fhandle(struct rpc_rqst *req, __be32 *p, struct nfs_fh *fh) { p = xdr_encode_fhandle(p, fh); req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); @@ -276,7 +276,7 @@ nfs3_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) * Encode SETATTR arguments */ static int -nfs3_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs3_sattrargs *args) +nfs3_xdr_sattrargs(struct rpc_rqst *req, __be32 *p, struct nfs3_sattrargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_sattr(p, args->sattr); @@ -291,7 +291,7 @@ nfs3_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs3_sattrargs *args) * Encode directory ops argument */ static int -nfs3_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs3_diropargs *args) +nfs3_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs3_diropargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -303,7 +303,7 @@ nfs3_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs3_diropargs *args) * Encode access() argument */ static int -nfs3_xdr_accessargs(struct rpc_rqst *req, u32 *p, struct nfs3_accessargs *args) +nfs3_xdr_accessargs(struct rpc_rqst *req, __be32 *p, struct nfs3_accessargs *args) { p = xdr_encode_fhandle(p, args->fh); *p++ = htonl(args->access); @@ -317,7 +317,7 @@ nfs3_xdr_accessargs(struct rpc_rqst *req, u32 *p, struct nfs3_accessargs *args) * exactly to the page we want to fetch. */ static int -nfs3_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) +nfs3_xdr_readargs(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -339,7 +339,7 @@ nfs3_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) * Write arguments. Splice the buffer to be written into the iovec. */ static int -nfs3_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) +nfs3_xdr_writeargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_buf *sndbuf = &req->rq_snd_buf; u32 count = args->count; @@ -360,7 +360,7 @@ nfs3_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) * Encode CREATE arguments */ static int -nfs3_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs3_createargs *args) +nfs3_xdr_createargs(struct rpc_rqst *req, __be32 *p, struct nfs3_createargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -380,7 +380,7 @@ nfs3_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs3_createargs *args) * Encode MKDIR arguments */ static int -nfs3_xdr_mkdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_mkdirargs *args) +nfs3_xdr_mkdirargs(struct rpc_rqst *req, __be32 *p, struct nfs3_mkdirargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -393,7 +393,7 @@ nfs3_xdr_mkdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_mkdirargs *args) * Encode SYMLINK arguments */ static int -nfs3_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_symlinkargs *args) +nfs3_xdr_symlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs3_symlinkargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); @@ -410,7 +410,7 @@ nfs3_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_symlinkargs *args * Encode MKNOD arguments */ static int -nfs3_xdr_mknodargs(struct rpc_rqst *req, u32 *p, struct nfs3_mknodargs *args) +nfs3_xdr_mknodargs(struct rpc_rqst *req, __be32 *p, struct nfs3_mknodargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -429,7 +429,7 @@ nfs3_xdr_mknodargs(struct rpc_rqst *req, u32 *p, struct nfs3_mknodargs *args) * Encode RENAME arguments */ static int -nfs3_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs3_renameargs *args) +nfs3_xdr_renameargs(struct rpc_rqst *req, __be32 *p, struct nfs3_renameargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); @@ -443,7 +443,7 @@ nfs3_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs3_renameargs *args) * Encode LINK arguments */ static int -nfs3_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs3_linkargs *args) +nfs3_xdr_linkargs(struct rpc_rqst *req, __be32 *p, struct nfs3_linkargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_fhandle(p, args->tofh); @@ -456,7 +456,7 @@ nfs3_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs3_linkargs *args) * Encode arguments to readdir call */ static int -nfs3_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_readdirargs *args) +nfs3_xdr_readdirargs(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -485,7 +485,7 @@ nfs3_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_readdirargs *args * We just check for syntactical correctness. */ static int -nfs3_xdr_readdirres(struct rpc_rqst *req, u32 *p, struct nfs3_readdirres *res) +nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -493,7 +493,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, u32 *p, struct nfs3_readdirres *res) int hdrlen, recvd; int status, nr; unsigned int len, pglen; - u32 *entry, *end, *kaddr; + __be32 *entry, *end, *kaddr; status = ntohl(*p++); /* Decode post_op_attrs */ @@ -523,8 +523,8 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, u32 *p, struct nfs3_readdirres *res) if (pglen > recvd) pglen = recvd; page = rcvbuf->pages; - kaddr = p = (u32 *)kmap_atomic(*page, KM_USER0); - end = (u32 *)((char *)p + pglen); + kaddr = p = kmap_atomic(*page, KM_USER0); + end = (__be32 *)((char *)p + pglen); entry = p; for (nr = 0; *p++; nr++) { if (p + 3 > end) @@ -626,7 +626,7 @@ nfs3_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) * Encode COMMIT arguments */ static int -nfs3_xdr_commitargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) +nfs3_xdr_commitargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_hyper(p, args->offset); @@ -640,7 +640,7 @@ nfs3_xdr_commitargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) * Encode GETACL arguments */ static int -nfs3_xdr_getaclargs(struct rpc_rqst *req, u32 *p, +nfs3_xdr_getaclargs(struct rpc_rqst *req, __be32 *p, struct nfs3_getaclargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; @@ -664,7 +664,7 @@ nfs3_xdr_getaclargs(struct rpc_rqst *req, u32 *p, * Encode SETACL arguments */ static int -nfs3_xdr_setaclargs(struct rpc_rqst *req, u32 *p, +nfs3_xdr_setaclargs(struct rpc_rqst *req, __be32 *p, struct nfs3_setaclargs *args) { struct xdr_buf *buf = &req->rq_snd_buf; @@ -711,7 +711,7 @@ nfs3_xdr_setaclargs(struct rpc_rqst *req, u32 *p, * Decode attrstat reply. */ static int -nfs3_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_attrstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status; @@ -726,7 +726,7 @@ nfs3_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * SATTR, REMOVE, RMDIR */ static int -nfs3_xdr_wccstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_wccstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status; @@ -740,7 +740,7 @@ nfs3_xdr_wccstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * Decode LOOKUP reply */ static int -nfs3_xdr_lookupres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) +nfs3_xdr_lookupres(struct rpc_rqst *req, __be32 *p, struct nfs3_diropres *res) { int status; @@ -759,7 +759,7 @@ nfs3_xdr_lookupres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) * Decode ACCESS reply */ static int -nfs3_xdr_accessres(struct rpc_rqst *req, u32 *p, struct nfs3_accessres *res) +nfs3_xdr_accessres(struct rpc_rqst *req, __be32 *p, struct nfs3_accessres *res) { int status = ntohl(*p++); @@ -771,7 +771,7 @@ nfs3_xdr_accessres(struct rpc_rqst *req, u32 *p, struct nfs3_accessres *res) } static int -nfs3_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_readlinkargs *args) +nfs3_xdr_readlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs3_readlinkargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -789,7 +789,7 @@ nfs3_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_readlinkargs *ar * Decode READLINK reply */ static int -nfs3_xdr_readlinkres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -837,7 +837,7 @@ nfs3_xdr_readlinkres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * Decode READ reply */ static int -nfs3_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) +nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; int status, count, ocount, recvd, hdrlen; @@ -888,7 +888,7 @@ nfs3_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) * Decode WRITE response */ static int -nfs3_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) +nfs3_xdr_writeres(struct rpc_rqst *req, __be32 *p, struct nfs_writeres *res) { int status; @@ -910,7 +910,7 @@ nfs3_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) * Decode a CREATE response */ static int -nfs3_xdr_createres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) +nfs3_xdr_createres(struct rpc_rqst *req, __be32 *p, struct nfs3_diropres *res) { int status; @@ -937,7 +937,7 @@ nfs3_xdr_createres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) * Decode RENAME reply */ static int -nfs3_xdr_renameres(struct rpc_rqst *req, u32 *p, struct nfs3_renameres *res) +nfs3_xdr_renameres(struct rpc_rqst *req, __be32 *p, struct nfs3_renameres *res) { int status; @@ -952,7 +952,7 @@ nfs3_xdr_renameres(struct rpc_rqst *req, u32 *p, struct nfs3_renameres *res) * Decode LINK reply */ static int -nfs3_xdr_linkres(struct rpc_rqst *req, u32 *p, struct nfs3_linkres *res) +nfs3_xdr_linkres(struct rpc_rqst *req, __be32 *p, struct nfs3_linkres *res) { int status; @@ -967,7 +967,7 @@ nfs3_xdr_linkres(struct rpc_rqst *req, u32 *p, struct nfs3_linkres *res) * Decode FSSTAT reply */ static int -nfs3_xdr_fsstatres(struct rpc_rqst *req, u32 *p, struct nfs_fsstat *res) +nfs3_xdr_fsstatres(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *res) { int status; @@ -992,7 +992,7 @@ nfs3_xdr_fsstatres(struct rpc_rqst *req, u32 *p, struct nfs_fsstat *res) * Decode FSINFO reply */ static int -nfs3_xdr_fsinfores(struct rpc_rqst *req, u32 *p, struct nfs_fsinfo *res) +nfs3_xdr_fsinfores(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *res) { int status; @@ -1020,7 +1020,7 @@ nfs3_xdr_fsinfores(struct rpc_rqst *req, u32 *p, struct nfs_fsinfo *res) * Decode PATHCONF reply */ static int -nfs3_xdr_pathconfres(struct rpc_rqst *req, u32 *p, struct nfs_pathconf *res) +nfs3_xdr_pathconfres(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *res) { int status; @@ -1040,7 +1040,7 @@ nfs3_xdr_pathconfres(struct rpc_rqst *req, u32 *p, struct nfs_pathconf *res) * Decode COMMIT reply */ static int -nfs3_xdr_commitres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) +nfs3_xdr_commitres(struct rpc_rqst *req, __be32 *p, struct nfs_writeres *res) { int status; @@ -1059,7 +1059,7 @@ nfs3_xdr_commitres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) * Decode GETACL reply */ static int -nfs3_xdr_getaclres(struct rpc_rqst *req, u32 *p, +nfs3_xdr_getaclres(struct rpc_rqst *req, __be32 *p, struct nfs3_getaclres *res) { struct xdr_buf *buf = &req->rq_rcv_buf; @@ -1091,7 +1091,7 @@ nfs3_xdr_getaclres(struct rpc_rqst *req, u32 *p, * Decode setacl reply. */ static int -nfs3_xdr_setaclres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_setaclres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status = ntohl(*p++); -- cgit v0.10.2 From 8687b63afbe42103730bff4d3f7bfff3463c303e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:48 -0700 Subject: [PATCH] xdr annotations: NFSv4 on-the-wire data is big-endian [in large part pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 3dd413f..e284123 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -471,7 +471,7 @@ struct compound_hdr { static void encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 4 + len); BUG_ON(p == NULL); @@ -480,7 +480,7 @@ static void encode_string(struct xdr_stream *xdr, unsigned int len, const char * static int encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) { - uint32_t *p; + __be32 *p; dprintk("encode_compound: tag=%.*s\n", (int)hdr->taglen, hdr->tag); BUG_ON(hdr->taglen > NFS4_MAXTAGLEN); @@ -494,7 +494,7 @@ static int encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) static void encode_nfs4_verifier(struct xdr_stream *xdr, const nfs4_verifier *verf) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, NFS4_VERIFIER_SIZE); BUG_ON(p == NULL); @@ -507,8 +507,8 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s char owner_group[IDMAP_NAMESZ]; int owner_namelen = 0; int owner_grouplen = 0; - uint32_t *p; - uint32_t *q; + __be32 *p; + __be32 *q; int len; uint32_t bmval0 = 0; uint32_t bmval1 = 0; @@ -630,7 +630,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s static int encode_access(struct xdr_stream *xdr, u32 access) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8); WRITE32(OP_ACCESS); @@ -641,7 +641,7 @@ static int encode_access(struct xdr_stream *xdr, u32 access) static int encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_CLOSE); @@ -653,7 +653,7 @@ static int encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg) static int encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(16); WRITE32(OP_COMMIT); @@ -665,7 +665,7 @@ static int encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *arg static int encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *create) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8); WRITE32(OP_CREATE); @@ -697,7 +697,7 @@ static int encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *c static int encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(12); WRITE32(OP_GETATTR); @@ -708,7 +708,7 @@ static int encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap) static int encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(16); WRITE32(OP_GETATTR); @@ -740,7 +740,7 @@ static int encode_fs_locations(struct xdr_stream *xdr, const u32* bitmask) static int encode_getfh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_GETFH); @@ -750,7 +750,7 @@ static int encode_getfh(struct xdr_stream *xdr) static int encode_link(struct xdr_stream *xdr, const struct qstr *name) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + name->len); WRITE32(OP_LINK); @@ -780,7 +780,7 @@ static inline uint64_t nfs4_lock_length(struct file_lock *fl) */ static int encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(32); WRITE32(OP_LOCK); @@ -809,7 +809,7 @@ static int encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args) static int encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(40); WRITE32(OP_LOCKT); @@ -825,7 +825,7 @@ static int encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *arg static int encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(44); WRITE32(OP_LOCKU); @@ -841,7 +841,7 @@ static int encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *arg static int encode_lookup(struct xdr_stream *xdr, const struct qstr *name) { int len = name->len; - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + len); WRITE32(OP_LOOKUP); @@ -853,7 +853,7 @@ static int encode_lookup(struct xdr_stream *xdr, const struct qstr *name) static void encode_share_access(struct xdr_stream *xdr, int open_flags) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8); switch (open_flags & (FMODE_READ|FMODE_WRITE)) { @@ -874,7 +874,7 @@ static void encode_share_access(struct xdr_stream *xdr, int open_flags) static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_openargs *arg) { - uint32_t *p; + __be32 *p; /* * opcode 4, seqid 4, share_access 4, share_deny 4, clientid 8, ownerlen 4, * owner 4 = 32 @@ -891,7 +891,7 @@ static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_opena static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_openargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); switch(arg->open_flags & O_EXCL) { @@ -907,7 +907,7 @@ static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_op static void encode_opentype(struct xdr_stream *xdr, const struct nfs_openargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); switch (arg->open_flags & O_CREAT) { @@ -923,7 +923,7 @@ static void encode_opentype(struct xdr_stream *xdr, const struct nfs_openargs *a static inline void encode_delegation_type(struct xdr_stream *xdr, int delegation_type) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); switch (delegation_type) { @@ -943,7 +943,7 @@ static inline void encode_delegation_type(struct xdr_stream *xdr, int delegation static inline void encode_claim_null(struct xdr_stream *xdr, const struct qstr *name) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(NFS4_OPEN_CLAIM_NULL); @@ -952,7 +952,7 @@ static inline void encode_claim_null(struct xdr_stream *xdr, const struct qstr * static inline void encode_claim_previous(struct xdr_stream *xdr, int type) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(NFS4_OPEN_CLAIM_PREVIOUS); @@ -961,7 +961,7 @@ static inline void encode_claim_previous(struct xdr_stream *xdr, int type) static inline void encode_claim_delegate_cur(struct xdr_stream *xdr, const struct qstr *name, const nfs4_stateid *stateid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4+sizeof(stateid->data)); WRITE32(NFS4_OPEN_CLAIM_DELEGATE_CUR); @@ -991,7 +991,7 @@ static int encode_open(struct xdr_stream *xdr, const struct nfs_openargs *arg) static int encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_confirmargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_OPEN_CONFIRM); @@ -1003,7 +1003,7 @@ static int encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_con static int encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closeargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_OPEN_DOWNGRADE); @@ -1017,7 +1017,7 @@ static int encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh) { int len = fh->size; - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + len); WRITE32(OP_PUTFH); @@ -1029,7 +1029,7 @@ encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh) static int encode_putrootfh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_PUTROOTFH); @@ -1040,7 +1040,7 @@ static int encode_putrootfh(struct xdr_stream *xdr) static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx) { nfs4_stateid stateid; - uint32_t *p; + __be32 *p; RESERVE_SPACE(16); if (ctx->state != NULL) { @@ -1052,7 +1052,7 @@ static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context static int encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_READ); @@ -1074,7 +1074,7 @@ static int encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg FATTR4_WORD1_MOUNTED_ON_FILEID, }; int replen; - uint32_t *p; + __be32 *p; RESERVE_SPACE(32+sizeof(nfs4_verifier)); WRITE32(OP_READDIR); @@ -1116,7 +1116,7 @@ static int encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *r { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_READLINK); @@ -1134,7 +1134,7 @@ static int encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *r static int encode_remove(struct xdr_stream *xdr, const struct qstr *name) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + name->len); WRITE32(OP_REMOVE); @@ -1146,7 +1146,7 @@ static int encode_remove(struct xdr_stream *xdr, const struct qstr *name) static int encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, const struct qstr *newname) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + oldname->len); WRITE32(OP_RENAME); @@ -1162,7 +1162,7 @@ static int encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, con static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(12); WRITE32(OP_RENEW); @@ -1174,7 +1174,7 @@ static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_ static int encode_restorefh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_RESTOREFH); @@ -1185,7 +1185,7 @@ encode_restorefh(struct xdr_stream *xdr) static int encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4+sizeof(zero_stateid.data)); WRITE32(OP_SETATTR); @@ -1204,7 +1204,7 @@ encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg) static int encode_savefh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_SAVEFH); @@ -1215,7 +1215,7 @@ encode_savefh(struct xdr_stream *xdr) static int encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs *arg, const struct nfs_server *server) { int status; - uint32_t *p; + __be32 *p; RESERVE_SPACE(4+sizeof(arg->stateid.data)); WRITE32(OP_SETATTR); @@ -1229,7 +1229,7 @@ static int encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs * static int encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclientid *setclientid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4 + sizeof(setclientid->sc_verifier->data)); WRITE32(OP_SETCLIENTID); @@ -1248,7 +1248,7 @@ static int encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclien static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(12 + sizeof(client_state->cl_confirm.data)); WRITE32(OP_SETCLIENTID_CONFIRM); @@ -1260,7 +1260,7 @@ static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_c static int encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_WRITE); @@ -1279,7 +1279,7 @@ static int encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args static int encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *stateid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(20); @@ -1295,7 +1295,7 @@ static int encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *statei /* * Encode an ACCESS request */ -static int nfs4_xdr_enc_access(struct rpc_rqst *req, uint32_t *p, const struct nfs4_accessargs *args) +static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs4_accessargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1313,7 +1313,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, uint32_t *p, const struct n /* * Encode LOOKUP request */ -static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, uint32_t *p, const struct nfs4_lookup_arg *args) +static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs4_lookup_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1337,7 +1337,7 @@ out: /* * Encode LOOKUP_ROOT request */ -static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, uint32_t *p, const struct nfs4_lookup_root_arg *args) +static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struct nfs4_lookup_root_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1358,7 +1358,7 @@ out: /* * Encode REMOVE request */ -static int nfs4_xdr_enc_remove(struct rpc_rqst *req, uint32_t *p, const struct nfs4_remove_arg *args) +static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs4_remove_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1380,7 +1380,7 @@ out: /* * Encode RENAME request */ -static int nfs4_xdr_enc_rename(struct rpc_rqst *req, uint32_t *p, const struct nfs4_rename_arg *args) +static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs4_rename_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1410,7 +1410,7 @@ out: /* * Encode LINK request */ -static int nfs4_xdr_enc_link(struct rpc_rqst *req, uint32_t *p, const struct nfs4_link_arg *args) +static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_link_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1440,7 +1440,7 @@ out: /* * Encode CREATE request */ -static int nfs4_xdr_enc_create(struct rpc_rqst *req, uint32_t *p, const struct nfs4_create_arg *args) +static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs4_create_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1470,7 +1470,7 @@ out: /* * Encode SYMLINK request */ -static int nfs4_xdr_enc_symlink(struct rpc_rqst *req, uint32_t *p, const struct nfs4_create_arg *args) +static int nfs4_xdr_enc_symlink(struct rpc_rqst *req, __be32 *p, const struct nfs4_create_arg *args) { return nfs4_xdr_enc_create(req, p, args); } @@ -1478,7 +1478,7 @@ static int nfs4_xdr_enc_symlink(struct rpc_rqst *req, uint32_t *p, const struct /* * Encode GETATTR request */ -static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, uint32_t *p, const struct nfs4_getattr_arg *args) +static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nfs4_getattr_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1496,7 +1496,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, uint32_t *p, const struct /* * Encode a CLOSE request */ -static int nfs4_xdr_enc_close(struct rpc_rqst *req, uint32_t *p, struct nfs_closeargs *args) +static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1520,7 +1520,7 @@ out: /* * Encode an OPEN request */ -static int nfs4_xdr_enc_open(struct rpc_rqst *req, uint32_t *p, struct nfs_openargs *args) +static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1556,7 +1556,7 @@ out: /* * Encode an OPEN_CONFIRM request */ -static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_open_confirmargs *args) +static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_open_confirmargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1577,7 +1577,7 @@ out: /* * Encode an OPEN request with no attributes. */ -static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, uint32_t *p, struct nfs_openargs *args) +static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_openargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1601,7 +1601,7 @@ out: /* * Encode an OPEN_DOWNGRADE request */ -static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, uint32_t *p, struct nfs_closeargs *args) +static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct nfs_closeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1625,7 +1625,7 @@ out: /* * Encode a LOCK request */ -static int nfs4_xdr_enc_lock(struct rpc_rqst *req, uint32_t *p, struct nfs_lock_args *args) +static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_args *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1646,7 +1646,7 @@ out: /* * Encode a LOCKT request */ -static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, uint32_t *p, struct nfs_lockt_args *args) +static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_args *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1667,7 +1667,7 @@ out: /* * Encode a LOCKU request */ -static int nfs4_xdr_enc_locku(struct rpc_rqst *req, uint32_t *p, struct nfs_locku_args *args) +static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_args *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1688,7 +1688,7 @@ out: /* * Encode a READLINK request */ -static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, uint32_t *p, const struct nfs4_readlink *args) +static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct nfs4_readlink *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1709,7 +1709,7 @@ out: /* * Encode a READDIR request */ -static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, uint32_t *p, const struct nfs4_readdir_arg *args) +static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nfs4_readdir_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1730,7 +1730,7 @@ out: /* * Encode a READ request */ -static int nfs4_xdr_enc_read(struct rpc_rqst *req, uint32_t *p, struct nfs_readargs *args) +static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; struct xdr_stream xdr; @@ -1762,7 +1762,7 @@ out: /* * Encode an SETATTR request */ -static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, uint32_t *p, struct nfs_setattrargs *args) +static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_setattrargs *args) { struct xdr_stream xdr; @@ -1788,7 +1788,7 @@ out: * Encode a GETACL request */ static int -nfs4_xdr_enc_getacl(struct rpc_rqst *req, uint32_t *p, +nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p, struct nfs_getaclargs *args) { struct xdr_stream xdr; @@ -1815,7 +1815,7 @@ out: /* * Encode a WRITE request */ -static int nfs4_xdr_enc_write(struct rpc_rqst *req, uint32_t *p, struct nfs_writeargs *args) +static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1839,7 +1839,7 @@ out: /* * a COMMIT request */ -static int nfs4_xdr_enc_commit(struct rpc_rqst *req, uint32_t *p, struct nfs_writeargs *args) +static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1863,7 +1863,7 @@ out: /* * FSINFO request */ -static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs4_fsinfo_arg *args) +static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsinfo_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1882,7 +1882,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs /* * a PATHCONF request */ -static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct nfs4_pathconf_arg *args) +static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct nfs4_pathconf_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1902,7 +1902,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct /* * a STATFS request */ -static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, uint32_t *p, const struct nfs4_statfs_arg *args) +static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs4_statfs_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1923,7 +1923,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, uint32_t *p, const struct n /* * GETATTR_BITMAP request */ -static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, uint32_t *p, const struct nfs_fh *fhandle) +static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struct nfs_fh *fhandle) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1945,7 +1945,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, uint32_t *p, const str /* * a RENEW request */ -static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp) +static int nfs4_xdr_enc_renew(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1960,7 +1960,7 @@ static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_clie /* * a SETCLIENTID request */ -static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, uint32_t *p, struct nfs4_setclientid *sc) +static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs4_setclientid *sc) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1975,7 +1975,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, uint32_t *p, struct nf /* * a SETCLIENTID_CONFIRM request */ -static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp) +static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1997,7 +1997,7 @@ static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, s /* * DELEGRETURN request */ -static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, uint32_t *p, const struct nfs4_delegreturnargs *args) +static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struct nfs4_delegreturnargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -2021,7 +2021,7 @@ out: /* * Encode FS_LOCATIONS request */ -static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs_locations_arg *args) +static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -2086,7 +2086,7 @@ out: static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char **string) { - uint32_t *p; + __be32 *p; READ_BUF(4); READ32(*len); @@ -2097,7 +2097,7 @@ static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) { - uint32_t *p; + __be32 *p; READ_BUF(8); READ32(hdr->status); @@ -2112,7 +2112,7 @@ static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) { - uint32_t *p; + __be32 *p; uint32_t opnum; int32_t nfserr; @@ -2134,7 +2134,7 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) /* Dummy routine */ static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp) { - uint32_t *p; + __be32 *p; unsigned int strlen; char *str; @@ -2144,7 +2144,8 @@ static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp) static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) { - uint32_t bmlen, *p; + uint32_t bmlen; + __be32 *p; READ_BUF(4); READ32(bmlen); @@ -2159,9 +2160,9 @@ static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) return 0; } -static inline int decode_attr_length(struct xdr_stream *xdr, uint32_t *attrlen, uint32_t **savep) +static inline int decode_attr_length(struct xdr_stream *xdr, uint32_t *attrlen, __be32 **savep) { - uint32_t *p; + __be32 *p; READ_BUF(4); READ32(*attrlen); @@ -2182,7 +2183,7 @@ static int decode_attr_supported(struct xdr_stream *xdr, uint32_t *bitmap, uint3 static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *type) { - uint32_t *p; + __be32 *p; *type = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_TYPE - 1U))) @@ -2202,7 +2203,7 @@ static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *change) { - uint32_t *p; + __be32 *p; *change = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_CHANGE - 1U))) @@ -2219,7 +2220,7 @@ static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *size) { - uint32_t *p; + __be32 *p; *size = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_SIZE - 1U))) @@ -2235,7 +2236,7 @@ static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t * static int decode_attr_link_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_LINK_SUPPORT - 1U))) @@ -2251,7 +2252,7 @@ static int decode_attr_link_support(struct xdr_stream *xdr, uint32_t *bitmap, ui static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_SYMLINK_SUPPORT - 1U))) @@ -2267,7 +2268,7 @@ static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fsid *fsid) { - uint32_t *p; + __be32 *p; fsid->major = 0; fsid->minor = 0; @@ -2287,7 +2288,7 @@ static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = 60; if (unlikely(bitmap[0] & (FATTR4_WORD0_LEASE_TIME - 1U))) @@ -2303,7 +2304,7 @@ static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = ACL4_SUPPORT_ALLOW_ACL|ACL4_SUPPORT_DENY_ACL; if (unlikely(bitmap[0] & (FATTR4_WORD0_ACLSUPPORT - 1U))) @@ -2319,7 +2320,7 @@ static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) { - uint32_t *p; + __be32 *p; *fileid = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_FILEID - 1U))) @@ -2335,7 +2336,7 @@ static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) { - uint32_t *p; + __be32 *p; *fileid = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_MOUNTED_ON_FILEID - 1U))) @@ -2351,7 +2352,7 @@ static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitma static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2368,7 +2369,7 @@ static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_files_free(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2385,7 +2386,7 @@ static int decode_attr_files_free(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2403,7 +2404,7 @@ static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_pathname(struct xdr_stream *xdr, struct nfs4_pathname *path) { int n; - uint32_t *p; + __be32 *p; int status = 0; READ_BUF(4); @@ -2448,7 +2449,7 @@ out_eio: static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_fs_locations *res) { int n; - uint32_t *p; + __be32 *p; int status = -EIO; if (unlikely(bitmap[0] & (FATTR4_WORD0_FS_LOCATIONS -1U))) @@ -2512,7 +2513,7 @@ out_eio: static int decode_attr_maxfilesize(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2529,7 +2530,7 @@ static int decode_attr_maxfilesize(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_maxlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *maxlink) { - uint32_t *p; + __be32 *p; int status = 0; *maxlink = 1; @@ -2546,7 +2547,7 @@ static int decode_attr_maxlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_ static int decode_attr_maxname(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *maxname) { - uint32_t *p; + __be32 *p; int status = 0; *maxname = 1024; @@ -2563,7 +2564,7 @@ static int decode_attr_maxname(struct xdr_stream *xdr, uint32_t *bitmap, uint32_ static int decode_attr_maxread(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 1024; @@ -2584,7 +2585,7 @@ static int decode_attr_maxread(struct xdr_stream *xdr, uint32_t *bitmap, uint32_ static int decode_attr_maxwrite(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 1024; @@ -2605,7 +2606,7 @@ static int decode_attr_maxwrite(struct xdr_stream *xdr, uint32_t *bitmap, uint32 static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *mode) { - uint32_t *p; + __be32 *p; *mode = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_MODE - 1U))) @@ -2622,7 +2623,7 @@ static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *nlink) { - uint32_t *p; + __be32 *p; *nlink = 1; if (unlikely(bitmap[1] & (FATTR4_WORD1_NUMLINKS - 1U))) @@ -2638,7 +2639,8 @@ static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *uid) { - uint32_t len, *p; + uint32_t len; + __be32 *p; *uid = -2; if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER - 1U))) @@ -2662,7 +2664,8 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *gid) { - uint32_t len, *p; + uint32_t len; + __be32 *p; *gid = -2; if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER_GROUP - 1U))) @@ -2686,7 +2689,8 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nf static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rdev) { - uint32_t major = 0, minor = 0, *p; + uint32_t major = 0, minor = 0; + __be32 *p; *rdev = MKDEV(0,0); if (unlikely(bitmap[1] & (FATTR4_WORD1_RAWDEV - 1U))) @@ -2708,7 +2712,7 @@ static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rde static int decode_attr_space_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2725,7 +2729,7 @@ static int decode_attr_space_avail(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_space_free(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2742,7 +2746,7 @@ static int decode_attr_space_free(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_space_total(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2759,7 +2763,7 @@ static int decode_attr_space_total(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *used) { - uint32_t *p; + __be32 *p; *used = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_USED - 1U))) @@ -2776,7 +2780,7 @@ static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_time(struct xdr_stream *xdr, struct timespec *time) { - uint32_t *p; + __be32 *p; uint64_t sec; uint32_t nsec; @@ -2836,7 +2840,7 @@ static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, str return status; } -static int verify_attr_len(struct xdr_stream *xdr, uint32_t *savep, uint32_t attrlen) +static int verify_attr_len(struct xdr_stream *xdr, __be32 *savep, uint32_t attrlen) { unsigned int attrwords = XDR_QUADLEN(attrlen); unsigned int nwords = xdr->p - savep; @@ -2854,7 +2858,7 @@ static int verify_attr_len(struct xdr_stream *xdr, uint32_t *savep, uint32_t att static int decode_change_info(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) { - uint32_t *p; + __be32 *p; READ_BUF(20); READ32(cinfo->atomic); @@ -2865,7 +2869,7 @@ static int decode_change_info(struct xdr_stream *xdr, struct nfs4_change_info *c static int decode_access(struct xdr_stream *xdr, struct nfs4_accessres *access) { - uint32_t *p; + __be32 *p; uint32_t supp, acc; int status; @@ -2882,7 +2886,7 @@ static int decode_access(struct xdr_stream *xdr, struct nfs4_accessres *access) static int decode_close(struct xdr_stream *xdr, struct nfs_closeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_CLOSE); @@ -2895,7 +2899,7 @@ static int decode_close(struct xdr_stream *xdr, struct nfs_closeres *res) static int decode_commit(struct xdr_stream *xdr, struct nfs_writeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_COMMIT); @@ -2908,7 +2912,7 @@ static int decode_commit(struct xdr_stream *xdr, struct nfs_writeres *res) static int decode_create(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) { - uint32_t *p; + __be32 *p; uint32_t bmlen; int status; @@ -2925,7 +2929,7 @@ static int decode_create(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) static int decode_server_caps(struct xdr_stream *xdr, struct nfs4_server_caps_res *res) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; int status; @@ -2952,7 +2956,7 @@ xdr_error: static int decode_statfs(struct xdr_stream *xdr, struct nfs_fsstat *fsstat) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; int status; @@ -2985,7 +2989,7 @@ xdr_error: static int decode_pathconf(struct xdr_stream *xdr, struct nfs_pathconf *pathconf) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; int status; @@ -3010,7 +3014,7 @@ xdr_error: static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, const struct nfs_server *server) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}, type; @@ -3079,7 +3083,7 @@ xdr_error: static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2]; int status; @@ -3111,7 +3115,7 @@ xdr_error: static int decode_getfh(struct xdr_stream *xdr, struct nfs_fh *fh) { - uint32_t *p; + __be32 *p; uint32_t len; int status; @@ -3147,7 +3151,7 @@ static int decode_link(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) static int decode_lock_denied (struct xdr_stream *xdr, struct file_lock *fl) { uint64_t offset, length, clientid; - uint32_t *p; + __be32 *p; uint32_t namelen, type; READ_BUF(32); @@ -3172,7 +3176,7 @@ static int decode_lock_denied (struct xdr_stream *xdr, struct file_lock *fl) static int decode_lock(struct xdr_stream *xdr, struct nfs_lock_res *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_LOCK); @@ -3195,7 +3199,7 @@ static int decode_lockt(struct xdr_stream *xdr, struct nfs_lockt_res *res) static int decode_locku(struct xdr_stream *xdr, struct nfs_locku_res *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_LOCKU); @@ -3214,7 +3218,7 @@ static int decode_lookup(struct xdr_stream *xdr) /* This is too sick! */ static int decode_space_limit(struct xdr_stream *xdr, u64 *maxsize) { - uint32_t *p; + __be32 *p; uint32_t limit_type, nblocks, blocksize; READ_BUF(12); @@ -3233,7 +3237,7 @@ static int decode_space_limit(struct xdr_stream *xdr, u64 *maxsize) static int decode_delegation(struct xdr_stream *xdr, struct nfs_openres *res) { - uint32_t *p; + __be32 *p; uint32_t delegation_type; READ_BUF(4); @@ -3259,7 +3263,7 @@ static int decode_delegation(struct xdr_stream *xdr, struct nfs_openres *res) static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res) { - uint32_t *p; + __be32 *p; uint32_t bmlen; int status; @@ -3287,7 +3291,7 @@ xdr_error: static int decode_open_confirm(struct xdr_stream *xdr, struct nfs_open_confirmres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_OPEN_CONFIRM); @@ -3300,7 +3304,7 @@ static int decode_open_confirm(struct xdr_stream *xdr, struct nfs_open_confirmre static int decode_open_downgrade(struct xdr_stream *xdr, struct nfs_closeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_OPEN_DOWNGRADE); @@ -3324,7 +3328,7 @@ static int decode_putrootfh(struct xdr_stream *xdr) static int decode_read(struct xdr_stream *xdr, struct rpc_rqst *req, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; - uint32_t *p; + __be32 *p; uint32_t count, eof, recvd, hdrlen; int status; @@ -3354,7 +3358,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n struct page *page = *rcvbuf->pages; struct kvec *iov = rcvbuf->head; unsigned int nr, pglen = rcvbuf->page_len; - uint32_t *end, *entry, *p, *kaddr; + __be32 *end, *entry, *p, *kaddr; uint32_t len, attrlen, xlen; int hdrlen, recvd, status; @@ -3376,7 +3380,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n xdr_read_pages(xdr, pglen); BUG_ON(pglen + readdir->pgbase > PAGE_CACHE_SIZE); - kaddr = p = (uint32_t *) kmap_atomic(page, KM_USER0); + kaddr = p = kmap_atomic(page, KM_USER0); end = p + ((pglen + readdir->pgbase) >> 2); entry = p; for (nr = 0; *p++; nr++) { @@ -3428,7 +3432,7 @@ static int decode_readlink(struct xdr_stream *xdr, struct rpc_rqst *req) struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; int hdrlen, len, recvd; - uint32_t *p; + __be32 *p; char *kaddr; int status; @@ -3505,7 +3509,7 @@ decode_restorefh(struct xdr_stream *xdr) static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, size_t *acl_len) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; struct kvec *iov = req->rq_rcv_buf.head; @@ -3551,7 +3555,7 @@ decode_savefh(struct xdr_stream *xdr) static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res) { - uint32_t *p; + __be32 *p; uint32_t bmlen; int status; @@ -3567,7 +3571,7 @@ static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res) static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp) { - uint32_t *p; + __be32 *p; uint32_t opnum; int32_t nfserr; @@ -3610,7 +3614,7 @@ static int decode_setclientid_confirm(struct xdr_stream *xdr) static int decode_write(struct xdr_stream *xdr, struct nfs_writeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_WRITE); @@ -3632,7 +3636,7 @@ static int decode_delegreturn(struct xdr_stream *xdr) /* * Decode OPEN_DOWNGRADE response */ -static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_closeres *res) +static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, __be32 *p, struct nfs_closeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3660,7 +3664,7 @@ out: /* * Decode ACCESS response */ -static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_accessres *res) +static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_accessres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3678,7 +3682,7 @@ out: /* * Decode LOOKUP response */ -static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_lookup_res *res) +static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lookup_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3701,7 +3705,7 @@ out: /* * Decode LOOKUP_ROOT response */ -static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_lookup_res *res) +static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lookup_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3721,7 +3725,7 @@ out: /* * Decode REMOVE response */ -static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_remove_res *res) +static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_remove_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3742,7 +3746,7 @@ out: /* * Decode RENAME response */ -static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_rename_res *res) +static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_rename_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3772,7 +3776,7 @@ out: /* * Decode LINK response */ -static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_link_res *res) +static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_link_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3805,7 +3809,7 @@ out: /* * Decode CREATE response */ -static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_create_res *res) +static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_create_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3834,7 +3838,7 @@ out: /* * Decode SYMLINK response */ -static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_create_res *res) +static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_create_res *res) { return nfs4_xdr_dec_create(rqstp, p, res); } @@ -3842,7 +3846,7 @@ static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4 /* * Decode GETATTR response */ -static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_getattr_res *res) +static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_getattr_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3865,7 +3869,7 @@ out: * Encode an SETACL request */ static int -nfs4_xdr_enc_setacl(struct rpc_rqst *req, uint32_t *p, struct nfs_setaclargs *args) +nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -3886,7 +3890,7 @@ out: * Decode SETACL response */ static int -nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, uint32_t *p, void *res) +nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3908,7 +3912,7 @@ out: * Decode GETACL response */ static int -nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, uint32_t *p, size_t *acl_len) +nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3930,7 +3934,7 @@ out: /* * Decode CLOSE response */ -static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_closeres *res) +static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, __be32 *p, struct nfs_closeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3960,7 +3964,7 @@ out: /* * Decode OPEN response */ -static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_openres *res) +static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3994,7 +3998,7 @@ out: /* * Decode OPEN_CONFIRM response */ -static int nfs4_xdr_dec_open_confirm(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_open_confirmres *res) +static int nfs4_xdr_dec_open_confirm(struct rpc_rqst *rqstp, __be32 *p, struct nfs_open_confirmres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4015,7 +4019,7 @@ out: /* * Decode OPEN response */ -static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_openres *res) +static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4039,7 +4043,7 @@ out: /* * Decode SETATTR response */ -static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_setattrres *res) +static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_setattrres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4065,7 +4069,7 @@ out: /* * Decode LOCK response */ -static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_lock_res *res) +static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4086,7 +4090,7 @@ out: /* * Decode LOCKT response */ -static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_lockt_res *res) +static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lockt_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4107,7 +4111,7 @@ out: /* * Decode LOCKU response */ -static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_locku_res *res) +static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, __be32 *p, struct nfs_locku_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4128,7 +4132,7 @@ out: /* * Decode READLINK response */ -static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, uint32_t *p, void *res) +static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, void *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4149,7 +4153,7 @@ out: /* * Decode READDIR response */ -static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_readdir_res *res) +static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_readdir_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4170,7 +4174,7 @@ out: /* * Decode Read response */ -static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_readres *res) +static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, __be32 *p, struct nfs_readres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4193,7 +4197,7 @@ out: /* * Decode WRITE response */ -static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_writeres *res) +static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4219,7 +4223,7 @@ out: /* * Decode COMMIT response */ -static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_writeres *res) +static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4243,7 +4247,7 @@ out: /* * FSINFO request */ -static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs_fsinfo *fsinfo) +static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4263,7 +4267,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs_fsi /* * PATHCONF request */ -static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, uint32_t *p, struct nfs_pathconf *pathconf) +static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *pathconf) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4281,7 +4285,7 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, uint32_t *p, struct nfs_p /* * STATFS request */ -static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, uint32_t *p, struct nfs_fsstat *fsstat) +static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *fsstat) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4299,7 +4303,7 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, uint32_t *p, struct nfs_fss /* * GETATTR_BITMAP request */ -static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, uint32_t *p, struct nfs4_server_caps_res *res) +static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4_server_caps_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4318,7 +4322,7 @@ out: /* * Decode RENEW response */ -static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, uint32_t *p, void *dummy) +static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, __be32 *p, void *dummy) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4334,7 +4338,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, uint32_t *p, void *dummy) /* * a SETCLIENTID request */ -static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p, +static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp) { struct xdr_stream xdr; @@ -4353,7 +4357,7 @@ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p, /* * a SETCLIENTID_CONFIRM request */ -static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_fsinfo *fsinfo) +static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4375,7 +4379,7 @@ static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, s /* * DELEGRETURN request */ -static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_delegreturnres *res) +static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_delegreturnres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4397,7 +4401,7 @@ out: /* * FS_LOCATIONS request */ -static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs_locations *res) +static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations *res) { struct xdr_stream xdr; struct compound_hdr hdr; -- cgit v0.10.2 From 0dbb4c6799cf8fa8c5ba1926153a30960117477d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:49 -0700 Subject: [PATCH] xdr annotations: NFS readdir entries on-the-wire data is big-endian [in large part pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index c86a1ea..4133ef5 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -142,12 +142,12 @@ nfs_opendir(struct inode *inode, struct file *filp) return res; } -typedef u32 * (*decode_dirent_t)(u32 *, struct nfs_entry *, int); +typedef __be32 * (*decode_dirent_t)(__be32 *, struct nfs_entry *, int); typedef struct { struct file *file; struct page *page; unsigned long page_index; - u32 *ptr; + __be32 *ptr; u64 *dir_cookie; loff_t current_index; struct nfs_entry *entry; @@ -220,7 +220,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) static inline int dir_decode(nfs_readdir_descriptor_t *desc) { - u32 *p = desc->ptr; + __be32 *p = desc->ptr; p = desc->decode(p, desc->entry, desc->plus); if (IS_ERR(p)) return PTR_ERR(p); diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index bea0b01..d205466 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -93,15 +93,15 @@ extern void nfs_destroy_directcache(void); /* nfs2xdr.c */ extern int nfs_stat_to_errno(int); extern struct rpc_procinfo nfs_procedures[]; -extern u32 * nfs_decode_dirent(u32 *, struct nfs_entry *, int); +extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int); /* nfs3xdr.c */ extern struct rpc_procinfo nfs3_procedures[]; -extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int); +extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int); /* nfs4xdr.c */ #ifdef CONFIG_NFS_V4 -extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); +extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); #endif /* nfs4proc.c */ diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 1d801e3..3be4e72 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -468,8 +468,8 @@ err_unmap: goto out; } -u32 * -nfs_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) +__be32 * +nfs_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { if (!*p++) { if (!*p) diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index b4e740e..0ace092 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -583,8 +583,8 @@ err_unmap: goto out; } -u32 * -nfs3_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) +__be32 * +nfs3_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { struct nfs_entry old = *entry; diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 61095fe..6f34667 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -212,7 +212,7 @@ extern void nfs_free_seqid(struct nfs_seqid *seqid); extern const nfs4_stateid zero_stateid; /* nfs4xdr.c */ -extern uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus); +extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); extern struct rpc_procinfo nfs4_procedures[]; struct nfs4_mount_data; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 7421bcb..8d09b47 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -141,7 +141,7 @@ const u32 nfs4_fs_locations_bitmap[2] = { static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { - u32 *start, *p; + __be32 *start, *p; BUG_ON(readdir->count < 80); if (cookie > 2) { @@ -162,7 +162,7 @@ static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, * when talking to the server, we always send cookie 0 * instead of 1 or 2. */ - start = p = (u32 *)kmap_atomic(*readdir->pages, KM_USER0); + start = p = kmap_atomic(*readdir->pages, KM_USER0); if (cookie == 0) { *p++ = xdr_one; /* next */ diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index e284123..0cf3fa3 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -4421,7 +4421,7 @@ out: return status; } -uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus) +__be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { uint32_t bitmap[2] = {0}; uint32_t len; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index dc5397d..ac8058b 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -811,7 +811,7 @@ struct nfs_rpc_ops { int (*pathconf) (struct nfs_server *, struct nfs_fh *, struct nfs_pathconf *); int (*set_capabilities)(struct nfs_server *, struct nfs_fh *); - u32 * (*decode_dirent)(u32 *, struct nfs_entry *, int plus); + __be32 *(*decode_dirent)(__be32 *, struct nfs_entry *, int plus); void (*read_setup) (struct nfs_read_data *); int (*read_done) (struct rpc_task *, struct nfs_read_data *); void (*write_setup) (struct nfs_write_data *, int how); -- cgit v0.10.2 From e6f684f6443dd37384c63d2f27571350e0b5c8aa Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:50 -0700 Subject: [PATCH] fs/nfs/callback* passes error values big-endian [pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index 5676163d..6921d82 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -44,7 +44,7 @@ struct cb_getattrargs { }; struct cb_getattrres { - uint32_t status; + __be32 status; uint32_t bitmap[2]; uint64_t size; uint64_t change_attr; @@ -59,8 +59,8 @@ struct cb_recallargs { uint32_t truncate; }; -extern unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res); -extern unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy); +extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res); +extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy); #ifdef CONFIG_NFS_V4 extern int nfs_callback_up(void); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 97cf8f7..72e55d8 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -14,7 +14,7 @@ #define NFSDBG_FACILITY NFSDBG_CALLBACK -unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res) +__be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res) { struct nfs_client *clp; struct nfs_delegation *delegation; @@ -55,11 +55,11 @@ out: return res->status; } -unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy) +__be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy) { struct nfs_client *clp; struct inode *inode; - unsigned res; + __be32 res; res = htonl(NFS4ERR_BADHANDLE); clp = nfs_find_client(args->addr, 4); diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 5998d0c..909a140 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -22,9 +22,9 @@ #define NFSDBG_FACILITY NFSDBG_CALLBACK -typedef unsigned (*callback_process_op_t)(void *, void *); -typedef unsigned (*callback_decode_arg_t)(struct svc_rqst *, struct xdr_stream *, void *); -typedef unsigned (*callback_encode_res_t)(struct svc_rqst *, struct xdr_stream *, void *); +typedef __be32 (*callback_process_op_t)(void *, void *); +typedef __be32 (*callback_decode_arg_t)(struct svc_rqst *, struct xdr_stream *, void *); +typedef __be32 (*callback_encode_res_t)(struct svc_rqst *, struct xdr_stream *, void *); struct callback_op { @@ -61,7 +61,7 @@ static uint32_t *read_buf(struct xdr_stream *xdr, int nbytes) return p; } -static unsigned decode_string(struct xdr_stream *xdr, unsigned int *len, const char **str) +static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len, const char **str) { uint32_t *p; @@ -81,7 +81,7 @@ static unsigned decode_string(struct xdr_stream *xdr, unsigned int *len, const c return 0; } -static unsigned decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) +static __be32 decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) { uint32_t *p; @@ -99,7 +99,7 @@ static unsigned decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) return 0; } -static unsigned decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) +static __be32 decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) { uint32_t *p; unsigned int attrlen; @@ -118,7 +118,7 @@ static unsigned decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) return 0; } -static unsigned decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +static __be32 decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) { uint32_t *p; @@ -129,11 +129,11 @@ static unsigned decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) return 0; } -static unsigned decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr) +static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr) { uint32_t *p; unsigned int minor_version; - unsigned status; + __be32 status; status = decode_string(xdr, &hdr->taglen, &hdr->tag); if (unlikely(status != 0)) @@ -159,7 +159,7 @@ static unsigned decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compou return 0; } -static unsigned decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) +static __be32 decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) { uint32_t *p; p = read_buf(xdr, 4); @@ -169,9 +169,9 @@ static unsigned decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) return 0; } -static unsigned decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_getattrargs *args) +static __be32 decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_getattrargs *args) { - unsigned status; + __be32 status; status = decode_fh(xdr, &args->fh); if (unlikely(status != 0)) @@ -183,10 +183,10 @@ out: return status; } -static unsigned decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_recallargs *args) +static __be32 decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_recallargs *args) { uint32_t *p; - unsigned status; + __be32 status; args->addr = &rqstp->rq_addr; status = decode_stateid(xdr, &args->stateid); @@ -204,7 +204,7 @@ out: return status; } -static unsigned encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) +static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) { uint32_t *p; @@ -217,7 +217,7 @@ static unsigned encode_string(struct xdr_stream *xdr, unsigned int len, const ch #define CB_SUPPORTED_ATTR0 (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) #define CB_SUPPORTED_ATTR1 (FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY) -static unsigned encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, uint32_t **savep) +static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, uint32_t **savep) { uint32_t bm[2]; uint32_t *p; @@ -247,7 +247,7 @@ static unsigned encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitma return 0; } -static unsigned encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t change) +static __be32 encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t change) { uint32_t *p; @@ -260,7 +260,7 @@ static unsigned encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitma return 0; } -static unsigned encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t size) +static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t size) { uint32_t *p; @@ -273,7 +273,7 @@ static unsigned encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, return 0; } -static unsigned encode_attr_time(struct xdr_stream *xdr, const struct timespec *time) +static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec *time) { uint32_t *p; @@ -285,23 +285,23 @@ static unsigned encode_attr_time(struct xdr_stream *xdr, const struct timespec * return 0; } -static unsigned encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) +static __be32 encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) { if (!(bitmap[1] & FATTR4_WORD1_TIME_METADATA)) return 0; return encode_attr_time(xdr,time); } -static unsigned encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) +static __be32 encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) { if (!(bitmap[1] & FATTR4_WORD1_TIME_MODIFY)) return 0; return encode_attr_time(xdr,time); } -static unsigned encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compound_hdr_res *hdr) +static __be32 encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compound_hdr_res *hdr) { - unsigned status; + __be32 status; hdr->status = xdr_reserve_space(xdr, 4); if (unlikely(hdr->status == NULL)) @@ -315,7 +315,7 @@ static unsigned encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compou return 0; } -static unsigned encode_op_hdr(struct xdr_stream *xdr, uint32_t op, uint32_t res) +static __be32 encode_op_hdr(struct xdr_stream *xdr, uint32_t op, __be32 res) { uint32_t *p; @@ -327,10 +327,10 @@ static unsigned encode_op_hdr(struct xdr_stream *xdr, uint32_t op, uint32_t res) return 0; } -static unsigned encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, const struct cb_getattrres *res) +static __be32 encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, const struct cb_getattrres *res) { uint32_t *savep = NULL; - unsigned status = res->status; + __be32 status = res->status; if (unlikely(status != 0)) goto out; @@ -353,15 +353,15 @@ out: return status; } -static unsigned process_op(struct svc_rqst *rqstp, +static __be32 process_op(struct svc_rqst *rqstp, struct xdr_stream *xdr_in, void *argp, struct xdr_stream *xdr_out, void *resp) { struct callback_op *op = &callback_ops[0]; unsigned int op_nr = OP_CB_ILLEGAL; - unsigned int status = 0; + __be32 status = 0; long maxlen; - unsigned res; + __be32 res; dprintk("%s: start\n", __FUNCTION__); status = decode_op_hdr(xdr_in, &op_nr); @@ -405,7 +405,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r struct cb_compound_hdr_res hdr_res; struct xdr_stream xdr_in, xdr_out; uint32_t *p; - unsigned int status; + __be32 status; unsigned int nops = 1; dprintk("%s: start\n", __FUNCTION__); -- cgit v0.10.2 From 5704fdeb41c9fb282ae576516f221ea0b8f64b2b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:51 -0700 Subject: [PATCH] xdr annotations: fs/nfs/callback* on-the-wire data is big-endian [mostly pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index 6921d82..db3d791 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -31,10 +31,10 @@ struct cb_compound_hdr_arg { }; struct cb_compound_hdr_res { - uint32_t *status; + __be32 *status; int taglen; const char *tag; - uint32_t *nops; + __be32 *nops; }; struct cb_getattrargs { diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 909a140..f8ea1f5 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -41,19 +41,19 @@ static __be32 nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) return htonl(NFS4_OK); } -static int nfs4_decode_void(struct svc_rqst *rqstp, uint32_t *p, void *dummy) +static int nfs4_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } -static int nfs4_encode_void(struct svc_rqst *rqstp, uint32_t *p, void *dummy) +static int nfs4_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } -static uint32_t *read_buf(struct xdr_stream *xdr, int nbytes) +static __be32 *read_buf(struct xdr_stream *xdr, int nbytes) { - uint32_t *p; + __be32 *p; p = xdr_inline_decode(xdr, nbytes); if (unlikely(p == NULL)) @@ -63,7 +63,7 @@ static uint32_t *read_buf(struct xdr_stream *xdr, int nbytes) static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len, const char **str) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 4); if (unlikely(p == NULL)) @@ -83,7 +83,7 @@ static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len, const cha static __be32 decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 4); if (unlikely(p == NULL)) @@ -101,7 +101,7 @@ static __be32 decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) static __be32 decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) { - uint32_t *p; + __be32 *p; unsigned int attrlen; p = read_buf(xdr, 4); @@ -120,7 +120,7 @@ static __be32 decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) static __be32 decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 16); if (unlikely(p == NULL)) @@ -131,7 +131,7 @@ static __be32 decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr) { - uint32_t *p; + __be32 *p; unsigned int minor_version; __be32 status; @@ -161,7 +161,7 @@ static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound static __be32 decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 4); if (unlikely(p == NULL)) return htonl(NFS4ERR_RESOURCE); @@ -185,7 +185,7 @@ out: static __be32 decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_recallargs *args) { - uint32_t *p; + __be32 *p; __be32 status; args->addr = &rqstp->rq_addr; @@ -206,7 +206,7 @@ out: static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 4 + len); if (unlikely(p == NULL)) @@ -217,10 +217,10 @@ static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char #define CB_SUPPORTED_ATTR0 (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) #define CB_SUPPORTED_ATTR1 (FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY) -static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, uint32_t **savep) +static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, __be32 **savep) { - uint32_t bm[2]; - uint32_t *p; + __be32 bm[2]; + __be32 *p; bm[0] = htonl(bitmap[0] & CB_SUPPORTED_ATTR0); bm[1] = htonl(bitmap[1] & CB_SUPPORTED_ATTR1); @@ -249,7 +249,7 @@ static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, static __be32 encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t change) { - uint32_t *p; + __be32 *p; if (!(bitmap[0] & FATTR4_WORD0_CHANGE)) return 0; @@ -262,7 +262,7 @@ static __be32 encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t size) { - uint32_t *p; + __be32 *p; if (!(bitmap[0] & FATTR4_WORD0_SIZE)) return 0; @@ -275,7 +275,7 @@ static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, u static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec *time) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 12); if (unlikely(p == 0)) @@ -317,7 +317,7 @@ static __be32 encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compound static __be32 encode_op_hdr(struct xdr_stream *xdr, uint32_t op, __be32 res) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 8); if (unlikely(p == NULL)) @@ -329,7 +329,7 @@ static __be32 encode_op_hdr(struct xdr_stream *xdr, uint32_t op, __be32 res) static __be32 encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, const struct cb_getattrres *res) { - uint32_t *savep = NULL; + __be32 *savep = NULL; __be32 status = res->status; if (unlikely(status != 0)) @@ -404,7 +404,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r struct cb_compound_hdr_arg hdr_arg; struct cb_compound_hdr_res hdr_res; struct xdr_stream xdr_in, xdr_out; - uint32_t *p; + __be32 *p; __be32 status; unsigned int nops = 1; @@ -412,7 +412,7 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r xdr_init_decode(&xdr_in, &rqstp->rq_arg, rqstp->rq_arg.head[0].iov_base); - p = (uint32_t*)((char *)rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len); + p = (__be32*)((char *)rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len); xdr_init_encode(&xdr_out, &rqstp->rq_res, p); decode_compound_hdr_arg(&xdr_in, &hdr_arg); -- cgit v0.10.2 From bc4785cd475a11ba125df7af674e16c6ea1cfc30 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:51 -0700 Subject: [PATCH] nfs: verifier is network-endian Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 3b234d4..e5f128f 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -668,7 +668,7 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, { struct inode *dir = dentry->d_inode; struct nfs_fattr dir_attr; - u32 *verf = NFS_COOKIEVERF(dir); + __be32 *verf = NFS_COOKIEVERF(dir); struct nfs3_readdirargs arg = { .fh = NFS_FH(dir), .cookie = cookie, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 8d09b47..8118036 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -138,7 +138,7 @@ const u32 nfs4_fs_locations_bitmap[2] = { | FATTR4_WORD1_MOUNTED_ON_FILEID }; -static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, +static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { __be32 *start, *p; @@ -2915,11 +2915,11 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po .rpc_resp = clp, .rpc_cred = cred, }; - u32 *p; + __be32 *p; int loop = 0; int status; - p = (u32*)sc_verifier.data; + p = (__be32*)sc_verifier.data; *p++ = htonl((u32)clp->cl_boot_time.tv_sec); *p = htonl((u32)clp->cl_boot_time.tv_nsec); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 6b2de1b..45228c1 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -157,7 +157,7 @@ struct nfs_inode { * This is the cookie verifier used for NFSv3 readdir * operations */ - __u32 cookieverf[2]; + __be32 cookieverf[2]; /* * This is the list of dirty unwritten pages. diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index ac8058b..768c1ad 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -266,7 +266,7 @@ struct nfs_writeargs { struct nfs_writeverf { enum nfs3_stable_how committed; - __u32 verifier[2]; + __be32 verifier[2]; }; struct nfs_writeres { @@ -420,7 +420,7 @@ struct nfs3_createargs { unsigned int len; struct iattr * sattr; enum nfs3_createmode createmode; - __u32 verifier[2]; + __be32 verifier[2]; }; struct nfs3_mkdirargs { @@ -467,7 +467,7 @@ struct nfs3_linkargs { struct nfs3_readdirargs { struct nfs_fh * fh; __u64 cookie; - __u32 verf[2]; + __be32 verf[2]; int plus; unsigned int count; struct page ** pages; @@ -503,7 +503,7 @@ struct nfs3_linkres { struct nfs3_readdirres { struct nfs_fattr * dir_attr; - __u32 * verf; + __be32 * verf; int plus; }; -- cgit v0.10.2 From d21ec0c33d0a9eb7a6f6c716008863a97797709e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:52 -0700 Subject: [PATCH] xdr annotations: mount_clnt [pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index d507b02..f75fe72 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -95,7 +95,7 @@ mnt_create(char *hostname, struct sockaddr_in *srvaddr, int version, * XDR encode/decode functions for MOUNT */ static int -xdr_encode_dirpath(struct rpc_rqst *req, u32 *p, const char *path) +xdr_encode_dirpath(struct rpc_rqst *req, __be32 *p, const char *path) { p = xdr_encode_string(p, path); @@ -104,7 +104,7 @@ xdr_encode_dirpath(struct rpc_rqst *req, u32 *p, const char *path) } static int -xdr_decode_fhstatus(struct rpc_rqst *req, u32 *p, struct mnt_fhstatus *res) +xdr_decode_fhstatus(struct rpc_rqst *req, __be32 *p, struct mnt_fhstatus *res) { struct nfs_fh *fh = res->fh; @@ -116,7 +116,7 @@ xdr_decode_fhstatus(struct rpc_rqst *req, u32 *p, struct mnt_fhstatus *res) } static int -xdr_decode_fhstatus3(struct rpc_rqst *req, u32 *p, struct mnt_fhstatus *res) +xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p, struct mnt_fhstatus *res) { struct nfs_fh *fh = res->fh; -- cgit v0.10.2 From 83bbe2ef63ec4f6a22aaaa0c03bd918b38300127 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:53 -0700 Subject: [PATCH] nfs_common endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index 0c2be8c..c11f537 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -46,7 +46,7 @@ xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem) { struct nfsacl_encode_desc *nfsacl_desc = (struct nfsacl_encode_desc *) desc; - u32 *p = (u32 *) elem; + __be32 *p = elem; struct posix_acl_entry *entry = &nfsacl_desc->acl->a_entries[nfsacl_desc->count++]; @@ -127,7 +127,7 @@ xdr_nfsace_decode(struct xdr_array2_desc *desc, void *elem) { struct nfsacl_decode_desc *nfsacl_desc = (struct nfsacl_decode_desc *) desc; - u32 *p = (u32 *) elem; + __be32 *p = elem; struct posix_acl_entry *entry; if (!nfsacl_desc->acl) { -- cgit v0.10.2 From 63f103111fdfc3cba00e4c94921d32362f375d93 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:54 -0700 Subject: [PATCH] nfsd: nfserrno() endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 09030af..03ab682 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -579,11 +579,11 @@ struct svc_version nfsd_version2 = { /* * Map errnos to NFS errnos. */ -int +__be32 nfserrno (int errno) { static struct { - int nfserr; + __be32 nfserr; int syserr; } nfs_errtbl[] = { { nfs_ok, 0 }, @@ -615,11 +615,10 @@ nfserrno (int errno) { nfserr_badname, -ESRCH }, { nfserr_io, -ETXTBSY }, { nfserr_notsupp, -EOPNOTSUPP }, - { -1, -EIO } }; int i; - for (i = 0; nfs_errtbl[i].nfserr != -1; i++) { + for (i = 0; i < ARRAY_SIZE(nfs_errtbl); i++) { if (nfs_errtbl[i].syserr == errno) return nfs_errtbl[i].nfserr; } diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h index 6e78ea9..27666f5b 100644 --- a/include/linux/nfsd/export.h +++ b/include/linux/nfsd/export.h @@ -118,7 +118,7 @@ struct svc_export * exp_parent(struct auth_domain *clp, int exp_rootfh(struct auth_domain *, char *path, struct knfsd_fh *, int maxsize); int exp_pseudoroot(struct auth_domain *, struct svc_fh *fhp, struct cache_req *creq); -int nfserrno(int errno); +__be32 nfserrno(int errno); extern struct cache_detail svc_export_cache; -- cgit v0.10.2 From 83b11340d683a67a77e35a5ffb5ad4afbf0be4e5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:55 -0700 Subject: [PATCH] nfsfh simple endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 501d838..727ab3b 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -76,7 +76,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) * comment in the NFSv3 spec says this is incorrect (implementation notes for * the write call). */ -static inline int +static inline __be32 nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type) { /* Type can be negative when creating hardlinks - not to a dir */ @@ -110,13 +110,13 @@ nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type) * This is only called at the start of an nfsproc call, so fhp points to * a svc_fh which is all 0 except for the over-the-wire file handle. */ -u32 +__be32 fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) { struct knfsd_fh *fh = &fhp->fh_handle; struct svc_export *exp = NULL; struct dentry *dentry; - u32 error = 0; + __be32 error = 0; dprintk("nfsd: fh_verify(%s)\n", SVCFH_fmt(fhp)); @@ -315,7 +315,7 @@ static inline void _fh_update_old(struct dentry *dentry, fh->ofh_dirino = 0; } -int +__be32 fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, struct svc_fh *ref_fh) { /* ref_fh is a reference file handle. @@ -451,7 +451,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, st * Update file handle information after changing a dentry. * This is only called by nfsd_create, nfsd_create_v3 and nfsd_proc_create */ -int +__be32 fh_update(struct svc_fh *fhp) { struct dentry *dentry; diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index 069257e..749bad1 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -209,9 +209,9 @@ extern char * SVCFH_fmt(struct svc_fh *fhp); /* * Function prototypes */ -u32 fh_verify(struct svc_rqst *, struct svc_fh *, int, int); -int fh_compose(struct svc_fh *, struct svc_export *, struct dentry *, struct svc_fh *); -int fh_update(struct svc_fh *); +__be32 fh_verify(struct svc_rqst *, struct svc_fh *, int, int); +__be32 fh_compose(struct svc_fh *, struct svc_export *, struct dentry *, struct svc_fh *); +__be32 fh_update(struct svc_fh *); void fh_put(struct svc_fh *); static __inline__ struct svc_fh * -- cgit v0.10.2 From ad451d389f46f699832da3e9ad95f610cb8c0fd2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:55 -0700 Subject: [PATCH] xdr annotations: nfsd_dispatch() Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 013b389..8067118 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -495,8 +495,8 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) { struct svc_procedure *proc; kxdrproc_t xdr; - u32 nfserr; - u32 *nfserrp; + __be32 nfserr; + __be32 *nfserrp; dprintk("nfsd_dispatch: vers %d proc %d\n", rqstp->rq_vers, rqstp->rq_proc); @@ -515,7 +515,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) /* Decode arguments */ xdr = proc->pc_decode; - if (xdr && !xdr(rqstp, (u32*)rqstp->rq_arg.head[0].iov_base, + if (xdr && !xdr(rqstp, (__be32*)rqstp->rq_arg.head[0].iov_base, rqstp->rq_argp)) { dprintk("nfsd: failed to decode arguments!\n"); nfsd_cache_update(rqstp, RC_NOCACHE, NULL); @@ -528,7 +528,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) */ nfserrp = rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len; - rqstp->rq_res.head[0].iov_len += sizeof(u32); + rqstp->rq_res.head[0].iov_len += sizeof(__be32); /* Now call the procedure handler, and encode NFS status. */ nfserr = proc->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp); -- cgit v0.10.2 From 131a21c2177c267ab259fcd06947c6f593a7de8e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:56 -0700 Subject: [PATCH] xdr annotations: NFSv2 server Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 8d48616..fd5397d 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -158,7 +158,7 @@ static __be32 nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessarg /* * XDR decode functions */ -static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclargs *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -169,7 +169,7 @@ static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, } -static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_setaclargs *argp) { struct kvec *head = rqstp->rq_arg.head; @@ -194,7 +194,7 @@ static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, return (n > 0); } -static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -202,7 +202,7 @@ static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, return xdr_argsize_check(rqstp, p); } -static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessargs *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -217,7 +217,7 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, */ /* GETACL */ -static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; @@ -259,7 +259,7 @@ static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, return 1; } -static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_attrstat *resp) { p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); @@ -267,7 +267,7 @@ static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, } /* ACCESS */ -static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessres *resp) { p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); @@ -278,7 +278,7 @@ static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, /* * XDR release functions */ -static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { fh_put(&resp->fh); @@ -287,7 +287,7 @@ static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, return 1; } -static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 1135c0d..56ebb14 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -37,8 +37,8 @@ static u32 nfs_ftypes[] = { /* * XDR functions for basic NFS types */ -static u32 * -decode_fh(u32 *p, struct svc_fh *fhp) +static __be32 * +decode_fh(__be32 *p, struct svc_fh *fhp) { fh_init(fhp, NFS_FHSIZE); memcpy(&fhp->fh_handle.fh_base, p, NFS_FHSIZE); @@ -50,13 +50,13 @@ decode_fh(u32 *p, struct svc_fh *fhp) } /* Helper function for NFSv2 ACL code */ -u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp) +__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp) { return decode_fh(p, fhp); } -static inline u32 * -encode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_fh(__be32 *p, struct svc_fh *fhp) { memcpy(p, &fhp->fh_handle.fh_base, NFS_FHSIZE); return p + (NFS_FHSIZE>> 2); @@ -66,8 +66,8 @@ encode_fh(u32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline u32 * -decode_filename(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_filename(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -82,8 +82,8 @@ decode_filename(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_pathname(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_pathname(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -98,8 +98,8 @@ decode_pathname(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_sattr(u32 *p, struct iattr *iap) +static inline __be32 * +decode_sattr(__be32 *p, struct iattr *iap) { u32 tmp, tmp1; @@ -151,8 +151,8 @@ decode_sattr(u32 *p, struct iattr *iap) return p; } -static u32 * -encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, +static __be32 * +encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { struct dentry *dentry = fhp->fh_dentry; @@ -195,7 +195,7 @@ encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, } /* Helper function for NFSv2 ACL code */ -u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +__be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct kstat stat; vfs_getattr(fhp->fh_export->ex_mnt, fhp->fh_dentry, &stat); @@ -206,13 +206,13 @@ u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * XDR decode functions */ int -nfssvc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfssvc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nfssvc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) +nfssvc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -220,7 +220,7 @@ nfssvc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) } int -nfssvc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_sattrargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -231,7 +231,7 @@ nfssvc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_diropargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -242,7 +242,7 @@ nfssvc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readargs *args) { unsigned int len; @@ -273,7 +273,7 @@ nfssvc_decode_readargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_writeargs *args) { unsigned int len; @@ -303,7 +303,7 @@ nfssvc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_createargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -315,7 +315,7 @@ nfssvc_decode_createargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_renameargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -328,7 +328,7 @@ nfssvc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, struct nfsd_readlinkargs *args) +nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readlinkargs *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -338,7 +338,7 @@ nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, struct nfsd_readlinka } int -nfssvc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_linkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -350,7 +350,7 @@ nfssvc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_symlinkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -363,7 +363,7 @@ nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readdirargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -382,13 +382,13 @@ nfssvc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, * XDR encode functions */ int -nfssvc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfssvc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } int -nfssvc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_attrstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd_attrstat *resp) { p = encode_fattr(rqstp, p, &resp->fh, &resp->stat); @@ -396,7 +396,7 @@ nfssvc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_diropres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_diropres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_diropres *resp) { p = encode_fh(p, &resp->fh); @@ -405,7 +405,7 @@ nfssvc_encode_diropres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readlinkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readlinkres *resp) { *p++ = htonl(resp->len); @@ -421,7 +421,7 @@ nfssvc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readres *resp) { p = encode_fattr(rqstp, p, &resp->fh, &resp->stat); @@ -440,7 +440,7 @@ nfssvc_encode_readres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readdirres *resp) { xdr_ressize_check(rqstp, p); @@ -453,7 +453,7 @@ nfssvc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_statfsres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_statfsres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_statfsres *resp) { struct kstatfs *stat = &resp->stats; @@ -471,7 +471,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int d_type) { struct nfsd_readdirres *cd = container_of(ccd, struct nfsd_readdirres, common); - u32 *p = cd->buffer; + __be32 *p = cd->buffer; int buflen, slen; /* @@ -497,7 +497,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, *p++ = htonl((u32) ino); /* file id */ p = xdr_encode_array(p, name, namlen);/* name length & name */ cd->offset = p; /* remember pointer */ - *p++ = ~(u32) 0; /* offset of next entry */ + *p++ = htonl(~0U); /* offset of next entry */ cd->buflen = buflen; cd->buffer = p; @@ -509,7 +509,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, * XDR release functions */ int -nfssvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +nfssvc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *resp) { fh_put(&resp->fh); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index d0d4aae..2f75160 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -50,7 +50,7 @@ * Callback function for readdir */ struct readdir_cd { - int err; /* 0, nfserr, or nfserr_eof */ + __be32 err; /* 0, nfserr, or nfserr_eof */ }; typedef int (*encode_dent_fn)(struct readdir_cd *, const char *, int, loff_t, ino_t, unsigned int); diff --git a/include/linux/nfsd/xdr.h b/include/linux/nfsd/xdr.h index 0e53de8..877192d 100644 --- a/include/linux/nfsd/xdr.h +++ b/include/linux/nfsd/xdr.h @@ -81,7 +81,7 @@ struct nfsd_readdirargs { struct svc_fh fh; __u32 cookie; __u32 count; - u32 * buffer; + __be32 * buffer; }; struct nfsd_attrstat { @@ -108,9 +108,9 @@ struct nfsd_readdirres { int count; struct readdir_cd common; - u32 * buffer; + __be32 * buffer; int buflen; - u32 * offset; + __be32 * offset; }; struct nfsd_statfsres { @@ -135,43 +135,43 @@ union nfsd_xdrstore { #define NFS2_SVC_XDRSIZE sizeof(union nfsd_xdrstore) -int nfssvc_decode_void(struct svc_rqst *, u32 *, void *); -int nfssvc_decode_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); -int nfssvc_decode_sattrargs(struct svc_rqst *, u32 *, +int nfssvc_decode_void(struct svc_rqst *, __be32 *, void *); +int nfssvc_decode_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); +int nfssvc_decode_sattrargs(struct svc_rqst *, __be32 *, struct nfsd_sattrargs *); -int nfssvc_decode_diropargs(struct svc_rqst *, u32 *, +int nfssvc_decode_diropargs(struct svc_rqst *, __be32 *, struct nfsd_diropargs *); -int nfssvc_decode_readargs(struct svc_rqst *, u32 *, +int nfssvc_decode_readargs(struct svc_rqst *, __be32 *, struct nfsd_readargs *); -int nfssvc_decode_writeargs(struct svc_rqst *, u32 *, +int nfssvc_decode_writeargs(struct svc_rqst *, __be32 *, struct nfsd_writeargs *); -int nfssvc_decode_createargs(struct svc_rqst *, u32 *, +int nfssvc_decode_createargs(struct svc_rqst *, __be32 *, struct nfsd_createargs *); -int nfssvc_decode_renameargs(struct svc_rqst *, u32 *, +int nfssvc_decode_renameargs(struct svc_rqst *, __be32 *, struct nfsd_renameargs *); -int nfssvc_decode_readlinkargs(struct svc_rqst *, u32 *, +int nfssvc_decode_readlinkargs(struct svc_rqst *, __be32 *, struct nfsd_readlinkargs *); -int nfssvc_decode_linkargs(struct svc_rqst *, u32 *, +int nfssvc_decode_linkargs(struct svc_rqst *, __be32 *, struct nfsd_linkargs *); -int nfssvc_decode_symlinkargs(struct svc_rqst *, u32 *, +int nfssvc_decode_symlinkargs(struct svc_rqst *, __be32 *, struct nfsd_symlinkargs *); -int nfssvc_decode_readdirargs(struct svc_rqst *, u32 *, +int nfssvc_decode_readdirargs(struct svc_rqst *, __be32 *, struct nfsd_readdirargs *); -int nfssvc_encode_void(struct svc_rqst *, u32 *, void *); -int nfssvc_encode_attrstat(struct svc_rqst *, u32 *, struct nfsd_attrstat *); -int nfssvc_encode_diropres(struct svc_rqst *, u32 *, struct nfsd_diropres *); -int nfssvc_encode_readlinkres(struct svc_rqst *, u32 *, struct nfsd_readlinkres *); -int nfssvc_encode_readres(struct svc_rqst *, u32 *, struct nfsd_readres *); -int nfssvc_encode_statfsres(struct svc_rqst *, u32 *, struct nfsd_statfsres *); -int nfssvc_encode_readdirres(struct svc_rqst *, u32 *, struct nfsd_readdirres *); +int nfssvc_encode_void(struct svc_rqst *, __be32 *, void *); +int nfssvc_encode_attrstat(struct svc_rqst *, __be32 *, struct nfsd_attrstat *); +int nfssvc_encode_diropres(struct svc_rqst *, __be32 *, struct nfsd_diropres *); +int nfssvc_encode_readlinkres(struct svc_rqst *, __be32 *, struct nfsd_readlinkres *); +int nfssvc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd_readres *); +int nfssvc_encode_statfsres(struct svc_rqst *, __be32 *, struct nfsd_statfsres *); +int nfssvc_encode_readdirres(struct svc_rqst *, __be32 *, struct nfsd_readdirres *); int nfssvc_encode_entry(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int); -int nfssvc_release_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); +int nfssvc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); /* Helper functions for NFSv2 ACL code */ -u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp); -u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp); +__be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp); +__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp); #endif /* LINUX_NFSD_H */ -- cgit v0.10.2 From 91f07168cef8e99dd16f608fbc703e7a5af0237f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:57 -0700 Subject: [PATCH] xdr annotations: NFSv3 server Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index ed6e2c2..78b2c83 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -122,7 +122,7 @@ static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp, /* * XDR decode functions */ -static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclargs *args) { if (!(p = nfs3svc_decode_fh(p, &args->fh))) @@ -133,7 +133,7 @@ static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, } -static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_setaclargs *args) { struct kvec *head = rqstp->rq_arg.head; @@ -163,7 +163,7 @@ static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, */ /* GETACL */ -static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; @@ -208,7 +208,7 @@ static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, } /* SETACL */ -static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh); @@ -219,7 +219,7 @@ static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, /* * XDR release functions */ -static int nfs3svc_release_getacl(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_release_getacl(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 247d518..b4baca3 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -42,23 +42,23 @@ static u32 nfs3_ftypes[] = { /* * XDR functions for basic NFS types */ -static inline u32 * -encode_time3(u32 *p, struct timespec *time) +static inline __be32 * +encode_time3(__be32 *p, struct timespec *time) { *p++ = htonl((u32) time->tv_sec); *p++ = htonl(time->tv_nsec); return p; } -static inline u32 * -decode_time3(u32 *p, struct timespec *time) +static inline __be32 * +decode_time3(__be32 *p, struct timespec *time) { time->tv_sec = ntohl(*p++); time->tv_nsec = ntohl(*p++); return p; } -static inline u32 * -decode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +decode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size; fh_init(fhp, NFS3_FHSIZE); @@ -72,13 +72,13 @@ decode_fh(u32 *p, struct svc_fh *fhp) } /* Helper function for NFSv3 ACL code */ -u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp) +__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp) { return decode_fh(p, fhp); } -static inline u32 * -encode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size = fhp->fh_handle.fh_size; *p++ = htonl(size); @@ -91,8 +91,8 @@ encode_fh(u32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline u32 * -decode_filename(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_filename(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -107,8 +107,8 @@ decode_filename(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_sattr3(u32 *p, struct iattr *iap) +static inline __be32 * +decode_sattr3(__be32 *p, struct iattr *iap) { u32 tmp; @@ -153,8 +153,8 @@ decode_sattr3(u32 *p, struct iattr *iap) return p; } -static inline u32 * -encode_fattr3(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, +static inline __be32 * +encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { struct dentry *dentry = fhp->fh_dentry; @@ -186,8 +186,8 @@ encode_fattr3(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, return p; } -static inline u32 * -encode_saved_post_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_saved_post_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct inode *inode = fhp->fh_dentry->d_inode; @@ -224,8 +224,8 @@ encode_saved_post_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * The inode may be NULL if the call failed because of a stale file * handle. In this case, no attributes are returned. */ -static u32 * -encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static __be32 * +encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; if (dentry && dentry->d_inode != NULL) { @@ -243,8 +243,8 @@ encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) } /* Helper for NFSv3 ACLs */ -u32 * -nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +__be32 * +nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { return encode_post_op_attr(rqstp, p, fhp); } @@ -252,8 +252,8 @@ nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) /* * Enocde weak cache consistency data */ -static u32 * -encode_wcc_data(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static __be32 * +encode_wcc_data(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; @@ -278,7 +278,7 @@ encode_wcc_data(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * XDR decode functions */ int -nfs3svc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) +nfs3svc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -286,7 +286,7 @@ nfs3svc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args } int -nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_sattrargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -303,7 +303,7 @@ nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -314,7 +314,7 @@ nfs3svc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -325,7 +325,7 @@ nfs3svc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readargs *args) { unsigned int len; @@ -355,7 +355,7 @@ nfs3svc_decode_readargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeargs *args) { unsigned int len, v, hdr; @@ -393,7 +393,7 @@ nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_createargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -417,7 +417,7 @@ nfs3svc_decode_createargs(struct svc_rqst *rqstp, u32 *p, return xdr_argsize_check(rqstp, p); } int -nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -429,7 +429,7 @@ nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_symlinkargs *args) { unsigned int len; @@ -481,7 +481,7 @@ nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_mknodargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -505,7 +505,7 @@ nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_renameargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -518,7 +518,7 @@ nfs3svc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readlinkargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -530,7 +530,7 @@ nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_linkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -542,7 +542,7 @@ nfs3svc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -562,7 +562,7 @@ nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirargs *args) { int len, pn; @@ -590,7 +590,7 @@ nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_commitargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_commitargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_commitargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -609,14 +609,14 @@ nfs3svc_decode_commitargs(struct svc_rqst *rqstp, u32 *p, * will work properly. */ int -nfs3svc_encode_voidres(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfs3svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } /* GETATTR */ int -nfs3svc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_attrstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { if (resp->status == 0) @@ -626,7 +626,7 @@ nfs3svc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, /* SETATTR, REMOVE, RMDIR */ int -nfs3svc_encode_wccstat(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_wccstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -635,7 +635,7 @@ nfs3svc_encode_wccstat(struct svc_rqst *rqstp, u32 *p, /* LOOKUP */ int -nfs3svc_encode_diropres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_diropres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropres *resp) { if (resp->status == 0) { @@ -648,7 +648,7 @@ nfs3svc_encode_diropres(struct svc_rqst *rqstp, u32 *p, /* ACCESS */ int -nfs3svc_encode_accessres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_accessres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -659,7 +659,7 @@ nfs3svc_encode_accessres(struct svc_rqst *rqstp, u32 *p, /* READLINK */ int -nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readlinkres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -680,7 +680,7 @@ nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, /* READ */ int -nfs3svc_encode_readres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -704,7 +704,7 @@ nfs3svc_encode_readres(struct svc_rqst *rqstp, u32 *p, /* WRITE */ int -nfs3svc_encode_writeres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_writeres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeres *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -719,7 +719,7 @@ nfs3svc_encode_writeres(struct svc_rqst *rqstp, u32 *p, /* CREATE, MKDIR, SYMLINK, MKNOD */ int -nfs3svc_encode_createres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_createres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropres *resp) { if (resp->status == 0) { @@ -733,7 +733,7 @@ nfs3svc_encode_createres(struct svc_rqst *rqstp, u32 *p, /* RENAME */ int -nfs3svc_encode_renameres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_renameres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_renameres *resp) { p = encode_wcc_data(rqstp, p, &resp->ffh); @@ -743,7 +743,7 @@ nfs3svc_encode_renameres(struct svc_rqst *rqstp, u32 *p, /* LINK */ int -nfs3svc_encode_linkres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_linkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_linkres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -753,7 +753,7 @@ nfs3svc_encode_linkres(struct svc_rqst *rqstp, u32 *p, /* READDIR */ int -nfs3svc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -776,8 +776,8 @@ nfs3svc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, return xdr_ressize_check(rqstp, p); } -static inline u32 * -encode_entry_baggage(struct nfsd3_readdirres *cd, u32 *p, const char *name, +static inline __be32 * +encode_entry_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen, ino_t ino) { *p++ = xdr_one; /* mark entry present */ @@ -790,8 +790,8 @@ encode_entry_baggage(struct nfsd3_readdirres *cd, u32 *p, const char *name, return p; } -static inline u32 * -encode_entryplus_baggage(struct nfsd3_readdirres *cd, u32 *p, +static inline __be32 * +encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, struct svc_fh *fhp) { p = encode_post_op_attr(cd->rqstp, p, fhp); @@ -853,7 +853,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, { struct nfsd3_readdirres *cd = container_of(ccd, struct nfsd3_readdirres, common); - u32 *p = cd->buffer; + __be32 *p = cd->buffer; caddr_t curr_page_addr = NULL; int pn; /* current page number */ int slen; /* string (name) length */ @@ -919,7 +919,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, } else if (cd->rqstp->rq_respages[pn+1] != NULL) { /* temporarily encode entry into next page, then move back to * current and next page in rq_respages[] */ - u32 *p1, *tmp; + __be32 *p1, *tmp; int len1, len2; /* grab next page for temporary storage of entry */ @@ -1009,7 +1009,7 @@ nfs3svc_encode_entry_plus(struct readdir_cd *cd, const char *name, /* FSSTAT */ int -nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fsstatres *resp) { struct kstatfs *s = &resp->stats; @@ -1031,7 +1031,7 @@ nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, u32 *p, /* FSINFO */ int -nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fsinfores *resp) { *p++ = xdr_zero; /* no post_op_attr */ @@ -1055,7 +1055,7 @@ nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, u32 *p, /* PATHCONF */ int -nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_pathconfres *resp) { *p++ = xdr_zero; /* no post_op_attr */ @@ -1074,7 +1074,7 @@ nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, u32 *p, /* COMMIT */ int -nfs3svc_encode_commitres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_commitres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_commitres *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -1090,7 +1090,7 @@ nfs3svc_encode_commitres(struct svc_rqst *rqstp, u32 *p, * XDR release functions */ int -nfs3svc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +nfs3svc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { fh_put(&resp->fh); @@ -1098,7 +1098,7 @@ nfs3svc_release_fhandle(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_release_fhandle2(struct svc_rqst *rqstp, u32 *p, +nfs3svc_release_fhandle2(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fhandle_pair *resp) { fh_put(&resp->fh1); diff --git a/include/linux/nfsd/xdr3.h b/include/linux/nfsd/xdr3.h index 474d882..7996386 100644 --- a/include/linux/nfsd/xdr3.h +++ b/include/linux/nfsd/xdr3.h @@ -51,7 +51,7 @@ struct nfsd3_createargs { int len; int createmode; struct iattr attrs; - __u32 * verf; + __be32 * verf; }; struct nfsd3_mknodargs { @@ -98,8 +98,8 @@ struct nfsd3_readdirargs { __u64 cookie; __u32 dircount; __u32 count; - __u32 * verf; - u32 * buffer; + __be32 * verf; + __be32 * buffer; }; struct nfsd3_commitargs { @@ -122,79 +122,79 @@ struct nfsd3_setaclargs { }; struct nfsd3_attrstat { - __u32 status; + __be32 status; struct svc_fh fh; struct kstat stat; }; /* LOOKUP, CREATE, MKDIR, SYMLINK, MKNOD */ struct nfsd3_diropres { - __u32 status; + __be32 status; struct svc_fh dirfh; struct svc_fh fh; }; struct nfsd3_accessres { - __u32 status; + __be32 status; struct svc_fh fh; __u32 access; }; struct nfsd3_readlinkres { - __u32 status; + __be32 status; struct svc_fh fh; __u32 len; }; struct nfsd3_readres { - __u32 status; + __be32 status; struct svc_fh fh; unsigned long count; int eof; }; struct nfsd3_writeres { - __u32 status; + __be32 status; struct svc_fh fh; unsigned long count; int committed; }; struct nfsd3_renameres { - __u32 status; + __be32 status; struct svc_fh ffh; struct svc_fh tfh; }; struct nfsd3_linkres { - __u32 status; + __be32 status; struct svc_fh tfh; struct svc_fh fh; }; struct nfsd3_readdirres { - __u32 status; + __be32 status; struct svc_fh fh; int count; - __u32 verf[2]; + __be32 verf[2]; struct readdir_cd common; - u32 * buffer; + __be32 * buffer; int buflen; - u32 * offset; - u32 * offset1; + __be32 * offset; + __be32 * offset1; struct svc_rqst * rqstp; }; struct nfsd3_fsstatres { - __u32 status; + __be32 status; struct kstatfs stats; __u32 invarsec; }; struct nfsd3_fsinfores { - __u32 status; + __be32 status; __u32 f_rtmax; __u32 f_rtpref; __u32 f_rtmult; @@ -207,7 +207,7 @@ struct nfsd3_fsinfores { }; struct nfsd3_pathconfres { - __u32 status; + __be32 status; __u32 p_link_max; __u32 p_name_max; __u32 p_no_trunc; @@ -217,12 +217,12 @@ struct nfsd3_pathconfres { }; struct nfsd3_commitres { - __u32 status; + __be32 status; struct svc_fh fh; }; struct nfsd3_getaclres { - __u32 status; + __be32 status; struct svc_fh fh; int mask; struct posix_acl *acl_access; @@ -266,70 +266,70 @@ union nfsd3_xdrstore { #define NFS3_SVC_XDRSIZE sizeof(union nfsd3_xdrstore) -int nfs3svc_decode_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); -int nfs3svc_decode_sattrargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); +int nfs3svc_decode_sattrargs(struct svc_rqst *, __be32 *, struct nfsd3_sattrargs *); -int nfs3svc_decode_diropargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_diropargs(struct svc_rqst *, __be32 *, struct nfsd3_diropargs *); -int nfs3svc_decode_accessargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_accessargs(struct svc_rqst *, __be32 *, struct nfsd3_accessargs *); -int nfs3svc_decode_readargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readargs(struct svc_rqst *, __be32 *, struct nfsd3_readargs *); -int nfs3svc_decode_writeargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_writeargs(struct svc_rqst *, __be32 *, struct nfsd3_writeargs *); -int nfs3svc_decode_createargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_createargs(struct svc_rqst *, __be32 *, struct nfsd3_createargs *); -int nfs3svc_decode_mkdirargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_mkdirargs(struct svc_rqst *, __be32 *, struct nfsd3_createargs *); -int nfs3svc_decode_mknodargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_mknodargs(struct svc_rqst *, __be32 *, struct nfsd3_mknodargs *); -int nfs3svc_decode_renameargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_renameargs(struct svc_rqst *, __be32 *, struct nfsd3_renameargs *); -int nfs3svc_decode_readlinkargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readlinkargs(struct svc_rqst *, __be32 *, struct nfsd3_readlinkargs *); -int nfs3svc_decode_linkargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_linkargs(struct svc_rqst *, __be32 *, struct nfsd3_linkargs *); -int nfs3svc_decode_symlinkargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_symlinkargs(struct svc_rqst *, __be32 *, struct nfsd3_symlinkargs *); -int nfs3svc_decode_readdirargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readdirargs(struct svc_rqst *, __be32 *, struct nfsd3_readdirargs *); -int nfs3svc_decode_readdirplusargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readdirplusargs(struct svc_rqst *, __be32 *, struct nfsd3_readdirargs *); -int nfs3svc_decode_commitargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_commitargs(struct svc_rqst *, __be32 *, struct nfsd3_commitargs *); -int nfs3svc_encode_voidres(struct svc_rqst *, u32 *, void *); -int nfs3svc_encode_attrstat(struct svc_rqst *, u32 *, +int nfs3svc_encode_voidres(struct svc_rqst *, __be32 *, void *); +int nfs3svc_encode_attrstat(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); -int nfs3svc_encode_wccstat(struct svc_rqst *, u32 *, +int nfs3svc_encode_wccstat(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); -int nfs3svc_encode_diropres(struct svc_rqst *, u32 *, +int nfs3svc_encode_diropres(struct svc_rqst *, __be32 *, struct nfsd3_diropres *); -int nfs3svc_encode_accessres(struct svc_rqst *, u32 *, +int nfs3svc_encode_accessres(struct svc_rqst *, __be32 *, struct nfsd3_accessres *); -int nfs3svc_encode_readlinkres(struct svc_rqst *, u32 *, +int nfs3svc_encode_readlinkres(struct svc_rqst *, __be32 *, struct nfsd3_readlinkres *); -int nfs3svc_encode_readres(struct svc_rqst *, u32 *, struct nfsd3_readres *); -int nfs3svc_encode_writeres(struct svc_rqst *, u32 *, struct nfsd3_writeres *); -int nfs3svc_encode_createres(struct svc_rqst *, u32 *, +int nfs3svc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd3_readres *); +int nfs3svc_encode_writeres(struct svc_rqst *, __be32 *, struct nfsd3_writeres *); +int nfs3svc_encode_createres(struct svc_rqst *, __be32 *, struct nfsd3_diropres *); -int nfs3svc_encode_renameres(struct svc_rqst *, u32 *, +int nfs3svc_encode_renameres(struct svc_rqst *, __be32 *, struct nfsd3_renameres *); -int nfs3svc_encode_linkres(struct svc_rqst *, u32 *, +int nfs3svc_encode_linkres(struct svc_rqst *, __be32 *, struct nfsd3_linkres *); -int nfs3svc_encode_readdirres(struct svc_rqst *, u32 *, +int nfs3svc_encode_readdirres(struct svc_rqst *, __be32 *, struct nfsd3_readdirres *); -int nfs3svc_encode_fsstatres(struct svc_rqst *, u32 *, +int nfs3svc_encode_fsstatres(struct svc_rqst *, __be32 *, struct nfsd3_fsstatres *); -int nfs3svc_encode_fsinfores(struct svc_rqst *, u32 *, +int nfs3svc_encode_fsinfores(struct svc_rqst *, __be32 *, struct nfsd3_fsinfores *); -int nfs3svc_encode_pathconfres(struct svc_rqst *, u32 *, +int nfs3svc_encode_pathconfres(struct svc_rqst *, __be32 *, struct nfsd3_pathconfres *); -int nfs3svc_encode_commitres(struct svc_rqst *, u32 *, +int nfs3svc_encode_commitres(struct svc_rqst *, __be32 *, struct nfsd3_commitres *); -int nfs3svc_release_fhandle(struct svc_rqst *, u32 *, +int nfs3svc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); -int nfs3svc_release_fhandle2(struct svc_rqst *, u32 *, +int nfs3svc_release_fhandle2(struct svc_rqst *, __be32 *, struct nfsd3_fhandle_pair *); int nfs3svc_encode_entry(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, @@ -338,9 +338,9 @@ int nfs3svc_encode_entry_plus(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int); /* Helper functions for NFSv3 ACL code */ -u32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, +__be32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp); -u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp); +__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp); #endif /* _LINUX_NFSD_XDR3_H */ -- cgit v0.10.2 From 2ebbc012a9433a252be7ab4ce54e94bf7b21e506 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:58 -0700 Subject: [PATCH] xdr annotations: NFSv4 server Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 795ad6c5..ca64142 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -664,7 +664,7 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ static int nfsd4_verify(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_verify *verify) { - u32 *buf, *p; + __be32 *buf, *p; int count; int status; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 77be0c4..3419d99 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -94,7 +94,7 @@ check_filename(char *str, int len, int err) * consistent with the style used in NFSv2/v3... */ #define DECODE_HEAD \ - u32 *p; \ + __be32 *p; \ int status #define DECODE_TAIL \ status = 0; \ @@ -144,13 +144,13 @@ xdr_error: \ } \ } while (0) -static u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) +static __be32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) { /* We want more bytes than seem to be available. * Maybe we need a new page, maybe we have just run out */ int avail = (char*)argp->end - (char*)argp->p; - u32 *p; + __be32 *p; if (avail + argp->pagelen < nbytes) return NULL; if (avail + PAGE_SIZE < nbytes) /* need more than a page !! */ @@ -197,7 +197,7 @@ defer_free(struct nfsd4_compoundargs *argp, return 0; } -static char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes) +static char *savemem(struct nfsd4_compoundargs *argp, __be32 *p, int nbytes) { void *new = NULL; if (p == argp->tmp) { @@ -951,8 +951,8 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) argp->pagelen -= len; } } - argp->end = (u32*) (argp->rqstp->rq_vec[v].iov_base + argp->rqstp->rq_vec[v].iov_len); - argp->p = (u32*) (argp->rqstp->rq_vec[v].iov_base + (XDR_QUADLEN(len) << 2)); + argp->end = (__be32*) (argp->rqstp->rq_vec[v].iov_base + argp->rqstp->rq_vec[v].iov_len); + argp->p = (__be32*) (argp->rqstp->rq_vec[v].iov_base + (XDR_QUADLEN(len) << 2)); argp->rqstp->rq_vec[v].iov_len = len; write->wr_vlen = v+1; @@ -1179,7 +1179,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * task to translate them into Linux-specific versions which are more * consistent with the style used in NFSv2/v3... */ -#define ENCODE_HEAD u32 *p +#define ENCODE_HEAD __be32 *p #define WRITE32(n) *p++ = htonl(n) #define WRITE64(n) do { \ @@ -1209,8 +1209,8 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * Header routine to setup seqid operation replay cache */ #define ENCODE_SEQID_OP_HEAD \ - u32 *p; \ - u32 *save; \ + __be32 *p; \ + __be32 *save; \ \ save = resp->p; @@ -1235,10 +1235,10 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * seperated @sep. */ static int nfsd4_encode_components(char sep, char *components, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { - u32 *p = *pp; - u32 *countp = p; + __be32 *p = *pp; + __be32 *countp = p; int strlen, count=0; char *str, *end; @@ -1272,10 +1272,10 @@ static int nfsd4_encode_components(char sep, char *components, * encode a location element of a fs_locations structure */ static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { int status; - u32 *p = *pp; + __be32 *p = *pp; status = nfsd4_encode_components(':', location->hosts, &p, buflen); if (status) @@ -1320,11 +1320,11 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *sta */ static int nfsd4_encode_fs_locations(struct svc_rqst *rqstp, struct svc_export *exp, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { u32 status; int i; - u32 *p = *pp; + __be32 *p = *pp; struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs; char *root = nfsd4_path(rqstp, exp, &status); @@ -1355,7 +1355,7 @@ static u32 nfs4_ftypes[16] = { static int nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, - u32 **p, int *buflen) + __be32 **p, int *buflen) { int status; @@ -1376,20 +1376,20 @@ nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, } static inline int -nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, u32 **p, int *buflen) +nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, uid, 0, p, buflen); } static inline int -nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, u32 **p, int *buflen) +nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, gid, 1, p, buflen); } static inline int nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, - u32 **p, int *buflen) + __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, whotype, id, group, p, buflen); } @@ -1423,7 +1423,7 @@ static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) */ int nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, - struct dentry *dentry, u32 *buffer, int *countp, u32 *bmval, + struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *rqstp) { u32 bmval0 = bmval[0]; @@ -1432,11 +1432,11 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, struct svc_fh tempfh; struct kstatfs statfs; int buflen = *countp << 2; - u32 *attrlenp; + __be32 *attrlenp; u32 dummy; u64 dummy64; u32 rdattr_err = 0; - u32 *p = buffer; + __be32 *p = buffer; int status; int aclsupport = 0; struct nfs4_acl *acl = NULL; @@ -1831,7 +1831,7 @@ out_serverfault: static int nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, - const char *name, int namlen, u32 *p, int *buflen) + const char *name, int namlen, __be32 *p, int *buflen) { struct svc_export *exp = cd->rd_fhp->fh_export; struct dentry *dentry; @@ -1864,10 +1864,10 @@ out_put: return nfserr; } -static u32 * -nfsd4_encode_rdattr_error(u32 *p, int buflen, int nfserr) +static __be32 * +nfsd4_encode_rdattr_error(__be32 *p, int buflen, int nfserr) { - u32 *attrlenp; + __be32 *attrlenp; if (buflen < 6) return NULL; @@ -1887,7 +1887,7 @@ nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, { struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; - u32 *p = cd->buffer; + __be32 *p = cd->buffer; int nfserr = nfserr_toosmall; /* In nfsv4, "." and ".." never make it onto the wire.. */ @@ -2321,7 +2321,7 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_re { int maxcount; loff_t offset; - u32 *page, *savep, *tailbase; + __be32 *page, *savep, *tailbase; ENCODE_HEAD; if (nfserr) @@ -2479,7 +2479,7 @@ nfsd4_encode_write(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_writ void nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) { - u32 *statp; + __be32 *statp; ENCODE_HEAD; RESERVE_SPACE(8); @@ -2617,7 +2617,7 @@ nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op) */ int -nfs4svc_encode_voidres(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfs4svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -2639,7 +2639,7 @@ void nfsd4_release_compoundargs(struct nfsd4_compoundargs *args) } int -nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoundargs *args) +nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundargs *args) { int status; @@ -2660,7 +2660,7 @@ nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoun } int -nfs4svc_encode_compoundres(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoundres *resp) +nfs4svc_encode_compoundres(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundres *resp) { /* * All that remains is to write the tag and operation count... diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index 66e6427..003193f 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -258,9 +258,9 @@ struct nfsd4_readdir { struct svc_fh * rd_fhp; /* response */ struct readdir_cd common; - u32 * buffer; + __be32 * buffer; int buflen; - u32 * offset; + __be32 * offset; }; struct nfsd4_release_lockowner { @@ -371,12 +371,12 @@ struct nfsd4_op { struct nfsd4_compoundargs { /* scratch variables for XDR decode */ - u32 * p; - u32 * end; + __be32 * p; + __be32 * end; struct page ** pagelist; int pagelen; - u32 tmp[8]; - u32 * tmpp; + __be32 tmp[8]; + __be32 * tmpp; struct tmpbuf { struct tmpbuf *next; void (*release)(const void *); @@ -395,15 +395,15 @@ struct nfsd4_compoundargs { struct nfsd4_compoundres { /* scratch variables for XDR encode */ - u32 * p; - u32 * end; + __be32 * p; + __be32 * end; struct xdr_buf * xbuf; struct svc_rqst * rqstp; u32 taglen; char * tag; u32 opcnt; - u32 * tagp; /* where to encode tag and opcount */ + __be32 * tagp; /* where to encode tag and opcount */ }; #define NFS4_SVC_XDRSIZE sizeof(struct nfsd4_compoundargs) @@ -419,10 +419,10 @@ set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp) cinfo->after_ctime_nsec = fhp->fh_post_ctime.tv_nsec; } -int nfs4svc_encode_voidres(struct svc_rqst *, u32 *, void *); -int nfs4svc_decode_compoundargs(struct svc_rqst *, u32 *, +int nfs4svc_encode_voidres(struct svc_rqst *, __be32 *, void *); +int nfs4svc_decode_compoundargs(struct svc_rqst *, __be32 *, struct nfsd4_compoundargs *); -int nfs4svc_encode_compoundres(struct svc_rqst *, u32 *, +int nfs4svc_encode_compoundres(struct svc_rqst *, __be32 *, struct nfsd4_compoundres *); void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *); void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op); -- cgit v0.10.2 From 6264d69d7df654ca64f625e9409189a0e50734e9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:58 -0700 Subject: [PATCH] nfsd: vfs.c endianness annotations don't use the same variable to store NFS and host error values Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 1141bd2..f21e917 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -110,7 +110,7 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, struct dentry *dentry = *dpp; struct vfsmount *mnt = mntget(exp->ex_mnt); struct dentry *mounts = dget(dentry); - int err = nfs_ok; + int err = 0; while (follow_down(&mnt,&mounts)&&d_mountpoint(mounts)); @@ -148,14 +148,15 @@ out: * clients and is explicitly disallowed for NFSv3 * NeilBrown */ -int +__be32 nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, int len, struct svc_fh *resfh) { struct svc_export *exp; struct dentry *dparent; struct dentry *dentry; - int err; + __be32 err; + int host_err; dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name); @@ -193,7 +194,7 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, exp2 = exp_parent(exp->ex_client, mnt, dentry, &rqstp->rq_chandle); if (IS_ERR(exp2)) { - err = PTR_ERR(exp2); + host_err = PTR_ERR(exp2); dput(dentry); mntput(mnt); goto out_nfserr; @@ -210,14 +211,14 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, } else { fh_lock(fhp); dentry = lookup_one_len(name, dparent, len); - err = PTR_ERR(dentry); + host_err = PTR_ERR(dentry); if (IS_ERR(dentry)) goto out_nfserr; /* * check if we have crossed a mount point ... */ if (d_mountpoint(dentry)) { - if ((err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { + if ((host_err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { dput(dentry); goto out_nfserr; } @@ -236,7 +237,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -244,7 +245,7 @@ out_nfserr: * Set various file attributes. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, int check_guard, time_t guardtime) { @@ -253,7 +254,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, int accmode = MAY_SATTR; int ftype = 0; int imode; - int err; + __be32 err; + int host_err; int size_change = 0; if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE)) @@ -319,19 +321,19 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, * If we are changing the size of the file, then * we need to break all leases. */ - err = break_lease(inode, FMODE_WRITE | O_NONBLOCK); - if (err == -EWOULDBLOCK) - err = -ETIMEDOUT; - if (err) /* ENOMEM or EWOULDBLOCK */ + host_err = break_lease(inode, FMODE_WRITE | O_NONBLOCK); + if (host_err == -EWOULDBLOCK) + host_err = -ETIMEDOUT; + if (host_err) /* ENOMEM or EWOULDBLOCK */ goto out_nfserr; - err = get_write_access(inode); - if (err) + host_err = get_write_access(inode); + if (host_err) goto out_nfserr; size_change = 1; - err = locks_verify_truncate(inode, NULL, iap->ia_size); - if (err) { + host_err = locks_verify_truncate(inode, NULL, iap->ia_size); + if (host_err) { put_write_access(inode); goto out_nfserr; } @@ -357,8 +359,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, err = nfserr_notsync; if (!check_guard || guardtime == inode->i_ctime.tv_sec) { fh_lock(fhp); - err = notify_change(dentry, iap); - err = nfserrno(err); + host_err = notify_change(dentry, iap); + err = nfserrno(host_err); fh_unlock(fhp); } if (size_change) @@ -370,7 +372,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -420,11 +422,12 @@ out: return error; } -int +__be32 nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfs4_acl *acl) { - int error; + __be32 error; + int host_error; struct dentry *dentry; struct inode *inode; struct posix_acl *pacl = NULL, *dpacl = NULL; @@ -440,20 +443,20 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, if (S_ISDIR(inode->i_mode)) flags = NFS4_ACL_DIR; - error = nfs4_acl_nfsv4_to_posix(acl, &pacl, &dpacl, flags); - if (error == -EINVAL) { + host_error = nfs4_acl_nfsv4_to_posix(acl, &pacl, &dpacl, flags); + if (host_error == -EINVAL) { error = nfserr_attrnotsupp; goto out; - } else if (error < 0) + } else if (host_error < 0) goto out_nfserr; - error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); - if (error < 0) + host_error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); + if (host_error < 0) goto out_nfserr; if (S_ISDIR(inode->i_mode)) { - error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); - if (error < 0) + host_error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); + if (host_error < 0) goto out_nfserr; } @@ -464,7 +467,7 @@ out: posix_acl_release(dpacl); return (error); out_nfserr: - error = nfserrno(error); + error = nfserrno(host_error); goto out; } @@ -571,14 +574,14 @@ static struct accessmap nfs3_anyaccess[] = { { 0, 0 } }; -int +__be32 nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *supported) { struct accessmap *map; struct svc_export *export; struct dentry *dentry; u32 query, result = 0, sresult = 0; - unsigned int error; + __be32 error; error = fh_verify(rqstp, fhp, 0, MAY_NOP); if (error) @@ -598,7 +601,7 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor query = *access; for (; map->access; map++) { if (map->access & query) { - unsigned int err2; + __be32 err2; sresult |= map->access; @@ -637,13 +640,15 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor * The access argument indicates the type of open (read/write/lock) * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access, struct file **filp) { struct dentry *dentry; struct inode *inode; - int flags = O_RDONLY|O_LARGEFILE, err; + int flags = O_RDONLY|O_LARGEFILE; + __be32 err; + int host_err; /* * If we get here, then the client has already done an "open", @@ -673,10 +678,10 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, * Check to see if there are any leases on this file. * This may block while leases are broken. */ - err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); - if (err == -EWOULDBLOCK) - err = -ETIMEDOUT; - if (err) /* NOMEM or WOULDBLOCK */ + host_err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); + if (host_err == -EWOULDBLOCK) + host_err = -ETIMEDOUT; + if (host_err) /* NOMEM or WOULDBLOCK */ goto out_nfserr; if (access & MAY_WRITE) { @@ -689,10 +694,9 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, } *filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_mnt), flags); if (IS_ERR(*filp)) - err = PTR_ERR(*filp); + host_err = PTR_ERR(*filp); out_nfserr: - if (err) - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -830,14 +834,15 @@ nfsd_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset return size; } -static int +static __be32 nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { struct inode *inode; struct raparms *ra; mm_segment_t oldfs; - int err; + __be32 err; + int host_err; err = nfserr_perm; inode = file->f_dentry->d_inode; @@ -855,12 +860,12 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (file->f_op->sendfile && rqstp->rq_sendfile_ok) { rqstp->rq_resused = 1; - err = file->f_op->sendfile(file, &offset, *count, + host_err = file->f_op->sendfile(file, &offset, *count, nfsd_read_actor, rqstp); } else { oldfs = get_fs(); set_fs(KERNEL_DS); - err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); + host_err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); } @@ -874,13 +879,13 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, spin_unlock(&rab->pb_lock); } - if (err >= 0) { - nfsdstats.io_read += err; - *count = err; + if (host_err >= 0) { + nfsdstats.io_read += host_err; + *count = host_err; err = 0; fsnotify_access(file->f_dentry); } else - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -895,7 +900,7 @@ static void kill_suid(struct dentry *dentry) mutex_unlock(&dentry->d_inode->i_mutex); } -static int +static __be32 nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) @@ -904,7 +909,8 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; - int err = 0; + __be32 err = 0; + int host_err; int stable = *stablep; #ifdef MSNFS @@ -940,18 +946,18 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, /* Write the data. */ oldfs = get_fs(); set_fs(KERNEL_DS); - err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); + host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); - if (err >= 0) { + if (host_err >= 0) { nfsdstats.io_write += cnt; fsnotify_modify(file->f_dentry); } /* clear setuid/setgid flag after write */ - if (err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) + if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) kill_suid(dentry); - if (err >= 0 && stable) { + if (host_err >= 0 && stable) { static ino_t last_ino; static dev_t last_dev; @@ -977,7 +983,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (inode->i_state & I_DIRTY) { dprintk("nfsd: write sync %d\n", current->pid); - err=nfsd_sync(file); + host_err=nfsd_sync(file); } #if 0 wake_up(&inode->i_wait); @@ -987,11 +993,11 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, last_dev = inode->i_sb->s_dev; } - dprintk("nfsd: write complete err=%d\n", err); - if (err >= 0) + dprintk("nfsd: write complete host_err=%d\n", host_err); + if (host_err >= 0) err = 0; else - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -1001,12 +1007,12 @@ out: * on entry. On return, *count contains the number of bytes actually read. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { - int err; + __be32 err; if (file) { err = nfsd_permission(fhp->fh_export, fhp->fh_dentry, @@ -1030,12 +1036,12 @@ out: * The stable flag requests synchronous writes. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) { - int err = 0; + __be32 err = 0; if (file) { err = nfsd_permission(fhp->fh_export, fhp->fh_dentry, @@ -1067,12 +1073,12 @@ out: * Unfortunately we cannot lock the file to make sure we return full WCC * data to the client, as locking happens lower down in the filesystem. */ -int +__be32 nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, unsigned long count) { struct file *file; - int err; + __be32 err; if ((u64)count > ~(u64)offset) return nfserr_inval; @@ -1100,14 +1106,15 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, * * N.B. Every call to nfsd_create needs an fh_put for _both_ fhp and resfhp */ -int +__be32 nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, int type, dev_t rdev, struct svc_fh *resfhp) { struct dentry *dentry, *dchild = NULL; struct inode *dirp; - int err; + __be32 err; + int host_err; err = nfserr_perm; if (!flen) @@ -1134,7 +1141,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, /* called from nfsd_proc_mkdir, or possibly nfsd3_proc_create */ fh_lock_nested(fhp, I_MUTEX_PARENT); dchild = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dchild); + host_err = PTR_ERR(dchild); if (IS_ERR(dchild)) goto out_nfserr; err = fh_compose(resfhp, fhp->fh_export, dchild, fhp); @@ -1173,22 +1180,22 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, err = nfserr_perm; switch (type) { case S_IFREG: - err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); break; case S_IFDIR: - err = vfs_mkdir(dirp, dchild, iap->ia_mode); + host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); break; case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK: - err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); + host_err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); break; default: printk("nfsd: bad file type %o in nfsd_create\n", type); - err = -EINVAL; + host_err = -EINVAL; } - if (err < 0) + if (host_err < 0) goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { @@ -1203,7 +1210,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, * directories via NFS. */ if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { - int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; } @@ -1218,7 +1225,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1226,7 +1233,7 @@ out_nfserr: /* * NFSv3 version of nfsd_create */ -int +__be32 nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, struct svc_fh *resfhp, int createmode, u32 *verifier, @@ -1234,7 +1241,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, { struct dentry *dentry, *dchild = NULL; struct inode *dirp; - int err; + __be32 err; + int host_err; __u32 v_mtime=0, v_atime=0; int v_mode=0; @@ -1264,7 +1272,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * Compose the response file handle. */ dchild = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dchild); + host_err = PTR_ERR(dchild); if (IS_ERR(dchild)) goto out_nfserr; @@ -1320,8 +1328,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; } - err = vfs_create(dirp, dchild, iap->ia_mode, NULL); - if (err < 0) + host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + if (host_err < 0) goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { @@ -1350,7 +1358,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, */ set_attr: if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) { - int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; } @@ -1368,7 +1376,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } #endif /* CONFIG_NFSD_V3 */ @@ -1378,13 +1386,14 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * fits into the buffer. On return, it contains the true length. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) { struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, fhp, S_IFLNK, MAY_NOP); if (err) @@ -1403,18 +1412,18 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) */ oldfs = get_fs(); set_fs(KERNEL_DS); - err = inode->i_op->readlink(dentry, buf, *lenp); + host_err = inode->i_op->readlink(dentry, buf, *lenp); set_fs(oldfs); - if (err < 0) + if (host_err < 0) goto out_nfserr; - *lenp = err; + *lenp = host_err; err = 0; out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1422,7 +1431,7 @@ out_nfserr: * Create a symlink and look up its inode * N.B. After this call _both_ fhp and resfhp need an fh_put */ -int +__be32 nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, char *path, int plen, @@ -1430,7 +1439,8 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap) { struct dentry *dentry, *dnew; - int err, cerr; + __be32 err, cerr; + int host_err; umode_t mode; err = nfserr_noent; @@ -1446,7 +1456,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, fh_lock(fhp); dentry = fhp->fh_dentry; dnew = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dnew); + host_err = PTR_ERR(dnew); if (IS_ERR(dnew)) goto out_nfserr; @@ -1458,21 +1468,21 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (unlikely(path[plen] != 0)) { char *path_alloced = kmalloc(plen+1, GFP_KERNEL); if (path_alloced == NULL) - err = -ENOMEM; + host_err = -ENOMEM; else { strncpy(path_alloced, path, plen); path_alloced[plen] = 0; - err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); kfree(path_alloced); } } else - err = vfs_symlink(dentry->d_inode, dnew, path, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path, mode); - if (!err) + if (!host_err) { if (EX_ISSYNC(fhp->fh_export)) - err = nfsd_sync_dir(dentry); - if (err) - err = nfserrno(err); + host_err = nfsd_sync_dir(dentry); + } + err = nfserrno(host_err); fh_unlock(fhp); cerr = fh_compose(resfhp, fhp->fh_export, dnew, fhp); @@ -1482,7 +1492,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1490,13 +1500,14 @@ out_nfserr: * Create a hardlink * N.B. After this call _both_ ffhp and tfhp need an fh_put */ -int +__be32 nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *name, int len, struct svc_fh *tfhp) { struct dentry *ddir, *dnew, *dold; struct inode *dirp, *dest; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE); if (err) @@ -1517,24 +1528,25 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, dirp = ddir->d_inode; dnew = lookup_one_len(name, ddir, len); - err = PTR_ERR(dnew); + host_err = PTR_ERR(dnew); if (IS_ERR(dnew)) goto out_nfserr; dold = tfhp->fh_dentry; dest = dold->d_inode; - err = vfs_link(dold, dirp, dnew); - if (!err) { + host_err = vfs_link(dold, dirp, dnew); + if (!host_err) { if (EX_ISSYNC(ffhp->fh_export)) { err = nfserrno(nfsd_sync_dir(ddir)); write_inode_now(dest, 1); } + err = 0; } else { - if (err == -EXDEV && rqstp->rq_vers == 2) + if (host_err == -EXDEV && rqstp->rq_vers == 2) err = nfserr_acces; else - err = nfserrno(err); + err = nfserrno(host_err); } dput(dnew); @@ -1544,7 +1556,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out_unlock; } @@ -1552,13 +1564,14 @@ out_nfserr: * Rename a file * N.B. After this call _both_ ffhp and tfhp need an fh_put */ -int +__be32 nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, struct svc_fh *tfhp, char *tname, int tlen) { struct dentry *fdentry, *tdentry, *odentry, *ndentry, *trap; struct inode *fdir, *tdir; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE); if (err) @@ -1589,22 +1602,22 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, fill_pre_wcc(tfhp); odentry = lookup_one_len(fname, fdentry, flen); - err = PTR_ERR(odentry); + host_err = PTR_ERR(odentry); if (IS_ERR(odentry)) goto out_nfserr; - err = -ENOENT; + host_err = -ENOENT; if (!odentry->d_inode) goto out_dput_old; - err = -EINVAL; + host_err = -EINVAL; if (odentry == trap) goto out_dput_old; ndentry = lookup_one_len(tname, tdentry, tlen); - err = PTR_ERR(ndentry); + host_err = PTR_ERR(ndentry); if (IS_ERR(ndentry)) goto out_dput_old; - err = -ENOTEMPTY; + host_err = -ENOTEMPTY; if (ndentry == trap) goto out_dput_new; @@ -1612,14 +1625,14 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, if ((ffhp->fh_export->ex_flags & NFSEXP_MSNFS) && ((atomic_read(&odentry->d_count) > 1) || (atomic_read(&ndentry->d_count) > 1))) { - err = -EPERM; + host_err = -EPERM; } else #endif - err = vfs_rename(fdir, odentry, tdir, ndentry); - if (!err && EX_ISSYNC(tfhp->fh_export)) { - err = nfsd_sync_dir(tdentry); - if (!err) - err = nfsd_sync_dir(fdentry); + host_err = vfs_rename(fdir, odentry, tdir, ndentry); + if (!host_err && EX_ISSYNC(tfhp->fh_export)) { + host_err = nfsd_sync_dir(tdentry); + if (!host_err) + host_err = nfsd_sync_dir(fdentry); } out_dput_new: @@ -1627,8 +1640,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, out_dput_old: dput(odentry); out_nfserr: - if (err) - err = nfserrno(err); + err = nfserrno(host_err); /* we cannot reply on fh_unlock on the two filehandles, * as that would do the wrong thing if the two directories @@ -1647,13 +1659,14 @@ out: * Unlink a file or directory * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, char *fname, int flen) { struct dentry *dentry, *rdentry; struct inode *dirp; - int err; + __be32 err; + int host_err; err = nfserr_acces; if (!flen || isdotent(fname, flen)) @@ -1667,7 +1680,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, dirp = dentry->d_inode; rdentry = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(rdentry); + host_err = PTR_ERR(rdentry); if (IS_ERR(rdentry)) goto out_nfserr; @@ -1684,22 +1697,23 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, #ifdef MSNFS if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && (atomic_read(&rdentry->d_count) > 1)) { - err = -EPERM; + host_err = -EPERM; } else #endif - err = vfs_unlink(dirp, rdentry); + host_err = vfs_unlink(dirp, rdentry); } else { /* It's RMDIR */ - err = vfs_rmdir(dirp, rdentry); + host_err = vfs_rmdir(dirp, rdentry); } dput(rdentry); - if (err == 0 && - EX_ISSYNC(fhp->fh_export)) - err = nfsd_sync_dir(dentry); + if (host_err) + goto out_nfserr; + if (EX_ISSYNC(fhp->fh_export)) + host_err = nfsd_sync_dir(dentry); out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -1708,11 +1722,12 @@ out: * Read entries from a directory. * The NFSv3/4 verifier we ignore for now. */ -int +__be32 nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, struct readdir_cd *cdp, encode_dent_fn func) { - int err; + __be32 err; + int host_err; struct file *file; loff_t offset = *offsetp; @@ -1734,10 +1749,10 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, do { cdp->err = nfserr_eof; /* will be cleared on successful read */ - err = vfs_readdir(file, (filldir_t) func, cdp); - } while (err >=0 && cdp->err == nfs_ok); - if (err) - err = nfserrno(err); + host_err = vfs_readdir(file, (filldir_t) func, cdp); + } while (host_err >=0 && cdp->err == nfs_ok); + if (host_err) + err = nfserrno(host_err); else err = cdp->err; *offsetp = vfs_llseek(file, 0, 1); @@ -1754,10 +1769,10 @@ out: * Get file system stats * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) { - int err = fh_verify(rqstp, fhp, 0, MAY_NOP); + __be32 err = fh_verify(rqstp, fhp, 0, MAY_NOP); if (!err && vfs_statfs(fhp->fh_dentry,stat)) err = nfserr_io; return err; @@ -1766,7 +1781,7 @@ nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) /* * Check for a user's access permissions to this inode. */ -int +__be32 nfsd_permission(struct svc_export *exp, struct dentry *dentry, int acc) { struct inode *inode = dentry->d_inode; diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 2f75160..19a3c83 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -72,57 +72,57 @@ int nfsd_racache_init(int); void nfsd_racache_shutdown(void); int nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, struct svc_export **expp); -int nfsd_lookup(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_lookup(struct svc_rqst *, struct svc_fh *, const char *, int, struct svc_fh *); -int nfsd_setattr(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_setattr(struct svc_rqst *, struct svc_fh *, struct iattr *, int, time_t); #ifdef CONFIG_NFSD_V4 -int nfsd4_set_nfs4_acl(struct svc_rqst *, struct svc_fh *, +__be32 nfsd4_set_nfs4_acl(struct svc_rqst *, struct svc_fh *, struct nfs4_acl *); int nfsd4_get_nfs4_acl(struct svc_rqst *, struct dentry *, struct nfs4_acl **); #endif /* CONFIG_NFSD_V4 */ -int nfsd_create(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_create(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, int type, dev_t rdev, struct svc_fh *res); #ifdef CONFIG_NFSD_V3 -int nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); -int nfsd_create_v3(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); +__be32 nfsd_create_v3(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, struct svc_fh *res, int createmode, u32 *verifier, int *truncp); -int nfsd_commit(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_commit(struct svc_rqst *, struct svc_fh *, loff_t, unsigned long); #endif /* CONFIG_NFSD_V3 */ -int nfsd_open(struct svc_rqst *, struct svc_fh *, int, +__be32 nfsd_open(struct svc_rqst *, struct svc_fh *, int, int, struct file **); void nfsd_close(struct file *); -int nfsd_read(struct svc_rqst *, struct svc_fh *, struct file *, +__be32 nfsd_read(struct svc_rqst *, struct svc_fh *, struct file *, loff_t, struct kvec *, int, unsigned long *); -int nfsd_write(struct svc_rqst *, struct svc_fh *,struct file *, +__be32 nfsd_write(struct svc_rqst *, struct svc_fh *,struct file *, loff_t, struct kvec *,int, unsigned long, int *); -int nfsd_readlink(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_readlink(struct svc_rqst *, struct svc_fh *, char *, int *); -int nfsd_symlink(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_symlink(struct svc_rqst *, struct svc_fh *, char *name, int len, char *path, int plen, struct svc_fh *res, struct iattr *); -int nfsd_link(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_link(struct svc_rqst *, struct svc_fh *, char *, int, struct svc_fh *); -int nfsd_rename(struct svc_rqst *, +__be32 nfsd_rename(struct svc_rqst *, struct svc_fh *, char *, int, struct svc_fh *, char *, int); -int nfsd_remove(struct svc_rqst *, +__be32 nfsd_remove(struct svc_rqst *, struct svc_fh *, char *, int); -int nfsd_unlink(struct svc_rqst *, struct svc_fh *, int type, +__be32 nfsd_unlink(struct svc_rqst *, struct svc_fh *, int type, char *name, int len); int nfsd_truncate(struct svc_rqst *, struct svc_fh *, unsigned long size); -int nfsd_readdir(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_readdir(struct svc_rqst *, struct svc_fh *, loff_t *, struct readdir_cd *, encode_dent_fn); -int nfsd_statfs(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_statfs(struct svc_rqst *, struct svc_fh *, struct kstatfs *); int nfsd_notify_change(struct inode *, struct iattr *); -int nfsd_permission(struct svc_export *, struct dentry *, int); +__be32 nfsd_permission(struct svc_export *, struct dentry *, int); int nfsd_sync_dir(struct dentry *dp); #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) -- cgit v0.10.2 From b37ad28bcaa7c486a4ff0fb6c3bdaaacd67b86ce Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:59 -0700 Subject: [PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index ca64142..6382394 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -67,10 +67,10 @@ fh_dup2(struct svc_fh *dst, struct svc_fh *src) *dst = *src; } -static int +static __be32 do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, int accmode) { - int status; + __be32 status; if (open->op_truncate && !(open->op_share_access & NFS4_SHARE_ACCESS_WRITE)) @@ -88,11 +88,11 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs return status; } -static int +static __be32 do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { struct svc_fh resfh; - int status; + __be32 status; fh_init(&resfh, NFS4_FHSIZE); open->op_truncate = 0; @@ -131,10 +131,10 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o return status; } -static int +static __be32 do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { - int status; + __be32 status; /* Only reclaims from previously confirmed clients are valid */ if ((status = nfs4_check_open_reclaim(&open->op_clientid))) @@ -161,10 +161,10 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ } -static inline int +static inline __be32 nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; dprintk("NFSD: nfsd4_open filename %.*s op_stateowner %p\n", (int)open->op_fname.len, open->op_fname.data, open->op_stateowner); @@ -261,7 +261,7 @@ out: /* * filehandle-manipulating ops. */ -static inline int +static inline __be32 nfsd4_getfh(struct svc_fh *current_fh, struct svc_fh **getfh) { if (!current_fh->fh_dentry) @@ -271,7 +271,7 @@ nfsd4_getfh(struct svc_fh *current_fh, struct svc_fh **getfh) return nfs_ok; } -static inline int +static inline __be32 nfsd4_putfh(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_putfh *putfh) { fh_put(current_fh); @@ -280,10 +280,10 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_putf return fh_verify(rqstp, current_fh, 0, MAY_NOP); } -static inline int +static inline __be32 nfsd4_putrootfh(struct svc_rqst *rqstp, struct svc_fh *current_fh) { - int status; + __be32 status; fh_put(current_fh); status = exp_pseudoroot(rqstp->rq_client, current_fh, @@ -291,7 +291,7 @@ nfsd4_putrootfh(struct svc_rqst *rqstp, struct svc_fh *current_fh) return status; } -static inline int +static inline __be32 nfsd4_restorefh(struct svc_fh *current_fh, struct svc_fh *save_fh) { if (!save_fh->fh_dentry) @@ -301,7 +301,7 @@ nfsd4_restorefh(struct svc_fh *current_fh, struct svc_fh *save_fh) return nfs_ok; } -static inline int +static inline __be32 nfsd4_savefh(struct svc_fh *current_fh, struct svc_fh *save_fh) { if (!current_fh->fh_dentry) @@ -314,7 +314,7 @@ nfsd4_savefh(struct svc_fh *current_fh, struct svc_fh *save_fh) /* * misc nfsv4 ops */ -static inline int +static inline __be32 nfsd4_access(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_access *access) { if (access->ac_req_access & ~NFS3_ACCESS_FULL) @@ -324,10 +324,10 @@ nfsd4_access(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_acc return nfsd_access(rqstp, current_fh, &access->ac_resp_access, &access->ac_supported); } -static inline int +static inline __be32 nfsd4_commit(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_commit *commit) { - int status; + __be32 status; u32 *p = (u32 *)commit->co_verf.data; *p++ = nfssvc_boot.tv_sec; @@ -339,11 +339,11 @@ nfsd4_commit(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_com return status; } -static int +static __be32 nfsd4_create(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_create *create) { struct svc_fh resfh; - int status; + __be32 status; dev_t rdev; fh_init(&resfh, NFS4_FHSIZE); @@ -423,10 +423,10 @@ nfsd4_create(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_cre return status; } -static inline int +static inline __be32 nfsd4_getattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_getattr *getattr) { - int status; + __be32 status; status = fh_verify(rqstp, current_fh, 0, MAY_NOP); if (status) @@ -442,11 +442,11 @@ nfsd4_getattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ge return nfs_ok; } -static inline int +static inline __be32 nfsd4_link(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct svc_fh *save_fh, struct nfsd4_link *link) { - int status = nfserr_nofilehandle; + __be32 status = nfserr_nofilehandle; if (!save_fh->fh_dentry) return status; @@ -456,11 +456,11 @@ nfsd4_link(struct svc_rqst *rqstp, struct svc_fh *current_fh, return status; } -static int +static __be32 nfsd4_lookupp(struct svc_rqst *rqstp, struct svc_fh *current_fh) { struct svc_fh tmp_fh; - int ret; + __be32 ret; fh_init(&tmp_fh, NFS4_FHSIZE); if((ret = exp_pseudoroot(rqstp->rq_client, &tmp_fh, @@ -474,16 +474,16 @@ nfsd4_lookupp(struct svc_rqst *rqstp, struct svc_fh *current_fh) return nfsd_lookup(rqstp, current_fh, "..", 2, current_fh); } -static inline int +static inline __be32 nfsd4_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lookup *lookup) { return nfsd_lookup(rqstp, current_fh, lookup->lo_name, lookup->lo_len, current_fh); } -static inline int +static inline __be32 nfsd4_read(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_read *read) { - int status; + __be32 status; /* no need to check permission - this will be done in nfsd_read() */ @@ -508,7 +508,7 @@ out: return status; } -static inline int +static inline __be32 nfsd4_readdir(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_readdir *readdir) { u64 cookie = readdir->rd_cookie; @@ -531,7 +531,7 @@ nfsd4_readdir(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_re return nfs_ok; } -static inline int +static inline __be32 nfsd4_readlink(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_readlink *readlink) { readlink->rl_rqstp = rqstp; @@ -539,10 +539,10 @@ nfsd4_readlink(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_r return nfs_ok; } -static inline int +static inline __be32 nfsd4_remove(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_remove *remove) { - int status; + __be32 status; if (nfs4_in_grace()) return nfserr_grace; @@ -556,11 +556,11 @@ nfsd4_remove(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_rem return status; } -static inline int +static inline __be32 nfsd4_rename(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct svc_fh *save_fh, struct nfsd4_rename *rename) { - int status = nfserr_nofilehandle; + __be32 status = nfserr_nofilehandle; if (!save_fh->fh_dentry) return status; @@ -589,10 +589,10 @@ nfsd4_rename(struct svc_rqst *rqstp, struct svc_fh *current_fh, return status; } -static inline int +static inline __be32 nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_setattr *setattr) { - int status = nfs_ok; + __be32 status = nfs_ok; if (setattr->sa_iattr.ia_valid & ATTR_SIZE) { nfs4_lock_state(); @@ -614,13 +614,13 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_se return status; } -static inline int +static inline __be32 nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_write *write) { stateid_t *stateid = &write->wr_stateid; struct file *filp = NULL; u32 *p; - int status = nfs_ok; + __be32 status = nfs_ok; /* no need to check permission - this will be done in nfsd_write() */ @@ -661,12 +661,12 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ * attributes matched. VERIFY is implemented by mapping NFSERR_SAME * to NFS_OK after the call; NVERIFY by mapping NFSERR_NOT_SAME to NFS_OK. */ -static int +static __be32 nfsd4_verify(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_verify *verify) { __be32 *buf, *p; int count; - int status; + __be32 status; status = fh_verify(rqstp, current_fh, 0, MAY_NOP); if (status) @@ -741,7 +741,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, struct svc_fh *save_fh = NULL; struct nfs4_stateowner *replay_owner = NULL; int slack_space; /* in words, not bytes! */ - int status; + __be32 status; status = nfserr_resource; current_fh = kmalloc(sizeof(*current_fh), GFP_KERNEL); diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 1cbd2e4..e9d0770 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -83,13 +83,13 @@ md5_to_hex(char *out, char *md5) *out = '\0'; } -int +__be32 nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname) { struct xdr_netobj cksum; struct hash_desc desc; struct scatterlist sg[1]; - int status = nfserr_resource; + __be32 status = nfserr_resource; dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n", clname->len, clname->data); @@ -193,7 +193,7 @@ nfsd4_build_dentrylist(void *arg, const char *name, int namlen, struct dentry_list *child; if (name && isdotent(name, namlen)) - return nfs_ok; + return 0; dentry = lookup_one_len(name, parent, namlen); if (IS_ERR(dentry)) return PTR_ERR(dentry); @@ -333,14 +333,14 @@ purge_old(struct dentry *parent, struct dentry *child) int status; if (nfs4_has_reclaimed_state(child->d_name.name)) - return nfs_ok; + return 0; status = nfsd4_clear_clid_dir(parent, child); if (status) printk("failed to remove client recovery directory %s\n", child->d_name.name); /* Keep trying, success or failure: */ - return nfs_ok; + return 0; } void @@ -365,10 +365,10 @@ load_recdir(struct dentry *parent, struct dentry *child) printk("nfsd4: illegal name %s in recovery directory\n", child->d_name.name); /* Keep trying; maybe the others are OK: */ - return nfs_ok; + return 0; } nfs4_client_to_reclaim(child->d_name.name); - return nfs_ok; + return 0; } int diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ebcf226..e5ca6d7 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -710,7 +710,7 @@ out_err: * as described above. * */ -int +__be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) { u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; @@ -721,7 +721,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) nfs4_verifier clverifier = setclid->se_verf; unsigned int strhashval; struct nfs4_client *conf, *unconf, *new; - int status; + __be32 status; char dname[HEXDIR_LEN]; if (!check_name(clname)) @@ -875,14 +875,14 @@ out: * * NOTE: callback information will be processed here in a future patch */ -int +__be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm) { u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct nfs4_client *conf, *unconf; nfs4_verifier confirm = setclientid_confirm->sc_confirm; clientid_t * clid = &setclientid_confirm->sc_clientid; - int status; + __be32 status; if (STALE_CLIENTID(clid)) return nfserr_stale_clientid; @@ -1280,13 +1280,13 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) { * Called to check deny when READ with all zero stateid or * WRITE with all zero or all one stateid */ -static int +static __be32 nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type) { struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_file *fp; struct nfs4_stateid *stp; - int ret; + __be32 ret; dprintk("NFSD: nfs4_share_conflict\n"); @@ -1444,7 +1444,7 @@ static struct lock_manager_operations nfsd_lease_mng_ops = { }; -int +__be32 nfsd4_process_open1(struct nfsd4_open *open) { clientid_t *clientid = &open->op_clientid; @@ -1501,7 +1501,7 @@ renew: return nfs_ok; } -static inline int +static inline __be32 nfs4_check_delegmode(struct nfs4_delegation *dp, int flags) { if ((flags & WR_STATE) && (dp->dl_type == NFS4_OPEN_DELEGATE_READ)) @@ -1522,12 +1522,12 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid) return NULL; } -static int +static __be32 nfs4_check_deleg(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_delegation **dp) { int flags; - int status = nfserr_bad_stateid; + __be32 status = nfserr_bad_stateid; *dp = find_delegation_file(fp, &open->op_delegate_stateid); if (*dp == NULL) @@ -1546,11 +1546,11 @@ out: return nfs_ok; } -static int +static __be32 nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_stateid **stpp) { struct nfs4_stateid *local; - int status = nfserr_share_denied; + __be32 status = nfserr_share_denied; struct nfs4_stateowner *sop = open->op_stateowner; list_for_each_entry(local, &fp->fi_stateids, st_perfile) { @@ -1575,7 +1575,7 @@ nfs4_alloc_stateid(void) return kmem_cache_alloc(stateid_slab, GFP_KERNEL); } -static int +static __be32 nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, struct nfs4_delegation *dp, struct svc_fh *cur_fh, int flags) @@ -1590,7 +1590,7 @@ nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, get_file(dp->dl_vfs_file); stp->st_vfs_file = dp->dl_vfs_file; } else { - int status; + __be32 status; status = nfsd_open(rqstp, cur_fh, S_IFREG, flags, &stp->st_vfs_file); if (status) { @@ -1604,7 +1604,7 @@ nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, return 0; } -static inline int +static inline __be32 nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, struct nfsd4_open *open) { @@ -1619,22 +1619,22 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, return nfsd_setattr(rqstp, fh, &iattr, 0, (time_t)0); } -static int +static __be32 nfs4_upgrade_open(struct svc_rqst *rqstp, struct svc_fh *cur_fh, struct nfs4_stateid *stp, struct nfsd4_open *open) { struct file *filp = stp->st_vfs_file; struct inode *inode = filp->f_dentry->d_inode; unsigned int share_access, new_writer; - int status; + __be32 status; set_access(&share_access, stp->st_access_bmap); new_writer = (~share_access) & open->op_share_access & NFS4_SHARE_ACCESS_WRITE; if (new_writer) { - status = get_write_access(inode); - if (status) - return nfserrno(status); + int err = get_write_access(inode); + if (err) + return nfserrno(err); } status = nfsd4_truncate(rqstp, cur_fh, open); if (status) { @@ -1738,14 +1738,14 @@ out: /* * called with nfs4_lock_state() held. */ -int +__be32 nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { struct nfs4_file *fp = NULL; struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_stateid *stp = NULL; struct nfs4_delegation *dp = NULL; - int status; + __be32 status; status = nfserr_inval; if (!access_valid(open->op_share_access) @@ -1833,11 +1833,11 @@ static struct work_struct laundromat_work; static void laundromat_main(void *); static DECLARE_WORK(laundromat_work, laundromat_main, NULL); -int +__be32 nfsd4_renew(clientid_t *clid) { struct nfs4_client *clp; - int status; + __be32 status; nfs4_lock_state(); dprintk("process_renew(%08x/%08x): starting\n", @@ -1996,9 +1996,9 @@ access_permit_write(unsigned long access_bmap) } static -int nfs4_check_openmode(struct nfs4_stateid *stp, int flags) +__be32 nfs4_check_openmode(struct nfs4_stateid *stp, int flags) { - int status = nfserr_openmode; + __be32 status = nfserr_openmode; if ((flags & WR_STATE) && (!access_permit_write(stp->st_access_bmap))) goto out; @@ -2009,7 +2009,7 @@ out: return status; } -static inline int +static inline __be32 check_special_stateids(svc_fh *current_fh, stateid_t *stateid, int flags) { /* Trying to call delegreturn with a special stateid? Yuch: */ @@ -2043,14 +2043,14 @@ io_during_grace_disallowed(struct inode *inode, int flags) /* * Checks for stateid operations */ -int +__be32 nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filpp) { struct nfs4_stateid *stp = NULL; struct nfs4_delegation *dp = NULL; stateid_t *stidp; struct inode *ino = current_fh->fh_dentry->d_inode; - int status; + __be32 status; dprintk("NFSD: preprocess_stateid_op: stateid = (%08x/%08x/%08x/%08x)\n", stateid->si_boot, stateid->si_stateownerid, @@ -2125,7 +2125,7 @@ setlkflg (int type) /* * Checks for sequence id mutating operations. */ -static int +static __be32 nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *stateid, int flags, struct nfs4_stateowner **sopp, struct nfs4_stateid **stpp, struct nfsd4_lock *lock) { struct nfs4_stateid *stp; @@ -2169,7 +2169,7 @@ nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *statei clientid_t *lockclid = &lock->v.new.clientid; struct nfs4_client *clp = sop->so_client; int lkflg = 0; - int status; + __be32 status; lkflg = setlkflg(lock->lk_type); @@ -2241,10 +2241,10 @@ check_replay: return nfserr_bad_seqid; } -int +__be32 nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateowner *sop; struct nfs4_stateid *stp; @@ -2310,10 +2310,10 @@ reset_union_bmap_deny(unsigned long deny, unsigned long *bmap) } } -int +__be32 nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateid *stp; unsigned int share_access; @@ -2365,10 +2365,10 @@ out: /* * nfs4_unlock_state() called after encode */ -int +__be32 nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateid *stp; dprintk("NFSD: nfsd4_close on file %.*s\n", @@ -2404,10 +2404,10 @@ out: return status; } -int +__be32 nfsd4_delegreturn(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_delegreturn *dr) { - int status; + __be32 status; if ((status = fh_verify(rqstp, current_fh, S_IFREG, 0))) goto out; @@ -2635,7 +2635,7 @@ check_lock_length(u64 offset, u64 length) /* * LOCK operation */ -int +__be32 nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock, struct nfs4_stateowner **replay_owner) { struct nfs4_stateowner *open_sop = NULL; @@ -2644,7 +2644,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock struct file *filp; struct file_lock file_lock; struct file_lock conflock; - int status = 0; + __be32 status = 0; unsigned int strhashval; dprintk("NFSD: nfsd4_lock: start=%Ld length=%Ld\n", @@ -2793,14 +2793,14 @@ out: /* * LOCKT operation */ -int +__be32 nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt) { struct inode *inode; struct file file; struct file_lock file_lock; struct file_lock conflock; - int status; + __be32 status; if (nfs4_in_grace()) return nfserr_grace; @@ -2873,13 +2873,13 @@ out: return status; } -int +__be32 nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku, struct nfs4_stateowner **replay_owner) { struct nfs4_stateid *stp; struct file *filp = NULL; struct file_lock file_lock; - int status; + __be32 status; dprintk("NFSD: nfsd4_locku: start=%Ld length=%Ld\n", (long long) locku->lu_offset, @@ -2965,7 +2965,7 @@ out: return status; } -int +__be32 nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *rlockowner) { clientid_t *clid = &rlockowner->rl_clientid; @@ -2974,7 +2974,7 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner * struct xdr_netobj *owner = &rlockowner->rl_owner; struct list_head matches; int i; - int status; + __be32 status; dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n", clid->cl_boot, clid->cl_id); @@ -3111,7 +3111,7 @@ nfs4_find_reclaim_client(clientid_t *clid) /* * Called from OPEN. Look for clientid in reclaim list. */ -int +__be32 nfs4_check_open_reclaim(clientid_t *clid) { return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 3419d99..d7b630f 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -68,8 +68,8 @@ #define NFS4_REFERRAL_FSID_MAJOR 0x8000000ULL #define NFS4_REFERRAL_FSID_MINOR 0x8000000ULL -static int -check_filename(char *str, int len, int err) +static __be32 +check_filename(char *str, int len, __be32 err) { int i; @@ -95,7 +95,7 @@ check_filename(char *str, int len, int err) */ #define DECODE_HEAD \ __be32 *p; \ - int status + __be32 status #define DECODE_TAIL \ status = 0; \ out: \ @@ -217,7 +217,7 @@ static char *savemem(struct nfsd4_compoundargs *argp, __be32 *p, int nbytes) } -static int +static __be32 nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval) { u32 bmlen; @@ -240,7 +240,7 @@ nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *iattr, struct nfs4_acl **acl) { @@ -418,7 +418,7 @@ out_nfserr: goto out; } -static int +static __be32 nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access) { DECODE_HEAD; @@ -429,7 +429,7 @@ nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access DECODE_TAIL; } -static int +static __be32 nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) { DECODE_HEAD; @@ -444,7 +444,7 @@ nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) } -static int +static __be32 nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit) { DECODE_HEAD; @@ -456,7 +456,7 @@ nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit DECODE_TAIL; } -static int +static __be32 nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create) { DECODE_HEAD; @@ -496,7 +496,7 @@ nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create DECODE_TAIL; } -static inline int +static inline __be32 nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegreturn *dr) { DECODE_HEAD; @@ -508,13 +508,13 @@ nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegretu DECODE_TAIL; } -static inline int +static inline __be32 nfsd4_decode_getattr(struct nfsd4_compoundargs *argp, struct nfsd4_getattr *getattr) { return nfsd4_decode_bitmap(argp, getattr->ga_bmval); } -static int +static __be32 nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link) { DECODE_HEAD; @@ -529,7 +529,7 @@ nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) { DECODE_HEAD; @@ -568,7 +568,7 @@ nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) { DECODE_HEAD; @@ -587,7 +587,7 @@ nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku) { DECODE_HEAD; @@ -606,7 +606,7 @@ nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup) { DECODE_HEAD; @@ -621,7 +621,7 @@ nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open) { DECODE_HEAD; @@ -699,7 +699,7 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_confirm *open_conf) { DECODE_HEAD; @@ -713,7 +713,7 @@ nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_con DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_downgrade *open_down) { DECODE_HEAD; @@ -729,7 +729,7 @@ nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_d DECODE_TAIL; } -static int +static __be32 nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh) { DECODE_HEAD; @@ -744,7 +744,7 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read) { DECODE_HEAD; @@ -758,7 +758,7 @@ nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *readdir) { DECODE_HEAD; @@ -774,7 +774,7 @@ nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *read DECODE_TAIL; } -static int +static __be32 nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove) { DECODE_HEAD; @@ -789,7 +789,7 @@ nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove DECODE_TAIL; } -static int +static __be32 nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename) { DECODE_HEAD; @@ -809,7 +809,7 @@ nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename DECODE_TAIL; } -static int +static __be32 nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid) { DECODE_HEAD; @@ -820,7 +820,7 @@ nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *setattr) { DECODE_HEAD; @@ -834,7 +834,7 @@ nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *seta DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid *setclientid) { DECODE_HEAD; @@ -859,7 +859,7 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid_confirm *scd_c) { DECODE_HEAD; @@ -872,7 +872,7 @@ nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_s } /* Also used for NVERIFY */ -static int +static __be32 nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify) { #if 0 @@ -908,7 +908,7 @@ nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify DECODE_TAIL; } -static int +static __be32 nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) { int avail; @@ -959,7 +959,7 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_release_lockowner *rlockowner) { DECODE_HEAD; @@ -973,7 +973,7 @@ nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_rel DECODE_TAIL; } -static int +static __be32 nfsd4_decode_compound(struct nfsd4_compoundargs *argp) { DECODE_HEAD; @@ -1234,7 +1234,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) /* Encode as an array of strings the string given with components * seperated @sep. */ -static int nfsd4_encode_components(char sep, char *components, +static __be32 nfsd4_encode_components(char sep, char *components, __be32 **pp, int *buflen) { __be32 *p = *pp; @@ -1271,10 +1271,10 @@ static int nfsd4_encode_components(char sep, char *components, /* * encode a location element of a fs_locations structure */ -static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, +static __be32 nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, __be32 **pp, int *buflen) { - int status; + __be32 status; __be32 *p = *pp; status = nfsd4_encode_components(':', location->hosts, &p, buflen); @@ -1292,7 +1292,7 @@ static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, * Returned string is safe to use as long as the caller holds a reference * to @exp. */ -static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *stat) +static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 *stat) { struct svc_fh tmp_fh; char *path, *rootpath; @@ -1318,11 +1318,11 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *sta /* * encode a fs_locations structure */ -static int nfsd4_encode_fs_locations(struct svc_rqst *rqstp, +static __be32 nfsd4_encode_fs_locations(struct svc_rqst *rqstp, struct svc_export *exp, __be32 **pp, int *buflen) { - u32 status; + __be32 status; int i; __be32 *p = *pp; struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs; @@ -1353,7 +1353,7 @@ static u32 nfs4_ftypes[16] = { NF4SOCK, NF4BAD, NF4LNK, NF4BAD, }; -static int +static __be32 nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, __be32 **p, int *buflen) { @@ -1375,19 +1375,19 @@ nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, return 0; } -static inline int +static inline __be32 nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, uid, 0, p, buflen); } -static inline int +static inline __be32 nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, gid, 1, p, buflen); } -static inline int +static inline __be32 nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, __be32 **p, int *buflen) { @@ -1398,7 +1398,7 @@ nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, FATTR4_WORD0_RDATTR_ERROR) #define WORD1_ABSENT_FS_ATTRS FATTR4_WORD1_MOUNTED_ON_FILEID -static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) +static __be32 fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) { /* As per referral draft: */ if (*bmval0 & ~WORD0_ABSENT_FS_ATTRS || @@ -1421,7 +1421,7 @@ static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) * @countp is the buffer size in _words_; upon successful return this becomes * replaced with the number of words written. */ -int +__be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *rqstp) @@ -1437,7 +1437,7 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, u64 dummy64; u32 rdattr_err = 0; __be32 *p = buffer; - int status; + __be32 status; int aclsupport = 0; struct nfs4_acl *acl = NULL; @@ -1829,13 +1829,13 @@ out_serverfault: goto out; } -static int +static __be32 nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, const char *name, int namlen, __be32 *p, int *buflen) { struct svc_export *exp = cd->rd_fhp->fh_export; struct dentry *dentry; - int nfserr; + __be32 nfserr; dentry = lookup_one_len(name, cd->rd_fhp->fh_dentry, namlen); if (IS_ERR(dentry)) @@ -1865,7 +1865,7 @@ out_put: } static __be32 * -nfsd4_encode_rdattr_error(__be32 *p, int buflen, int nfserr) +nfsd4_encode_rdattr_error(__be32 *p, int buflen, __be32 nfserr) { __be32 *attrlenp; @@ -1888,7 +1888,7 @@ nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; __be32 *p = cd->buffer; - int nfserr = nfserr_toosmall; + __be32 nfserr = nfserr_toosmall; /* In nfsv4, "." and ".." never make it onto the wire.. */ if (name && isdotent(name, namlen)) { @@ -1944,7 +1944,7 @@ fail: } static void -nfsd4_encode_access(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_access *access) +nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_access *access) { ENCODE_HEAD; @@ -1957,7 +1957,7 @@ nfsd4_encode_access(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_acc } static void -nfsd4_encode_close(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_close *close) +nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_close *close) { ENCODE_SEQID_OP_HEAD; @@ -1972,7 +1972,7 @@ nfsd4_encode_close(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_clos static void -nfsd4_encode_commit(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_commit *commit) +nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_commit *commit) { ENCODE_HEAD; @@ -1984,7 +1984,7 @@ nfsd4_encode_commit(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_com } static void -nfsd4_encode_create(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_create *create) +nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_create *create) { ENCODE_HEAD; @@ -1998,8 +1998,8 @@ nfsd4_encode_create(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_cre } } -static int -nfsd4_encode_getattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_getattr *getattr) +static __be32 +nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_getattr *getattr) { struct svc_fh *fhp = getattr->ga_fhp; int buflen; @@ -2017,7 +2017,7 @@ nfsd4_encode_getattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_ge } static void -nfsd4_encode_getfh(struct nfsd4_compoundres *resp, int nfserr, struct svc_fh *fhp) +nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh *fhp) { unsigned int len; ENCODE_HEAD; @@ -2057,7 +2057,7 @@ nfsd4_encode_lock_denied(struct nfsd4_compoundres *resp, struct nfsd4_lock_denie } static void -nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock *lock) +nfsd4_encode_lock(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lock *lock) { ENCODE_SEQID_OP_HEAD; @@ -2073,14 +2073,14 @@ nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock } static void -nfsd4_encode_lockt(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lockt *lockt) +nfsd4_encode_lockt(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lockt *lockt) { if (nfserr == nfserr_denied) nfsd4_encode_lock_denied(resp, &lockt->lt_denied); } static void -nfsd4_encode_locku(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_locku *locku) +nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_locku *locku) { ENCODE_SEQID_OP_HEAD; @@ -2096,7 +2096,7 @@ nfsd4_encode_locku(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock static void -nfsd4_encode_link(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_link *link) +nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_link *link) { ENCODE_HEAD; @@ -2109,7 +2109,7 @@ nfsd4_encode_link(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_link static void -nfsd4_encode_open(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open *open) +nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open *open) { ENCODE_SEQID_OP_HEAD; @@ -2174,7 +2174,7 @@ out: } static void -nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open_confirm *oc) +nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_confirm *oc) { ENCODE_SEQID_OP_HEAD; @@ -2189,7 +2189,7 @@ nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, int nfserr, struct nfs } static void -nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open_downgrade *od) +nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_downgrade *od) { ENCODE_SEQID_OP_HEAD; @@ -2203,8 +2203,8 @@ nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, int nfserr, struct n ENCODE_SEQID_OP_TAIL(od->od_stateowner); } -static int -nfsd4_encode_read(struct nfsd4_compoundres *resp, int nfserr, +static __be32 +nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_read *read) { u32 eof; @@ -2268,8 +2268,8 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, int nfserr, return 0; } -static int -nfsd4_encode_readlink(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_readlink *readlink) +static __be32 +nfsd4_encode_readlink(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readlink *readlink) { int maxcount; char *page; @@ -2316,8 +2316,8 @@ nfsd4_encode_readlink(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_r return 0; } -static int -nfsd4_encode_readdir(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_readdir *readdir) +static __be32 +nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readdir *readdir) { int maxcount; loff_t offset; @@ -2396,7 +2396,7 @@ err_no_verf: } static void -nfsd4_encode_remove(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_remove *remove) +nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_remove *remove) { ENCODE_HEAD; @@ -2408,7 +2408,7 @@ nfsd4_encode_remove(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_rem } static void -nfsd4_encode_rename(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_rename *rename) +nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_rename *rename) { ENCODE_HEAD; @@ -2425,7 +2425,7 @@ nfsd4_encode_rename(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_ren * regardless of the error status. */ static void -nfsd4_encode_setattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_setattr *setattr) +nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setattr *setattr) { ENCODE_HEAD; @@ -2444,7 +2444,7 @@ nfsd4_encode_setattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_se } static void -nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_setclientid *scd) +nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setclientid *scd) { ENCODE_HEAD; @@ -2463,7 +2463,7 @@ nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, int nfserr, struct nfsd } static void -nfsd4_encode_write(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_write *write) +nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_write *write) { ENCODE_HEAD; @@ -2641,7 +2641,7 @@ void nfsd4_release_compoundargs(struct nfsd4_compoundargs *args) int nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundargs *args) { - int status; + __be32 status; args->p = p; args->end = rqstp->rq_arg.head[0].iov_base + rqstp->rq_arg.head[0].iov_len; diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 8bf23cf..a597e2e 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -164,7 +164,7 @@ update_stateid(stateid_t *stateid) * is cached. */ struct nfs4_replay { - u32 rp_status; + __be32 rp_status; unsigned int rp_buflen; char *rp_buf; unsigned intrp_allocated; @@ -273,19 +273,19 @@ struct nfs4_stateid { ((err) != nfserr_stale_stateid) && \ ((err) != nfserr_bad_stateid)) -extern int nfsd4_renew(clientid_t *clid); -extern int nfs4_preprocess_stateid_op(struct svc_fh *current_fh, +extern __be32 nfsd4_renew(clientid_t *clid); +extern __be32 nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filp); extern void nfs4_lock_state(void); extern void nfs4_unlock_state(void); extern int nfs4_in_grace(void); -extern int nfs4_check_open_reclaim(clientid_t *clid); +extern __be32 nfs4_check_open_reclaim(clientid_t *clid); extern void put_nfs4_client(struct nfs4_client *clp); extern void nfs4_free_stateowner(struct kref *kref); extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_cb_recall(struct nfs4_delegation *dp); extern void nfs4_put_delegation(struct nfs4_delegation *dp); -extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); +extern __be32 nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); extern void nfsd4_init_recdir(char *recdir_name); extern int nfsd4_recdir_load(void); extern void nfsd4_shutdown_recdir(void); diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index 003193f..45ca01b 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -334,7 +334,7 @@ struct nfsd4_write { struct nfsd4_op { int opnum; - int status; + __be32 status; union { struct nfsd4_access access; struct nfsd4_close close; @@ -426,38 +426,38 @@ int nfs4svc_encode_compoundres(struct svc_rqst *, __be32 *, struct nfsd4_compoundres *); void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *); void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op); -int nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, - struct dentry *dentry, u32 *buffer, int *countp, +__be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, + struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *); -extern int nfsd4_setclientid(struct svc_rqst *rqstp, +extern __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid); -extern int nfsd4_setclientid_confirm(struct svc_rqst *rqstp, +extern __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm); -extern int nfsd4_process_open1(struct nfsd4_open *open); -extern int nfsd4_process_open2(struct svc_rqst *rqstp, +extern __be32 nfsd4_process_open1(struct nfsd4_open *open); +extern __be32 nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open); -extern int nfsd4_open_confirm(struct svc_rqst *rqstp, +extern __be32 nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc, struct nfs4_stateowner **); -extern int nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close, struct nfs4_stateowner **replay_owner); -extern int nfsd4_open_downgrade(struct svc_rqst *rqstp, +extern __be32 nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od, struct nfs4_stateowner **replay_owner); -extern int nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock, struct nfs4_stateowner **replay_owner); -extern int nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt); -extern int nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku, struct nfs4_stateowner **replay_owner); -extern int +extern __be32 nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *rlockowner); extern void nfsd4_release_compoundargs(struct nfsd4_compoundargs *); -extern int nfsd4_delegreturn(struct svc_rqst *rqstp, +extern __be32 nfsd4_delegreturn(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_delegreturn *dr); #endif -- cgit v0.10.2 From c4d987ba841dff4b2fc768e52d1d95af83f9f157 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:00 -0700 Subject: [PATCH] nfsd: NFSv{2,3} trivial endianness annotations for error values Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index fd5397d..e3eca08 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -35,7 +35,7 @@ static __be32 nfsacld_proc_getacl(struct svc_rqst * rqstp, { svc_fh *fh; struct posix_acl *acl; - int nfserr = 0; + __be32 nfserr = 0; dprintk("nfsd: GETACL(2acl) %s\n", SVCFH_fmt(&argp->fh)); @@ -102,7 +102,7 @@ static __be32 nfsacld_proc_setacl(struct svc_rqst * rqstp, struct nfsd_attrstat *resp) { svc_fh *fh; - int nfserr = 0; + __be32 nfserr = 0; dprintk("nfsd: SETACL(2acl) %s\n", SVCFH_fmt(&argp->fh)); @@ -143,7 +143,7 @@ static __be32 nfsacld_proc_getattr(struct svc_rqst * rqstp, static __be32 nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: ACCESS(2acl) %s 0x%x\n", SVCFH_fmt(&argp->fh), diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index 78b2c83..fcad289 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -33,7 +33,7 @@ static __be32 nfsd3_proc_getacl(struct svc_rqst * rqstp, { svc_fh *fh; struct posix_acl *acl; - int nfserr = 0; + __be32 nfserr = 0; fh = fh_copy(&resp->fh, &argp->fh); if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP))) @@ -98,7 +98,7 @@ static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_attrstat *resp) { svc_fh *fh; - int nfserr = 0; + __be32 nfserr = 0; fh = fh_copy(&resp->fh, &argp->fh); nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_SATTR); diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index a12663f..64db601 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -56,7 +56,8 @@ static __be32 nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd3_attrstat *resp) { - int err, nfserr; + int err; + __be32 nfserr; dprintk("nfsd: GETATTR(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -80,7 +81,7 @@ static __be32 nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, struct nfsd3_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: SETATTR(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -98,7 +99,7 @@ static __be32 nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LOOKUP(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -122,7 +123,7 @@ static __be32 nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: ACCESS(3) %s 0x%x\n", SVCFH_fmt(&argp->fh), @@ -141,7 +142,7 @@ static __be32 nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, struct nfsd3_readlinkres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: READLINK(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -159,7 +160,7 @@ static __be32 nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, struct nfsd3_readres *resp) { - int nfserr; + __be32 nfserr; u32 max_blocksize = svc_max_payload(rqstp); dprintk("nfsd: READ(3) %s %lu bytes at %lu\n", @@ -199,7 +200,7 @@ static __be32 nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, struct nfsd3_writeres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: WRITE(3) %s %d bytes at %ld%s\n", SVCFH_fmt(&argp->fh), @@ -229,7 +230,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, { svc_fh *dirfhp, *newfhp = NULL; struct iattr *attr; - u32 nfserr; + __be32 nfserr; dprintk("nfsd: CREATE(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -269,7 +270,7 @@ static __be32 nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: MKDIR(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -289,7 +290,7 @@ static __be32 nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, struct nfsd3_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: SYMLINK(3) %s %.*s -> %.*s\n", SVCFH_fmt(&argp->ffh), @@ -311,7 +312,8 @@ static __be32 nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, struct nfsd3_diropres *resp) { - int nfserr, type; + __be32 nfserr; + int type; dev_t rdev = 0; dprintk("nfsd: MKNOD(3) %s %.*s\n", @@ -347,7 +349,7 @@ static __be32 nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: REMOVE(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -367,7 +369,7 @@ static __be32 nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RMDIR(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -383,7 +385,7 @@ static __be32 nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, struct nfsd3_renameres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RENAME(3) %s %.*s ->\n", SVCFH_fmt(&argp->ffh), @@ -405,7 +407,7 @@ static __be32 nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, struct nfsd3_linkres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LINK(3) %s ->\n", SVCFH_fmt(&argp->ffh)); @@ -428,7 +430,8 @@ static __be32 nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { - int nfserr, count; + __be32 nfserr; + int count; dprintk("nfsd: READDIR(3) %s %d bytes at %d\n", SVCFH_fmt(&argp->fh), @@ -463,7 +466,8 @@ static __be32 nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { - int nfserr, count = 0; + __be32 nfserr; + int count = 0; loff_t offset; int i; caddr_t page_addr = NULL; @@ -521,7 +525,7 @@ static __be32 nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsstatres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: FSSTAT(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -538,7 +542,7 @@ static __be32 nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsinfores *resp) { - int nfserr; + __be32 nfserr; u32 max_blocksize = svc_max_payload(rqstp); dprintk("nfsd: FSINFO(3) %s\n", @@ -580,7 +584,7 @@ static __be32 nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_pathconfres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: PATHCONF(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -623,7 +627,7 @@ static __be32 nfsd3_proc_commit(struct svc_rqst * rqstp, struct nfsd3_commitargs *argp, struct nfsd3_commitres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: COMMIT(3) %s %u@%Lu\n", SVCFH_fmt(&argp->fh), diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 03ab682..ec983b7 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -36,16 +36,16 @@ nfsd_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) return nfs_ok; } -static int -nfsd_return_attrs(int err, struct nfsd_attrstat *resp) +static __be32 +nfsd_return_attrs(__be32 err, struct nfsd_attrstat *resp) { if (err) return err; return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, resp->fh.fh_dentry, &resp->stat)); } -static int -nfsd_return_dirop(int err, struct nfsd_diropres *resp) +static __be32 +nfsd_return_dirop(__be32 err, struct nfsd_diropres *resp) { if (err) return err; return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, @@ -60,7 +60,7 @@ static __be32 nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); fh_copy(&resp->fh, &argp->fh); @@ -76,7 +76,7 @@ static __be32 nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, struct nfsd_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: SETATTR %s, valid=%x, size=%ld\n", SVCFH_fmt(&argp->fh), argp->attrs.ia_valid, (long) argp->attrs.ia_size); @@ -96,7 +96,7 @@ static __be32 nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, struct nfsd_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LOOKUP %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -116,7 +116,7 @@ static __be32 nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, struct nfsd_readlinkres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: READLINK %s\n", SVCFH_fmt(&argp->fh)); @@ -136,7 +136,7 @@ static __be32 nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, struct nfsd_readres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: READ %s %d bytes at %d\n", SVCFH_fmt(&argp->fh), @@ -176,7 +176,7 @@ static __be32 nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, struct nfsd_attrstat *resp) { - int nfserr; + __be32 nfserr; int stable = 1; dprintk("nfsd: WRITE %s %d bytes at %d\n", @@ -206,7 +206,8 @@ nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct iattr *attr = &argp->attrs; struct inode *inode; struct dentry *dchild; - int nfserr, type, mode; + int type, mode; + __be32 nfserr; dev_t rdev = 0, wanted = new_decode_dev(attr->ia_size); dprintk("nfsd: CREATE %s %.*s\n", @@ -352,7 +353,7 @@ static __be32 nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: REMOVE %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -367,7 +368,7 @@ static __be32 nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RENAME %s %.*s -> \n", SVCFH_fmt(&argp->ffh), argp->flen, argp->fname); @@ -385,7 +386,7 @@ static __be32 nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LINK %s ->\n", SVCFH_fmt(&argp->ffh)); @@ -406,7 +407,7 @@ nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, void *resp) { struct svc_fh newfh; - int nfserr; + __be32 nfserr; dprintk("nfsd: SYMLINK %s %.*s -> %.*s\n", SVCFH_fmt(&argp->ffh), argp->flen, argp->fname, @@ -434,7 +435,7 @@ static __be32 nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: MKDIR %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -458,7 +459,7 @@ static __be32 nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RMDIR %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -474,7 +475,8 @@ static __be32 nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, struct nfsd_readdirres *resp) { - int nfserr, count; + int count; + __be32 nfserr; loff_t offset; dprintk("nfsd: READDIR %s %d bytes at %d\n", @@ -513,7 +515,7 @@ static __be32 nfsd_proc_statfs(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_statfsres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: STATFS %s\n", SVCFH_fmt(&argp->fh)); -- cgit v0.10.2 From b8dd7b9ab194d9ab322881f49fde42954757efae Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:01 -0700 Subject: [PATCH] nfsd: NFSv4 errno endianness annotations don't use the same variable to store NFS and host error values Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e5ca6d7..ae1d477 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2646,6 +2646,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock struct file_lock conflock; __be32 status = 0; unsigned int strhashval; + int err; dprintk("NFSD: nfsd4_lock: start=%Ld length=%Ld\n", (long long) lock->lk_offset, @@ -2758,13 +2759,14 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock * locks_copy_lock: */ conflock.fl_ops = NULL; conflock.fl_lmops = NULL; - status = posix_lock_file_conf(filp, &file_lock, &conflock); + err = posix_lock_file_conf(filp, &file_lock, &conflock); dprintk("NFSD: nfsd4_lock: posix_lock_file_conf status %d\n",status); - switch (-status) { + switch (-err) { case 0: /* success! */ update_stateid(&lock_stp->st_stateid); memcpy(&lock->lk_resp_stateid, &lock_stp->st_stateid, sizeof(stateid_t)); + status = 0; break; case (EAGAIN): /* conflock holds conflicting lock */ status = nfserr_denied; @@ -2775,7 +2777,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock status = nfserr_deadlock; break; default: - dprintk("NFSD: nfsd4_lock: posix_lock_file_conf() failed! status %d\n",status); + dprintk("NFSD: nfsd4_lock: posix_lock_file_conf() failed! status %d\n",err); status = nfserr_resource; break; } @@ -2880,6 +2882,7 @@ nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock struct file *filp = NULL; struct file_lock file_lock; __be32 status; + int err; dprintk("NFSD: nfsd4_locku: start=%Ld length=%Ld\n", (long long) locku->lu_offset, @@ -2917,8 +2920,8 @@ nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock /* * Try to unlock the file in the VFS. */ - status = posix_lock_file(filp, &file_lock); - if (status) { + err = posix_lock_file(filp, &file_lock); + if (err) { dprintk("NFSD: nfs4_locku: posix_lock_file failed!\n"); goto out_nfserr; } @@ -2937,7 +2940,7 @@ out: return status; out_nfserr: - status = nfserrno(status); + status = nfserrno(err); goto out; } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index d7b630f..f3f239d 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -247,6 +247,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia int expected_len, len = 0; u32 dummy32; char *buf; + int host_err; DECODE_HEAD; iattr->ia_valid = 0; @@ -280,7 +281,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia *acl = nfs4_acl_new(); if (*acl == NULL) { - status = -ENOMEM; + host_err = -ENOMEM; goto out_nfserr; } defer_free(argp, (void (*)(const void *))nfs4_acl_free, *acl); @@ -295,20 +296,20 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia len += XDR_QUADLEN(dummy32) << 2; READMEM(buf, dummy32); ace.whotype = nfs4_acl_get_whotype(buf, dummy32); - status = 0; + host_err = 0; if (ace.whotype != NFS4_ACL_WHO_NAMED) ace.who = 0; else if (ace.flag & NFS4_ACE_IDENTIFIER_GROUP) - status = nfsd_map_name_to_gid(argp->rqstp, + host_err = nfsd_map_name_to_gid(argp->rqstp, buf, dummy32, &ace.who); else - status = nfsd_map_name_to_uid(argp->rqstp, + host_err = nfsd_map_name_to_uid(argp->rqstp, buf, dummy32, &ace.who); - if (status) + if (host_err) goto out_nfserr; - status = nfs4_acl_add_ace(*acl, ace.type, ace.flag, + host_err = nfs4_acl_add_ace(*acl, ace.type, ace.flag, ace.access_mask, ace.whotype, ace.who); - if (status) + if (host_err) goto out_nfserr; } } else @@ -327,7 +328,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia READ_BUF(dummy32); len += (XDR_QUADLEN(dummy32) << 2); READMEM(buf, dummy32); - if ((status = nfsd_map_name_to_uid(argp->rqstp, buf, dummy32, &iattr->ia_uid))) + if ((host_err = nfsd_map_name_to_uid(argp->rqstp, buf, dummy32, &iattr->ia_uid))) goto out_nfserr; iattr->ia_valid |= ATTR_UID; } @@ -338,7 +339,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia READ_BUF(dummy32); len += (XDR_QUADLEN(dummy32) << 2); READMEM(buf, dummy32); - if ((status = nfsd_map_name_to_gid(argp->rqstp, buf, dummy32, &iattr->ia_gid))) + if ((host_err = nfsd_map_name_to_gid(argp->rqstp, buf, dummy32, &iattr->ia_gid))) goto out_nfserr; iattr->ia_valid |= ATTR_GID; } @@ -414,7 +415,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia DECODE_TAIL; out_nfserr: - status = nfserrno(status); + status = nfserrno(host_err); goto out; } @@ -1438,6 +1439,7 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, u32 rdattr_err = 0; __be32 *p = buffer; __be32 status; + int err; int aclsupport = 0; struct nfs4_acl *acl = NULL; @@ -1451,14 +1453,14 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, goto out; } - status = vfs_getattr(exp->ex_mnt, dentry, &stat); - if (status) + err = vfs_getattr(exp->ex_mnt, dentry, &stat); + if (err) goto out_nfserr; if ((bmval0 & (FATTR4_WORD0_FILES_FREE | FATTR4_WORD0_FILES_TOTAL)) || (bmval1 & (FATTR4_WORD1_SPACE_AVAIL | FATTR4_WORD1_SPACE_FREE | FATTR4_WORD1_SPACE_TOTAL))) { - status = vfs_statfs(dentry, &statfs); - if (status) + err = vfs_statfs(dentry, &statfs); + if (err) goto out_nfserr; } if ((bmval0 & (FATTR4_WORD0_FILEHANDLE | FATTR4_WORD0_FSID)) && !fhp) { @@ -1470,15 +1472,15 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, } if (bmval0 & (FATTR4_WORD0_ACL | FATTR4_WORD0_ACLSUPPORT | FATTR4_WORD0_SUPPORTED_ATTRS)) { - status = nfsd4_get_nfs4_acl(rqstp, dentry, &acl); - aclsupport = (status == 0); + err = nfsd4_get_nfs4_acl(rqstp, dentry, &acl); + aclsupport = (err == 0); if (bmval0 & FATTR4_WORD0_ACL) { - if (status == -EOPNOTSUPP) + if (err == -EOPNOTSUPP) bmval0 &= ~FATTR4_WORD0_ACL; - else if (status == -EINVAL) { + else if (err == -EINVAL) { status = nfserr_attrnotsupp; goto out; - } else if (status != 0) + } else if (err != 0) goto out_nfserr; } } @@ -1818,7 +1820,7 @@ out: fh_put(&tempfh); return status; out_nfserr: - status = nfserrno(status); + status = nfserrno(err); goto out; out_resource: *countp = 0; -- cgit v0.10.2 From f00f328fda1eeec575cd0f360da81b66bf4133a1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:01 -0700 Subject: [PATCH] xdr annotations: nfsd callback* Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 324a278..8497ed4 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -85,8 +85,8 @@ enum nfs_cb_opnum4 { /* * Generic encode routines from fs/nfs/nfs4xdr.c */ -static inline u32 * -xdr_writemem(u32 *p, const void *ptr, int nbytes) +static inline __be32 * +xdr_writemem(__be32 *p, const void *ptr, int nbytes) { int tmp = XDR_QUADLEN(nbytes); if (!tmp) @@ -205,7 +205,7 @@ nfs_cb_stat_to_errno(int stat) static int encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr) { - u32 * p; + __be32 * p; RESERVE_SPACE(16); WRITE32(0); /* tag length is always 0 */ @@ -218,7 +218,7 @@ encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr) static int encode_cb_recall(struct xdr_stream *xdr, struct nfs4_cb_recall *cb_rec) { - u32 *p; + __be32 *p; int len = cb_rec->cbr_fhlen; RESERVE_SPACE(12+sizeof(cb_rec->cbr_stateid) + len); @@ -231,7 +231,7 @@ encode_cb_recall(struct xdr_stream *xdr, struct nfs4_cb_recall *cb_rec) } static int -nfs4_xdr_enc_cb_null(struct rpc_rqst *req, u32 *p) +nfs4_xdr_enc_cb_null(struct rpc_rqst *req, __be32 *p) { struct xdr_stream xdrs, *xdr = &xdrs; @@ -241,7 +241,7 @@ nfs4_xdr_enc_cb_null(struct rpc_rqst *req, u32 *p) } static int -nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, u32 *p, struct nfs4_cb_recall *args) +nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, __be32 *p, struct nfs4_cb_recall *args) { struct xdr_stream xdr; struct nfs4_cb_compound_hdr hdr = { @@ -257,7 +257,7 @@ nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, u32 *p, struct nfs4_cb_recall *args static int decode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr){ - u32 *p; + __be32 *p; READ_BUF(8); READ32(hdr->status); @@ -272,7 +272,7 @@ decode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr) static int decode_cb_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) { - u32 *p; + __be32 *p; u32 op; int32_t nfserr; @@ -291,13 +291,13 @@ decode_cb_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) } static int -nfs4_xdr_dec_cb_null(struct rpc_rqst *req, u32 *p) +nfs4_xdr_dec_cb_null(struct rpc_rqst *req, __be32 *p) { return 0; } static int -nfs4_xdr_dec_cb_recall(struct rpc_rqst *rqstp, u32 *p) +nfs4_xdr_dec_cb_recall(struct rpc_rqst *rqstp, __be32 *p) { struct xdr_stream xdr; struct nfs4_cb_compound_hdr hdr; -- cgit v0.10.2 From c7afef1f963bec198b186cc34b9e8c9b9ce2e266 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:02 -0700 Subject: [PATCH] nfsd: misc endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index e13fa23..f37df46 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1148,12 +1148,12 @@ exp_find(struct auth_domain *clp, int fsid_type, u32 *fsidv, * for a given NFSv4 client. The root is defined to be the * export point with fsid==0 */ -int +__be32 exp_pseudoroot(struct auth_domain *clp, struct svc_fh *fhp, struct cache_req *creq) { struct svc_export *exp; - int rv; + __be32 rv; u32 fsidv[2]; mk_fsid_v1(fsidv, 0); diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c index 9b9e7e1..11fdaf7 100644 --- a/fs/nfsd/lockd.c +++ b/fs/nfsd/lockd.c @@ -25,7 +25,7 @@ static u32 nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) { - u32 nfserr; + __be32 nfserr; struct svc_fh fh; /* must initialize before using! but maxsize doesn't matter */ diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 8497ed4..f57655a 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -461,7 +461,7 @@ nfs4_cb_null(struct rpc_task *task, void *dummy) { struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp; struct nfs4_callback *cb = &clp->cl_callback; - u32 addr = htonl(cb->cb_addr); + __be32 addr = htonl(cb->cb_addr); dprintk("NFSD: nfs4_cb_null task->tk_status %d\n", task->tk_status); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ae1d477..2e468c9 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -713,7 +713,7 @@ out_err: __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) { - u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; + __be32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct xdr_netobj clname = { .len = setclid->se_namelen, .data = setclid->se_name, @@ -878,7 +878,7 @@ out: __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm) { - u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; + __be32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct nfs4_client *conf, *unconf; nfs4_verifier confirm = setclientid_confirm->sc_confirm; clientid_t * clid = &setclientid_confirm->sc_clientid; diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index fdf7cf3..6100bbe 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -29,7 +29,7 @@ */ #define CACHESIZE 1024 #define HASHSIZE 64 -#define REQHASH(xid) ((((xid) >> 24) ^ (xid)) & (HASHSIZE-1)) +#define REQHASH(xid) (((((__force __u32)xid) >> 24) ^ ((__force __u32)xid)) & (HASHSIZE-1)) static struct hlist_head * hash_list; static struct list_head lru_head; @@ -127,8 +127,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp, int type) struct hlist_node *hn; struct hlist_head *rh; struct svc_cacherep *rp; - u32 xid = rqstp->rq_xid, - proto = rqstp->rq_prot, + __be32 xid = rqstp->rq_xid; + u32 proto = rqstp->rq_prot, vers = rqstp->rq_vers, proc = rqstp->rq_proc; unsigned long age; @@ -258,7 +258,7 @@ found_entry: * In this case, nfsd_cache_update is called with statp == NULL. */ void -nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, u32 *statp) +nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp) { struct svc_cacherep *rp; struct kvec *resv = &rqstp->rq_res.head[0], *cachv; diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 8067118..0aaccb0 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -491,7 +491,7 @@ out: } int -nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) +nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp) { struct svc_procedure *proc; kxdrproc_t xdr; diff --git a/include/linux/nfsd/cache.h b/include/linux/nfsd/cache.h index c3a3557..007480c 100644 --- a/include/linux/nfsd/cache.h +++ b/include/linux/nfsd/cache.h @@ -26,14 +26,14 @@ struct svc_cacherep { c_type, /* status, buffer */ c_secure : 1; /* req came from port < 1024 */ struct sockaddr_in c_addr; - u32 c_xid; + __be32 c_xid; u32 c_prot; u32 c_proc; u32 c_vers; unsigned long c_timestamp; union { struct kvec u_vec; - u32 u_status; + __be32 u_status; } c_u; }; @@ -75,7 +75,7 @@ enum { void nfsd_cache_init(void); void nfsd_cache_shutdown(void); int nfsd_cache_lookup(struct svc_rqst *, int); -void nfsd_cache_update(struct svc_rqst *, int, u32 *); +void nfsd_cache_update(struct svc_rqst *, int, __be32 *); #endif /* __KERNEL__ */ #endif /* NFSCACHE_H */ diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h index 27666f5b..045e38c 100644 --- a/include/linux/nfsd/export.h +++ b/include/linux/nfsd/export.h @@ -117,7 +117,7 @@ struct svc_export * exp_parent(struct auth_domain *clp, struct cache_req *reqp); int exp_rootfh(struct auth_domain *, char *path, struct knfsd_fh *, int maxsize); -int exp_pseudoroot(struct auth_domain *, struct svc_fh *fhp, struct cache_req *creq); +__be32 exp_pseudoroot(struct auth_domain *, struct svc_fh *fhp, struct cache_req *creq); __be32 nfserrno(int errno); extern struct cache_detail svc_export_cache; diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 19a3c83..68d29b6 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -64,7 +64,7 @@ extern struct svc_serv *nfsd_serv; * Function prototypes. */ int nfsd_svc(unsigned short port, int nrservs); -int nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp); +int nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp); /* nfsd/vfs.c */ int fh_lock_parent(struct svc_fh *, struct dentry *); diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index 749bad1..f3b51d6 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -157,7 +157,7 @@ typedef struct svc_fh { __u64 fh_post_size; /* i_size */ unsigned long fh_post_blocks; /* i_blocks */ unsigned long fh_post_blksize;/* i_blksize */ - __u32 fh_post_rdev[2];/* i_rdev */ + __be32 fh_post_rdev[2];/* i_rdev */ struct timespec fh_post_atime; /* i_atime */ struct timespec fh_post_mtime; /* i_mtime */ struct timespec fh_post_ctime; /* i_ctime */ diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index a597e2e..c3673f4 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -125,7 +125,7 @@ struct nfs4_client { char cl_recdir[HEXDIR_LEN]; /* recovery dir */ nfs4_verifier cl_verifier; /* generated by client */ time_t cl_time; /* time of last lease renewal */ - u32 cl_addr; /* client ipaddress */ + __be32 cl_addr; /* client ipaddress */ struct svc_cred cl_cred; /* setclientid principal */ clientid_t cl_clientid; /* generated by server */ nfs4_verifier cl_confirm; /* generated by server */ -- cgit v0.10.2 From a90b061c0bf712961cea40d9c916b300073d12e5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:03 -0700 Subject: [PATCH] nfsd: nfs_replay_me We are using NFS_REPLAY_ME as a special error value that is never leaked to clients. That works fine; the only problem is mixing host- and network- endian values in the same objects. Network-endian equivalent would work just as fine; switch to it. Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 6382394..0a7bbdc 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -177,7 +177,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open /* check seqid for replay. set nfs4_owner */ status = nfsd4_process_open1(open); - if (status == NFSERR_REPLAY_ME) { + if (status == nfserr_replay_me) { struct nfs4_replay *rp = &open->op_stateowner->so_replay; fh_put(current_fh); current_fh->fh_handle.fh_size = rp->rp_openfh_len; @@ -188,7 +188,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open dprintk("nfsd4_open: replay failed" " restoring previous filehandle\n"); else - status = NFSERR_REPLAY_ME; + status = nfserr_replay_me; } if (status) goto out; @@ -937,7 +937,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, } encode_op: - if (op->status == NFSERR_REPLAY_ME) { + if (op->status == nfserr_replay_me) { op->replay = &replay_owner->so_replay; nfsd4_encode_replay(resp, op); status = op->status = op->replay->rp_status; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 2e468c9..293b649 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1477,7 +1477,7 @@ nfsd4_process_open1(struct nfsd4_open *open) } if (open->op_seqid == sop->so_seqid - 1) { if (sop->so_replay.rp_buflen) - return NFSERR_REPLAY_ME; + return nfserr_replay_me; /* The original OPEN failed so spectacularly * that we don't even have replay data saved! * Therefore, we have no choice but to continue @@ -2233,7 +2233,7 @@ check_replay: if (seqid == sop->so_seqid - 1) { dprintk("NFSD: preprocess_seqid_op: retransmission?\n"); /* indicate replay to calling function */ - return NFSERR_REPLAY_ME; + return nfserr_replay_me; } printk("NFSD: preprocess_seqid_op: bad seqid (expected %d, got %d)\n", sop->so_seqid, seqid); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 68d29b6..eb23114 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -238,6 +238,7 @@ void nfsd_lockd_shutdown(void); #define nfserr_badname __constant_htonl(NFSERR_BADNAME) #define nfserr_cb_path_down __constant_htonl(NFSERR_CB_PATH_DOWN) #define nfserr_locked __constant_htonl(NFSERR_LOCKED) +#define nfserr_replay_me __constant_htonl(NFSERR_REPLAY_ME) /* error codes for internal use */ /* if a request fails due to kmalloc failure, it gets dropped. -- cgit v0.10.2 From e51959faa61278c762389802faf8ba1a40676628 Mon Sep 17 00:00:00 2001 From: Zachary Amsden Date: Thu, 19 Oct 2006 23:29:04 -0700 Subject: [PATCH] Fix potential interrupts during alternative patching Interrupts must be disabled during alternative instruction patching. On systems with high timer IRQ rates, or when running in an emulator, timing differences can result in random kernel panics because of running partially patched instructions. This doesn't yet fix NMIs, which requires extricating the patch code from the late bug checking and is logically separate (and also less likely to cause problems). Signed-off-by: Zachary Amsden Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c index 28ab806..583c238 100644 --- a/arch/i386/kernel/alternative.c +++ b/arch/i386/kernel/alternative.c @@ -344,6 +344,7 @@ void alternatives_smp_switch(int smp) void __init alternative_instructions(void) { + unsigned long flags; if (no_replacement) { printk(KERN_INFO "(SMP-)alternatives turned off\n"); free_init_pages("SMP alternatives", @@ -351,6 +352,8 @@ void __init alternative_instructions(void) (unsigned long)__smp_alt_end); return; } + + local_irq_save(flags); apply_alternatives(__alt_instructions, __alt_instructions_end); /* switch to patch-once-at-boottime-only mode and free the @@ -386,4 +389,5 @@ void __init alternative_instructions(void) alternatives_smp_switch(0); } #endif + local_irq_restore(flags); } -- cgit v0.10.2 From 6220ec7844fda2686496013a66b5b9169976b991 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 19 Oct 2006 23:29:05 -0700 Subject: [PATCH] highest_possible_node_id() linkage fix Qooting Adrian: - net/sunrpc/svc.c uses highest_possible_node_id() - include/linux/nodemask.h says highest_possible_node_id() is out-of-line #if MAX_NUMNODES > 1 - the out-of-line highest_possible_node_id() is in lib/cpumask.c - lib/Makefile: lib-$(CONFIG_SMP) += cpumask.o CONFIG_ARCH_DISCONTIGMEM_ENABLE=y, CONFIG_SMP=n, CONFIG_SUNRPC=y -> highest_possible_node_id() is used in net/sunrpc/svc.c CONFIG_NODES_SHIFT defined and > 0 -> include/linux/numa.h: MAX_NUMNODES > 1 -> compile error The bug is not present on architectures where ARCH_DISCONTIGMEM_ENABLE depends on NUMA (but m32r isn't the only affected architecture). So move the function into page_alloc.c Cc: Adrian Bunk Cc: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/cpumask.c b/lib/cpumask.c index 7a2a73f..3a67dc5 100644 --- a/lib/cpumask.c +++ b/lib/cpumask.c @@ -43,19 +43,3 @@ int __any_online_cpu(const cpumask_t *mask) return cpu; } EXPORT_SYMBOL(__any_online_cpu); - -#if MAX_NUMNODES > 1 -/* - * Find the highest possible node id. - */ -int highest_possible_node_id(void) -{ - unsigned int node; - unsigned int highest = 0; - - for_each_node_mask(node, node_possible_map) - highest = node; - return highest; -} -EXPORT_SYMBOL(highest_possible_node_id); -#endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index afee38f..ebd425c 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3120,3 +3120,19 @@ unsigned long page_to_pfn(struct page *page) EXPORT_SYMBOL(pfn_to_page); EXPORT_SYMBOL(page_to_pfn); #endif /* CONFIG_OUT_OF_LINE_PFN_TO_PAGE */ + +#if MAX_NUMNODES > 1 +/* + * Find the highest possible node id. + */ +int highest_possible_node_id(void) +{ + unsigned int node; + unsigned int highest = 0; + + for_each_node_mask(node, node_possible_map) + highest = node; + return highest; +} +EXPORT_SYMBOL(highest_possible_node_id); +#endif -- cgit v0.10.2 From fca4edb48b693d9df7de4c42d91b4158d8b7e347 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Thu, 19 Oct 2006 23:29:06 -0700 Subject: [PATCH] drivers/isdn: ioremap balanced with iounmap ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. Signed-off-by: Amol Lad Acked-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/hisax/diva.c b/drivers/isdn/hisax/diva.c index 7e95f04..3dacfff 100644 --- a/drivers/isdn/hisax/diva.c +++ b/drivers/isdn/hisax/diva.c @@ -716,8 +716,10 @@ release_io_diva(struct IsdnCardState *cs) *cfg = 0; /* disable INT0/1 */ *cfg = 2; /* reset pending INT0 */ - iounmap((void *)cs->hw.diva.cfg_reg); - iounmap((void *)cs->hw.diva.pci_cfg); + if (cs->hw.diva.cfg_reg) + iounmap((void *)cs->hw.diva.cfg_reg); + if (cs->hw.diva.pci_cfg) + iounmap((void *)cs->hw.diva.pci_cfg); return; } else if (cs->subtyp != DIVA_IPAC_ISA) { del_timer(&cs->hw.diva.tl); @@ -734,6 +736,23 @@ release_io_diva(struct IsdnCardState *cs) } static void +iounmap_diva(struct IsdnCardState *cs) +{ + if ((cs->subtyp == DIVA_IPAC_PCI) || (cs->subtyp == DIVA_IPACX_PCI)) { + if (cs->hw.diva.cfg_reg) { + iounmap((void *)cs->hw.diva.cfg_reg); + cs->hw.diva.cfg_reg = 0; + } + if (cs->hw.diva.pci_cfg) { + iounmap((void *)cs->hw.diva.pci_cfg); + cs->hw.diva.pci_cfg = 0; + } + } + + return; +} + +static void reset_diva(struct IsdnCardState *cs) { if (cs->subtyp == DIVA_IPAC_ISA) { @@ -1069,11 +1088,13 @@ setup_diva(struct IsdnCard *card) if (!cs->irq) { printk(KERN_WARNING "Diva: No IRQ for PCI card found\n"); + iounmap_diva(cs); return(0); } if (!cs->hw.diva.cfg_reg) { printk(KERN_WARNING "Diva: No IO-Adr for PCI card found\n"); + iounmap_diva(cs); return(0); } cs->irq_flags |= IRQF_SHARED; @@ -1123,6 +1144,7 @@ ready: CardType[card->typ], cs->hw.diva.cfg_reg, cs->hw.diva.cfg_reg + bytecnt); + iounmap_diva(cs); return (0); } } diff --git a/drivers/isdn/hysdn/boardergo.c b/drivers/isdn/hysdn/boardergo.c index 8bbe33a..82e42a8 100644 --- a/drivers/isdn/hysdn/boardergo.c +++ b/drivers/isdn/hysdn/boardergo.c @@ -403,7 +403,7 @@ ergo_releasehardware(hysdn_card * card) free_irq(card->irq, card); /* release interrupt */ release_region(card->iobase + PCI9050_INTR_REG, 1); /* release all io ports */ release_region(card->iobase + PCI9050_USER_IO, 1); - vfree(card->dpram); + iounmap(card->dpram); card->dpram = NULL; /* release shared mem */ } /* ergo_releasehardware */ -- cgit v0.10.2 From ca926e80dcfd18adaf7c4304935da4cc8ded8364 Mon Sep 17 00:00:00 2001 From: Satoru Takeuchi Date: Thu, 19 Oct 2006 23:29:06 -0700 Subject: [PATCH] doc: fixing cpu-hotplug documentation Fixing cpu-hotplug documentation as follows: - moving confusing asterisk on additional_cpus descrition - fixing some typos - unifying indentation for source code and command line example Signed-off-by: Satoru Takeuchi Cc: Ashok Raj Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt index bc107cb..4868c34 100644 --- a/Documentation/cpu-hotplug.txt +++ b/Documentation/cpu-hotplug.txt @@ -46,7 +46,7 @@ maxcpus=n Restrict boot time cpus to n. Say if you have 4 cpus, using maxcpus=2 will only boot 2. You can choose to bring the other cpus later online, read FAQ's for more info. -additional_cpus*=n Use this to limit hotpluggable cpus. This option sets +additional_cpus=n (*) Use this to limit hotpluggable cpus. This option sets cpu_possible_map = cpu_present_map + additional_cpus (*) Option valid only for following architectures @@ -101,15 +101,15 @@ cpu_possible_map/for_each_possible_cpu() to iterate. Never use anything other than cpumask_t to represent bitmap of CPUs. -#include + #include -for_each_possible_cpu - Iterate over cpu_possible_map -for_each_online_cpu - Iterate over cpu_online_map -for_each_present_cpu - Iterate over cpu_present_map -for_each_cpu_mask(x,mask) - Iterate over some random collection of cpu mask. + for_each_possible_cpu - Iterate over cpu_possible_map + for_each_online_cpu - Iterate over cpu_online_map + for_each_present_cpu - Iterate over cpu_present_map + for_each_cpu_mask(x,mask) - Iterate over some random collection of cpu mask. -#include -lock_cpu_hotplug() and unlock_cpu_hotplug(): + #include + lock_cpu_hotplug() and unlock_cpu_hotplug(): The above calls are used to inhibit cpu hotplug operations. While holding the cpucontrol mutex, cpu_online_map will not change. If you merely need to avoid @@ -120,7 +120,7 @@ will work as long as stop_machine_run() is used to take a cpu down. CPU Hotplug - Frequently Asked Questions. -Q: How to i enable my kernel to support CPU hotplug? +Q: How to enable my kernel to support CPU hotplug? A: When doing make defconfig, Enable CPU hotplug support "Processor type and Features" -> Support for Hotpluggable CPUs @@ -141,39 +141,39 @@ A: You should now notice an entry in sysfs. Check if sysfs is mounted, using the "mount" command. You should notice an entry as shown below in the output. -.... -none on /sys type sysfs (rw) -.... + .... + none on /sys type sysfs (rw) + .... -if this is not mounted, do the following. +If this is not mounted, do the following. -#mkdir /sysfs -#mount -t sysfs sys /sys + #mkdir /sysfs + #mount -t sysfs sys /sys -now you should see entries for all present cpu, the following is an example +Now you should see entries for all present cpu, the following is an example in a 8-way system. -#pwd -#/sys/devices/system/cpu -#ls -l -total 0 -drwxr-xr-x 10 root root 0 Sep 19 07:44 . -drwxr-xr-x 13 root root 0 Sep 19 07:45 .. -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu0 -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu1 -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu2 -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu3 -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu4 -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu5 -drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu6 -drwxr-xr-x 3 root root 0 Sep 19 07:48 cpu7 + #pwd + #/sys/devices/system/cpu + #ls -l + total 0 + drwxr-xr-x 10 root root 0 Sep 19 07:44 . + drwxr-xr-x 13 root root 0 Sep 19 07:45 .. + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu0 + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu1 + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu2 + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu3 + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu4 + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu5 + drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu6 + drwxr-xr-x 3 root root 0 Sep 19 07:48 cpu7 Under each directory you would find an "online" file which is the control file to logically online/offline a processor. Q: Does hot-add/hot-remove refer to physical add/remove of cpus? A: The usage of hot-add/remove may not be very consistently used in the code. -CONFIG_CPU_HOTPLUG enables logical online/offline capability in the kernel. +CONFIG_HOTPLUG_CPU enables logical online/offline capability in the kernel. To support physical addition/removal, one would need some BIOS hooks and the platform should have something like an attention button in PCI hotplug. CONFIG_ACPI_HOTPLUG_CPU enables ACPI support for physical add/remove of CPUs. @@ -181,17 +181,17 @@ CONFIG_ACPI_HOTPLUG_CPU enables ACPI support for physical add/remove of CPUs. Q: How do i logically offline a CPU? A: Do the following. -#echo 0 > /sys/devices/system/cpu/cpuX/online + #echo 0 > /sys/devices/system/cpu/cpuX/online -once the logical offline is successful, check +Once the logical offline is successful, check -#cat /proc/interrupts + #cat /proc/interrupts -you should now not see the CPU that you removed. Also online file will report +You should now not see the CPU that you removed. Also online file will report the state as 0 when a cpu if offline and 1 when its online. -#To display the current cpu state. -#cat /sys/devices/system/cpu/cpuX/online + #To display the current cpu state. + #cat /sys/devices/system/cpu/cpuX/online Q: Why cant i remove CPU0 on some systems? A: Some architectures may have some special dependency on a certain CPU. @@ -234,8 +234,8 @@ Q: If i have some kernel code that needs to be aware of CPU arrival and departure, how to i arrange for proper notification? A: This is what you would need in your kernel code to receive notifications. - #include - static int __cpuinit foobar_cpu_callback(struct notifier_block *nfb, + #include + static int __cpuinit foobar_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -279,10 +279,10 @@ Q: I don't see my action being called for all CPUs already up and running? A: Yes, CPU notifiers are called only when new CPUs are on-lined or offlined. If you need to perform some action for each cpu already in the system, then - for_each_online_cpu(i) { + for_each_online_cpu(i) { foobar_cpu_callback(&foobar_cpu_notifier, CPU_UP_PREPARE, i); - foobar_cpu_callback(&foobar-cpu_notifier, CPU_ONLINE, i); - } + foobar_cpu_callback(&foobar_cpu_notifier, CPU_ONLINE, i); + } Q: If i would like to develop cpu hotplug support for a new architecture, what do i need at a minimum? @@ -307,38 +307,38 @@ Q: I need to ensure that a particular cpu is not removed when there is some work specific to this cpu is in progress. A: First switch the current thread context to preferred cpu - int my_func_on_cpu(int cpu) - { - cpumask_t saved_mask, new_mask = CPU_MASK_NONE; - int curr_cpu, err = 0; - - saved_mask = current->cpus_allowed; - cpu_set(cpu, new_mask); - err = set_cpus_allowed(current, new_mask); - - if (err) - return err; - - /* - * If we got scheduled out just after the return from - * set_cpus_allowed() before running the work, this ensures - * we stay locked. - */ - curr_cpu = get_cpu(); - - if (curr_cpu != cpu) { - err = -EAGAIN; - goto ret; - } else { - /* - * Do work : But cant sleep, since get_cpu() disables preempt - */ - } - ret: - put_cpu(); - set_cpus_allowed(current, saved_mask); - return err; - } + int my_func_on_cpu(int cpu) + { + cpumask_t saved_mask, new_mask = CPU_MASK_NONE; + int curr_cpu, err = 0; + + saved_mask = current->cpus_allowed; + cpu_set(cpu, new_mask); + err = set_cpus_allowed(current, new_mask); + + if (err) + return err; + + /* + * If we got scheduled out just after the return from + * set_cpus_allowed() before running the work, this ensures + * we stay locked. + */ + curr_cpu = get_cpu(); + + if (curr_cpu != cpu) { + err = -EAGAIN; + goto ret; + } else { + /* + * Do work : But cant sleep, since get_cpu() disables preempt + */ + } + ret: + put_cpu(); + set_cpus_allowed(current, saved_mask); + return err; + } Q: How do we determine how many CPUs are available for hotplug. -- cgit v0.10.2 From c4ec7b0de4bc18ccb4380de638550984d9a65c25 Mon Sep 17 00:00:00 2001 From: Dmitriy Monakhov Date: Thu, 19 Oct 2006 23:29:08 -0700 Subject: [PATCH] mm: D-cache aliasing issue in cow_user_page --=-=-= from mm/memory.c: 1434 static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va) 1435 { 1436 /* 1437 * If the source page was a PFN mapping, we don't have 1438 * a "struct page" for it. We do a best-effort copy by 1439 * just copying from the original user address. If that 1440 * fails, we just zero-fill it. Live with it. 1441 */ 1442 if (unlikely(!src)) { 1443 void *kaddr = kmap_atomic(dst, KM_USER0); 1444 void __user *uaddr = (void __user *)(va & PAGE_MASK); 1445 1446 /* 1447 * This really shouldn't fail, because the page is there 1448 * in the page tables. But it might just be unreadable, 1449 * in which case we just give up and fill the result with 1450 * zeroes. 1451 */ 1452 if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE)) 1453 memset(kaddr, 0, PAGE_SIZE); 1454 kunmap_atomic(kaddr, KM_USER0); #### D-cache have to be flushed here. #### It seems it is just forgotten. 1455 return; 1456 1457 } 1458 copy_user_highpage(dst, src, va); #### Ok here. flush_dcache_page() called from this func if arch need it 1459 } Following is the patch fix this issue: Signed-off-by: Dmitriy Monakhov Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memory.c b/mm/memory.c index b5a4aad..156861f 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1452,6 +1452,7 @@ static inline void cow_user_page(struct page *dst, struct page *src, unsigned lo if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE)) memset(kaddr, 0, PAGE_SIZE); kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(dst); return; } -- cgit v0.10.2 From 1cd441f99819abdd2d919ff13e8c75af58a0fd9c Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 19 Oct 2006 23:29:09 -0700 Subject: [PATCH] ipmi: fix return codes in failure case These returns should be negative, like the others in this function. Signed-off-by: Dave Jones Acked-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 24825bd..e5cfb1f 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -1789,7 +1789,7 @@ static int __devinit ipmi_pci_probe(struct pci_dev *pdev, info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) - return ENOMEM; + return -ENOMEM; info->addr_source = "PCI"; @@ -1810,7 +1810,7 @@ static int __devinit ipmi_pci_probe(struct pci_dev *pdev, kfree(info); printk(KERN_INFO "ipmi_si: %s: Unknown IPMI type: %d\n", pci_name(pdev), class_type); - return ENOMEM; + return -ENOMEM; } rv = pci_enable_device(pdev); -- cgit v0.10.2 From b95936cb9267e4c90a0b92361609ef5fd85a0a5f Mon Sep 17 00:00:00 2001 From: Doug Warzecha Date: Thu, 19 Oct 2006 23:29:09 -0700 Subject: [PATCH] firmware/dcdbas: add size check in smi_data_write Add a size check in smi_data_write to prevent possible wrapping problems with large pos values when calling smi_data_buf_realloc on 32-bit. Signed-off-by: Doug Warzecha Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c index 8bcb58c..1865b56 100644 --- a/drivers/firmware/dcdbas.c +++ b/drivers/firmware/dcdbas.c @@ -8,7 +8,7 @@ * * See Documentation/dcdbas.txt for more information. * - * Copyright (C) 1995-2005 Dell Inc. + * Copyright (C) 1995-2006 Dell Inc. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License v2.0 as published by @@ -40,7 +40,7 @@ #include "dcdbas.h" #define DRIVER_NAME "dcdbas" -#define DRIVER_VERSION "5.6.0-2" +#define DRIVER_VERSION "5.6.0-3.2" #define DRIVER_DESCRIPTION "Dell Systems Management Base Driver" static struct platform_device *dcdbas_pdev; @@ -175,6 +175,9 @@ static ssize_t smi_data_write(struct kobject *kobj, char *buf, loff_t pos, { ssize_t ret; + if ((pos + count) > MAX_SMI_DATA_BUF_SIZE) + return -EINVAL; + mutex_lock(&smi_data_lock); ret = smi_data_buf_realloc(pos + count); -- cgit v0.10.2 From 82591e6ea234762eeaa8b2337fe060ed438c18dc Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Thu, 19 Oct 2006 23:29:10 -0700 Subject: [PATCH] mm: more commenting on lock ordering Clarify lockorder comments now that sys_msync dropps mmap_sem before calling do_fsync. Signed-off-by: Nick Piggin Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/filemap.c b/mm/filemap.c index 57faa8d..8558732 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -75,8 +75,8 @@ generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * ->mmap_sem * ->lock_page (access_process_vm) * - * ->mmap_sem - * ->i_mutex (msync) + * ->i_mutex (generic_file_buffered_write) + * ->mmap_sem (fault_in_pages_readable->do_page_fault) * * ->i_mutex * ->i_alloc_sem (various) diff --git a/mm/rmap.c b/mm/rmap.c index a9136d8..d8a842a 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -21,27 +21,21 @@ * Lock ordering in mm: * * inode->i_mutex (while writing or truncating, not reading or faulting) - * inode->i_alloc_sem - * - * When a page fault occurs in writing from user to file, down_read - * of mmap_sem nests within i_mutex; in sys_msync, i_mutex nests within - * down_read of mmap_sem; i_mutex and down_write of mmap_sem are never - * taken together; in truncation, i_mutex is taken outermost. - * - * mm->mmap_sem - * page->flags PG_locked (lock_page) - * mapping->i_mmap_lock - * anon_vma->lock - * mm->page_table_lock or pte_lock - * zone->lru_lock (in mark_page_accessed, isolate_lru_page) - * swap_lock (in swap_duplicate, swap_info_get) - * mmlist_lock (in mmput, drain_mmlist and others) - * mapping->private_lock (in __set_page_dirty_buffers) - * inode_lock (in set_page_dirty's __mark_inode_dirty) - * sb_lock (within inode_lock in fs/fs-writeback.c) - * mapping->tree_lock (widely used, in set_page_dirty, - * in arch-dependent flush_dcache_mmap_lock, - * within inode_lock in __sync_single_inode) + * inode->i_alloc_sem (vmtruncate_range) + * mm->mmap_sem + * page->flags PG_locked (lock_page) + * mapping->i_mmap_lock + * anon_vma->lock + * mm->page_table_lock or pte_lock + * zone->lru_lock (in mark_page_accessed, isolate_lru_page) + * swap_lock (in swap_duplicate, swap_info_get) + * mmlist_lock (in mmput, drain_mmlist and others) + * mapping->private_lock (in __set_page_dirty_buffers) + * inode_lock (in set_page_dirty's __mark_inode_dirty) + * sb_lock (within inode_lock in fs/fs-writeback.c) + * mapping->tree_lock (widely used, in set_page_dirty, + * in arch-dependent flush_dcache_mmap_lock, + * within inode_lock in __sync_single_inode) */ #include -- cgit v0.10.2 From 3e2a532b26b491706bd8b5c7cfc8d767b43b8f36 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 19 Oct 2006 23:29:11 -0700 Subject: [PATCH] ext3/4: fix J_ASSERT(transaction->t_updates > 0) in journal_stop() A disk generated some I/O error, after it, I hitted J_ASSERT(transaction->t_updates > 0) in journal_stop(). It seems to happened on ext3_truncate() path from stack trace. Then, maybe the following case may trigger J_ASSERT(transaction->t_updates > 0). ext3_truncate() -> ext3_free_branches() -> ext3_journal_test_restart() -> ext3_journal_restart() -> journal_restart() transaction->t_updates--; /* another process aborted journal */ -> start_this_handle() returns -EROFS without transaction->t_updates++; -> ext3_journal_stop() -> journal_stop() J_ASSERT(transaction->t_updates > 0) If journal was aborted in middle of journal_restart(), ext3_truncate() may trigger J_ASSERT(). Signed-off-by: OGAWA Hirofumi Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index e1b3c8a..d5c6304 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -1314,13 +1314,14 @@ int journal_stop(handle_t *handle) int old_handle_count, err; pid_t pid; - J_ASSERT(transaction->t_updates > 0); J_ASSERT(journal_current_handle() == handle); if (is_handle_aborted(handle)) err = -EIO; - else + else { + J_ASSERT(transaction->t_updates > 0); err = 0; + } if (--handle->h_ref > 0) { jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1, diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 149957b..b6cf2be 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1314,13 +1314,14 @@ int jbd2_journal_stop(handle_t *handle) int old_handle_count, err; pid_t pid; - J_ASSERT(transaction->t_updates > 0); J_ASSERT(journal_current_handle() == handle); if (is_handle_aborted(handle)) err = -EIO; - else + else { + J_ASSERT(transaction->t_updates > 0); err = 0; + } if (--handle->h_ref > 0) { jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1, -- cgit v0.10.2 From e05d722e4555cd54677b4c8431d9e81fd047ef7a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 19 Oct 2006 23:29:12 -0700 Subject: [PATCH] kernel/nsproxy.c: use kmemdup() Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 6ebdb82..674aceb 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -44,11 +44,9 @@ static inline struct nsproxy *clone_namespaces(struct nsproxy *orig) { struct nsproxy *ns; - ns = kmalloc(sizeof(struct nsproxy), GFP_KERNEL); - if (ns) { - memcpy(ns, orig, sizeof(struct nsproxy)); + ns = kmemdup(orig, sizeof(struct nsproxy), GFP_KERNEL); + if (ns) atomic_set(&ns->count, 1); - } return ns; } -- cgit v0.10.2 From 1a047060a99f274a7c52cfea8159e4142a14b8a7 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 19 Oct 2006 23:29:13 -0700 Subject: [PATCH] knfsd: fix race that can disable NFS server This patch is suitable for just about any 2.6 kernel. It should go in 2.6.19 and 2.6.18.2 and possible even the .17 and .16 stable series. This is a long standing bug that seems to have only recently become apparent, presumably due to increasing use of NFS over TCP - many distros seem to be making it the default. The SK_CONN bit gets set when a listening socket may be ready for an accept, just as SK_DATA is set when data may be available. It is entirely possible for svc_tcp_accept to be called with neither of these set. It doesn't happen often but there is a small race in svc_sock_enqueue as SK_CONN and SK_DATA are tested outside the spin_lock. They could be cleared immediately after the test and before the lock is gained. This normally shouldn't be a problem. The sockets are non-blocking so trying to read() or accept() when ther is nothing to do is not a problem. However: svc_tcp_recvfrom makes the decision "Should I accept() or should I read()" based on whether SK_CONN is set or not. This usually works but is not safe. The decision should be based on whether it is a TCP_LISTEN socket or a TCP_CONNECTED socket. Signed-off-by: Neil Brown Cc: Adrian Bunk Cc: Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 61e307c..96521f1 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -973,7 +973,7 @@ svc_tcp_recvfrom(struct svc_rqst *rqstp) return 0; } - if (test_bit(SK_CONN, &svsk->sk_flags)) { + if (svsk->sk_sk->sk_state == TCP_LISTEN) { svc_tcp_accept(svsk); svc_sock_received(svsk); return 0; -- cgit v0.10.2 From ac4e0aba7daf0a7c6ac20974070428481dc940f5 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 19 Oct 2006 23:29:14 -0700 Subject: [PATCH] one more ARM IRQ fix Fix one more compile breakage caused by the post -rc1 IRQ changes. Signed-off-by: Adrian Bunk Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/arm/mach-lh7a40x/common.h b/arch/arm/mach-lh7a40x/common.h index 18e8bb4..0ca20c6 100644 --- a/arch/arm/mach-lh7a40x/common.h +++ b/arch/arm/mach-lh7a40x/common.h @@ -15,4 +15,4 @@ extern void lh7a404_init_irq (void); extern void lh7a40x_clcd_init (void); extern void lh7a40x_init_board_irq (void); -#define IRQ_DISPATCH(irq) desc_handle_irq((irq),(irq_desc + irq), regs) +#define IRQ_DISPATCH(irq) desc_handle_irq((irq),(irq_desc + irq)) -- cgit v0.10.2 From b2ef7858db6394b758818358a43c7dd5f232bbcc Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 20 Oct 2006 10:40:48 -0700 Subject: Revert "[mv643xx] Add pci device table for auto module loading." This reverts commit 4596c75c23dde2623cbeec69357d5eb13d28387e as requested by Olaf Hering. It causes compile errors, and says Olaf: "This change is also wrong, the autoloading works perfect with 2.6.18, no need to add random PCI ids. See commit a0245f7ad5214cb00131d7cd176446e067c913dc, platform devices have now a modalias entry in sysfs. The network card is not a PCI device." Signed-off-by: Linus Torvalds diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index a4f861b..9997081 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -1557,12 +1557,6 @@ static void __exit mv643xx_cleanup_module(void) module_init(mv643xx_init_module); module_exit(mv643xx_cleanup_module); -static struct pci_device_id pci_marvell_mv64360[] = { - { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, PCI_DEVICE_ID_MARVELL_MV64360) }, - {} -}; -MODULE_DEVICE_TABLE(pci, pci_marvell_mv64360); - MODULE_LICENSE("GPL"); MODULE_AUTHOR( "Rabeeh Khoury, Assaf Hoffman, Matthew Dharm, Manish Lachwani" " and Dale Farnsworth"); -- cgit v0.10.2 From 79cd22d3ac921b9209bf813c7e75e6b69e74896c Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 12 Oct 2006 14:29:33 +0900 Subject: ocfs2: delete redundant memcmp() This patch deletes redundant memcmp() while looking up in rb tree. Signed-off-by: Akinbou Mita Signed-off-by: Mark Fasheh diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c index e1fceb8..d11753c 100644 --- a/fs/ocfs2/cluster/nodemanager.c +++ b/fs/ocfs2/cluster/nodemanager.c @@ -152,14 +152,16 @@ static struct o2nm_node *o2nm_node_ip_tree_lookup(struct o2nm_cluster *cluster, struct o2nm_node *node, *ret = NULL; while (*p) { + int cmp; + parent = *p; node = rb_entry(parent, struct o2nm_node, nd_ip_node); - if (memcmp(&ip_needle, &node->nd_ipv4_address, - sizeof(ip_needle)) < 0) + cmp = memcmp(&ip_needle, &node->nd_ipv4_address, + sizeof(ip_needle)); + if (cmp < 0) p = &(*p)->rb_left; - else if (memcmp(&ip_needle, &node->nd_ipv4_address, - sizeof(ip_needle)) > 0) + else if (cmp > 0) p = &(*p)->rb_right; else { ret = node; -- cgit v0.10.2 From 711a40fcaa83bfad87736544b69f6fdd6527482d Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Wed, 11 Oct 2006 12:23:02 -0700 Subject: ocfs2: remove spurious d_count check in ocfs2_rename() This was causing some folks to incorrectly get -EBUSY during rename. Signed-off-by: Sunil Mushran Signed-off-by: Mark Fasheh diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 259155f..a57b751 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -1085,14 +1085,6 @@ static int ocfs2_rename(struct inode *old_dir, BUG(); } - if (atomic_read(&old_dentry->d_count) > 2) { - shrink_dcache_parent(old_dentry); - if (atomic_read(&old_dentry->d_count) > 2) { - status = -EBUSY; - goto bail; - } - } - /* Assume a directory heirarchy thusly: * a/b/c * a/d -- cgit v0.10.2 From 0effef776ff95b7a6d6e48a2ef407ecaa8c21f96 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 3 Oct 2006 17:44:42 -0700 Subject: ocfs2: fix page zeroing during simple extends The page zeroing code was missing the region between old i_size and new i_size for those extends that didn't actually require a change in space allocation. Signed-off-by: Mark Fasheh diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index d9ba0a9..b499c32 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -728,31 +728,36 @@ static int ocfs2_extend_file(struct inode *inode, clusters_to_add = ocfs2_clusters_for_bytes(inode->i_sb, new_i_size) - OCFS2_I(inode)->ip_clusters; - if (clusters_to_add) { - /* - * protect the pages that ocfs2_zero_extend is going to - * be pulling into the page cache.. we do this before the - * metadata extend so that we don't get into the situation - * where we've extended the metadata but can't get the data - * lock to zero. - */ - ret = ocfs2_data_lock(inode, 1); - if (ret < 0) { - mlog_errno(ret); - goto out; - } + /* + * protect the pages that ocfs2_zero_extend is going to be + * pulling into the page cache.. we do this before the + * metadata extend so that we don't get into the situation + * where we've extended the metadata but can't get the data + * lock to zero. + */ + ret = ocfs2_data_lock(inode, 1); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + if (clusters_to_add) { ret = ocfs2_extend_allocation(inode, clusters_to_add); if (ret < 0) { mlog_errno(ret); goto out_unlock; } + } - ret = ocfs2_zero_extend(inode, (u64)new_i_size - tail_to_skip); - if (ret < 0) { - mlog_errno(ret); - goto out_unlock; - } + /* + * Call this even if we don't add any clusters to the tree. We + * still need to zero the area between the old i_size and the + * new i_size. + */ + ret = ocfs2_zero_extend(inode, (u64)new_i_size - tail_to_skip); + if (ret < 0) { + mlog_errno(ret); + goto out_unlock; } if (!tail_to_skip) { @@ -764,8 +769,7 @@ static int ocfs2_extend_file(struct inode *inode, } out_unlock: - if (clusters_to_add) /* this is the only case in which we lock */ - ocfs2_data_unlock(inode, 1); + ocfs2_data_unlock(inode, 1); out: return ret; -- cgit v0.10.2 From e2057c5a63821e17c8a54dab6db680c77ce7ee6c Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 3 Oct 2006 17:53:05 -0700 Subject: ocfs2: cond_resched() in ocfs2_zero_extend() The loop within ocfs2_zero_extend() can execute for a long time, causing spurious soft lockup warnings. Signed-off-by: Mark Fasheh diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index b499c32..1be74c4 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -30,6 +30,7 @@ #include #include #include +#include #define MLOG_MASK_PREFIX ML_INODE #include @@ -691,6 +692,12 @@ static int ocfs2_zero_extend(struct inode *inode, } start_off += sb->s_blocksize; + + /* + * Very large extends have the potential to lock up + * the cpu for extended periods of time. + */ + cond_resched(); } out: -- cgit v0.10.2 From 559c9ac391c046710bdeee5581dc5d9dda794881 Mon Sep 17 00:00:00 2001 From: Chandra Seetharaman Date: Tue, 10 Oct 2006 15:15:55 -0700 Subject: configfs: handle kzalloc() failure in check_perm() check_perm() does not drop the reference to the module when kzalloc() failure occurs. Signed-Off-By: Chandra Seetharaman Signed-off-by: Joel Becker Signed-off-by: Mark Fasheh diff --git a/fs/configfs/file.c b/fs/configfs/file.c index e6d5754..cf33fac 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -275,13 +275,14 @@ static int check_perm(struct inode * inode, struct file * file) * it in file->private_data for easy access. */ buffer = kzalloc(sizeof(struct configfs_buffer),GFP_KERNEL); - if (buffer) { - init_MUTEX(&buffer->sem); - buffer->needs_read_fill = 1; - buffer->ops = ops; - file->private_data = buffer; - } else + if (!buffer) { error = -ENOMEM; + goto Enomem; + } + init_MUTEX(&buffer->sem); + buffer->needs_read_fill = 1; + buffer->ops = ops; + file->private_data = buffer; goto Done; Einval: @@ -289,6 +290,7 @@ static int check_perm(struct inode * inode, struct file * file) goto Done; Eaccess: error = -EACCES; + Enomem: module_put(attr->ca_owner); Done: if (error && item) -- cgit v0.10.2 From e17e31e388b2e49ff1c9a2bdb39d7aeb2975c19a Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Fri, 20 Oct 2006 09:23:05 +0200 Subject: [PATCH] Fix up rpaphp driver for pci hotplug header move Use grep instead of make during interface changes. Signed-off-by: Olaf Hering Signed-off-by: Linus Torvalds diff --git a/drivers/pci/hotplug/rpaphp.h b/drivers/pci/hotplug/rpaphp.h index 310b618..2e7accf 100644 --- a/drivers/pci/hotplug/rpaphp.h +++ b/drivers/pci/hotplug/rpaphp.h @@ -28,7 +28,7 @@ #define _PPC64PHP_H #include -#include "pci_hotplug.h" +#include #define DR_INDICATOR 9002 #define DR_ENTITY_SENSE 9003 -- cgit v0.10.2 From c7a3bd177f248d01ee18a01d22048c80e071c331 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Fri, 20 Oct 2006 14:20:17 -0700 Subject: [PATCH] fix PXA2xx UDC compilation error This was apparently missed by the move to the generic IRQ code. Signed-off-by: Nicolas Pitre Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/usb/gadget/pxa2xx_udc.c b/drivers/usb/gadget/pxa2xx_udc.c index f42c00e..671c24b 100644 --- a/drivers/usb/gadget/pxa2xx_udc.c +++ b/drivers/usb/gadget/pxa2xx_udc.c @@ -43,11 +43,11 @@ #include #include #include +#include #include #include #include -#include #include #include #include -- cgit v0.10.2 From f248b6a34fdd726dd12b549fceba907b6df2771c Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:00 +0200 Subject: [PATCH] x86-64: Update defconfig Signed-off-by: Andi Kleen diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig index 47bfba6e..0f5d44e 100644 --- a/arch/x86_64/defconfig +++ b/arch/x86_64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc1 -# Thu Oct 5 13:04:43 2006 +# Linux kernel version: 2.6.19-rc2-git4 +# Sat Oct 21 03:38:52 2006 # CONFIG_X86_64=y CONFIG_64BIT=y @@ -335,8 +335,8 @@ CONFIG_IPV6=y # CONFIG_INET6_XFRM_MODE_TUNNEL is not set # CONFIG_INET6_XFRM_MODE_BEET is not set # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=y # CONFIG_IPV6_TUNNEL is not set -# CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set # CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set @@ -438,6 +438,13 @@ CONFIG_BLK_DEV_INITRD=y # CONFIG_ATA_OVER_ETH is not set # +# Misc devices +# +# CONFIG_IBM_ASM is not set +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y @@ -1008,6 +1015,7 @@ CONFIG_I2C_ISA=m # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -1059,12 +1067,6 @@ CONFIG_SENSORS_SMSC47B397=m # CONFIG_HWMON_DEBUG_CHIP is not set # -# Misc devices -# -# CONFIG_IBM_ASM is not set -# CONFIG_TIFM_CORE is not set - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set @@ -1196,7 +1198,6 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set -# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -1242,6 +1243,7 @@ CONFIG_USB_MON=y # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1318,6 +1320,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y CONFIG_EXT3_FS_POSIX_ACL=y # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1341,6 +1344,7 @@ CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set CONFIG_AUTOFS4_FS=y # CONFIG_FUSE_FS is not set +CONFIG_GENERIC_ACL=y # # CD-ROM/DVD Filesystems @@ -1418,7 +1422,6 @@ CONFIG_SUNRPC=y # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set # CONFIG_9P_FS is not set -CONFIG_GENERIC_ACL=y # # Partition Types @@ -1471,10 +1474,6 @@ CONFIG_NLS_ISO8859_15=y CONFIG_NLS_UTF8=y # -# Distributed Lock Manager -# - -# # Instrumentation Support # CONFIG_PROFILING=y @@ -1512,6 +1511,7 @@ CONFIG_DEBUG_FS=y CONFIG_UNWIND_INFO=y CONFIG_STACK_UNWIND=y # CONFIG_FORCED_INLINING is not set +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_LKDTM is not set # CONFIG_DEBUG_RODATA is not set -- cgit v0.10.2 From 13892de19eb9007ea47430c701bdbf69df71d883 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] i386: Update defconfig Signed-off-by: Andi Kleen diff --git a/arch/i386/defconfig b/arch/i386/defconfig index 60c0c02..97aacd6 100644 --- a/arch/i386/defconfig +++ b/arch/i386/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc1 -# Thu Oct 5 13:04:53 2006 +# Linux kernel version: 2.6.19-rc2-git4 +# Sat Oct 21 03:38:56 2006 # CONFIG_X86_32=y CONFIG_GENERIC_TIME=y @@ -380,8 +380,8 @@ CONFIG_INET6_XFRM_MODE_TRANSPORT=y CONFIG_INET6_XFRM_MODE_TUNNEL=y # CONFIG_INET6_XFRM_MODE_BEET is not set # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=y # CONFIG_IPV6_TUNNEL is not set -# CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set # CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set @@ -483,6 +483,13 @@ CONFIG_BLK_DEV_INITRD=y # CONFIG_ATA_OVER_ETH is not set # +# Misc devices +# +# CONFIG_IBM_ASM is not set +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y @@ -1024,6 +1031,7 @@ CONFIG_HANGCHECK_TIMER=y # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -1032,12 +1040,6 @@ CONFIG_HANGCHECK_TIMER=y # CONFIG_HWMON_VID is not set # -# Misc devices -# -# CONFIG_IBM_ASM is not set -# CONFIG_TIFM_CORE is not set - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set @@ -1169,7 +1171,6 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set -# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -1215,6 +1216,7 @@ CONFIG_USB_MON=y # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1284,6 +1286,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y CONFIG_EXT3_FS_POSIX_ACL=y # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1307,6 +1310,7 @@ CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set CONFIG_AUTOFS4_FS=y # CONFIG_FUSE_FS is not set +CONFIG_GENERIC_ACL=y # # CD-ROM/DVD Filesystems @@ -1384,7 +1388,6 @@ CONFIG_SUNRPC=y # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set # CONFIG_9P_FS is not set -CONFIG_GENERIC_ACL=y # # Partition Types @@ -1437,10 +1440,6 @@ CONFIG_NLS_ISO8859_15=y CONFIG_NLS_UTF8=y # -# Distributed Lock Manager -# - -# # Instrumentation Support # CONFIG_PROFILING=y @@ -1480,6 +1479,7 @@ CONFIG_DEBUG_BUGVERBOSE=y CONFIG_UNWIND_INFO=y CONFIG_STACK_UNWIND=y # CONFIG_FORCED_INLINING is not set +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_LKDTM is not set CONFIG_EARLY_PRINTK=y -- cgit v0.10.2 From 926fafebc48a3218fac675f12974f9a46473bd40 Mon Sep 17 00:00:00 2001 From: keith mannthey Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: x86_64 hot-add memory srat.c fix This patch corrects the logic used in srat.c to figure out what parsing what action to take when registering hot-add areas. Hot-add areas should only be added to the node information for the MEMORY_HOTPLUG_RESERVE case. When booting MEMORY_HOTPLUG_SPARSE hot-add areas on everything but the last node are getting include in the node data and during kernel boot the pages are setup then the kernel dies when the pages are used. This patch fixes this issue. Signed-off-by: Keith Mannthey Signed-off-by: Andi Kleen diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c index 3cc0544..1087e15 100644 --- a/arch/x86_64/mm/srat.c +++ b/arch/x86_64/mm/srat.c @@ -207,7 +207,7 @@ static inline int save_add_info(void) return hotadd_percent > 0; } #else -int update_end_of_memory(unsigned long end) {return 0;} +int update_end_of_memory(unsigned long end) {return -1;} static int hotadd_enough_memory(struct bootnode *nd) {return 1;} #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE static inline int save_add_info(void) {return 1;} @@ -337,7 +337,7 @@ acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma) push_node_boundaries(node, nd->start >> PAGE_SHIFT, nd->end >> PAGE_SHIFT); - if (ma->flags.hot_pluggable && !reserve_hotadd(node, start, end) < 0) { + if (ma->flags.hot_pluggable && (reserve_hotadd(node, start, end) < 0)) { /* Ignore hotadd region. Undo damage */ printk(KERN_NOTICE "SRAT: Hotplug region ignored\n"); *nd = oldnode; -- cgit v0.10.2 From 45edfd1db02f818b3dc7e4743ee8585af6b78f78 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: typo in __assign_irq_vector when updating pos for vector and offset typo with cpu instead of new_cpu Signed-off-by: Yinghai Lu Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 49e94f7..b848f48 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -651,12 +651,12 @@ next: if (vector == IA32_SYSCALL_VECTOR) goto next; for_each_cpu_mask(new_cpu, domain) - if (per_cpu(vector_irq, cpu)[vector] != -1) + if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; /* Found one! */ for_each_cpu_mask(new_cpu, domain) { - pos[cpu].vector = vector; - pos[cpu].offset = offset; + pos[new_cpu].vector = vector; + pos[new_cpu].offset = offset; } if (old_vector >= 0) { int old_cpu; -- cgit v0.10.2 From da8604cc2d6cd4cbde873c7ba308365e402ac7bf Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] i386: fix .cfi_signal_frame copy-n-paste error This was copied, pasted but not edited. Cc: Andi Kleen Cc: Jan Beulich Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 7cc0b18..2d9d756 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -51,8 +51,8 @@ cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) # is .cfi_signal_frame supported too? -cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) -AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) +cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) +AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) CFLAGS += $(cflags-y) -- cgit v0.10.2 From 469b1d8741a5970ad49f2b5a837811579ba0b6f2 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: Fix for arch/x86_64/pci/Makefile CFLAGS The arch/x86_64/pci directory was giving problems in a wierd cross-compile environment. The exact cause is unknown, but the Makefile used CFLAGS instead of EXTRA_CFLAGS. From what I can tell from Documentation/kbuild/makefiles.txt, CFLAGS should not be used for this, it should be EXTRA_CFLAGS. And it solves the cross-compile problem. Signed-off-by: Corey Minyard Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: Vojtech Pavlik Cc: Sam Ravnborg Signed-off-by: Andrew Morton diff --git a/arch/x86_64/pci/Makefile b/arch/x86_64/pci/Makefile index 1eb18f4..149aba0 100644 --- a/arch/x86_64/pci/Makefile +++ b/arch/x86_64/pci/Makefile @@ -3,7 +3,7 @@ # # Reuse the i386 PCI subsystem # -CFLAGS += -Iarch/i386/pci +EXTRA_CFLAGS += -Iarch/i386/pci obj-y := i386.o obj-$(CONFIG_PCI_DIRECT)+= direct.o -- cgit v0.10.2 From 73bb8919b33d42cf75a0ed89bc9ca6a7128665be Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: fix page align in e820 allocator Currently some code pieces assume that address returned by find_e820_area() are page aligned. But looks like find_e820_area() had no such intention and hence one might end up stomping over some of the data. One such case is bootmem allocator initialization code stomped over bss. This patch modified find_e820_area() to return page aligned address. This might be little wasteful of memory but at the same time probably it is easier to handle page aligned memory. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index b3f0908..a75c829 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -54,13 +54,13 @@ static inline int bad_addr(unsigned long *addrp, unsigned long size) /* various gunk below that needed for SMP startup */ if (addr < 0x8000) { - *addrp = 0x8000; + *addrp = PAGE_ALIGN(0x8000); return 1; } /* direct mapping tables of the kernel */ if (last >= table_start<= INITRD_START && addr < INITRD_START+INITRD_SIZE) { - *addrp = INITRD_START + INITRD_SIZE; + *addrp = PAGE_ALIGN(INITRD_START + INITRD_SIZE); return 1; } #endif /* kernel code */ - if (last >= __pa_symbol(&_text) && last < __pa_symbol(&_end)) { - *addrp = __pa_symbol(&_end); + if (last >= __pa_symbol(&_text) && addr < __pa_symbol(&_end)) { + *addrp = PAGE_ALIGN(__pa_symbol(&_end)); return 1; } if (last >= ebda_addr && addr < ebda_addr + ebda_size) { - *addrp = ebda_addr + ebda_size; + *addrp = PAGE_ALIGN(ebda_addr + ebda_size); return 1; } @@ -152,7 +152,7 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end, unsi continue; while (bad_addr(&addr, size) && addr+size <= ei->addr+ei->size) ; - last = addr + size; + last = PAGE_ALIGN(addr) + size; if (last > ei->addr + ei->size) continue; if (last > end) -- cgit v0.10.2 From cdfce1f5714fec7b24715f569b2fee1607350a6d Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86: Use -maccumulate-outgoing-args This avoids some problems with gcc 4.x and earlier generating invalid unwind information. In 4.1 the option is default when unwind information is enabled. And it seems to generate smaller code too, so it's probably a good thing on its own. With gcc 4.0: i386: 4683198 902112 480868 6066178 5c9002 vmlinux (before) 4449895 902112 480868 5832875 5900ab vmlinux (after) x86-64: 4939761 1449584 648216 7037561 6b6279 vmlinux (before) 4854193 1449584 648216 6951993 6a1439 vmlinux (after) On 4.1 it shouldn't make much difference because it is default when unwind is enabled anyways. Suggested by Michael Matz and Jan Beulich Cc: jbeulich@novell.com Signed-off-by: Andi Kleen diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 2d9d756..0677908 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -42,6 +42,10 @@ cflags-$(CONFIG_REGPARM) += -mregparm=3 # temporary until string.h is fixed cflags-y += -ffreestanding +# this works around some issues with generating unwind tables in older gccs +# newer gccs do it by default +cflags-y += -maccumulate-outgoing-args + # Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use # a lot more stack due to the lack of sharing of stacklots: CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-option,-fno-unit-at-a-time); fi ;) diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 1c0f18d..1397214 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -54,6 +54,10 @@ endif cflags-y += $(call cc-option,-funit-at-a-time) # prevent gcc from generating any FP code by mistake cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,) +# this works around some issues with generating unwind tables in older gccs +# newer gccs do it by default +cflags-y += -maccumulate-outgoing-args + # do binutils support CFI? cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) -- cgit v0.10.2 From 690a973f48b6ba2954465992c08e65059c8374fe Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: Speed up dwarf2 unwinder This changes the dwarf2 unwinder to do a binary search for CIEs instead of a linear work. The linker is unfortunately not able to build a proper lookup table at link time, instead it creates one at runtime as soon as the bootmem allocator is usable (so you'll continue using the linear lookup for the first [hopefully] few calls). The code should be ready to utilize a build-time created table once a fixed linker becomes available. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen diff --git a/Makefile b/Makefile index 62a1343..389ff0c 100644 --- a/Makefile +++ b/Makefile @@ -499,6 +499,7 @@ endif ifdef CONFIG_UNWIND_INFO CFLAGS += -fasynchronous-unwind-tables +LDFLAGS_vmlinux += --eh-frame-hdr endif ifdef CONFIG_DEBUG_INFO diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 69240b5..9d0d11c 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -125,6 +125,10 @@ *(__param) \ VMLINUX_SYMBOL(__stop___param) = .; \ } \ + \ + /* Unwind data binary search table */ \ + EH_FRAME_HDR \ + \ __end_rodata = .; \ . = ALIGN(4096); @@ -157,6 +161,18 @@ *(.kprobes.text) \ VMLINUX_SYMBOL(__kprobes_text_end) = .; +#ifdef CONFIG_STACK_UNWIND + /* Unwind data binary search table */ +#define EH_FRAME_HDR \ + .eh_frame_hdr : AT(ADDR(.eh_frame_hdr) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_unwind_hdr) = .; \ + *(.eh_frame_hdr) \ + VMLINUX_SYMBOL(__end_unwind_hdr) = .; \ + } +#else +#define EH_FRAME_HDR +#endif + /* DWARF debug sections. Symbols in the DWARF debugging sections are relative to the beginning of the section so we begin them at 0. */ diff --git a/include/linux/unwind.h b/include/linux/unwind.h index 73e1751..749928c 100644 --- a/include/linux/unwind.h +++ b/include/linux/unwind.h @@ -26,6 +26,7 @@ struct module; * Initialize unwind support. */ extern void unwind_init(void); +extern void unwind_setup(void); #ifdef CONFIG_MODULES @@ -73,6 +74,7 @@ extern int unwind_to_user(struct unwind_frame_info *); struct unwind_frame_info {}; static inline void unwind_init(void) {} +static inline void unwind_setup(void) {} #ifdef CONFIG_MODULES diff --git a/init/main.c b/init/main.c index ee12324..36f608a 100644 --- a/init/main.c +++ b/init/main.c @@ -503,6 +503,7 @@ asmlinkage void __init start_kernel(void) printk(KERN_NOTICE); printk(linux_banner); setup_arch(&command_line); + unwind_setup(); setup_per_cpu_areas(); smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */ diff --git a/kernel/unwind.c b/kernel/unwind.c index 2e23686..f7e50d1 100644 --- a/kernel/unwind.c +++ b/kernel/unwind.c @@ -11,13 +11,15 @@ #include #include -#include +#include +#include #include #include #include #include extern char __start_unwind[], __end_unwind[]; +extern const u8 __start_unwind_hdr[], __end_unwind_hdr[]; #define MAX_STACK_DEPTH 8 @@ -100,6 +102,8 @@ static struct unwind_table { } core, init; const void *address; unsigned long size; + const unsigned char *header; + unsigned long hdrsz; struct unwind_table *link; const char *name; } root_table; @@ -145,6 +149,10 @@ static struct unwind_table *find_table(unsigned long pc) return table; } +static unsigned long read_pointer(const u8 **pLoc, + const void *end, + signed ptrType); + static void init_unwind_table(struct unwind_table *table, const char *name, const void *core_start, @@ -152,14 +160,30 @@ static void init_unwind_table(struct unwind_table *table, const void *init_start, unsigned long init_size, const void *table_start, - unsigned long table_size) + unsigned long table_size, + const u8 *header_start, + unsigned long header_size) { + const u8 *ptr = header_start + 4; + const u8 *end = header_start + header_size; + table->core.pc = (unsigned long)core_start; table->core.range = core_size; table->init.pc = (unsigned long)init_start; table->init.range = init_size; table->address = table_start; table->size = table_size; + /* See if the linker provided table looks valid. */ + if (header_size <= 4 + || header_start[0] != 1 + || (void *)read_pointer(&ptr, end, header_start[1]) != table_start + || header_start[2] == DW_EH_PE_omit + || read_pointer(&ptr, end, header_start[2]) <= 0 + || header_start[3] == DW_EH_PE_omit) + header_start = NULL; + table->hdrsz = header_size; + smp_wmb(); + table->header = header_start; table->link = NULL; table->name = name; } @@ -169,7 +193,143 @@ void __init unwind_init(void) init_unwind_table(&root_table, "kernel", _text, _end - _text, NULL, 0, - __start_unwind, __end_unwind - __start_unwind); + __start_unwind, __end_unwind - __start_unwind, + __start_unwind_hdr, __end_unwind_hdr - __start_unwind_hdr); +} + +static const u32 bad_cie, not_fde; +static const u32 *cie_for_fde(const u32 *fde, const struct unwind_table *); +static signed fde_pointer_type(const u32 *cie); + +struct eh_frame_hdr_table_entry { + unsigned long start, fde; +}; + +static int cmp_eh_frame_hdr_table_entries(const void *p1, const void *p2) +{ + const struct eh_frame_hdr_table_entry *e1 = p1; + const struct eh_frame_hdr_table_entry *e2 = p2; + + return (e1->start > e2->start) - (e1->start < e2->start); +} + +static void swap_eh_frame_hdr_table_entries(void *p1, void *p2, int size) +{ + struct eh_frame_hdr_table_entry *e1 = p1; + struct eh_frame_hdr_table_entry *e2 = p2; + unsigned long v; + + v = e1->start; + e1->start = e2->start; + e2->start = v; + v = e1->fde; + e1->fde = e2->fde; + e2->fde = v; +} + +static void __init setup_unwind_table(struct unwind_table *table, + void *(*alloc)(unsigned long)) +{ + const u8 *ptr; + unsigned long tableSize = table->size, hdrSize; + unsigned n; + const u32 *fde; + struct { + u8 version; + u8 eh_frame_ptr_enc; + u8 fde_count_enc; + u8 table_enc; + unsigned long eh_frame_ptr; + unsigned int fde_count; + struct eh_frame_hdr_table_entry table[]; + } __attribute__((__packed__)) *header; + + if (table->header) + return; + + if (table->hdrsz) + printk(KERN_WARNING ".eh_frame_hdr for '%s' present but unusable\n", + table->name); + + if (tableSize & (sizeof(*fde) - 1)) + return; + + for (fde = table->address, n = 0; + tableSize > sizeof(*fde) && tableSize - sizeof(*fde) >= *fde; + tableSize -= sizeof(*fde) + *fde, fde += 1 + *fde / sizeof(*fde)) { + const u32 *cie = cie_for_fde(fde, table); + signed ptrType; + + if (cie == ¬_fde) + continue; + if (cie == NULL + || cie == &bad_cie + || (ptrType = fde_pointer_type(cie)) < 0) + return; + ptr = (const u8 *)(fde + 2); + if (!read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType)) + return; + ++n; + } + + if (tableSize || !n) + return; + + hdrSize = 4 + sizeof(unsigned long) + sizeof(unsigned int) + + 2 * n * sizeof(unsigned long); + header = alloc(hdrSize); + if (!header) + return; + header->version = 1; + header->eh_frame_ptr_enc = DW_EH_PE_abs|DW_EH_PE_native; + header->fde_count_enc = DW_EH_PE_abs|DW_EH_PE_data4; + header->table_enc = DW_EH_PE_abs|DW_EH_PE_native; + put_unaligned((unsigned long)table->address, &header->eh_frame_ptr); + BUILD_BUG_ON(offsetof(typeof(*header), fde_count) + % __alignof(typeof(header->fde_count))); + header->fde_count = n; + + BUILD_BUG_ON(offsetof(typeof(*header), table) + % __alignof(typeof(*header->table))); + for (fde = table->address, tableSize = table->size, n = 0; + tableSize; + tableSize -= sizeof(*fde) + *fde, fde += 1 + *fde / sizeof(*fde)) { + const u32 *cie = fde + 1 - fde[1] / sizeof(*fde); + + if (!fde[1]) + continue; /* this is a CIE */ + ptr = (const u8 *)(fde + 2); + header->table[n].start = read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + fde_pointer_type(cie)); + header->table[n].fde = (unsigned long)fde; + ++n; + } + WARN_ON(n != header->fde_count); + + sort(header->table, + n, + sizeof(*header->table), + cmp_eh_frame_hdr_table_entries, + swap_eh_frame_hdr_table_entries); + + table->hdrsz = hdrSize; + smp_wmb(); + table->header = (const void *)header; +} + +static void *__init balloc(unsigned long sz) +{ + return __alloc_bootmem_nopanic(sz, + sizeof(unsigned int), + __pa(MAX_DMA_ADDRESS)); +} + +void __init unwind_setup(void) +{ + setup_unwind_table(&root_table, balloc); } #ifdef CONFIG_MODULES @@ -193,7 +353,8 @@ void *unwind_add_table(struct module *module, init_unwind_table(table, module->name, module->module_core, module->core_size, module->module_init, module->init_size, - table_start, table_size); + table_start, table_size, + NULL, 0); if (last_table) last_table->link = table; @@ -303,6 +464,26 @@ static sleb128_t get_sleb128(const u8 **pcur, const u8 *end) return value; } +static const u32 *cie_for_fde(const u32 *fde, const struct unwind_table *table) +{ + const u32 *cie; + + if (!*fde || (*fde & (sizeof(*fde) - 1))) + return &bad_cie; + if (!fde[1]) + return ¬_fde; /* this is a CIE */ + if ((fde[1] & (sizeof(*fde) - 1)) + || fde[1] > (unsigned long)(fde + 1) - (unsigned long)table->address) + return NULL; /* this is not a valid FDE */ + cie = fde + 1 - fde[1] / sizeof(*fde); + if (*cie <= sizeof(*cie) + 4 + || *cie >= fde[1] - sizeof(*fde) + || (*cie & (sizeof(*cie) - 1)) + || cie[1]) + return NULL; /* this is not a (valid) CIE */ + return cie; +} + static unsigned long read_pointer(const u8 **pLoc, const void *end, signed ptrType) @@ -610,49 +791,108 @@ int unwind(struct unwind_frame_info *frame) unsigned i; signed ptrType = -1; uleb128_t retAddrReg = 0; - struct unwind_table *table; + const struct unwind_table *table; struct unwind_state state; if (UNW_PC(frame) == 0) return -EINVAL; if ((table = find_table(pc)) != NULL && !(table->size & (sizeof(*fde) - 1))) { - unsigned long tableSize = table->size; - - for (fde = table->address; - tableSize > sizeof(*fde) && tableSize - sizeof(*fde) >= *fde; - tableSize -= sizeof(*fde) + *fde, - fde += 1 + *fde / sizeof(*fde)) { - if (!*fde || (*fde & (sizeof(*fde) - 1))) - break; - if (!fde[1]) - continue; /* this is a CIE */ - if ((fde[1] & (sizeof(*fde) - 1)) - || fde[1] > (unsigned long)(fde + 1) - - (unsigned long)table->address) - continue; /* this is not a valid FDE */ - cie = fde + 1 - fde[1] / sizeof(*fde); - if (*cie <= sizeof(*cie) + 4 - || *cie >= fde[1] - sizeof(*fde) - || (*cie & (sizeof(*cie) - 1)) - || cie[1] - || (ptrType = fde_pointer_type(cie)) < 0) { - cie = NULL; /* this is not a (valid) CIE */ - continue; + const u8 *hdr = table->header; + unsigned long tableSize; + + smp_rmb(); + if (hdr && hdr[0] == 1) { + switch(hdr[3] & DW_EH_PE_FORM) { + case DW_EH_PE_native: tableSize = sizeof(unsigned long); break; + case DW_EH_PE_data2: tableSize = 2; break; + case DW_EH_PE_data4: tableSize = 4; break; + case DW_EH_PE_data8: tableSize = 8; break; + default: tableSize = 0; break; + } + ptr = hdr + 4; + end = hdr + table->hdrsz; + if (tableSize + && read_pointer(&ptr, end, hdr[1]) + == (unsigned long)table->address + && (i = read_pointer(&ptr, end, hdr[2])) > 0 + && i == (end - ptr) / (2 * tableSize) + && !((end - ptr) % (2 * tableSize))) { + do { + const u8 *cur = ptr + (i / 2) * (2 * tableSize); + + startLoc = read_pointer(&cur, + cur + tableSize, + hdr[3]); + if (pc < startLoc) + i /= 2; + else { + ptr = cur - tableSize; + i = (i + 1) / 2; + } + } while (startLoc && i > 1); + if (i == 1 + && (startLoc = read_pointer(&ptr, + ptr + tableSize, + hdr[3])) != 0 + && pc >= startLoc) + fde = (void *)read_pointer(&ptr, + ptr + tableSize, + hdr[3]); } + } + + if (fde != NULL) { + cie = cie_for_fde(fde, table); ptr = (const u8 *)(fde + 2); - startLoc = read_pointer(&ptr, - (const u8 *)(fde + 1) + *fde, - ptrType); - endLoc = startLoc - + read_pointer(&ptr, - (const u8 *)(fde + 1) + *fde, - ptrType & DW_EH_PE_indirect - ? ptrType - : ptrType & (DW_EH_PE_FORM|DW_EH_PE_signed)); - if (pc >= startLoc && pc < endLoc) - break; - cie = NULL; + if(cie != NULL + && cie != &bad_cie + && cie != ¬_fde + && (ptrType = fde_pointer_type(cie)) >= 0 + && read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType) == startLoc) { + if (!(ptrType & DW_EH_PE_indirect)) + ptrType &= DW_EH_PE_FORM|DW_EH_PE_signed; + endLoc = startLoc + + read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType); + if(pc >= endLoc) + fde = NULL; + } else + fde = NULL; + } + if (fde == NULL) { + for (fde = table->address, tableSize = table->size; + cie = NULL, tableSize > sizeof(*fde) + && tableSize - sizeof(*fde) >= *fde; + tableSize -= sizeof(*fde) + *fde, + fde += 1 + *fde / sizeof(*fde)) { + cie = cie_for_fde(fde, table); + if (cie == &bad_cie) { + cie = NULL; + break; + } + if (cie == NULL + || cie == ¬_fde + || (ptrType = fde_pointer_type(cie)) < 0) + continue; + ptr = (const u8 *)(fde + 2); + startLoc = read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType); + if (!startLoc) + continue; + if (!(ptrType & DW_EH_PE_indirect)) + ptrType &= DW_EH_PE_FORM|DW_EH_PE_signed; + endLoc = startLoc + + read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType); + if (pc >= startLoc && pc < endLoc) + break; + } } } if (cie != NULL) { -- cgit v0.10.2 From 7a71cef780404e8c90d23b1131142e158d94354b Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: x86_64 add NX mask for PTE entry If function change_page_attr_addr calls revert_page to revert to original pte value, mk_pte_phys does not mask NX bit. If NX bit is set on no NX hardware supported x86_64 machine, there is will be RSVD type page fault and system will crash. This patch adds NX mask bit for PTE entry. Signed-off-by: bibo,mao Signed-off-by: Andi Kleen diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h index 6899e77..0555c1c 100644 --- a/include/asm-x86_64/pgtable.h +++ b/include/asm-x86_64/pgtable.h @@ -366,6 +366,7 @@ static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot) { pte_t pte; pte_val(pte) = physpage | pgprot_val(pgprot); + pte_val(pte) &= __supported_pte_mask; return pte; } -- cgit v0.10.2 From 26fd5e084e470dbe8edc6f726fc918e89b9f988c Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] i386: Fix fake return address The fake return address was being set to __KERNEL_PDA, rather than 0. Push it earlier while %eax still equals 0. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: Andrew Morton diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index be9d883..ca31f18 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -317,7 +317,7 @@ is386: movl $2,%ecx # set MP movl %eax,%gs lldt %ax cld # gcc2 wants the direction flag cleared at all times - pushl %eax # fake return address + pushl $0 # fake return address for unwinder #ifdef CONFIG_SMP movb ready, %cl movb $1, ready -- cgit v0.10.2 From cc7d479fe56133e79840beffe9cb4fd193af93aa Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Fix ENOSYS in system call tracing This patch: - out of range system calls failing to return -ENOSYS under system call tracing [AK: split out from another patch by Jan as separate bugfix] Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 38a7b2d..038dcf7 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -315,6 +315,8 @@ tracesys: LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */ RESTORE_REST cmpq $__NR_syscall_max,%rax + movq $-ENOSYS,%rcx + cmova %rcx,%rax ja 1f movq %r10,%rcx /* fixup for C */ call *sys_call_table(,%rax,8) -- cgit v0.10.2 From 581910e2eb952e541b8ca9b5f551d6c124903b61 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Revert interrupt backlink changes They break more than they fix Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 038dcf7..e3eddde 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -537,8 +537,6 @@ END(stub_rt_sigreturn) 1: incl %gs:pda_irqcount cmoveq %gs:pda_irqstackptr,%rsp push %rbp # backlink for old unwinder - CFI_ADJUST_CFA_OFFSET 8 - CFI_REL_OFFSET rbp,0 /* * We entered an interrupt context - irqs are off: */ @@ -1178,7 +1176,6 @@ ENTRY(call_softirq) incl %gs:pda_irqcount cmove %gs:pda_irqstackptr,%rsp push %rbp # backlink for old unwinder - CFI_ADJUST_CFA_OFFSET 8 call __do_softirq leaveq CFI_DEF_CFA_REGISTER rsp -- cgit v0.10.2 From a1bae67243512ca30ceda48e3e24e25b543f8ab7 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] i386: Disable nmi watchdog on all ThinkPads Even newer Thinkpads have bugs in SMM code that causes hangs with NMI watchdog. Signed-off-by: Andi Kleen diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 3e8e3ad..eaafe23 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -219,11 +219,11 @@ static int __init check_nmi_watchdog(void) int cpu; /* Enable NMI watchdog for newer systems. - Actually it should be safe for most systems before 2004 too except - for some IBM systems that corrupt registers when NMI happens - during SMM. Unfortunately we don't have more exact information - on these and use this coarse check. */ - if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004) + Probably safe on most older systems too, but let's be careful. + IBM ThinkPads use INT10 inside SMM and that allows early NMI inside SMM + which hangs the system. Disable watchdog for all thinkpads */ + if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004 && + !dmi_name_in_vendors("ThinkPad")) nmi_watchdog = NMI_LOCAL_APIC; if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT)) diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c index b8b596d..37deee6 100644 --- a/drivers/firmware/dmi_scan.c +++ b/drivers/firmware/dmi_scan.c @@ -326,6 +326,26 @@ char *dmi_get_system_info(int field) } EXPORT_SYMBOL(dmi_get_system_info); + +/** + * dmi_name_in_vendors - Check if string is anywhere in the DMI vendor information. + * @str: Case sensitive Name + */ +int dmi_name_in_vendors(char *str) +{ + static int fields[] = { DMI_BIOS_VENDOR, DMI_BIOS_VERSION, DMI_SYS_VENDOR, + DMI_PRODUCT_NAME, DMI_PRODUCT_VERSION, DMI_BOARD_VENDOR, + DMI_BOARD_NAME, DMI_BOARD_VERSION, DMI_NONE }; + int i; + for (i = 0; fields[i] != DMI_NONE; i++) { + int f = fields[i]; + if (dmi_ident[f] && strstr(dmi_ident[f], str)) + return 1; + } + return 0; +} +EXPORT_SYMBOL(dmi_name_in_vendors); + /** * dmi_find_device - find onboard device by type/name * @type: device type or %DMI_DEV_TYPE_ANY to match all device types diff --git a/include/linux/dmi.h b/include/linux/dmi.h index 38dc403..904bf3d 100644 --- a/include/linux/dmi.h +++ b/include/linux/dmi.h @@ -69,6 +69,7 @@ extern struct dmi_device * dmi_find_device(int type, const char *name, struct dmi_device *from); extern void dmi_scan_machine(void); extern int dmi_get_year(int field); +extern int dmi_name_in_vendors(char *str); #else @@ -77,6 +78,7 @@ static inline char * dmi_get_system_info(int field) { return NULL; } static inline struct dmi_device * dmi_find_device(int type, const char *name, struct dmi_device *from) { return NULL; } static inline int dmi_get_year(int year) { return 0; } +static inline int dmi_name_in_vendors(char *s) { return 0; } #endif -- cgit v0.10.2 From 6bf2dafad18c119beb534cbb3d882fe7a6c3f529 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Use irq_domain in ioapic_retrigger_irq Thanks to YH Lu for spotting this. It appears I missed this function when I refactored allocate_irq_vector and introduced irq_domain, with the result that all retriggered irqs would go to cpu 0 even if we were not prepared to receive them there. While reviewing YH's patch I also noticed that this function was missing locking, and since I am now reading two values from two diffrent arrays that looks like a race we might be able to hit in the real world. Cc: Yinghai Lu Signed-off-by: Eric W. Biederman Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index b848f48..8a9a357 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1255,12 +1255,15 @@ static int ioapic_retrigger_irq(unsigned int irq) { cpumask_t mask; unsigned vector; + unsigned long flags; + spin_lock_irqsave(&vector_lock, flags); vector = irq_vector[irq]; cpus_clear(mask); - cpu_set(vector >> 8, mask); + cpu_set(first_cpu(irq_domain[irq]), mask); - send_IPI_mask(mask, vector & 0xff); + send_IPI_mask(mask, vector); + spin_unlock_irqrestore(&vector_lock, flags); return 1; } -- cgit v0.10.2 From 8cf2c51927bbeefafc25193d01b91f9ed3806e96 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86: Revert new unwind kernel stack termination Jan convinced me that it was unnecessary because the assembly stubs do this already on the stack. Cc: jbeulich@novell.com Signed-off-by: Andi Kleen diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 57d3759..1e1fa3e 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -336,7 +336,6 @@ extern void kernel_thread_helper(void); int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) { struct pt_regs regs; - int err; memset(®s, 0, sizeof(regs)); @@ -351,10 +350,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2; /* Ok, create the new process.. */ - err = do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL); - if (err == 0) /* terminate kernel stack */ - task_pt_regs(current)->eip = 0; - return err; + return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL); } EXPORT_SYMBOL(kernel_thread); diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index e3eddde..7d401b0 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -980,11 +980,6 @@ ENTRY(kernel_thread) call do_fork movq %rax,RAX(%rsp) xorl %edi,%edi - test %rax,%rax - jnz 1f - /* terminate stack in child */ - movq %rdi,RIP(%rsp) -1: /* * It isn't worth to check for reschedule here, -- cgit v0.10.2 From 84f404f695b16bd142c8dd9910d5a398f54fb044 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Put more than one cpu in TARGET_CPUS TARGET_CPUS is the default irq routing poicy. It specifies which cpus the kernel should aim an irq at. In physflat delivery mode we can route an irq to a single cpu. But that doesn't mean our default policy should only be a single cpu is allowed. By allowing the irq routing code to select from multiple cpus this enables systems with more irqs then we can service on a single processor to actually work. I just audited and tested the code and irqbalance doesn't care, and the io_apic.c doesn't care if we have extra cpus in the mask. Everything will use or assume we are using the lowest numbered cpu in the mask if we can't use them all. So this should result in no behavior changes except on systems that need it. Thanks for YH Lu for spotting this problem in his testing. Cc: Yinghai Lu Signed-off-by: Eric W. Biederman Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c index 0dfc223..7c01db8 100644 --- a/arch/x86_64/kernel/genapic_flat.c +++ b/arch/x86_64/kernel/genapic_flat.c @@ -153,7 +153,7 @@ struct genapic apic_flat = { static cpumask_t physflat_target_cpus(void) { - return cpumask_of_cpu(0); + return cpu_online_map; } static cpumask_t physflat_vector_allocation_domain(int cpu) -- cgit v0.10.2 From dbaab49f92ff6ae6255762a948375e4036cbdbd2 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 21 Oct 2006 18:37:03 +0200 Subject: [PATCH] x86-64: Overlapping program headers in physical addr space fix o A recent change to vmlinux.ld.S file broke kexec as now resulting vmlinux program headers are overlapping in physical address space. o Now all the vsyscall related sections are placed after data and after that mostly init data sections are placed. To avoid physical overlap among phdrs, there are three possible solutions. - Place vsyscall sections also in data phdrs instead of user - move vsyscal sections after init data in bss. - create another phdrs say data.init and move all the sections after vsyscall into this new phdr. o This patch implements the third solution. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Cc: Magnus Damm Cc: Andi Kleen Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index b9df2ab..1283614 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -17,6 +17,7 @@ PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ user PT_LOAD FLAGS(7); /* RWE */ + data.init PT_LOAD FLAGS(7); /* RWE */ note PT_NOTE FLAGS(4); /* R__ */ } SECTIONS @@ -131,7 +132,7 @@ SECTIONS . = ALIGN(8192); /* init_task */ .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { *(.data.init_task) - } :data + }:data.init . = ALIGN(4096); .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { -- cgit v0.10.2 From e70ea8c09db0e25ab58f84ba7f393e5c9125a8ee Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:03 +0200 Subject: [PATCH] x86-64: Revert timer routing behaviour back to 2.6.16 state By default route the 8254 over the 8259 and only disable it on ATI boards where this causes double timer interrupts. This should unbreak some Nvidia boards where the timer doesn't seem to tick of it isn't enabled in the 8259. At least one VIA board also seemed to have a little trouble with the disabled 8259. For 2.6.20 we'll try both dynamically without black listing, but I think for .19 this is the safer approach because it has been already well tested in earlier kernels. This also makes the x86-64 behaviour the same as i386. Command line options can change all this of course. Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index 208e38a..2b1245d 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -61,10 +61,11 @@ static void nvidia_bugs(void) static void ati_bugs(void) { -#if 1 /* for testing */ - printk("ATI board detected\n"); -#endif - /* No bugs right now */ + if (timer_over_8254 == 1) { + timer_over_8254 = 0; + printk(KERN_INFO + "ATI board detected. Disabling timer routing over 8254.\n"); + } } struct chipset { diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 8a9a357..b000017 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -57,7 +57,7 @@ static int no_timer_check; static int disable_timer_pin_1 __initdata; -int timer_over_8254 __initdata = 0; +int timer_over_8254 __initdata = 1; /* Where if anywhere is the i8259 connect in external int mode */ static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h index c181fef..e72cfcd 100644 --- a/include/asm-x86_64/proto.h +++ b/include/asm-x86_64/proto.h @@ -122,6 +122,8 @@ extern int fix_aperture; extern int reboot_force; extern int notsc_setup(char *); +extern int timer_over_8254; + extern int gsi_irq_sharing(int gsi); extern void smp_local_timer_interrupt(void); -- cgit v0.10.2 From 718ecac2ed7ae1b3d61388ddbff2938a377b1a11 Mon Sep 17 00:00:00 2001 From: Deepak Saxena Date: Fri, 20 Oct 2006 14:42:04 -0700 Subject: [PATCH] Update smc91x driver with ARM Versatile board info We need to specify a Versatile-specific SMC_IRQ_FLAGS value or the new generic IRQ layer will complain thusly: No IRQF_TRIGGER set_type function for IRQ 25 () Signed-off-by: Deepak Saxena Cc: Jeff Garzik Cc: Russell King Cc: Nicolas Pitre Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h index 0c9f1e7..a864016 100644 --- a/drivers/net/smc91x.h +++ b/drivers/net/smc91x.h @@ -416,6 +416,24 @@ static inline void LPD7_SMC_outsw (unsigned char* a, int r, #define SMC_IRQ_FLAGS (0) +#elif defined(CONFIG_ARCH_VERSATILE) + +#define SMC_CAN_USE_8BIT 1 +#define SMC_CAN_USE_16BIT 1 +#define SMC_CAN_USE_32BIT 1 +#define SMC_NOWAIT 1 + +#define SMC_inb(a, r) readb((a) + (r)) +#define SMC_inw(a, r) readw((a) + (r)) +#define SMC_inl(a, r) readl((a) + (r)) +#define SMC_outb(v, a, r) writeb(v, (a) + (r)) +#define SMC_outw(v, a, r) writew(v, (a) + (r)) +#define SMC_outl(v, a, r) writel(v, (a) + (r)) +#define SMC_insl(a, r, p, l) readsl((a) + (r), p, l) +#define SMC_outsl(a, r, p, l) writesl((a) + (r), p, l) + +#define SMC_IRQ_FLAGS (0) + #else #define SMC_CAN_USE_8BIT 1 -- cgit v0.10.2 From 6f0f6d87a2a5fc96fc54e90961d5244d668e5fbb Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Fri, 20 Oct 2006 14:43:15 -0700 Subject: [PATCH] WAN/pc300: handle, propagate minor errors - move definition of 'tmc' and 'br' locals closer to usage - handle clock_rate_calc() error - propagate errors back to upper level open routine Signed-off-by: Jeff Garzik Cc: Krzysztof Halasa Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/wan/pc300_drv.c b/drivers/net/wan/pc300_drv.c index 5823e3b..36d1c3f 100644 --- a/drivers/net/wan/pc300_drv.c +++ b/drivers/net/wan/pc300_drv.c @@ -2867,7 +2867,6 @@ static int ch_config(pc300dev_t * d) uclong clktype = chan->conf.phys_settings.clock_type; ucshort encoding = chan->conf.proto_settings.encoding; ucshort parity = chan->conf.proto_settings.parity; - int tmc, br; ucchar md0, md2; /* Reset the channel */ @@ -2940,8 +2939,12 @@ static int ch_config(pc300dev_t * d) case PC300_RSV: case PC300_X21: if (clktype == CLOCK_INT || clktype == CLOCK_TXINT) { + int tmc, br; + /* Calculate the clkrate parameters */ tmc = clock_rate_calc(clkrate, card->hw.clock, &br); + if (tmc < 0) + return -EIO; cpc_writeb(scabase + M_REG(TMCT, ch), tmc); cpc_writeb(scabase + M_REG(TXS, ch), (TXS_DTRXC | TXS_IBRG | br)); @@ -3097,14 +3100,16 @@ static int cpc_attach(struct net_device *dev, unsigned short encoding, return 0; } -static void cpc_opench(pc300dev_t * d) +static int cpc_opench(pc300dev_t * d) { pc300ch_t *chan = (pc300ch_t *) d->chan; pc300_t *card = (pc300_t *) chan->card; - int ch = chan->channel; + int ch = chan->channel, rc; void __iomem *scabase = card->hw.scabase; - ch_config(d); + rc = ch_config(d); + if (rc) + return rc; rx_config(d); @@ -3113,6 +3118,8 @@ static void cpc_opench(pc300dev_t * d) /* Assert RTS and DTR */ cpc_writeb(scabase + M_REG(CTL, ch), cpc_readb(scabase + M_REG(CTL, ch)) & ~(CTL_RTS | CTL_DTR)); + + return 0; } static void cpc_closech(pc300dev_t * d) @@ -3168,9 +3175,16 @@ int cpc_open(struct net_device *dev) } sprintf(ifr.ifr_name, "%s", dev->name); - cpc_opench(d); + result = cpc_opench(d); + if (result) + goto err_out; + netif_start_queue(dev); return 0; + +err_out: + hdlc_close(dev); + return result; } static int cpc_close(struct net_device *dev) -- cgit v0.10.2 From 7347b03d25ad7d7f001373cf64f709457c6af618 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Fri, 20 Oct 2006 14:42:14 -0700 Subject: [PATCH] e1000: Reset all functions after a PCI error During the handling of the PCI error recovery sequence, the current e1000 driver erroneously blocks a device reset for any but the first PCI function. It shouldn't -- this is a cut-n-paste error from a different driver (which tolerated only one hardware reset per hardware card). Signed-off-by: Linas Vepstas Cc: Jesse Brandeburg Acked-by: Auke Kok Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index ce0d35f..fa84983 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -4914,10 +4914,6 @@ static pci_ers_result_t e1000_io_slot_reset(struct pci_dev *pdev) pci_enable_wake(pdev, PCI_D3hot, 0); pci_enable_wake(pdev, PCI_D3cold, 0); - /* Perform card reset only on one instance of the card */ - if (PCI_FUNC (pdev->devfn) != 0) - return PCI_ERS_RESULT_RECOVERED; - e1000_reset(adapter); E1000_WRITE_REG(&adapter->hw, WUS, ~0); -- cgit v0.10.2 From 470ea7eba4aaa517533f9b02ac9a104e77264548 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 20 Oct 2006 17:06:11 -0700 Subject: [PATCH] sky2: 88E803X transmit lockup The reason sky2 driver was locking up on transmit on the Yukon-FE chipset is that it was misconfiguring the internal RAM buffer so the transmitter and receiver were sharing the same space. The code assumed there was 16K of RAM on Yukon-FE (taken from vendor driver sk98lin which is even more f*cked up on this). Then it assigned based on that. The giveaway was that the registers would only hold 9bits so both RX/TX had 0..1ff for space. It is a wonder it worked at all! This patch addresses this, and fixes an easily reproducible hang on Transmit. Only the Yukon-FE chip is Marvell 88E803X (10/100 only) are affected. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 67ecd66..95efdb5 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -699,16 +699,10 @@ static void sky2_mac_init(struct sky2_hw *hw, unsigned port) } -/* Assign Ram Buffer allocation. - * start and end are in units of 4k bytes - * ram registers are in units of 64bit words - */ -static void sky2_ramset(struct sky2_hw *hw, u16 q, u8 startk, u8 endk) +/* Assign Ram Buffer allocation in units of 64bit (8 bytes) */ +static void sky2_ramset(struct sky2_hw *hw, u16 q, u32 start, u32 end) { - u32 start, end; - - start = startk * 4096/8; - end = (endk * 4096/8) - 1; + pr_debug(PFX "q %d %#x %#x\n", q, start, end); sky2_write8(hw, RB_ADDR(q, RB_CTRL), RB_RST_CLR); sky2_write32(hw, RB_ADDR(q, RB_START), start); @@ -717,7 +711,7 @@ static void sky2_ramset(struct sky2_hw *hw, u16 q, u8 startk, u8 endk) sky2_write32(hw, RB_ADDR(q, RB_RP), start); if (q == Q_R1 || q == Q_R2) { - u32 space = (endk - startk) * 4096/8; + u32 space = end - start + 1; u32 tp = space - space/4; /* On receive queue's set the thresholds @@ -1199,19 +1193,16 @@ static int sky2_up(struct net_device *dev) sky2_mac_init(hw, port); - /* Determine available ram buffer space (in 4K blocks). - * Note: not sure about the FE setting below yet - */ - if (hw->chip_id == CHIP_ID_YUKON_FE) - ramsize = 4; - else - ramsize = sky2_read8(hw, B2_E_0); + /* Determine available ram buffer space in qwords. */ + ramsize = sky2_read8(hw, B2_E_0) * 4096/8; - /* Give transmitter one third (rounded up) */ - rxspace = ramsize - (ramsize + 2) / 3; + if (ramsize > 6*1024/8) + rxspace = ramsize - (ramsize + 2) / 3; + else + rxspace = ramsize / 2; - sky2_ramset(hw, rxqaddr[port], 0, rxspace); - sky2_ramset(hw, txqaddr[port], rxspace, ramsize); + sky2_ramset(hw, rxqaddr[port], 0, rxspace-1); + sky2_ramset(hw, txqaddr[port], rxspace, ramsize-1); /* Make sure SyncQ is disabled */ sky2_write8(hw, RB_ADDR(port == 0 ? Q_XS1 : Q_XS2, RB_CTRL), -- cgit v0.10.2 From 18a8e8649d2687283da51fbcf8218372dc5a8f6f Mon Sep 17 00:00:00 2001 From: Li Yang Date: Thu, 19 Oct 2006 21:07:34 -0500 Subject: [PATCH] ucc_geth: changes to ucc_geth driver as a result of qe_lib changes and bugfixes changes due to qe_lib changes include: o removed inclusion of platform header file o removed platform_device code, replaced with of_device o removed typedefs o uint -> u32 conversions o removed following defines: QE_SIZEOF_BD, BD_BUFFER_ARG, BD_BUFFER_CLEAR, BD_BUFFER, BD_STATUS_AND_LENGTH_SET, BD_STATUS_AND_LENGTH, and BD_BUFFER_SET because they hid sizeof/in_be32/out_be32 operations from the reader. o removed irrelevant comments, added others to resemble removed BD_ defines o const'd and uncasted all get_property() assignments bugfixes, courtesy of Scott Wood, include: - Read phy_address as a u32, not u8. - Match on type == "network" as well as compatible == "ucc_geth", as device_is_compatible() will only compare up to the length of the test string, allowing "ucc_geth_phy" to match as well. - fixes the MAC setting code in ucc_geth.c. The old code was overwriting and dereferencing random stack contents. Signed-off-by: Li Yang Signed-off-by: Kim Phillips Signed-off-by: Scott Wood Signed-off-by: Jeff Garzik diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index ab92cc7..e2ed249 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2288,7 +2288,7 @@ config UGETH_TX_ON_DEMOND config UGETH_HAS_GIGA bool - depends on UCC_GETH && MPC836x + depends on UCC_GETH && PPC_MPC836x config MV643XX_ETH tristate "MV-643XX Ethernet support" diff --git a/drivers/net/ucc_geth.c b/drivers/net/ucc_geth.c index 12cd7b5..b378880 100644 --- a/drivers/net/ucc_geth.c +++ b/drivers/net/ucc_geth.c @@ -2,14 +2,11 @@ * Copyright (C) Freescale Semicondutor, Inc. 2006. All rights reserved. * * Author: Shlomi Gridish + * Li Yang * * Description: * QE UCC Gigabit Ethernet Driver * - * Changelog: - * Jul 6, 2006 Li Yang - * - Rearrange code and style fixes - * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the * Free Software Foundation; either version 2 of the License, or (at your @@ -31,9 +28,9 @@ #include #include #include -#include #include +#include #include #include #include @@ -70,7 +67,7 @@ static DEFINE_SPINLOCK(ugeth_lock); -static ucc_geth_info_t ugeth_primary_info = { +static struct ucc_geth_info ugeth_primary_info = { .uf_info = { .bd_mem_part = MEM_PART_SYSTEM, .rtsm = UCC_FAST_SEND_IDLES_BETWEEN_FRAMES, @@ -163,7 +160,7 @@ static ucc_geth_info_t ugeth_primary_info = { .riscRx = QE_RISC_ALLOCATION_RISC1_AND_RISC2, }; -static ucc_geth_info_t ugeth_info[8]; +static struct ucc_geth_info ugeth_info[8]; #ifdef DEBUG static void mem_disp(u8 *addr, int size) @@ -219,8 +216,8 @@ static struct list_head *dequeue(struct list_head *lh) } } -static int get_interface_details(enet_interface_e enet_interface, - enet_speed_e *speed, +static int get_interface_details(enum enet_interface enet_interface, + enum enet_speed *speed, int *r10m, int *rmm, int *rpm, @@ -283,7 +280,7 @@ static int get_interface_details(enet_interface_e enet_interface, return 0; } -static struct sk_buff *get_new_skb(ucc_geth_private_t *ugeth, u8 *bd) +static struct sk_buff *get_new_skb(struct ucc_geth_private *ugeth, u8 *bd) { struct sk_buff *skb = NULL; @@ -303,21 +300,19 @@ static struct sk_buff *get_new_skb(ucc_geth_private_t *ugeth, u8 *bd) skb->dev = ugeth->dev; - BD_BUFFER_SET(bd, + out_be32(&((struct qe_bd *)bd)->buf, dma_map_single(NULL, skb->data, ugeth->ug_info->uf_info.max_rx_buf_length + UCC_GETH_RX_DATA_BUF_ALIGNMENT, DMA_FROM_DEVICE)); - BD_STATUS_AND_LENGTH_SET(bd, - (R_E | R_I | - (BD_STATUS_AND_LENGTH(bd) & R_W))); + out_be32((u32 *)bd, (R_E | R_I | (in_be32((u32 *)bd) & R_W))); return skb; } -static int rx_bd_buffer_set(ucc_geth_private_t *ugeth, u8 rxQ) +static int rx_bd_buffer_set(struct ucc_geth_private *ugeth, u8 rxQ) { u8 *bd; u32 bd_status; @@ -328,7 +323,7 @@ static int rx_bd_buffer_set(ucc_geth_private_t *ugeth, u8 rxQ) i = 0; do { - bd_status = BD_STATUS_AND_LENGTH(bd); + bd_status = in_be32((u32*)bd); skb = get_new_skb(ugeth, bd); if (!skb) /* If can not allocate data buffer, @@ -338,19 +333,19 @@ static int rx_bd_buffer_set(ucc_geth_private_t *ugeth, u8 rxQ) ugeth->rx_skbuff[rxQ][i] = skb; /* advance the BD pointer */ - bd += UCC_GETH_SIZE_OF_BD; + bd += sizeof(struct qe_bd); i++; } while (!(bd_status & R_W)); return 0; } -static int fill_init_enet_entries(ucc_geth_private_t *ugeth, +static int fill_init_enet_entries(struct ucc_geth_private *ugeth, volatile u32 *p_start, u8 num_entries, u32 thread_size, u32 thread_alignment, - qe_risc_allocation_e risc, + enum qe_risc_allocation risc, int skip_page_for_first_entry) { u32 init_enet_offset; @@ -383,10 +378,10 @@ static int fill_init_enet_entries(ucc_geth_private_t *ugeth, return 0; } -static int return_init_enet_entries(ucc_geth_private_t *ugeth, +static int return_init_enet_entries(struct ucc_geth_private *ugeth, volatile u32 *p_start, u8 num_entries, - qe_risc_allocation_e risc, + enum qe_risc_allocation risc, int skip_page_for_first_entry) { u32 init_enet_offset; @@ -416,11 +411,11 @@ static int return_init_enet_entries(ucc_geth_private_t *ugeth, } #ifdef DEBUG -static int dump_init_enet_entries(ucc_geth_private_t *ugeth, +static int dump_init_enet_entries(struct ucc_geth_private *ugeth, volatile u32 *p_start, u8 num_entries, u32 thread_size, - qe_risc_allocation_e risc, + enum qe_risc_allocation risc, int skip_page_for_first_entry) { u32 init_enet_offset; @@ -456,14 +451,14 @@ static int dump_init_enet_entries(ucc_geth_private_t *ugeth, #endif #ifdef CONFIG_UGETH_FILTERING -static enet_addr_container_t *get_enet_addr_container(void) +static struct enet_addr_container *get_enet_addr_container(void) { - enet_addr_container_t *enet_addr_cont; + struct enet_addr_container *enet_addr_cont; /* allocate memory */ - enet_addr_cont = kmalloc(sizeof(enet_addr_container_t), GFP_KERNEL); + enet_addr_cont = kmalloc(sizeof(struct enet_addr_container), GFP_KERNEL); if (!enet_addr_cont) { - ugeth_err("%s: No memory for enet_addr_container_t object.", + ugeth_err("%s: No memory for enet_addr_container object.", __FUNCTION__); return NULL; } @@ -472,45 +467,43 @@ static enet_addr_container_t *get_enet_addr_container(void) } #endif /* CONFIG_UGETH_FILTERING */ -static void put_enet_addr_container(enet_addr_container_t *enet_addr_cont) +static void put_enet_addr_container(struct enet_addr_container *enet_addr_cont) { kfree(enet_addr_cont); } +static int set_mac_addr(__be16 __iomem *reg, u8 *mac) +{ + out_be16(®[0], ((u16)mac[5] << 8) | mac[4]); + out_be16(®[1], ((u16)mac[3] << 8) | mac[2]); + out_be16(®[2], ((u16)mac[1] << 8) | mac[0]); +} + #ifdef CONFIG_UGETH_FILTERING -static int hw_add_addr_in_paddr(ucc_geth_private_t *ugeth, - enet_addr_t *p_enet_addr, u8 paddr_num) +static int hw_add_addr_in_paddr(struct ucc_geth_private *ugeth, + u8 *p_enet_addr, u8 paddr_num) { - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; if (!(paddr_num < NUM_OF_PADDRS)) { - ugeth_warn("%s: Illagel paddr_num.", __FUNCTION__); + ugeth_warn("%s: Illegal paddr_num.", __FUNCTION__); return -EINVAL; } p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> addressfiltering; /* Ethernet frames are defined in Little Endian mode, */ /* therefore to insert the address we reverse the bytes. */ - out_be16(&p_82xx_addr_filt->paddr[paddr_num].h, - (u16) (((u16) (((u16) ((*p_enet_addr)[5])) << 8)) | - (u16) (*p_enet_addr)[4])); - out_be16(&p_82xx_addr_filt->paddr[paddr_num].m, - (u16) (((u16) (((u16) ((*p_enet_addr)[3])) << 8)) | - (u16) (*p_enet_addr)[2])); - out_be16(&p_82xx_addr_filt->paddr[paddr_num].l, - (u16) (((u16) (((u16) ((*p_enet_addr)[1])) << 8)) | - (u16) (*p_enet_addr)[0])); - + set_mac_addr(&p_82xx_addr_filt->paddr[paddr_num].h, p_enet_addr); return 0; } #endif /* CONFIG_UGETH_FILTERING */ -static int hw_clear_addr_in_paddr(ucc_geth_private_t *ugeth, u8 paddr_num) +static int hw_clear_addr_in_paddr(struct ucc_geth_private *ugeth, u8 paddr_num) { - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; if (!(paddr_num < NUM_OF_PADDRS)) { ugeth_warn("%s: Illagel paddr_num.", __FUNCTION__); @@ -518,7 +511,7 @@ static int hw_clear_addr_in_paddr(ucc_geth_private_t *ugeth, u8 paddr_num) } p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> addressfiltering; /* Writing address ff.ff.ff.ff.ff.ff disables address @@ -530,14 +523,14 @@ static int hw_clear_addr_in_paddr(ucc_geth_private_t *ugeth, u8 paddr_num) return 0; } -static void hw_add_addr_in_hash(ucc_geth_private_t *ugeth, - enet_addr_t *p_enet_addr) +static void hw_add_addr_in_hash(struct ucc_geth_private *ugeth, + u8 *p_enet_addr) { - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; u32 cecr_subblock; p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> addressfiltering; cecr_subblock = @@ -546,25 +539,18 @@ static void hw_add_addr_in_hash(ucc_geth_private_t *ugeth, /* Ethernet frames are defined in Little Endian mode, therefor to insert */ /* the address to the hash (Big Endian mode), we reverse the bytes.*/ - out_be16(&p_82xx_addr_filt->taddr.h, - (u16) (((u16) (((u16) ((*p_enet_addr)[5])) << 8)) | - (u16) (*p_enet_addr)[4])); - out_be16(&p_82xx_addr_filt->taddr.m, - (u16) (((u16) (((u16) ((*p_enet_addr)[3])) << 8)) | - (u16) (*p_enet_addr)[2])); - out_be16(&p_82xx_addr_filt->taddr.l, - (u16) (((u16) (((u16) ((*p_enet_addr)[1])) << 8)) | - (u16) (*p_enet_addr)[0])); + + set_mac_addr(&p_82xx_addr_filt->taddr.h, p_enet_addr); qe_issue_cmd(QE_SET_GROUP_ADDRESS, cecr_subblock, - (u8) QE_CR_PROTOCOL_ETHERNET, 0); + QE_CR_PROTOCOL_ETHERNET, 0); } #ifdef CONFIG_UGETH_MAGIC_PACKET -static void magic_packet_detection_enable(ucc_geth_private_t *ugeth) +static void magic_packet_detection_enable(struct ucc_geth_private *ugeth) { - ucc_fast_private_t *uccf; - ucc_geth_t *ug_regs; + struct ucc_fast_private *uccf; + struct ucc_geth *ug_regs; u32 maccfg2, uccm; uccf = ugeth->uccf; @@ -581,10 +567,10 @@ static void magic_packet_detection_enable(ucc_geth_private_t *ugeth) out_be32(&ug_regs->maccfg2, maccfg2); } -static void magic_packet_detection_disable(ucc_geth_private_t *ugeth) +static void magic_packet_detection_disable(struct ucc_geth_private *ugeth) { - ucc_fast_private_t *uccf; - ucc_geth_t *ug_regs; + struct ucc_fast_private *uccf; + struct ucc_geth *ug_regs; u32 maccfg2, uccm; uccf = ugeth->uccf; @@ -602,26 +588,26 @@ static void magic_packet_detection_disable(ucc_geth_private_t *ugeth) } #endif /* MAGIC_PACKET */ -static inline int compare_addr(enet_addr_t *addr1, enet_addr_t *addr2) +static inline int compare_addr(u8 **addr1, u8 **addr2) { return memcmp(addr1, addr2, ENET_NUM_OCTETS_PER_ADDRESS); } #ifdef DEBUG -static void get_statistics(ucc_geth_private_t *ugeth, - ucc_geth_tx_firmware_statistics_t * +static void get_statistics(struct ucc_geth_private *ugeth, + struct ucc_geth_tx_firmware_statistics * tx_firmware_statistics, - ucc_geth_rx_firmware_statistics_t * + struct ucc_geth_rx_firmware_statistics * rx_firmware_statistics, - ucc_geth_hardware_statistics_t *hardware_statistics) + struct ucc_geth_hardware_statistics *hardware_statistics) { - ucc_fast_t *uf_regs; - ucc_geth_t *ug_regs; - ucc_geth_tx_firmware_statistics_pram_t *p_tx_fw_statistics_pram; - ucc_geth_rx_firmware_statistics_pram_t *p_rx_fw_statistics_pram; + struct ucc_fast *uf_regs; + struct ucc_geth *ug_regs; + struct ucc_geth_tx_firmware_statistics_pram *p_tx_fw_statistics_pram; + struct ucc_geth_rx_firmware_statistics_pram *p_rx_fw_statistics_pram; ug_regs = ugeth->ug_regs; - uf_regs = (ucc_fast_t *) ug_regs; + uf_regs = (struct ucc_fast *) ug_regs; p_tx_fw_statistics_pram = ugeth->p_tx_fw_statistics_pram; p_rx_fw_statistics_pram = ugeth->p_rx_fw_statistics_pram; @@ -727,7 +713,7 @@ static void get_statistics(ucc_geth_private_t *ugeth, } } -static void dump_bds(ucc_geth_private_t *ugeth) +static void dump_bds(struct ucc_geth_private *ugeth) { int i; int length; @@ -736,7 +722,7 @@ static void dump_bds(ucc_geth_private_t *ugeth) if (ugeth->p_tx_bd_ring[i]) { length = (ugeth->ug_info->bdRingLenTx[i] * - UCC_GETH_SIZE_OF_BD); + sizeof(struct qe_bd)); ugeth_info("TX BDs[%d]", i); mem_disp(ugeth->p_tx_bd_ring[i], length); } @@ -745,14 +731,14 @@ static void dump_bds(ucc_geth_private_t *ugeth) if (ugeth->p_rx_bd_ring[i]) { length = (ugeth->ug_info->bdRingLenRx[i] * - UCC_GETH_SIZE_OF_BD); + sizeof(struct qe_bd)); ugeth_info("RX BDs[%d]", i); mem_disp(ugeth->p_rx_bd_ring[i], length); } } } -static void dump_regs(ucc_geth_private_t *ugeth) +static void dump_regs(struct ucc_geth_private *ugeth) { int i; @@ -893,7 +879,7 @@ static void dump_regs(ucc_geth_private_t *ugeth) ugeth_info("Base address: 0x%08x", (u32) & ugeth->p_thread_data_tx[i]); mem_disp((u8 *) & ugeth->p_thread_data_tx[i], - sizeof(ucc_geth_thread_data_tx_t)); + sizeof(struct ucc_geth_thread_data_tx)); } } if (ugeth->p_thread_data_rx) { @@ -927,7 +913,7 @@ static void dump_regs(ucc_geth_private_t *ugeth) ugeth_info("Base address: 0x%08x", (u32) & ugeth->p_thread_data_rx[i]); mem_disp((u8 *) & ugeth->p_thread_data_rx[i], - sizeof(ucc_geth_thread_data_rx_t)); + sizeof(struct ucc_geth_thread_data_rx)); } } if (ugeth->p_exf_glbl_param) { @@ -1105,7 +1091,7 @@ static void dump_regs(ucc_geth_private_t *ugeth) ugeth_info("Base address: 0x%08x", (u32) & ugeth->p_send_q_mem_reg->sqqd[i]); mem_disp((u8 *) & ugeth->p_send_q_mem_reg->sqqd[i], - sizeof(ucc_geth_send_queue_qd_t)); + sizeof(struct ucc_geth_send_queue_qd)); } } if (ugeth->p_scheduler) { @@ -1187,7 +1173,7 @@ static void dump_regs(ucc_geth_private_t *ugeth) qe_muram_addr(in_be32 (&ugeth->p_rx_bd_qs_tbl[i]. bdbaseptr)), - sizeof(ucc_geth_rx_prefetched_bds_t)); + sizeof(struct ucc_geth_rx_prefetched_bds)); } } if (ugeth->p_init_enet_param_shadow) { @@ -1198,7 +1184,7 @@ static void dump_regs(ucc_geth_private_t *ugeth) mem_disp((u8 *) ugeth->p_init_enet_param_shadow, sizeof(*ugeth->p_init_enet_param_shadow)); - size = sizeof(ucc_geth_thread_rx_pram_t); + size = sizeof(struct ucc_geth_thread_rx_pram); if (ugeth->ug_info->rxExtendedFiltering) { size += THREAD_RX_PRAM_ADDITIONAL_FOR_EXTENDED_FILTERING; @@ -1216,7 +1202,7 @@ static void dump_regs(ucc_geth_private_t *ugeth) &(ugeth->p_init_enet_param_shadow-> txthread[0]), ENET_INIT_PARAM_MAX_ENTRIES_TX, - sizeof(ucc_geth_thread_tx_pram_t), + sizeof(struct ucc_geth_thread_tx_pram), ugeth->ug_info->riscTx, 0); dump_init_enet_entries(ugeth, &(ugeth->p_init_enet_param_shadow-> @@ -1578,12 +1564,12 @@ static int init_min_frame_len(u16 min_frame_length, return 0; } -static int adjust_enet_interface(ucc_geth_private_t *ugeth) +static int adjust_enet_interface(struct ucc_geth_private *ugeth) { - ucc_geth_info_t *ug_info; - ucc_geth_t *ug_regs; - ucc_fast_t *uf_regs; - enet_speed_e speed; + struct ucc_geth_info *ug_info; + struct ucc_geth *ug_regs; + struct ucc_fast *uf_regs; + enum enet_speed speed; int ret_val, rpm = 0, tbi = 0, r10m = 0, rmm = 0, limited_to_full_duplex = 0; u32 upsmr, maccfg2, utbipar, tbiBaseAddress; @@ -1691,8 +1677,8 @@ static int adjust_enet_interface(ucc_geth_private_t *ugeth) */ static void adjust_link(struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); - ucc_geth_t *ug_regs; + struct ucc_geth_private *ugeth = netdev_priv(dev); + struct ucc_geth *ug_regs; u32 tempval; struct ugeth_mii_info *mii_info = ugeth->mii_info; @@ -1722,7 +1708,7 @@ static void adjust_link(struct net_device *dev) if (mii_info->speed != ugeth->oldspeed) { switch (mii_info->speed) { case 1000: -#ifdef CONFIG_MPC836x +#ifdef CONFIG_PPC_MPC836x /* FIXME: This code is for 100Mbs BUG fixing, remove this when it is fixed!!! */ if (ugeth->ug_info->enet_interface == @@ -1768,7 +1754,7 @@ remove this when it is fixed!!! */ break; case 100: case 10: -#ifdef CONFIG_MPC836x +#ifdef CONFIG_PPC_MPC836x /* FIXME: This code is for 100Mbs BUG fixing, remove this lines when it will be fixed!!! */ ugeth->ug_info->enet_interface = ENET_100_RGMII; @@ -1827,9 +1813,9 @@ remove this lines when it will be fixed!!! */ */ static int init_phy(struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); struct phy_info *curphy; - ucc_mii_mng_t *mii_regs; + struct ucc_mii_mng *mii_regs; struct ugeth_mii_info *mii_info; int err; @@ -1914,17 +1900,17 @@ static int init_phy(struct net_device *dev) } #ifdef CONFIG_UGETH_TX_ON_DEMOND -static int ugeth_transmit_on_demand(ucc_geth_private_t *ugeth) +static int ugeth_transmit_on_demand(struct ucc_geth_private *ugeth) { - ucc_fast_transmit_on_demand(ugeth->uccf); + struct ucc_fastransmit_on_demand(ugeth->uccf); return 0; } #endif -static int ugeth_graceful_stop_tx(ucc_geth_private_t *ugeth) +static int ugeth_graceful_stop_tx(struct ucc_geth_private *ugeth) { - ucc_fast_private_t *uccf; + struct ucc_fast_private *uccf; u32 cecr_subblock; u32 temp; @@ -1940,7 +1926,7 @@ static int ugeth_graceful_stop_tx(ucc_geth_private_t *ugeth) cecr_subblock = ucc_fast_get_qe_cr_subblock(ugeth->ug_info->uf_info.ucc_num); qe_issue_cmd(QE_GRACEFUL_STOP_TX, cecr_subblock, - (u8) QE_CR_PROTOCOL_ETHERNET, 0); + QE_CR_PROTOCOL_ETHERNET, 0); /* Wait for command to complete */ do { @@ -1952,9 +1938,9 @@ static int ugeth_graceful_stop_tx(ucc_geth_private_t *ugeth) return 0; } -static int ugeth_graceful_stop_rx(ucc_geth_private_t * ugeth) +static int ugeth_graceful_stop_rx(struct ucc_geth_private * ugeth) { - ucc_fast_private_t *uccf; + struct ucc_fast_private *uccf; u32 cecr_subblock; u8 temp; @@ -1973,7 +1959,7 @@ static int ugeth_graceful_stop_rx(ucc_geth_private_t * ugeth) ucc_fast_get_qe_cr_subblock(ugeth->ug_info->uf_info. ucc_num); qe_issue_cmd(QE_GRACEFUL_STOP_RX, cecr_subblock, - (u8) QE_CR_PROTOCOL_ETHERNET, 0); + QE_CR_PROTOCOL_ETHERNET, 0); temp = ugeth->p_rx_glbl_pram->rxgstpack; } while (!(temp & GRACEFUL_STOP_ACKNOWLEDGE_RX)); @@ -1983,41 +1969,40 @@ static int ugeth_graceful_stop_rx(ucc_geth_private_t * ugeth) return 0; } -static int ugeth_restart_tx(ucc_geth_private_t *ugeth) +static int ugeth_restart_tx(struct ucc_geth_private *ugeth) { - ucc_fast_private_t *uccf; + struct ucc_fast_private *uccf; u32 cecr_subblock; uccf = ugeth->uccf; cecr_subblock = ucc_fast_get_qe_cr_subblock(ugeth->ug_info->uf_info.ucc_num); - qe_issue_cmd(QE_RESTART_TX, cecr_subblock, (u8) QE_CR_PROTOCOL_ETHERNET, - 0); + qe_issue_cmd(QE_RESTART_TX, cecr_subblock, QE_CR_PROTOCOL_ETHERNET, 0); uccf->stopped_tx = 0; return 0; } -static int ugeth_restart_rx(ucc_geth_private_t *ugeth) +static int ugeth_restart_rx(struct ucc_geth_private *ugeth) { - ucc_fast_private_t *uccf; + struct ucc_fast_private *uccf; u32 cecr_subblock; uccf = ugeth->uccf; cecr_subblock = ucc_fast_get_qe_cr_subblock(ugeth->ug_info->uf_info.ucc_num); - qe_issue_cmd(QE_RESTART_RX, cecr_subblock, (u8) QE_CR_PROTOCOL_ETHERNET, + qe_issue_cmd(QE_RESTART_RX, cecr_subblock, QE_CR_PROTOCOL_ETHERNET, 0); uccf->stopped_rx = 0; return 0; } -static int ugeth_enable(ucc_geth_private_t *ugeth, comm_dir_e mode) +static int ugeth_enable(struct ucc_geth_private *ugeth, enum comm_dir mode) { - ucc_fast_private_t *uccf; + struct ucc_fast_private *uccf; int enabled_tx, enabled_rx; uccf = ugeth->uccf; @@ -2044,9 +2029,9 @@ static int ugeth_enable(ucc_geth_private_t *ugeth, comm_dir_e mode) } -static int ugeth_disable(ucc_geth_private_t * ugeth, comm_dir_e mode) +static int ugeth_disable(struct ucc_geth_private * ugeth, enum comm_dir mode) { - ucc_fast_private_t *uccf; + struct ucc_fast_private *uccf; uccf = ugeth->uccf; @@ -2069,7 +2054,7 @@ static int ugeth_disable(ucc_geth_private_t * ugeth, comm_dir_e mode) return 0; } -static void ugeth_dump_regs(ucc_geth_private_t *ugeth) +static void ugeth_dump_regs(struct ucc_geth_private *ugeth) { #ifdef DEBUG ucc_fast_dump_regs(ugeth->uccf); @@ -2079,9 +2064,9 @@ static void ugeth_dump_regs(ucc_geth_private_t *ugeth) } #ifdef CONFIG_UGETH_FILTERING -static int ugeth_ext_filtering_serialize_tad(ucc_geth_tad_params_t * +static int ugeth_ext_filtering_serialize_tad(struct ucc_geth_tad_params * p_UccGethTadParams, - qe_fltr_tad_t *qe_fltr_tad) + struct qe_fltr_tad *qe_fltr_tad) { u16 temp; @@ -2119,11 +2104,11 @@ static int ugeth_ext_filtering_serialize_tad(ucc_geth_tad_params_t * return 0; } -static enet_addr_container_t - *ugeth_82xx_filtering_get_match_addr_in_hash(ucc_geth_private_t *ugeth, - enet_addr_t *p_enet_addr) +static struct enet_addr_container_t + *ugeth_82xx_filtering_get_match_addr_in_hash(struct ucc_geth_private *ugeth, + struct enet_addr *p_enet_addr) { - enet_addr_container_t *enet_addr_cont; + struct enet_addr_container *enet_addr_cont; struct list_head *p_lh; u16 i, num; int32_t j; @@ -2144,7 +2129,7 @@ static enet_addr_container_t for (i = 0; i < num; i++) { enet_addr_cont = - (enet_addr_container_t *) + (struct enet_addr_container *) ENET_ADDR_CONT_ENTRY(dequeue(p_lh)); for (j = ENET_NUM_OCTETS_PER_ADDRESS - 1; j >= 0; j--) { if ((*p_enet_addr)[j] != (enet_addr_cont->address)[j]) @@ -2157,11 +2142,11 @@ static enet_addr_container_t return NULL; } -static int ugeth_82xx_filtering_add_addr_in_hash(ucc_geth_private_t *ugeth, - enet_addr_t *p_enet_addr) +static int ugeth_82xx_filtering_add_addr_in_hash(struct ucc_geth_private *ugeth, + struct enet_addr *p_enet_addr) { - ucc_geth_enet_address_recognition_location_e location; - enet_addr_container_t *enet_addr_cont; + enum ucc_geth_enet_address_recognition_location location; + struct enet_addr_container *enet_addr_cont; struct list_head *p_lh; u8 i; u32 limit; @@ -2196,18 +2181,17 @@ static int ugeth_82xx_filtering_add_addr_in_hash(ucc_geth_private_t *ugeth, enqueue(p_lh, &enet_addr_cont->node); /* Put it back */ ++(*p_counter); - hw_add_addr_in_hash(ugeth, &(enet_addr_cont->address)); - + hw_add_addr_in_hash(ugeth, enet_addr_cont->address); return 0; } -static int ugeth_82xx_filtering_clear_addr_in_hash(ucc_geth_private_t *ugeth, - enet_addr_t *p_enet_addr) +static int ugeth_82xx_filtering_clear_addr_in_hash(struct ucc_geth_private *ugeth, + struct enet_addr *p_enet_addr) { - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; - enet_addr_container_t *enet_addr_cont; - ucc_fast_private_t *uccf; - comm_dir_e comm_dir; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct enet_addr_container *enet_addr_cont; + struct ucc_fast_private *uccf; + enum comm_dir comm_dir; u16 i, num; struct list_head *p_lh; u32 *addr_h, *addr_l; @@ -2216,7 +2200,7 @@ static int ugeth_82xx_filtering_clear_addr_in_hash(ucc_geth_private_t *ugeth, uccf = ugeth->uccf; p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> addressfiltering; if (! @@ -2256,9 +2240,9 @@ static int ugeth_82xx_filtering_clear_addr_in_hash(ucc_geth_private_t *ugeth, num = --(*p_counter); for (i = 0; i < num; i++) { enet_addr_cont = - (enet_addr_container_t *) + (struct enet_addr_container *) ENET_ADDR_CONT_ENTRY(dequeue(p_lh)); - hw_add_addr_in_hash(ugeth, &(enet_addr_cont->address)); + hw_add_addr_in_hash(ugeth, enet_addr_cont->address); enqueue(p_lh, &enet_addr_cont->node); /* Put it back */ } @@ -2269,14 +2253,14 @@ static int ugeth_82xx_filtering_clear_addr_in_hash(ucc_geth_private_t *ugeth, } #endif /* CONFIG_UGETH_FILTERING */ -static int ugeth_82xx_filtering_clear_all_addr_in_hash(ucc_geth_private_t * +static int ugeth_82xx_filtering_clear_all_addr_in_hash(struct ucc_geth_private * ugeth, - enet_addr_type_e + enum enet_addr_type enet_addr_type) { - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; - ucc_fast_private_t *uccf; - comm_dir_e comm_dir; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct ucc_fast_private *uccf; + enum comm_dir comm_dir; struct list_head *p_lh; u16 i, num; u32 *addr_h, *addr_l; @@ -2285,7 +2269,7 @@ static int ugeth_82xx_filtering_clear_all_addr_in_hash(ucc_geth_private_t * uccf = ugeth->uccf; p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> addressfiltering; if (enet_addr_type == ENET_ADDR_TYPE_GROUP) { @@ -2331,8 +2315,8 @@ static int ugeth_82xx_filtering_clear_all_addr_in_hash(ucc_geth_private_t * } #ifdef CONFIG_UGETH_FILTERING -static int ugeth_82xx_filtering_add_addr_in_paddr(ucc_geth_private_t *ugeth, - enet_addr_t *p_enet_addr, +static int ugeth_82xx_filtering_add_addr_in_paddr(struct ucc_geth_private *ugeth, + struct enet_addr *p_enet_addr, u8 paddr_num) { int i; @@ -2352,14 +2336,14 @@ static int ugeth_82xx_filtering_add_addr_in_paddr(ucc_geth_private_t *ugeth, } #endif /* CONFIG_UGETH_FILTERING */ -static int ugeth_82xx_filtering_clear_addr_in_paddr(ucc_geth_private_t *ugeth, +static int ugeth_82xx_filtering_clear_addr_in_paddr(struct ucc_geth_private *ugeth, u8 paddr_num) { ugeth->indAddrRegUsed[paddr_num] = 0; /* mark this paddr as not used */ return hw_clear_addr_in_paddr(ugeth, paddr_num);/* clear in hardware */ } -static void ucc_geth_memclean(ucc_geth_private_t *ugeth) +static void ucc_geth_memclean(struct ucc_geth_private *ugeth) { u16 i, j; u8 *bd; @@ -2433,8 +2417,8 @@ static void ucc_geth_memclean(ucc_geth_private_t *ugeth) for (j = 0; j < ugeth->ug_info->bdRingLenTx[i]; j++) { if (ugeth->tx_skbuff[i][j]) { dma_unmap_single(NULL, - BD_BUFFER_ARG(bd), - (BD_STATUS_AND_LENGTH(bd) & + ((qe_bd_t *)bd)->buf, + (in_be32((u32 *)bd) & BD_LENGTH_MASK), DMA_TO_DEVICE); dev_kfree_skb_any(ugeth->tx_skbuff[i][j]); @@ -2460,18 +2444,17 @@ static void ucc_geth_memclean(ucc_geth_private_t *ugeth) bd = ugeth->p_rx_bd_ring[i]; for (j = 0; j < ugeth->ug_info->bdRingLenRx[i]; j++) { if (ugeth->rx_skbuff[i][j]) { - dma_unmap_single(NULL, BD_BUFFER(bd), - ugeth->ug_info-> - uf_info. - max_rx_buf_length + - UCC_GETH_RX_DATA_BUF_ALIGNMENT, - DMA_FROM_DEVICE); - - dev_kfree_skb_any(ugeth-> - rx_skbuff[i][j]); + dma_unmap_single(NULL, + ((struct qe_bd *)bd)->buf, + ugeth->ug_info-> + uf_info.max_rx_buf_length + + UCC_GETH_RX_DATA_BUF_ALIGNMENT, + DMA_FROM_DEVICE); + dev_kfree_skb_any( + ugeth->rx_skbuff[i][j]); ugeth->rx_skbuff[i][j] = NULL; } - bd += UCC_GETH_SIZE_OF_BD; + bd += sizeof(struct qe_bd); } kfree(ugeth->rx_skbuff[i]); @@ -2496,11 +2479,11 @@ static void ucc_geth_memclean(ucc_geth_private_t *ugeth) static void ucc_geth_set_multi(struct net_device *dev) { - ucc_geth_private_t *ugeth; + struct ucc_geth_private *ugeth; struct dev_mc_list *dmi; - ucc_fast_t *uf_regs; - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; - enet_addr_t tempaddr; + struct ucc_fast *uf_regs; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + u8 tempaddr[6]; u8 *mcptr, *tdptr; int i, j; @@ -2517,7 +2500,7 @@ static void ucc_geth_set_multi(struct net_device *dev) uf_regs->upsmr &= ~UPSMR_PRO; p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth-> p_rx_glbl_pram->addressfiltering; if (dev->flags & IFF_ALLMULTI) { @@ -2546,23 +2529,22 @@ static void ucc_geth_set_multi(struct net_device *dev) * copy bytes MSB first from dmi_addr. */ mcptr = (u8 *) dmi->dmi_addr + 5; - tdptr = (u8 *) & tempaddr; + tdptr = (u8 *) tempaddr; for (j = 0; j < 6; j++) *tdptr++ = *mcptr--; /* Ask CPM to run CRC and set bit in * filter mask. */ - hw_add_addr_in_hash(ugeth, &tempaddr); - + hw_add_addr_in_hash(ugeth, tempaddr); } } } } -static void ucc_geth_stop(ucc_geth_private_t *ugeth) +static void ucc_geth_stop(struct ucc_geth_private *ugeth) { - ucc_geth_t *ug_regs = ugeth->ug_regs; + struct ucc_geth *ug_regs = ugeth->ug_regs; u32 tempval; ugeth_vdbg("%s: IN", __FUNCTION__); @@ -2605,15 +2587,15 @@ static void ucc_geth_stop(ucc_geth_private_t *ugeth) ucc_geth_memclean(ugeth); } -static int ucc_geth_startup(ucc_geth_private_t *ugeth) +static int ucc_geth_startup(struct ucc_geth_private *ugeth) { - ucc_geth_82xx_address_filtering_pram_t *p_82xx_addr_filt; - ucc_geth_init_pram_t *p_init_enet_pram; - ucc_fast_private_t *uccf; - ucc_geth_info_t *ug_info; - ucc_fast_info_t *uf_info; - ucc_fast_t *uf_regs; - ucc_geth_t *ug_regs; + struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct ucc_geth_init_pram *p_init_enet_pram; + struct ucc_fast_private *uccf; + struct ucc_geth_info *ug_info; + struct ucc_fast_info *uf_info; + struct ucc_fast *uf_regs; + struct ucc_geth *ug_regs; int ret_val = -EINVAL; u32 remoder = UCC_GETH_REMODER_INIT; u32 init_enet_pram_offset, cecr_subblock, command, maccfg1; @@ -2788,7 +2770,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) UCC_GETH_VLAN_OPERATION_NON_TAGGED_NOP); uf_regs = uccf->uf_regs; - ug_regs = (ucc_geth_t *) (uccf->uf_regs); + ug_regs = (struct ucc_geth *) (uccf->uf_regs); ugeth->ug_regs = ug_regs; init_default_reg_vals(&uf_regs->upsmr, @@ -2869,10 +2851,10 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Allocate in multiple of UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT, according to spec */ - length = ((ug_info->bdRingLenTx[j] * UCC_GETH_SIZE_OF_BD) + length = ((ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)) / UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT) * UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT; - if ((ug_info->bdRingLenTx[j] * UCC_GETH_SIZE_OF_BD) % + if ((ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)) % UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT) length += UCC_GETH_TX_BD_RING_SIZE_MEMORY_ALIGNMENT; if (uf_info->bd_mem_part == MEM_PART_SYSTEM) { @@ -2904,13 +2886,13 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } /* Zero unused end of bd ring, according to spec */ memset(ugeth->p_tx_bd_ring[j] + - ug_info->bdRingLenTx[j] * UCC_GETH_SIZE_OF_BD, 0, - length - ug_info->bdRingLenTx[j] * UCC_GETH_SIZE_OF_BD); + ug_info->bdRingLenTx[j] * sizeof(struct qe_bd), 0, + length - ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)); } /* Allocate Rx bds */ for (j = 0; j < ug_info->numQueuesRx; j++) { - length = ug_info->bdRingLenRx[j] * UCC_GETH_SIZE_OF_BD; + length = ug_info->bdRingLenRx[j] * sizeof(struct qe_bd); if (uf_info->bd_mem_part == MEM_PART_SYSTEM) { u32 align = 4; if (UCC_GETH_RX_BD_RING_ALIGNMENT > 4) @@ -2960,12 +2942,15 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) ugeth->skb_curtx[j] = ugeth->skb_dirtytx[j] = 0; bd = ugeth->confBd[j] = ugeth->txBd[j] = ugeth->p_tx_bd_ring[j]; for (i = 0; i < ug_info->bdRingLenTx[j]; i++) { - BD_BUFFER_CLEAR(bd); - BD_STATUS_AND_LENGTH_SET(bd, 0); - bd += UCC_GETH_SIZE_OF_BD; + /* clear bd buffer */ + out_be32(&((struct qe_bd *)bd)->buf, 0); + /* set bd status and length */ + out_be32((u32 *)bd, 0); + bd += sizeof(struct qe_bd); } - bd -= UCC_GETH_SIZE_OF_BD; - BD_STATUS_AND_LENGTH_SET(bd, T_W);/* for last BD set Wrap bit */ + bd -= sizeof(struct qe_bd); + /* set bd status and length */ + out_be32((u32 *)bd, T_W); /* for last BD set Wrap bit */ } /* Init Rx bds */ @@ -2989,12 +2974,15 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) ugeth->skb_currx[j] = 0; bd = ugeth->rxBd[j] = ugeth->p_rx_bd_ring[j]; for (i = 0; i < ug_info->bdRingLenRx[j]; i++) { - BD_STATUS_AND_LENGTH_SET(bd, R_I); - BD_BUFFER_CLEAR(bd); - bd += UCC_GETH_SIZE_OF_BD; + /* set bd status and length */ + out_be32((u32 *)bd, R_I); + /* clear bd buffer */ + out_be32(&((struct qe_bd *)bd)->buf, 0); + bd += sizeof(struct qe_bd); } - bd -= UCC_GETH_SIZE_OF_BD; - BD_STATUS_AND_LENGTH_SET(bd, R_W);/* for last BD set Wrap bit */ + bd -= sizeof(struct qe_bd); + /* set bd status and length */ + out_be32((u32 *)bd, R_W); /* for last BD set Wrap bit */ } /* @@ -3003,7 +2991,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Tx global PRAM */ /* Allocate global tx parameter RAM page */ ugeth->tx_glbl_pram_offset = - qe_muram_alloc(sizeof(ucc_geth_tx_global_pram_t), + qe_muram_alloc(sizeof(struct ucc_geth_tx_global_pram), UCC_GETH_TX_GLOBAL_PRAM_ALIGNMENT); if (IS_MURAM_ERR(ugeth->tx_glbl_pram_offset)) { ugeth_err @@ -3013,10 +3001,10 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) return -ENOMEM; } ugeth->p_tx_glbl_pram = - (ucc_geth_tx_global_pram_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_tx_global_pram *) qe_muram_addr(ugeth-> tx_glbl_pram_offset); /* Zero out p_tx_glbl_pram */ - memset(ugeth->p_tx_glbl_pram, 0, sizeof(ucc_geth_tx_global_pram_t)); + memset(ugeth->p_tx_glbl_pram, 0, sizeof(struct ucc_geth_tx_global_pram)); /* Fill global PRAM */ @@ -3024,7 +3012,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Size varies with number of Tx threads */ ugeth->thread_dat_tx_offset = qe_muram_alloc(numThreadsTxNumerical * - sizeof(ucc_geth_thread_data_tx_t) + + sizeof(struct ucc_geth_thread_data_tx) + 32 * (numThreadsTxNumerical == 1), UCC_GETH_THREAD_DATA_ALIGNMENT); if (IS_MURAM_ERR(ugeth->thread_dat_tx_offset)) { @@ -3036,7 +3024,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_thread_data_tx = - (ucc_geth_thread_data_tx_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_thread_data_tx *) qe_muram_addr(ugeth-> thread_dat_tx_offset); out_be32(&ugeth->p_tx_glbl_pram->tqptr, ugeth->thread_dat_tx_offset); @@ -3053,7 +3041,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Size varies with number of Tx queues */ ugeth->send_q_mem_reg_offset = qe_muram_alloc(ug_info->numQueuesTx * - sizeof(ucc_geth_send_queue_qd_t), + sizeof(struct ucc_geth_send_queue_qd), UCC_GETH_SEND_QUEUE_QUEUE_DESCRIPTOR_ALIGNMENT); if (IS_MURAM_ERR(ugeth->send_q_mem_reg_offset)) { ugeth_err @@ -3064,7 +3052,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_send_q_mem_reg = - (ucc_geth_send_queue_mem_region_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_send_queue_mem_region *) qe_muram_addr(ugeth-> send_q_mem_reg_offset); out_be32(&ugeth->p_tx_glbl_pram->sqptr, ugeth->send_q_mem_reg_offset); @@ -3073,7 +3061,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) for (i = 0; i < ug_info->numQueuesTx; i++) { endOfRing = ugeth->p_tx_bd_ring[i] + (ug_info->bdRingLenTx[i] - - 1) * UCC_GETH_SIZE_OF_BD; + 1) * sizeof(struct qe_bd); if (ugeth->ug_info->uf_info.bd_mem_part == MEM_PART_SYSTEM) { out_be32(&ugeth->p_send_q_mem_reg->sqqd[i].bd_ring_base, (u32) virt_to_phys(ugeth->p_tx_bd_ring[i])); @@ -3096,7 +3084,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) if (ug_info->numQueuesTx > 1) { /* scheduler exists only if more than 1 tx queue */ ugeth->scheduler_offset = - qe_muram_alloc(sizeof(ucc_geth_scheduler_t), + qe_muram_alloc(sizeof(struct ucc_geth_scheduler), UCC_GETH_SCHEDULER_ALIGNMENT); if (IS_MURAM_ERR(ugeth->scheduler_offset)) { ugeth_err @@ -3107,12 +3095,12 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_scheduler = - (ucc_geth_scheduler_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_scheduler *) qe_muram_addr(ugeth-> scheduler_offset); out_be32(&ugeth->p_tx_glbl_pram->schedulerbasepointer, ugeth->scheduler_offset); /* Zero out p_scheduler */ - memset(ugeth->p_scheduler, 0, sizeof(ucc_geth_scheduler_t)); + memset(ugeth->p_scheduler, 0, sizeof(struct ucc_geth_scheduler)); /* Set values in scheduler */ out_be32(&ugeth->p_scheduler->mblinterval, @@ -3144,7 +3132,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) statisticsMode & UCC_GETH_STATISTICS_GATHERING_MODE_FIRMWARE_TX) { ugeth->tx_fw_statistics_pram_offset = qe_muram_alloc(sizeof - (ucc_geth_tx_firmware_statistics_pram_t), + (struct ucc_geth_tx_firmware_statistics_pram), UCC_GETH_TX_STATISTICS_ALIGNMENT); if (IS_MURAM_ERR(ugeth->tx_fw_statistics_pram_offset)) { ugeth_err @@ -3154,11 +3142,11 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) return -ENOMEM; } ugeth->p_tx_fw_statistics_pram = - (ucc_geth_tx_firmware_statistics_pram_t *) + (struct ucc_geth_tx_firmware_statistics_pram *) qe_muram_addr(ugeth->tx_fw_statistics_pram_offset); /* Zero out p_tx_fw_statistics_pram */ memset(ugeth->p_tx_fw_statistics_pram, - 0, sizeof(ucc_geth_tx_firmware_statistics_pram_t)); + 0, sizeof(struct ucc_geth_tx_firmware_statistics_pram)); } /* temoder */ @@ -3183,7 +3171,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Rx global PRAM */ /* Allocate global rx parameter RAM page */ ugeth->rx_glbl_pram_offset = - qe_muram_alloc(sizeof(ucc_geth_rx_global_pram_t), + qe_muram_alloc(sizeof(struct ucc_geth_rx_global_pram), UCC_GETH_RX_GLOBAL_PRAM_ALIGNMENT); if (IS_MURAM_ERR(ugeth->rx_glbl_pram_offset)) { ugeth_err @@ -3193,10 +3181,10 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) return -ENOMEM; } ugeth->p_rx_glbl_pram = - (ucc_geth_rx_global_pram_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_rx_global_pram *) qe_muram_addr(ugeth-> rx_glbl_pram_offset); /* Zero out p_rx_glbl_pram */ - memset(ugeth->p_rx_glbl_pram, 0, sizeof(ucc_geth_rx_global_pram_t)); + memset(ugeth->p_rx_glbl_pram, 0, sizeof(struct ucc_geth_rx_global_pram)); /* Fill global PRAM */ @@ -3204,7 +3192,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Size varies with number of Rx threads */ ugeth->thread_dat_rx_offset = qe_muram_alloc(numThreadsRxNumerical * - sizeof(ucc_geth_thread_data_rx_t), + sizeof(struct ucc_geth_thread_data_rx), UCC_GETH_THREAD_DATA_ALIGNMENT); if (IS_MURAM_ERR(ugeth->thread_dat_rx_offset)) { ugeth_err @@ -3215,7 +3203,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_thread_data_rx = - (ucc_geth_thread_data_rx_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_thread_data_rx *) qe_muram_addr(ugeth-> thread_dat_rx_offset); out_be32(&ugeth->p_rx_glbl_pram->rqptr, ugeth->thread_dat_rx_offset); @@ -3227,7 +3215,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) statisticsMode & UCC_GETH_STATISTICS_GATHERING_MODE_FIRMWARE_RX) { ugeth->rx_fw_statistics_pram_offset = qe_muram_alloc(sizeof - (ucc_geth_rx_firmware_statistics_pram_t), + (struct ucc_geth_rx_firmware_statistics_pram), UCC_GETH_RX_STATISTICS_ALIGNMENT); if (IS_MURAM_ERR(ugeth->rx_fw_statistics_pram_offset)) { ugeth_err @@ -3237,11 +3225,11 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) return -ENOMEM; } ugeth->p_rx_fw_statistics_pram = - (ucc_geth_rx_firmware_statistics_pram_t *) + (struct ucc_geth_rx_firmware_statistics_pram *) qe_muram_addr(ugeth->rx_fw_statistics_pram_offset); /* Zero out p_rx_fw_statistics_pram */ memset(ugeth->p_rx_fw_statistics_pram, 0, - sizeof(ucc_geth_rx_firmware_statistics_pram_t)); + sizeof(struct ucc_geth_rx_firmware_statistics_pram)); } /* intCoalescingPtr */ @@ -3249,7 +3237,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Size varies with number of Rx queues */ ugeth->rx_irq_coalescing_tbl_offset = qe_muram_alloc(ug_info->numQueuesRx * - sizeof(ucc_geth_rx_interrupt_coalescing_entry_t), + sizeof(struct ucc_geth_rx_interrupt_coalescing_entry), UCC_GETH_RX_INTERRUPT_COALESCING_ALIGNMENT); if (IS_MURAM_ERR(ugeth->rx_irq_coalescing_tbl_offset)) { ugeth_err @@ -3260,7 +3248,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_rx_irq_coalescing_tbl = - (ucc_geth_rx_interrupt_coalescing_table_t *) + (struct ucc_geth_rx_interrupt_coalescing_table *) qe_muram_addr(ugeth->rx_irq_coalescing_tbl_offset); out_be32(&ugeth->p_rx_glbl_pram->intcoalescingptr, ugeth->rx_irq_coalescing_tbl_offset); @@ -3300,7 +3288,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) l3qt = 0; for (i = 0; i < 8; i++) l3qt |= (ug_info->l3qt[j + i] << (28 - 4 * i)); - out_be32(&ugeth->p_rx_glbl_pram->l3qt[j], l3qt); + out_be32(&ugeth->p_rx_glbl_pram->l3qt[j/8], l3qt); } /* vlantype */ @@ -3316,8 +3304,8 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Size varies with number of Rx queues */ ugeth->rx_bd_qs_tbl_offset = qe_muram_alloc(ug_info->numQueuesRx * - (sizeof(ucc_geth_rx_bd_queues_entry_t) + - sizeof(ucc_geth_rx_prefetched_bds_t)), + (sizeof(struct ucc_geth_rx_bd_queues_entry) + + sizeof(struct ucc_geth_rx_prefetched_bds)), UCC_GETH_RX_BD_QUEUES_ALIGNMENT); if (IS_MURAM_ERR(ugeth->rx_bd_qs_tbl_offset)) { ugeth_err @@ -3328,14 +3316,14 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_rx_bd_qs_tbl = - (ucc_geth_rx_bd_queues_entry_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_rx_bd_queues_entry *) qe_muram_addr(ugeth-> rx_bd_qs_tbl_offset); out_be32(&ugeth->p_rx_glbl_pram->rbdqptr, ugeth->rx_bd_qs_tbl_offset); /* Zero out p_rx_bd_qs_tbl */ memset(ugeth->p_rx_bd_qs_tbl, 0, - ug_info->numQueuesRx * (sizeof(ucc_geth_rx_bd_queues_entry_t) + - sizeof(ucc_geth_rx_prefetched_bds_t))); + ug_info->numQueuesRx * (sizeof(struct ucc_geth_rx_bd_queues_entry) + + sizeof(struct ucc_geth_rx_prefetched_bds))); /* Setup the table */ /* Assume BD rings are already established */ @@ -3406,7 +3394,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Allocate memory for extended filtering Mode Global Parameters */ ugeth->exf_glbl_param_offset = - qe_muram_alloc(sizeof(ucc_geth_exf_global_pram_t), + qe_muram_alloc(sizeof(struct ucc_geth_exf_global_pram), UCC_GETH_RX_EXTENDED_FILTERING_GLOBAL_PARAMETERS_ALIGNMENT); if (IS_MURAM_ERR(ugeth->exf_glbl_param_offset)) { ugeth_err @@ -3417,7 +3405,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_exf_glbl_param = - (ucc_geth_exf_global_pram_t *) qe_muram_addr(ugeth-> + (struct ucc_geth_exf_global_pram *) qe_muram_addr(ugeth-> exf_glbl_param_offset); out_be32(&ugeth->p_rx_glbl_pram->exfGlobalParam, ugeth->exf_glbl_param_offset); @@ -3439,7 +3427,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) INIT_LIST_HEAD(&ugeth->ind_hash_q); } p_82xx_addr_filt = - (ucc_geth_82xx_address_filtering_pram_t *) ugeth-> + (struct ucc_geth_82xx_address_filtering_pram *) ugeth-> p_rx_glbl_pram->addressfiltering; ugeth_82xx_filtering_clear_all_addr_in_hash(ugeth, @@ -3462,7 +3450,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) * allocated resources can be released when the channel is freed. */ if (!(ugeth->p_init_enet_param_shadow = - (ucc_geth_init_pram_t *) kmalloc(sizeof(ucc_geth_init_pram_t), + (struct ucc_geth_init_pram *) kmalloc(sizeof(struct ucc_geth_init_pram), GFP_KERNEL))) { ugeth_err ("%s: Can not allocate memory for" @@ -3472,7 +3460,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } /* Zero out *p_init_enet_param_shadow */ memset((char *)ugeth->p_init_enet_param_shadow, - 0, sizeof(ucc_geth_init_pram_t)); + 0, sizeof(struct ucc_geth_init_pram)); /* Fill shadow InitEnet command parameter structure */ @@ -3506,7 +3494,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } ugeth->p_init_enet_param_shadow->largestexternallookupkeysize = ug_info->largestexternallookupkeysize; - size = sizeof(ucc_geth_thread_rx_pram_t); + size = sizeof(struct ucc_geth_thread_rx_pram); if (ug_info->rxExtendedFiltering) { size += THREAD_RX_PRAM_ADDITIONAL_FOR_EXTENDED_FILTERING; if (ug_info->largestexternallookupkeysize == @@ -3537,7 +3525,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) fill_init_enet_entries(ugeth, &(ugeth->p_init_enet_param_shadow-> txthread[0]), numThreadsTxNumerical, - sizeof(ucc_geth_thread_tx_pram_t), + sizeof(struct ucc_geth_thread_tx_pram), UCC_GETH_THREAD_TX_PRAM_ALIGNMENT, ug_info->riscTx, 0)) != 0) { ugeth_err("%s: Can not fill p_init_enet_param_shadow.", @@ -3557,7 +3545,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) } /* Allocate InitEnet command parameter structure */ - init_enet_pram_offset = qe_muram_alloc(sizeof(ucc_geth_init_pram_t), 4); + init_enet_pram_offset = qe_muram_alloc(sizeof(struct ucc_geth_init_pram), 4); if (IS_MURAM_ERR(init_enet_pram_offset)) { ugeth_err ("%s: Can not allocate DPRAM memory for p_init_enet_pram.", @@ -3566,7 +3554,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) return -ENOMEM; } p_init_enet_pram = - (ucc_geth_init_pram_t *) qe_muram_addr(init_enet_pram_offset); + (struct ucc_geth_init_pram *) qe_muram_addr(init_enet_pram_offset); /* Copy shadow InitEnet command parameter structure into PRAM */ p_init_enet_pram->resinit1 = ugeth->p_init_enet_param_shadow->resinit1; @@ -3591,7 +3579,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* Issue QE command */ cecr_subblock = ucc_fast_get_qe_cr_subblock(ugeth->ug_info->uf_info.ucc_num); - qe_issue_cmd(command, cecr_subblock, (u8) QE_CR_PROTOCOL_ETHERNET, + qe_issue_cmd(command, cecr_subblock, QE_CR_PROTOCOL_ETHERNET, init_enet_pram_offset); /* Free InitEnet command parameter */ @@ -3603,7 +3591,7 @@ static int ucc_geth_startup(ucc_geth_private_t *ugeth) /* returns a net_device_stats structure pointer */ static struct net_device_stats *ucc_geth_get_stats(struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); return &(ugeth->stats); } @@ -3614,7 +3602,7 @@ static struct net_device_stats *ucc_geth_get_stats(struct net_device *dev) * starting over will fix the problem. */ static void ucc_geth_timeout(struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); ugeth_vdbg("%s: IN", __FUNCTION__); @@ -3634,7 +3622,7 @@ static void ucc_geth_timeout(struct net_device *dev) /* It is pointed to by the dev->hard_start_xmit function pointer */ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); u8 *bd; /* BD pointer */ u32 bd_status; u8 txQ = 0; @@ -3647,7 +3635,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Start from the next BD that should be filled */ bd = ugeth->txBd[txQ]; - bd_status = BD_STATUS_AND_LENGTH(bd); + bd_status = in_be32((u32 *)bd); /* Save the skb pointer so we can free it later */ ugeth->tx_skbuff[txQ][ugeth->skb_curtx[txQ]] = skb; @@ -3657,20 +3645,21 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) 1) & TX_RING_MOD_MASK(ugeth->ug_info->bdRingLenTx[txQ]); /* set up the buffer descriptor */ - BD_BUFFER_SET(bd, + out_be32(&((struct qe_bd *)bd)->buf, dma_map_single(NULL, skb->data, skb->len, DMA_TO_DEVICE)); - //printk(KERN_DEBUG"skb->data is 0x%x\n",skb->data); + /* printk(KERN_DEBUG"skb->data is 0x%x\n",skb->data); */ bd_status = (bd_status & T_W) | T_R | T_I | T_L | skb->len; - BD_STATUS_AND_LENGTH_SET(bd, bd_status); + /* set bd status and length */ + out_be32((u32 *)bd, bd_status); dev->trans_start = jiffies; /* Move to next BD in the ring */ if (!(bd_status & T_W)) - ugeth->txBd[txQ] = bd + UCC_GETH_SIZE_OF_BD; + ugeth->txBd[txQ] = bd + sizeof(struct qe_bd); else ugeth->txBd[txQ] = ugeth->p_tx_bd_ring[txQ]; @@ -3695,7 +3684,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) return 0; } -static int ucc_geth_rx(ucc_geth_private_t *ugeth, u8 rxQ, int rx_work_limit) +static int ucc_geth_rx(struct ucc_geth_private *ugeth, u8 rxQ, int rx_work_limit) { struct sk_buff *skb; u8 *bd; @@ -3709,11 +3698,11 @@ static int ucc_geth_rx(ucc_geth_private_t *ugeth, u8 rxQ, int rx_work_limit) /* collect received buffers */ bd = ugeth->rxBd[rxQ]; - bd_status = BD_STATUS_AND_LENGTH(bd); + bd_status = in_be32((u32 *)bd); /* while there are received buffers and BD is full (~R_E) */ while (!((bd_status & (R_E)) || (--rx_work_limit < 0))) { - bdBuffer = (u8 *) BD_BUFFER(bd); + bdBuffer = (u8 *) in_be32(&((struct qe_bd *)bd)->buf); length = (u16) ((bd_status & BD_LENGTH_MASK) - 4); skb = ugeth->rx_skbuff[rxQ][ugeth->skb_currx[rxQ]]; @@ -3768,9 +3757,9 @@ static int ucc_geth_rx(ucc_geth_private_t *ugeth, u8 rxQ, int rx_work_limit) if (bd_status & R_W) bd = ugeth->p_rx_bd_ring[rxQ]; else - bd += UCC_GETH_SIZE_OF_BD; + bd += sizeof(struct qe_bd); - bd_status = BD_STATUS_AND_LENGTH(bd); + bd_status = in_be32((u32 *)bd); } ugeth->rxBd[rxQ] = bd; @@ -3781,12 +3770,12 @@ static int ucc_geth_rx(ucc_geth_private_t *ugeth, u8 rxQ, int rx_work_limit) static int ucc_geth_tx(struct net_device *dev, u8 txQ) { /* Start from the next BD that should be filled */ - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); u8 *bd; /* BD pointer */ u32 bd_status; bd = ugeth->confBd[txQ]; - bd_status = BD_STATUS_AND_LENGTH(bd); + bd_status = in_be32((u32 *)bd); /* Normal processing. */ while ((bd_status & T_R) == 0) { @@ -3813,7 +3802,7 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ) /* Advance the confirmation BD pointer */ if (!(bd_status & T_W)) - ugeth->confBd[txQ] += UCC_GETH_SIZE_OF_BD; + ugeth->confBd[txQ] += sizeof(struct qe_bd); else ugeth->confBd[txQ] = ugeth->p_tx_bd_ring[txQ]; } @@ -3823,7 +3812,7 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ) #ifdef CONFIG_UGETH_NAPI static int ucc_geth_poll(struct net_device *dev, int *budget) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); int howmany; int rx_work_limit = *budget; u8 rxQ = 0; @@ -3847,9 +3836,9 @@ static int ucc_geth_poll(struct net_device *dev, int *budget) static irqreturn_t ucc_geth_irq_handler(int irq, void *info) { struct net_device *dev = (struct net_device *)info; - ucc_geth_private_t *ugeth = netdev_priv(dev); - ucc_fast_private_t *uccf; - ucc_geth_info_t *ug_info; + struct ucc_geth_private *ugeth = netdev_priv(dev); + struct ucc_fast_private *uccf; + struct ucc_geth_info *ug_info; register u32 ucce = 0; register u32 bit_mask = UCCE_RXBF_SINGLE_MASK; register u32 tx_mask = UCCE_TXBF_SINGLE_MASK; @@ -3912,7 +3901,7 @@ static irqreturn_t ucc_geth_irq_handler(int irq, void *info) static irqreturn_t phy_interrupt(int irq, void *dev_id) { struct net_device *dev = (struct net_device *)dev_id; - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); ugeth_vdbg("%s: IN", __FUNCTION__); @@ -3932,8 +3921,8 @@ static irqreturn_t phy_interrupt(int irq, void *dev_id) static void ugeth_phy_change(void *data) { struct net_device *dev = (struct net_device *)data; - ucc_geth_private_t *ugeth = netdev_priv(dev); - ucc_geth_t *ug_regs; + struct ucc_geth_private *ugeth = netdev_priv(dev); + struct ucc_geth *ug_regs; int result = 0; ugeth_vdbg("%s: IN", __FUNCTION__); @@ -3963,7 +3952,7 @@ static void ugeth_phy_change(void *data) static void ugeth_phy_timer(unsigned long data) { struct net_device *dev = (struct net_device *)data; - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); schedule_work(&ugeth->tq); @@ -3979,7 +3968,7 @@ static void ugeth_phy_timer(unsigned long data) static void ugeth_phy_startup_timer(unsigned long data) { struct ugeth_mii_info *mii_info = (struct ugeth_mii_info *)data; - ucc_geth_private_t *ugeth = netdev_priv(mii_info->dev); + struct ucc_geth_private *ugeth = netdev_priv(mii_info->dev); static int secondary = UGETH_AN_TIMEOUT; int result; @@ -4034,7 +4023,7 @@ static void ugeth_phy_startup_timer(unsigned long data) /* Returns 0 for success. */ static int ucc_geth_open(struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); int err; ugeth_vdbg("%s: IN", __FUNCTION__); @@ -4111,7 +4100,7 @@ static int ucc_geth_open(struct net_device *dev) /* Stops the kernel queue, and halts the controller */ static int ucc_geth_close(struct net_device *dev) { - ucc_geth_private_t *ugeth = netdev_priv(dev); + struct ucc_geth_private *ugeth = netdev_priv(dev); ugeth_vdbg("%s: IN", __FUNCTION__); @@ -4130,30 +4119,53 @@ static int ucc_geth_close(struct net_device *dev) const struct ethtool_ops ucc_geth_ethtool_ops = { }; -static int ucc_geth_probe(struct device *device) +static int ucc_geth_probe(struct of_device* ofdev, const struct of_device_id *match) { - struct platform_device *pdev = to_platform_device(device); - struct ucc_geth_platform_data *ugeth_pdata; + struct device *device = &ofdev->dev; + struct device_node *np = ofdev->node; struct net_device *dev = NULL; struct ucc_geth_private *ugeth = NULL; struct ucc_geth_info *ug_info; - int err; + struct resource res; + struct device_node *phy; + int err, ucc_num, phy_interface; static int mii_mng_configured = 0; + const phandle *ph; + const unsigned int *prop; ugeth_vdbg("%s: IN", __FUNCTION__); - ugeth_pdata = (struct ucc_geth_platform_data *)pdev->dev.platform_data; + prop = get_property(np, "device-id", NULL); + ucc_num = *prop - 1; + if ((ucc_num < 0) || (ucc_num > 7)) + return -ENODEV; + + ug_info = &ugeth_info[ucc_num]; + ug_info->uf_info.ucc_num = ucc_num; + prop = get_property(np, "rx-clock", NULL); + ug_info->uf_info.rx_clock = *prop; + prop = get_property(np, "tx-clock", NULL); + ug_info->uf_info.tx_clock = *prop; + err = of_address_to_resource(np, 0, &res); + if (err) + return -EINVAL; + + ug_info->uf_info.regs = res.start; + ug_info->uf_info.irq = irq_of_parse_and_map(np, 0); + + ph = get_property(np, "phy-handle", NULL); + phy = of_find_node_by_phandle(*ph); - ug_info = &ugeth_info[pdev->id]; - ug_info->uf_info.ucc_num = pdev->id; - ug_info->uf_info.rx_clock = ugeth_pdata->rx_clock; - ug_info->uf_info.tx_clock = ugeth_pdata->tx_clock; - ug_info->uf_info.regs = ugeth_pdata->phy_reg_addr; - ug_info->uf_info.irq = platform_get_irq(pdev, 0); - ug_info->phy_address = ugeth_pdata->phy_id; - ug_info->enet_interface = ugeth_pdata->phy_interface; - ug_info->board_flags = ugeth_pdata->board_flags; - ug_info->phy_interrupt = ugeth_pdata->phy_interrupt; + if (phy == NULL) + return -ENODEV; + + prop = get_property(phy, "reg", NULL); + ug_info->phy_address = *prop; + prop = get_property(phy, "interface", NULL); + ug_info->enet_interface = *prop; + ug_info->phy_interrupt = irq_of_parse_and_map(phy, 0); + ug_info->board_flags = (ug_info->phy_interrupt == NO_IRQ)? + 0:FSL_UGETH_BRD_HAS_PHY_INTR; printk(KERN_INFO "ucc_geth: UCC%1d at 0x%8x (irq = %d) \n", ug_info->uf_info.ucc_num + 1, ug_info->uf_info.regs, @@ -4161,12 +4173,44 @@ static int ucc_geth_probe(struct device *device) if (ug_info == NULL) { ugeth_err("%s: [%d] Missing additional data!", __FUNCTION__, - pdev->id); + ucc_num); return -ENODEV; } + /* FIXME: Work around for early chip rev. */ + /* There's a bug in initial chip rev(s) in the RGMII ac */ + /* timing. */ + /* The following compensates by writing to the reserved */ + /* QE Port Output Hold Registers (CPOH1?). */ + prop = get_property(phy, "interface", NULL); + phy_interface = *prop; + if ((phy_interface == ENET_1000_RGMII) || + (phy_interface == ENET_100_RGMII) || + (phy_interface == ENET_10_RGMII)) { + struct device_node *soc; + phys_addr_t immrbase = -1; + u32 *tmp_reg; + u32 tmp_val; + + soc = of_find_node_by_type(NULL, "soc"); + if (soc) { + unsigned int size; + const void *prop = get_property(soc, "reg", &size); + immrbase = of_translate_address(soc, prop); + of_node_put(soc); + }; + + tmp_reg = (u32 *) ioremap(immrbase + 0x14A8, 0x4); + tmp_val = in_be32(tmp_reg); + if (ucc_num == 1) + out_be32(tmp_reg, tmp_val | 0x00003000); + else if (ucc_num == 2) + out_be32(tmp_reg, tmp_val | 0x0c000000); + iounmap(tmp_reg); + } + if (!mii_mng_configured) { - ucc_set_qe_mux_mii_mng(ug_info->uf_info.ucc_num); + ucc_set_qe_mux_mii_mng(ucc_num); mii_mng_configured = 1; } @@ -4213,13 +4257,14 @@ static int ucc_geth_probe(struct device *device) ugeth->ug_info = ug_info; ugeth->dev = dev; - memcpy(dev->dev_addr, ugeth_pdata->mac_addr, 6); + memcpy(dev->dev_addr, get_property(np, "mac-address", NULL), 6); return 0; } -static int ucc_geth_remove(struct device *device) +static int ucc_geth_remove(struct of_device* ofdev) { + struct device *device = &ofdev->dev; struct net_device *dev = dev_get_drvdata(device); struct ucc_geth_private *ugeth = netdev_priv(dev); @@ -4230,28 +4275,38 @@ static int ucc_geth_remove(struct device *device) return 0; } -/* Structure for a device driver */ -static struct device_driver ucc_geth_driver = { - .name = DRV_NAME, - .bus = &platform_bus_type, - .probe = ucc_geth_probe, - .remove = ucc_geth_remove, +static struct of_device_id ucc_geth_match[] = { + { + .type = "network", + .compatible = "ucc_geth", + }, + {}, +}; + +MODULE_DEVICE_TABLE(of, ucc_geth_match); + +static struct of_platform_driver ucc_geth_driver = { + .name = DRV_NAME, + .match_table = ucc_geth_match, + .probe = ucc_geth_probe, + .remove = ucc_geth_remove, }; static int __init ucc_geth_init(void) { int i; + printk(KERN_INFO "ucc_geth: " DRV_DESC "\n"); for (i = 0; i < 8; i++) memcpy(&(ugeth_info[i]), &ugeth_primary_info, sizeof(ugeth_primary_info)); - return driver_register(&ucc_geth_driver); + return of_register_driver(&ucc_geth_driver); } static void __exit ucc_geth_exit(void) { - driver_unregister(&ucc_geth_driver); + of_unregister_driver(&ucc_geth_driver); } module_init(ucc_geth_init); diff --git a/drivers/net/ucc_geth.h b/drivers/net/ucc_geth.h index 005965f..a665612 100644 --- a/drivers/net/ucc_geth.h +++ b/drivers/net/ucc_geth.h @@ -36,24 +36,24 @@ #define ENET_INIT_PARAM_MAX_ENTRIES_RX 9 #define ENET_INIT_PARAM_MAX_ENTRIES_TX 8 -typedef struct ucc_mii_mng { +struct ucc_mii_mng { u32 miimcfg; /* MII management configuration reg */ u32 miimcom; /* MII management command reg */ u32 miimadd; /* MII management address reg */ u32 miimcon; /* MII management control reg */ u32 miimstat; /* MII management status reg */ u32 miimind; /* MII management indication reg */ -} __attribute__ ((packed)) ucc_mii_mng_t; +} __attribute__ ((packed)); -typedef struct ucc_geth { - ucc_fast_t uccf; +struct ucc_geth { + struct ucc_fast uccf; u32 maccfg1; /* mac configuration reg. 1 */ u32 maccfg2; /* mac configuration reg. 2 */ u32 ipgifg; /* interframe gap reg. */ u32 hafdup; /* half-duplex reg. */ u8 res1[0x10]; - ucc_mii_mng_t miimng; /* MII management structure */ + struct ucc_mii_mng miimng; /* MII management structure */ u32 ifctl; /* interface control reg */ u32 ifstat; /* interface statux reg */ u32 macstnaddr1; /* mac station address part 1 reg */ @@ -111,7 +111,7 @@ typedef struct ucc_geth { u32 scar; /* Statistics carry register */ u32 scam; /* Statistics caryy mask register */ u8 res5[0x200 - 0x1c4]; -} __attribute__ ((packed)) ucc_geth_t; +} __attribute__ ((packed)); /* UCC GETH TEMODR Register */ #define TEMODER_TX_RMON_STATISTICS_ENABLE 0x0100 /* enable Tx statistics @@ -508,39 +508,39 @@ typedef struct ucc_geth { /* UCC GETH UDSR (Data Synchronization Register) */ #define UDSR_MAGIC 0x067E -typedef struct ucc_geth_thread_data_tx { +struct ucc_geth_thread_data_tx { u8 res0[104]; -} __attribute__ ((packed)) ucc_geth_thread_data_tx_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_thread_data_rx { +struct ucc_geth_thread_data_rx { u8 res0[40]; -} __attribute__ ((packed)) ucc_geth_thread_data_rx_t; +} __attribute__ ((packed)); /* Send Queue Queue-Descriptor */ -typedef struct ucc_geth_send_queue_qd { +struct ucc_geth_send_queue_qd { u32 bd_ring_base; /* pointer to BD ring base address */ u8 res0[0x8]; u32 last_bd_completed_address;/* initialize to last entry in BD ring */ u8 res1[0x30]; -} __attribute__ ((packed)) ucc_geth_send_queue_qd_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_send_queue_mem_region { - ucc_geth_send_queue_qd_t sqqd[NUM_TX_QUEUES]; -} __attribute__ ((packed)) ucc_geth_send_queue_mem_region_t; +struct ucc_geth_send_queue_mem_region { + struct ucc_geth_send_queue_qd sqqd[NUM_TX_QUEUES]; +} __attribute__ ((packed)); -typedef struct ucc_geth_thread_tx_pram { +struct ucc_geth_thread_tx_pram { u8 res0[64]; -} __attribute__ ((packed)) ucc_geth_thread_tx_pram_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_thread_rx_pram { +struct ucc_geth_thread_rx_pram { u8 res0[128]; -} __attribute__ ((packed)) ucc_geth_thread_rx_pram_t; +} __attribute__ ((packed)); #define THREAD_RX_PRAM_ADDITIONAL_FOR_EXTENDED_FILTERING 64 #define THREAD_RX_PRAM_ADDITIONAL_FOR_EXTENDED_FILTERING_8 64 #define THREAD_RX_PRAM_ADDITIONAL_FOR_EXTENDED_FILTERING_16 96 -typedef struct ucc_geth_scheduler { +struct ucc_geth_scheduler { u16 cpucount0; /* CPU packet counter */ u16 cpucount1; /* CPU packet counter */ u16 cecount0; /* QE packet counter */ @@ -574,9 +574,9 @@ typedef struct ucc_geth_scheduler { /**< weight factor for queues */ u32 minw; /* temporary variable handled by QE */ u8 res1[0x70 - 0x64]; -} __attribute__ ((packed)) ucc_geth_scheduler_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_tx_firmware_statistics_pram { +struct ucc_geth_tx_firmware_statistics_pram { u32 sicoltx; /* single collision */ u32 mulcoltx; /* multiple collision */ u32 latecoltxfr; /* late collision */ @@ -596,9 +596,9 @@ typedef struct ucc_geth_tx_firmware_statistics_pram { and 1518 octets */ u32 txpktsjumbo; /* total packets (including bad) between 1024 and MAXLength octets */ -} __attribute__ ((packed)) ucc_geth_tx_firmware_statistics_pram_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_rx_firmware_statistics_pram { +struct ucc_geth_rx_firmware_statistics_pram { u32 frrxfcser; /* frames with crc error */ u32 fraligner; /* frames with alignment error */ u32 inrangelenrxer; /* in range length error */ @@ -630,33 +630,33 @@ typedef struct ucc_geth_rx_firmware_statistics_pram { replaced */ u32 insertvlan; /* total frames that had their VLAN tag inserted */ -} __attribute__ ((packed)) ucc_geth_rx_firmware_statistics_pram_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_rx_interrupt_coalescing_entry { +struct ucc_geth_rx_interrupt_coalescing_entry { u32 interruptcoalescingmaxvalue; /* interrupt coalescing max value */ u32 interruptcoalescingcounter; /* interrupt coalescing counter, initialize to interruptcoalescingmaxvalue */ -} __attribute__ ((packed)) ucc_geth_rx_interrupt_coalescing_entry_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_rx_interrupt_coalescing_table { - ucc_geth_rx_interrupt_coalescing_entry_t coalescingentry[NUM_RX_QUEUES]; +struct ucc_geth_rx_interrupt_coalescing_table { + struct ucc_geth_rx_interrupt_coalescing_entry coalescingentry[NUM_RX_QUEUES]; /**< interrupt coalescing entry */ -} __attribute__ ((packed)) ucc_geth_rx_interrupt_coalescing_table_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_rx_prefetched_bds { - qe_bd_t bd[NUM_BDS_IN_PREFETCHED_BDS]; /* prefetched bd */ -} __attribute__ ((packed)) ucc_geth_rx_prefetched_bds_t; +struct ucc_geth_rx_prefetched_bds { + struct qe_bd bd[NUM_BDS_IN_PREFETCHED_BDS]; /* prefetched bd */ +} __attribute__ ((packed)); -typedef struct ucc_geth_rx_bd_queues_entry { +struct ucc_geth_rx_bd_queues_entry { u32 bdbaseptr; /* BD base pointer */ u32 bdptr; /* BD pointer */ u32 externalbdbaseptr; /* external BD base pointer */ u32 externalbdptr; /* external BD pointer */ -} __attribute__ ((packed)) ucc_geth_rx_bd_queues_entry_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_tx_global_pram { +struct ucc_geth_tx_global_pram { u16 temoder; u8 res0[0x38 - 0x02]; u32 sqptr; /* a base pointer to send queue memory region */ @@ -670,15 +670,15 @@ typedef struct ucc_geth_tx_global_pram { u32 tqptr; /* a base pointer to the Tx Queues Memory Region */ u8 res2[0x80 - 0x74]; -} __attribute__ ((packed)) ucc_geth_tx_global_pram_t; +} __attribute__ ((packed)); /* structure representing Extended Filtering Global Parameters in PRAM */ -typedef struct ucc_geth_exf_global_pram { +struct ucc_geth_exf_global_pram { u32 l2pcdptr; /* individual address filter, high */ u8 res0[0x10 - 0x04]; -} __attribute__ ((packed)) ucc_geth_exf_global_pram_t; +} __attribute__ ((packed)); -typedef struct ucc_geth_rx_global_pram { +struct ucc_geth_rx_global_pram { u32 remoder; /* ethernet mode reg. */ u32 rqptr; /* base pointer to the Rx Queues Memory Region*/ u32 res0[0x1]; @@ -710,12 +710,12 @@ typedef struct ucc_geth_rx_global_pram { u32 exfGlobalParam; /* base address for extended filtering global parameters */ u8 res6[0x100 - 0xC4]; /* Initialize to zero */ -} __attribute__ ((packed)) ucc_geth_rx_global_pram_t; +} __attribute__ ((packed)); #define GRACEFUL_STOP_ACKNOWLEDGE_RX 0x01 /* structure representing InitEnet command */ -typedef struct ucc_geth_init_pram { +struct ucc_geth_init_pram { u8 resinit1; u8 resinit2; u8 resinit3; @@ -729,7 +729,7 @@ typedef struct ucc_geth_init_pram { u32 txglobal; /* tx global */ u32 txthread[ENET_INIT_PARAM_MAX_ENTRIES_TX]; /* tx threads */ u8 res3[0x1]; -} __attribute__ ((packed)) ucc_geth_init_pram_t; +} __attribute__ ((packed)); #define ENET_INIT_PARAM_RGF_SHIFT (32 - 4) #define ENET_INIT_PARAM_TGF_SHIFT (32 - 8) @@ -746,27 +746,27 @@ typedef struct ucc_geth_init_pram { #define ENET_INIT_PARAM_MAGIC_RES_INIT5 0x0400 /* structure representing 82xx Address Filtering Enet Address in PRAM */ -typedef struct ucc_geth_82xx_enet_address { +struct ucc_geth_82xx_enet_address { u8 res1[0x2]; u16 h; /* address (MSB) */ u16 m; /* address */ u16 l; /* address (LSB) */ -} __attribute__ ((packed)) ucc_geth_82xx_enet_address_t; +} __attribute__ ((packed)); /* structure representing 82xx Address Filtering PRAM */ -typedef struct ucc_geth_82xx_address_filtering_pram { +struct ucc_geth_82xx_address_filtering_pram { u32 iaddr_h; /* individual address filter, high */ u32 iaddr_l; /* individual address filter, low */ u32 gaddr_h; /* group address filter, high */ u32 gaddr_l; /* group address filter, low */ - ucc_geth_82xx_enet_address_t taddr; - ucc_geth_82xx_enet_address_t paddr[NUM_OF_PADDRS]; + struct ucc_geth_82xx_enet_address taddr; + struct ucc_geth_82xx_enet_address paddr[NUM_OF_PADDRS]; u8 res0[0x40 - 0x38]; -} __attribute__ ((packed)) ucc_geth_82xx_address_filtering_pram_t; +} __attribute__ ((packed)); /* GETH Tx firmware statistics structure, used when calling UCC_GETH_GetStatistics. */ -typedef struct ucc_geth_tx_firmware_statistics { +struct ucc_geth_tx_firmware_statistics { u32 sicoltx; /* single collision */ u32 mulcoltx; /* multiple collision */ u32 latecoltxfr; /* late collision */ @@ -786,11 +786,11 @@ typedef struct ucc_geth_tx_firmware_statistics { and 1518 octets */ u32 txpktsjumbo; /* total packets (including bad) between 1024 and MAXLength octets */ -} __attribute__ ((packed)) ucc_geth_tx_firmware_statistics_t; +} __attribute__ ((packed)); /* GETH Rx firmware statistics structure, used when calling UCC_GETH_GetStatistics. */ -typedef struct ucc_geth_rx_firmware_statistics { +struct ucc_geth_rx_firmware_statistics { u32 frrxfcser; /* frames with crc error */ u32 fraligner; /* frames with alignment error */ u32 inrangelenrxer; /* in range length error */ @@ -822,11 +822,11 @@ typedef struct ucc_geth_rx_firmware_statistics { replaced */ u32 insertvlan; /* total frames that had their VLAN tag inserted */ -} __attribute__ ((packed)) ucc_geth_rx_firmware_statistics_t; +} __attribute__ ((packed)); /* GETH hardware statistics structure, used when calling UCC_GETH_GetStatistics. */ -typedef struct ucc_geth_hardware_statistics { +struct ucc_geth_hardware_statistics { u32 tx64; /* Total number of frames (including bad frames) transmitted that were exactly of the minimal length (64 for un tagged, 68 for @@ -871,7 +871,7 @@ typedef struct ucc_geth_hardware_statistics { u32 rbca; /* Total number of frames received succesfully that had destination address equal to the broadcast address */ -} __attribute__ ((packed)) ucc_geth_hardware_statistics_t; +} __attribute__ ((packed)); /* UCC GETH Tx errors returned via TxConf callback */ #define TX_ERRORS_DEF 0x0200 @@ -1013,21 +1013,21 @@ typedef struct ucc_geth_hardware_statistics { (MIIMCFG_MANAGEMENT_CLOCK_DIVIDE_BY_112) /* Ethernet speed */ -typedef enum enet_speed { +enum enet_speed { ENET_SPEED_10BT, /* 10 Base T */ ENET_SPEED_100BT, /* 100 Base T */ ENET_SPEED_1000BT /* 1000 Base T */ -} enet_speed_e; +}; /* Ethernet Address Type. */ -typedef enum enet_addr_type { +enum enet_addr_type { ENET_ADDR_TYPE_INDIVIDUAL, ENET_ADDR_TYPE_GROUP, ENET_ADDR_TYPE_BROADCAST -} enet_addr_type_e; +}; /* TBI / MII Set Register */ -typedef enum enet_tbi_mii_reg { +enum enet_tbi_mii_reg { ENET_TBI_MII_CR = 0x00, /* Control (CR ) */ ENET_TBI_MII_SR = 0x01, /* Status (SR ) */ ENET_TBI_MII_ANA = 0x04, /* AN advertisement (ANA ) */ @@ -1040,10 +1040,10 @@ typedef enum enet_tbi_mii_reg { ENET_TBI_MII_EXST = 0x0F, /* Extended status (EXST ) */ ENET_TBI_MII_JD = 0x10, /* Jitter diagnostics (JD ) */ ENET_TBI_MII_TBICON = 0x11 /* TBI control (TBICON ) */ -} enet_tbi_mii_reg_e; +}; /* UCC GETH 82xx Ethernet Address Recognition Location */ -typedef enum ucc_geth_enet_address_recognition_location { +enum ucc_geth_enet_address_recognition_location { UCC_GETH_ENET_ADDRESS_RECOGNITION_LOCATION_STATION_ADDRESS,/* station address */ UCC_GETH_ENET_ADDRESS_RECOGNITION_LOCATION_PADDR_FIRST, /* additional @@ -1065,10 +1065,10 @@ typedef enum ucc_geth_enet_address_recognition_location { UCC_GETH_ENET_ADDRESS_RECOGNITION_LOCATION_GROUP_HASH, /* group hash */ UCC_GETH_ENET_ADDRESS_RECOGNITION_LOCATION_INDIVIDUAL_HASH /* individual hash */ -} ucc_geth_enet_address_recognition_location_e; +}; /* UCC GETH vlan operation tagged */ -typedef enum ucc_geth_vlan_operation_tagged { +enum ucc_geth_vlan_operation_tagged { UCC_GETH_VLAN_OPERATION_TAGGED_NOP = 0x0, /* Tagged - nop */ UCC_GETH_VLAN_OPERATION_TAGGED_REPLACE_VID_PORTION_OF_Q_TAG = 0x1, /* Tagged - replace vid portion of q tag */ @@ -1076,18 +1076,18 @@ typedef enum ucc_geth_vlan_operation_tagged { = 0x2, /* Tagged - if vid0 replace vid with default value */ UCC_GETH_VLAN_OPERATION_TAGGED_EXTRACT_Q_TAG_FROM_FRAME = 0x3 /* Tagged - extract q tag from frame */ -} ucc_geth_vlan_operation_tagged_e; +}; /* UCC GETH vlan operation non-tagged */ -typedef enum ucc_geth_vlan_operation_non_tagged { +enum ucc_geth_vlan_operation_non_tagged { UCC_GETH_VLAN_OPERATION_NON_TAGGED_NOP = 0x0, /* Non tagged - nop */ UCC_GETH_VLAN_OPERATION_NON_TAGGED_Q_TAG_INSERT = 0x1 /* Non tagged - q tag insert */ -} ucc_geth_vlan_operation_non_tagged_e; +}; /* UCC GETH Rx Quality of Service Mode */ -typedef enum ucc_geth_qos_mode { +enum ucc_geth_qos_mode { UCC_GETH_QOS_MODE_DEFAULT = 0x0, /* default queue */ UCC_GETH_QOS_MODE_QUEUE_NUM_FROM_L2_CRITERIA = 0x1, /* queue determined @@ -1097,11 +1097,11 @@ typedef enum ucc_geth_qos_mode { determined by L3 criteria */ -} ucc_geth_qos_mode_e; +}; /* UCC GETH Statistics Gathering Mode - These are bit flags, 'or' them together for combined functionality */ -typedef enum ucc_geth_statistics_gathering_mode { +enum ucc_geth_statistics_gathering_mode { UCC_GETH_STATISTICS_GATHERING_MODE_NONE = 0x00000000, /* No statistics gathering */ @@ -1122,10 +1122,10 @@ typedef enum ucc_geth_statistics_gathering_mode { statistics gathering */ -} ucc_geth_statistics_gathering_mode_e; +}; /* UCC GETH Pad and CRC Mode - Note, Padding without CRC is not possible */ -typedef enum ucc_geth_maccfg2_pad_and_crc_mode { +enum ucc_geth_maccfg2_pad_and_crc_mode { UCC_GETH_PAD_AND_CRC_MODE_NONE = MACCFG2_PAD_AND_CRC_MODE_NONE, /* Neither Padding short frames @@ -1135,61 +1135,59 @@ typedef enum ucc_geth_maccfg2_pad_and_crc_mode { CRC only */ UCC_GETH_PAD_AND_CRC_MODE_PAD_AND_CRC = MACCFG2_PAD_AND_CRC_MODE_PAD_AND_CRC -} ucc_geth_maccfg2_pad_and_crc_mode_e; +}; /* UCC GETH upsmr Flow Control Mode */ -typedef enum ucc_geth_flow_control_mode { +enum ucc_geth_flow_control_mode { UPSMR_AUTOMATIC_FLOW_CONTROL_MODE_NONE = 0x00000000, /* No automatic flow control */ UPSMR_AUTOMATIC_FLOW_CONTROL_MODE_PAUSE_WHEN_EMERGENCY = 0x00004000 /* Send pause frame when RxFIFO reaches its emergency threshold */ -} ucc_geth_flow_control_mode_e; +}; /* UCC GETH number of threads */ -typedef enum ucc_geth_num_of_threads { +enum ucc_geth_num_of_threads { UCC_GETH_NUM_OF_THREADS_1 = 0x1, /* 1 */ UCC_GETH_NUM_OF_THREADS_2 = 0x2, /* 2 */ UCC_GETH_NUM_OF_THREADS_4 = 0x0, /* 4 */ UCC_GETH_NUM_OF_THREADS_6 = 0x3, /* 6 */ UCC_GETH_NUM_OF_THREADS_8 = 0x4 /* 8 */ -} ucc_geth_num_of_threads_e; +}; /* UCC GETH number of station addresses */ -typedef enum ucc_geth_num_of_station_addresses { +enum ucc_geth_num_of_station_addresses { UCC_GETH_NUM_OF_STATION_ADDRESSES_1, /* 1 */ UCC_GETH_NUM_OF_STATION_ADDRESSES_5 /* 5 */ -} ucc_geth_num_of_station_addresses_e; - -typedef u8 enet_addr_t[ENET_NUM_OCTETS_PER_ADDRESS]; +}; /* UCC GETH 82xx Ethernet Address Container */ -typedef struct enet_addr_container { - enet_addr_t address; /* ethernet address */ - ucc_geth_enet_address_recognition_location_e location; /* location in +struct enet_addr_container { + u8 address[ENET_NUM_OCTETS_PER_ADDRESS]; /* ethernet address */ + enum ucc_geth_enet_address_recognition_location location; /* location in 82xx address recognition hardware */ struct list_head node; -} enet_addr_container_t; +}; -#define ENET_ADDR_CONT_ENTRY(ptr) list_entry(ptr, enet_addr_container_t, node) +#define ENET_ADDR_CONT_ENTRY(ptr) list_entry(ptr, struct enet_addr_container, node) /* UCC GETH Termination Action Descriptor (TAD) structure. */ -typedef struct ucc_geth_tad_params { +struct ucc_geth_tad_params { int rx_non_dynamic_extended_features_mode; int reject_frame; - ucc_geth_vlan_operation_tagged_e vtag_op; - ucc_geth_vlan_operation_non_tagged_e vnontag_op; - ucc_geth_qos_mode_e rqos; + enum ucc_geth_vlan_operation_tagged vtag_op; + enum ucc_geth_vlan_operation_non_tagged vnontag_op; + enum ucc_geth_qos_mode rqos; u8 vpri; u16 vid; -} ucc_geth_tad_params_t; +}; /* GETH protocol initialization structure */ -typedef struct ucc_geth_info { - ucc_fast_info_t uf_info; +struct ucc_geth_info { + struct ucc_fast_info uf_info; u8 numQueuesTx; u8 numQueuesRx; int ipCheckSumCheck; @@ -1251,51 +1249,51 @@ typedef struct ucc_geth_info { u8 iphoffset[TX_IP_OFFSET_ENTRY_MAX]; u16 bdRingLenTx[NUM_TX_QUEUES]; u16 bdRingLenRx[NUM_RX_QUEUES]; - enet_interface_e enet_interface; - ucc_geth_num_of_station_addresses_e numStationAddresses; - qe_fltr_largest_external_tbl_lookup_key_size_e + enum enet_interface enet_interface; + enum ucc_geth_num_of_station_addresses numStationAddresses; + enum qe_fltr_largest_external_tbl_lookup_key_size largestexternallookupkeysize; - ucc_geth_statistics_gathering_mode_e statisticsMode; - ucc_geth_vlan_operation_tagged_e vlanOperationTagged; - ucc_geth_vlan_operation_non_tagged_e vlanOperationNonTagged; - ucc_geth_qos_mode_e rxQoSMode; - ucc_geth_flow_control_mode_e aufc; - ucc_geth_maccfg2_pad_and_crc_mode_e padAndCrc; - ucc_geth_num_of_threads_e numThreadsTx; - ucc_geth_num_of_threads_e numThreadsRx; - qe_risc_allocation_e riscTx; - qe_risc_allocation_e riscRx; -} ucc_geth_info_t; + enum ucc_geth_statistics_gathering_mode statisticsMode; + enum ucc_geth_vlan_operation_tagged vlanOperationTagged; + enum ucc_geth_vlan_operation_non_tagged vlanOperationNonTagged; + enum ucc_geth_qos_mode rxQoSMode; + enum ucc_geth_flow_control_mode aufc; + enum ucc_geth_maccfg2_pad_and_crc_mode padAndCrc; + enum ucc_geth_num_of_threads numThreadsTx; + enum ucc_geth_num_of_threads numThreadsRx; + enum qe_risc_allocation riscTx; + enum qe_risc_allocation riscRx; +}; /* structure representing UCC GETH */ -typedef struct ucc_geth_private { - ucc_geth_info_t *ug_info; - ucc_fast_private_t *uccf; +struct ucc_geth_private { + struct ucc_geth_info *ug_info; + struct ucc_fast_private *uccf; struct net_device *dev; struct net_device_stats stats; /* linux network statistics */ - ucc_geth_t *ug_regs; - ucc_geth_init_pram_t *p_init_enet_param_shadow; - ucc_geth_exf_global_pram_t *p_exf_glbl_param; + struct ucc_geth *ug_regs; + struct ucc_geth_init_pram *p_init_enet_param_shadow; + struct ucc_geth_exf_global_pram *p_exf_glbl_param; u32 exf_glbl_param_offset; - ucc_geth_rx_global_pram_t *p_rx_glbl_pram; + struct ucc_geth_rx_global_pram *p_rx_glbl_pram; u32 rx_glbl_pram_offset; - ucc_geth_tx_global_pram_t *p_tx_glbl_pram; + struct ucc_geth_tx_global_pram *p_tx_glbl_pram; u32 tx_glbl_pram_offset; - ucc_geth_send_queue_mem_region_t *p_send_q_mem_reg; + struct ucc_geth_send_queue_mem_region *p_send_q_mem_reg; u32 send_q_mem_reg_offset; - ucc_geth_thread_data_tx_t *p_thread_data_tx; + struct ucc_geth_thread_data_tx *p_thread_data_tx; u32 thread_dat_tx_offset; - ucc_geth_thread_data_rx_t *p_thread_data_rx; + struct ucc_geth_thread_data_rx *p_thread_data_rx; u32 thread_dat_rx_offset; - ucc_geth_scheduler_t *p_scheduler; + struct ucc_geth_scheduler *p_scheduler; u32 scheduler_offset; - ucc_geth_tx_firmware_statistics_pram_t *p_tx_fw_statistics_pram; + struct ucc_geth_tx_firmware_statistics_pram *p_tx_fw_statistics_pram; u32 tx_fw_statistics_pram_offset; - ucc_geth_rx_firmware_statistics_pram_t *p_rx_fw_statistics_pram; + struct ucc_geth_rx_firmware_statistics_pram *p_rx_fw_statistics_pram; u32 rx_fw_statistics_pram_offset; - ucc_geth_rx_interrupt_coalescing_table_t *p_rx_irq_coalescing_tbl; + struct ucc_geth_rx_interrupt_coalescing_table *p_rx_irq_coalescing_tbl; u32 rx_irq_coalescing_tbl_offset; - ucc_geth_rx_bd_queues_entry_t *p_rx_bd_qs_tbl; + struct ucc_geth_rx_bd_queues_entry *p_rx_bd_qs_tbl; u32 rx_bd_qs_tbl_offset; u8 *p_tx_bd_ring[NUM_TX_QUEUES]; u32 tx_bd_ring_offset[NUM_TX_QUEUES]; @@ -1308,7 +1306,7 @@ typedef struct ucc_geth_private { u16 cpucount[NUM_TX_QUEUES]; volatile u16 *p_cpucount[NUM_TX_QUEUES]; int indAddrRegUsed[NUM_OF_PADDRS]; - enet_addr_t paddr[NUM_OF_PADDRS]; + u8 paddr[NUM_OF_PADDRS][ENET_NUM_OCTETS_PER_ADDRESS]; /* ethernet address */ u8 numGroupAddrInHash; u8 numIndAddrInHash; u8 numIndAddrInReg; @@ -1334,6 +1332,6 @@ typedef struct ucc_geth_private { int oldspeed; int oldduplex; int oldlink; -} ucc_geth_private_t; +}; #endif /* __UCC_GETH_H__ */ diff --git a/drivers/net/ucc_geth_phy.c b/drivers/net/ucc_geth_phy.c index 67260eb..5360ec0 100644 --- a/drivers/net/ucc_geth_phy.c +++ b/drivers/net/ucc_geth_phy.c @@ -42,7 +42,6 @@ #include "ucc_geth.h" #include "ucc_geth_phy.h" -#include #define ugphy_printk(level, format, arg...) \ printk(level format "\n", ## arg) @@ -72,16 +71,14 @@ static int genmii_read_status(struct ugeth_mii_info *mii_info); u16 phy_read(struct ugeth_mii_info *mii_info, u16 regnum); void phy_write(struct ugeth_mii_info *mii_info, u16 regnum, u16 val); -static u8 *bcsr_regs = NULL; - /* Write value to the PHY for this device to the register at regnum, */ /* waiting until the write is done before it returns. All PHY */ /* configuration has to be done through the TSEC1 MIIM regs */ void write_phy_reg(struct net_device *dev, int mii_id, int regnum, int value) { - ucc_geth_private_t *ugeth = netdev_priv(dev); - ucc_mii_mng_t *mii_regs; - enet_tbi_mii_reg_e mii_reg = (enet_tbi_mii_reg_e) regnum; + struct ucc_geth_private *ugeth = netdev_priv(dev); + struct ucc_mii_mng *mii_regs; + enum enet_tbi_mii_reg mii_reg = (enum enet_tbi_mii_reg) regnum; u32 tmp_reg; ugphy_vdbg("%s: IN", __FUNCTION__); @@ -116,9 +113,9 @@ void write_phy_reg(struct net_device *dev, int mii_id, int regnum, int value) /* configuration has to be done through the TSEC1 MIIM regs */ int read_phy_reg(struct net_device *dev, int mii_id, int regnum) { - ucc_geth_private_t *ugeth = netdev_priv(dev); - ucc_mii_mng_t *mii_regs; - enet_tbi_mii_reg_e mii_reg = (enet_tbi_mii_reg_e) regnum; + struct ucc_geth_private *ugeth = netdev_priv(dev); + struct ucc_mii_mng *mii_regs; + enum enet_tbi_mii_reg mii_reg = (enum enet_tbi_mii_reg) regnum; u32 tmp_reg; u16 value; @@ -634,11 +631,6 @@ static void dm9161_close(struct ugeth_mii_info *mii_info) static int dm9161_ack_interrupt(struct ugeth_mii_info *mii_info) { -/* FIXME: This lines are for BUG fixing in the mpc8325. -Remove this from here when it's fixed */ - if (bcsr_regs == NULL) - bcsr_regs = (u8 *) ioremap(BCSR_PHYS_ADDR, BCSR_SIZE); - bcsr_regs[14] |= 0x40; ugphy_vdbg("%s: IN", __FUNCTION__); /* Clear the interrupts by reading the reg */ @@ -650,12 +642,6 @@ Remove this from here when it's fixed */ static int dm9161_config_intr(struct ugeth_mii_info *mii_info) { -/* FIXME: This lines are for BUG fixing in the mpc8325. -Remove this from here when it's fixed */ - if (bcsr_regs == NULL) { - bcsr_regs = (u8 *) ioremap(BCSR_PHYS_ADDR, BCSR_SIZE); - bcsr_regs[14] &= ~0x40; - } ugphy_vdbg("%s: IN", __FUNCTION__); if (mii_info->interrupts == MII_INTERRUPT_ENABLED) diff --git a/drivers/net/ucc_geth_phy.h b/drivers/net/ucc_geth_phy.h index 2f98b8f..f574078 100644 --- a/drivers/net/ucc_geth_phy.h +++ b/drivers/net/ucc_geth_phy.h @@ -126,7 +126,7 @@ struct ugeth_mii_info { /* And management functions */ struct phy_info *phyinfo; - ucc_mii_mng_t *mii_regs; + struct ucc_mii_mng *mii_regs; /* forced speed & duplex (no autoneg) * partner speed & duplex & pause (autoneg) -- cgit v0.10.2 From 089fff2aa8cc2a0383ea9fce17afd10bfab9ac7c Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Wed, 18 Oct 2006 00:30:27 -0400 Subject: [PATCH] Remove useless comment from sb1250 Signed-off-by: Dave Jones Signed-off-by: Jeff Garzik diff --git a/drivers/net/sb1250-mac.c b/drivers/net/sb1250-mac.c index db23249..1eae16b 100644 --- a/drivers/net/sb1250-mac.c +++ b/drivers/net/sb1250-mac.c @@ -2903,7 +2903,7 @@ sbmac_init_module(void) dev = alloc_etherdev(sizeof(struct sbmac_softc)); if (!dev) - return -ENOMEM; /* return ENOMEM */ + return -ENOMEM; printk(KERN_DEBUG "sbmac: configuring MAC at %lx\n", port); -- cgit v0.10.2 From cfadbd298e8b3e7f2e324696b653bb74094590db Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 18 Oct 2006 02:15:37 +0100 Subject: [PATCH] Fix timer race When closing the driver or reinitializing the hardware there is the usual del_timer() race condition that exists when timers re-add themselves. Fix by conversion to del_timer_sync(). Signed-off-by: Ralf Baechle Signed-off-by: Jeff Garzik diff --git a/drivers/net/ioc3-eth.c b/drivers/net/ioc3-eth.c index e963dbf..f56b00e 100644 --- a/drivers/net/ioc3-eth.c +++ b/drivers/net/ioc3-eth.c @@ -1017,7 +1017,7 @@ static void ioc3_init(struct net_device *dev) struct ioc3_private *ip = netdev_priv(dev); struct ioc3 *ioc3 = ip->regs; - del_timer(&ip->ioc3_timer); /* Kill if running */ + del_timer_sync(&ip->ioc3_timer); /* Kill if running */ ioc3_w_emcr(EMCR_RST); /* Reset */ (void) ioc3_r_emcr(); /* Flush WB */ @@ -1081,7 +1081,7 @@ static int ioc3_close(struct net_device *dev) { struct ioc3_private *ip = netdev_priv(dev); - del_timer(&ip->ioc3_timer); + del_timer_sync(&ip->ioc3_timer); netif_stop_queue(dev); -- cgit v0.10.2 From 5826cade4341a6298eb10d476dccc5f403ca7ad8 Mon Sep 17 00:00:00 2001 From: David Gibson Date: Fri, 13 Oct 2006 14:20:59 +1000 Subject: [PATCH] ibmveth: Fix index increment calculation On Thu, Oct 12, 2006 at 06:22:14PM +1000, David Gibson wrote: > Your recent ibmveth commit, 751ae21c6cd1493e3d0a4935b08fb298b9d89773 > ("fix int rollover panic"), causes a rapid oops on my test machine > (POWER5 LPAR). > > I've bisected it down to that commit, but am still investigating the > cause of the crash itself. Found the problem, I believe: an object lesson in the need for great caution using ++. [...] @@ -213,6 +213,7 @@ static void ibmveth_replenish_buffer_poo } free_index = pool->consumer_index++ % pool->size; + pool->consumer_index = free_index; index = pool->free_map[free_index]; ibmveth_assert(index != IBM_VETH_INVALID_MAP); Since the ++ is used as post-increment, the increment is not included in free_index, and so the added line effectively reverts the increment. The produced_index side has an analagous bug. The following change corrects this: The recent commit 751ae21c6cd1493e3d0a4935b08fb298b9d89773 introduced a bug in the producer/consumer index calculation in the ibmveth driver - incautious use of the post-increment ++ operator resulted in an increment being immediately reverted. This patch corrects the logic. Without this patch, the driver oopses almost immediately after activation on at least some machines. Signed-off-by: David Gibson Signed-off-by: Jeff Garzik diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 2802db2..44c9f99 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -212,8 +212,8 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc break; } - free_index = pool->consumer_index++ % pool->size; - pool->consumer_index = free_index; + free_index = pool->consumer_index; + pool->consumer_index = (pool->consumer_index + 1) % pool->size; index = pool->free_map[free_index]; ibmveth_assert(index != IBM_VETH_INVALID_MAP); @@ -329,8 +329,10 @@ static void ibmveth_remove_buffer_from_pool(struct ibmveth_adapter *adapter, u64 adapter->rx_buff_pool[pool].buff_size, DMA_FROM_DEVICE); - free_index = adapter->rx_buff_pool[pool].producer_index++ % adapter->rx_buff_pool[pool].size; - adapter->rx_buff_pool[pool].producer_index = free_index; + free_index = adapter->rx_buff_pool[pool].producer_index; + adapter->rx_buff_pool[pool].producer_index + = (adapter->rx_buff_pool[pool].producer_index + 1) + % adapter->rx_buff_pool[pool].size; adapter->rx_buff_pool[pool].free_map[free_index] = index; mb(); -- cgit v0.10.2 From 158f30c8945fea7cf0d0161cd9463cf2f3d2c19e Mon Sep 17 00:00:00 2001 From: Kristen Carlson Accardi Date: Thu, 19 Oct 2006 13:27:39 -0700 Subject: [PATCH] libata: use correct map_db values for ICH8 Use valid values for ICH8 map_db. With the old values, when the controller was in Native mode, and SCC was 1 (drives configured for IDE), any drive plugged into a slave port was not recognized. For Combined Mode (and SCC is still 1), 2 is a value value for MAP.map_value, and needs to be recognized. Signed-off-by: Kristen Carlson Accardi Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 5719704..5250187 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -432,9 +432,9 @@ static const struct piix_map_db ich8_map_db = { .present_shift = 8, .map = { /* PM PS SM SS MAP */ - { P0, NA, P1, NA }, /* 00b (hardwired) */ + { P0, P2, P1, P3 }, /* 00b (hardwired when in AHCI) */ { RV, RV, RV, RV }, - { RV, RV, RV, RV }, /* 10b (never) */ + { IDE, IDE, NA, NA }, /* 10b (IDE mode) */ { RV, RV, RV, RV }, }, }; -- cgit v0.10.2 From bf2d401bca3681f5380f711be65f2026255cc166 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 20 Oct 2006 14:39:35 -0700 Subject: [PATCH] ATA must depend on BLOCK MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the following compile error with CONFIG_ATA=y, CONFIG_BLOCK=n: ... CC drivers/ata/libata-scsi.o /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c: In function ‘ata_scsi_dev_config’: /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:791: warning: implicit declaration of function ‘blk_queue_max_sectors’ /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:799: error: ‘request_queue_t’ undeclared (first use in this function) /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:799: error: (Each undeclared identifier is reported only once /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:799: error: for each function it appears in.) /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:799: error: ‘q’ undeclared (first use in this function) /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:800: warning: implicit declaration of function ‘blk_queue_max_hw_segments’ /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c: In function ‘ata_scsi_slave_config’: /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ata/libata-scsi.c:831: warning: implicit declaration of function ‘blk_queue_max_phys_segments’ make[3]: *** [drivers/ata/libata-scsi.o] Error 1 Bug report by Jesper Juhl. Signed-off-by: Adrian Bunk Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig index 3f4aa0c..03f6338 100644 --- a/drivers/ata/Kconfig +++ b/drivers/ata/Kconfig @@ -6,6 +6,7 @@ menu "Serial ATA (prod) and Parallel ATA (experimental) drivers" config ATA tristate "ATA device support" + depends on BLOCK depends on !(M32R || M68K) || BROKEN depends on !SUN4 || BROKEN select SCSI -- cgit v0.10.2 From 3343571d9f88a0de542d33aea9ab881f00ff866d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 19 Oct 2006 14:44:53 +0900 Subject: [PATCH] libata: typo fix Typo fix in commment. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik diff --git a/include/linux/libata.h b/include/linux/libata.h index d0a7ad5..b03d5a3 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -143,7 +143,7 @@ enum { ATA_DFLAG_CFG_MASK = (1 << 8) - 1, ATA_DFLAG_PIO = (1 << 8), /* device limited to PIO mode */ - ATA_DFLAG_NCQ_OFF = (1 << 9), /* devied limited to non-NCQ mode */ + ATA_DFLAG_NCQ_OFF = (1 << 9), /* device limited to non-NCQ mode */ ATA_DFLAG_SUSPENDED = (1 << 10), /* device suspended */ ATA_DFLAG_INIT_MASK = (1 << 16) - 1, -- cgit v0.10.2 From 12a87d36b3c5cb76a182c35f40d959a615d1c862 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:21:40 +0100 Subject: [PATCH] ahci: readability tweak Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 2592912..cef2e70 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -1041,7 +1041,7 @@ static void ahci_host_intr(struct ata_port *ap) /* hmmm... a spurious interupt */ /* some devices send D2H reg with I bit set during NCQ command phase */ - if (ap->sactive && status & PORT_IRQ_D2H_REG_FIS) + if (ap->sactive && (status & PORT_IRQ_D2H_REG_FIS)) return; /* ignore interim PIO setup fis interrupts */ -- cgit v0.10.2 From 8eb166bf805cc1c1d38d57211e8737631376b9ba Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:24:50 +0100 Subject: [PATCH] libata-sff: Allow for wacky systems There are some Linux supported platforms that simply cannot hit the low I/O addresses used by ATA legacy mode PCI mappings. These platforms have a window for PCI space that is fixed by the board logic and doesn't include the neccessary locations. Provide a config option so that such platforms faced with a controller that they cannot support simply error it and punt Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index 06daaa3..7645f2b 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -981,6 +981,15 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info, mask = (1 << 2) | (1 << 0); if ((tmp8 & mask) != mask) legacy_mode = (1 << 3); +#if defined(CONFIG_NO_ATA_LEGACY) + /* Some platforms with PCI limits cannot address compat + port space. In that case we punt if their firmware has + left a device in compatibility mode */ + if (legacy_mode) { + printk(KERN_ERR "ata: Compatibility mode ATA is not supported on this platform, skipping.\n"); + return -EOPNOTSUPP; + } +#endif } rc = pci_request_regions(pdev, DRV_NAME); -- cgit v0.10.2 From 86fbf1486a44a4bce4fdcbe3665a7d8a62ba958a Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Sat, 21 Oct 2006 10:24:01 -0700 Subject: [PATCH] Char: correct pci_get_device changes Commits 881a8c120acf7ec09c90289e2996b7c70f51e996 and efe1ec27837d6639eae82e1f5876910ba6433c3f corrects pci device matching in only one way; it no longer oopses/crashes, despite hotplug is not solved in these changes. Whenever pci_find_device -> pci_get_device change is performed, also pci_dev_get and pci_dev_put should be in most cases called to properly handle hotplug. This patch does exactly this thing -- increase refcount to let kernel know, that we are using this piece of HW just now. It affects moxa and rio char drivers. Cc: Acked-by: Amit Gud Acked-by: Greg Kroah-Hartman Acked-by: Alan Cox Signed-off-by: Jiri Slaby Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c index b401383..96cb1f0 100644 --- a/drivers/char/moxa.c +++ b/drivers/char/moxa.c @@ -130,6 +130,7 @@ static moxa_isa_board_conf moxa_isa_boards[] = typedef struct _moxa_pci_devinfo { ushort busNum; ushort devNum; + struct pci_dev *pdev; } moxa_pci_devinfo; typedef struct _moxa_board_conf { @@ -324,6 +325,9 @@ static int moxa_get_PCI_conf(struct pci_dev *p, int board_type, moxa_board_conf board->busType = MOXA_BUS_TYPE_PCI; board->pciInfo.busNum = p->bus->number; board->pciInfo.devNum = p->devfn >> 3; + board->pciInfo.pdev = p; + /* don't lose the reference in the next pci_get_device iteration */ + pci_dev_get(p); return (0); } @@ -493,6 +497,11 @@ static void __exit moxa_exit(void) if (tty_unregister_driver(moxaDriver)) printk("Couldn't unregister MOXA Intellio family serial driver\n"); put_tty_driver(moxaDriver); + + for (i = 0; i < MAX_BOARDS; i++) + if (moxa_boards[i].busType == MOXA_BUS_TYPE_PCI) + pci_dev_put(moxa_boards[i].pciInfo.pdev); + if (verbose) printk("Done\n"); } diff --git a/drivers/char/rio/host.h b/drivers/char/rio/host.h index ee2ddea..23d0681 100644 --- a/drivers/char/rio/host.h +++ b/drivers/char/rio/host.h @@ -44,6 +44,7 @@ ** the host. */ struct Host { + struct pci_dev *pdev; unsigned char Type; /* RIO_EISA, RIO_MCA, ... */ unsigned char Ivec; /* POLLED or ivec number */ unsigned char Mode; /* Control stuff */ diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c index c382df0..7ac68cb 100644 --- a/drivers/char/rio/rio_linux.c +++ b/drivers/char/rio/rio_linux.c @@ -1017,6 +1017,10 @@ static int __init rio_init(void) rio_dprintk(RIO_DEBUG_PROBE, "Hmm Tested ok, uniqid = %x.\n", p->RIOHosts[p->RIONumHosts].UniqueNum); fix_rio_pci(pdev); + + p->RIOHosts[p->RIONumHosts].pdev = pdev; + pci_dev_get(pdev); + p->RIOLastPCISearch = 0; p->RIONumHosts++; found++; @@ -1066,6 +1070,9 @@ static int __init rio_init(void) ((readb(&p->RIOHosts[p->RIONumHosts].Unique[1]) & 0xFF) << 8) | ((readb(&p->RIOHosts[p->RIONumHosts].Unique[2]) & 0xFF) << 16) | ((readb(&p->RIOHosts[p->RIONumHosts].Unique[3]) & 0xFF) << 24); rio_dprintk(RIO_DEBUG_PROBE, "Hmm Tested ok, uniqid = %x.\n", p->RIOHosts[p->RIONumHosts].UniqueNum); + p->RIOHosts[p->RIONumHosts].pdev = pdev; + pci_dev_get(pdev); + p->RIOLastPCISearch = 0; p->RIONumHosts++; found++; @@ -1181,6 +1188,8 @@ static void __exit rio_exit(void) } /* It is safe/allowed to del_timer a non-active timer */ del_timer(&hp->timer); + if (hp->Type == RIO_PCI) + pci_dev_put(hp->pdev); } if (misc_deregister(&rio_fw_device) < 0) { -- cgit v0.10.2 From 3c5473f80770768ab5712eb5a7492c16e97209fe Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sat, 21 Oct 2006 10:24:06 -0700 Subject: [PATCH] drivers/ide/pci/generic.c: re-add the __setup("all-generic-ide",...) The change from __setup() to module_param_named() requires users to prefix the option with "generic.". This patch re-adds the __setup() additionally to the module_param_named(). Usually it would make sense getting rid of such an obsolete __setup() at some time, but considering that drivers/ide/ is slowly approaching a RIP status it's already implicitely scheduled for removal. This patch fixes kernel Bugzilla #7353. Signed-off-by: Adrian Bunk Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c index 5b77a5b..ad418ce 100644 --- a/drivers/ide/pci/generic.c +++ b/drivers/ide/pci/generic.c @@ -40,6 +40,19 @@ static int ide_generic_all; /* Set to claim all devices */ +/* + * the module_param_named() was added for the modular case + * the __setup() is left as compatibility for existing setups + */ +#ifndef MODULE +static int __init ide_generic_all_on(char *unused) +{ + ide_generic_all = 1; + printk(KERN_INFO "IDE generic will claim all unknown PCI IDE storage controllers."); + return 1; +} +__setup("all-generic-ide", ide_generic_all_on); +#endif module_param_named(all_generic_ide, ide_generic_all, bool, 0444); MODULE_PARM_DESC(all_generic_ide, "IDE generic will claim all unknown PCI IDE storage controllers."); -- cgit v0.10.2 From 2e333e89860431d22816c1bdaa2ea72c2753396e Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:07 -0700 Subject: [PATCH] md: fix calculation of ->degraded for multipath and raid10 Two less-used md personalities have bugs in the calculation of ->degraded (the extent to which the array is degraded). Signed-off-by: Neil Brown Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index 171ff41..a6260f0 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -501,7 +501,7 @@ static int multipath_run (mddev_t *mddev) mdname(mddev)); goto out_free_conf; } - mddev->degraded = conf->raid_disks = conf->working_disks; + mddev->degraded = conf->raid_disks - conf->working_disks; conf->pool = mempool_create_kzalloc_pool(NR_RESERVED_BUFS, sizeof(struct multipath_bh)); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 1250f0e..74f17a9 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -2079,7 +2079,7 @@ static int run(mddev_t *mddev) disk = conf->mirrors + i; if (!disk->rdev || - !test_bit(In_sync, &rdev->flags)) { + !test_bit(In_sync, &disk->rdev->flags)) { disk->head_position = 0; mddev->degraded++; } -- cgit v0.10.2 From da3ed32fe568148ede256975d40825ffcdac767b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:08 -0700 Subject: [PATCH] md: add another COMPAT_IOCTL for md .. so that you can use bitmaps with 32bit userspace on a 64 bit kernel. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h index cfdb4f6a..c26c3ad 100644 --- a/include/linux/compat_ioctl.h +++ b/include/linux/compat_ioctl.h @@ -131,6 +131,7 @@ COMPATIBLE_IOCTL(RUN_ARRAY) COMPATIBLE_IOCTL(STOP_ARRAY) COMPATIBLE_IOCTL(STOP_ARRAY_RO) COMPATIBLE_IOCTL(RESTART_ARRAY_RW) +COMPATIBLE_IOCTL(GET_BITMAP_FILE) ULONG_IOCTL(SET_BITMAP_FILE) /* DM */ COMPATIBLE_IOCTL(DM_VERSION_32) -- cgit v0.10.2 From 1c05b4bc22cd640d3a534bd2851a8413d5df3709 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:08 -0700 Subject: [PATCH] md: endian annotation for v1 superblock access Includes a couple of bugfixes found by sparse. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index f7f1908..7daa7b1 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -974,12 +974,13 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev) * version 1 superblock */ -static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb) +static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb) { - unsigned int disk_csum, csum; + __le32 disk_csum; + u32 csum; unsigned long long newcsum; int size = 256 + le32_to_cpu(sb->max_dev)*2; - unsigned int *isuper = (unsigned int*)sb; + __le32 *isuper = (__le32*)sb; int i; disk_csum = sb->sb_csum; @@ -989,7 +990,7 @@ static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb) newcsum += le32_to_cpu(*isuper++); if (size == 2) - newcsum += le16_to_cpu(*(unsigned short*) isuper); + newcsum += le16_to_cpu(*(__le16*) isuper); csum = (newcsum & 0xffffffff) + (newcsum >> 32); sb->sb_csum = disk_csum; @@ -1106,7 +1107,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version) if (le32_to_cpu(sb->chunksize)) rdev->size &= ~((sector_t)le32_to_cpu(sb->chunksize)/2 - 1); - if (le32_to_cpu(sb->size) > rdev->size*2) + if (le64_to_cpu(sb->size) > rdev->size*2) return -EINVAL; return ret; } @@ -1228,7 +1229,7 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) else sb->resync_offset = cpu_to_le64(0); - sb->cnt_corrected_read = atomic_read(&rdev->corrected_errors); + sb->cnt_corrected_read = cpu_to_le32(atomic_read(&rdev->corrected_errors)); sb->raid_disks = cpu_to_le32(mddev->raid_disks); sb->size = cpu_to_le64(mddev->size<<1); diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h index b6ebc69..3f2cd98 100644 --- a/include/linux/raid/md_p.h +++ b/include/linux/raid/md_p.h @@ -206,52 +206,52 @@ static inline __u64 md_event(mdp_super_t *sb) { */ struct mdp_superblock_1 { /* constant array information - 128 bytes */ - __u32 magic; /* MD_SB_MAGIC: 0xa92b4efc - little endian */ - __u32 major_version; /* 1 */ - __u32 feature_map; /* bit 0 set if 'bitmap_offset' is meaningful */ - __u32 pad0; /* always set to 0 when writing */ + __le32 magic; /* MD_SB_MAGIC: 0xa92b4efc - little endian */ + __le32 major_version; /* 1 */ + __le32 feature_map; /* bit 0 set if 'bitmap_offset' is meaningful */ + __le32 pad0; /* always set to 0 when writing */ __u8 set_uuid[16]; /* user-space generated. */ char set_name[32]; /* set and interpreted by user-space */ - __u64 ctime; /* lo 40 bits are seconds, top 24 are microseconds or 0*/ - __u32 level; /* -4 (multipath), -1 (linear), 0,1,4,5 */ - __u32 layout; /* only for raid5 and raid10 currently */ - __u64 size; /* used size of component devices, in 512byte sectors */ + __le64 ctime; /* lo 40 bits are seconds, top 24 are microseconds or 0*/ + __le32 level; /* -4 (multipath), -1 (linear), 0,1,4,5 */ + __le32 layout; /* only for raid5 and raid10 currently */ + __le64 size; /* used size of component devices, in 512byte sectors */ - __u32 chunksize; /* in 512byte sectors */ - __u32 raid_disks; - __u32 bitmap_offset; /* sectors after start of superblock that bitmap starts + __le32 chunksize; /* in 512byte sectors */ + __le32 raid_disks; + __le32 bitmap_offset; /* sectors after start of superblock that bitmap starts * NOTE: signed, so bitmap can be before superblock * only meaningful of feature_map[0] is set. */ /* These are only valid with feature bit '4' */ - __u32 new_level; /* new level we are reshaping to */ - __u64 reshape_position; /* next address in array-space for reshape */ - __u32 delta_disks; /* change in number of raid_disks */ - __u32 new_layout; /* new layout */ - __u32 new_chunk; /* new chunk size (bytes) */ + __le32 new_level; /* new level we are reshaping to */ + __le64 reshape_position; /* next address in array-space for reshape */ + __le32 delta_disks; /* change in number of raid_disks */ + __le32 new_layout; /* new layout */ + __le32 new_chunk; /* new chunk size (bytes) */ __u8 pad1[128-124]; /* set to 0 when written */ /* constant this-device information - 64 bytes */ - __u64 data_offset; /* sector start of data, often 0 */ - __u64 data_size; /* sectors in this device that can be used for data */ - __u64 super_offset; /* sector start of this superblock */ - __u64 recovery_offset;/* sectors before this offset (from data_offset) have been recovered */ - __u32 dev_number; /* permanent identifier of this device - not role in raid */ - __u32 cnt_corrected_read; /* number of read errors that were corrected by re-writing */ + __le64 data_offset; /* sector start of data, often 0 */ + __le64 data_size; /* sectors in this device that can be used for data */ + __le64 super_offset; /* sector start of this superblock */ + __le64 recovery_offset;/* sectors before this offset (from data_offset) have been recovered */ + __le32 dev_number; /* permanent identifier of this device - not role in raid */ + __le32 cnt_corrected_read; /* number of read errors that were corrected by re-writing */ __u8 device_uuid[16]; /* user-space setable, ignored by kernel */ __u8 devflags; /* per-device flags. Only one defined...*/ #define WriteMostly1 1 /* mask for writemostly flag in above */ __u8 pad2[64-57]; /* set to 0 when writing */ /* array state information - 64 bytes */ - __u64 utime; /* 40 bits second, 24 btes microseconds */ - __u64 events; /* incremented when superblock updated */ - __u64 resync_offset; /* data before this offset (from data_offset) known to be in sync */ - __u32 sb_csum; /* checksum upto devs[max_dev] */ - __u32 max_dev; /* size of devs[] array to consider */ + __le64 utime; /* 40 bits second, 24 btes microseconds */ + __le64 events; /* incremented when superblock updated */ + __le64 resync_offset; /* data before this offset (from data_offset) known to be in sync */ + __le32 sb_csum; /* checksum upto devs[max_dev] */ + __le32 max_dev; /* size of devs[] array to consider */ __u8 pad3[64-32]; /* set to 0 when writing */ /* device state information. Indexed by dev_number. @@ -260,7 +260,7 @@ struct mdp_superblock_1 { * into the 'roles' value. If a device is spare or faulty, then it doesn't * have a meaningful role. */ - __u16 dev_roles[0]; /* role in array, or 0xffff for a spare, or 0xfffe for faulty */ + __le16 dev_roles[0]; /* role in array, or 0xffff for a spare, or 0xfffe for faulty */ }; /* feature_map bits */ -- cgit v0.10.2 From 4f2e639af4bd5e152fc79256e333643d3dd6c10f Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:09 -0700 Subject: [PATCH] md: endian annotations for the bitmap superblock And a couple of bug fixes found by sparse. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index d47d38a..d6f6147 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -536,7 +536,7 @@ static int bitmap_read_sb(struct bitmap *bitmap) printk(KERN_INFO "%s: bitmap file is out of date (%llu < %llu) " "-- forcing full recovery\n", bmname(bitmap), events, (unsigned long long) bitmap->mddev->events); - sb->state |= BITMAP_STALE; + sb->state |= cpu_to_le32(BITMAP_STALE); } success: /* assign fields using values from superblock */ @@ -544,11 +544,11 @@ success: bitmap->daemon_sleep = daemon_sleep; bitmap->daemon_lastrun = jiffies; bitmap->max_write_behind = write_behind; - bitmap->flags |= sb->state; + bitmap->flags |= le32_to_cpu(sb->state); if (le32_to_cpu(sb->version) == BITMAP_MAJOR_HOSTENDIAN) bitmap->flags |= BITMAP_HOSTENDIAN; bitmap->events_cleared = le64_to_cpu(sb->events_cleared); - if (sb->state & BITMAP_STALE) + if (sb->state & cpu_to_le32(BITMAP_STALE)) bitmap->events_cleared = bitmap->mddev->events; err = 0; out: @@ -578,9 +578,9 @@ static void bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits, spin_unlock_irqrestore(&bitmap->lock, flags); sb = (bitmap_super_t *)kmap_atomic(bitmap->sb_page, KM_USER0); switch (op) { - case MASK_SET: sb->state |= bits; + case MASK_SET: sb->state |= cpu_to_le32(bits); break; - case MASK_UNSET: sb->state &= ~bits; + case MASK_UNSET: sb->state &= cpu_to_le32(~bits); break; default: BUG(); } diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h index 84d8877..ebd42a3 100644 --- a/include/linux/raid/bitmap.h +++ b/include/linux/raid/bitmap.h @@ -146,16 +146,16 @@ enum bitmap_state { /* the superblock at the front of the bitmap file -- little endian */ typedef struct bitmap_super_s { - __u32 magic; /* 0 BITMAP_MAGIC */ - __u32 version; /* 4 the bitmap major for now, could change... */ - __u8 uuid[16]; /* 8 128 bit uuid - must match md device uuid */ - __u64 events; /* 24 event counter for the bitmap (1)*/ - __u64 events_cleared;/*32 event counter when last bit cleared (2) */ - __u64 sync_size; /* 40 the size of the md device's sync range(3) */ - __u32 state; /* 48 bitmap state information */ - __u32 chunksize; /* 52 the bitmap chunk size in bytes */ - __u32 daemon_sleep; /* 56 seconds between disk flushes */ - __u32 write_behind; /* 60 number of outstanding write-behind writes */ + __le32 magic; /* 0 BITMAP_MAGIC */ + __le32 version; /* 4 the bitmap major for now, could change... */ + __u8 uuid[16]; /* 8 128 bit uuid - must match md device uuid */ + __le64 events; /* 24 event counter for the bitmap (1)*/ + __le64 events_cleared;/*32 event counter when last bit cleared (2) */ + __le64 sync_size; /* 40 the size of the md device's sync range(3) */ + __le32 state; /* 48 bitmap state information */ + __le32 chunksize; /* 52 the bitmap chunk size in bytes */ + __le32 daemon_sleep; /* 56 seconds between disk flushes */ + __le32 write_behind; /* 60 number of outstanding write-behind writes */ __u8 pad[256 - 64]; /* set to zero */ } bitmap_super_t; -- cgit v0.10.2 From 78f32668e64caea8f638b9133da7b97c5aec20d1 Mon Sep 17 00:00:00 2001 From: Daniel Walker Date: Sat, 21 Oct 2006 10:24:10 -0700 Subject: [PATCH] clocksource: acpi_pm: add another greylist chipset I have an acpi_pm that goes backwards, but it's not intel. I tested the verified read and my acpi_pm started to function properly. So I added it to the greylist. I'm assuming that's the right spot. I also added an unlikely() to the while, cause it seems appropriate. Signed-off-by: Daniel Walker Acked-by: John Stultz Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/clocksource/acpi_pm.c b/drivers/clocksource/acpi_pm.c index 7ad3be8..7fcb77a 100644 --- a/drivers/clocksource/acpi_pm.c +++ b/drivers/clocksource/acpi_pm.c @@ -54,8 +54,8 @@ static cycle_t acpi_pm_read_verified(void) v1 = read_pmtmr(); v2 = read_pmtmr(); v3 = read_pmtmr(); - } while ((v1 > v2 && v1 < v3) || (v2 > v3 && v2 < v1) - || (v3 > v1 && v3 < v2)); + } while (unlikely((v1 > v2 && v1 < v3) || (v2 > v3 && v2 < v1) + || (v3 > v1 && v3 < v2))); return (cycle_t)v2; } @@ -138,6 +138,8 @@ static void __devinit acpi_pm_check_graylist(struct pci_dev *dev) } DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_0, acpi_pm_check_graylist); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_LE, + acpi_pm_check_graylist); #endif -- cgit v0.10.2 From d42552c3ace1fa1f16ae02ce642f4c733cec40ca Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 21 Oct 2006 10:24:12 -0700 Subject: [PATCH] pci: declare pci_get_device_reverse() We seem to have lost the declaration of pci_get_device_reverse(), if we ever had one. Add a CONFIG_PCI=0 stub too. Acked-by: Alan Cox Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/pci.h b/include/linux/pci.h index 4689e2a..09be0f8 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -455,7 +455,11 @@ int pci_find_next_capability (struct pci_dev *dev, u8 pos, int cap); int pci_find_ext_capability (struct pci_dev *dev, int cap); struct pci_bus *pci_find_next_bus(const struct pci_bus *from); -struct pci_dev *pci_get_device (unsigned int vendor, unsigned int device, struct pci_dev *from); +struct pci_dev *pci_get_device(unsigned int vendor, unsigned int device, + struct pci_dev *from); +struct pci_dev *pci_get_device_reverse(unsigned int vendor, unsigned int device, + struct pci_dev *from); + struct pci_dev *pci_get_subsys (unsigned int vendor, unsigned int device, unsigned int ss_vendor, unsigned int ss_device, struct pci_dev *from); @@ -660,7 +664,12 @@ static inline struct pci_dev *pci_find_device(unsigned int vendor, unsigned int static inline struct pci_dev *pci_find_slot(unsigned int bus, unsigned int devfn) { return NULL; } -static inline struct pci_dev *pci_get_device (unsigned int vendor, unsigned int device, struct pci_dev *from) +static inline struct pci_dev *pci_get_device(unsigned int vendor, + unsigned int device, struct pci_dev *from) +{ return NULL; } + +static inline struct pci_dev *pci_get_device_reverse(unsigned int vendor, + unsigned int device, struct pci_dev *from) { return NULL; } static inline struct pci_dev *pci_get_subsys (unsigned int vendor, unsigned int device, -- cgit v0.10.2 From 047a66d4bb24aaf19f41d620f8f0534c2153cd0b Mon Sep 17 00:00:00 2001 From: David Gibson Date: Sat, 21 Oct 2006 10:24:13 -0700 Subject: [PATCH] ibmveth: Fix index increment calculation The recent commit 751ae21c6cd1493e3d0a4935b08fb298b9d89773 introduced a bug in the producer/consumer index calculation in the ibmveth driver - incautious use of the post-increment ++ operator resulted in an increment being immediately reverted. This patch corrects the logic. Without this patch, the driver oopses almost immediately after activation on at least some machines. Signed-off-by: David Gibson Acked-by: Santiago Leon Cc: Jeff Garzik Cc: Martin Schwidefsky Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c index 2802db2..44c9f99 100644 --- a/drivers/net/ibmveth.c +++ b/drivers/net/ibmveth.c @@ -212,8 +212,8 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc break; } - free_index = pool->consumer_index++ % pool->size; - pool->consumer_index = free_index; + free_index = pool->consumer_index; + pool->consumer_index = (pool->consumer_index + 1) % pool->size; index = pool->free_map[free_index]; ibmveth_assert(index != IBM_VETH_INVALID_MAP); @@ -329,8 +329,10 @@ static void ibmveth_remove_buffer_from_pool(struct ibmveth_adapter *adapter, u64 adapter->rx_buff_pool[pool].buff_size, DMA_FROM_DEVICE); - free_index = adapter->rx_buff_pool[pool].producer_index++ % adapter->rx_buff_pool[pool].size; - adapter->rx_buff_pool[pool].producer_index = free_index; + free_index = adapter->rx_buff_pool[pool].producer_index; + adapter->rx_buff_pool[pool].producer_index + = (adapter->rx_buff_pool[pool].producer_index + 1) + % adapter->rx_buff_pool[pool].size; adapter->rx_buff_pool[pool].free_map[free_index] = index; mb(); -- cgit v0.10.2 From 7516795739bd53175629b90fab0ad488d7a6a9f7 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Sat, 21 Oct 2006 10:24:14 -0700 Subject: [PATCH] Reintroduce NODES_SPAN_OTHER_NODES for powerpc Reintroduce NODES_SPAN_OTHER_NODES for powerpc Revert "[PATCH] Remove SPAN_OTHER_NODES config definition" This reverts commit f62859bb6871c5e4a8e591c60befc8caaf54db8c. Revert "[PATCH] mm: remove arch independent NODES_SPAN_OTHER_NODES" This reverts commit a94b3ab7eab4edcc9b2cb474b188f774c331adf7. Also update the comments to indicate that this is still required and where its used. Signed-off-by: Andy Whitcroft Cc: Paul Mackerras Cc: Mike Kravetz Cc: Benjamin Herrenschmidt Acked-by: Mel Gorman Acked-by: Will Schmidt Cc: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 8b69104..2bd9b7f 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -751,6 +751,15 @@ config ARCH_MEMORY_PROBE def_bool y depends on MEMORY_HOTPLUG +# Some NUMA nodes have memory ranges that span +# other nodes. Even though a pfn is valid and +# between a node's start and end pfns, it may not +# reside on that node. See memmap_init_zone() +# for details. +config NODES_SPAN_OTHER_NODES + def_bool y + depends on NEED_MULTIPLE_NODES + config PPC_64K_PAGES bool "64k page size" depends on PPC64 diff --git a/arch/powerpc/configs/pseries_defconfig b/arch/powerpc/configs/pseries_defconfig index 9828663..d2833c1 100644 --- a/arch/powerpc/configs/pseries_defconfig +++ b/arch/powerpc/configs/pseries_defconfig @@ -184,6 +184,7 @@ CONFIG_SPLIT_PTLOCK_CPUS=4 CONFIG_MIGRATION=y CONFIG_RESOURCES_64BIT=y CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y +CONFIG_NODES_SPAN_OTHER_NODES=y # CONFIG_PPC_64K_PAGES is not set CONFIG_SCHED_SMT=y CONFIG_PROC_DEVICETREE=y diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 59855b8..ed0762b 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -674,6 +674,12 @@ void sparse_init(void); #define sparse_index_init(_sec, _nid) do {} while (0) #endif /* CONFIG_SPARSEMEM */ +#ifdef CONFIG_NODES_SPAN_OTHER_NODES +#define early_pfn_in_nid(pfn, nid) (early_pfn_to_nid(pfn) == (nid)) +#else +#define early_pfn_in_nid(pfn, nid) (1) +#endif + #ifndef early_pfn_valid #define early_pfn_valid(pfn) (1) #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ebd425c..f5fc454 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1689,6 +1689,8 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, for (pfn = start_pfn; pfn < end_pfn; pfn++) { if (!early_pfn_valid(pfn)) continue; + if (!early_pfn_in_nid(pfn, nid)) + continue; page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); init_page_count(page); -- cgit v0.10.2 From aedb0eb107961a234f7c38e53da65a8f7ea992a9 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Sat, 21 Oct 2006 10:24:16 -0700 Subject: [PATCH] Slab: Do not fallback to nodes that have not been bootstrapped yet The zonelist may contain zones of nodes that have not been bootstrapped and we will oops if we try to allocate from those zones. So check if the node information for the slab and the node have been setup before attempting an allocation. If it has not been setup then skip that zone. Usually we will not encounter this situation since the slab bootstrap code avoids falling back before we have setup the respective nodes but we seem to have a special needs for pppc. Signed-off-by: Christoph Lameter Acked-by: Andy Whitcroft Cc: Paul Mackerras Cc: Mike Kravetz Cc: Benjamin Herrenschmidt Acked-by: Mel Gorman Acked-by: Will Schmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/slab.c b/mm/slab.c index 266449d..84c631f 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -3152,12 +3152,15 @@ void *fallback_alloc(struct kmem_cache *cache, gfp_t flags) struct zone **z; void *obj = NULL; - for (z = zonelist->zones; *z && !obj; z++) + for (z = zonelist->zones; *z && !obj; z++) { + int nid = zone_to_nid(*z); + if (zone_idx(*z) <= ZONE_NORMAL && - cpuset_zone_allowed(*z, flags)) + cpuset_zone_allowed(*z, flags) && + cache->nodelists[nid]) obj = __cache_alloc_node(cache, - flags | __GFP_THISNODE, - zone_to_nid(*z)); + flags | __GFP_THISNODE, nid); + } return obj; } -- cgit v0.10.2 From faf6bbcf94caee10ba34adb86db4ecca96bfd3bf Mon Sep 17 00:00:00 2001 From: Paul Jackson Date: Sat, 21 Oct 2006 10:24:17 -0700 Subject: [PATCH] cpuset: mempolicy migration typo fix Mistyped an ifdef CONFIG_CPUSETS - fixed. I doubt that anyone ever noticed. The impact of this typo was that if someone: 1) was using MPOL_BIND to force off node allocations 2) while using cpusets to constrain memory placement 3) when that cpuset was migrating that jobs memory 4) while the tasks in that job were actively forking then there was a rare chance that future allocations using that MPOL_BIND policy would be node local, not off node. Signed-off-by: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h index 09f0f57..daabb3a 100644 --- a/include/linux/mempolicy.h +++ b/include/linux/mempolicy.h @@ -150,7 +150,7 @@ extern void mpol_rebind_mm(struct mm_struct *mm, nodemask_t *new); extern void mpol_fix_fork_child_flag(struct task_struct *p); #define set_cpuset_being_rebound(x) (cpuset_being_rebound = (x)) -#ifdef CONFIG_CPUSET +#ifdef CONFIG_CPUSETS #define current_cpuset_is_being_rebound() \ (cpuset_being_rebound == current->cpuset) #else -- cgit v0.10.2 From 8a7822a61ca9c22f464c0b79f455e62cccee747e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 21 Oct 2006 10:24:17 -0700 Subject: [PATCH] i2o/exec-osm.c: use "unsigned long flags;" Just like everyone else. Signed-off-by: Alexey Dobriyan Cc: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 91f95d1..01a5a70 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -127,7 +127,7 @@ int i2o_msg_post_wait_mem(struct i2o_controller *c, struct i2o_message *msg, DECLARE_WAIT_QUEUE_HEAD(wq); struct i2o_exec_wait *wait; static u32 tcntxt = 0x80000000; - long flags; + unsigned long flags; int rc = 0; wait = i2o_exec_wait_alloc(); -- cgit v0.10.2 From 3f7705eab6722ad1a346d748c4aad55755d6c241 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Sat, 21 Oct 2006 10:24:19 -0700 Subject: [PATCH] cciss: Fix warnings (and bug on 1TB discs) CCISS was producing warnings about shifts being greater than the size of the type and pointers being of incompatible type. Turns out this is because it's calling do_div on a 32-bit quantity. Upon further investigation, the sector_t total_size is being assigned to an int, and then we're calling do_div on that int. Obviously, sector_div is called for here, and I took the chance to refactor the code a little. Signed-off-by: Matthew Wilcox Acked-by: Mike Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index dcccaf2..bc66026 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -1923,7 +1923,6 @@ static void cciss_geometry_inquiry(int ctlr, int logvol, { int return_code; unsigned long t; - unsigned long rem; memset(inq_buff, 0, sizeof(InquiryData_struct)); if (withirq) @@ -1939,26 +1938,23 @@ static void cciss_geometry_inquiry(int ctlr, int logvol, printk(KERN_WARNING "cciss: reading geometry failed, volume " "does not support reading geometry\n"); - drv->block_size = block_size; - drv->nr_blocks = total_size; drv->heads = 255; drv->sectors = 32; // Sectors per track - t = drv->heads * drv->sectors; - drv->cylinders = total_size; - rem = do_div(drv->cylinders, t); } else { - drv->block_size = block_size; - drv->nr_blocks = total_size; drv->heads = inq_buff->data_byte[6]; drv->sectors = inq_buff->data_byte[7]; drv->cylinders = (inq_buff->data_byte[4] & 0xff) << 8; drv->cylinders += inq_buff->data_byte[5]; drv->raid_level = inq_buff->data_byte[8]; - t = drv->heads * drv->sectors; - if (t > 1) { - drv->cylinders = total_size; - rem = do_div(drv->cylinders, t); - } + } + drv->block_size = block_size; + drv->nr_blocks = total_size; + t = drv->heads * drv->sectors; + if (t > 1) { + unsigned rem = sector_div(total_size, t); + if (rem) + total_size++; + drv->cylinders = total_size; } } else { /* Get geometry failed */ printk(KERN_WARNING "cciss: reading geometry failed\n"); -- cgit v0.10.2 From 9eaef27b36a6b716384948da94b8fc5bfba7b712 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 21 Oct 2006 10:24:20 -0700 Subject: [PATCH] VFS: Make d_materialise_unique() enforce directory uniqueness If the caller tries to instantiate a directory using an inode that already has a dentry alias, then we attempt to rename the existing dentry instead of instantiating a new one. Fail with an ELOOP error if the rename would affect one of our parent directories. This behaviour is needed in order to avoid issues such as http://bugzilla.kernel.org/show_bug.cgi?id=7178 Signed-off-by: Trond Myklebust Cc: Miklos Szeredi Cc: Maneesh Soni Cc: Dipankar Sarma Cc: Neil Brown Cc: Al Viro Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/dcache.c b/fs/dcache.c index 2bac4ba..a1ff91e 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1469,23 +1469,21 @@ static void switch_names(struct dentry *dentry, struct dentry *target) * deleted it. */ -/** - * d_move - move a dentry +/* + * d_move_locked - move a dentry * @dentry: entry to move * @target: new dentry * * Update the dcache to reflect the move of a file name. Negative * dcache entries should not be moved in this way. */ - -void d_move(struct dentry * dentry, struct dentry * target) +static void d_move_locked(struct dentry * dentry, struct dentry * target) { struct hlist_head *list; if (!dentry->d_inode) printk(KERN_WARNING "VFS: moving negative dcache entry\n"); - spin_lock(&dcache_lock); write_seqlock(&rename_lock); /* * XXXX: do we really need to take target->d_lock? @@ -1536,10 +1534,84 @@ already_unhashed: fsnotify_d_move(dentry); spin_unlock(&dentry->d_lock); write_sequnlock(&rename_lock); +} + +/** + * d_move - move a dentry + * @dentry: entry to move + * @target: new dentry + * + * Update the dcache to reflect the move of a file name. Negative + * dcache entries should not be moved in this way. + */ + +void d_move(struct dentry * dentry, struct dentry * target) +{ + spin_lock(&dcache_lock); + d_move_locked(dentry, target); spin_unlock(&dcache_lock); } /* + * Helper that returns 1 if p1 is a parent of p2, else 0 + */ +static int d_isparent(struct dentry *p1, struct dentry *p2) +{ + struct dentry *p; + + for (p = p2; p->d_parent != p; p = p->d_parent) { + if (p->d_parent == p1) + return 1; + } + return 0; +} + +/* + * This helper attempts to cope with remotely renamed directories + * + * It assumes that the caller is already holding + * dentry->d_parent->d_inode->i_mutex and the dcache_lock + * + * Note: If ever the locking in lock_rename() changes, then please + * remember to update this too... + * + * On return, dcache_lock will have been unlocked. + */ +static struct dentry *__d_unalias(struct dentry *dentry, struct dentry *alias) +{ + struct mutex *m1 = NULL, *m2 = NULL; + struct dentry *ret; + + /* If alias and dentry share a parent, then no extra locks required */ + if (alias->d_parent == dentry->d_parent) + goto out_unalias; + + /* Check for loops */ + ret = ERR_PTR(-ELOOP); + if (d_isparent(alias, dentry)) + goto out_err; + + /* See lock_rename() */ + ret = ERR_PTR(-EBUSY); + if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex)) + goto out_err; + m1 = &dentry->d_sb->s_vfs_rename_mutex; + if (!mutex_trylock(&alias->d_parent->d_inode->i_mutex)) + goto out_err; + m2 = &alias->d_parent->d_inode->i_mutex; +out_unalias: + d_move_locked(alias, dentry); + ret = alias; +out_err: + spin_unlock(&dcache_lock); + if (m2) + mutex_unlock(m2); + if (m1) + mutex_unlock(m1); + return ret; +} + +/* * Prepare an anonymous dentry for life in the superblock's dentry tree as a * named dentry in place of the dentry to be replaced. */ @@ -1581,7 +1653,7 @@ static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon) */ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) { - struct dentry *alias, *actual; + struct dentry *actual; BUG_ON(!d_unhashed(dentry)); @@ -1593,26 +1665,27 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) goto found_lock; } - /* See if a disconnected directory already exists as an anonymous root - * that we should splice into the tree instead */ - if (S_ISDIR(inode->i_mode) && (alias = __d_find_alias(inode, 1))) { - spin_lock(&alias->d_lock); - - /* Is this a mountpoint that we could splice into our tree? */ - if (IS_ROOT(alias)) - goto connect_mountpoint; - - if (alias->d_name.len == dentry->d_name.len && - alias->d_parent == dentry->d_parent && - memcmp(alias->d_name.name, - dentry->d_name.name, - dentry->d_name.len) == 0) - goto replace_with_alias; - - spin_unlock(&alias->d_lock); - - /* Doh! Seem to be aliasing directories for some reason... */ - dput(alias); + if (S_ISDIR(inode->i_mode)) { + struct dentry *alias; + + /* Does an aliased dentry already exist? */ + alias = __d_find_alias(inode, 0); + if (alias) { + actual = alias; + /* Is this an anonymous mountpoint that we could splice + * into our tree? */ + if (IS_ROOT(alias)) { + spin_lock(&alias->d_lock); + __d_materialise_dentry(dentry, alias); + __d_drop(alias); + goto found; + } + /* Nope, but we must(!) avoid directory aliasing */ + actual = __d_unalias(dentry, alias); + if (IS_ERR(actual)) + dput(alias); + goto out_nolock; + } } /* Add a unique reference */ @@ -1628,7 +1701,7 @@ found: _d_rehash(actual); spin_unlock(&actual->d_lock); spin_unlock(&dcache_lock); - +out_nolock: if (actual == dentry) { security_d_instantiate(dentry, inode); return NULL; @@ -1637,16 +1710,6 @@ found: iput(inode); return actual; - /* Convert the anonymous/root alias into an ordinary dentry */ -connect_mountpoint: - __d_materialise_dentry(dentry, alias); - - /* Replace the candidate dentry with the alias in the tree */ -replace_with_alias: - __d_drop(alias); - actual = alias; - goto found; - shouldnt_be_hashed: spin_unlock(&dcache_lock); BUG(); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 4133ef5..27b5a10 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -935,8 +935,11 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru no_entry: res = d_materialise_unique(dentry, inode); - if (res != NULL) + if (res != NULL) { + if (IS_ERR(res)) + goto out_unlock; dentry = res; + } nfs_renew_times(dentry); nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); out_unlock: @@ -1132,6 +1135,8 @@ static struct dentry *nfs_readdir_lookup(nfs_readdir_descriptor_t *desc) alias = d_materialise_unique(dentry, inode); if (alias != NULL) { dput(dentry); + if (IS_ERR(alias)) + return NULL; dentry = alias; } -- cgit v0.10.2 From fc22617e451f23b466d4d63bb016f5f6111b69e4 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 21 Oct 2006 10:24:24 -0700 Subject: [PATCH] NFS: Cache invalidation fixup If someone has renamed a directory on the server, triggering the d_move code in d_materialise_unique(), then we need to invalidate the cached directory information in the source parent directory. Signed-off-by: Trond Myklebust Cc: Miklos Szeredi Cc: Maneesh Soni Cc: Dipankar Sarma Cc: Neil Brown Cc: Al Viro Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 27b5a10..b34cd16 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -936,8 +936,14 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru no_entry: res = d_materialise_unique(dentry, inode); if (res != NULL) { + struct dentry *parent; if (IS_ERR(res)) goto out_unlock; + /* Was a directory renamed! */ + parent = dget_parent(res); + if (!IS_ROOT(parent)) + nfs_mark_for_revalidate(parent->d_inode); + dput(parent); dentry = res; } nfs_renew_times(dentry); -- cgit v0.10.2 From 224dc50ece1b40f8cff5ecadd42a6b2691e231de Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 21 Oct 2006 02:05:20 +0100 Subject: [MIPS] Cleanup remaining references to mips_counter_frequency. Noticed by Samium Gromoff but his patch got stale in flight ... Signed-off-by: Ralf Baechle diff --git a/Documentation/mips/time.README b/Documentation/mips/time.README index 69ddc5c..e1304b6 100644 --- a/Documentation/mips/time.README +++ b/Documentation/mips/time.README @@ -63,7 +63,7 @@ the following functions or values: a) board_time_init - a function pointer. Invoked at the beginnig of time_init(). It is optional. 1. (optional) set up RTC routines - 2. (optional) calibrate and set the mips_counter_frequency + 2. (optional) calibrate and set the mips_hpt_frequency b) plat_timer_setup - a function pointer. Invoked at the end of time_init() 1. (optional) over-ride any decisions made in time_init() @@ -72,7 +72,7 @@ the following functions or values: c) (optional) board-specific RTC routines. - d) (optional) mips_counter_frequency - It must be definied if the board + d) (optional) mips_hpt_frequency - It must be definied if the board is using CPU counter for timer interrupt or it is using fixed rate gettimeoffset(). @@ -104,7 +104,7 @@ Step 1: decide how you like to implement the time services. or use an exnternal timer? In order to use CPU counter register as the timer interrupt source, you - must know the counter speed (mips_counter_frequency). It is usually the + must know the counter speed (mips_hpt_frequency). It is usually the same as the CPU speed or an integral divisor of it. d) decide on whether you want to use high-level or low-level timer @@ -121,8 +121,8 @@ Step 3: implement rtc routines, board_time_init() and plat_timer_setup() if needed. board_time_init() - - a) (optional) set up RTC routines, - b) (optional) calibrate and set the mips_counter_frequency + a) (optional) set up RTC routines, + b) (optional) calibrate and set the mips_hpt_frequency (only needed if you intended to use fixed_rate_gettimeoffset or use cpu counter as timer interrupt source) diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c index 6f8a9fe..c079e2a 100644 --- a/arch/mips/mips-boards/generic/time.c +++ b/arch/mips/mips-boards/generic/time.c @@ -187,7 +187,7 @@ out: } /* - * Estimate CPU frequency. Sets mips_counter_frequency as a side-effect + * Estimate CPU frequency. Sets mips_hpt_frequency as a side-effect */ static unsigned int __init estimate_cpu_frequency(void) { diff --git a/arch/mips/mips-boards/sim/sim_time.c b/arch/mips/mips-boards/sim/sim_time.c index c566b9b..24a4ed0 100644 --- a/arch/mips/mips-boards/sim/sim_time.c +++ b/arch/mips/mips-boards/sim/sim_time.c @@ -102,7 +102,7 @@ irqreturn_t sim_timer_interrupt(int irq, void *dev_id) /* - * Estimate CPU frequency. Sets mips_counter_frequency as a side-effect + * Estimate CPU frequency. Sets mips_hpt_frequency as a side-effect */ static unsigned int __init estimate_cpu_frequency(void) { -- cgit v0.10.2 From f8829caee311207afbc882794bdc5aa0db5caf33 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 21 Oct 2006 23:17:35 +0100 Subject: [MIPS] Fix aliasing bug in copy_to_user_page / copy_from_user_page The current implementation uses a sequence of a cacheflush and a copy. This is racy in case of a multithreaded debuggee and renders GDB virtually unusable. Aside this fixes a performance hog rendering access to /proc/cmdline very slow and resulting in a enough cache stalls for the 34K AP/SP programming model to make the bare metal code on the non-Linux VPE miss RT deadlines. The main part of this patch was originally written by Ralf Baechle; Atushi Nemoto did the the debugging. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 88b72c9..2de4d3c 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -30,11 +30,34 @@ #include #include #include +#include #include #include #include #include #include +#include + +/* Atomicity and interruptability */ +#ifdef CONFIG_MIPS_MT_SMTC + +#include + +#define ENTER_CRITICAL(flags) \ + { \ + unsigned int mvpflags; \ + local_irq_save(flags);\ + mvpflags = dvpe() +#define EXIT_CRITICAL(flags) \ + evpe(mvpflags); \ + local_irq_restore(flags); \ + } +#else + +#define ENTER_CRITICAL(flags) local_irq_save(flags) +#define EXIT_CRITICAL(flags) local_irq_restore(flags) + +#endif /* CONFIG_MIPS_MT_SMTC */ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); @@ -80,13 +103,142 @@ unsigned long setup_zero_pages(void) return 1UL << order; } -#ifdef CONFIG_HIGHMEM -pte_t *kmap_pte; -pgprot_t kmap_prot; +/* + * These are almost like kmap_atomic / kunmap_atmic except they take an + * additional address argument as the hint. + */ #define kmap_get_fixmap_pte(vaddr) \ pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)), (vaddr)), (vaddr)) +#ifdef CONFIG_MIPS_MT_SMTC +static pte_t *kmap_coherent_pte; +static void __init kmap_coherent_init(void) +{ + unsigned long vaddr; + + /* cache the first coherent kmap pte */ + vaddr = __fix_to_virt(FIX_CMAP_BEGIN); + kmap_coherent_pte = kmap_get_fixmap_pte(vaddr); +} +#else +static inline void kmap_coherent_init(void) {} +#endif + +static inline void *kmap_coherent(struct page *page, unsigned long addr) +{ + enum fixed_addresses idx; + unsigned long vaddr, flags, entrylo; + unsigned long old_ctx; + pte_t pte; + int tlbidx; + + inc_preempt_count(); + idx = (addr >> PAGE_SHIFT) & (FIX_N_COLOURS - 1); +#ifdef CONFIG_MIPS_MT_SMTC + idx += FIX_N_COLOURS * smp_processor_id(); +#endif + vaddr = __fix_to_virt(FIX_CMAP_END - idx); + pte = mk_pte(page, PAGE_KERNEL); +#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32_R1) + entrylo = pte.pte_high; +#else + entrylo = pte_val(pte) >> 6; +#endif + + ENTER_CRITICAL(flags); + old_ctx = read_c0_entryhi(); + write_c0_entryhi(vaddr & (PAGE_MASK << 1)); + write_c0_entrylo0(entrylo); + write_c0_entrylo1(entrylo); +#ifdef CONFIG_MIPS_MT_SMTC + set_pte(kmap_coherent_pte - (FIX_CMAP_END - idx), pte); + /* preload TLB instead of local_flush_tlb_one() */ + mtc0_tlbw_hazard(); + tlb_probe(); + tlb_probe_hazard(); + tlbidx = read_c0_index(); + mtc0_tlbw_hazard(); + if (tlbidx < 0) + tlb_write_random(); + else + tlb_write_indexed(); +#else + tlbidx = read_c0_wired(); + write_c0_wired(tlbidx + 1); + write_c0_index(tlbidx); + mtc0_tlbw_hazard(); + tlb_write_indexed(); +#endif + tlbw_use_hazard(); + write_c0_entryhi(old_ctx); + EXIT_CRITICAL(flags); + + return (void*) vaddr; +} + +#define UNIQUE_ENTRYHI(idx) (CKSEG0 + ((idx) << (PAGE_SHIFT + 1))) + +static inline void kunmap_coherent(struct page *page) +{ +#ifndef CONFIG_MIPS_MT_SMTC + unsigned int wired; + unsigned long flags, old_ctx; + + ENTER_CRITICAL(flags); + old_ctx = read_c0_entryhi(); + wired = read_c0_wired() - 1; + write_c0_wired(wired); + write_c0_index(wired); + write_c0_entryhi(UNIQUE_ENTRYHI(wired)); + write_c0_entrylo0(0); + write_c0_entrylo1(0); + mtc0_tlbw_hazard(); + tlb_write_indexed(); + tlbw_use_hazard(); + write_c0_entryhi(old_ctx); + EXIT_CRITICAL(flags); +#endif + dec_preempt_count(); + preempt_check_resched(); +} + +void copy_to_user_page(struct vm_area_struct *vma, + struct page *page, unsigned long vaddr, void *dst, const void *src, + unsigned long len) +{ + if (cpu_has_dc_aliases) { + void *vto = kmap_coherent(page, vaddr) + (vaddr & ~PAGE_MASK); + memcpy(vto, src, len); + kunmap_coherent(page); + } else + memcpy(dst, src, len); + if ((vma->vm_flags & VM_EXEC) && !cpu_has_ic_fills_f_dc) + flush_cache_page(vma, vaddr, page_to_pfn(page)); +} + +EXPORT_SYMBOL(copy_to_user_page); + +void copy_from_user_page(struct vm_area_struct *vma, + struct page *page, unsigned long vaddr, void *dst, const void *src, + unsigned long len) +{ + if (cpu_has_dc_aliases) { + void *vfrom = + kmap_coherent(page, vaddr) + (vaddr & ~PAGE_MASK); + memcpy(dst, vfrom, len); + kunmap_coherent(page); + } else + memcpy(dst, src, len); +} + +EXPORT_SYMBOL(copy_from_user_page); + + +#ifdef CONFIG_HIGHMEM +pte_t *kmap_pte; +pgprot_t kmap_prot; + static void __init kmap_init(void) { unsigned long kmap_vstart; @@ -97,11 +249,12 @@ static void __init kmap_init(void) kmap_prot = PAGE_KERNEL; } +#endif /* CONFIG_HIGHMEM */ -#ifdef CONFIG_32BIT void __init fixrange_init(unsigned long start, unsigned long end, pgd_t *pgd_base) { +#if defined(CONFIG_HIGHMEM) || defined(CONFIG_MIPS_MT_SMTC) pgd_t *pgd; pud_t *pud; pmd_t *pmd; @@ -122,7 +275,7 @@ void __init fixrange_init(unsigned long start, unsigned long end, for (; (k < PTRS_PER_PMD) && (vaddr != end); pmd++, k++) { if (pmd_none(*pmd)) { pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); - set_pmd(pmd, __pmd(pte)); + set_pmd(pmd, __pmd((unsigned long)pte)); if (pte != pte_offset_kernel(pmd, 0)) BUG(); } @@ -132,9 +285,8 @@ void __init fixrange_init(unsigned long start, unsigned long end, } j = 0; } +#endif } -#endif /* CONFIG_32BIT */ -#endif /* CONFIG_HIGHMEM */ #ifndef CONFIG_NEED_MULTIPLE_NODES extern void pagetable_init(void); @@ -175,6 +327,7 @@ void __init paging_init(void) #ifdef CONFIG_HIGHMEM kmap_init(); #endif + kmap_coherent_init(); max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT; low = max_low_pfn; diff --git a/arch/mips/mm/pgtable-32.c b/arch/mips/mm/pgtable-32.c index 4bdaa05..4a61e62 100644 --- a/arch/mips/mm/pgtable-32.c +++ b/arch/mips/mm/pgtable-32.c @@ -31,9 +31,10 @@ void pgd_init(unsigned long page) void __init pagetable_init(void) { -#ifdef CONFIG_HIGHMEM unsigned long vaddr; - pgd_t *pgd, *pgd_base; + pgd_t *pgd_base; +#ifdef CONFIG_HIGHMEM + pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t *pte; @@ -44,7 +45,6 @@ void __init pagetable_init(void) pgd_init((unsigned long)swapper_pg_dir + sizeof(pgd_t) * USER_PTRS_PER_PGD); -#ifdef CONFIG_HIGHMEM pgd_base = swapper_pg_dir; /* @@ -53,6 +53,7 @@ void __init pagetable_init(void) vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; fixrange_init(vaddr, 0, pgd_base); +#ifdef CONFIG_HIGHMEM /* * Permanent kmaps: */ diff --git a/arch/mips/mm/pgtable-64.c b/arch/mips/mm/pgtable-64.c index 44b5e97f..8d600d3 100644 --- a/arch/mips/mm/pgtable-64.c +++ b/arch/mips/mm/pgtable-64.c @@ -8,6 +8,7 @@ */ #include #include +#include #include void pgd_init(unsigned long page) @@ -52,7 +53,17 @@ void pmd_init(unsigned long addr, unsigned long pagetable) void __init pagetable_init(void) { + unsigned long vaddr; + pgd_t *pgd_base; + /* Initialize the entire pgd. */ pgd_init((unsigned long)swapper_pg_dir); pmd_init((unsigned long)invalid_pmd_table, (unsigned long)invalid_pte_table); + + pgd_base = swapper_pg_dir; + /* + * Fixed mappings: + */ + vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; + fixrange_init(vaddr, 0, pgd_base); } diff --git a/include/asm-mips/cacheflush.h b/include/asm-mips/cacheflush.h index 9ab59e2..e3c9925 100644 --- a/include/asm-mips/cacheflush.h +++ b/include/asm-mips/cacheflush.h @@ -55,24 +55,13 @@ extern void (*flush_icache_range)(unsigned long start, unsigned long end); #define flush_cache_vmap(start, end) flush_cache_all() #define flush_cache_vunmap(start, end) flush_cache_all() -static inline void copy_to_user_page(struct vm_area_struct *vma, +extern void copy_to_user_page(struct vm_area_struct *vma, struct page *page, unsigned long vaddr, void *dst, const void *src, - unsigned long len) -{ - if (cpu_has_dc_aliases) - flush_cache_page(vma, vaddr, page_to_pfn(page)); - memcpy(dst, src, len); - __flush_icache_page(vma, page); -} + unsigned long len); -static inline void copy_from_user_page(struct vm_area_struct *vma, +extern void copy_from_user_page(struct vm_area_struct *vma, struct page *page, unsigned long vaddr, void *dst, const void *src, - unsigned long len) -{ - if (cpu_has_dc_aliases) - flush_cache_page(vma, vaddr, page_to_pfn(page)); - memcpy(dst, src, len); -} + unsigned long len); extern void (*flush_cache_sigtramp)(unsigned long addr); extern void (*flush_icache_all)(void); diff --git a/include/asm-mips/fixmap.h b/include/asm-mips/fixmap.h index 6959bdb..02c8a13 100644 --- a/include/asm-mips/fixmap.h +++ b/include/asm-mips/fixmap.h @@ -45,8 +45,16 @@ * fix-mapped? */ enum fixed_addresses { +#define FIX_N_COLOURS 8 + FIX_CMAP_BEGIN, +#ifdef CONFIG_MIPS_MT_SMTC + FIX_CMAP_END = FIX_CMAP_BEGIN + (FIX_N_COLOURS * NR_CPUS), +#else + FIX_CMAP_END = FIX_CMAP_BEGIN + FIX_N_COLOURS, +#endif #ifdef CONFIG_HIGHMEM - FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */ + /* reserved pte's for temporary kernel mappings */ + FIX_KMAP_BEGIN = FIX_CMAP_END + 1, FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1, #endif __end_of_fixed_addresses @@ -70,9 +78,9 @@ extern void __set_fixmap (enum fixed_addresses idx, * at the top of mem.. */ #if defined(CONFIG_CPU_TX39XX) || defined(CONFIG_CPU_TX49XX) -#define FIXADDR_TOP (0xff000000UL - 0x2000) +#define FIXADDR_TOP ((unsigned long)(long)(int)(0xff000000 - 0x20000)) #else -#define FIXADDR_TOP (0xffffe000UL) +#define FIXADDR_TOP ((unsigned long)(long)(int)0xfffe0000) #endif #define FIXADDR_SIZE (__end_of_fixed_addresses << PAGE_SHIFT) #define FIXADDR_START (FIXADDR_TOP - FIXADDR_SIZE) -- cgit v0.10.2 From 37b1bccfe75691f5f42bf210d8a349f931896887 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sun, 22 Oct 2006 00:38:23 +0200 Subject: [PATCH] x86-64: Fix C3 timer test There was a typo in the C3 latency test to decide of the TSC should be used or not. It used the C2 latency threshold, not the C3 one. Fix that. This should fix the time on various dual core laptops. Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 1ba5a44..88722f1 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -948,7 +948,7 @@ __cpuinit int unsynchronized_tsc(void) if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) { #ifdef CONFIG_ACPI /* But TSC doesn't tick in C3 so don't use it there */ - if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 100) + if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 1000) return 1; #endif return 0; -- cgit v0.10.2 From b3edc9cec07ade41aaf1804f7c9e876afa90c862 Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Sun, 22 Oct 2006 00:38:23 +0200 Subject: [PATCH] x86-64: increase PHB1 split transaction timeout This patch increases the timeout for PCI split transactions on PHB1 on the first Calgary to work around an issue with the aic94xx adapter. Fixes kernel.org bugzilla #7180 (http://bugzilla.kernel.org/show_bug.cgi?id=7180) Based on excellent debugging and a patch by Darrick J. Wong Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen Acked-by: Darrick J. Wong diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index b3296cc..37a7708 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -52,7 +52,8 @@ #define ONE_BASED_CHASSIS_NUM 1 /* register offsets inside the host bridge space */ -#define PHB_CSR_OFFSET 0x0110 +#define CALGARY_CONFIG_REG 0x0108 +#define PHB_CSR_OFFSET 0x0110 /* Channel Status */ #define PHB_PLSSR_OFFSET 0x0120 #define PHB_CONFIG_RW_OFFSET 0x0160 #define PHB_IOBASE_BAR_LOW 0x0170 @@ -83,6 +84,8 @@ #define TAR_VALID 0x0000000000000008UL /* CSR (Channel/DMA Status Register) */ #define CSR_AGENT_MASK 0xffe0ffff +/* CCR (Calgary Configuration Register) */ +#define CCR_2SEC_TIMEOUT 0x000000000000000EUL #define MAX_NUM_OF_PHBS 8 /* how many PHBs in total? */ #define MAX_NUM_CHASSIS 8 /* max number of chassis */ @@ -732,6 +735,38 @@ static void calgary_watchdog(unsigned long data) } } +static void __init calgary_increase_split_completion_timeout(void __iomem *bbar, + unsigned char busnum) +{ + u64 val64; + void __iomem *target; + unsigned long phb_shift = -1; + u64 mask; + + switch (busno_to_phbid(busnum)) { + case 0: phb_shift = (63 - 19); + break; + case 1: phb_shift = (63 - 23); + break; + case 2: phb_shift = (63 - 27); + break; + case 3: phb_shift = (63 - 35); + break; + default: + BUG_ON(busno_to_phbid(busnum)); + } + + target = calgary_reg(bbar, CALGARY_CONFIG_REG); + val64 = be64_to_cpu(readq(target)); + + /* zero out this PHB's timer bits */ + mask = ~(0xFUL << phb_shift); + val64 &= mask; + val64 |= (CCR_2SEC_TIMEOUT << phb_shift); + writeq(cpu_to_be64(val64), target); + readq(target); /* flush */ +} + static void __init calgary_enable_translation(struct pci_dev *dev) { u32 val32; @@ -756,6 +791,13 @@ static void __init calgary_enable_translation(struct pci_dev *dev) writel(cpu_to_be32(val32), target); readl(target); /* flush */ + /* + * Give split completion a longer timeout on bus 1 for aic94xx + * http://bugzilla.kernel.org/show_bug.cgi?id=7180 + */ + if (busnum == 1) + calgary_increase_split_completion_timeout(bbar, busnum); + init_timer(&tbl->watchdog_timer); tbl->watchdog_timer.function = &calgary_watchdog; tbl->watchdog_timer.data = (unsigned long)dev; -- cgit v0.10.2 From aa026ede513b7d672fa7d9106b2f2a475455dcf2 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sun, 22 Oct 2006 00:41:15 +0200 Subject: [PATCH] x86-64: Fix C3 timer test There was a typo in the C3 latency test to decide of the TSC should be used or not. It used the C2 latency threshold, not the C3 one. Fix that. This should fix the time on various dual core laptops. Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 1ba5a44..88722f1 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -948,7 +948,7 @@ __cpuinit int unsynchronized_tsc(void) if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) { #ifdef CONFIG_ACPI /* But TSC doesn't tick in C3 so don't use it there */ - if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 100) + if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 1000) return 1; #endif return 0; -- cgit v0.10.2 From cb01fc720c629261b9c616b2d5fcc3d93cd8bb09 Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Sun, 22 Oct 2006 00:41:15 +0200 Subject: [PATCH] x86-64: increase PHB1 split transaction timeout This patch increases the timeout for PCI split transactions on PHB1 on the first Calgary to work around an issue with the aic94xx adapter. Fixes kernel.org bugzilla #7180 (http://bugzilla.kernel.org/show_bug.cgi?id=7180) Based on excellent debugging and a patch by Darrick J. Wong Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen Acked-by: Darrick J. Wong diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index b3296cc..37a7708 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -52,7 +52,8 @@ #define ONE_BASED_CHASSIS_NUM 1 /* register offsets inside the host bridge space */ -#define PHB_CSR_OFFSET 0x0110 +#define CALGARY_CONFIG_REG 0x0108 +#define PHB_CSR_OFFSET 0x0110 /* Channel Status */ #define PHB_PLSSR_OFFSET 0x0120 #define PHB_CONFIG_RW_OFFSET 0x0160 #define PHB_IOBASE_BAR_LOW 0x0170 @@ -83,6 +84,8 @@ #define TAR_VALID 0x0000000000000008UL /* CSR (Channel/DMA Status Register) */ #define CSR_AGENT_MASK 0xffe0ffff +/* CCR (Calgary Configuration Register) */ +#define CCR_2SEC_TIMEOUT 0x000000000000000EUL #define MAX_NUM_OF_PHBS 8 /* how many PHBs in total? */ #define MAX_NUM_CHASSIS 8 /* max number of chassis */ @@ -732,6 +735,38 @@ static void calgary_watchdog(unsigned long data) } } +static void __init calgary_increase_split_completion_timeout(void __iomem *bbar, + unsigned char busnum) +{ + u64 val64; + void __iomem *target; + unsigned long phb_shift = -1; + u64 mask; + + switch (busno_to_phbid(busnum)) { + case 0: phb_shift = (63 - 19); + break; + case 1: phb_shift = (63 - 23); + break; + case 2: phb_shift = (63 - 27); + break; + case 3: phb_shift = (63 - 35); + break; + default: + BUG_ON(busno_to_phbid(busnum)); + } + + target = calgary_reg(bbar, CALGARY_CONFIG_REG); + val64 = be64_to_cpu(readq(target)); + + /* zero out this PHB's timer bits */ + mask = ~(0xFUL << phb_shift); + val64 &= mask; + val64 |= (CCR_2SEC_TIMEOUT << phb_shift); + writeq(cpu_to_be64(val64), target); + readq(target); /* flush */ +} + static void __init calgary_enable_translation(struct pci_dev *dev) { u32 val32; @@ -756,6 +791,13 @@ static void __init calgary_enable_translation(struct pci_dev *dev) writel(cpu_to_be32(val32), target); readl(target); /* flush */ + /* + * Give split completion a longer timeout on bus 1 for aic94xx + * http://bugzilla.kernel.org/show_bug.cgi?id=7180 + */ + if (busnum == 1) + calgary_increase_split_completion_timeout(bbar, busnum); + init_timer(&tbl->watchdog_timer); tbl->watchdog_timer.function = &calgary_watchdog; tbl->watchdog_timer.data = (unsigned long)dev; -- cgit v0.10.2 From 82709531a800fcf8de71bb8c5d8e92462fb81f84 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Wed, 11 Oct 2006 16:26:54 +0100 Subject: [DCCP]: Fix Oops in DCCPv6 I think I got the cause for the Oops observed in http://www.mail-archive.com/dccp@vger.kernel.org/msg00578.html The problem is always with applications listening on PF_INET6 sockets. Apart from the mentioned oops, I observed another one one, triggered at irregular intervals via timer interrupt: run_timer_softirq -> dccp_keepalive_timer -> inet_csk_reqsk_queue_prune -> reqsk_free -> dccp_v6_reqsk_destructor The latter function is the problem and is also the last function to be called in said kernel panic. In any case, there is a real problem with allocating the right request_sock which is what this patch tackles. It fixes the following problem: - application listens on PF_INET6 - DCCPv4 packet comes in, is handed over to dccp_v4_do_rcv, from there to dccp_v4_conn_request Now: socket is PF_INET6, packet is IPv4. The following code then furnishes the connection with IPv6 - request_sock operations: req = reqsk_alloc(sk->sk_prot->rsk_prot); The first problem is that all further incoming packets will get a Reset since the connection can not be looked up. The second problem is worse: --> reqsk_alloc is called instead of inet6_reqsk_alloc --> consequently inet6_rsk_offset is never set (dangling pointer) --> the request_sock_ops are nevertheless still dccp6_request_ops --> destructor is called via reqsk_free --> dccp_v6_reqsk_destructor tries to free random memory location (inet6_rsk_offset not set) --> panic I have tested this for a while, DCCP sockets are now handled correctly in all three scenarios (v4/v6 only/v4-mapped). Commiter note: I've added the dccp_request_sock_ops forward declaration to keep the tree building and to reduce the size of the patch for 2.6.19, later I'll move the functions to the top of the affected source code to match what we have in the TCP counterpart, where this problem hasn't existed in the first place, dumb me not to have done the same thing on DCCP land 8) Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 7e746c4..aaaf4d0 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -449,6 +449,8 @@ static inline u64 dccp_v4_init_sequence(const struct sock *sk, dccp_hdr(skb)->dccph_sport); } +static struct request_sock_ops dccp_request_sock_ops; + int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { struct inet_request_sock *ireq; @@ -489,7 +491,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1) goto drop; - req = reqsk_alloc(sk->sk_prot->rsk_prot); + req = reqsk_alloc(&dccp_request_sock_ops); if (req == NULL) goto drop; diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 7171a78..91e7b12 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -672,7 +672,6 @@ static struct sock *dccp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) { - struct inet_request_sock *ireq; struct dccp_sock dp; struct request_sock *req; struct dccp_request_sock *dreq; @@ -701,7 +700,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1) goto drop; - req = inet6_reqsk_alloc(sk->sk_prot->rsk_prot); + req = inet6_reqsk_alloc(&dccp6_request_sock_ops); if (req == NULL) goto drop; @@ -713,7 +712,6 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) goto drop_and_free; ireq6 = inet6_rsk(req); - ireq = inet_rsk(req); ipv6_addr_copy(&ireq6->rmt_addr, &skb->nh.ipv6h->saddr); ipv6_addr_copy(&ireq6->loc_addr, &skb->nh.ipv6h->daddr); req->rcv_wnd = dccp_feat_default_sequence_window; -- cgit v0.10.2 From fd169f15a67b47f23bd1704919c719a8e8409a73 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 20 Oct 2006 19:44:17 -0700 Subject: [DCCP] ipv6: Fix opt_skb leak. Based upon a patch from Jesper Juhl. Try to match the TCP IPv6 code this was copied from as much as possible, so that it's easy to see where to add the ipv6 pktoptions support code. Signed-off-by: David S. Miller diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 91e7b12..c8bf89b 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -995,6 +995,10 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) if (sk->sk_state == DCCP_OPEN) { /* Fast path */ if (dccp_rcv_established(sk, skb, dccp_hdr(skb), skb->len)) goto reset; + if (opt_skb) { + /* This is where we would goto ipv6_pktoptions. */ + __kfree_skb(opt_skb); + } return 0; } @@ -1019,6 +1023,10 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) if (dccp_rcv_state_process(sk, skb, dccp_hdr(skb), skb->len)) goto reset; + if (opt_skb) { + /* This is where we would goto ipv6_pktoptions. */ + __kfree_skb(opt_skb); + } return 0; reset: -- cgit v0.10.2 From 97f80bc66f5c6384e3aab70c67340116b8c4284b Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Fri, 20 Oct 2006 19:48:42 -0700 Subject: [ATM]: handle sysfs errors Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/net/atm/atm_sysfs.c b/net/atm/atm_sysfs.c index c0a4ae2..62f6ed1 100644 --- a/net/atm/atm_sysfs.c +++ b/net/atm/atm_sysfs.c @@ -141,7 +141,7 @@ static struct class atm_class = { int atm_register_sysfs(struct atm_dev *adev) { struct class_device *cdev = &adev->class_dev; - int i, err; + int i, j, err; cdev->class = &atm_class; class_set_devdata(cdev, adev); @@ -151,10 +151,19 @@ int atm_register_sysfs(struct atm_dev *adev) if (err < 0) return err; - for (i = 0; atm_attrs[i]; i++) - class_device_create_file(cdev, atm_attrs[i]); + for (i = 0; atm_attrs[i]; i++) { + err = class_device_create_file(cdev, atm_attrs[i]); + if (err) + goto err_out; + } return 0; + +err_out: + for (j = 0; j < i; j++) + class_device_remove_file(cdev, atm_attrs[j]); + class_device_del(cdev); + return err; } void atm_unregister_sysfs(struct atm_dev *adev) -- cgit v0.10.2 From b45eccdb51c102e3c5ff9eaecc36200ab2eb09c0 Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Fri, 20 Oct 2006 19:49:45 -0700 Subject: [ATM]: No need to return void The module_exit function has return-type void and pci_unregister_driver() returns void anyway. Signed-off-by: Tobias Klauser Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/drivers/atm/ambassador.c b/drivers/atm/ambassador.c index 323592d..9fffa7af 100644 --- a/drivers/atm/ambassador.c +++ b/drivers/atm/ambassador.c @@ -2452,8 +2452,8 @@ static int __init amb_module_init (void) static void __exit amb_module_exit (void) { PRINTD (DBG_FLOW|DBG_INIT, "cleanup_module"); - - return pci_unregister_driver(&amb_driver); + + pci_unregister_driver(&amb_driver); } module_init(amb_module_init); diff --git a/drivers/atm/horizon.c b/drivers/atm/horizon.c index f593492..44268cb 100644 --- a/drivers/atm/horizon.c +++ b/drivers/atm/horizon.c @@ -2932,8 +2932,8 @@ static int __init hrz_module_init (void) { static void __exit hrz_module_exit (void) { PRINTD (DBG_FLOW, "cleanup_module"); - - return pci_unregister_driver(&hrz_driver); + + pci_unregister_driver(&hrz_driver); } module_init(hrz_module_init); -- cgit v0.10.2 From 663bab6fd097c18ae0c7a7fd1b4a44558b998cdb Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Fri, 20 Oct 2006 19:50:50 -0700 Subject: [ATM] firestream: handle thrown error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc emits the following warning: drivers/atm/firestream.c: In function ‘fs_open’: drivers/atm/firestream.c:870: warning: ‘tmc0’ may be used uninitialized in this function This indicates a real bug. We should check make_rate() return value for potential errors. Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/drivers/atm/firestream.c b/drivers/atm/firestream.c index 40ab9b6..697ad82 100644 --- a/drivers/atm/firestream.c +++ b/drivers/atm/firestream.c @@ -1002,6 +1002,10 @@ static int fs_open(struct atm_vcc *atm_vcc) r = ROUND_UP; } error = make_rate (pcr, r, &tmc0, NULL); + if (error) { + kfree(tc); + return error; + } } fs_dprintk (FS_DEBUG_OPEN, "pcr = %d.\n", pcr); } -- cgit v0.10.2 From 69c3014763966b0ae2bd190dac3654dd6cebdd45 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 20 Oct 2006 19:51:46 -0700 Subject: [ATM] nicstar: Fix a bogus casting warning Not enough to make Nicstar 64bit friendly but got squashed in passing so might as well be applied Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/drivers/atm/nicstar.c b/drivers/atm/nicstar.c index 632ede5..bd09045 100644 --- a/drivers/atm/nicstar.c +++ b/drivers/atm/nicstar.c @@ -2759,7 +2759,7 @@ static int ns_ioctl(struct atm_dev *dev, unsigned int cmd, void __user *arg) { ns_dev *card; pool_levels pl; - int btype; + long btype; unsigned long flags; card = dev->dev_data; @@ -2859,7 +2859,7 @@ static int ns_ioctl(struct atm_dev *dev, unsigned int cmd, void __user *arg) case NS_ADJBUFLEV: if (!capable(CAP_NET_ADMIN)) return -EPERM; - btype = (int) arg; /* an int is the same size as a pointer */ + btype = (long) arg; /* a long is the same size as a pointer or bigger */ switch (btype) { case NS_BUFTYPE_SMALL: -- cgit v0.10.2 From 375216ad0c303adeed45281ce82e153d41de679a Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 21 Oct 2006 20:20:54 -0700 Subject: [IPv6] fib: initialize tb6_lock in common place to give lockdep a key Signed-off-by: Thomas Graf Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 8fcae7a..f98ca30 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -169,7 +169,6 @@ static __inline__ void rt6_release(struct rt6_info *rt) static struct fib6_table fib6_main_tbl = { .tb6_id = RT6_TABLE_MAIN, - .tb6_lock = RW_LOCK_UNLOCKED, .tb6_root = { .leaf = &ip6_null_entry, .fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO, @@ -187,6 +186,12 @@ static void fib6_link_table(struct fib6_table *tb) { unsigned int h; + /* + * Initialize table lock at a single place to give lockdep a key, + * tables aren't visible prior to being linked to the list. + */ + rwlock_init(&tb->tb6_lock); + h = tb->tb6_id & (FIB_TABLE_HASHSZ - 1); /* @@ -199,7 +204,6 @@ static void fib6_link_table(struct fib6_table *tb) #ifdef CONFIG_IPV6_MULTIPLE_TABLES static struct fib6_table fib6_local_tbl = { .tb6_id = RT6_TABLE_LOCAL, - .tb6_lock = RW_LOCK_UNLOCKED, .tb6_root = { .leaf = &ip6_null_entry, .fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO, @@ -213,7 +217,6 @@ static struct fib6_table *fib6_alloc_table(u32 id) table = kzalloc(sizeof(*table), GFP_ATOMIC); if (table != NULL) { table->tb6_id = id; - table->tb6_lock = RW_LOCK_UNLOCKED; table->tb6_root.leaf = &ip6_null_entry; table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; } -- cgit v0.10.2 From 185b1aa122f87052d9154bb74990bc785372a750 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 21 Oct 2006 20:24:01 -0700 Subject: [NET]: Reduce sizeof(struct flowi) by 20 bytes. As suggested by David, just kill off some unused fields in dnports to reduce sizef(struct flowi). If they come back, they should be moved to nl_u.dn_u in order not to enlarge again struct flowi [ Modified to really delete this stuff instead of using #if 0. -DaveM ] Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/dn.h b/include/net/dn.h index 465b783..ac4ce90 100644 --- a/include/net/dn.h +++ b/include/net/dn.h @@ -199,11 +199,6 @@ static inline void dn_sk_ports_copy(struct flowi *fl, struct dn_scp *scp) { fl->uli_u.dnports.sport = scp->addrloc; fl->uli_u.dnports.dport = scp->addrrem; - fl->uli_u.dnports.objnum = scp->addr.sdn_objnum; - if (fl->uli_u.dnports.objnum == 0) { - fl->uli_u.dnports.objnamel = (__u8)dn_ntohs(scp->addr.sdn_objnamel); - memcpy(fl->uli_u.dnports.objname, scp->addr.sdn_objname, 16); - } } extern unsigned dn_mss_from_pmtu(struct net_device *dev, int mtu); diff --git a/include/net/flow.h b/include/net/flow.h index 3b44d72..5cda27c 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -68,9 +68,6 @@ struct flowi { struct { __le16 sport; __le16 dport; - __u8 objnum; - __u8 objnamel; /* Not 16 bits since max val is 16 */ - __u8 objname[16]; /* Not zero terminated */ } dnports; __be32 spi; -- cgit v0.10.2 From 4a81a21b6095997f50ac9ca37e4fdfe48d789ad7 Mon Sep 17 00:00:00 2001 From: Amit Choudhary Date: Mon, 9 Oct 2006 16:02:49 +0200 Subject: [ALSA] sound/isa/gus/interwave.c: check kmalloc() return value Check the return value of kmalloc() in function snd_interwave_pnp(), in file sound/isa/gus/interwave.c. Signed-off-by: Amit Choudhary Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/isa/gus/interwave.c b/sound/isa/gus/interwave.c index f12cd09..4ec2d79 100644 --- a/sound/isa/gus/interwave.c +++ b/sound/isa/gus/interwave.c @@ -564,6 +564,8 @@ static int __devinit snd_interwave_pnp(int dev, struct snd_interwave *iwcard, struct pnp_resource_table * cfg = kmalloc(sizeof(struct pnp_resource_table), GFP_KERNEL); int err; + if (!cfg) + return -ENOMEM; iwcard->dev = pnp_request_card_device(card, id->devs[0].id, NULL); if (iwcard->dev == NULL) { kfree(cfg); -- cgit v0.10.2 From fbdbb2205a7ff3c3e19941477cde6f9a693637ef Mon Sep 17 00:00:00 2001 From: Amit Choudhary Date: Mon, 9 Oct 2006 16:03:23 +0200 Subject: [ALSA] sound/isa/cmi8330.c: check kmalloc() return value Check the return value of kmalloc() in function snd_cmi8330_pnp(), in file sound/isa/cmi8330.c. Signed-off-by: Amit Choudhary Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/isa/cmi8330.c b/sound/isa/cmi8330.c index 3c1e9fd..d1f6dfc 100644 --- a/sound/isa/cmi8330.c +++ b/sound/isa/cmi8330.c @@ -289,6 +289,8 @@ static int __devinit snd_cmi8330_pnp(int dev, struct snd_cmi8330 *acard, struct pnp_resource_table * cfg = kmalloc(sizeof(struct pnp_resource_table), GFP_KERNEL); int err; + if (!cfg) + return -ENOMEM; acard->cap = pnp_request_card_device(card, id->devs[0].id, NULL); if (acard->cap == NULL) { kfree(cfg); -- cgit v0.10.2 From b1e8a791c77bd35f88b678ce761928583efd8cd0 Mon Sep 17 00:00:00 2001 From: Amit Choudhary Date: Mon, 9 Oct 2006 16:03:52 +0200 Subject: [ALSA] sound/isa/ad1816a/ad1816a.c: check kmalloc() return value Check the return value of kmalloc() in function snd_card_ad1816a_pnp(), in file sound/isa/ad1816a/ad1816a.c. Signed-off-by: Amit Choudhary Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/isa/ad1816a/ad1816a.c b/sound/isa/ad1816a/ad1816a.c index b33a5fb..5903450 100644 --- a/sound/isa/ad1816a/ad1816a.c +++ b/sound/isa/ad1816a/ad1816a.c @@ -120,6 +120,8 @@ static int __devinit snd_card_ad1816a_pnp(int dev, struct snd_card_ad1816a *acar struct pnp_resource_table *cfg = kmalloc(sizeof(*cfg), GFP_KERNEL); int err; + if (!cfg) + return -ENOMEM; acard->dev = pnp_request_card_device(card, id->devs[0].id, NULL); if (acard->dev == NULL) { kfree(cfg); -- cgit v0.10.2 From 5019f75ea1a98c36e9139ffb2bf8614a2e9f0a03 Mon Sep 17 00:00:00 2001 From: Amit Choudhary Date: Mon, 9 Oct 2006 16:04:34 +0200 Subject: [ALSA] sound/isa/opti9xx/opti92x-ad1848.c: check kmalloc() return value Check the return value of kmalloc() in function snd_card_opti9xx_pnp(), in file sound/isa/opti9xx/opti92x-ad1848.c. Signed-off-by: Amit Choudhary Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/isa/opti9xx/opti92x-ad1848.c b/sound/isa/opti9xx/opti92x-ad1848.c index a1ad39a..df22737 100644 --- a/sound/isa/opti9xx/opti92x-ad1848.c +++ b/sound/isa/opti9xx/opti92x-ad1848.c @@ -1683,6 +1683,8 @@ static int __init snd_card_opti9xx_pnp(struct snd_opti9xx *chip, struct pnp_card struct pnp_resource_table *cfg = kmalloc(sizeof(*cfg), GFP_KERNEL); int err; + if (!cfg) + return -ENOMEM; chip->dev = pnp_request_card_device(card, pid->devs[0].id, NULL); if (chip->dev == NULL) { kfree(cfg); -- cgit v0.10.2 From c06134d73cdc02bb8ab1fad180f6da1f28d2e049 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 11 Oct 2006 18:49:13 +0200 Subject: [ALSA] hda-codec - Fix assignment of PCM devices for Realtek codecs Fixed the assignment of PCM devices for Realtek codecs. The secondary analog capture should be statically asigned to the third device regardless whether SPDIF exists or not. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 84a3eb8..0d728c6 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -1799,7 +1799,7 @@ static int alc_build_pcms(struct hda_codec *codec) /* SPDIF for stream index #1 */ if (spec->multiout.dig_out_nid || spec->dig_in_nid) { codec->num_pcms = 2; - info++; + info = spec->pcm_rec + 1; info->name = spec->stream_name_digital; if (spec->multiout.dig_out_nid && spec->stream_digital_playback) { @@ -1820,7 +1820,7 @@ static int alc_build_pcms(struct hda_codec *codec) if (spec->num_adc_nids > 1 && spec->stream_analog_capture && spec->adc_nids) { codec->num_pcms = 3; - info++; + info = spec->pcm_rec + 2; info->name = spec->stream_name_analog; /* No playback stream for second PCM */ info->stream[SNDRV_PCM_STREAM_PLAYBACK] = alc_pcm_null_playback; -- cgit v0.10.2 From 30b35399ceb2398d05837863476dcb12f12f3a82 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 11 Oct 2006 18:52:53 +0200 Subject: [ALSA] Various fixes for suspend/resume of ALSA PCI drivers - Check the return value of pci_enable_device() and request_irq() in the suspend. If any error occurs there, disable the device using snd_card_disconnect(). - Call pci_set_power_state() properly with pci_choose_state(). - Fix the order to call pci_set_power_state(). - Removed obsolete house-made PM codes in some drivers. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/pci/ali5451/ali5451.c b/sound/pci/ali5451/ali5451.c index 13a8cef..a7edd56 100644 --- a/sound/pci/ali5451/ali5451.c +++ b/sound/pci/ali5451/ali5451.c @@ -2032,8 +2032,10 @@ static int ali_suspend(struct pci_dev *pci, pm_message_t state) outl(0xffffffff, ALI_REG(chip, ALI_STOP)); spin_unlock_irq(&chip->reg_lock); + pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2048,8 +2050,15 @@ static int ali_resume(struct pci_dev *pci) if (! im) return 0; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "ali5451: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } + pci_set_master(pci); spin_lock_irq(&chip->reg_lock); diff --git a/sound/pci/als300.c b/sound/pci/als300.c index 9b16c29..95f70f3 100644 --- a/sound/pci/als300.c +++ b/sound/pci/als300.c @@ -768,9 +768,9 @@ static int snd_als300_suspend(struct pci_dev *pci, pm_message_t state) snd_pcm_suspend_all(chip->pcm); snd_ac97_suspend(chip->ac97); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -779,9 +779,14 @@ static int snd_als300_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct snd_als300 *chip = card->private_data; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "als300: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_als300_init(chip); diff --git a/sound/pci/als4000.c b/sound/pci/als4000.c index 15fc392..8fb55d3 100644 --- a/sound/pci/als4000.c +++ b/sound/pci/als4000.c @@ -804,9 +804,9 @@ static int snd_als4000_suspend(struct pci_dev *pci, pm_message_t state) snd_pcm_suspend_all(chip->pcm); snd_sbmixer_suspend(chip); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -816,9 +816,14 @@ static int snd_als4000_resume(struct pci_dev *pci) struct snd_card_als4000 *acard = card->private_data; struct snd_sb *chip = acard->chip; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "als4000: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_als4000_configure(chip); diff --git a/sound/pci/atiixp.c b/sound/pci/atiixp.c index 3e8fc5a..e3e99f3 100644 --- a/sound/pci/atiixp.c +++ b/sound/pci/atiixp.c @@ -1442,9 +1442,9 @@ static int snd_atiixp_suspend(struct pci_dev *pci, pm_message_t state) snd_atiixp_aclink_down(chip); snd_atiixp_chip_stop(chip); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1454,9 +1454,14 @@ static int snd_atiixp_resume(struct pci_dev *pci) struct atiixp *chip = card->private_data; int i; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "atiixp: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_atiixp_aclink_reset(chip); diff --git a/sound/pci/atiixp_modem.c b/sound/pci/atiixp_modem.c index c5dda1b..dc54f2c 100644 --- a/sound/pci/atiixp_modem.c +++ b/sound/pci/atiixp_modem.c @@ -1128,9 +1128,9 @@ static int snd_atiixp_suspend(struct pci_dev *pci, pm_message_t state) snd_atiixp_aclink_down(chip); snd_atiixp_chip_stop(chip); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1140,9 +1140,14 @@ static int snd_atiixp_resume(struct pci_dev *pci) struct atiixp_modem *chip = card->private_data; int i; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "atiixp-modem: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_atiixp_aclink_reset(chip); diff --git a/sound/pci/azt3328.c b/sound/pci/azt3328.c index 692f203..2414ee6 100644 --- a/sound/pci/azt3328.c +++ b/sound/pci/azt3328.c @@ -1903,9 +1903,9 @@ snd_azf3328_suspend(struct pci_dev *pci, pm_message_t state) for (reg = 0; reg < AZF_IO_SIZE_SYNTH_PM / 2; reg++) chip->saved_regs_synth[reg] = inw(chip->synth_port + reg * 2); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1916,9 +1916,14 @@ snd_azf3328_resume(struct pci_dev *pci) struct snd_azf3328 *chip = card->private_data; int reg; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "azt3328: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); for (reg = 0; reg < AZF_IO_SIZE_IO2_PM / 2; reg++) diff --git a/sound/pci/cmipci.c b/sound/pci/cmipci.c index 1f7e710..0093cd1 100644 --- a/sound/pci/cmipci.c +++ b/sound/pci/cmipci.c @@ -3122,9 +3122,9 @@ static int snd_cmipci_suspend(struct pci_dev *pci, pm_message_t state) /* disable ints */ snd_cmipci_write(cm, CM_REG_INT_HLDCLR, 0); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -3134,9 +3134,14 @@ static int snd_cmipci_resume(struct pci_dev *pci) struct cmipci *cm = card->private_data; int i; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "cmipci: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); /* reset / initialize to a sane state */ diff --git a/sound/pci/cs4281.c b/sound/pci/cs4281.c index d54924e..0905fa8 100644 --- a/sound/pci/cs4281.c +++ b/sound/pci/cs4281.c @@ -2050,6 +2050,7 @@ static int cs4281_suspend(struct pci_dev *pci, pm_message_t state) pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2060,8 +2061,14 @@ static int cs4281_resume(struct pci_dev *pci) unsigned int i; u32 ulCLK; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "cs4281: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); ulCLK = snd_cs4281_peekBA0(chip, BA0_CLKCR1); diff --git a/sound/pci/cs46xx/cs46xx_lib.c b/sound/pci/cs46xx/cs46xx_lib.c index 16d4ebf..2807b97 100644 --- a/sound/pci/cs46xx/cs46xx_lib.c +++ b/sound/pci/cs46xx/cs46xx_lib.c @@ -3687,8 +3687,10 @@ int snd_cs46xx_suspend(struct pci_dev *pci, pm_message_t state) /* disable CLKRUN */ chip->active_ctrl(chip, -chip->amplifier); chip->amplifier = amp_saved; /* restore the status */ + pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -3698,9 +3700,16 @@ int snd_cs46xx_resume(struct pci_dev *pci) struct snd_cs46xx *chip = card->private_data; int amp_saved; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "cs46xx: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); + amp_saved = chip->amplifier; chip->amplifier = 0; chip->active_ctrl(chip, 1); /* force to on */ diff --git a/sound/pci/cs5535audio/cs5535audio_pm.c b/sound/pci/cs5535audio/cs5535audio_pm.c index aad0e69..3e4d198 100644 --- a/sound/pci/cs5535audio/cs5535audio_pm.c +++ b/sound/pci/cs5535audio/cs5535audio_pm.c @@ -73,9 +73,10 @@ int snd_cs5535audio_suspend(struct pci_dev *pci, pm_message_t state) snd_ac97_suspend(cs5535au->ac97); /* save important regs, then disable aclink in hw */ snd_cs5535audio_stop_hardware(cs5535au); + pci_disable_device(pci); pci_save_state(pci); - + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -87,8 +88,14 @@ int snd_cs5535audio_resume(struct pci_dev *pci) int timeout; int i; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "cs5535audio: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); /* set LNK_WRM_RST to reset AC link */ diff --git a/sound/pci/emu10k1/emu10k1.c b/sound/pci/emu10k1/emu10k1.c index 493ec08..55caf34 100644 --- a/sound/pci/emu10k1/emu10k1.c +++ b/sound/pci/emu10k1/emu10k1.c @@ -226,9 +226,9 @@ static int snd_emu10k1_suspend(struct pci_dev *pci, pm_message_t state) snd_emu10k1_done(emu); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -237,11 +237,16 @@ static int snd_emu10k1_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct snd_emu10k1 *emu = card->private_data; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "emu10k1: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); - + snd_emu10k1_resume_init(emu); snd_emu10k1_efx_resume(emu); snd_ac97_resume(emu->ac97); diff --git a/sound/pci/ens1370.c b/sound/pci/ens1370.c index 8cb4fb2..d2a811f 100644 --- a/sound/pci/ens1370.c +++ b/sound/pci/ens1370.c @@ -2072,9 +2072,10 @@ static int snd_ensoniq_suspend(struct pci_dev *pci, pm_message_t state) udelay(100); snd_ak4531_suspend(ensoniq->u.es1370.ak4531); #endif - pci_set_power_state(pci, PCI_D3hot); + pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2083,9 +2084,14 @@ static int snd_ensoniq_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct ensoniq *ensoniq = card->private_data; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR DRIVER_NAME ": pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_ensoniq_chip_init(ensoniq); diff --git a/sound/pci/es1938.c b/sound/pci/es1938.c index 2da988f..1a8d36d 100644 --- a/sound/pci/es1938.c +++ b/sound/pci/es1938.c @@ -1481,10 +1481,14 @@ static int es1938_suspend(struct pci_dev *pci, pm_message_t state) *d = snd_es1938_reg_read(chip, *s); outb(0x00, SLIO_REG(chip, IRQCONTROL)); /* disable irqs */ - if (chip->irq >= 0) + if (chip->irq >= 0) { + synchronize_irq(chip->irq); free_irq(chip->irq, chip); + chip->irq = -1; + } pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1494,10 +1498,22 @@ static int es1938_resume(struct pci_dev *pci) struct es1938 *chip = card->private_data; unsigned char *s, *d; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); - request_irq(pci->irq, snd_es1938_interrupt, - IRQF_DISABLED|IRQF_SHARED, "ES1938", chip); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "es1938: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } + + if (request_irq(pci->irq, snd_es1938_interrupt, + IRQF_DISABLED|IRQF_SHARED, "ES1938", chip)) { + printk(KERN_ERR "es1938: unable to grab IRQ %d, " + "disabling device\n", pci->irq); + snd_card_disconnect(card); + return -EIO; + } chip->irq = pci->irq; snd_es1938_chip_init(chip); @@ -1556,8 +1572,10 @@ static int snd_es1938_free(struct es1938 *chip) snd_es1938_free_gameport(chip); - if (chip->irq >= 0) + if (chip->irq >= 0) { + synchronize_irq(chip->irq); free_irq(chip->irq, chip); + } pci_release_regions(chip->pci); pci_disable_device(chip->pci); kfree(chip); @@ -1602,6 +1620,7 @@ static int __devinit snd_es1938_create(struct snd_card *card, spin_lock_init(&chip->mixer_lock); chip->card = card; chip->pci = pci; + chip->irq = -1; if ((err = pci_request_regions(pci, "ESS Solo-1")) < 0) { kfree(chip); pci_disable_device(pci); diff --git a/sound/pci/es1968.c b/sound/pci/es1968.c index b9d723c..092da53e 100644 --- a/sound/pci/es1968.c +++ b/sound/pci/es1968.c @@ -432,46 +432,6 @@ MODULE_PARM_DESC(joystick, "Enable joystick."); #define ESM_MODE_PLAY 0 #define ESM_MODE_CAPTURE 1 -/* acpi states */ -enum { - ACPI_D0=0, - ACPI_D1, - ACPI_D2, - ACPI_D3 -}; - -/* bits in the acpi masks */ -#define ACPI_12MHZ ( 1 << 15) -#define ACPI_24MHZ ( 1 << 14) -#define ACPI_978 ( 1 << 13) -#define ACPI_SPDIF ( 1 << 12) -#define ACPI_GLUE ( 1 << 11) -#define ACPI__10 ( 1 << 10) /* reserved */ -#define ACPI_PCIINT ( 1 << 9) -#define ACPI_HV ( 1 << 8) /* hardware volume */ -#define ACPI_GPIO ( 1 << 7) -#define ACPI_ASSP ( 1 << 6) -#define ACPI_SB ( 1 << 5) /* sb emul */ -#define ACPI_FM ( 1 << 4) /* fm emul */ -#define ACPI_RB ( 1 << 3) /* ringbus / aclink */ -#define ACPI_MIDI ( 1 << 2) -#define ACPI_GP ( 1 << 1) /* game port */ -#define ACPI_WP ( 1 << 0) /* wave processor */ - -#define ACPI_ALL (0xffff) -#define ACPI_SLEEP (~(ACPI_SPDIF|ACPI_ASSP|ACPI_SB|ACPI_FM| \ - ACPI_MIDI|ACPI_GP|ACPI_WP)) -#define ACPI_NONE (ACPI__10) - -/* these masks indicate which units we care about at - which states */ -static u16 acpi_state_mask[] = { - [ACPI_D0] = ACPI_ALL, - [ACPI_D1] = ACPI_SLEEP, - [ACPI_D2] = ACPI_SLEEP, - [ACPI_D3] = ACPI_NONE -}; - /* APU use in the driver */ enum snd_enum_apu_type { @@ -2160,21 +2120,6 @@ static void snd_es1968_reset(struct es1968 *chip) } /* - * power management - */ -static void snd_es1968_set_acpi(struct es1968 *chip, int state) -{ - u16 active_mask = acpi_state_mask[state]; - - pci_set_power_state(chip->pci, state); - /* make sure the units we care about are on - XXX we might want to do this before state flipping? */ - pci_write_config_word(chip->pci, 0x54, ~ active_mask); - pci_write_config_word(chip->pci, 0x56, ~ active_mask); -} - - -/* * initialize maestro chip */ static void snd_es1968_chip_init(struct es1968 *chip) @@ -2196,9 +2141,6 @@ static void snd_es1968_chip_init(struct es1968 *chip) * IRQs. */ - /* do config work at full power */ - snd_es1968_set_acpi(chip, ACPI_D0); - /* Config Reg A */ pci_read_config_word(pci, ESM_CONFIG_A, &w); @@ -2397,9 +2339,10 @@ static int es1968_suspend(struct pci_dev *pci, pm_message_t state) snd_pcm_suspend_all(chip->pcm); snd_ac97_suspend(chip->ac97); snd_es1968_bob_stop(chip); - snd_es1968_set_acpi(chip, ACPI_D3); + pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2413,9 +2356,16 @@ static int es1968_resume(struct pci_dev *pci) return 0; /* restore all our config */ + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "es1968: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); + snd_es1968_chip_init(chip); /* need to restore the base pointers.. */ @@ -2514,7 +2464,6 @@ static int snd_es1968_free(struct es1968 *chip) if (chip->irq >= 0) free_irq(chip->irq, (void *)chip); snd_es1968_free_gameport(chip); - snd_es1968_set_acpi(chip, ACPI_D3); chip->master_switch = NULL; chip->master_volume = NULL; pci_release_regions(chip->pci); diff --git a/sound/pci/fm801.c b/sound/pci/fm801.c index 3ec7d7e..77e3d5c 100644 --- a/sound/pci/fm801.c +++ b/sound/pci/fm801.c @@ -1531,9 +1531,9 @@ static int snd_fm801_suspend(struct pci_dev *pci, pm_message_t state) chip->saved_regs[i] = inw(chip->port + saved_regs[i]); /* FIXME: tea575x suspend */ - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1543,9 +1543,14 @@ static int snd_fm801_resume(struct pci_dev *pci) struct fm801 *chip = card->private_data; int i; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "fm801: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_fm801_chip_init(chip, 1); diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index feeed12..7c96361 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -1379,12 +1379,16 @@ static int azx_suspend(struct pci_dev *pci, pm_message_t state) snd_pcm_suspend_all(chip->pcm[i]); snd_hda_suspend(chip->bus, state); azx_free_cmd_io(chip); - if (chip->irq >= 0) + if (chip->irq >= 0) { + synchronize_irq(chip->irq); free_irq(chip->irq, chip); + chip->irq = -1; + } if (!disable_msi) pci_disable_msi(chip->pci); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1393,15 +1397,25 @@ static int azx_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct azx *chip = card->private_data; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "hda-intel: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } + pci_set_master(pci); if (!disable_msi) pci_enable_msi(pci); - /* FIXME: need proper error handling */ - request_irq(pci->irq, azx_interrupt, IRQF_DISABLED|IRQF_SHARED, - "HDA Intel", chip); + if (request_irq(pci->irq, azx_interrupt, IRQF_DISABLED|IRQF_SHARED, + "HDA Intel", chip)) { + printk(KERN_ERR "hda-intel: unable to grab IRQ %d, " + "disabling device\n", pci->irq); + snd_card_disconnect(card); + return -EIO; + } chip->irq = pci->irq; - pci_set_master(pci); azx_init_chip(chip); snd_hda_resume(chip->bus); snd_power_change_state(card, SNDRV_CTL_POWER_D0); @@ -1431,15 +1445,14 @@ static int azx_free(struct azx *chip) /* disable position buffer */ azx_writel(chip, DPLBASE, 0); azx_writel(chip, DPUBASE, 0); - - synchronize_irq(chip->irq); } if (chip->irq >= 0) { + synchronize_irq(chip->irq); free_irq(chip->irq, (void*)chip); - if (!disable_msi) - pci_disable_msi(chip->pci); } + if (!disable_msi) + pci_disable_msi(chip->pci); if (chip->remap_addr) iounmap(chip->remap_addr); diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c index f4319b8..7f22dab 100644 --- a/sound/pci/intel8x0.c +++ b/sound/pci/intel8x0.c @@ -2476,10 +2476,14 @@ static int intel8x0_suspend(struct pci_dev *pci, pm_message_t state) if (chip->device_type == DEVICE_INTEL_ICH4) chip->sdm_saved = igetbyte(chip, ICHREG(SDM)); - if (chip->irq >= 0) + if (chip->irq >= 0) { + synchronize_irq(chip->irq); free_irq(chip->irq, chip); + chip->irq = -1; + } pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2489,11 +2493,22 @@ static int intel8x0_resume(struct pci_dev *pci) struct intel8x0 *chip = card->private_data; int i; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "intel8x0: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); - request_irq(pci->irq, snd_intel8x0_interrupt, IRQF_DISABLED|IRQF_SHARED, - card->shortname, chip); + if (request_irq(pci->irq, snd_intel8x0_interrupt, + IRQF_DISABLED|IRQF_SHARED, card->shortname, chip)) { + printk(KERN_ERR "intel8x0: unable to grab IRQ %d, " + "disabling device\n", pci->irq); + snd_card_disconnect(card); + return -EIO; + } chip->irq = pci->irq; synchronize_irq(chip->irq); snd_intel8x0_chip_init(chip, 0); diff --git a/sound/pci/intel8x0m.c b/sound/pci/intel8x0m.c index 6703f5c..bd467c5 100644 --- a/sound/pci/intel8x0m.c +++ b/sound/pci/intel8x0m.c @@ -1045,10 +1045,14 @@ static int intel8x0m_suspend(struct pci_dev *pci, pm_message_t state) for (i = 0; i < chip->pcm_devs; i++) snd_pcm_suspend_all(chip->pcm[i]); snd_ac97_suspend(chip->ac97); - if (chip->irq >= 0) + if (chip->irq >= 0) { + synchronize_irq(chip->irq); free_irq(chip->irq, chip); + chip->irq = -1; + } pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1057,11 +1061,22 @@ static int intel8x0m_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct intel8x0m *chip = card->private_data; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "intel8x0m: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); - request_irq(pci->irq, snd_intel8x0_interrupt, IRQF_DISABLED|IRQF_SHARED, - card->shortname, chip); + if (request_irq(pci->irq, snd_intel8x0_interrupt, + IRQF_DISABLED|IRQF_SHARED, card->shortname, chip)) { + printk(KERN_ERR "intel8x0m: unable to grab IRQ %d, " + "disabling device\n", pci->irq); + snd_card_disconnect(card); + return -EIO; + } chip->irq = pci->irq; snd_intel8x0_chip_init(chip, 0); snd_ac97_resume(chip->ac97); diff --git a/sound/pci/maestro3.c b/sound/pci/maestro3.c index 05605f4..8cab342 100644 --- a/sound/pci/maestro3.c +++ b/sound/pci/maestro3.c @@ -2589,12 +2589,9 @@ static int m3_suspend(struct pci_dev *pci, pm_message_t state) chip->suspend_mem[index++] = snd_m3_assp_read(chip, MEMTYPE_INTERNAL_DATA, i); - /* power down apci registers */ - snd_m3_outw(chip, 0xffff, 0x54); - snd_m3_outw(chip, 0xffff, 0x56); - pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2607,8 +2604,14 @@ static int m3_resume(struct pci_dev *pci) if (chip->suspend_mem == NULL) return 0; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "maestor3: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); /* first lets just bring everything back. .*/ diff --git a/sound/pci/nm256/nm256.c b/sound/pci/nm256/nm256.c index b1bbdb9..945d21b 100644 --- a/sound/pci/nm256/nm256.c +++ b/sound/pci/nm256/nm256.c @@ -1390,6 +1390,7 @@ static int nm256_suspend(struct pci_dev *pci, pm_message_t state) chip->coeffs_current = 0; pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1401,8 +1402,17 @@ static int nm256_resume(struct pci_dev *pci) /* Perform a full reset on the hardware */ chip->in_resume = 1; + + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "nm256: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } + pci_set_master(pci); + snd_nm256_init_chip(chip); /* restore ac97 */ diff --git a/sound/pci/riptide/riptide.c b/sound/pci/riptide/riptide.c index ec48991..56e0c01 100644 --- a/sound/pci/riptide/riptide.c +++ b/sound/pci/riptide/riptide.c @@ -1178,9 +1178,9 @@ static int riptide_suspend(struct pci_dev *pci, pm_message_t state) snd_power_change_state(card, SNDRV_CTL_POWER_D3hot); snd_pcm_suspend_all(chip->pcm); snd_ac97_suspend(chip->ac97); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1189,9 +1189,14 @@ static int riptide_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct snd_riptide *chip = card->private_data; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "riptide: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_riptide_initialize(chip); snd_ac97_resume(chip->ac97); diff --git a/sound/pci/trident/trident_main.c b/sound/pci/trident/trident_main.c index 0d47887..1fbc432 100644 --- a/sound/pci/trident/trident_main.c +++ b/sound/pci/trident/trident_main.c @@ -3966,15 +3966,9 @@ int snd_trident_suspend(struct pci_dev *pci, pm_message_t state) snd_ac97_suspend(trident->ac97); snd_ac97_suspend(trident->ac97_sec); - switch (trident->device) { - case TRIDENT_DEVICE_ID_DX: - case TRIDENT_DEVICE_ID_NX: - break; /* TODO */ - case TRIDENT_DEVICE_ID_SI7018: - break; - } pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -3983,9 +3977,15 @@ int snd_trident_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct snd_trident *trident = card->private_data; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); - pci_set_master(pci); /* to be sure */ + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "trident: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } + pci_set_master(pci); switch (trident->device) { case TRIDENT_DEVICE_ID_DX: diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c index e6990e0..92b0736 100644 --- a/sound/pci/via82xx.c +++ b/sound/pci/via82xx.c @@ -2185,9 +2185,9 @@ static int snd_via82xx_suspend(struct pci_dev *pci, pm_message_t state) chip->capture_src_saved[1] = inb(chip->port + VIA_REG_CAPTURE_CHANNEL + 0x10); } - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2197,9 +2197,15 @@ static int snd_via82xx_resume(struct pci_dev *pci) struct via82xx *chip = card->private_data; int i; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "via82xx: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } + pci_set_master(pci); snd_via82xx_chip_init(chip); diff --git a/sound/pci/via82xx_modem.c b/sound/pci/via82xx_modem.c index 5ab1cf3..feb27c9 100644 --- a/sound/pci/via82xx_modem.c +++ b/sound/pci/via82xx_modem.c @@ -1032,9 +1032,10 @@ static int snd_via82xx_suspend(struct pci_dev *pci, pm_message_t state) snd_via82xx_channel_reset(chip, &chip->devs[i]); synchronize_irq(chip->irq); snd_ac97_suspend(chip->ac97); - pci_set_power_state(pci, PCI_D3hot); + pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -1044,9 +1045,14 @@ static int snd_via82xx_resume(struct pci_dev *pci) struct via82xx_modem *chip = card->private_data; int i; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "via82xx-modem: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_via82xx_chip_init(chip); diff --git a/sound/pci/vx222/vx222.c b/sound/pci/vx222/vx222.c index e7cd8ac..af49e8a 100644 --- a/sound/pci/vx222/vx222.c +++ b/sound/pci/vx222/vx222.c @@ -266,9 +266,9 @@ static int snd_vx222_suspend(struct pci_dev *pci, pm_message_t state) int err; err = snd_vx_suspend(&vx->core, state); - pci_set_power_state(pci, PCI_D3hot); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return err; } @@ -277,9 +277,14 @@ static int snd_vx222_resume(struct pci_dev *pci) struct snd_card *card = pci_get_drvdata(pci); struct snd_vx222 *vx = card->private_data; - pci_restore_state(pci); - pci_enable_device(pci); pci_set_power_state(pci, PCI_D0); + pci_restore_state(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "vx222: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); return snd_vx_resume(&vx->core); } diff --git a/sound/pci/ymfpci/ymfpci_main.c b/sound/pci/ymfpci/ymfpci_main.c index ebc6da8..a40c108 100644 --- a/sound/pci/ymfpci/ymfpci_main.c +++ b/sound/pci/ymfpci/ymfpci_main.c @@ -2218,6 +2218,7 @@ int snd_ymfpci_suspend(struct pci_dev *pci, pm_message_t state) snd_ymfpci_disable_dsp(chip); pci_disable_device(pci); pci_save_state(pci); + pci_set_power_state(pci, pci_choose_state(pci, state)); return 0; } @@ -2227,8 +2228,14 @@ int snd_ymfpci_resume(struct pci_dev *pci) struct snd_ymfpci *chip = card->private_data; unsigned int i; + pci_set_power_state(pci, PCI_D0); pci_restore_state(pci); - pci_enable_device(pci); + if (pci_enable_device(pci) < 0) { + printk(KERN_ERR "ymfpci: pci_enable_device failed, " + "disabling device\n"); + snd_card_disconnect(card); + return -EIO; + } pci_set_master(pci); snd_ymfpci_aclink_reset(pci); snd_ymfpci_codec_ready(chip, 0); -- cgit v0.10.2 From c9949452f221077c22a4765ef380841e169f034b Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 12 Oct 2006 21:10:21 +0200 Subject: [ALSA] Fix dependency of snd-adlib driver in Kconfig Added the missing dependency on CONFIG_SND for snd-adlib driver. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/isa/Kconfig b/sound/isa/Kconfig index 557c4de..57371f1 100644 --- a/sound/isa/Kconfig +++ b/sound/isa/Kconfig @@ -13,6 +13,7 @@ config SND_CS4231_LIB config SND_ADLIB tristate "AdLib FM card" + depends on SND select SND_OPL3_LIB help Say Y here to include support for AdLib FM cards. -- cgit v0.10.2 From 264e6e3b0c5af44d8975153bf0c88ccf2cb10a2f Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 13 Oct 2006 12:40:51 +0200 Subject: [ALSA] hda-codec - Add model entry for ASUS U5F laptop Added a model entry for ASUS U5F laptop with AD1986A codec. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index 511df07..edd22de 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -818,6 +818,8 @@ static struct hda_board_config ad1986a_cfg_tbl[] = { .config = AD1986A_LAPTOP_EAPD }, /* ASUS A6J */ { .pci_subvendor = 0x1043, .pci_subdevice = 0x11f7, .config = AD1986A_LAPTOP_EAPD }, /* ASUS U5A */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1263, + .config = AD1986A_LAPTOP_EAPD }, /* ASUS U5F */ { .pci_subvendor = 0x1043, .pci_subdevice = 0x1297, .config = AD1986A_LAPTOP_EAPD }, /* ASUS Z62F */ { .pci_subvendor = 0x103c, .pci_subdevice = 0x30af, -- cgit v0.10.2 From 5b15c95f889c7bc43cb759c27211c597c0ad9f05 Mon Sep 17 00:00:00 2001 From: Felix Kuehling Date: Mon, 16 Oct 2006 12:49:47 +0200 Subject: [ALSA] hda_intel: add ATI RS690 HDMI audio support This patch adds support for the HDMI codec of the ATI RS690 IGP northbridge. Signed-off-by: Felix Kuehling Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 7c96361..c177192 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -86,6 +86,7 @@ MODULE_SUPPORTED_DEVICE("{{Intel, ICH6}," "{ATI, SB450}," "{ATI, SB600}," "{ATI, RS600}," + "{ATI, RS690}," "{VIA, VT8251}," "{VIA, VT8237A}," "{SiS, SIS966}," @@ -1690,6 +1691,7 @@ static struct pci_device_id azx_ids[] = { { 0x1002, 0x437b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ATI }, /* ATI SB450 */ { 0x1002, 0x4383, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ATI }, /* ATI SB600 */ { 0x1002, 0x793b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ATIHDMI }, /* ATI RS600 HDMI */ + { 0x1002, 0x7919, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ATIHDMI }, /* ATI RS690 HDMI */ { 0x1106, 0x3288, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_VIA }, /* VIA VT8251/VT8237A */ { 0x1039, 0x7502, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_SIS }, /* SIS966 */ { 0x10b9, 0x5461, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ULI }, /* ULI M5461 */ diff --git a/sound/pci/hda/patch_atihdmi.c b/sound/pci/hda/patch_atihdmi.c index a27440f..7333f27 100644 --- a/sound/pci/hda/patch_atihdmi.c +++ b/sound/pci/hda/patch_atihdmi.c @@ -161,5 +161,6 @@ static int patch_atihdmi(struct hda_codec *codec) */ struct hda_codec_preset snd_hda_preset_atihdmi[] = { { .id = 0x1002793c, .name = "ATI RS600 HDMI", .patch = patch_atihdmi }, + { .id = 0x1002791a, .name = "ATI RS690 HDMI", .patch = patch_atihdmi }, {} /* terminator */ }; -- cgit v0.10.2 From dbedca39fe30db87b6401e4396f959c63d90082e Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 18 Oct 2006 19:09:46 +0200 Subject: [ALSA] Fix re-use of va_list The va_list is designed to be used only once. The current code may pass va_list arguments multiple times and may cause Oops. Copy/release the arguments temporarily to avoid this problem. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/core/info.c b/sound/core/info.c index e43662b..0b4aab3 100644 --- a/sound/core/info.c +++ b/sound/core/info.c @@ -120,7 +120,10 @@ int snd_iprintf(struct snd_info_buffer *buffer, char *fmt,...) len = buffer->len - buffer->size; va_start(args, fmt); for (;;) { - res = vsnprintf(buffer->buffer + buffer->curr, len, fmt, args); + va_list ap; + va_copy(ap, args); + res = vsnprintf(buffer->buffer + buffer->curr, len, fmt, ap); + va_end(ap); if (res < len) break; err = resize_info_buffer(buffer, buffer->len + PAGE_SIZE); -- cgit v0.10.2 From da43deb116c0cea9ca2174f2ac64985c4d53077e Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 19 Oct 2006 15:20:08 +0200 Subject: [ALSA] Fix AC97 power-saving mode Fix the bug in AC97 power-saving mode that the power isn't turned on when power_save is set to 1 via sysfs during the power off state. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c index a79e918..6577b23 100644 --- a/sound/pci/ac97/ac97_codec.c +++ b/sound/pci/ac97/ac97_codec.c @@ -570,8 +570,7 @@ int snd_ac97_put_volsw(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value ac97->power_up &= ~(1 << (reg>>1)); else ac97->power_up |= 1 << (reg>>1); - if (power_save) - update_power_regs(ac97); + update_power_regs(ac97); } #endif return err; @@ -2337,10 +2336,7 @@ int snd_ac97_update_power(struct snd_ac97 *ac97, int reg, int powerup) } } - if (! power_save) - return 0; - - if (! powerup && ac97->power_workq) + if (power_save && !powerup && ac97->power_workq) /* adjust power-down bits after two seconds delay * (for avoiding loud click noises for many (OSS) apps * that open/close frequently) -- cgit v0.10.2 From ef35be7f52669cc00c780fa5a640abf0580d1ecf Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 19 Oct 2006 17:38:31 +0200 Subject: [ALSA] Fix addition of user-defined boolean controls Fixed the addition of user-defined boolean controls, the private data size is corrected to be handled properly. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/core/control.c b/sound/core/control.c index 6973a96..48ef0a0 100644 --- a/sound/core/control.c +++ b/sound/core/control.c @@ -1018,10 +1018,6 @@ static int snd_ctl_elem_add(struct snd_ctl_file *file, } switch (info->type) { case SNDRV_CTL_ELEM_TYPE_BOOLEAN: - private_size = sizeof(char); - if (info->count > 128) - return -EINVAL; - break; case SNDRV_CTL_ELEM_TYPE_INTEGER: private_size = sizeof(long); if (info->count > 128) -- cgit v0.10.2 From a5c81b648476f5b0594daeefb38bb98409da5340 Mon Sep 17 00:00:00 2001 From: Jaroslav Kysela Date: Sun, 22 Oct 2006 10:56:24 +0200 Subject: [ALSA] version 1.0.13 Signed-off-by: Jaroslav Kysela diff --git a/include/sound/version.h b/include/sound/version.h index 4ad86eb..52fd687 100644 --- a/include/sound/version.h +++ b/include/sound/version.h @@ -1,3 +1,3 @@ /* include/version.h. Generated by alsa/ksync script. */ #define CONFIG_SND_VERSION "1.0.13" -#define CONFIG_SND_DATE " (Fri Oct 06 18:28:19 2006 UTC)" +#define CONFIG_SND_DATE " (Sun Oct 22 08:56:16 2006 UTC)" -- cgit v0.10.2 From 6a43487f43fbd4e03c606dcb62b98374a3af88fc Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 22 Oct 2006 20:38:00 -0700 Subject: [NET]: kernel-doc fix for sock.h Fix kernel-doc warning in include/net/sock.h: Warning(/var/linsrc/linux-2619-rc1-pv//include/net/sock.h:894): No description found for parameter 'rcu' Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller diff --git a/include/net/sock.h b/include/net/sock.h index 40bb90e..ac286a3 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -884,8 +884,7 @@ static inline int sk_filter(struct sock *sk, struct sk_buff *skb) /** * sk_filter_release: Release a socket filter - * @sk: socket - * @fp: filter to remove + * @rcu: rcu_head that contains the sk_filter info to remove * * Remove a filter from a socket and release its resources. */ -- cgit v0.10.2 From 4e8a5201506423e0241202de1349422af4260296 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 22 Oct 2006 21:00:33 -0700 Subject: [PKT_SCHED] netem: Orphan SKB when adding to queue. The networking emulator can queue SKBs for a very long time, so if you're using netem on the sender side for large bandwidth/delay product testing, the SKB socket send queue sizes become artificially larger. Correct this by calling skb_orphan() in netem_enqueue(). Signed-off-by: David S. Miller diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index e8a7f1b..ecf8e4d 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1634,7 +1634,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1 * is marked here since the boot video device will be the only enabled * video device at this point. */ - +#if 0 static void __devinit fixup_video(struct pci_dev *pdev) { struct pci_dev *bridge; @@ -1663,7 +1663,7 @@ static void __devinit fixup_video(struct pci_dev *pdev) } } DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, fixup_video); - +#endif static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 45939ba..ef8874b 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -170,6 +170,8 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) return NET_XMIT_BYPASS; } + skb_orphan(skb); + /* * If we need to duplicate packet, then re-insert at top of the * qdisc tree, since parent queuer expects that only one -- cgit v0.10.2 From a94b1d1fd7ca3129e1d38d38167779fa6ee69780 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 22 Oct 2006 21:53:30 -0700 Subject: [SPARC64]: 8-byte align return value from compat_alloc_user_space() Otherwise we get a ton of unaligned exceptions, for cases such as compat_sys_msgrcv() which go: p = compat_alloc_user_space(second + sizeof(struct msgbuf)); and here 'second' can for example be an arbitrary odd value. Based upon a bug report from Jurij Smakov. Signed-off-by: David S. Miller diff --git a/include/asm-sparc64/compat.h b/include/asm-sparc64/compat.h index c73935d..36511ca 100644 --- a/include/asm-sparc64/compat.h +++ b/include/asm-sparc64/compat.h @@ -164,7 +164,7 @@ static inline compat_uptr_t ptr_to_compat(void __user *uptr) return (u32)(unsigned long)uptr; } -static __inline__ void __user *compat_alloc_user_space(long len) +static inline void __user *compat_alloc_user_space(long len) { struct pt_regs *regs = current_thread_info()->kregs; unsigned long usp = regs->u_regs[UREG_I6]; @@ -174,7 +174,10 @@ static __inline__ void __user *compat_alloc_user_space(long len) else usp &= 0xffffffffUL; - return (void __user *) (usp - len); + usp -= len; + usp &= ~0x7UL; + + return (void __user *) usp; } struct compat_ipc64_perm { -- cgit v0.10.2 From cbcdb93d4443568f17e93610d240043ec5ac067b Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Tue, 17 Oct 2006 23:08:35 +1000 Subject: [POWERPC] Simplify stolen time calculation In calculating stolen time, we were trying to actually account for time spent in the hypervisor. We don't really have enough information to do that accurately, so don't try. Instead, we now calculate stolen time as time that the current cpu thread is not actually dispatching instructions. On chips without a PURR, we cannot do this, so stolen time will always be zero. On chips with a PURR, this is merely the difference between the elapsed PURR values and the elapsed TB values. This gives us much more sane vaules from tools such as mpstat, even if they are still a bit strange e.g. 2 busy threads on one cpu will both appear to have 50% user time and 50% stolen time while 1 busy thread on a cpu will look like 100% user on one of them and 100% idle on the other. Signed-off-by: Stephen Rothwell Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 5b59bc1..a1b5e4b 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -220,11 +220,8 @@ static void account_process_time(struct pt_regs *regs) */ struct cpu_purr_data { int initialized; /* thread is running */ - u64 tb0; /* timebase at origin time */ - u64 purr0; /* PURR at origin time */ u64 tb; /* last TB value read */ u64 purr; /* last PURR value read */ - u64 stolen; /* stolen time so far */ spinlock_t lock; }; @@ -234,10 +231,8 @@ static void snapshot_tb_and_purr(void *data) { struct cpu_purr_data *p = &__get_cpu_var(cpu_purr_data); - p->tb0 = mftb(); - p->purr0 = mfspr(SPRN_PURR); - p->tb = p->tb0; - p->purr = 0; + p->tb = mftb(); + p->purr = mfspr(SPRN_PURR); wmb(); p->initialized = 1; } @@ -258,37 +253,24 @@ void snapshot_timebases(void) void calculate_steal_time(void) { - u64 tb, purr, t0; + u64 tb, purr; s64 stolen; - struct cpu_purr_data *p0, *pme, *phim; - int cpu; + struct cpu_purr_data *pme; if (!cpu_has_feature(CPU_FTR_PURR)) return; - cpu = smp_processor_id(); - pme = &per_cpu(cpu_purr_data, cpu); + pme = &per_cpu(cpu_purr_data, smp_processor_id()); if (!pme->initialized) return; /* this can happen in early boot */ - p0 = &per_cpu(cpu_purr_data, cpu & ~1); - phim = &per_cpu(cpu_purr_data, cpu ^ 1); - spin_lock(&p0->lock); + spin_lock(&pme->lock); tb = mftb(); - purr = mfspr(SPRN_PURR) - pme->purr0; - if (!phim->initialized || !cpu_online(cpu ^ 1)) { - stolen = (tb - pme->tb) - (purr - pme->purr); - } else { - t0 = pme->tb0; - if (phim->tb0 < t0) - t0 = phim->tb0; - stolen = phim->tb - t0 - phim->purr - purr - p0->stolen; - } - if (stolen > 0) { + purr = mfspr(SPRN_PURR); + stolen = (tb - pme->tb) - (purr - pme->purr); + if (stolen > 0) account_steal_time(current, stolen); - p0->stolen += stolen; - } pme->tb = tb; pme->purr = purr; - spin_unlock(&p0->lock); + spin_unlock(&pme->lock); } /* @@ -297,30 +279,17 @@ void calculate_steal_time(void) */ static void snapshot_purr(void) { - int cpu; - u64 purr; - struct cpu_purr_data *p0, *pme, *phim; + struct cpu_purr_data *pme; unsigned long flags; if (!cpu_has_feature(CPU_FTR_PURR)) return; - cpu = smp_processor_id(); - pme = &per_cpu(cpu_purr_data, cpu); - p0 = &per_cpu(cpu_purr_data, cpu & ~1); - phim = &per_cpu(cpu_purr_data, cpu ^ 1); - spin_lock_irqsave(&p0->lock, flags); - pme->tb = pme->tb0 = mftb(); - purr = mfspr(SPRN_PURR); - if (!phim->initialized) { - pme->purr = 0; - pme->purr0 = purr; - } else { - /* set p->purr and p->purr0 for no change in p0->stolen */ - pme->purr = phim->tb - phim->tb0 - phim->purr - p0->stolen; - pme->purr0 = purr - pme->purr; - } + pme = &per_cpu(cpu_purr_data, smp_processor_id()); + spin_lock_irqsave(&pme->lock, flags); + pme->tb = mftb(); + pme->purr = mfspr(SPRN_PURR); pme->initialized = 1; - spin_unlock_irqrestore(&p0->lock, flags); + spin_unlock_irqrestore(&pme->lock, flags); } #endif /* CONFIG_PPC_SPLPAR */ -- cgit v0.10.2 From c3386e40142e9d0c077460c2a548c4653fecaf15 Mon Sep 17 00:00:00 2001 From: Zang Roy-r61911 Date: Wed, 18 Oct 2006 11:18:58 +0800 Subject: [POWERPC] Fix compiler warning message on get_property call This fixes the warning message from the return value of function get_property(), by making sure that the variable that receives the value is marked as const. Signed-off-by: Roy Zang -- Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/sysdev/tsi108_dev.c b/arch/powerpc/sysdev/tsi108_dev.c index 11de090..97f37ef 100644 --- a/arch/powerpc/sysdev/tsi108_dev.c +++ b/arch/powerpc/sysdev/tsi108_dev.c @@ -48,7 +48,7 @@ phys_addr_t get_csrbase(void) tsi = of_find_node_by_type(NULL, "tsi-bridge"); if (tsi) { unsigned int size; - void *prop = get_property(tsi, "reg", &size); + const void *prop = get_property(tsi, "reg", &size); tsi108_csr_base = of_translate_address(tsi, prop); of_node_put(tsi); }; @@ -79,7 +79,7 @@ static int __init tsi108_eth_of_init(void) hw_info tsi_eth_data; unsigned int *id; unsigned int *phy_id; - void *mac_addr; + const void *mac_addr; phandle *ph; memset(r, 0, sizeof(r)); -- cgit v0.10.2 From 7f8c4c50bda13d27afc03679d25aa1fcac8df551 Mon Sep 17 00:00:00 2001 From: Srinivasa Ds Date: Wed, 18 Oct 2006 17:34:49 +0530 Subject: [POWERPC] Fix build breakage with CONFIG_PPC32 low_cpu_die is called from the CPU hotplug code on 32-bit powermacs, but it is only defined if CONFIG_PM || CONFIG_CPU_FREQ_PMAC. This changes the ifdef so it is defined for CONFIG_HOTPLUG_CPU on 32-bit machines. Signed-off-by: Srinivasa DS Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/powermac/sleep.S b/arch/powerpc/platforms/powermac/sleep.S index 1174ca1..adee28d 100644 --- a/arch/powerpc/platforms/powermac/sleep.S +++ b/arch/powerpc/platforms/powermac/sleep.S @@ -45,7 +45,8 @@ .section .text .align 5 -#if defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ_PMAC) +#if defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ_PMAC) || \ + (defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32)) /* This gets called by via-pmu.c late during the sleep process. * The PMU was already send the sleep command and will shut us down -- cgit v0.10.2 From 362ff7b2ac0234152b4a334dd006b77f4fa2ab23 Mon Sep 17 00:00:00 2001 From: Jake Moilanen Date: Wed, 18 Oct 2006 10:47:22 -0500 Subject: [POWERPC] Add 970GX cputable entry 970GX cputable entry from Steve Winiecki. Signed-off-by: Jake Moilanen arch/powerpc/kernel/cputable.c | 15 +++++++++++++++ arch/powerpc/oprofile/op_model_power4.c | 2 +- include/asm-powerpc/reg.h | 1 + 3 files changed, 17 insertions(+), 1 deletion(-) Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 95382f9..f23aad6 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -227,6 +227,21 @@ struct cpu_spec cpu_specs[] = { .oprofile_type = PPC_OPROFILE_POWER4, .platform = "ppc970", }, + { /* PPC970GX */ + .pvr_mask = 0xffff0000, + .pvr_value = 0x00450000, + .cpu_name = "PPC970GX", + .cpu_features = CPU_FTRS_PPC970, + .cpu_user_features = COMMON_USER_POWER4 | + PPC_FEATURE_HAS_ALTIVEC_COMP, + .icache_bsize = 128, + .dcache_bsize = 128, + .num_pmcs = 8, + .cpu_setup = __setup_cpu_ppc970, + .oprofile_cpu_type = "ppc64/970", + .oprofile_type = PPC_OPROFILE_POWER4, + .platform = "ppc970", + }, { /* Power5 GR */ .pvr_mask = 0xffff0000, .pvr_value = 0x003a0000, diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c index 506f6b7..6a927ef 100644 --- a/arch/powerpc/oprofile/op_model_power4.c +++ b/arch/powerpc/oprofile/op_model_power4.c @@ -76,7 +76,7 @@ static inline int mmcra_must_set_sample(void) { if (__is_processor(PV_POWER4) || __is_processor(PV_POWER4p) || __is_processor(PV_970) || __is_processor(PV_970FX) || - __is_processor(PV_970MP)) + __is_processor(PV_970MP) || __is_processor(PV_970GX)) return 1; return 0; diff --git a/include/asm-powerpc/reg.h b/include/asm-powerpc/reg.h index 8fb9681..fde5c80 100644 --- a/include/asm-powerpc/reg.h +++ b/include/asm-powerpc/reg.h @@ -591,6 +591,7 @@ #define PV_630 0x0040 #define PV_630p 0x0041 #define PV_970MP 0x0044 +#define PV_970GX 0x0045 #define PV_BE 0x0070 #define PV_PA6T 0x0090 -- cgit v0.10.2 From 3688a0f484e980771c078ab90f74a2656c339106 Mon Sep 17 00:00:00 2001 From: "Mark A. Greer" Date: Wed, 18 Oct 2006 14:38:09 -0700 Subject: [POWERPC] Don't require execute perms on wrapper when building zImage.initrd Don't require that the wrapper script be executable when building zImage.initrds. This has already been fixed for zImages. Signed-off-by: Mark A. Greer Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 37ddfca..4b2be61 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -115,7 +115,7 @@ endif quiet_cmd_wrap = WRAP $@ cmd_wrap =$(CONFIG_SHELL) $(wrapper) -c -o $@ -p $2 $(CROSSWRAP) vmlinux quiet_cmd_wrap_initrd = WRAP $@ - cmd_wrap_initrd =$(wrapper) -c -o $@ -p $2 $(CROSSWRAP) \ + cmd_wrap_initrd =$(CONFIG_SHELL) $(wrapper) -c -o $@ -p $2 $(CROSSWRAP) \ -i $(obj)/ramdisk.image.gz vmlinux $(obj)/zImage.chrp: vmlinux $(wrapperbits) -- cgit v0.10.2 From aa7a32cbdac50bb46a75722faa359993dab07c61 Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Wed, 18 Oct 2006 17:27:32 -0500 Subject: [POWERPC] Fix spelling errors in ucc_fast.c and ucc_slow.c In ucc_fast.c and ucc_slow.c, "illegal" is twice spelled "illagal". Signed-off-by: Timur Tabi Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/sysdev/qe_lib/ucc_fast.c b/arch/powerpc/sysdev/qe_lib/ucc_fast.c index c2be734..75fa310 100644 --- a/arch/powerpc/sysdev/qe_lib/ucc_fast.c +++ b/arch/powerpc/sysdev/qe_lib/ucc_fast.c @@ -163,7 +163,7 @@ int ucc_fast_init(struct ucc_fast_info * uf_info, struct ucc_fast_private ** ucc /* check if the UCC port number is in range. */ if ((uf_info->ucc_num < 0) || (uf_info->ucc_num > UCC_MAX_NUM - 1)) { - uccf_err("ucc_fast_init: Illagal UCC number!"); + uccf_err("ucc_fast_init: Illegal UCC number!"); return -EINVAL; } diff --git a/arch/powerpc/sysdev/qe_lib/ucc_slow.c b/arch/powerpc/sysdev/qe_lib/ucc_slow.c index 1fb88ef..a49da6b 100644 --- a/arch/powerpc/sysdev/qe_lib/ucc_slow.c +++ b/arch/powerpc/sysdev/qe_lib/ucc_slow.c @@ -152,7 +152,7 @@ int ucc_slow_init(struct ucc_slow_info * us_info, struct ucc_slow_private ** ucc /* check if the UCC port number is in range. */ if ((us_info->ucc_num < 0) || (us_info->ucc_num > UCC_MAX_NUM - 1)) { - uccs_err("ucc_slow_init: Illagal UCC number!"); + uccs_err("ucc_slow_init: Illegal UCC number!"); return -EINVAL; } -- cgit v0.10.2 From 68e7fffc0f3e95063ba5bd94ee6f9b8927247297 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 23 Oct 2006 13:40:59 +0200 Subject: [ALSA] hda-intel - Add check of MSI availabity Check the availability of MSI and turn off MSI automatically when it's not available on the hardware. MSI seems broken on some hardwares but the kernel doesn't know exactly, thus we have to turn the MSI feature off on the sound driver manually. Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index c177192..0e292dc 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -337,6 +337,7 @@ struct azx { unsigned int initialized :1; unsigned int single_cmd :1; unsigned int polling_mode :1; + unsigned int msi :1; }; /* driver types */ @@ -397,6 +398,7 @@ static char *driver_short_names[] __devinitdata = { */ #define upper_32bit(addr) (sizeof(addr) > 4 ? (u32)((addr) >> 32) : (u32)0) +static int azx_acquire_irq(struct azx *chip, int do_disconnect); /* * Interface for HD codec @@ -536,6 +538,18 @@ static unsigned int azx_rirb_get_response(struct hda_codec *codec) schedule_timeout_interruptible(1); } while (time_after_eq(timeout, jiffies)); + if (chip->msi) { + snd_printk(KERN_WARNING "hda_intel: No response from codec, " + "disabling MSI...\n"); + free_irq(chip->irq, chip); + chip->irq = -1; + pci_disable_msi(chip->pci); + chip->msi = 0; + if (azx_acquire_irq(chip, 1) < 0) + return -1; + goto again; + } + if (!chip->polling_mode) { snd_printk(KERN_WARNING "hda_intel: azx_get_response timeout, " "switching to polling mode...\n"); @@ -1364,6 +1378,20 @@ static int __devinit azx_init_stream(struct azx *chip) return 0; } +static int azx_acquire_irq(struct azx *chip, int do_disconnect) +{ + if (request_irq(chip->pci->irq, azx_interrupt, IRQF_DISABLED|IRQF_SHARED, + "HDA Intel", chip)) { + printk(KERN_ERR "hda-intel: unable to grab IRQ %d, " + "disabling device\n", chip->pci->irq); + if (do_disconnect) + snd_card_disconnect(chip->card); + return -1; + } + chip->irq = chip->pci->irq; + return 0; +} + #ifdef CONFIG_PM /* @@ -1385,7 +1413,7 @@ static int azx_suspend(struct pci_dev *pci, pm_message_t state) free_irq(chip->irq, chip); chip->irq = -1; } - if (!disable_msi) + if (chip->msi) pci_disable_msi(chip->pci); pci_disable_device(pci); pci_save_state(pci); @@ -1407,16 +1435,11 @@ static int azx_resume(struct pci_dev *pci) return -EIO; } pci_set_master(pci); - if (!disable_msi) - pci_enable_msi(pci); - if (request_irq(pci->irq, azx_interrupt, IRQF_DISABLED|IRQF_SHARED, - "HDA Intel", chip)) { - printk(KERN_ERR "hda-intel: unable to grab IRQ %d, " - "disabling device\n", pci->irq); - snd_card_disconnect(card); + if (chip->msi) + if (pci_enable_msi(pci) < 0) + chip->msi = 0; + if (azx_acquire_irq(chip, 1) < 0) return -EIO; - } - chip->irq = pci->irq; azx_init_chip(chip); snd_hda_resume(chip->bus); snd_power_change_state(card, SNDRV_CTL_POWER_D0); @@ -1452,7 +1475,7 @@ static int azx_free(struct azx *chip) synchronize_irq(chip->irq); free_irq(chip->irq, (void*)chip); } - if (!disable_msi) + if (chip->msi) pci_disable_msi(chip->pci); if (chip->remap_addr) iounmap(chip->remap_addr); @@ -1508,6 +1531,7 @@ static int __devinit azx_create(struct snd_card *card, struct pci_dev *pci, chip->pci = pci; chip->irq = -1; chip->driver_type = driver_type; + chip->msi = !disable_msi; chip->position_fix = position_fix; chip->single_cmd = single_cmd; @@ -1537,16 +1561,14 @@ static int __devinit azx_create(struct snd_card *card, struct pci_dev *pci, goto errout; } - if (!disable_msi) - pci_enable_msi(pci); + if (chip->msi) + if (pci_enable_msi(pci) < 0) + chip->msi = 0; - if (request_irq(pci->irq, azx_interrupt, IRQF_DISABLED|IRQF_SHARED, - "HDA Intel", (void*)chip)) { - snd_printk(KERN_ERR SFX "unable to grab IRQ %d\n", pci->irq); + if (azx_acquire_irq(chip, 0) < 0) { err = -EBUSY; goto errout; } - chip->irq = pci->irq; pci_set_master(pci); synchronize_irq(chip->irq); -- cgit v0.10.2 From 04fed361dadb7921507a470947ac23d2f26352cf Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 22 Oct 2006 15:57:18 +0100 Subject: [PATCH] Remove __must_check for device_for_each_child() Eliminate more __must_check madness. The return code from device_for_each_child() depends on the values which the helper function returns. If the helper function always returns zero, it's utterly pointless to check the return code from device_for_each_child(). The only code which knows if the return value should be checked is the caller itself, so forcing the return code to always be checked is silly. Hence, remove the __must_check annotation. Signed-off-by: Russell King Signed-off-by: Linus Torvalds diff --git a/include/linux/device.h b/include/linux/device.h index 662e6a1..9d4f6a9 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -393,7 +393,7 @@ extern void device_unregister(struct device * dev); extern void device_initialize(struct device * dev); extern int __must_check device_add(struct device * dev); extern void device_del(struct device * dev); -extern int __must_check device_for_each_child(struct device *, void *, +extern int device_for_each_child(struct device *, void *, int (*fn)(struct device *, void *)); extern int device_rename(struct device *dev, char *new_name); -- cgit v0.10.2 From 0c0e4668e0e65dd1404e8cf066d147235f95561d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 23 Oct 2006 14:25:30 -0700 Subject: Revert unintentional and bogus change to drivers/pci/quirks.c In commit 4e8a5201506423e0241202de1349422af4260296 ("[PKT_SCHED] netem: Orphan SKB when adding to queue.") Davem mistakenly also included a temporary diff in his tree that disabled the pci_fixup_video VGA quirk, which broke sparc64. This reverts that part of the commit. Sayeth Davem: "Greg KH has a patch coming to you soon which will move that VGA code back into x86/x86_64/IA64 specific areas and will fix the sparc64 problem properly." Special thanks to Claudio Martins for noticing the error in the first place. Cc: Claudio Martins Cc: David Miller Signed-off-by: Linus Torvalds diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index ecf8e4d..e8a7f1b 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1634,7 +1634,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1 * is marked here since the boot video device will be the only enabled * video device at this point. */ -#if 0 + static void __devinit fixup_video(struct pci_dev *pdev) { struct pci_dev *bridge; @@ -1663,7 +1663,7 @@ static void __devinit fixup_video(struct pci_dev *pdev) } } DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, fixup_video); -#endif + static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { -- cgit v0.10.2 From 7059abedd2f04b68bd7e1a79c9c72f7aeee134c0 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 23 Oct 2006 16:02:02 -0700 Subject: Linux 2.6.19-rc3 diff --git a/Makefile b/Makefile index 389ff0c..25b3599 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 19 -EXTRAVERSION =-rc2 +EXTRAVERSION =-rc3 NAME=Avast! A bilge rat! # *DOCUMENTATION* -- cgit v0.10.2 From a1aa28970316d7fb606321d5ab7fb3873641ab54 Mon Sep 17 00:00:00 2001 From: Roland Scheidegger Date: Tue, 24 Oct 2006 21:45:00 +1000 Subject: drm: radeon: only allow specific type-3 packetss through verifier only allow specific type-3 packets to pass the verifier instead of all for r100/r200 as others might be unsafe (r300 already does this), and add checking for these we need but aren't safe. Check the RADEON_CP_INDX_BUFFER packet on both r200 and r300 as it isn't safe neither. Signed-off-by: Dave Airlie diff --git a/drivers/char/drm/r300_cmdbuf.c b/drivers/char/drm/r300_cmdbuf.c index 26bdf2c..d14477b 100644 --- a/drivers/char/drm/r300_cmdbuf.c +++ b/drivers/char/drm/r300_cmdbuf.c @@ -538,6 +538,36 @@ static __inline__ int r300_emit_bitblt_multi(drm_radeon_private_t *dev_priv, return 0; } +static __inline__ int r300_emit_indx_buffer(drm_radeon_private_t *dev_priv, + drm_radeon_kcmd_buffer_t *cmdbuf) +{ + u32 *cmd = (u32 *) cmdbuf->buf; + int count, ret; + RING_LOCALS; + + count=(cmd[0]>>16) & 0x3fff; + + if ((cmd[1] & 0x8000ffff) != 0x80000810) { + DRM_ERROR("Invalid indx_buffer reg address %08X\n", cmd[1]); + return DRM_ERR(EINVAL); + } + ret = r300_check_offset(dev_priv, cmd[2]); + if (ret) { + DRM_ERROR("Invalid indx_buffer offset is %08X\n", cmd[2]); + return DRM_ERR(EINVAL); + } + + BEGIN_RING(count+2); + OUT_RING(cmd[0]); + OUT_RING_TABLE((int *)(cmdbuf->buf + 4), count + 1); + ADVANCE_RING(); + + cmdbuf->buf += (count+2)*4; + cmdbuf->bufsz -= (count+2)*4; + + return 0; +} + static __inline__ int r300_emit_raw_packet3(drm_radeon_private_t *dev_priv, drm_radeon_kcmd_buffer_t *cmdbuf) { @@ -578,10 +608,11 @@ static __inline__ int r300_emit_raw_packet3(drm_radeon_private_t *dev_priv, case RADEON_CNTL_BITBLT_MULTI: return r300_emit_bitblt_multi(dev_priv, cmdbuf); + case RADEON_CP_INDX_BUFFER: /* DRAW_INDX_2 without INDX_BUFFER seems to lock up the gpu */ + return r300_emit_indx_buffer(dev_priv, cmdbuf); case RADEON_CP_3D_DRAW_IMMD_2: /* triggers drawing using in-packet vertex data */ case RADEON_CP_3D_DRAW_VBUF_2: /* triggers drawing of vertex buffers setup elsewhere */ case RADEON_CP_3D_DRAW_INDX_2: /* triggers drawing using indices to vertex buffer */ - case RADEON_CP_INDX_BUFFER: /* DRAW_INDX_2 without INDX_BUFFER seems to lock up the gpu */ case RADEON_WAIT_FOR_IDLE: case RADEON_CP_NOP: /* these packets are safe */ diff --git a/drivers/char/drm/radeon_state.c b/drivers/char/drm/radeon_state.c index feac5f0..6e04fdd 100644 --- a/drivers/char/drm/radeon_state.c +++ b/drivers/char/drm/radeon_state.c @@ -275,6 +275,8 @@ static __inline__ int radeon_check_and_fixup_packet3(drm_radeon_private_t * unsigned int *cmdsz) { u32 *cmd = (u32 *) cmdbuf->buf; + u32 offset, narrays; + int count, i, k; *cmdsz = 2 + ((cmd[0] & RADEON_CP_PACKET_COUNT_MASK) >> 16); @@ -288,10 +290,106 @@ static __inline__ int radeon_check_and_fixup_packet3(drm_radeon_private_t * return DRM_ERR(EINVAL); } - /* Check client state and fix it up if necessary */ - if (cmd[0] & 0x8000) { /* MSB of opcode: next DWORD GUI_CNTL */ - u32 offset; + switch(cmd[0] & 0xff00) { + /* XXX Are there old drivers needing other packets? */ + case RADEON_3D_DRAW_IMMD: + case RADEON_3D_DRAW_VBUF: + case RADEON_3D_DRAW_INDX: + case RADEON_WAIT_FOR_IDLE: + case RADEON_CP_NOP: + case RADEON_3D_CLEAR_ZMASK: +/* case RADEON_CP_NEXT_CHAR: + case RADEON_CP_PLY_NEXTSCAN: + case RADEON_CP_SET_SCISSORS: */ /* probably safe but will never need them? */ + /* these packets are safe */ + break; + + case RADEON_CP_3D_DRAW_IMMD_2: + case RADEON_CP_3D_DRAW_VBUF_2: + case RADEON_CP_3D_DRAW_INDX_2: + case RADEON_3D_CLEAR_HIZ: + /* safe but r200 only */ + if (dev_priv->microcode_version != UCODE_R200) { + DRM_ERROR("Invalid 3d packet for r100-class chip\n"); + return DRM_ERR(EINVAL); + } + break; + + case RADEON_3D_LOAD_VBPNTR: + count = (cmd[0] >> 16) & 0x3fff; + + if (count > 18) { /* 12 arrays max */ + DRM_ERROR("Too large payload in 3D_LOAD_VBPNTR (count=%d)\n", + count); + return DRM_ERR(EINVAL); + } + + /* carefully check packet contents */ + narrays = cmd[1] & ~0xc000; + k = 0; + i = 2; + while ((k < narrays) && (i < (count + 2))) { + i++; /* skip attribute field */ + if (radeon_check_and_fixup_offset(dev_priv, filp_priv, &cmd[i])) { + DRM_ERROR + ("Invalid offset (k=%d i=%d) in 3D_LOAD_VBPNTR packet.\n", + k, i); + return DRM_ERR(EINVAL); + } + k++; + i++; + if (k == narrays) + break; + /* have one more to process, they come in pairs */ + if (radeon_check_and_fixup_offset(dev_priv, filp_priv, &cmd[i])) { + DRM_ERROR + ("Invalid offset (k=%d i=%d) in 3D_LOAD_VBPNTR packet.\n", + k, i); + return DRM_ERR(EINVAL); + } + k++; + i++; + } + /* do the counts match what we expect ? */ + if ((k != narrays) || (i != (count + 2))) { + DRM_ERROR + ("Malformed 3D_LOAD_VBPNTR packet (k=%d i=%d narrays=%d count+1=%d).\n", + k, i, narrays, count + 1); + return DRM_ERR(EINVAL); + } + break; + + case RADEON_3D_RNDR_GEN_INDX_PRIM: + if (dev_priv->microcode_version != UCODE_R100) { + DRM_ERROR("Invalid 3d packet for r200-class chip\n"); + return DRM_ERR(EINVAL); + } + if (radeon_check_and_fixup_offset(dev_priv, filp_priv, &cmd[1])) { + DRM_ERROR("Invalid rndr_gen_indx offset\n"); + return DRM_ERR(EINVAL); + } + break; + + case RADEON_CP_INDX_BUFFER: + if (dev_priv->microcode_version != UCODE_R200) { + DRM_ERROR("Invalid 3d packet for r100-class chip\n"); + return DRM_ERR(EINVAL); + } + if ((cmd[1] & 0x8000ffff) != 0x80000810) { + DRM_ERROR("Invalid indx_buffer reg address %08X\n", cmd[1]); + return DRM_ERR(EINVAL); + } + if (radeon_check_and_fixup_offset(dev_priv, filp_priv, &cmd[2])) { + DRM_ERROR("Invalid indx_buffer offset is %08X\n", cmd[2]); + return DRM_ERR(EINVAL); + } + break; + + case RADEON_CNTL_HOSTDATA_BLT: + case RADEON_CNTL_PAINT_MULTI: + case RADEON_CNTL_BITBLT_MULTI: + /* MSB of opcode: next DWORD GUI_CNTL */ if (cmd[1] & (RADEON_GMC_SRC_PITCH_OFFSET_CNTL | RADEON_GMC_DST_PITCH_OFFSET_CNTL)) { offset = cmd[2] << 10; @@ -313,6 +411,11 @@ static __inline__ int radeon_check_and_fixup_packet3(drm_radeon_private_t * } cmd[3] = (cmd[3] & 0xffc00000) | offset >> 10; } + break; + + default: + DRM_ERROR("Invalid packet type %x\n", cmd[0] & 0xff00); + return DRM_ERR(EINVAL); } return 0; -- cgit v0.10.2 From 10eee0fe9114694401c7ae154e8cfb2ab2f59c10 Mon Sep 17 00:00:00 2001 From: Michael Karcher Date: Tue, 24 Oct 2006 21:46:55 +1000 Subject: drm: savage: dev->agp_buffer_map is not initialized for AGP DMA on savages fd.o bug 8662 Signed-off-by: Dave Airlie diff --git a/drivers/char/drm/savage_bci.c b/drivers/char/drm/savage_bci.c index 59c7520..a9a84f8 100644 --- a/drivers/char/drm/savage_bci.c +++ b/drivers/char/drm/savage_bci.c @@ -728,6 +728,7 @@ static int savage_do_init_bci(drm_device_t * dev, drm_savage_init_t * init) dev_priv->status = NULL; } if (dev_priv->dma_type == SAVAGE_DMA_AGP && init->buffers_offset) { + dev->agp_buffer_token = init->buffers_offset; dev->agp_buffer_map = drm_core_findmap(dev, init->buffers_offset); if (!dev->agp_buffer_map) { -- cgit v0.10.2 From 958de71b1ab01c20c1b385035235746c9227b24f Mon Sep 17 00:00:00 2001 From: Tilman Sauerbeck Date: Tue, 24 Oct 2006 21:52:23 +1000 Subject: drm: mga: set dev_priv_size fd.o bug 1746 Signed-off-by: Dave Airlie diff --git a/drivers/char/drm/mga_drv.c b/drivers/char/drm/mga_drv.c index e30f556..be49dbb 100644 --- a/drivers/char/drm/mga_drv.c +++ b/drivers/char/drm/mga_drv.c @@ -47,6 +47,7 @@ static struct drm_driver driver = { DRIVER_USE_AGP | DRIVER_USE_MTRR | DRIVER_PCI_DMA | DRIVER_HAVE_DMA | DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_IRQ_VBL, + .dev_priv_size = sizeof(drm_mga_buf_priv_t), .load = mga_driver_load, .unload = mga_driver_unload, .lastclose = mga_driver_lastclose, -- cgit v0.10.2 From cf610ca226a484f3182a59f168402cb27dcc1c53 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 20 Oct 2006 20:16:24 +0100 Subject: [ARM] Fix breakage in 7281c248f797723f66244b7ecef204620f664648 A couple of missing semicolons. Signed-off-by: Russell King diff --git a/arch/arm/mach-ixp2000/ixdp2400.c b/arch/arm/mach-ixp2000/ixdp2400.c index 9ee6383..0fdd03a 100644 --- a/arch/arm/mach-ixp2000/ixdp2400.c +++ b/arch/arm/mach-ixp2000/ixdp2400.c @@ -135,11 +135,11 @@ static void ixdp2400_pci_postinit(void) if (ixdp2x00_master_npu()) { dev = pci_get_bus_and_slot(1, IXDP2400_SLAVE_ENET_DEVFN); pci_remove_bus_device(dev); - pci_dev_put(dev) + pci_dev_put(dev); } else { dev = pci_get_bus_and_slot(1, IXDP2400_MASTER_ENET_DEVFN); pci_remove_bus_device(dev); - pci_dev_put(dev) + pci_dev_put(dev); ixdp2x00_slave_pci_postinit(); } -- cgit v0.10.2 From f586fbd0ef273a80d88a07f911d9f2f2a8ac1679 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 20 Oct 2006 00:56:28 +0100 Subject: [ARM] 3897/1: corgi_bl fix module compiling Fix module compiling: WARNING: drivers/video/backlight/corgi_bl.o - Section mismatch: reference to .init.text: from .data between '$d' (at offset 0x0) and 'bl_mutex' Cc: Richard Purdie Signed-off-by: Anton Vorontsov Signed-off-by: Russell King diff --git a/drivers/video/backlight/corgi_bl.c b/drivers/video/backlight/corgi_bl.c index 2ebbfd9..6697778 100644 --- a/drivers/video/backlight/corgi_bl.c +++ b/drivers/video/backlight/corgi_bl.c @@ -111,7 +111,7 @@ static struct backlight_properties corgibl_data = { .update_status = corgibl_set_intensity, }; -static int __init corgibl_probe(struct platform_device *pdev) +static int corgibl_probe(struct platform_device *pdev) { struct corgibl_machinfo *machinfo = pdev->dev.platform_data; -- cgit v0.10.2 From 62c877b9b7c463aa16ffbc9a322cb094fdb5827a Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 20 Oct 2006 00:58:49 +0100 Subject: [ARM] 3898/1: corgi_bl fix module loading Fix module loading: corgi_bl: module license 'GPLv2' taints kernel. corgi_bl: Unknown symbol platform_driver_unregister corgi_bl: Unknown symbol __symbol_get corgi_bl: Unknown symbol platform_driver_register Cc: Richard Purdie Signed-off-by: pHilipp Zabel Signed-off-by: Anton Vorontsov Signed-off-by: Russell King diff --git a/drivers/video/backlight/corgi_bl.c b/drivers/video/backlight/corgi_bl.c index 6697778..d07ecb5 100644 --- a/drivers/video/backlight/corgi_bl.c +++ b/drivers/video/backlight/corgi_bl.c @@ -166,4 +166,4 @@ module_exit(corgibl_exit); MODULE_AUTHOR("Richard Purdie "); MODULE_DESCRIPTION("Corgi Backlight Driver"); -MODULE_LICENSE("GPLv2"); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 52f0c67340ca306d5802b52140a186fcfa4b340e Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 22 Oct 2006 16:04:29 +0100 Subject: [ARM] Comment out missing configuration symbols HAS_TOUCHSCREEN_ADS7843_LH7 and HAS_TOUCHSCREEN_ADC_LH7 are referenced but not defined in the LH7A40x configuration. Comment them out to prevent them causing warnings. Marc Singer said: Feel free to remove the Kconfig lines. I'll add it back with the rest of the config entries. Signed-off-by: Russell King diff --git a/arch/arm/mach-lh7a40x/Kconfig b/arch/arm/mach-lh7a40x/Kconfig index 558a34f..147b019 100644 --- a/arch/arm/mach-lh7a40x/Kconfig +++ b/arch/arm/mach-lh7a40x/Kconfig @@ -14,7 +14,7 @@ config MACH_LPD7A400 bool "LPD7A400 Card Engine" select ARCH_LH7A400 # select IDE_POLL - select HAS_TOUCHSCREEN_ADS7843_LH7 +# select HAS_TOUCHSCREEN_ADS7843_LH7 help Say Y here if you are using Logic Product Development's LPD7A400 CardEngine. For the time being, the LPD7A400 and @@ -24,7 +24,7 @@ config MACH_LPD7A404 bool "LPD7A404 Card Engine" select ARCH_LH7A404 # select IDE_POLL - select HAS_TOUCHSCREEN_ADC_LH7 +# select HAS_TOUCHSCREEN_ADC_LH7 help Say Y here if you are using Logic Product Development's LPD7A404 CardEngine. For the time being, the LPD7A400 and -- cgit v0.10.2 From 64271c4d909a15bc588f053a739db2e6df336d7d Mon Sep 17 00:00:00 2001 From: Craig Hughes Date: Tue, 24 Oct 2006 00:47:35 +0100 Subject: [ARM] 3902/1: Enable GPIO81-84 on PXA255 The PXA255 has 84 GPIO lines available. This patch allows access to 81-84 Signed-off-by: Craig Hughes Signed-off-by: Russell King diff --git a/include/asm-arm/arch-pxa/irqs.h b/include/asm-arm/arch-pxa/irqs.h index f3bc70e..67ed436 100644 --- a/include/asm-arm/arch-pxa/irqs.h +++ b/include/asm-arm/arch-pxa/irqs.h @@ -73,7 +73,7 @@ #define IRQ_TO_GPIO(i) (((i) < IRQ_GPIO(2)) ? ((i) - IRQ_GPIO0) : IRQ_TO_GPIO_2_x(i)) #if defined(CONFIG_PXA25x) -#define PXA_LAST_GPIO 80 +#define PXA_LAST_GPIO 84 #elif defined(CONFIG_PXA27x) #define PXA_LAST_GPIO 127 #endif -- cgit v0.10.2 From 4ccc12aeece8ab14ad96461c4db269aea080715d Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Tue, 24 Oct 2006 14:45:53 -0700 Subject: e1000: FIX: don't poke at manageability registers for incompatible adapters The MANC register should not be read for PCI-E adapters at all, as well as 82543 and older where 82543 would master abort when this register was accessed. Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_ethtool.c b/drivers/net/e1000/e1000_ethtool.c index 773821e..71fb278 100644 --- a/drivers/net/e1000/e1000_ethtool.c +++ b/drivers/net/e1000/e1000_ethtool.c @@ -461,7 +461,8 @@ e1000_get_regs(struct net_device *netdev, regs_buff[24] = (uint32_t)phy_data; /* phy local receiver status */ regs_buff[25] = regs_buff[24]; /* phy remote receiver status */ if (hw->mac_type >= e1000_82540 && - hw->media_type == e1000_media_type_copper) { + hw->mac_type < e1000_82571 && + hw->media_type == e1000_media_type_copper) { regs_buff[26] = E1000_READ_REG(hw, MANC); } } diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index fa84983..fb83082 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -699,7 +699,10 @@ e1000_reset(struct e1000_adapter *adapter) phy_data); } - if ((adapter->en_mng_pt) && (adapter->hw.mac_type < e1000_82571)) { + if ((adapter->en_mng_pt) && + (adapter->hw.mac_type >= e1000_82540) && + (adapter->hw.mac_type < e1000_82571) && + (adapter->hw.media_type == e1000_media_type_copper)) { manc = E1000_READ_REG(&adapter->hw, MANC); manc |= (E1000_MANC_ARP_EN | E1000_MANC_EN_MNG2HOST); E1000_WRITE_REG(&adapter->hw, MANC, manc); @@ -1076,8 +1079,9 @@ e1000_remove(struct pci_dev *pdev) flush_scheduled_work(); - if (adapter->hw.mac_type < e1000_82571 && - adapter->hw.media_type == e1000_media_type_copper) { + if (adapter->hw.mac_type >= e1000_82540 && + adapter->hw.mac_type < e1000_82571 && + adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); if (manc & E1000_MANC_SMBUS_EN) { manc |= E1000_MANC_ARP_EN; @@ -4773,8 +4777,9 @@ e1000_suspend(struct pci_dev *pdev, pm_message_t state) pci_enable_wake(pdev, PCI_D3cold, 0); } - if (adapter->hw.mac_type < e1000_82571 && - adapter->hw.media_type == e1000_media_type_copper) { + if (adapter->hw.mac_type >= e1000_82540 && + adapter->hw.mac_type < e1000_82571 && + adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); if (manc & E1000_MANC_SMBUS_EN) { manc |= E1000_MANC_ARP_EN; @@ -4825,8 +4830,9 @@ e1000_resume(struct pci_dev *pdev) netif_device_attach(netdev); - if (adapter->hw.mac_type < e1000_82571 && - adapter->hw.media_type == e1000_media_type_copper) { + if (adapter->hw.mac_type >= e1000_82540 && + adapter->hw.mac_type < e1000_82571 && + adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); manc &= ~(E1000_MANC_ARP_EN); E1000_WRITE_REG(&adapter->hw, MANC, manc); @@ -4944,6 +4950,7 @@ static void e1000_io_resume(struct pci_dev *pdev) netif_device_attach(netdev); if (adapter->hw.mac_type >= e1000_82540 && + adapter->hw.mac_type < e1000_82571 && adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); manc &= ~(E1000_MANC_ARP_EN); -- cgit v0.10.2 From dc1f71f6b30c258704885cd488582eb3d68b3e8e Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Tue, 24 Oct 2006 14:45:55 -0700 Subject: e1000: FIX: 82542 doesn't support WoL Exclude 82542 when setting up WoL. This card does not do WoL at all. Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_ethtool.c b/drivers/net/e1000/e1000_ethtool.c index 71fb278..c564adb 100644 --- a/drivers/net/e1000/e1000_ethtool.c +++ b/drivers/net/e1000/e1000_ethtool.c @@ -1691,6 +1691,7 @@ static int e1000_wol_exclusion(struct e1000_adapter *adapter, struct ethtool_wol int retval = 1; /* fail by default */ switch (hw->device_id) { + case E1000_DEV_ID_82542: case E1000_DEV_ID_82543GC_FIBER: case E1000_DEV_ID_82543GC_COPPER: case E1000_DEV_ID_82544EI_FIBER: -- cgit v0.10.2 From 225a5dbd68f5271b7425f2f783ae64a1f6863b51 Mon Sep 17 00:00:00 2001 From: Bruce Allan Date: Tue, 24 Oct 2006 14:45:58 -0700 Subject: e1000: FIX: fix wrong txdctl threshold bitmasks Threshold bitmasks for prefetch, host and writeback were clearing bits that they were not supposed to. The leftmost 2 bits in the byte for each threshold are reserved. Signed-off-by: Bruce Allan Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_hw.h b/drivers/net/e1000/e1000_hw.h index 112447f..449a603 100644 --- a/drivers/net/e1000/e1000_hw.h +++ b/drivers/net/e1000/e1000_hw.h @@ -1961,9 +1961,9 @@ struct e1000_hw { #define E1000_RXDCTL_GRAN 0x01000000 /* RXDCTL Granularity */ /* Transmit Descriptor Control */ -#define E1000_TXDCTL_PTHRESH 0x000000FF /* TXDCTL Prefetch Threshold */ -#define E1000_TXDCTL_HTHRESH 0x0000FF00 /* TXDCTL Host Threshold */ -#define E1000_TXDCTL_WTHRESH 0x00FF0000 /* TXDCTL Writeback Threshold */ +#define E1000_TXDCTL_PTHRESH 0x0000003F /* TXDCTL Prefetch Threshold */ +#define E1000_TXDCTL_HTHRESH 0x00003F00 /* TXDCTL Host Threshold */ +#define E1000_TXDCTL_WTHRESH 0x003F0000 /* TXDCTL Writeback Threshold */ #define E1000_TXDCTL_GRAN 0x01000000 /* TXDCTL Granularity */ #define E1000_TXDCTL_LWTHRESH 0xFE000000 /* TXDCTL Low Threshold */ #define E1000_TXDCTL_FULL_TX_DESC_WB 0x01010000 /* GRAN=1, WTHRESH=1 */ -- cgit v0.10.2 From e64d7d02090e475cfd7efbc830146d0c6dd579bc Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Tue, 24 Oct 2006 14:46:01 -0700 Subject: e1000: FIX: Disable Packet Split for non jumbo frames Allocations using alloc_page are taking too long for normal MTU, so use LPE only for jumbo frames. Signed-off-bu: Jesse Brandeburg Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index fb83082..7362e12 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -1808,9 +1808,11 @@ e1000_setup_rctl(struct e1000_adapter *adapter) * followed by the page buffers. Therefore, skb->data is * sized to hold the largest protocol header. */ + /* allocations using alloc_page take too long for regular MTU + * so only enable packet split for jumbo frames */ pages = PAGE_USE_COUNT(adapter->netdev->mtu); - if ((adapter->hw.mac_type > e1000_82547_rev_2) && (pages <= 3) && - PAGE_SIZE <= 16384) + if ((adapter->hw.mac_type >= e1000_82571) && (pages <= 3) && + PAGE_SIZE <= 16384 && (rctl & E1000_RCTL_LPE)) adapter->rx_ps_pages = pages; else adapter->rx_ps_pages = 0; -- cgit v0.10.2 From 032fe6e9e253ebb37a0df0893844674dea9210fd Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Tue, 24 Oct 2006 14:46:04 -0700 Subject: e1000: FIX: Don't limit descriptor size to 4kb for PCI-E adapters 82571 and newer chispets don't need to limit desc. length to 4kb and can handle 8kb sizes. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 7362e12..e75909e 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2992,6 +2992,11 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) return NETDEV_TX_OK; } + /* 82571 and newer doesn't need the workaround that limited descriptor + * length to 4kB */ + if (adapter->hw.mac_type >= e1000_82571) + max_per_txd = 8192; + #ifdef NETIF_F_TSO mss = skb_shinfo(skb)->gso_size; /* The controller does a simple calculation to -- cgit v0.10.2 From d2a1e2131aee7b3feb815636dc7917a96e49fd8e Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Tue, 24 Oct 2006 14:46:06 -0700 Subject: e1000: FIX: move length adjustment due to crc stripping disabled. Move the length (rx_bytes counter) adjustment of 4 bytes down to after the TBI_ACCEPT workaround. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index e75909e..66ed920 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -3786,9 +3786,6 @@ e1000_clean_rx_irq(struct e1000_adapter *adapter, length = le16_to_cpu(rx_desc->length); - /* adjust length to remove Ethernet CRC */ - length -= 4; - if (unlikely(!(status & E1000_RXD_STAT_EOP))) { /* All receives must fit into a single buffer */ E1000_DBG("%s: Receive packet consumed multiple" @@ -3816,6 +3813,10 @@ e1000_clean_rx_irq(struct e1000_adapter *adapter, } } + /* adjust length to remove Ethernet CRC, this must be + * done after the TBI_ACCEPT workaround above */ + length -= 4; + /* code added for copybreak, this should improve * performance for small packets with large amounts * of reassembly being done in the stack */ -- cgit v0.10.2 From ff1e55b078676d3c449ace6b99d95c0e22c905d6 Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Tue, 24 Oct 2006 14:46:09 -0700 Subject: e1000: Increment version to 7.2.9-k4 Significant fixes -> increment driver version. Signed-off-by: Auke Kok diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 66ed920..8d04752 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -35,7 +35,7 @@ static char e1000_driver_string[] = "Intel(R) PRO/1000 Network Driver"; #else #define DRIVERNAPI "-NAPI" #endif -#define DRV_VERSION "7.2.9-k2"DRIVERNAPI +#define DRV_VERSION "7.2.9-k4"DRIVERNAPI char e1000_driver_version[] = DRV_VERSION; static char e1000_copyright[] = "Copyright (c) 1999-2006 Intel Corporation."; -- cgit v0.10.2 From 824545e7031541f83245d254caca012bf6bdc6cd Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Tue, 24 Oct 2006 14:49:44 -0700 Subject: e100: account for closed interface when shutting down Account for the interface being closed before disabling polling on a device, to fix shutdown on some systems that explcitly close the netdevice before calling shutdown. Signed-off-by: Auke Kok diff --git a/drivers/net/e100.c b/drivers/net/e100.c index a3a08a5d..19ab344 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -2719,7 +2719,10 @@ static int e100_suspend(struct pci_dev *pdev, pm_message_t state) struct net_device *netdev = pci_get_drvdata(pdev); struct nic *nic = netdev_priv(netdev); - netif_poll_disable(nic->netdev); +#ifdef CONFIG_E100_NAPI + if (netif_running(netdev)) + netif_poll_disable(nic->netdev); +#endif del_timer_sync(&nic->watchdog); netif_carrier_off(nic->netdev); @@ -2763,7 +2766,10 @@ static void e100_shutdown(struct pci_dev *pdev) struct net_device *netdev = pci_get_drvdata(pdev); struct nic *nic = netdev_priv(netdev); - netif_poll_disable(nic->netdev); +#ifdef CONFIG_E100_NAPI + if (netif_running(netdev)) + netif_poll_disable(nic->netdev); +#endif del_timer_sync(&nic->watchdog); netif_carrier_off(nic->netdev); -- cgit v0.10.2 From 82571026b9771c4035b0c62abbbe588fe73373ea Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 24 Oct 2006 15:18:36 -0700 Subject: [IPV4] ipconfig: fix RARP ic_servaddr breakage memcpy 4 bytes to address of auto unsigned long variable followed by comparison with u32 is a bloody bad idea. Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index f8ce847..955a07a 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -420,7 +420,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt { struct arphdr *rarp; unsigned char *rarp_ptr; - unsigned long sip, tip; + u32 sip, tip; unsigned char *sha, *tha; /* s for "source", t for "target" */ struct ic_device *d; -- cgit v0.10.2 From 2fab22f2d3290ff7c602fe62f22e825c48e97a06 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 24 Oct 2006 15:34:00 -0700 Subject: [XFRM]: Fix xfrm_state accounting xfrm_state_num needs to be increased for XFRM_STATE_ACQ states created by xfrm_state_find() to prevent the counter from going negative when the state is destroyed. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 84bbf84..899de9e 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -505,6 +505,14 @@ __xfrm_state_locate(struct xfrm_state *x, int use_spi, int family) x->id.proto, family); } +static void xfrm_hash_grow_check(int have_hash_collision) +{ + if (have_hash_collision && + (xfrm_state_hmask + 1) < xfrm_state_hashmax && + xfrm_state_num > xfrm_state_hmask) + schedule_work(&xfrm_hash_work); +} + struct xfrm_state * xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, struct flowi *fl, struct xfrm_tmpl *tmpl, @@ -598,6 +606,8 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, x->lft.hard_add_expires_seconds = XFRM_ACQ_EXPIRES; x->timer.expires = jiffies + XFRM_ACQ_EXPIRES*HZ; add_timer(&x->timer); + xfrm_state_num++; + xfrm_hash_grow_check(x->bydst.next != NULL); } else { x->km.state = XFRM_STATE_DEAD; xfrm_state_put(x); @@ -614,14 +624,6 @@ out: return x; } -static void xfrm_hash_grow_check(int have_hash_collision) -{ - if (have_hash_collision && - (xfrm_state_hmask + 1) < xfrm_state_hashmax && - xfrm_state_num > xfrm_state_hmask) - schedule_work(&xfrm_hash_work); -} - static void __xfrm_state_insert(struct xfrm_state *x) { unsigned int h; -- cgit v0.10.2 From d1752aa884ec0ac3027c1a3d456bf69bf765c8b8 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 25 Oct 2006 01:00:22 +0200 Subject: [PATCH] x86-64: Simplify the vector allocator. There is no reason to remember a per cpu position of which vector to try. Keeping a global position is simpler and more likely to result in a global vector allocation even if I don't need or require it. For level triggered interrupts this means we are less likely to acknowledge another cpus irq, and cause the level triggered irq to harmlessly refire. This simplification makes it easier to only access data structures of online cpus, by having fewer special cases to deal with. Signed-off-by: Eric W. Biederman Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index b000017..0e89ae7 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -612,10 +612,7 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) * Also, we've got to be careful not to trash gate * 0x80, because int 0x80 is hm, kind of importantish. ;) */ - static struct { - int vector; - int offset; - } pos[NR_CPUS] = { [ 0 ... NR_CPUS - 1] = {FIRST_DEVICE_VECTOR, 0} }; + static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0; int old_vector = -1; int cpu; @@ -631,14 +628,13 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) for_each_cpu_mask(cpu, mask) { cpumask_t domain; - int first, new_cpu; + int new_cpu; int vector, offset; domain = vector_allocation_domain(cpu); - first = first_cpu(domain); - vector = pos[first].vector; - offset = pos[first].offset; + vector = current_vector; + offset = current_offset; next: vector += 8; if (vector >= FIRST_SYSTEM_VECTOR) { @@ -646,7 +642,7 @@ next: offset = (offset + 1) % 8; vector = FIRST_DEVICE_VECTOR + offset; } - if (unlikely(pos[first].vector == vector)) + if (unlikely(current_vector == vector)) continue; if (vector == IA32_SYSCALL_VECTOR) goto next; @@ -654,10 +650,8 @@ next: if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; /* Found one! */ - for_each_cpu_mask(new_cpu, domain) { - pos[new_cpu].vector = vector; - pos[new_cpu].offset = offset; - } + current_vector = vector; + current_offset = offset; if (old_vector >= 0) { int old_cpu; for_each_cpu_mask(old_cpu, irq_domain[irq]) -- cgit v0.10.2 From 70a0a5357db20c291d46c04011d646d5d84c868c Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 25 Oct 2006 01:00:23 +0200 Subject: [PATCH] x86-64: Only look at per_cpu data for online cpus. When I generalized __assign_irq_vector I failed to pay attention to what happens when you access a per cpu data structure for a cpu that is not online. It is an undefined case making any code that does it have undefined behavior as well. The code still needs to be able to allocate a vector across cpus that are not online to properly handle combinations like lowest priority interrupt delivery and cpu_hotplug. Not that we can do that today but the infrastructure shouldn't prevent it. So this patch updates the places where we touch per cpu data to only touch online cpus, it makes cpu vector allocation an atomic operation with respect to cpu hotplug, and it updates the cpu start code to properly initialize vector_irq so we don't have inconsistencies. Signed-off-by: Eric W. Biederman Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 0e89ae7..fe429e5 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -63,7 +63,7 @@ int timer_over_8254 __initdata = 1; static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; static DEFINE_SPINLOCK(ioapic_lock); -static DEFINE_SPINLOCK(vector_lock); +DEFINE_SPINLOCK(vector_lock); /* * # of IRQ routing registers @@ -618,6 +618,9 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) BUG_ON((unsigned)irq >= NR_IRQ_VECTORS); + /* Only try and allocate irqs on cpus that are present */ + cpus_and(mask, mask, cpu_online_map); + if (irq_vector[irq] > 0) old_vector = irq_vector[irq]; if (old_vector > 0) { @@ -627,11 +630,12 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) } for_each_cpu_mask(cpu, mask) { - cpumask_t domain; + cpumask_t domain, new_mask; int new_cpu; int vector, offset; domain = vector_allocation_domain(cpu); + cpus_and(new_mask, domain, cpu_online_map); vector = current_vector; offset = current_offset; @@ -646,18 +650,20 @@ next: continue; if (vector == IA32_SYSCALL_VECTOR) goto next; - for_each_cpu_mask(new_cpu, domain) + for_each_cpu_mask(new_cpu, new_mask) if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; /* Found one! */ current_vector = vector; current_offset = offset; if (old_vector >= 0) { + cpumask_t old_mask; int old_cpu; - for_each_cpu_mask(old_cpu, irq_domain[irq]) + cpus_and(old_mask, irq_domain[irq], cpu_online_map); + for_each_cpu_mask(old_cpu, old_mask) per_cpu(vector_irq, old_cpu)[old_vector] = -1; } - for_each_cpu_mask(new_cpu, domain) + for_each_cpu_mask(new_cpu, new_mask) per_cpu(vector_irq, new_cpu)[vector] = irq; irq_vector[irq] = vector; irq_domain[irq] = domain; @@ -678,6 +684,32 @@ static int assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) return vector; } +void __setup_vector_irq(int cpu) +{ + /* Initialize vector_irq on a new cpu */ + /* This function must be called with vector_lock held */ + unsigned long flags; + int irq, vector; + + + /* Mark the inuse vectors */ + for (irq = 0; irq < NR_IRQ_VECTORS; ++irq) { + if (!cpu_isset(cpu, irq_domain[irq])) + continue; + vector = irq_vector[irq]; + per_cpu(vector_irq, cpu)[vector] = irq; + } + /* Mark the free vectors */ + for (vector = 0; vector < NR_VECTORS; ++vector) { + irq = per_cpu(vector_irq, cpu)[vector]; + if (irq < 0) + continue; + if (!cpu_isset(cpu, irq_domain[irq])) + per_cpu(vector_irq, cpu)[vector] = -1; + } +} + + extern void (*interrupt[NR_IRQS])(void); static struct irq_chip ioapic_chip; diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 7b7a687..62c2e74 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -581,12 +581,16 @@ void __cpuinit start_secondary(void) * smp_call_function(). */ lock_ipi_call_lock(); + spin_lock(&vector_lock); + /* Setup the per cpu irq handling data structures */ + __setup_vector_irq(smp_processor_id()); /* * Allow the master to continue. */ cpu_set(smp_processor_id(), cpu_online_map); per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; + spin_unlock(&vector_lock); unlock_ipi_call_lock(); cpu_idle(); @@ -799,7 +803,6 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid) cpu, node); } - alternatives_smp_switch(1); c_idle.idle = get_idle_for_cpu(cpu); @@ -1246,8 +1249,10 @@ int __cpu_disable(void) local_irq_disable(); remove_siblinginfo(cpu); + spin_lock(&vector_lock); /* It's now safe to remove this processor from the online map */ cpu_clear(cpu, cpu_online_map); + spin_unlock(&vector_lock); remove_cpu_from_maps(); fixup_irqs(cpu_online_map); return 0; diff --git a/include/asm-x86_64/hw_irq.h b/include/asm-x86_64/hw_irq.h index 792dd52..179cce7 100644 --- a/include/asm-x86_64/hw_irq.h +++ b/include/asm-x86_64/hw_irq.h @@ -76,6 +76,8 @@ #ifndef __ASSEMBLY__ typedef int vector_irq_t[NR_VECTORS]; DECLARE_PER_CPU(vector_irq_t, vector_irq); +extern void __setup_vector_irq(int cpu); +extern spinlock_t vector_lock; /* * Various low-level irq details needed by irq.c, process.c, -- cgit v0.10.2 From 51d8b1a65291a6956b79374b6adbbadc2263bcf6 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 24 Oct 2006 16:14:04 -0700 Subject: [NETFILTER]: Fix ip6_tables protocol bypass bug As reported by Mark Dowd , ip6_tables is susceptible to a fragmentation attack causing false negatives on protocol matches. When the protocol header doesn't follow the fragment header immediately, the fragment header contains the protocol number of the next extension header. When the extension header and the protocol header are sent in a second fragment a rule like "ip6tables .. -p udp -j DROP" will never match. Drop fragments that are at offset 0 and don't contain the final protocol header regardless of the ruleset, since this should not happen normally. With help from Yasuyuki KOZAKAI . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 4ab368f..f0328c7 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -111,7 +111,7 @@ ip6_packet_match(const struct sk_buff *skb, const char *outdev, const struct ip6t_ip6 *ip6info, unsigned int *protoff, - int *fragoff) + int *fragoff, int *hotdrop) { size_t i; unsigned long ret; @@ -169,9 +169,11 @@ ip6_packet_match(const struct sk_buff *skb, unsigned short _frag_off; protohdr = ipv6_find_hdr(skb, protoff, -1, &_frag_off); - if (protohdr < 0) + if (protohdr < 0) { + if (_frag_off == 0) + *hotdrop = 1; return 0; - + } *fragoff = _frag_off; dprintf("Packet protocol %hi ?= %s%hi.\n", @@ -290,7 +292,7 @@ ip6t_do_table(struct sk_buff **pskb, IP_NF_ASSERT(e); IP_NF_ASSERT(back); if (ip6_packet_match(*pskb, indev, outdev, &e->ipv6, - &protoff, &offset)) { + &protoff, &offset, &hotdrop)) { struct ip6t_entry_target *t; if (IP6T_MATCH_ITERATE(e, do_match, -- cgit v0.10.2 From 6d381634d213580d40d431e7664dfb45f641b884 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 24 Oct 2006 16:15:10 -0700 Subject: [NETFILTER]: Fix ip6_tables extension header bypass bug As reported by Mark Dowd , ip6_tables is susceptible to a fragmentation attack causing false negatives on extension header matches. When extension headers occur in the non-first fragment after the fragment header (possibly with an incorrect nexthdr value in the fragment header) a rule looking for this extension header will never match. Drop fragments that are at offset 0 and don't contain the final protocol header regardless of the ruleset, since this should not happen normally. Since all extension headers are before the protocol header this makes sure an extension header is either not present or in the first fragment, where we can properly parse it. With help from Yasuyuki KOZAKAI . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index f0328c7..53bf977 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1440,6 +1440,9 @@ static void __exit ip6_tables_fini(void) * If target header is found, its offset is set in *offset and return protocol * number. Otherwise, return -1. * + * If the first fragment doesn't contain the final protocol header or + * NEXTHDR_NONE it is considered invalid. + * * Note that non-1st fragment is special case that "the protocol number * of last header" is "next header" field in Fragment header. In this case, * *offset is meaningless and fragment offset is stored in *fragoff if fragoff @@ -1463,12 +1466,12 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, if ((!ipv6_ext_hdr(nexthdr)) || nexthdr == NEXTHDR_NONE) { if (target < 0) break; - return -1; + return -ENOENT; } hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr); if (hp == NULL) - return -1; + return -EBADMSG; if (nexthdr == NEXTHDR_FRAGMENT) { unsigned short _frag_off, *fp; fp = skb_header_pointer(skb, @@ -1477,7 +1480,7 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, sizeof(_frag_off), &_frag_off); if (fp == NULL) - return -1; + return -EBADMSG; _frag_off = ntohs(*fp) & ~0x7; if (_frag_off) { @@ -1488,7 +1491,7 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, *fragoff = _frag_off; return hp->nexthdr; } - return -1; + return -ENOENT; } hdrlen = 8; } else if (nexthdr == NEXTHDR_AUTH) diff --git a/net/ipv6/netfilter/ip6t_ah.c b/net/ipv6/netfilter/ip6t_ah.c index ec1b160..4648664 100644 --- a/net/ipv6/netfilter/ip6t_ah.c +++ b/net/ipv6/netfilter/ip6t_ah.c @@ -54,9 +54,14 @@ match(const struct sk_buff *skb, const struct ip6t_ah *ahinfo = matchinfo; unsigned int ptr; unsigned int hdrlen = 0; + int err; - if (ipv6_find_hdr(skb, &ptr, NEXTHDR_AUTH, NULL) < 0) + err = ipv6_find_hdr(skb, &ptr, NEXTHDR_AUTH, NULL); + if (err < 0) { + if (err != -ENOENT) + *hotdrop = 1; return 0; + } ah = skb_header_pointer(skb, ptr, sizeof(_ah), &_ah); if (ah == NULL) { diff --git a/net/ipv6/netfilter/ip6t_frag.c b/net/ipv6/netfilter/ip6t_frag.c index 78d9c8b..cd22eaa 100644 --- a/net/ipv6/netfilter/ip6t_frag.c +++ b/net/ipv6/netfilter/ip6t_frag.c @@ -52,9 +52,14 @@ match(const struct sk_buff *skb, struct frag_hdr _frag, *fh; const struct ip6t_frag *fraginfo = matchinfo; unsigned int ptr; + int err; - if (ipv6_find_hdr(skb, &ptr, NEXTHDR_FRAGMENT, NULL) < 0) + err = ipv6_find_hdr(skb, &ptr, NEXTHDR_FRAGMENT, NULL); + if (err < 0) { + if (err != -ENOENT) + *hotdrop = 1; return 0; + } fh = skb_header_pointer(skb, ptr, sizeof(_frag), &_frag); if (fh == NULL) { diff --git a/net/ipv6/netfilter/ip6t_hbh.c b/net/ipv6/netfilter/ip6t_hbh.c index d32a205..3f25bab 100644 --- a/net/ipv6/netfilter/ip6t_hbh.c +++ b/net/ipv6/netfilter/ip6t_hbh.c @@ -65,9 +65,14 @@ match(const struct sk_buff *skb, u8 _opttype, *tp = NULL; u8 _optlen, *lp = NULL; unsigned int optlen; + int err; - if (ipv6_find_hdr(skb, &ptr, match->data, NULL) < 0) + err = ipv6_find_hdr(skb, &ptr, match->data, NULL); + if (err < 0) { + if (err != -ENOENT) + *hotdrop = 1; return 0; + } oh = skb_header_pointer(skb, ptr, sizeof(_optsh), &_optsh); if (oh == NULL) { diff --git a/net/ipv6/netfilter/ip6t_rt.c b/net/ipv6/netfilter/ip6t_rt.c index bcb2e16..54d7d14 100644 --- a/net/ipv6/netfilter/ip6t_rt.c +++ b/net/ipv6/netfilter/ip6t_rt.c @@ -58,9 +58,14 @@ match(const struct sk_buff *skb, unsigned int hdrlen = 0; unsigned int ret = 0; struct in6_addr *ap, _addr; + int err; - if (ipv6_find_hdr(skb, &ptr, NEXTHDR_ROUTING, NULL) < 0) + err = ipv6_find_hdr(skb, &ptr, NEXTHDR_ROUTING, NULL); + if (err < 0) { + if (err != -ENOENT) + *hotdrop = 1; return 0; + } rh = skb_header_pointer(skb, ptr, sizeof(_route), &_route); if (rh == NULL) { -- cgit v0.10.2 From 977a415f2b70b5693aaa23b1a16ad57ea20a1dce Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 24 Oct 2006 16:16:39 -0700 Subject: [ATM] horizon: read_bia() needs to be __devinit Thanks to Randy Dunlap. Signed-off-by: David S. Miller diff --git a/drivers/atm/horizon.c b/drivers/atm/horizon.c index 44268cb..4dc1010 100644 --- a/drivers/atm/horizon.c +++ b/drivers/atm/horizon.c @@ -1789,7 +1789,7 @@ static inline void CLOCK_IT (const hrz_dev *dev, u32 ctrl) WRITE_IT_WAIT(dev, ctrl | SEEPROM_SK); } -static u16 __init read_bia (const hrz_dev * dev, u16 addr) +static u16 __devinit read_bia (const hrz_dev * dev, u16 addr) { u32 ctrl = rd_regl (dev, CONTROL_0_REG); -- cgit v0.10.2 From 0e64e94e477f8ed04e9295b11a5898d443c28a47 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 24 Oct 2006 16:17:51 -0700 Subject: [DCCP]: Update documentation references. Updates the references to spec documents throughout the code, taking into account that * the DCCP, CCID 2, and CCID 3 drafts all became RFCs in March this year * RFC 1063 was obsoleted by RFC 1191 * draft-ietf-tcpimpl-pmtud-0x.txt was published as an Informational RFC, RFC 2923 on 2000-09-22. All references verified. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller diff --git a/net/dccp/Kconfig b/net/dccp/Kconfig index e2a095d..ef8919c 100644 --- a/net/dccp/Kconfig +++ b/net/dccp/Kconfig @@ -4,15 +4,15 @@ menu "DCCP Configuration (EXPERIMENTAL)" config IP_DCCP tristate "The DCCP Protocol (EXPERIMENTAL)" ---help--- - Datagram Congestion Control Protocol + Datagram Congestion Control Protocol (RFC 4340) - From draft-ietf-dccp-spec-11 . + From http://www.ietf.org/rfc/rfc4340.txt: The Datagram Congestion Control Protocol (DCCP) is a transport protocol that implements bidirectional, unicast connections of congestion-controlled, unreliable datagrams. It should be suitable for use by applications such as streaming media, Internet telephony, - and on-line games + and on-line games. To compile this protocol support as a module, choose M here: the module will be called dccp. diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index 4d176d3..f820887 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -113,7 +113,7 @@ int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb) memcpy(to, from, len); /* - * From draft-ietf-dccp-spec-11.txt: + * From RFC 4340, A.2: * * For each acknowledgement it sends, the HC-Receiver will add an * acknowledgement record. ack_seqno will equal the HC-Receiver @@ -224,7 +224,7 @@ static inline int dccp_ackvec_set_buf_head_state(struct dccp_ackvec *av, } /* - * Implements the draft-ietf-dccp-spec-11.txt Appendix A + * Implements the RFC 4340, Appendix A */ int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, const u64 ackno, const u8 state) @@ -237,7 +237,7 @@ int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, * We may well decide to do buffer compression, etc, but for now lets * just drop. * - * From Appendix A: + * From Appendix A.1.1 (`New Packets'): * * Of course, the circular buffer may overflow, either when the * HC-Sender is sending data at a very high rate, when the @@ -274,9 +274,9 @@ int dccp_ackvec_add(struct dccp_ackvec *av, const struct sock *sk, /* * A.1.2. Old Packets * - * When a packet with Sequence Number S arrives, and - * S <= buf_ackno, the HC-Receiver will scan the table - * for the byte corresponding to S. (Indexing structures + * When a packet with Sequence Number S <= buf_ackno + * arrives, the HC-Receiver will scan the table for + * the byte corresponding to S. (Indexing structures * could reduce the complexity of this scan.) */ u64 delta = dccp_delta_seqno(ackno, av->dccpav_buf_ackno); diff --git a/net/dccp/ackvec.h b/net/dccp/ackvec.h index 2424eff..cf8f20c 100644 --- a/net/dccp/ackvec.h +++ b/net/dccp/ackvec.h @@ -28,8 +28,7 @@ /** struct dccp_ackvec - ack vector * - * This data structure is the one defined in the DCCP draft - * Appendix A. + * This data structure is the one defined in RFC 4340, Appendix A. * * @dccpav_buf_head - circular buffer head * @dccpav_buf_tail - circular buffer tail diff --git a/net/dccp/ccids/Kconfig b/net/dccp/ccids/Kconfig index 32752f7..8533dab 100644 --- a/net/dccp/ccids/Kconfig +++ b/net/dccp/ccids/Kconfig @@ -22,11 +22,11 @@ config IP_DCCP_CCID2 for lost packets, would prefer CCID 2 to CCID 3. On-line games may also prefer CCID 2. - CCID 2 is further described in: - http://www.icir.org/kohler/dccp/draft-ietf-dccp-ccid2-10.txt + CCID 2 is further described in RFC 4341, + http://www.ietf.org/rfc/rfc4341.txt - This text was extracted from: - http://www.icir.org/kohler/dccp/draft-ietf-dccp-spec-13.txt + This text was extracted from RFC 4340 (sec. 10.1), + http://www.ietf.org/rfc/rfc4340.txt If in doubt, say M. @@ -53,15 +53,14 @@ config IP_DCCP_CCID3 suitable than CCID 2 for applications such streaming media where a relatively smooth sending rate is of importance. - CCID 3 is further described in: - - http://www.icir.org/kohler/dccp/draft-ietf-dccp-ccid3-11.txt. + CCID 3 is further described in RFC 4342, + http://www.ietf.org/rfc/rfc4342.txt The TFRC congestion control algorithms were initially described in RFC 3448. - This text was extracted from: - http://www.icir.org/kohler/dccp/draft-ietf-dccp-spec-13.txt + This text was extracted from RFC 4340 (sec. 10.2), + http://www.ietf.org/rfc/rfc4340.txt If in doubt, say M. diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 2efb505..2fbb84b 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -23,7 +23,7 @@ */ /* - * This implementation should follow: draft-ietf-dccp-ccid2-10.txt + * This implementation should follow RFC 4341 * * BUGS: * - sequence number wrapping diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 67d2dc0..cec23ad 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -379,8 +379,7 @@ static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, int len) packet->dccphtx_seqno = dp->dccps_gss; /* * Check if win_count have changed - * Algorithm in "8.1. Window Counter Valuer" in - * draft-ietf-dccp-ccid3-11.txt + * Algorithm in "8.1. Window Counter Value" in RFC 4342. */ quarter_rtt = timeval_delta(&now, &hctx->ccid3hctx_t_last_win_count); if (likely(hctx->ccid3hctx_rtt > 8)) diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 0a21be4..272e858 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -50,7 +50,7 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo); #define DCCP_TIMEWAIT_LEN (60 * HZ) /* how long to wait to destroy TIME-WAIT * state, about 60 seconds */ -/* draft-ietf-dccp-spec-11.txt initial RTO value */ +/* RFC 1122, 4.2.3.1 initial RTO value */ #define DCCP_TIMEOUT_INIT ((unsigned)(3 * HZ)) /* Maximal interval between probes for local resources. */ diff --git a/net/dccp/input.c b/net/dccp/input.c index 7f9dc6a..1d24881 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -216,11 +216,11 @@ send_sync: dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq, DCCP_PKT_SYNCACK); /* - * From the draft: + * From RFC 4340, sec. 5.7 * * As with DCCP-Ack packets, DCCP-Sync and DCCP-SyncAck packets * MAY have non-zero-length application data areas, whose - * contents * receivers MUST ignore. + * contents receivers MUST ignore. */ goto discard; } diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index aaaf4d0..e08e768 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -183,7 +183,7 @@ static inline void dccp_do_pmtu_discovery(struct sock *sk, dccp_sync_mss(sk, mtu); /* - * From: draft-ietf-dccp-spec-11.txt + * From RFC 4340, sec. 14.1: * * DCCP-Sync packets are the best choice for upward * probing, since DCCP-Sync probes do not risk application @@ -733,7 +733,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) dccp_hdr_reset(skb)->dccph_reset_code = DCCP_SKB_CB(rxskb)->dccpd_reset_code; - /* See "8.3.1. Abnormal Termination" in draft-ietf-dccp-spec-11 */ + /* See "8.3.1. Abnormal Termination" in RFC 4340 */ seqno = 0; if (DCCP_SKB_CB(rxskb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_set_seqno(&seqno, DCCP_SKB_CB(rxskb)->dccpd_ack_seq + 1); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index c8bf89b..eb0ff7a 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -550,7 +550,7 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) dccp_hdr_reset(skb)->dccph_reset_code = DCCP_SKB_CB(rxskb)->dccpd_reset_code; - /* See "8.3.1. Abnormal Termination" in draft-ietf-dccp-spec-11 */ + /* See "8.3.1. Abnormal Termination" in RFC 4340 */ seqno = 0; if (DCCP_SKB_CB(rxskb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_set_seqno(&seqno, DCCP_SKB_CB(rxskb)->dccpd_ack_seq + 1); diff --git a/net/dccp/options.c b/net/dccp/options.c index 07a3469..fb0db1f 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -215,7 +215,7 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) elapsed_time); break; /* - * From draft-ietf-dccp-spec-11.txt: + * From RFC 4340, sec. 10.3: * * Option numbers 128 through 191 are for * options sent from the HC-Sender to the -- cgit v0.10.2 From fb20f65a01a97bdf4bb746eecfc24a08561e2648 Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Fri, 20 Oct 2006 15:57:15 +0200 Subject: [POWERPC] Fix hang in start_ldr if _end or _edata is unaligned Quick fix for lack of memset(__bss_start, 0, _end-__bss_start) in load_kernel(). If edata is unaligned, the loop will overwrite all memory because r3 and r4 will never be equal. Signed-off-by: Olaf Hering Signed-off-by: Paul Mackerras diff --git a/arch/ppc/boot/simple/relocate.S b/arch/ppc/boot/simple/relocate.S index 0c02155..1bbbcd2 100644 --- a/arch/ppc/boot/simple/relocate.S +++ b/arch/ppc/boot/simple/relocate.S @@ -154,8 +154,8 @@ do_relocate_out: start_ldr: /* Clear all of BSS and set up stack for C calls */ - lis r3,edata@h - ori r3,r3,edata@l + lis r3,__bss_start@h + ori r3,r3,__bss_start@l lis r4,end@h ori r4,r4,end@l subi r3,r3,4 @@ -163,7 +163,7 @@ start_ldr: li r0,0 50: stwu r0,4(r3) cmpw cr0,r3,r4 - bne 50b + blt 50b 90: mr r9,r1 /* Save old stack pointer (in case it matters) */ lis r1,.stack@h ori r1,r1,.stack@l -- cgit v0.10.2 From 42c4aaadb737e0e672b3fb86b2c41ff59f0fb8bc Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 24 Oct 2006 16:42:40 +1000 Subject: [POWERPC] Consolidate feature fixup code There are currently two versions of the functions for applying the feature fixups, one for CPU features and one for firmware features. In addition, they are both in assembly and with separate implementations for 32 and 64 bits. identify_cpu() is also implemented in assembly and separately for 32 and 64 bits. This patch replaces them with a pair of C functions. The call sites are slightly moved on ppc64 as well to be called from C instead of from assembly, though it's a very small change, and thus shouldn't cause any problem. Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index f23aad6..6fdfaa4 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -18,6 +18,7 @@ #include #include +#include /* for PTRRELOC on ARCH=ppc */ struct cpu_spec* cur_cpu_spec = NULL; EXPORT_SYMBOL(cur_cpu_spec); @@ -73,7 +74,7 @@ extern void __restore_cpu_ppc970(void); #define PPC_FEATURE_SPE_COMP 0 #endif -struct cpu_spec cpu_specs[] = { +static struct cpu_spec cpu_specs[] = { #ifdef CONFIG_PPC64 { /* Power3 */ .pvr_mask = 0xffff0000, @@ -1167,3 +1168,72 @@ struct cpu_spec cpu_specs[] = { #endif /* !CLASSIC_PPC */ #endif /* CONFIG_PPC32 */ }; + +struct cpu_spec *identify_cpu(unsigned long offset) +{ + struct cpu_spec *s = cpu_specs; + struct cpu_spec **cur = &cur_cpu_spec; + unsigned int pvr = mfspr(SPRN_PVR); + int i; + + s = PTRRELOC(s); + cur = PTRRELOC(cur); + + if (*cur != NULL) + return PTRRELOC(*cur); + + for (i = 0; i < ARRAY_SIZE(cpu_specs); i++,s++) + if ((pvr & s->pvr_mask) == s->pvr_value) { + *cur = cpu_specs + i; +#ifdef CONFIG_PPC64 + /* ppc64 expects identify_cpu to also call setup_cpu + * for that processor. I will consolidate that at a + * later time, for now, just use our friend #ifdef. + * we also don't need to PTRRELOC the function pointer + * on ppc64 as we are running at 0 in real mode. + */ + if (s->cpu_setup) { + s->cpu_setup(offset, s); + } +#endif /* CONFIG_PPC64 */ + return s; + } + BUG(); + return NULL; +} + +void do_feature_fixups(unsigned long offset, unsigned long value, + void *fixup_start, void *fixup_end) +{ + struct fixup_entry { + unsigned long mask; + unsigned long value; + unsigned int *start; + unsigned int *end; + } *fcur, *fend; + + fcur = fixup_start; + fend = fixup_end; + + for (; fcur < fend; fcur++) { + unsigned int *pstart, *pend, *p; + + if ((value & fcur->mask) == fcur->value) + continue; + + /* These PTRRELOCs will disappear once the new scheme for + * modules and vdso is implemented + */ + pstart = PTRRELOC(fcur->start); + pend = PTRRELOC(fcur->end); + + for (p = pstart; p < pend; p++) { + *p = 0x60000000u; + asm volatile ("dcbst 0, %0" : : "r" (p)); + } + asm volatile ("sync" : : : "memory"); + for (p = pstart; p < pend; p++) + asm volatile ("icbi 0,%0" : : "r" (p)); + asm volatile ("sync; isync" : : : "memory"); + } +} diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 645c7f1..f12e3c5 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -1580,11 +1580,6 @@ _STATIC(__start_initialization_iSeries) li r0,0 stdu r0,-STACK_FRAME_OVERHEAD(r1) - LOAD_REG_IMMEDIATE(r3,cpu_specs) - LOAD_REG_IMMEDIATE(r4,cur_cpu_spec) - li r5,0 - bl .identify_cpu - LOAD_REG_IMMEDIATE(r2,__toc_start) addi r2,r2,0x4000 addi r2,r2,0x4000 @@ -1964,13 +1959,6 @@ _STATIC(start_here_multiplatform) addi r2,r2,0x4000 add r2,r2,r26 - LOAD_REG_IMMEDIATE(r3, cpu_specs) - add r3,r3,r26 - LOAD_REG_IMMEDIATE(r4,cur_cpu_spec) - add r4,r4,r26 - mr r5,r26 - bl .identify_cpu - /* Do very early kernel initializations, including initial hash table, * stab and slb setup before we turn on relocation. */ @@ -2000,13 +1988,6 @@ _STATIC(start_here_common) li r0,0 stdu r0,-STACK_FRAME_OVERHEAD(r1) - /* Apply the CPUs-specific fixups (nop out sections not relevant - * to this CPU - */ - li r3,0 - bl .do_cpu_ftr_fixups - bl .do_fw_ftr_fixups - /* ptr to current */ LOAD_REG_IMMEDIATE(r4, init_task) std r4,PACACURRENT(r13) diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 88fd73f..412bea3 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -102,80 +102,6 @@ _GLOBAL(reloc_got2) blr /* - * identify_cpu, - * called with r3 = data offset and r4 = CPU number - * doesn't change r3 - */ -_GLOBAL(identify_cpu) - addis r8,r3,cpu_specs@ha - addi r8,r8,cpu_specs@l - mfpvr r7 -1: - lwz r5,CPU_SPEC_PVR_MASK(r8) - and r5,r5,r7 - lwz r6,CPU_SPEC_PVR_VALUE(r8) - cmplw 0,r6,r5 - beq 1f - addi r8,r8,CPU_SPEC_ENTRY_SIZE - b 1b -1: - addis r6,r3,cur_cpu_spec@ha - addi r6,r6,cur_cpu_spec@l - sub r8,r8,r3 - stw r8,0(r6) - blr - -/* - * do_cpu_ftr_fixups - goes through the list of CPU feature fixups - * and writes nop's over sections of code that don't apply for this cpu. - * r3 = data offset (not changed) - */ -_GLOBAL(do_cpu_ftr_fixups) - /* Get CPU 0 features */ - addis r6,r3,cur_cpu_spec@ha - addi r6,r6,cur_cpu_spec@l - lwz r4,0(r6) - add r4,r4,r3 - lwz r4,CPU_SPEC_FEATURES(r4) - - /* Get the fixup table */ - addis r6,r3,__start___ftr_fixup@ha - addi r6,r6,__start___ftr_fixup@l - addis r7,r3,__stop___ftr_fixup@ha - addi r7,r7,__stop___ftr_fixup@l - - /* Do the fixup */ -1: cmplw 0,r6,r7 - bgelr - addi r6,r6,16 - lwz r8,-16(r6) /* mask */ - and r8,r8,r4 - lwz r9,-12(r6) /* value */ - cmplw 0,r8,r9 - beq 1b - lwz r8,-8(r6) /* section begin */ - lwz r9,-4(r6) /* section end */ - subf. r9,r8,r9 - beq 1b - /* write nops over the section of code */ - /* todo: if large section, add a branch at the start of it */ - srwi r9,r9,2 - mtctr r9 - add r8,r8,r3 - lis r0,0x60000000@h /* nop */ -3: stw r0,0(r8) - andi. r10,r4,CPU_FTR_SPLIT_ID_CACHE@l - beq 2f - dcbst 0,r8 /* suboptimal, but simpler */ - sync - icbi 0,r8 -2: addi r8,r8,4 - bdnz 3b - sync /* additional sync needed on g4 */ - isync - b 1b - -/* * call_setup_cpu - call the setup_cpu function for this cpu * r3 = data offset, r24 = cpu number * diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S index c70e207..21fd2c6 100644 --- a/arch/powerpc/kernel/misc_64.S +++ b/arch/powerpc/kernel/misc_64.S @@ -246,130 +246,6 @@ _GLOBAL(__flush_dcache_icache) isync blr -/* - * identify_cpu and calls setup_cpu - * In: r3 = base of the cpu_specs array - * r4 = address of cur_cpu_spec - * r5 = relocation offset - */ -_GLOBAL(identify_cpu) - mfpvr r7 -1: - lwz r8,CPU_SPEC_PVR_MASK(r3) - and r8,r8,r7 - lwz r9,CPU_SPEC_PVR_VALUE(r3) - cmplw 0,r9,r8 - beq 1f - addi r3,r3,CPU_SPEC_ENTRY_SIZE - b 1b -1: - sub r0,r3,r5 - std r0,0(r4) - ld r4,CPU_SPEC_SETUP(r3) - cmpdi 0,r4,0 - add r4,r4,r5 - beqlr - ld r4,0(r4) - add r4,r4,r5 - mtctr r4 - /* Calling convention for cpu setup is r3=offset, r4=cur_cpu_spec */ - mr r4,r3 - mr r3,r5 - bctr - -/* - * do_cpu_ftr_fixups - goes through the list of CPU feature fixups - * and writes nop's over sections of code that don't apply for this cpu. - * r3 = data offset (not changed) - */ -_GLOBAL(do_cpu_ftr_fixups) - /* Get CPU 0 features */ - LOAD_REG_IMMEDIATE(r6,cur_cpu_spec) - sub r6,r6,r3 - ld r4,0(r6) - sub r4,r4,r3 - ld r4,CPU_SPEC_FEATURES(r4) - /* Get the fixup table */ - LOAD_REG_IMMEDIATE(r6,__start___ftr_fixup) - sub r6,r6,r3 - LOAD_REG_IMMEDIATE(r7,__stop___ftr_fixup) - sub r7,r7,r3 - /* Do the fixup */ -1: cmpld r6,r7 - bgelr - addi r6,r6,32 - ld r8,-32(r6) /* mask */ - and r8,r8,r4 - ld r9,-24(r6) /* value */ - cmpld r8,r9 - beq 1b - ld r8,-16(r6) /* section begin */ - ld r9,-8(r6) /* section end */ - subf. r9,r8,r9 - beq 1b - /* write nops over the section of code */ - /* todo: if large section, add a branch at the start of it */ - srwi r9,r9,2 - mtctr r9 - sub r8,r8,r3 - lis r0,0x60000000@h /* nop */ -3: stw r0,0(r8) - andi. r10,r4,CPU_FTR_SPLIT_ID_CACHE@l - beq 2f - dcbst 0,r8 /* suboptimal, but simpler */ - sync - icbi 0,r8 -2: addi r8,r8,4 - bdnz 3b - sync /* additional sync needed on g4 */ - isync - b 1b - -/* - * do_fw_ftr_fixups - goes through the list of firmware feature fixups - * and writes nop's over sections of code that don't apply for this firmware. - * r3 = data offset (not changed) - */ -_GLOBAL(do_fw_ftr_fixups) - /* Get firmware features */ - LOAD_REG_IMMEDIATE(r6,powerpc_firmware_features) - sub r6,r6,r3 - ld r4,0(r6) - /* Get the fixup table */ - LOAD_REG_IMMEDIATE(r6,__start___fw_ftr_fixup) - sub r6,r6,r3 - LOAD_REG_IMMEDIATE(r7,__stop___fw_ftr_fixup) - sub r7,r7,r3 - /* Do the fixup */ -1: cmpld r6,r7 - bgelr - addi r6,r6,32 - ld r8,-32(r6) /* mask */ - and r8,r8,r4 - ld r9,-24(r6) /* value */ - cmpld r8,r9 - beq 1b - ld r8,-16(r6) /* section begin */ - ld r9,-8(r6) /* section end */ - subf. r9,r8,r9 - beq 1b - /* write nops over the section of code */ - /* todo: if large section, add a branch at the start of it */ - srwi r9,r9,2 - mtctr r9 - sub r8,r8,r3 - lis r0,0x60000000@h /* nop */ -3: stw r0,0(r8) -BEGIN_FTR_SECTION - dcbst 0,r8 /* suboptimal, but simpler */ - sync - icbi 0,r8 -END_FTR_SECTION_IFSET(CPU_FTR_SPLIT_ID_CACHE) - addi r8,r8,4 - bdnz 3b - sync /* additional sync needed on g4 */ - isync - b 1b #if defined(CONFIG_PPC_PMAC) || defined(CONFIG_PPC_MAPLE) /* diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index 191d0ab..769e511 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -91,6 +91,7 @@ int ucache_bsize; unsigned long __init early_init(unsigned long dt_ptr) { unsigned long offset = reloc_offset(); + struct cpu_spec *spec; /* First zero the BSS -- use memset_io, some platforms don't have * caches on yet */ @@ -100,8 +101,11 @@ unsigned long __init early_init(unsigned long dt_ptr) * Identify the CPU type and fix up code sections * that depend on which cpu we have. */ - identify_cpu(offset, 0); - do_cpu_ftr_fixups(offset); + spec = identify_cpu(offset); + + do_feature_fixups(offset, spec->cpu_features, + PTRRELOC(&__start___ftr_fixup), + PTRRELOC(&__stop___ftr_fixup)); return KERNELBASE + offset; } diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 4b2e32e..1969b56 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -170,6 +170,9 @@ void __init setup_paca(int cpu) void __init early_setup(unsigned long dt_ptr) { + /* Identify CPU type */ + identify_cpu(0); + /* Assume we're on cpu 0 for now. Don't write to the paca yet! */ setup_paca(0); @@ -348,6 +351,14 @@ void __init setup_system(void) { DBG(" -> setup_system()\n"); + /* Apply the CPUs-specific and firmware specific fixups to kernel + * text (nop out sections not relevant to this CPU or this firmware) + */ + do_feature_fixups(0, cur_cpu_spec->cpu_features, + &__start___ftr_fixup, &__stop___ftr_fixup); + do_feature_fixups(0, powerpc_firmware_features, + &__start___fw_ftr_fixup, &__stop___fw_ftr_fixup); + /* * Unflatten the device-tree passed by prom_init or kexec */ diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c index a0ff7ba..6f73469 100644 --- a/arch/powerpc/platforms/iseries/setup.c +++ b/arch/powerpc/platforms/iseries/setup.c @@ -694,6 +694,11 @@ void * __init iSeries_early_setup(void) { unsigned long phys_mem_size; + /* Identify CPU type. This is done again by the common code later + * on but calling this function multiple times is fine. + */ + identify_cpu(0); + powerpc_firmware_features |= FW_FEATURE_ISERIES; powerpc_firmware_features |= FW_FEATURE_LPAR; diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S index 5f66840..d319f9b 100644 --- a/arch/ppc/kernel/misc.S +++ b/arch/ppc/kernel/misc.S @@ -110,80 +110,6 @@ _GLOBAL(reloc_got2) blr /* - * identify_cpu, - * called with r3 = data offset and r4 = CPU number - * doesn't change r3 - */ -_GLOBAL(identify_cpu) - addis r8,r3,cpu_specs@ha - addi r8,r8,cpu_specs@l - mfpvr r7 -1: - lwz r5,CPU_SPEC_PVR_MASK(r8) - and r5,r5,r7 - lwz r6,CPU_SPEC_PVR_VALUE(r8) - cmplw 0,r6,r5 - beq 1f - addi r8,r8,CPU_SPEC_ENTRY_SIZE - b 1b -1: - addis r6,r3,cur_cpu_spec@ha - addi r6,r6,cur_cpu_spec@l - sub r8,r8,r3 - stw r8,0(r6) - blr - -/* - * do_cpu_ftr_fixups - goes through the list of CPU feature fixups - * and writes nop's over sections of code that don't apply for this cpu. - * r3 = data offset (not changed) - */ -_GLOBAL(do_cpu_ftr_fixups) - /* Get CPU 0 features */ - addis r6,r3,cur_cpu_spec@ha - addi r6,r6,cur_cpu_spec@l - lwz r4,0(r6) - add r4,r4,r3 - lwz r4,CPU_SPEC_FEATURES(r4) - - /* Get the fixup table */ - addis r6,r3,__start___ftr_fixup@ha - addi r6,r6,__start___ftr_fixup@l - addis r7,r3,__stop___ftr_fixup@ha - addi r7,r7,__stop___ftr_fixup@l - - /* Do the fixup */ -1: cmplw 0,r6,r7 - bgelr - addi r6,r6,16 - lwz r8,-16(r6) /* mask */ - and r8,r8,r4 - lwz r9,-12(r6) /* value */ - cmplw 0,r8,r9 - beq 1b - lwz r8,-8(r6) /* section begin */ - lwz r9,-4(r6) /* section end */ - subf. r9,r8,r9 - beq 1b - /* write nops over the section of code */ - /* todo: if large section, add a branch at the start of it */ - srwi r9,r9,2 - mtctr r9 - add r8,r8,r3 - lis r0,0x60000000@h /* nop */ -3: stw r0,0(r8) - andi. r10,r4,CPU_FTR_SPLIT_ID_CACHE@l - beq 2f - dcbst 0,r8 /* suboptimal, but simpler */ - sync - icbi 0,r8 -2: addi r8,r8,4 - bdnz 3b - sync /* additional sync needed on g4 */ - isync - b 1b - -/* * call_setup_cpu - call the setup_cpu function for this cpu * r3 = data offset, r24 = cpu number * diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c index 75fe138..41a640f 100644 --- a/arch/ppc/kernel/setup.c +++ b/arch/ppc/kernel/setup.c @@ -38,6 +38,7 @@ #include #include #include +#include #define USES_PPC_SYS (defined(CONFIG_85xx) || defined(CONFIG_83xx) || \ defined(CONFIG_MPC10X_BRIDGE) || defined(CONFIG_8260) || \ @@ -53,8 +54,6 @@ extern void platform_init(unsigned long r3, unsigned long r4, unsigned long r5, unsigned long r6, unsigned long r7); -extern void identify_cpu(unsigned long offset, unsigned long cpu); -extern void do_cpu_ftr_fixups(unsigned long offset); extern void reloc_got2(unsigned long offset); extern void ppc6xx_idle(void); @@ -301,6 +300,7 @@ early_init(int r3, int r4, int r5) { unsigned long phys; unsigned long offset = reloc_offset(); + struct cpu_spec *spec; /* Default */ phys = offset + KERNELBASE; @@ -313,8 +313,10 @@ early_init(int r3, int r4, int r5) * Identify the CPU type and fix up code sections * that depend on which cpu we have. */ - identify_cpu(offset, 0); - do_cpu_ftr_fixups(offset); + spec = identify_cpu(offset); + do_feature_fixups(offset, spec->cpu_features, + PTRRELOC(&__start___ftr_fixup), + PTRRELOC(&__stop___ftr_fixup)); return phys; } diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h index 12707ab..4d22218 100644 --- a/include/asm-powerpc/cputable.h +++ b/include/asm-powerpc/cputable.h @@ -89,8 +89,11 @@ struct cpu_spec { extern struct cpu_spec *cur_cpu_spec; -extern void identify_cpu(unsigned long offset, unsigned long cpu); -extern void do_cpu_ftr_fixups(unsigned long offset); +extern unsigned int __start___ftr_fixup, __stop___ftr_fixup; + +extern struct cpu_spec *identify_cpu(unsigned long offset); +extern void do_feature_fixups(unsigned long offset, unsigned long value, + void *fixup_start, void *fixup_end); #endif /* __ASSEMBLY__ */ diff --git a/include/asm-powerpc/firmware.h b/include/asm-powerpc/firmware.h index 1022737..c16e0a6 100644 --- a/include/asm-powerpc/firmware.h +++ b/include/asm-powerpc/firmware.h @@ -96,6 +96,8 @@ extern void machine_check_fwnmi(void); /* This is true if we are using the firmware NMI handler (typically LPAR) */ extern int fwnmi_active; +extern unsigned int __start___fw_ftr_fixup, __stop___fw_ftr_fixup; + #else /* __ASSEMBLY__ */ #define BEGIN_FW_FTR_SECTION 96: -- cgit v0.10.2 From 7aeb732428fc8e2ecae6d432873770c12f04a979 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 20 Oct 2006 11:47:16 +1000 Subject: [POWERPC] Support nested cpu feature sections This patch adds some macros that can be used with an explicit label in order to nest cpu features. This should be used very careful but is necessary for the upcoming cell TB fixup. Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Signed-off-by: Paul Mackerras diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h index 4d22218..65faf32 100644 --- a/include/asm-powerpc/cputable.h +++ b/include/asm-powerpc/cputable.h @@ -434,30 +434,34 @@ static inline int cpu_has_feature(unsigned long feature) #ifdef __ASSEMBLY__ -#define BEGIN_FTR_SECTION 98: +#define BEGIN_FTR_SECTION_NESTED(label) label: +#define BEGIN_FTR_SECTION BEGIN_FTR_SECTION_NESTED(98) #ifndef __powerpc64__ -#define END_FTR_SECTION(msk, val) \ +#define END_FTR_SECTION_NESTED(msk, val, label) \ 99: \ .section __ftr_fixup,"a"; \ .align 2; \ .long msk; \ .long val; \ - .long 98b; \ + .long label##b; \ .long 99b; \ .previous #else /* __powerpc64__ */ -#define END_FTR_SECTION(msk, val) \ +#define END_FTR_SECTION_NESTED(msk, val, label) \ 99: \ .section __ftr_fixup,"a"; \ .align 3; \ .llong msk; \ .llong val; \ - .llong 98b; \ + .llong label##b; \ .llong 99b; \ .previous #endif /* __powerpc64__ */ +#define END_FTR_SECTION(msk, val) \ + END_FTR_SECTION_NESTED(msk, val, 98) + #define END_FTR_SECTION_IFSET(msk) END_FTR_SECTION((msk), (msk)) #define END_FTR_SECTION_IFCLR(msk) END_FTR_SECTION((msk), 0) #endif /* __ASSEMBLY__ */ -- cgit v0.10.2 From 0909c8c2d547e45ca50e2492b08ec93a37b35237 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 20 Oct 2006 11:47:18 +1000 Subject: [POWERPC] Support feature fixups in vdso's This patch reworks the feature fixup mecanism so vdso's can be fixed up. The main issue was that the construct: .long label (or .llong on 64 bits) will not work in the case of a shared library like the vdso. It will generate an empty placeholder in the fixup table along with a reloc, which is not something we can deal with in the vdso. The idea here (thanks Alan Modra !) is to instead use something like: 1: .long label - 1b That is, the feature fixup tables no longer contain addresses of bits of code to patch, but offsets of such code from the fixup table entry itself. That is properly resolved by ld when building the .so's. I've modified the fixup mecanism generically to use that method for the rest of the kernel as well. Another trick is that the 32 bits vDSO included in the 64 bits kernel need to have a table in the 64 bits format. However, gas does not support 32 bits code with a statement of the form: .llong label - 1b (Or even just .llong label) That is, it cannot emit the right fixup/relocation for the linker to use to assign a 32 bits address to an .llong field. Thus, in the specific case of the 32 bits vdso built as part of the 64 bits kernel, we are using a modified macro that generates: .long 0xffffffff .llong label - 1b Note that is assumes that the value is negative which is enforced by the .lds (those offsets are always negative as the .text is always before the fixup table and gas doesn't support emiting the reloc the other way around). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 6fdfaa4..bfd499e 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -1202,14 +1202,13 @@ struct cpu_spec *identify_cpu(unsigned long offset) return NULL; } -void do_feature_fixups(unsigned long offset, unsigned long value, - void *fixup_start, void *fixup_end) +void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) { struct fixup_entry { unsigned long mask; unsigned long value; - unsigned int *start; - unsigned int *end; + long start_off; + long end_off; } *fcur, *fend; fcur = fixup_start; @@ -1224,8 +1223,8 @@ void do_feature_fixups(unsigned long offset, unsigned long value, /* These PTRRELOCs will disappear once the new scheme for * modules and vdso is implemented */ - pstart = PTRRELOC(fcur->start); - pend = PTRRELOC(fcur->end); + pstart = ((unsigned int *)fcur) + (fcur->start_off / 4); + pend = ((unsigned int *)fcur) + (fcur->end_off / 4); for (p = pstart; p < pend; p++) { *p = 0x60000000u; diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index 769e511..a4c2964 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -103,7 +103,7 @@ unsigned long __init early_init(unsigned long dt_ptr) */ spec = identify_cpu(offset); - do_feature_fixups(offset, spec->cpu_features, + do_feature_fixups(spec->cpu_features, PTRRELOC(&__start___ftr_fixup), PTRRELOC(&__stop___ftr_fixup)); diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 1969b56..1627896 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -354,9 +354,9 @@ void __init setup_system(void) /* Apply the CPUs-specific and firmware specific fixups to kernel * text (nop out sections not relevant to this CPU or this firmware) */ - do_feature_fixups(0, cur_cpu_spec->cpu_features, + do_feature_fixups(cur_cpu_spec->cpu_features, &__start___ftr_fixup, &__stop___ftr_fixup); - do_feature_fixups(0, powerpc_firmware_features, + do_feature_fixups(powerpc_firmware_features, &__start___fw_ftr_fixup, &__stop___fw_ftr_fixup); /* diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c index 1a7e19c..c913ad5 100644 --- a/arch/powerpc/kernel/vdso.c +++ b/arch/powerpc/kernel/vdso.c @@ -36,6 +36,8 @@ #include #include +#include "setup.h" + #undef DEBUG #ifdef DEBUG @@ -586,6 +588,43 @@ static __init int vdso_fixup_datapage(struct lib32_elfinfo *v32, return 0; } + +static __init int vdso_fixup_features(struct lib32_elfinfo *v32, + struct lib64_elfinfo *v64) +{ + void *start32; + unsigned long size32; + +#ifdef CONFIG_PPC64 + void *start64; + unsigned long size64; + + start64 = find_section64(v64->hdr, "__ftr_fixup", &size64); + if (start64) + do_feature_fixups(cur_cpu_spec->cpu_features, + start64, start64 + size64); + + start64 = find_section64(v64->hdr, "__fw_ftr_fixup", &size64); + if (start64) + do_feature_fixups(powerpc_firmware_features, + start64, start64 + size64); +#endif /* CONFIG_PPC64 */ + + start32 = find_section32(v32->hdr, "__ftr_fixup", &size32); + if (start32) + do_feature_fixups(cur_cpu_spec->cpu_features, + start32, start32 + size32); + +#ifdef CONFIG_PPC64 + start32 = find_section32(v32->hdr, "__fw_ftr_fixup", &size32); + if (start32) + do_feature_fixups(powerpc_firmware_features, + start32, start32 + size32); +#endif /* CONFIG_PPC64 */ + + return 0; +} + static __init int vdso_fixup_alt_funcs(struct lib32_elfinfo *v32, struct lib64_elfinfo *v64) { @@ -634,6 +673,9 @@ static __init int vdso_setup(void) if (vdso_fixup_datapage(&v32, &v64)) return -1; + if (vdso_fixup_features(&v32, &v64)) + return -1; + if (vdso_fixup_alt_funcs(&v32, &v64)) return -1; @@ -714,6 +756,7 @@ void __init vdso_init(void) * Setup the syscall map in the vDOS */ vdso_setup_syscall_map(); + /* * Initialize the vDSO images in memory, that is do necessary * fixups of vDSO symbols, locate trampolines, etc... diff --git a/arch/powerpc/kernel/vdso32/vdso32.lds.S b/arch/powerpc/kernel/vdso32/vdso32.lds.S index 6187af2..26e138c 100644 --- a/arch/powerpc/kernel/vdso32/vdso32.lds.S +++ b/arch/powerpc/kernel/vdso32/vdso32.lds.S @@ -32,6 +32,18 @@ SECTIONS PROVIDE (_etext = .); PROVIDE (etext = .); + . = ALIGN(8); + __ftr_fixup : { + *(__ftr_fixup) + } + +#ifdef CONFIG_PPC64 + . = ALIGN(8); + __fw_ftr_fixup : { + *(__fw_ftr_fixup) + } +#endif + /* Other stuff is appended to the text segment: */ .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } .rodata1 : { *(.rodata1) } diff --git a/arch/powerpc/kernel/vdso64/vdso64.lds.S b/arch/powerpc/kernel/vdso64/vdso64.lds.S index 4a2b6dc..2d70f35 100644 --- a/arch/powerpc/kernel/vdso64/vdso64.lds.S +++ b/arch/powerpc/kernel/vdso64/vdso64.lds.S @@ -31,6 +31,16 @@ SECTIONS PROVIDE (_etext = .); PROVIDE (etext = .); + . = ALIGN(8); + __ftr_fixup : { + *(__ftr_fixup) + } + + . = ALIGN(8); + __fw_ftr_fixup : { + *(__fw_ftr_fixup) + } + /* Other stuff is appended to the text segment: */ .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } .rodata1 : { *(.rodata1) } diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c index 41a640f..27faeca 100644 --- a/arch/ppc/kernel/setup.c +++ b/arch/ppc/kernel/setup.c @@ -314,7 +314,7 @@ early_init(int r3, int r4, int r5) * that depend on which cpu we have. */ spec = identify_cpu(offset); - do_feature_fixups(offset, spec->cpu_features, + do_feature_fixups(spec->cpu_features, PTRRELOC(&__start___ftr_fixup), PTRRELOC(&__stop___ftr_fixup)); diff --git a/include/asm-powerpc/asm-compat.h b/include/asm-powerpc/asm-compat.h index 8e64be0..c89bd58 100644 --- a/include/asm-powerpc/asm-compat.h +++ b/include/asm-powerpc/asm-compat.h @@ -14,6 +14,58 @@ # define ASM_CONST(x) __ASM_CONST(x) #endif + +/* + * Feature section common macros + * + * Note that the entries now contain offsets between the table entry + * and the code rather than absolute code pointers in order to be + * useable with the vdso shared library. There is also an assumption + * that values will be negative, that is, the fixup table has to be + * located after the code it fixes up. + */ +#ifdef CONFIG_PPC64 +#ifdef __powerpc64__ +/* 64 bits kernel, 64 bits code */ +#define MAKE_FTR_SECTION_ENTRY(msk, val, label, sect) \ +99: \ + .section sect,"a"; \ + .align 3; \ +98: \ + .llong msk; \ + .llong val; \ + .llong label##b-98b; \ + .llong 99b-98b; \ + .previous +#else /* __powerpc64__ */ +/* 64 bits kernel, 32 bits code (ie. vdso32) */ +#define MAKE_FTR_SECTION_ENTRY(msk, val, label, sect) \ +99: \ + .section sect,"a"; \ + .align 3; \ +98: \ + .llong msk; \ + .llong val; \ + .long 0xffffffff; \ + .long label##b-98b; \ + .long 0xffffffff; \ + .long 99b-98b; \ + .previous +#endif /* !__powerpc64__ */ +#else /* CONFIG_PPC64 */ +/* 32 bits kernel, 32 bits code */ +#define MAKE_FTR_SECTION_ENTRY(msk, val, label, sect) \ +99: \ + .section sect,"a"; \ + .align 2; \ +98: \ + .long msk; \ + .long val; \ + .long label##b-98b; \ + .long 99b-98b; \ + .previous +#endif /* !CONFIG_PPC64 */ + #ifdef __powerpc64__ /* operations for longs and pointers */ diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h index 65faf32..02e52d6 100644 --- a/include/asm-powerpc/cputable.h +++ b/include/asm-powerpc/cputable.h @@ -92,8 +92,8 @@ extern struct cpu_spec *cur_cpu_spec; extern unsigned int __start___ftr_fixup, __stop___ftr_fixup; extern struct cpu_spec *identify_cpu(unsigned long offset); -extern void do_feature_fixups(unsigned long offset, unsigned long value, - void *fixup_start, void *fixup_end); +extern void do_feature_fixups(unsigned long value, void *fixup_start, + void *fixup_end); #endif /* __ASSEMBLY__ */ @@ -435,32 +435,11 @@ static inline int cpu_has_feature(unsigned long feature) #ifdef __ASSEMBLY__ #define BEGIN_FTR_SECTION_NESTED(label) label: -#define BEGIN_FTR_SECTION BEGIN_FTR_SECTION_NESTED(98) - -#ifndef __powerpc64__ -#define END_FTR_SECTION_NESTED(msk, val, label) \ -99: \ - .section __ftr_fixup,"a"; \ - .align 2; \ - .long msk; \ - .long val; \ - .long label##b; \ - .long 99b; \ - .previous -#else /* __powerpc64__ */ +#define BEGIN_FTR_SECTION BEGIN_FTR_SECTION_NESTED(97) #define END_FTR_SECTION_NESTED(msk, val, label) \ -99: \ - .section __ftr_fixup,"a"; \ - .align 3; \ - .llong msk; \ - .llong val; \ - .llong label##b; \ - .llong 99b; \ - .previous -#endif /* __powerpc64__ */ - + MAKE_FTR_SECTION_ENTRY(msk, val, label, __ftr_fixup) #define END_FTR_SECTION(msk, val) \ - END_FTR_SECTION_NESTED(msk, val, 98) + END_FTR_SECTION_NESTED(msk, val, 97) #define END_FTR_SECTION_IFSET(msk) END_FTR_SECTION((msk), (msk)) #define END_FTR_SECTION_IFCLR(msk) END_FTR_SECTION((msk), 0) diff --git a/include/asm-powerpc/firmware.h b/include/asm-powerpc/firmware.h index c16e0a6..fdf9aff 100644 --- a/include/asm-powerpc/firmware.h +++ b/include/asm-powerpc/firmware.h @@ -100,17 +100,12 @@ extern unsigned int __start___fw_ftr_fixup, __stop___fw_ftr_fixup; #else /* __ASSEMBLY__ */ -#define BEGIN_FW_FTR_SECTION 96: - +#define BEGIN_FW_FTR_SECTION_NESTED(label) label: +#define BEGIN_FW_FTR_SECTION BEGIN_FW_FTR_SECTION_NESTED(97) +#define END_FW_FTR_SECTION_NESTED(msk, val, label) \ + MAKE_FTR_SECTION_ENTRY(msk, val, label, __fw_ftr_fixup) #define END_FW_FTR_SECTION(msk, val) \ -97: \ - .section __fw_ftr_fixup,"a"; \ - .align 3; \ - .llong msk; \ - .llong val; \ - .llong 96b; \ - .llong 97b; \ - .previous + END_FW_FTR_SECTION_NESTED(msk, val, 97) #define END_FW_FTR_SECTION_IFSET(msk) END_FW_FTR_SECTION((msk), (msk)) #define END_FW_FTR_SECTION_IFCLR(msk) END_FW_FTR_SECTION((msk), 0) diff --git a/include/asm-powerpc/timex.h b/include/asm-powerpc/timex.h index 3b9a8e7..e3f08cf 100644 --- a/include/asm-powerpc/timex.h +++ b/include/asm-powerpc/timex.h @@ -30,13 +30,15 @@ static inline cycles_t get_cycles(void) ret = 0; __asm__ __volatile__( - "98: mftb %0\n" + "97: mftb %0\n" "99:\n" ".section __ftr_fixup,\"a\"\n" + ".align 2\n" + "98:\n" " .long %1\n" " .long 0\n" - " .long 98b\n" - " .long 99b\n" + " .long 97b-98b\n" + " .long 99b-98b\n" ".previous" : "=r" (ret) : "i" (CPU_FTR_601)); #endif -- cgit v0.10.2 From 21c4ff80cba5e24932f3ef79c8482c0491630b2b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 20 Oct 2006 11:47:19 +1000 Subject: [POWERPC] Support feature fixups in modules This patch adds support for feature fixups in modules. This involves adding support for R_PPC64_REL64 relocs to the 64 bits module loader. It also modifies modpost.c to ignore the powerpc fixup sections (or it would warn when used in .init.text). Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/module_32.c b/arch/powerpc/kernel/module_32.c index 92f4e5f..e2c3c6a 100644 --- a/arch/powerpc/kernel/module_32.c +++ b/arch/powerpc/kernel/module_32.c @@ -24,6 +24,8 @@ #include #include +#include "setup.h" + #if 0 #define DEBUGP printk #else @@ -269,33 +271,50 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, return 0; } +static const Elf_Shdr *find_section(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + const char *name) +{ + char *secstrings; + unsigned int i; + + secstrings = (char *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; + for (i = 1; i < hdr->e_shnum; i++) + if (strcmp(secstrings+sechdrs[i].sh_name, name) == 0) + return &sechdrs[i]; + return NULL; +} + int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *me) { - char *secstrings; - unsigned int i; + const Elf_Shdr *sect; me->arch.bug_table = NULL; me->arch.num_bugs = 0; /* Find the __bug_table section, if present */ - secstrings = (char *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; - for (i = 1; i < hdr->e_shnum; i++) { - if (strcmp(secstrings+sechdrs[i].sh_name, "__bug_table")) - continue; - me->arch.bug_table = (void *) sechdrs[i].sh_addr; - me->arch.num_bugs = sechdrs[i].sh_size / sizeof(struct bug_entry); - break; + sect = find_section(hdr, sechdrs, "__bug_table"); + if (sect != NULL) { + me->arch.bug_table = (void *) sect->sh_addr; + me->arch.num_bugs = sect->sh_size / sizeof(struct bug_entry); } - /* + /* * Strictly speaking this should have a spinlock to protect against * traversals, but since we only traverse on BUG()s, a spinlock * could potentially lead to deadlock and thus be counter-productive. */ list_add(&me->arch.bug_list, &module_bug_list); + /* Apply feature fixups */ + sect = find_section(hdr, sechdrs, "__ftr_fixup"); + if (sect != NULL) + do_feature_fixups(cur_cpu_spec->cpu_features, + (void *)sect->sh_addr, + (void *)sect->sh_addr + sect->sh_size); + return 0; } diff --git a/arch/powerpc/kernel/module_64.c b/arch/powerpc/kernel/module_64.c index ba34001..8dd1f0a 100644 --- a/arch/powerpc/kernel/module_64.c +++ b/arch/powerpc/kernel/module_64.c @@ -22,6 +22,9 @@ #include #include #include +#include + +#include "setup.h" /* FIXME: We don't do .init separately. To do this, we'd need to have a separate r2 value in the init and core section, and stub between @@ -400,6 +403,11 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, | (value & 0x03fffffc); break; + case R_PPC64_REL64: + /* 64 bits relative (used by features fixups) */ + *location = value - (unsigned long)location; + break; + default: printk("%s: Unknown ADD relocation: %lu\n", me->name, @@ -413,23 +421,33 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, LIST_HEAD(module_bug_list); -int module_finalize(const Elf_Ehdr *hdr, - const Elf_Shdr *sechdrs, struct module *me) +static const Elf_Shdr *find_section(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + const char *name) { char *secstrings; unsigned int i; + secstrings = (char *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; + for (i = 1; i < hdr->e_shnum; i++) + if (strcmp(secstrings+sechdrs[i].sh_name, name) == 0) + return &sechdrs[i]; + return NULL; +} + +int module_finalize(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, struct module *me) +{ + const Elf_Shdr *sect; + me->arch.bug_table = NULL; me->arch.num_bugs = 0; /* Find the __bug_table section, if present */ - secstrings = (char *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; - for (i = 1; i < hdr->e_shnum; i++) { - if (strcmp(secstrings+sechdrs[i].sh_name, "__bug_table")) - continue; - me->arch.bug_table = (void *) sechdrs[i].sh_addr; - me->arch.num_bugs = sechdrs[i].sh_size / sizeof(struct bug_entry); - break; + sect = find_section(hdr, sechdrs, "__bug_table"); + if (sect != NULL) { + me->arch.bug_table = (void *) sect->sh_addr; + me->arch.num_bugs = sect->sh_size / sizeof(struct bug_entry); } /* @@ -439,6 +457,19 @@ int module_finalize(const Elf_Ehdr *hdr, */ list_add(&me->arch.bug_list, &module_bug_list); + /* Apply feature fixups */ + sect = find_section(hdr, sechdrs, "__ftr_fixup"); + if (sect != NULL) + do_feature_fixups(cur_cpu_spec->cpu_features, + (void *)sect->sh_addr, + (void *)sect->sh_addr + sect->sh_size); + + sect = find_section(hdr, sechdrs, "__fw_ftr_fixup"); + if (sect != NULL) + do_feature_fixups(powerpc_firmware_features, + (void *)sect->sh_addr, + (void *)sect->sh_addr + sect->sh_size); + return 0; } diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index 4127796..2e11416 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -921,6 +921,8 @@ static int init_section_ref_ok(const char *name) ".fixup", ".smp_locks", ".plt", /* seen on ARCH=um build on x86_64. Harmless */ + "__ftr_fixup", /* powerpc cpu feature fixup */ + "__fw_ftr_fixup", /* powerpc firmware feature fixup */ NULL }; /* Start of section names */ -- cgit v0.10.2 From 859deea949c382d9ccb6397fe33df3703ecef45d Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 20 Oct 2006 14:37:05 +1000 Subject: [POWERPC] Cell timebase bug workaround The Cell CPU timebase has an erratum. When reading the entire 64 bits of the timebase with one mftb instruction, there is a handful of cycles window during which one might read a value with the low order 32 bits already reset to 0x00000000 but the high order bits not yet incremeted by one. This fixes it by reading the timebase again until the low order 32 bits is no longer 0. That might introduce occasional latencies if hitting mftb just at the wrong time, but no more than 70ns on a cell blade, and that was considered acceptable. Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/vdso64/gettimeofday.S b/arch/powerpc/kernel/vdso64/gettimeofday.S index 56e76ff..40ffd9b 100644 --- a/arch/powerpc/kernel/vdso64/gettimeofday.S +++ b/arch/powerpc/kernel/vdso64/gettimeofday.S @@ -229,8 +229,10 @@ V_FUNCTION_BEGIN(__do_get_xsec) xor r0,r8,r8 /* create dependency */ add r3,r3,r0 - /* Get TB & offset it */ - mftb r7 + /* Get TB & offset it. We use the MFTB macro which will generate + * workaround code for Cell. + */ + MFTB(r7) ld r9,CFG_TB_ORIG_STAMP(r3) subf r7,r9,r7 diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h index 02e52d6..a9a4014 100644 --- a/include/asm-powerpc/cputable.h +++ b/include/asm-powerpc/cputable.h @@ -147,6 +147,7 @@ extern void do_feature_fixups(unsigned long value, void *fixup_start, #define CPU_FTR_CI_LARGE_PAGE LONG_ASM_CONST(0x0000100000000000) #define CPU_FTR_PAUSE_ZERO LONG_ASM_CONST(0x0000200000000000) #define CPU_FTR_PURR LONG_ASM_CONST(0x0000400000000000) +#define CPU_FTR_CELL_TB_BUG LONG_ASM_CONST(0x0000800000000000) #ifndef __ASSEMBLY__ @@ -335,7 +336,7 @@ extern void do_feature_fixups(unsigned long value, void *fixup_start, #define CPU_FTRS_CELL (CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \ CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_CTRL | \ CPU_FTR_ALTIVEC_COMP | CPU_FTR_MMCRA | CPU_FTR_SMT | \ - CPU_FTR_PAUSE_ZERO | CPU_FTR_CI_LARGE_PAGE) + CPU_FTR_PAUSE_ZERO | CPU_FTR_CI_LARGE_PAGE | CPU_FTR_CELL_TB_BUG) #define CPU_FTRS_PA6T (CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | \ CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 | \ CPU_FTR_ALTIVEC_COMP | CPU_FTR_CI_LARGE_PAGE | \ diff --git a/include/asm-powerpc/ppc_asm.h b/include/asm-powerpc/ppc_asm.h index a940cfe..fa083d8 100644 --- a/include/asm-powerpc/ppc_asm.h +++ b/include/asm-powerpc/ppc_asm.h @@ -30,9 +30,9 @@ BEGIN_FTR_SECTION; \ mfspr ra,SPRN_PURR; /* get processor util. reg */ \ END_FTR_SECTION_IFSET(CPU_FTR_PURR); \ BEGIN_FTR_SECTION; \ - mftb ra; /* or get TB if no PURR */ \ + MFTB(ra); /* or get TB if no PURR */ \ END_FTR_SECTION_IFCLR(CPU_FTR_PURR); \ - ld rb,PACA_STARTPURR(r13); \ + ld rb,PACA_STARTPURR(r13); \ std ra,PACA_STARTPURR(r13); \ subf rb,rb,ra; /* subtract start value */ \ ld ra,PACA_USER_TIME(r13); \ @@ -45,9 +45,9 @@ BEGIN_FTR_SECTION; \ mfspr ra,SPRN_PURR; /* get processor util. reg */ \ END_FTR_SECTION_IFSET(CPU_FTR_PURR); \ BEGIN_FTR_SECTION; \ - mftb ra; /* or get TB if no PURR */ \ + MFTB(ra); /* or get TB if no PURR */ \ END_FTR_SECTION_IFCLR(CPU_FTR_PURR); \ - ld rb,PACA_STARTPURR(r13); \ + ld rb,PACA_STARTPURR(r13); \ std ra,PACA_STARTPURR(r13); \ subf rb,rb,ra; /* subtract start value */ \ ld ra,PACA_SYSTEM_TIME(r13); \ @@ -274,6 +274,16 @@ END_FTR_SECTION_IFSET(CPU_FTR_601) #define ISYNC_601 #endif +#ifdef CONFIG_PPC_CELL +#define MFTB(dest) \ +90: mftb dest; \ +BEGIN_FTR_SECTION_NESTED(96); \ + cmpwi dest,0; \ + beq- 90b; \ +END_FTR_SECTION_NESTED(CPU_FTR_CELL_TB_BUG, CPU_FTR_CELL_TB_BUG, 96) +#else +#define MFTB(dest) mftb dest +#endif #ifndef CONFIG_SMP #define TLBSYNC diff --git a/include/asm-powerpc/reg.h b/include/asm-powerpc/reg.h index fde5c80..6faae7b 100644 --- a/include/asm-powerpc/reg.h +++ b/include/asm-powerpc/reg.h @@ -619,10 +619,35 @@ : "=r" (rval)); rval;}) #define mtspr(rn, v) asm volatile("mtspr " __stringify(rn) ",%0" : : "r" (v)) +#ifdef __powerpc64__ +#ifdef CONFIG_PPC_CELL +#define mftb() ({unsigned long rval; \ + asm volatile( \ + "90: mftb %0;\n" \ + "97: cmpwi %0,0;\n" \ + " beq- 90b;\n" \ + "99:\n" \ + ".section __ftr_fixup,\"a\"\n" \ + ".align 3\n" \ + "98:\n" \ + " .llong %1\n" \ + " .llong %1\n" \ + " .llong 97b-98b\n" \ + " .llong 99b-98b\n" \ + ".previous" \ + : "=r" (rval) : "i" (CPU_FTR_CELL_TB_BUG)); rval;}) +#else #define mftb() ({unsigned long rval; \ asm volatile("mftb %0" : "=r" (rval)); rval;}) +#endif /* !CONFIG_PPC_CELL */ + +#else /* __powerpc64__ */ + #define mftbl() ({unsigned long rval; \ asm volatile("mftbl %0" : "=r" (rval)); rval;}) +#define mftbu() ({unsigned long rval; \ + asm volatile("mftbu %0" : "=r" (rval)); rval;}) +#endif /* !__powerpc64__ */ #define mttbl(v) asm volatile("mttbl %0":: "r"(v)) #define mttbu(v) asm volatile("mttbu %0":: "r"(v)) diff --git a/include/asm-powerpc/time.h b/include/asm-powerpc/time.h index b051d4c..a782850 100644 --- a/include/asm-powerpc/time.h +++ b/include/asm-powerpc/time.h @@ -82,30 +82,35 @@ struct div_result { #define __USE_RTC() 0 #endif -/* On ppc64 this gets us the whole timebase; on ppc32 just the lower half */ +#ifdef CONFIG_PPC64 + +/* For compatibility, get_tbl() is defined as get_tb() on ppc64 */ +#define get_tbl get_tb + +#else + static inline unsigned long get_tbl(void) { - unsigned long tbl; - #if defined(CONFIG_403GCX) + unsigned long tbl; asm volatile("mfspr %0, 0x3dd" : "=r" (tbl)); + return tbl; #else - asm volatile("mftb %0" : "=r" (tbl)); + return mftbl(); #endif - return tbl; } static inline unsigned int get_tbu(void) { +#ifdef CONFIG_403GCX unsigned int tbu; - -#if defined(CONFIG_403GCX) asm volatile("mfspr %0, 0x3dc" : "=r" (tbu)); + return tbu; #else - asm volatile("mftbu %0" : "=r" (tbu)); + return mftbu(); #endif - return tbu; } +#endif /* !CONFIG_PPC64 */ static inline unsigned int get_rtcl(void) { @@ -131,7 +136,7 @@ static inline u64 get_tb(void) { return mftb(); } -#else +#else /* CONFIG_PPC64 */ static inline u64 get_tb(void) { unsigned int tbhi, tblo, tbhi2; @@ -144,7 +149,7 @@ static inline u64 get_tb(void) return ((u64)tbhi << 32) | tblo; } -#endif +#endif /* !CONFIG_PPC64 */ static inline void set_tb(unsigned int upper, unsigned int lower) { diff --git a/include/asm-powerpc/timex.h b/include/asm-powerpc/timex.h index e3f08cf..92dedde 100644 --- a/include/asm-powerpc/timex.h +++ b/include/asm-powerpc/timex.h @@ -8,6 +8,7 @@ */ #include +#include #define CLOCK_TICK_RATE 1024000 /* Underlying HZ */ @@ -15,13 +16,11 @@ typedef unsigned long cycles_t; static inline cycles_t get_cycles(void) { - cycles_t ret; - #ifdef __powerpc64__ - - __asm__ __volatile__("mftb %0" : "=r" (ret) : ); - + return mftb(); #else + cycles_t ret; + /* * For the "cycle" counter we use the timebase lower half. * Currently only used on SMP. @@ -41,9 +40,8 @@ static inline cycles_t get_cycles(void) " .long 99b-98b\n" ".previous" : "=r" (ret) : "i" (CPU_FTR_601)); -#endif - return ret; +#endif } #endif /* __KERNEL__ */ -- cgit v0.10.2 From e2100efb266c9335925191afe79f81f8d0a5807e Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 20 Oct 2006 11:49:54 +1000 Subject: [POWERPC] Fix device_is_compatible() const warning Fix a const'ification related warning with device_is_compatible() and friends related to get_property() not properly having const on it's input device node argument. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 865b964..bdb412d 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -1014,7 +1014,7 @@ EXPORT_SYMBOL(find_all_nodes); /** Checks if the given "compat" string matches one of the strings in * the device's "compatible" property */ -int device_is_compatible(struct device_node *device, const char *compat) +int device_is_compatible(const struct device_node *device, const char *compat) { const char* cp; int cplen, l; @@ -1491,7 +1491,8 @@ static int __init prom_reconfig_setup(void) __initcall(prom_reconfig_setup); #endif -struct property *of_find_property(struct device_node *np, const char *name, +struct property *of_find_property(const struct device_node *np, + const char *name, int *lenp) { struct property *pp; @@ -1512,7 +1513,8 @@ struct property *of_find_property(struct device_node *np, const char *name, * Find a property with a given name for a given node * and return the value. */ -const void *get_property(struct device_node *np, const char *name, int *lenp) +const void *get_property(const struct device_node *np, const char *name, + int *lenp) { struct property *pp = of_find_property(np,name,lenp); return pp ? pp->value : NULL; diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h index 5246297..ec11d44 100644 --- a/include/asm-powerpc/prom.h +++ b/include/asm-powerpc/prom.h @@ -134,7 +134,7 @@ extern struct device_node *of_find_all_nodes(struct device_node *prev); extern struct device_node *of_get_parent(const struct device_node *node); extern struct device_node *of_get_next_child(const struct device_node *node, struct device_node *prev); -extern struct property *of_find_property(struct device_node *np, +extern struct property *of_find_property(const struct device_node *np, const char *name, int *lenp); extern struct device_node *of_node_get(struct device_node *node); @@ -158,10 +158,12 @@ extern void of_detach_node(const struct device_node *); extern void finish_device_tree(void); extern void unflatten_device_tree(void); extern void early_init_devtree(void *); -extern int device_is_compatible(struct device_node *device, const char *); +extern int device_is_compatible(const struct device_node *device, + const char *); extern int machine_is_compatible(const char *compat); -extern const void *get_property(struct device_node *node, const char *name, - int *lenp); +extern const void *get_property(const struct device_node *node, + const char *name, + int *lenp); extern void print_properties(struct device_node *node); extern int prom_n_addr_cells(struct device_node* np); extern int prom_n_size_cells(struct device_node* np); -- cgit v0.10.2 From f6b301b89b7bf0bb872da4f37dc28240413cbae7 Mon Sep 17 00:00:00 2001 From: Dwayne Grant Mcconnell Date: Tue, 24 Oct 2006 18:27:30 +0200 Subject: [POWERPC] spufs: fix signal2 file to report signal2 This fixes the /signal2 file to actually give signal2 data. Signed-off-by: Dwayne Grant Mcconnell Signed-off-by: Arnd Bergmann Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/cell/spufs/hw_ops.c b/arch/powerpc/platforms/cell/spufs/hw_ops.c index efc452e..d805ffe 100644 --- a/arch/powerpc/platforms/cell/spufs/hw_ops.c +++ b/arch/powerpc/platforms/cell/spufs/hw_ops.c @@ -147,7 +147,7 @@ static void spu_hw_signal1_write(struct spu_context *ctx, u32 data) static u32 spu_hw_signal2_read(struct spu_context *ctx) { - return in_be32(&ctx->spu->problem->signal_notify1); + return in_be32(&ctx->spu->problem->signal_notify2); } static void spu_hw_signal2_write(struct spu_context *ctx, u32 data) -- cgit v0.10.2 From 274cef5e9d0e494ad84dbc28513b0bb6e3d847ae Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 24 Oct 2006 18:01:42 +0200 Subject: [POWERPC] spufs: fix another off-by-one bug in spufs_mbox_read Currently, spufs_mbox_read transfers more bytes than requested on a read. If you ask for four bytes, you get eight. This fixes it to transfer the largest multiple of four bytes that is less than or equal to the number you asked for. Note: one nasty property of this file in spufs is that you can only read multiples of four bytes in the first place, since there is no way to atomically put back a few bytes into the hardware register. Thus, reading less than four bytes returns -EINVAL. Asking for more than four returns the largest possible multiple of four. Signed-off-by: Arnd Bergmann Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c index 0de8e11..533e272 100644 --- a/arch/powerpc/platforms/cell/spufs/file.c +++ b/arch/powerpc/platforms/cell/spufs/file.c @@ -385,7 +385,7 @@ static ssize_t spufs_mbox_read(struct file *file, char __user *buf, udata = (void __user *)buf; spu_acquire(ctx); - for (count = 0; count <= len; count += 4, udata++) { + for (count = 0; (count + 4) <= len; count += 4, udata++) { int ret; ret = ctx->ops->mbox_read(ctx, &mbox_data); if (ret == 0) -- cgit v0.10.2 From b910ecf6bf221bb06f37e44765307c42b20db205 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 24 Oct 2006 18:01:43 +0200 Subject: [POWERPC] cell: update defconfig =================================================================== Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/configs/cell_defconfig b/arch/powerpc/configs/cell_defconfig index 892d5dd..0aba06d 100644 --- a/arch/powerpc/configs/cell_defconfig +++ b/arch/powerpc/configs/cell_defconfig @@ -254,6 +254,7 @@ CONFIG_SYN_COOKIES=y CONFIG_INET_TUNNEL=y CONFIG_INET_XFRM_MODE_TRANSPORT=y CONFIG_INET_XFRM_MODE_TUNNEL=y +# CONFIG_INET_XFRM_MODE_BEET is not set CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set @@ -275,7 +276,9 @@ CONFIG_INET6_XFRM_TUNNEL=m CONFIG_INET6_TUNNEL=m CONFIG_INET6_XFRM_MODE_TRANSPORT=y CONFIG_INET6_XFRM_MODE_TUNNEL=y +# CONFIG_INET6_XFRM_MODE_BEET is not set # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +# CONFIG_IPV6_SIT is not set CONFIG_IPV6_TUNNEL=m # CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set @@ -406,6 +409,12 @@ CONFIG_BLK_DEV_INITRD=y # CONFIG_ATA_OVER_ETH is not set # +# Misc devices +# +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # CONFIG_IDE=y @@ -738,7 +747,6 @@ CONFIG_GEN_RTC=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -802,6 +810,7 @@ CONFIG_I2C_ALGOBIT=y # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -810,14 +819,9 @@ CONFIG_I2C_ALGOBIT=y # CONFIG_HWMON_VID is not set # -# Misc devices -# - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -923,6 +927,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -930,6 +935,7 @@ CONFIG_FS_MBCACHE=y # CONFIG_JFS_FS is not set CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set @@ -1129,6 +1135,7 @@ CONFIG_DEBUG_FS=y # CONFIG_DEBUG_VM is not set # CONFIG_DEBUG_LIST is not set # CONFIG_FORCED_INLINING is not set +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_DEBUG_STACKOVERFLOW is not set # CONFIG_DEBUG_STACK_USAGE is not set -- cgit v0.10.2 From f84c39da766b4c8f13872282f58286a57ad05b3e Mon Sep 17 00:00:00 2001 From: Liu Dave-r63238 Date: Wed, 18 Oct 2006 16:36:56 +0800 Subject: [POWERPC] Fix the UCC rx/tx clock of QE MPC8323EMDS board ethernet interface with RMII uses the CLK16 divisor for the rx and tx clock, but the ucc_set_qe_mux_rxtx() function doesn't handle the CLK16 setting of the CMXUCR3 and CMXUCR4 registers. This fixes it. Signed-off-by: Dave Liu Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/sysdev/qe_lib/ucc.c b/arch/powerpc/sysdev/qe_lib/ucc.c index 916c9e5..ac12a44 100644 --- a/arch/powerpc/sysdev/qe_lib/ucc.c +++ b/arch/powerpc/sysdev/qe_lib/ucc.c @@ -207,6 +207,7 @@ int ucc_set_qe_mux_rxtx(int ucc_num, enum qe_clock clock, enum comm_dir mode) case QE_CLK18: source = 8; break; case QE_CLK7: source = 9; break; case QE_CLK8: source = 10; break; + case QE_CLK16: source = 11; break; default: source = -1; break; } break; @@ -222,6 +223,7 @@ int ucc_set_qe_mux_rxtx(int ucc_num, enum qe_clock clock, enum comm_dir mode) case QE_CLK22: source = 8; break; case QE_CLK7: source = 9; break; case QE_CLK8: source = 10; break; + case QE_CLK16: source = 11; break; default: source = -1; break; } break; -- cgit v0.10.2 From f49196a5f53aa62a964b08ffa2c59699a0c8eb53 Mon Sep 17 00:00:00 2001 From: Scott Wood Date: Mon, 23 Oct 2006 11:35:22 -0500 Subject: [POWERPC] IPIC: Fix spinlock recursion in set_irq_handler This causes ipic_set_irq_type to set the handler directly rather than call set_irq_handler, which causes spinlock recursion because the lock is already held when ipic_set_irq_type is called. I'm also not convinced that ipic_set_irq_type should be changing the handler at all. There seem to be several controllers that don't and several that do. Those that do would break what appears to be a common usage of calling set_irq_chip_and_handler followed by set_irq_type, if a non-standard handler were to be used. OTOH, irq_create_of_mapping() doesn't set the handler, but only calls set_irq_type(). This patch gets things working in the spinlock-debugging-enabled case, but I'm curious as to where the handler setting is ideally supposed to be done. I don't see any documentation on set_irq_type() that clarifies what the semantics are supposed to be. Signed-off-by: Scott Wood Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/sysdev/ipic.c b/arch/powerpc/sysdev/ipic.c index bc4d4a7..746f78c 100644 --- a/arch/powerpc/sysdev/ipic.c +++ b/arch/powerpc/sysdev/ipic.c @@ -473,9 +473,9 @@ static int ipic_set_irq_type(unsigned int virq, unsigned int flow_type) desc->status |= flow_type & IRQ_TYPE_SENSE_MASK; if (flow_type & IRQ_TYPE_LEVEL_LOW) { desc->status |= IRQ_LEVEL; - set_irq_handler(virq, handle_level_irq); + desc->handle_irq = handle_level_irq; } else { - set_irq_handler(virq, handle_edge_irq); + desc->handle_irq = handle_edge_irq; } /* only EXT IRQ senses are programmable on ipic -- cgit v0.10.2 From f4d4c354bca18210296cc0a8f592c0cdb720bf20 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 25 Oct 2006 13:22:27 +1000 Subject: [POWERPC] Fix CHRP platforms with only 8259 On CHRP platforms with only a 8259 controller, we should set the default IRQ host to the 8259 driver's one for the IRQ probing fallbacks to work in case the IRQ tree is incorrect (like on Pegasos for example). Without this fix, we get a bunch of WARN_ON's during boot. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c index cae3d13..49b8dab 100644 --- a/arch/powerpc/platforms/chrp/setup.c +++ b/arch/powerpc/platforms/chrp/setup.c @@ -477,8 +477,10 @@ static void __init chrp_find_8259(void) " address, polling\n"); i8259_init(pic, chrp_int_ack); - if (ppc_md.get_irq == NULL) + if (ppc_md.get_irq == NULL) { ppc_md.get_irq = i8259_irq; + irq_set_default_host(i8259_get_host()); + } if (chrp_mpic != NULL) { cascade_irq = irq_of_parse_and_map(pic, 0); if (cascade_irq == NO_IRQ) diff --git a/arch/powerpc/sysdev/i8259.c b/arch/powerpc/sysdev/i8259.c index 0450265..ad87adc9 100644 --- a/arch/powerpc/sysdev/i8259.c +++ b/arch/powerpc/sysdev/i8259.c @@ -224,6 +224,11 @@ static struct irq_host_ops i8259_host_ops = { .xlate = i8259_host_xlate, }; +struct irq_host *i8259_get_host(void) +{ + return i8259_host; +} + /** * i8259_init - Initialize the legacy controller * @node: device node of the legacy PIC (can be NULL, but then, it will match diff --git a/include/asm-powerpc/i8259.h b/include/asm-powerpc/i8259.h index 78489fb..db1362f 100644 --- a/include/asm-powerpc/i8259.h +++ b/include/asm-powerpc/i8259.h @@ -7,6 +7,7 @@ #ifdef CONFIG_PPC_MERGE extern void i8259_init(struct device_node *node, unsigned long intack_addr); extern unsigned int i8259_irq(void); +extern struct irq_host *i8259_get_host(void); #else extern void i8259_init(unsigned long intack_addr, int offset); extern int i8259_irq(void); -- cgit v0.10.2 From 6c50444723f2e6487b3377450f90d813a88e6c31 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Tue, 24 Oct 2006 13:34:11 -0700 Subject: [PATCH] Merge headphone and speaker volume controls for Panasonic R4 laptop Signed-off-by: Keith Packard Signed-off-by: Linus Torvalds diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c index 7f22dab..9c1bce7 100644 --- a/sound/pci/intel8x0.c +++ b/sound/pci/intel8x0.c @@ -1962,6 +1962,12 @@ static struct ac97_quirk ac97_quirks[] __devinitdata = { .type = AC97_TUNE_HP_ONLY }, { + .subvendor = 0x10f7, + .subdevice = 0x834c, + .name = "Panasonic CF-R4", + .type = AC97_TUNE_HP_ONLY, + }, + { .subvendor = 0x110a, .subdevice = 0x0056, .name = "Fujitsu-Siemens Scenic", /* AD1981? */ -- cgit v0.10.2 From 3a51237dc11efe283b40ea0248f4e26ab935dbd1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 24 Oct 2006 11:15:29 +0100 Subject: [PATCH] uml: mconsole fixes * when we have stop/sysrq/go, we get pt_regs of whatever executes mc_work_proc(). Would be better to see what we had at the time of interrupt that got us stop. * stop/stop/stop..... will give stack overflow. Shouldn't allow stop from mconsole_stop(). * stop/stop/go leaves us inside mconsole_stop() with os_set_fd_block(req->originating_fd, 0); reactivate_fd(req->originating_fd, MCONSOLE_IRQ); just done by nested mconsole_stop(). Ditto. * once we'd seen stop, there's a period when INTR commands are executed out of order (as they should; we might have the things stuck badly enough to never reach mconsole_stop(), but still not badly enough to block mconsole_interrupt(); in that situation we _want_ things like "cad" to be executed immediately). Once we enter monsole_stop(), all INTR commands will be executed in order, mixed with PROC ones. We'd better let user see that such change of behaviour has happened. (Suggested by lennert). * stack footprint of monsole_interrupt() is an atrocity; AFAICS we can safely make struct mc_request req; static in function there. Signed-off-by: Al Viro Acked-by: Jeff Dike Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index d08bd03..7b17216 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -79,7 +79,7 @@ static irqreturn_t mconsole_interrupt(int irq, void *dev_id) /* long to avoid size mismatch warnings from gcc */ long fd; struct mconsole_entry *new; - struct mc_request req; + static struct mc_request req; /* that's OK */ fd = (long) dev_id; while (mconsole_get_request(fd, &req)){ @@ -91,6 +91,7 @@ static irqreturn_t mconsole_interrupt(int irq, void *dev_id) mconsole_reply(&req, "Out of memory", 1, 0); else { new->request = req; + new->request.regs = get_irq_regs()->regs; list_add(&new->list, &mc_requests); } } @@ -314,9 +315,21 @@ void mconsole_stop(struct mc_request *req) { deactivate_fd(req->originating_fd, MCONSOLE_IRQ); os_set_fd_block(req->originating_fd, 1); - mconsole_reply(req, "", 0, 0); - while(mconsole_get_request(req->originating_fd, req)){ - if(req->cmd->handler == mconsole_go) break; + mconsole_reply(req, "stopped", 0, 0); + while (mconsole_get_request(req->originating_fd, req)) { + if (req->cmd->handler == mconsole_go) + break; + if (req->cmd->handler == mconsole_stop) { + mconsole_reply(req, "Already stopped", 1, 0); + continue; + } + if (req->cmd->handler == mconsole_sysrq) { + struct pt_regs *old_regs; + old_regs = set_irq_regs((struct pt_regs *)&req->regs); + mconsole_sysrq(req); + set_irq_regs(old_regs); + continue; + } (*req->cmd->handler)(req); } os_set_fd_block(req->originating_fd, 0); @@ -673,9 +686,7 @@ static void with_console(struct mc_request *req, void (*proc)(void *), static void sysrq_proc(void *arg) { char *op = arg; - struct pt_regs *old_regs = set_irq_regs(¤t->thread.regs); handle_sysrq(*op, NULL); - set_irq_regs(old_regs); } void mconsole_sysrq(struct mc_request *req) diff --git a/arch/um/drivers/mconsole_user.c b/arch/um/drivers/mconsole_user.c index 17068eb..75aef6f 100644 --- a/arch/um/drivers/mconsole_user.c +++ b/arch/um/drivers/mconsole_user.c @@ -14,6 +14,7 @@ #include #include #include "user.h" +#include "sysdep/ptrace.h" #include "mconsole.h" #include "umid.h" #include "user_util.h" diff --git a/arch/um/include/mconsole.h b/arch/um/include/mconsole.h index 58f67d3..2666815 100644 --- a/arch/um/include/mconsole.h +++ b/arch/um/include/mconsole.h @@ -61,6 +61,7 @@ struct mc_request struct mconsole_request request; struct mconsole_command *cmd; + union uml_pt_regs regs; }; extern char mconsole_socket_name[]; -- cgit v0.10.2 From aa6c2e62bbe7a20ccc85906f75bc63465d899227 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 24 Oct 2006 11:16:29 +0100 Subject: [PATCH] IOC4 should depend on PCI Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index b6c045d..00db31c 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -30,6 +30,7 @@ config IBM_ASM config SGI_IOC4 tristate "SGI IOC4 Base IO support" + depends on PCI ---help--- This option enables basic support for the IOC4 chip on certain SGI IO controller cards (IO9, IO10, and PCI-RT). This option -- cgit v0.10.2 From 016002312d50004908a79df37174b336e3682e18 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 24 Oct 2006 11:17:37 +0100 Subject: [PATCH] missing include of dma-mapping.h Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_dma.h b/drivers/net/wireless/bcm43xx/bcm43xx_dma.h index ea16078..d1105e5 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_dma.h +++ b/drivers/net/wireless/bcm43xx/bcm43xx_dma.h @@ -4,6 +4,7 @@ #include #include #include +#include #include #include -- cgit v0.10.2 From 2099c99e3b24f86b131566aa9854249189ae9ea2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 24 Oct 2006 11:17:06 +0100 Subject: [PATCH] missing includes of io.h Signed-off-by: Al Viro Signed-off-by: Linus Torvalds diff --git a/drivers/misc/ioc4.c b/drivers/misc/ioc4.c index 1c3c14a..79354bb 100644 --- a/drivers/misc/ioc4.c +++ b/drivers/misc/ioc4.c @@ -32,6 +32,7 @@ #include #include #include +#include /*************** * Definitions * diff --git a/drivers/mmc/tifm_sd.c b/drivers/mmc/tifm_sd.c index 2bacff6..0fdc55b 100644 --- a/drivers/mmc/tifm_sd.c +++ b/drivers/mmc/tifm_sd.c @@ -14,6 +14,7 @@ #include #include #include +#include #define DRIVER_NAME "tifm_sd" #define DRIVER_VERSION "0.6" -- cgit v0.10.2 From bcbaecbb9968750d4bfb2686a97e396f681f88ef Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 25 Oct 2006 16:49:36 +1000 Subject: [CRYPTO] users: Select ECB/CBC where needed CRYPTO_MANAGER is selected automatically by CONFIG_ECB and CONFIG_CBC. config CRYPTO_ECB tristate "ECB support" select CRYPTO_BLKCIPHER select CRYPTO_MANAGER I've added CONFIG_ECB to the ones you mentioned and CONFIG_CBC to gssapi. Signed-off-by: Patrick McHardy Signed-off-by: Herbert Xu diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index e2ed249..e38846e 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2717,6 +2717,7 @@ config PPP_MPPE select CRYPTO select CRYPTO_SHA1 select CRYPTO_ARC4 + select CRYPTO_ECB ---help--- Support for the MPPE Encryption protocol, as employed by the Microsoft Point-to-Point Tunneling Protocol. diff --git a/fs/Kconfig b/fs/Kconfig index fee318e..133dcc8 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1777,6 +1777,7 @@ config RPCSEC_GSS_KRB5 select CRYPTO select CRYPTO_MD5 select CRYPTO_DES + select CRYPTO_CBC help Provides for secure RPC calls by means of a gss-api mechanism based on Kerberos V5. This is required for @@ -1795,6 +1796,7 @@ config RPCSEC_GSS_SPKM3 select CRYPTO_MD5 select CRYPTO_DES select CRYPTO_CAST5 + select CRYPTO_CBC help Provides for secure RPC calls by means of a gss-api mechanism based on the SPKM3 public-key mechanism. diff --git a/net/ieee80211/Kconfig b/net/ieee80211/Kconfig index f7e84e9..a64be6c 100644 --- a/net/ieee80211/Kconfig +++ b/net/ieee80211/Kconfig @@ -32,6 +32,7 @@ config IEEE80211_CRYPT_WEP depends on IEEE80211 select CRYPTO select CRYPTO_ARC4 + select CRYPTO_ECB select CRC32 ---help--- Include software based cipher suites in support of IEEE @@ -58,6 +59,7 @@ config IEEE80211_CRYPT_TKIP depends on IEEE80211 && NET_RADIO select CRYPTO select CRYPTO_MICHAEL_MIC + select CRYPTO_ECB select CRC32 ---help--- Include software based cipher suites in support of IEEE 802.11i -- cgit v0.10.2 From 0d960d26c42888cf327df7faa1a8aa62bab53fa4 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Wed, 18 Oct 2006 00:26:39 -0400 Subject: fix return code in error case. The other failure returns in this function are negative, so make this one do the same. Signed-off-by: Dave Jones Signed-off-by: Dave Airlie diff --git a/drivers/char/drm/savage_state.c b/drivers/char/drm/savage_state.c index ef2581d..1ca1e9c 100644 --- a/drivers/char/drm/savage_state.c +++ b/drivers/char/drm/savage_state.c @@ -994,7 +994,7 @@ int savage_bci_cmdbuf(DRM_IOCTL_ARGS) if (cmdbuf.size) { kcmd_addr = drm_alloc(cmdbuf.size * 8, DRM_MEM_DRIVER); if (kcmd_addr == NULL) - return ENOMEM; + return DRM_ERR(ENOMEM); if (DRM_COPY_FROM_USER(kcmd_addr, cmdbuf.cmd_addr, cmdbuf.size * 8)) -- cgit v0.10.2 From 24f73c92a990ecd3d1bb846267780a264d830065 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 10 Oct 2006 14:23:37 -0700 Subject: drm: fix error returns, sysfs error handling - callers of drm_sysfs_create() and drm_sysfs_device_add() looked for errors using IS_ERR(), but the functions themselves only ever returned NULL on error. Fixed. - unwind from, and propagate sysfs errors Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Dave Airlie diff --git a/drivers/char/drm/drm_sysfs.c b/drivers/char/drm/drm_sysfs.c index 51ad98c..ba4b8de 100644 --- a/drivers/char/drm/drm_sysfs.c +++ b/drivers/char/drm/drm_sysfs.c @@ -42,13 +42,24 @@ static CLASS_ATTR(version, S_IRUGO, version_show, NULL); struct class *drm_sysfs_create(struct module *owner, char *name) { struct class *class; + int err; class = class_create(owner, name); - if (!class) - return class; + if (!class) { + err = -ENOMEM; + goto err_out; + } + + err = class_create_file(class, &class_attr_version); + if (err) + goto err_out_class; - class_create_file(class, &class_attr_version); return class; + +err_out_class: + class_destroy(class); +err_out: + return ERR_PTR(err); } /** @@ -96,20 +107,36 @@ static struct class_device_attribute class_device_attrs[] = { struct class_device *drm_sysfs_device_add(struct class *cs, drm_head_t *head) { struct class_device *class_dev; - int i; + int i, j, err; class_dev = class_device_create(cs, NULL, MKDEV(DRM_MAJOR, head->minor), &(head->dev->pdev)->dev, "card%d", head->minor); - if (!class_dev) - return NULL; + if (!class_dev) { + err = -ENOMEM; + goto err_out; + } class_set_devdata(class_dev, head); - for (i = 0; i < ARRAY_SIZE(class_device_attrs); i++) - class_device_create_file(class_dev, &class_device_attrs[i]); + for (i = 0; i < ARRAY_SIZE(class_device_attrs); i++) { + err = class_device_create_file(class_dev, + &class_device_attrs[i]); + if (err) + goto err_out_files; + } + return class_dev; + +err_out_files: + if (i > 0) + for (j = 0; j < i; j++) + class_device_remove_file(class_dev, + &class_device_attrs[i]); + class_device_unregister(class_dev); +err_out: + return ERR_PTR(err); } /** -- cgit v0.10.2 From 85abb3f95010b277a6efbc9b8031a7854af87e10 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Wed, 25 Oct 2006 09:55:34 -0700 Subject: drm: ioremap balanced with iounmap for drivers/char/drm ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. Tested (compilation only) to make sure the files are compiling without any warning/error due to new changes Signed-off-by: Amol Lad Signed-off-by: Dave Airlie diff --git a/drivers/char/drm/drm_bufs.c b/drivers/char/drm/drm_bufs.c index 029baea..6eafff1 100644 --- a/drivers/char/drm/drm_bufs.c +++ b/drivers/char/drm/drm_bufs.c @@ -237,6 +237,8 @@ static int drm_addmap_core(drm_device_t * dev, unsigned int offset, list = drm_alloc(sizeof(*list), DRM_MEM_MAPS); if (!list) { + if (map->type == _DRM_REGISTERS) + drm_ioremapfree(map->handle, map->size, dev); drm_free(map, sizeof(*map), DRM_MEM_MAPS); return -EINVAL; } @@ -252,6 +254,8 @@ static int drm_addmap_core(drm_device_t * dev, unsigned int offset, map->offset; ret = drm_map_handle(dev, &list->hash, user_token, 0); if (ret) { + if (map->type == _DRM_REGISTERS) + drm_ioremapfree(map->handle, map->size, dev); drm_free(map, sizeof(*map), DRM_MEM_MAPS); drm_free(list, sizeof(*list), DRM_MEM_MAPS); mutex_unlock(&dev->struct_mutex); -- cgit v0.10.2 From a77b8950019289611f836c8fc19f91592822efcd Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 20 Oct 2006 14:36:00 -0700 Subject: intel fb: switch to pci_get API Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Dave Airlie diff --git a/drivers/video/intelfb/intelfbhw.c b/drivers/video/intelfb/intelfbhw.c index eeeeff9..eae60f9 100644 --- a/drivers/video/intelfb/intelfbhw.c +++ b/drivers/video/intelfb/intelfbhw.c @@ -161,7 +161,7 @@ intelfbhw_get_memory(struct pci_dev *pdev, int *aperture_size, return 1; /* Find the bridge device. It is always 0:0.0 */ - if (!(bridge_dev = pci_find_slot(0, PCI_DEVFN(0, 0)))) { + if (!(bridge_dev = pci_get_bus_and_slot(0, PCI_DEVFN(0, 0)))) { ERR_MSG("cannot find bridge device\n"); return 1; } @@ -169,6 +169,8 @@ intelfbhw_get_memory(struct pci_dev *pdev, int *aperture_size, /* Get the fb aperture size and "stolen" memory amount. */ tmp = 0; pci_read_config_word(bridge_dev, INTEL_GMCH_CTRL, &tmp); + pci_dev_put(bridge_dev); + switch (pdev->device) { case PCI_DEVICE_ID_INTEL_915G: case PCI_DEVICE_ID_INTEL_915GM: -- cgit v0.10.2 From f84fcb06a1f7ab4ac0444ece82b25b0701369641 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Fri, 20 Oct 2006 14:35:59 -0700 Subject: Remove unnecessary check in drivers/video/intelfb/intelfbhw.c All callers and the function itself dereference dinfo, so we can remove the check. (coverity id #1371) Signed-off-by: Eric Sesterhenn Signed-off-by: Andrew Morton Signed-off-by: Dave Airlie diff --git a/drivers/video/intelfb/intelfbhw.c b/drivers/video/intelfb/intelfbhw.c index eae60f9..a958368 100644 --- a/drivers/video/intelfb/intelfbhw.c +++ b/drivers/video/intelfb/intelfbhw.c @@ -664,7 +664,7 @@ intelfbhw_print_hw_state(struct intelfb_info *dinfo, struct intelfb_hwstate *hw) int index = dinfo->pll_index; DBG_MSG("intelfbhw_print_hw_state\n"); - if (!hw || !dinfo) + if (!hw) return; /* Read in as much of the HW state as possible. */ printk("hw state dump start\n"); -- cgit v0.10.2 From 811c93666c3f4a0e99382c24a84480b03c7262f6 Mon Sep 17 00:00:00 2001 From: Henne Date: Tue, 3 Oct 2006 19:51:59 +0200 Subject: [SCSI] Scsi_Cmnd convertion in sun3-driver Change the obsolete Scsi_Cmnd to struct scsi_cmnd in the sun3-driver. Signed-off-by: Henrik Kretzschmar Signed-off-by: James Bottomley diff --git a/drivers/scsi/sun3_NCR5380.c b/drivers/scsi/sun3_NCR5380.c index 5ec5af8..3b3f305 100644 --- a/drivers/scsi/sun3_NCR5380.c +++ b/drivers/scsi/sun3_NCR5380.c @@ -266,8 +266,8 @@ static struct scsi_host_template *the_template = NULL; (struct NCR5380_hostdata *)(in)->hostdata #define HOSTDATA(in) ((struct NCR5380_hostdata *)(in)->hostdata) -#define NEXT(cmd) ((Scsi_Cmnd *)((cmd)->host_scribble)) -#define NEXTADDR(cmd) ((Scsi_Cmnd **)&((cmd)->host_scribble)) +#define NEXT(cmd) ((struct scsi_cmnd *)((cmd)->host_scribble)) +#define NEXTADDR(cmd) ((struct scsi_cmnd **)&((cmd)->host_scribble)) #define HOSTNO instance->host_no #define H_NO(cmd) (cmd)->device->host->host_no @@ -360,7 +360,7 @@ static void __init init_tags( void ) * conditions. */ -static int is_lun_busy( Scsi_Cmnd *cmd, int should_be_tagged ) +static int is_lun_busy(struct scsi_cmnd *cmd, int should_be_tagged) { SETUP_HOSTDATA(cmd->device->host); @@ -384,7 +384,7 @@ static int is_lun_busy( Scsi_Cmnd *cmd, int should_be_tagged ) * untagged. */ -static void cmd_get_tag( Scsi_Cmnd *cmd, int should_be_tagged ) +static void cmd_get_tag(struct scsi_cmnd *cmd, int should_be_tagged) { SETUP_HOSTDATA(cmd->device->host); @@ -416,7 +416,7 @@ static void cmd_get_tag( Scsi_Cmnd *cmd, int should_be_tagged ) * unlock the LUN. */ -static void cmd_free_tag( Scsi_Cmnd *cmd ) +static void cmd_free_tag(struct scsi_cmnd *cmd) { SETUP_HOSTDATA(cmd->device->host); @@ -460,18 +460,18 @@ static void free_all_tags( void ) /* - * Function: void merge_contiguous_buffers( Scsi_Cmnd *cmd ) + * Function: void merge_contiguous_buffers(struct scsi_cmnd *cmd) * * Purpose: Try to merge several scatter-gather requests into one DMA * transfer. This is possible if the scatter buffers lie on * physical contiguous addresses. * - * Parameters: Scsi_Cmnd *cmd + * Parameters: struct scsi_cmnd *cmd * The command to work on. The first scatter buffer's data are * assumed to be already transfered into ptr/this_residual. */ -static void merge_contiguous_buffers( Scsi_Cmnd *cmd ) +static void merge_contiguous_buffers(struct scsi_cmnd *cmd) { unsigned long endaddr; #if (NDEBUG & NDEBUG_MERGING) @@ -501,15 +501,15 @@ static void merge_contiguous_buffers( Scsi_Cmnd *cmd ) } /* - * Function : void initialize_SCp(Scsi_Cmnd *cmd) + * Function : void initialize_SCp(struct scsi_cmnd *cmd) * * Purpose : initialize the saved data pointers for cmd to point to the * start of the buffer. * - * Inputs : cmd - Scsi_Cmnd structure to have pointers reset. + * Inputs : cmd - struct scsi_cmnd structure to have pointers reset. */ -static __inline__ void initialize_SCp(Scsi_Cmnd *cmd) +static __inline__ void initialize_SCp(struct scsi_cmnd *cmd) { /* * Initialize the Scsi Pointer field so that all of the commands in the @@ -753,14 +753,15 @@ static void NCR5380_print_status (struct Scsi_Host *instance) do { if (pos + strlen(fmt) + 20 /* slop */ < buffer + length) \ pos += sprintf(pos, fmt , ## args); } while(0) static -char *lprint_Scsi_Cmnd (Scsi_Cmnd *cmd, char *pos, char *buffer, int length); +char *lprint_Scsi_Cmnd(struct scsi_cmnd *cmd, char *pos, char *buffer, + int length); -static int NCR5380_proc_info (struct Scsi_Host *instance, char *buffer, char **start, - off_t offset, int length, int inout) +static int NCR5380_proc_info(struct Scsi_Host *instance, char *buffer, + char **start, off_t offset, int length, int inout) { char *pos = buffer; struct NCR5380_hostdata *hostdata; - Scsi_Cmnd *ptr; + struct scsi_cmnd *ptr; unsigned long flags; off_t begin = 0; #define check_offset() \ @@ -784,18 +785,19 @@ static int NCR5380_proc_info (struct Scsi_Host *instance, char *buffer, char **s if (!hostdata->connected) SPRINTF("scsi%d: no currently connected command\n", HOSTNO); else - pos = lprint_Scsi_Cmnd ((Scsi_Cmnd *) hostdata->connected, + pos = lprint_Scsi_Cmnd ((struct scsi_cmnd *) hostdata->connected, pos, buffer, length); SPRINTF("scsi%d: issue_queue\n", HOSTNO); check_offset(); - for (ptr = (Scsi_Cmnd *) hostdata->issue_queue; ptr; ptr = NEXT(ptr)) { + for (ptr = (struct scsi_cmnd *) hostdata->issue_queue; ptr; ptr = NEXT(ptr)) + { pos = lprint_Scsi_Cmnd (ptr, pos, buffer, length); check_offset(); } SPRINTF("scsi%d: disconnected_queue\n", HOSTNO); check_offset(); - for (ptr = (Scsi_Cmnd *) hostdata->disconnected_queue; ptr; + for (ptr = (struct scsi_cmnd *) hostdata->disconnected_queue; ptr; ptr = NEXT(ptr)) { pos = lprint_Scsi_Cmnd (ptr, pos, buffer, length); check_offset(); @@ -810,8 +812,8 @@ static int NCR5380_proc_info (struct Scsi_Host *instance, char *buffer, char **s return length; } -static char * -lprint_Scsi_Cmnd (Scsi_Cmnd *cmd, char *pos, char *buffer, int length) +static char *lprint_Scsi_Cmnd(struct scsi_cmnd *cmd, char *pos, char *buffer, + int length) { int i, s; unsigned char *command; @@ -888,8 +890,8 @@ static int NCR5380_init (struct Scsi_Host *instance, int flags) } /* - * Function : int NCR5380_queue_command (Scsi_Cmnd *cmd, - * void (*done)(Scsi_Cmnd *)) + * Function : int NCR5380_queue_command (struct scsi_cmnd *cmd, + * void (*done)(struct scsi_cmnd *)) * * Purpose : enqueues a SCSI command * @@ -906,10 +908,11 @@ static int NCR5380_init (struct Scsi_Host *instance, int flags) */ /* Only make static if a wrapper function is used */ -static int NCR5380_queue_command (Scsi_Cmnd *cmd, void (*done)(Scsi_Cmnd *)) +static int NCR5380_queue_command(struct scsi_cmnd *cmd, + void (*done)(struct scsi_cmnd *)) { SETUP_HOSTDATA(cmd->device->host); - Scsi_Cmnd *tmp; + struct scsi_cmnd *tmp; unsigned long flags; #if (NDEBUG & NDEBUG_NO_WRITE) @@ -990,7 +993,7 @@ static int NCR5380_queue_command (Scsi_Cmnd *cmd, void (*done)(Scsi_Cmnd *)) NEXT(cmd) = hostdata->issue_queue; hostdata->issue_queue = cmd; } else { - for (tmp = (Scsi_Cmnd *)hostdata->issue_queue; + for (tmp = (struct scsi_cmnd *)hostdata->issue_queue; NEXT(tmp); tmp = NEXT(tmp)) ; LIST(cmd, tmp); @@ -1030,7 +1033,7 @@ static int NCR5380_queue_command (Scsi_Cmnd *cmd, void (*done)(Scsi_Cmnd *)) static void NCR5380_main (void *bl) { - Scsi_Cmnd *tmp, *prev; + struct scsi_cmnd *tmp, *prev; struct Scsi_Host *instance = first_instance; struct NCR5380_hostdata *hostdata = HOSTDATA(instance); int done; @@ -1073,12 +1076,12 @@ static void NCR5380_main (void *bl) * for a target that's not busy. */ #if (NDEBUG & NDEBUG_LISTS) - for (tmp = (Scsi_Cmnd *) hostdata->issue_queue, prev = NULL; + for (tmp = (struct scsi_cmnd *) hostdata->issue_queue, prev = NULL; tmp && (tmp != prev); prev = tmp, tmp = NEXT(tmp)) ; if ((tmp == prev) && tmp) printk(" LOOP\n");/* else printk("\n");*/ #endif - for (tmp = (Scsi_Cmnd *) hostdata->issue_queue, + for (tmp = (struct scsi_cmnd *) hostdata->issue_queue, prev = NULL; tmp; prev = tmp, tmp = NEXT(tmp) ) { #if (NDEBUG & NDEBUG_LISTS) @@ -1339,7 +1342,8 @@ static irqreturn_t NCR5380_intr (int irq, void *dev_id) } #ifdef NCR5380_STATS -static void collect_stats(struct NCR5380_hostdata* hostdata, Scsi_Cmnd* cmd) +static void collect_stats(struct NCR5380_hostdata *hostdata, + struct scsi_cmnd *cmd) { # ifdef NCR5380_STAT_LIMIT if (cmd->request_bufflen > NCR5380_STAT_LIMIT) @@ -1365,8 +1369,8 @@ static void collect_stats(struct NCR5380_hostdata* hostdata, Scsi_Cmnd* cmd) #endif /* - * Function : int NCR5380_select (struct Scsi_Host *instance, Scsi_Cmnd *cmd, - * int tag); + * Function : int NCR5380_select(struct Scsi_Host *instance, + * struct scsi_cmnd *cmd, int tag); * * Purpose : establishes I_T_L or I_T_L_Q nexus for new or existing command, * including ARBITRATION, SELECTION, and initial message out for @@ -1395,7 +1399,8 @@ static void collect_stats(struct NCR5380_hostdata* hostdata, Scsi_Cmnd* cmd) * cmd->result host byte set to DID_BAD_TARGET. */ -static int NCR5380_select (struct Scsi_Host *instance, Scsi_Cmnd *cmd, int tag) +static int NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd, + int tag) { SETUP_HOSTDATA(instance); unsigned char tmp[3], phase; @@ -1985,7 +1990,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance) #endif unsigned char *data; unsigned char phase, tmp, extended_msg[10], old_phase=0xff; - Scsi_Cmnd *cmd = (Scsi_Cmnd *) hostdata->connected; + struct scsi_cmnd *cmd = (struct scsi_cmnd *) hostdata->connected; #ifdef SUN3_SCSI_VME dregs->csr |= CSR_INTR; @@ -2272,7 +2277,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance) local_irq_save(flags); LIST(cmd,hostdata->issue_queue); NEXT(cmd) = hostdata->issue_queue; - hostdata->issue_queue = (Scsi_Cmnd *) cmd; + hostdata->issue_queue = (struct scsi_cmnd *) cmd; local_irq_restore(flags); QU_PRINTK("scsi%d: REQUEST SENSE added to head of " "issue queue\n", H_NO(cmd)); @@ -2502,7 +2507,7 @@ static void NCR5380_information_transfer (struct Scsi_Host *instance) * Function : void NCR5380_reselect (struct Scsi_Host *instance) * * Purpose : does reselection, initializing the instance->connected - * field to point to the Scsi_Cmnd for which the I_T_L or I_T_L_Q + * field to point to the struct scsi_cmnd for which the I_T_L or I_T_L_Q * nexus has been reestablished, * * Inputs : instance - this instance of the NCR5380. @@ -2521,7 +2526,7 @@ static void NCR5380_reselect (struct Scsi_Host *instance) unsigned char tag; #endif unsigned char msg[3]; - Scsi_Cmnd *tmp = NULL, *prev; + struct scsi_cmnd *tmp = NULL, *prev; /* unsigned long flags; */ /* @@ -2577,7 +2582,7 @@ static void NCR5380_reselect (struct Scsi_Host *instance) * just reestablished, and remove it from the disconnected queue. */ - for (tmp = (Scsi_Cmnd *) hostdata->disconnected_queue, prev = NULL; + for (tmp = (struct scsi_cmnd *) hostdata->disconnected_queue, prev = NULL; tmp; prev = tmp, tmp = NEXT(tmp) ) { if ((target_mask == (1 << tmp->device->id)) && (lun == tmp->device->lun) #ifdef SUPPORT_TAGS @@ -2668,11 +2673,11 @@ static void NCR5380_reselect (struct Scsi_Host *instance) /* - * Function : int NCR5380_abort (Scsi_Cmnd *cmd) + * Function : int NCR5380_abort(struct scsi_cmnd *cmd) * * Purpose : abort a command * - * Inputs : cmd - the Scsi_Cmnd to abort, code - code to set the + * Inputs : cmd - the struct scsi_cmnd to abort, code - code to set the * host byte of the result field to, if zero DID_ABORTED is * used. * @@ -2684,11 +2689,11 @@ static void NCR5380_reselect (struct Scsi_Host *instance) * called where the loop started in NCR5380_main(). */ -static int NCR5380_abort (Scsi_Cmnd *cmd) +static int NCR5380_abort(struct scsi_cmnd *cmd) { struct Scsi_Host *instance = cmd->device->host; SETUP_HOSTDATA(instance); - Scsi_Cmnd *tmp, **prev; + struct scsi_cmnd *tmp, **prev; unsigned long flags; printk(KERN_NOTICE "scsi%d: aborting command\n", HOSTNO); @@ -2753,9 +2758,9 @@ static int NCR5380_abort (Scsi_Cmnd *cmd) * Case 2 : If the command hasn't been issued yet, we simply remove it * from the issue queue. */ - for (prev = (Scsi_Cmnd **) &(hostdata->issue_queue), - tmp = (Scsi_Cmnd *) hostdata->issue_queue; - tmp; prev = NEXTADDR(tmp), tmp = NEXT(tmp) ) + for (prev = (struct scsi_cmnd **) &(hostdata->issue_queue), + tmp = (struct scsi_cmnd *) hostdata->issue_queue; + tmp; prev = NEXTADDR(tmp), tmp = NEXT(tmp)) if (cmd == tmp) { REMOVE(5, *prev, tmp, NEXT(tmp)); (*prev) = NEXT(tmp); @@ -2812,7 +2817,7 @@ static int NCR5380_abort (Scsi_Cmnd *cmd) * it from the disconnected queue. */ - for (tmp = (Scsi_Cmnd *) hostdata->disconnected_queue; tmp; + for (tmp = (struct scsi_cmnd *) hostdata->disconnected_queue; tmp; tmp = NEXT(tmp)) if (cmd == tmp) { local_irq_restore(flags); @@ -2826,8 +2831,8 @@ static int NCR5380_abort (Scsi_Cmnd *cmd) do_abort (instance); local_irq_save(flags); - for (prev = (Scsi_Cmnd **) &(hostdata->disconnected_queue), - tmp = (Scsi_Cmnd *) hostdata->disconnected_queue; + for (prev = (struct scsi_cmnd **) &(hostdata->disconnected_queue), + tmp = (struct scsi_cmnd *) hostdata->disconnected_queue; tmp; prev = NEXTADDR(tmp), tmp = NEXT(tmp) ) if (cmd == tmp) { REMOVE(5, *prev, tmp, NEXT(tmp)); @@ -2868,7 +2873,7 @@ static int NCR5380_abort (Scsi_Cmnd *cmd) /* - * Function : int NCR5380_bus_reset (Scsi_Cmnd *cmd) + * Function : int NCR5380_bus_reset(struct scsi_cmnd *cmd) * * Purpose : reset the SCSI bus. * @@ -2876,13 +2881,13 @@ static int NCR5380_abort (Scsi_Cmnd *cmd) * */ -static int NCR5380_bus_reset( Scsi_Cmnd *cmd) +static int NCR5380_bus_reset(struct scsi_cmnd *cmd) { SETUP_HOSTDATA(cmd->device->host); int i; unsigned long flags; #if 1 - Scsi_Cmnd *connected, *disconnected_queue; + struct scsi_cmnd *connected, *disconnected_queue; #endif @@ -2914,9 +2919,9 @@ static int NCR5380_bus_reset( Scsi_Cmnd *cmd) * remembered in local variables first. */ local_irq_save(flags); - connected = (Scsi_Cmnd *)hostdata->connected; + connected = (struct scsi_cmnd *)hostdata->connected; hostdata->connected = NULL; - disconnected_queue = (Scsi_Cmnd *)hostdata->disconnected_queue; + disconnected_queue = (struct scsi_cmnd *)hostdata->disconnected_queue; hostdata->disconnected_queue = NULL; #ifdef SUPPORT_TAGS free_all_tags(); diff --git a/drivers/scsi/sun3_scsi.c b/drivers/scsi/sun3_scsi.c index e625b4c..d56d85d 100644 --- a/drivers/scsi/sun3_scsi.c +++ b/drivers/scsi/sun3_scsi.c @@ -119,7 +119,7 @@ module_param(setup_use_tagged_queuing, int, 0); static int setup_hostid = -1; module_param(setup_hostid, int, 0); -static Scsi_Cmnd *sun3_dma_setup_done = NULL; +static struct scsi_cmnd *sun3_dma_setup_done = NULL; #define AFTER_RESET_DELAY (HZ/2) @@ -521,8 +521,9 @@ static inline unsigned long sun3scsi_dma_residual(struct Scsi_Host *instance) return last_residual; } -static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted, Scsi_Cmnd *cmd, - int write_flag) +static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted, + struct scsi_cmnd *cmd, + int write_flag) { if(blk_fs_request(cmd->request)) return wanted; diff --git a/drivers/scsi/sun3_scsi.h b/drivers/scsi/sun3_scsi.h index 834dab4..a1103b3 100644 --- a/drivers/scsi/sun3_scsi.h +++ b/drivers/scsi/sun3_scsi.h @@ -47,11 +47,12 @@ #define IOBASE_SUN3_VMESCSI 0xff200000 -static int sun3scsi_abort (Scsi_Cmnd *); +static int sun3scsi_abort(struct scsi_cmnd *); static int sun3scsi_detect (struct scsi_host_template *); static const char *sun3scsi_info (struct Scsi_Host *); -static int sun3scsi_bus_reset(Scsi_Cmnd *); -static int sun3scsi_queue_command (Scsi_Cmnd *, void (*done)(Scsi_Cmnd *)); +static int sun3scsi_bus_reset(struct scsi_cmnd *); +static int sun3scsi_queue_command(struct scsi_cmnd *, + void (*done)(struct scsi_cmnd *)); static int sun3scsi_release (struct Scsi_Host *); #ifndef CMD_PER_LUN diff --git a/drivers/scsi/sun3_scsi_vme.c b/drivers/scsi/sun3_scsi_vme.c index e8faab1..92def31 100644 --- a/drivers/scsi/sun3_scsi_vme.c +++ b/drivers/scsi/sun3_scsi_vme.c @@ -84,7 +84,7 @@ module_param(setup_use_tagged_queuing, int, 0); static int setup_hostid = -1; module_param(setup_hostid, int, 0); -static Scsi_Cmnd *sun3_dma_setup_done = NULL; +static struct scsi_cmnd *sun3_dma_setup_done = NULL; #define AFTER_RESET_DELAY (HZ/2) @@ -455,8 +455,9 @@ static inline unsigned long sun3scsi_dma_residual(struct Scsi_Host *instance) return last_residual; } -static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted, Scsi_Cmnd *cmd, - int write_flag) +static inline unsigned long sun3scsi_dma_xfer_len(unsigned long wanted, + struct scsi_cmnd *cmd, + int write_flag) { if(blk_fs_request(cmd->request)) return wanted; -- cgit v0.10.2 From a24342b90c9c829fc5fea9ee01b127f81bca18ef Mon Sep 17 00:00:00 2001 From: Henne Date: Tue, 3 Oct 2006 21:31:14 +0200 Subject: [SCSI] Scsi_Cmnd conversion in qlogicfas408 driver Change obsolete Scsi_Cmnd to struct scsi_cmnd in the Qlocic FAS408 driver. Signed-off-by: Henrik Kretzschmar rejections fixed and Signed-off-by: James Bottomley diff --git a/drivers/scsi/qlogicfas408.c b/drivers/scsi/qlogicfas408.c index e072535..2e7db18 100644 --- a/drivers/scsi/qlogicfas408.c +++ b/drivers/scsi/qlogicfas408.c @@ -209,7 +209,7 @@ static int ql_wai(struct qlogicfas408_priv *priv) * caller must hold host lock */ -static void ql_icmd(Scsi_Cmnd * cmd) +static void ql_icmd(struct scsi_cmnd *cmd) { struct qlogicfas408_priv *priv = get_priv_by_cmd(cmd); int qbase = priv->qbase; @@ -256,7 +256,7 @@ static void ql_icmd(Scsi_Cmnd * cmd) * Process scsi command - usually after interrupt */ -static unsigned int ql_pcmd(Scsi_Cmnd * cmd) +static unsigned int ql_pcmd(struct scsi_cmnd *cmd) { unsigned int i, j; unsigned long k; @@ -407,7 +407,7 @@ static unsigned int ql_pcmd(Scsi_Cmnd * cmd) static void ql_ihandl(void *dev_id) { - Scsi_Cmnd *icmd; + struct scsi_cmnd *icmd; struct Scsi_Host *host = dev_id; struct qlogicfas408_priv *priv = get_priv_by_host(host); int qbase = priv->qbase; @@ -447,7 +447,8 @@ irqreturn_t qlogicfas408_ihandl(int irq, void *dev_id) * Queued command */ -int qlogicfas408_queuecommand(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)) +int qlogicfas408_queuecommand(struct scsi_cmnd *cmd, + void (*done) (struct scsi_cmnd *)) { struct qlogicfas408_priv *priv = get_priv_by_cmd(cmd); if (scmd_id(cmd) == priv->qinitid) { @@ -470,9 +471,8 @@ int qlogicfas408_queuecommand(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)) * Return bios parameters */ -int qlogicfas408_biosparam(struct scsi_device * disk, - struct block_device *dev, - sector_t capacity, int ip[]) +int qlogicfas408_biosparam(struct scsi_device *disk, struct block_device *dev, + sector_t capacity, int ip[]) { /* This should mimic the DOS Qlogic driver's behavior exactly */ ip[0] = 0x40; @@ -494,7 +494,7 @@ int qlogicfas408_biosparam(struct scsi_device * disk, * Abort a command in progress */ -int qlogicfas408_abort(Scsi_Cmnd * cmd) +int qlogicfas408_abort(struct scsi_cmnd *cmd) { struct qlogicfas408_priv *priv = get_priv_by_cmd(cmd); priv->qabort = 1; @@ -508,7 +508,7 @@ int qlogicfas408_abort(Scsi_Cmnd * cmd) * the PCMCIA qlogic_stub code. This wants fixing */ -int qlogicfas408_bus_reset(Scsi_Cmnd * cmd) +int qlogicfas408_bus_reset(struct scsi_cmnd *cmd) { struct qlogicfas408_priv *priv = get_priv_by_cmd(cmd); unsigned long flags; diff --git a/drivers/scsi/qlogicfas408.h b/drivers/scsi/qlogicfas408.h index 8fd5555..2606264 100644 --- a/drivers/scsi/qlogicfas408.h +++ b/drivers/scsi/qlogicfas408.h @@ -75,15 +75,15 @@ /*----------------------------------------------------------------*/ struct qlogicfas408_priv { - int qbase; /* Port */ - int qinitid; /* initiator ID */ - int qabort; /* Flag to cause an abort */ - int qlirq; /* IRQ being used */ - int int_type; /* type of irq, 2 for ISA board, 0 for PCMCIA */ - char qinfo[80]; /* description */ - Scsi_Cmnd *qlcmd; /* current command being processed */ - struct Scsi_Host *shost; /* pointer back to host */ - struct qlogicfas408_priv *next; /* next private struct */ + int qbase; /* Port */ + int qinitid; /* initiator ID */ + int qabort; /* Flag to cause an abort */ + int qlirq; /* IRQ being used */ + int int_type; /* type of irq, 2 for ISA board, 0 for PCMCIA */ + char qinfo[80]; /* description */ + struct scsi_cmnd *qlcmd; /* current command being processed */ + struct Scsi_Host *shost; /* pointer back to host */ + struct qlogicfas408_priv *next; /* next private struct */ }; /* The qlogic card uses two register maps - These macros select which one */ @@ -103,12 +103,13 @@ struct qlogicfas408_priv { #define get_priv_by_host(x) (struct qlogicfas408_priv *)&((x)->hostdata[0]) irqreturn_t qlogicfas408_ihandl(int irq, void *dev_id); -int qlogicfas408_queuecommand(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)); +int qlogicfas408_queuecommand(struct scsi_cmnd * cmd, + void (*done) (struct scsi_cmnd *)); int qlogicfas408_biosparam(struct scsi_device * disk, - struct block_device *dev, - sector_t capacity, int ip[]); -int qlogicfas408_abort(Scsi_Cmnd * cmd); -int qlogicfas408_bus_reset(Scsi_Cmnd * cmd); + struct block_device *dev, + sector_t capacity, int ip[]); +int qlogicfas408_abort(struct scsi_cmnd * cmd); +int qlogicfas408_bus_reset(struct scsi_cmnd * cmd); const char *qlogicfas408_info(struct Scsi_Host *host); int qlogicfas408_get_chip_type(int qbase, int int_type); void qlogicfas408_setup(int qbase, int id, int int_type); -- cgit v0.10.2 From 413f73272090a69e35a03c938272ec661b1d3d4c Mon Sep 17 00:00:00 2001 From: Kai Makisara Date: Thu, 5 Oct 2006 22:59:46 +0300 Subject: [SCSI] st: Fixup -ENOMEDIUM Based on the original patch from Hannes Reinecke Fix st_open() to return -ENOMEDIUM instead of -EIO if no medium is found. Signed-off-by: Kai Makisara Signed-off-by: James Bottomley diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c index 3babdc7..e1a52c5 100644 --- a/drivers/scsi/st.c +++ b/drivers/scsi/st.c @@ -1177,7 +1177,10 @@ static int st_open(struct inode *inode, struct file *filp) goto err_out; if ((filp->f_flags & O_NONBLOCK) == 0 && retval != CHKRES_READY) { - retval = (-EIO); + if (STp->ready == NO_TAPE) + retval = (-ENOMEDIUM); + else + retval = (-EIO); goto err_out; } return 0; -- cgit v0.10.2 From 75c28851c9eee889ef4347ff6f55b2dd1e1ceb81 Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Fri, 6 Oct 2006 00:11:17 +0200 Subject: [SCSI] tmscsim: set max_sectors AM53C974A's Start Transfer Counter register has 24 bits, thus maximum transfer length is 16MiB. But the maximum I can test is 8MiB, so use that until somebody tests 16MiB. Signed-off-by: Guennadi Liakhovetski Signed-off-by: James Bottomley diff --git a/drivers/scsi/tmscsim.c b/drivers/scsi/tmscsim.c index d03aa6c..fa5382e 100644 --- a/drivers/scsi/tmscsim.c +++ b/drivers/scsi/tmscsim.c @@ -2304,6 +2304,7 @@ static struct scsi_host_template driver_template = { .sg_tablesize = SG_ALL, .cmd_per_lun = 1, .use_clustering = ENABLE_CLUSTERING, + .max_sectors = 0x4000, /* 8MiB = 16 * 1024 * 512 */ }; /*********************************************************************** -- cgit v0.10.2 From 5ae16db36988e811410494fb5d07c81e14453e7b Mon Sep 17 00:00:00 2001 From: Doug Maxey Date: Thu, 5 Oct 2006 23:50:07 -0500 Subject: [SCSI] qla4xxx: fix double printk on load There is a dup printk at the tail of qla4xxx_module_init(). Remove the first instance as it's before the complete success of the function. Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 178fcdd..4fa0153 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -1724,13 +1724,13 @@ static int __init qla4xxx_module_init(void) goto release_srb_cache; } - printk(KERN_INFO "QLogic iSCSI HBA Driver\n"); ret = pci_register_driver(&qla4xxx_pci_driver); if (ret) goto unregister_transport; printk(KERN_INFO "QLogic iSCSI HBA Driver\n"); return 0; + unregister_transport: iscsi_unregister_transport(&qla4xxx_iscsi_transport); release_srb_cache: -- cgit v0.10.2 From 80f1443c66de3ec42e28d151bd43a80de398877e Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Fri, 6 Oct 2006 09:22:41 +0200 Subject: [SCSI] aic7xxx: Adjust .max_sectors According to the adaptec sources aic7xxx / aic79xx really can do 4MB transfers. So we should adjust .max_sectors. Signed-off-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.c b/drivers/scsi/aic7xxx/aic79xx_osm.c index f8e6048..d8d6687 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm.c +++ b/drivers/scsi/aic7xxx/aic79xx_osm.c @@ -773,6 +773,7 @@ struct scsi_host_template aic79xx_driver_template = { #endif .can_queue = AHD_MAX_QUEUE, .this_id = -1, + .max_sectors = 8192, .cmd_per_lun = 2, .use_clustering = ENABLE_CLUSTERING, .slave_alloc = ahd_linux_slave_alloc, diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.c b/drivers/scsi/aic7xxx/aic7xxx_osm.c index 43ab753..ad8578e 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.c @@ -777,6 +777,7 @@ struct scsi_host_template aic7xxx_driver_template = { #endif .can_queue = AHC_MAX_QUEUE, .this_id = -1, + .max_sectors = 8192, .cmd_per_lun = 2, .use_clustering = ENABLE_CLUSTERING, .slave_alloc = ahc_linux_slave_alloc, -- cgit v0.10.2 From 11010fecd2a1fdae684237b61709367ae6a93289 Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Fri, 6 Oct 2006 09:54:59 -0700 Subject: [SCSI] Maintain module-parameter name consistency with qla2xxx/qla4xxx. Signed-off-by: Andrew Vasquez Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_dbg.h b/drivers/scsi/qla2xxx/qla_dbg.h index 90dad7e..5b12278 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.h +++ b/drivers/scsi/qla2xxx/qla_dbg.h @@ -38,7 +38,7 @@ * Macros use for debugging the driver. */ -#define DEBUG(x) do { if (qla2_extended_error_logging) { x; } } while (0) +#define DEBUG(x) do { if (ql2xextended_error_logging) { x; } } while (0) #if defined(QL_DEBUG_LEVEL_1) #define DEBUG1(x) do {x;} while (0) @@ -46,12 +46,12 @@ #define DEBUG1(x) do {} while (0) #endif -#define DEBUG2(x) do { if (qla2_extended_error_logging) { x; } } while (0) -#define DEBUG2_3(x) do { if (qla2_extended_error_logging) { x; } } while (0) -#define DEBUG2_3_11(x) do { if (qla2_extended_error_logging) { x; } } while (0) -#define DEBUG2_9_10(x) do { if (qla2_extended_error_logging) { x; } } while (0) -#define DEBUG2_11(x) do { if (qla2_extended_error_logging) { x; } } while (0) -#define DEBUG2_13(x) do { if (qla2_extended_error_logging) { x; } } while (0) +#define DEBUG2(x) do { if (ql2xextended_error_logging) { x; } } while (0) +#define DEBUG2_3(x) do { if (ql2xextended_error_logging) { x; } } while (0) +#define DEBUG2_3_11(x) do { if (ql2xextended_error_logging) { x; } } while (0) +#define DEBUG2_9_10(x) do { if (ql2xextended_error_logging) { x; } } while (0) +#define DEBUG2_11(x) do { if (ql2xextended_error_logging) { x; } } while (0) +#define DEBUG2_13(x) do { if (ql2xextended_error_logging) { x; } } while (0) #if defined(QL_DEBUG_LEVEL_3) #define DEBUG3(x) do {x;} while (0) diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h index 7da6983..b51ce8f 100644 --- a/drivers/scsi/qla2xxx/qla_gbl.h +++ b/drivers/scsi/qla2xxx/qla_gbl.h @@ -60,7 +60,7 @@ extern int ql2xplogiabsentdevice; extern int ql2xloginretrycount; extern int ql2xfdmienable; extern int ql2xallocfwdump; -extern int qla2_extended_error_logging; +extern int ql2xextended_error_logging; extern void qla2x00_sp_compl(scsi_qla_host_t *, srb_t *); diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index 833b930..d5e0a12 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -1644,7 +1644,7 @@ qla2x00_nvram_config(scsi_qla_host_t *ha) * Set host adapter parameters. */ if (nv->host_p[0] & BIT_7) - qla2_extended_error_logging = 1; + ql2xextended_error_logging = 1; ha->flags.disable_risc_code_load = ((nv->host_p[0] & BIT_4) ? 1 : 0); /* Always load RISC code on non ISP2[12]00 chips. */ if (!IS_QLA2100(ha) && !IS_QLA2200(ha)) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 3f20d76..34b6eb7 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -61,9 +61,9 @@ MODULE_PARM_DESC(ql2xallocfwdump, "during HBA initialization. Memory allocation requirements " "vary by ISP type. Default is 1 - allocate memory."); -int qla2_extended_error_logging; -module_param(qla2_extended_error_logging, int, S_IRUGO|S_IRUSR); -MODULE_PARM_DESC(qla2_extended_error_logging, +int ql2xextended_error_logging; +module_param(ql2xextended_error_logging, int, S_IRUGO|S_IRUSR); +MODULE_PARM_DESC(ql2xextended_error_logging, "Option to enable extended error logging, " "Default is 0 - no logging. 1 - log errors."); @@ -2697,7 +2697,7 @@ qla2x00_module_init(void) /* Derive version string. */ strcpy(qla2x00_version_str, QLA2XXX_VERSION); - if (qla2_extended_error_logging) + if (ql2xextended_error_logging) strcat(qla2x00_version_str, "-debug"); qla2xxx_transport_template = diff --git a/drivers/scsi/qla4xxx/ql4_dbg.h b/drivers/scsi/qla4xxx/ql4_dbg.h index 3e99dcf..d861c3b 100644 --- a/drivers/scsi/qla4xxx/ql4_dbg.h +++ b/drivers/scsi/qla4xxx/ql4_dbg.h @@ -22,14 +22,14 @@ #endif #if defined(QL_DEBUG_LEVEL_2) -#define DEBUG2(x) do {if(qla4_extended_error_logging == 2) x;} while (0); +#define DEBUG2(x) do {if(ql4xextended_error_logging == 2) x;} while (0); #define DEBUG2_3(x) do {x;} while (0); #else /* */ #define DEBUG2(x) do {} while (0); #endif /* */ #if defined(QL_DEBUG_LEVEL_3) -#define DEBUG3(x) do {if(qla4_extended_error_logging == 3) x;} while (0); +#define DEBUG3(x) do {if(ql4xextended_error_logging == 3) x;} while (0); #else /* */ #define DEBUG3(x) do {} while (0); #if !defined(QL_DEBUG_LEVEL_2) diff --git a/drivers/scsi/qla4xxx/ql4_glbl.h b/drivers/scsi/qla4xxx/ql4_glbl.h index 2c803ed..1b221ff 100644 --- a/drivers/scsi/qla4xxx/ql4_glbl.h +++ b/drivers/scsi/qla4xxx/ql4_glbl.h @@ -72,7 +72,7 @@ int qla4xxx_reinitialize_ddb_list(struct scsi_qla_host * ha); int qla4xxx_process_ddb_changed(struct scsi_qla_host * ha, uint32_t fw_ddb_index, uint32_t state); -extern int qla4_extended_error_logging; +extern int ql4xextended_error_logging; extern int ql4xdiscoverywait; extern int ql4xdontresethba; #endif /* _QLA4x_GBL_H */ diff --git a/drivers/scsi/qla4xxx/ql4_mbx.c b/drivers/scsi/qla4xxx/ql4_mbx.c index ef82399..b721dc5 100644 --- a/drivers/scsi/qla4xxx/ql4_mbx.c +++ b/drivers/scsi/qla4xxx/ql4_mbx.c @@ -701,7 +701,7 @@ void qla4xxx_get_conn_event_log(struct scsi_qla_host * ha) DEBUG3(printk("scsi%ld: Connection Event Log Dump (%d entries):\n", ha->host_no, num_valid_entries)); - if (qla4_extended_error_logging == 3) { + if (ql4xextended_error_logging == 3) { if (oldest_entry == 0) { /* Circular Buffer has not wrapped around */ for (i=0; i < num_valid_entries; i++) { diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 4fa0153..5b8db61 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -34,9 +34,9 @@ MODULE_PARM_DESC(ql4xdontresethba, " default it will reset hba :0" " set to 1 to avoid resetting HBA"); -int qla4_extended_error_logging = 0; /* 0 = off, 1 = log errors */ -module_param(qla4_extended_error_logging, int, S_IRUGO | S_IRUSR); -MODULE_PARM_DESC(qla4_extended_error_logging, +int ql4xextended_error_logging = 0; /* 0 = off, 1 = log errors */ +module_param(ql4xextended_error_logging, int, S_IRUGO | S_IRUSR); +MODULE_PARM_DESC(ql4xextended_error_logging, "Option to enable extended error logging, " "Default is 0 - no logging, 1 - debug logging"); @@ -1714,7 +1714,7 @@ static int __init qla4xxx_module_init(void) /* Derive version string. */ strcpy(qla4xxx_version_str, QLA4XXX_DRIVER_VERSION); - if (qla4_extended_error_logging) + if (ql4xextended_error_logging) strcat(qla4xxx_version_str, "-debug"); qla4xxx_scsi_transport = -- cgit v0.10.2 From 35508e46aae4b57bd07d095eb11533e296b254dc Mon Sep 17 00:00:00 2001 From: Michael Reed Date: Fri, 6 Oct 2006 15:39:25 -0500 Subject: [SCSI] mptfc: stall eh handlers if resetting while rport blocked Thanks to James Smart for the inspiration. Stall error handler if attempting recovery while an rport is blocked. This avoids device offline scenarios due to errors in the error handler. Also verify that VirtDevice is available before issuing scsi command. VirtDevice is removed when fc transport removes a target. See James Smart's patch of 08/17/2006 for greater detail. http://marc.theaimsgroup.com/?l=linux-scsi&m=115583213624803&w=2 Also bump version number per Eric's request. Signed-off-by: Michael Reed Acked-by: Eric Moore Signed-off-by: James Bottomley diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h index c537d71..a4afad4 100644 --- a/drivers/message/fusion/mptbase.h +++ b/drivers/message/fusion/mptbase.h @@ -75,8 +75,8 @@ #define COPYRIGHT "Copyright (c) 1999-2005 " MODULEAUTHOR #endif -#define MPT_LINUX_VERSION_COMMON "3.04.01" -#define MPT_LINUX_PACKAGE_NAME "@(#)mptlinux-3.04.01" +#define MPT_LINUX_VERSION_COMMON "3.04.02" +#define MPT_LINUX_PACKAGE_NAME "@(#)mptlinux-3.04.02" #define WHAT_MAGIC_STRING "@" "(" "#" ")" #define show_mptmod_ver(s,ver) \ diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c index e57bb03..1dd4917 100644 --- a/drivers/message/fusion/mptfc.c +++ b/drivers/message/fusion/mptfc.c @@ -96,6 +96,10 @@ static int mptfc_qcmd(struct scsi_cmnd *SCpnt, static void mptfc_target_destroy(struct scsi_target *starget); static void mptfc_set_rport_loss_tmo(struct fc_rport *rport, uint32_t timeout); static void __devexit mptfc_remove(struct pci_dev *pdev); +static int mptfc_abort(struct scsi_cmnd *SCpnt); +static int mptfc_dev_reset(struct scsi_cmnd *SCpnt); +static int mptfc_bus_reset(struct scsi_cmnd *SCpnt); +static int mptfc_host_reset(struct scsi_cmnd *SCpnt); static struct scsi_host_template mptfc_driver_template = { .module = THIS_MODULE, @@ -110,10 +114,10 @@ static struct scsi_host_template mptfc_driver_template = { .target_destroy = mptfc_target_destroy, .slave_destroy = mptscsih_slave_destroy, .change_queue_depth = mptscsih_change_queue_depth, - .eh_abort_handler = mptscsih_abort, - .eh_device_reset_handler = mptscsih_dev_reset, - .eh_bus_reset_handler = mptscsih_bus_reset, - .eh_host_reset_handler = mptscsih_host_reset, + .eh_abort_handler = mptfc_abort, + .eh_device_reset_handler = mptfc_dev_reset, + .eh_bus_reset_handler = mptfc_bus_reset, + .eh_host_reset_handler = mptfc_host_reset, .bios_param = mptscsih_bios_param, .can_queue = MPT_FC_CAN_QUEUE, .this_id = -1, @@ -171,6 +175,77 @@ static struct fc_function_template mptfc_transport_functions = { .show_host_symbolic_name = 1, }; +static int +mptfc_block_error_handler(struct scsi_cmnd *SCpnt, + int (*func)(struct scsi_cmnd *SCpnt), + const char *caller) +{ + struct scsi_device *sdev = SCpnt->device; + struct Scsi_Host *shost = sdev->host; + struct fc_rport *rport = starget_to_rport(scsi_target(sdev)); + unsigned long flags; + int ready; + + spin_lock_irqsave(shost->host_lock, flags); + while ((ready = fc_remote_port_chkready(rport) >> 16) == DID_IMM_RETRY) { + spin_unlock_irqrestore(shost->host_lock, flags); + dfcprintk ((MYIOC_s_INFO_FMT + "mptfc_block_error_handler.%d: %d:%d, port status is " + "DID_IMM_RETRY, deferring %s recovery.\n", + ((MPT_SCSI_HOST *) shost->hostdata)->ioc->name, + ((MPT_SCSI_HOST *) shost->hostdata)->ioc->sh->host_no, + SCpnt->device->id,SCpnt->device->lun,caller)); + msleep(1000); + spin_lock_irqsave(shost->host_lock, flags); + } + spin_unlock_irqrestore(shost->host_lock, flags); + + if (ready == DID_NO_CONNECT || !SCpnt->device->hostdata) { + dfcprintk ((MYIOC_s_INFO_FMT + "%s.%d: %d:%d, failing recovery, " + "port state %d, vdev %p.\n", caller, + ((MPT_SCSI_HOST *) shost->hostdata)->ioc->name, + ((MPT_SCSI_HOST *) shost->hostdata)->ioc->sh->host_no, + SCpnt->device->id,SCpnt->device->lun,ready, + SCpnt->device->hostdata)); + return FAILED; + } + dfcprintk ((MYIOC_s_INFO_FMT + "%s.%d: %d:%d, executing recovery.\n", caller, + ((MPT_SCSI_HOST *) shost->hostdata)->ioc->name, + ((MPT_SCSI_HOST *) shost->hostdata)->ioc->sh->host_no, + SCpnt->device->id,SCpnt->device->lun)); + return (*func)(SCpnt); +} + +static int +mptfc_abort(struct scsi_cmnd *SCpnt) +{ + return + mptfc_block_error_handler(SCpnt, mptscsih_abort, __FUNCTION__); +} + +static int +mptfc_dev_reset(struct scsi_cmnd *SCpnt) +{ + return + mptfc_block_error_handler(SCpnt, mptscsih_dev_reset, __FUNCTION__); +} + +static int +mptfc_bus_reset(struct scsi_cmnd *SCpnt) +{ + return + mptfc_block_error_handler(SCpnt, mptscsih_bus_reset, __FUNCTION__); +} + +static int +mptfc_host_reset(struct scsi_cmnd *SCpnt) +{ + return + mptfc_block_error_handler(SCpnt, mptscsih_host_reset, __FUNCTION__); +} + static void mptfc_set_rport_loss_tmo(struct fc_rport *rport, uint32_t timeout) { @@ -562,6 +637,12 @@ mptfc_qcmd(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd *)) return 0; } + if (!SCpnt->device->hostdata) { /* vdev */ + SCpnt->result = DID_NO_CONNECT << 16; + done(SCpnt); + return 0; + } + /* dd_data is null until finished adding target */ ri = *((struct mptfc_rport_info **)rport->dd_data); if (unlikely(!ri)) { -- cgit v0.10.2 From 8e394aec14f24e3b41a315a2dc53537024190c8a Mon Sep 17 00:00:00 2001 From: Henne Date: Mon, 9 Oct 2006 15:38:34 +0200 Subject: [SCSI] fix typo in previous Scsi_Cmnd convertion in aic7xxx_old.c Fixes a typo in the aic7xxx_old.c. Signed-off-by: Olaf Hering Signed-off-by: Henrik Kretzschmar Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx_old.c b/drivers/scsi/aic7xxx_old.c index bcd7fff..46eed10 100644 --- a/drivers/scsi/aic7xxx_old.c +++ b/drivers/scsi/aic7xxx_old.c @@ -2646,7 +2646,7 @@ static void aic7xxx_done_cmds_complete(struct aic7xxx_host *p) while (p->completeq.head != NULL) { cmd = p->completeq.head; - p->completeq.head = (struct scsi_Cmnd *) cmd->host_scribble; + p->completeq.head = (struct scsi_cmnd *) cmd->host_scribble; cmd->host_scribble = NULL; cmd->scsi_done(cmd); } -- cgit v0.10.2 From 46c43db1eabcdc46ad9a3d711edff1d698ecd21f Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sun, 8 Oct 2006 15:55:55 +0400 Subject: [SCSI] scsi_lib.c: use BUILD_BUG_ON Signed-off-by: Alexey Dobriyan Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 743f67e..d2c02df 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1084,7 +1084,7 @@ static void scsi_setup_blk_pc_cmnd(struct scsi_cmnd *cmd) { struct request *req = cmd->request; - BUG_ON(sizeof(req->cmd) > sizeof(cmd->cmnd)); + BUILD_BUG_ON(sizeof(req->cmd) > sizeof(cmd->cmnd)); memcpy(cmd->cmnd, req->cmd, sizeof(cmd->cmnd)); cmd->cmd_len = req->cmd_len; if (!req->data_len) -- cgit v0.10.2 From 0b3a82d391563da15df2b3a0d245d41748822489 Mon Sep 17 00:00:00 2001 From: Eric Sesterhenn Date: Tue, 10 Oct 2006 14:41:43 -0700 Subject: [SCSI] lpfc: check before dereference in lpfc_ct.c If we fail to allocate mp->virt during the first while loop iteration, mlist is still uninitialized, therefore we should check if before dereferencing. Signed-off-by: Eric Sesterhenn Acked-by: James Smart Signed-off-by: Andrew Morton Signed-off-by: James Bottomley diff --git a/drivers/scsi/lpfc/lpfc_ct.c b/drivers/scsi/lpfc/lpfc_ct.c index 1b53afb..3add7c2 100644 --- a/drivers/scsi/lpfc/lpfc_ct.c +++ b/drivers/scsi/lpfc/lpfc_ct.c @@ -188,7 +188,8 @@ lpfc_alloc_ct_rsp(struct lpfc_hba * phba, int cmdcode, struct ulp_bde64 * bpl, if (!mp->virt) { kfree(mp); - lpfc_free_ct_rsp(phba, mlist); + if (mlist) + lpfc_free_ct_rsp(phba, mlist); return NULL; } -- cgit v0.10.2 From c543a3739c2a3034c80d77a189bd187c43a00feb Mon Sep 17 00:00:00 2001 From: Henrik Kretzschmar Date: Tue, 10 Oct 2006 14:41:45 -0700 Subject: [SCSI] Scsi_Cmnd conversion in psi240i driver Changes the obsolete Scsi_Cmnd to struct scsi_cmnd in psi240i-driver. Signed-off-by: Henrik Kretzschmar Signed-off-by: Andrew Morton Signed-off-by: James Bottomley diff --git a/drivers/scsi/psi240i.c b/drivers/scsi/psi240i.c index a720c92..ac0419e 100644 --- a/drivers/scsi/psi240i.c +++ b/drivers/scsi/psi240i.c @@ -87,11 +87,11 @@ typedef struct { USHORT ports[13]; OUR_DEVICE device[8]; - Scsi_Cmnd *pSCmnd; + struct scsi_cmnd *pSCmnd; IDE_STRUCT ide; ULONG startSector; USHORT sectorCount; - Scsi_Cmnd *SCpnt; + struct scsi_cmnd *SCpnt; VOID *buffer; USHORT expectingIRQ; } ADAPTER240I, *PADAPTER240I; @@ -253,12 +253,12 @@ static ULONG DecodeError (struct Scsi_Host *pshost, UCHAR status) ****************************************************************/ static void Irq_Handler (int irq, void *dev_id) { - struct Scsi_Host *shost; // Pointer to host data block - PADAPTER240I padapter; // Pointer to adapter control structure - USHORT *pports; // I/O port array - Scsi_Cmnd *SCpnt; - UCHAR status; - int z; + struct Scsi_Host *shost; // Pointer to host data block + PADAPTER240I padapter; // Pointer to adapter control structure + USHORT *pports; // I/O port array + struct scsi_cmnd *SCpnt; + UCHAR status; + int z; DEB(printk ("\npsi240i received interrupt\n")); @@ -389,12 +389,17 @@ static irqreturn_t do_Irq_Handler (int irq, void *dev_id) * Returns: Status code. * ****************************************************************/ -static int Psi240i_QueueCommand (Scsi_Cmnd *SCpnt, void (*done)(Scsi_Cmnd *)) +static int Psi240i_QueueCommand(struct scsi_cmnd *SCpnt, + void (*done)(struct scsi_cmnd *)) { - UCHAR *cdb = (UCHAR *)SCpnt->cmnd; // Pointer to SCSI CDB - PADAPTER240I padapter = HOSTDATA (SCpnt->device->host); // Pointer to adapter control structure - POUR_DEVICE pdev = &padapter->device [SCpnt->device->id];// Pointer to device information - UCHAR rc; // command return code + UCHAR *cdb = (UCHAR *)SCpnt->cmnd; + // Pointer to SCSI CDB + PADAPTER240I padapter = HOSTDATA (SCpnt->device->host); + // Pointer to adapter control structure + POUR_DEVICE pdev = &padapter->device [SCpnt->device->id]; + // Pointer to device information + UCHAR rc; + // command return code SCpnt->scsi_done = done; padapter->ide.ide.ides.spigot = pdev->spigot; diff --git a/drivers/scsi/psi240i.h b/drivers/scsi/psi240i.h index 6a59876..21ebb92 100644 --- a/drivers/scsi/psi240i.h +++ b/drivers/scsi/psi240i.h @@ -309,7 +309,7 @@ typedef struct _IDENTIFY_DATA2 { #endif // PSI_EIDE_SCSIOP // function prototypes -int Psi240i_Command (Scsi_Cmnd *SCpnt); -int Psi240i_Abort (Scsi_Cmnd *SCpnt); -int Psi240i_Reset (Scsi_Cmnd *SCpnt, unsigned int flags); +int Psi240i_Command(struct scsi_cmnd *SCpnt); +int Psi240i_Abort(struct scsi_cmnd *SCpnt); +int Psi240i_Reset(struct scsi_cmnd *SCpnt, unsigned int flags); #endif -- cgit v0.10.2 From 0fc82d5e84825ab43006f40935633120d23c2e15 Mon Sep 17 00:00:00 2001 From: Henrik Kretzschmar Date: Tue, 10 Oct 2006 14:41:41 -0700 Subject: [SCSI] convert ninja driver to struct scsi_cmnd Changes the obsolete typedefd Scsi_Cmnd to struct scsi_cmnd in the ninja scsi pcmcia driver. Signed-off-by: Henrik Kretzschmar Signed-off-by: Andrew Morton Signed-off-by: James Bottomley diff --git a/drivers/scsi/pcmcia/nsp_cs.c b/drivers/scsi/pcmcia/nsp_cs.c index b1d3460..f2d79c3 100644 --- a/drivers/scsi/pcmcia/nsp_cs.c +++ b/drivers/scsi/pcmcia/nsp_cs.c @@ -183,7 +183,7 @@ static void nsp_cs_dmessage(const char *func, int line, int mask, char *fmt, ... * Clenaup parameters and call done() functions. * You must be set SCpnt->result before call this function. */ -static void nsp_scsi_done(Scsi_Cmnd *SCpnt) +static void nsp_scsi_done(struct scsi_cmnd *SCpnt) { nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; @@ -192,7 +192,8 @@ static void nsp_scsi_done(Scsi_Cmnd *SCpnt) SCpnt->scsi_done(SCpnt); } -static int nsp_queuecommand(Scsi_Cmnd *SCpnt, void (*done)(Scsi_Cmnd *)) +static int nsp_queuecommand(struct scsi_cmnd *SCpnt, + void (*done)(struct scsi_cmnd *)) { #ifdef NSP_DEBUG /*unsigned int host_id = SCpnt->device->host->this_id;*/ @@ -365,7 +366,7 @@ static int nsphw_init(nsp_hw_data *data) /* * Start selection phase */ -static int nsphw_start_selection(Scsi_Cmnd *SCpnt) +static int nsphw_start_selection(struct scsi_cmnd *SCpnt) { unsigned int host_id = SCpnt->device->host->this_id; unsigned int base = SCpnt->device->host->io_port; @@ -446,7 +447,7 @@ static struct nsp_sync_table nsp_sync_table_20M[] = { /* * setup synchronous data transfer mode */ -static int nsp_analyze_sdtr(Scsi_Cmnd *SCpnt) +static int nsp_analyze_sdtr(struct scsi_cmnd *SCpnt) { unsigned char target = scmd_id(SCpnt); // unsigned char lun = SCpnt->device->lun; @@ -504,7 +505,7 @@ static int nsp_analyze_sdtr(Scsi_Cmnd *SCpnt) /* * start ninja hardware timer */ -static void nsp_start_timer(Scsi_Cmnd *SCpnt, int time) +static void nsp_start_timer(struct scsi_cmnd *SCpnt, int time) { unsigned int base = SCpnt->device->host->io_port; nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; @@ -517,7 +518,8 @@ static void nsp_start_timer(Scsi_Cmnd *SCpnt, int time) /* * wait for bus phase change */ -static int nsp_negate_signal(Scsi_Cmnd *SCpnt, unsigned char mask, char *str) +static int nsp_negate_signal(struct scsi_cmnd *SCpnt, unsigned char mask, + char *str) { unsigned int base = SCpnt->device->host->io_port; unsigned char reg; @@ -544,9 +546,9 @@ static int nsp_negate_signal(Scsi_Cmnd *SCpnt, unsigned char mask, char *str) /* * expect Ninja Irq */ -static int nsp_expect_signal(Scsi_Cmnd *SCpnt, - unsigned char current_phase, - unsigned char mask) +static int nsp_expect_signal(struct scsi_cmnd *SCpnt, + unsigned char current_phase, + unsigned char mask) { unsigned int base = SCpnt->device->host->io_port; int time_out; @@ -579,7 +581,7 @@ static int nsp_expect_signal(Scsi_Cmnd *SCpnt, /* * transfer SCSI message */ -static int nsp_xfer(Scsi_Cmnd *SCpnt, int phase) +static int nsp_xfer(struct scsi_cmnd *SCpnt, int phase) { unsigned int base = SCpnt->device->host->io_port; nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; @@ -619,7 +621,7 @@ static int nsp_xfer(Scsi_Cmnd *SCpnt, int phase) /* * get extra SCSI data from fifo */ -static int nsp_dataphase_bypass(Scsi_Cmnd *SCpnt) +static int nsp_dataphase_bypass(struct scsi_cmnd *SCpnt) { nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; unsigned int count; @@ -651,7 +653,7 @@ static int nsp_dataphase_bypass(Scsi_Cmnd *SCpnt) /* * accept reselection */ -static int nsp_reselected(Scsi_Cmnd *SCpnt) +static int nsp_reselected(struct scsi_cmnd *SCpnt) { unsigned int base = SCpnt->device->host->io_port; unsigned int host_id = SCpnt->device->host->this_id; @@ -690,7 +692,7 @@ static int nsp_reselected(Scsi_Cmnd *SCpnt) /* * count how many data transferd */ -static int nsp_fifo_count(Scsi_Cmnd *SCpnt) +static int nsp_fifo_count(struct scsi_cmnd *SCpnt) { unsigned int base = SCpnt->device->host->io_port; unsigned int count; @@ -717,7 +719,7 @@ static int nsp_fifo_count(Scsi_Cmnd *SCpnt) /* * read data in DATA IN phase */ -static void nsp_pio_read(Scsi_Cmnd *SCpnt) +static void nsp_pio_read(struct scsi_cmnd *SCpnt) { unsigned int base = SCpnt->device->host->io_port; unsigned long mmio_base = SCpnt->device->host->base; @@ -812,7 +814,7 @@ static void nsp_pio_read(Scsi_Cmnd *SCpnt) /* * write data in DATA OUT phase */ -static void nsp_pio_write(Scsi_Cmnd *SCpnt) +static void nsp_pio_write(struct scsi_cmnd *SCpnt) { unsigned int base = SCpnt->device->host->io_port; unsigned long mmio_base = SCpnt->device->host->base; @@ -905,7 +907,7 @@ static void nsp_pio_write(Scsi_Cmnd *SCpnt) /* * setup synchronous/asynchronous data transfer mode */ -static int nsp_nexus(Scsi_Cmnd *SCpnt) +static int nsp_nexus(struct scsi_cmnd *SCpnt) { unsigned int base = SCpnt->device->host->io_port; unsigned char target = scmd_id(SCpnt); @@ -952,7 +954,7 @@ static irqreturn_t nspintr(int irq, void *dev_id) { unsigned int base; unsigned char irq_status, irq_phase, phase; - Scsi_Cmnd *tmpSC; + struct scsi_cmnd *tmpSC; unsigned char target, lun; unsigned int *sync_neg; int i, tmp; @@ -1530,7 +1532,7 @@ nsp_proc_info( /*---------------------------------------------------------------*/ /* -static int nsp_eh_abort(Scsi_Cmnd *SCpnt) +static int nsp_eh_abort(struct scsi_cmnd *SCpnt) { nsp_dbg(NSP_DEBUG_BUSRESET, "SCpnt=0x%p", SCpnt); @@ -1558,7 +1560,7 @@ static int nsp_bus_reset(nsp_hw_data *data) return SUCCESS; } -static int nsp_eh_bus_reset(Scsi_Cmnd *SCpnt) +static int nsp_eh_bus_reset(struct scsi_cmnd *SCpnt) { nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; @@ -1567,7 +1569,7 @@ static int nsp_eh_bus_reset(Scsi_Cmnd *SCpnt) return nsp_bus_reset(data); } -static int nsp_eh_host_reset(Scsi_Cmnd *SCpnt) +static int nsp_eh_host_reset(struct scsi_cmnd *SCpnt) { nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; diff --git a/drivers/scsi/pcmcia/nsp_cs.h b/drivers/scsi/pcmcia/nsp_cs.h index a88714f..625ca97 100644 --- a/drivers/scsi/pcmcia/nsp_cs.h +++ b/drivers/scsi/pcmcia/nsp_cs.h @@ -266,7 +266,7 @@ typedef struct _nsp_hw_data { int TimerCount; int SelectionTimeOut; - Scsi_Cmnd *CurrentSC; + struct scsi_cmnd *CurrentSC; //int CurrnetTarget; int FifoCount; @@ -319,30 +319,34 @@ static int nsp_proc_info ( int hostno, #endif int inout); -static int nsp_queuecommand(Scsi_Cmnd *SCpnt, void (* done)(Scsi_Cmnd *SCpnt)); +static int nsp_queuecommand(struct scsi_cmnd *SCpnt, + void (* done)(struct scsi_cmnd *SCpnt)); /* Error handler */ -/*static int nsp_eh_abort (Scsi_Cmnd *SCpnt);*/ -/*static int nsp_eh_device_reset(Scsi_Cmnd *SCpnt);*/ -static int nsp_eh_bus_reset (Scsi_Cmnd *SCpnt); -static int nsp_eh_host_reset (Scsi_Cmnd *SCpnt); +/*static int nsp_eh_abort (struct scsi_cmnd *SCpnt);*/ +/*static int nsp_eh_device_reset(struct scsi_cmnd *SCpnt);*/ +static int nsp_eh_bus_reset (struct scsi_cmnd *SCpnt); +static int nsp_eh_host_reset (struct scsi_cmnd *SCpnt); static int nsp_bus_reset (nsp_hw_data *data); /* */ static int nsphw_init (nsp_hw_data *data); -static int nsphw_start_selection(Scsi_Cmnd *SCpnt); -static void nsp_start_timer (Scsi_Cmnd *SCpnt, int time); -static int nsp_fifo_count (Scsi_Cmnd *SCpnt); -static void nsp_pio_read (Scsi_Cmnd *SCpnt); -static void nsp_pio_write (Scsi_Cmnd *SCpnt); -static int nsp_nexus (Scsi_Cmnd *SCpnt); -static void nsp_scsi_done (Scsi_Cmnd *SCpnt); -static int nsp_analyze_sdtr (Scsi_Cmnd *SCpnt); -static int nsp_negate_signal (Scsi_Cmnd *SCpnt, unsigned char mask, char *str); -static int nsp_expect_signal (Scsi_Cmnd *SCpnt, unsigned char current_phase, unsigned char mask); -static int nsp_xfer (Scsi_Cmnd *SCpnt, int phase); -static int nsp_dataphase_bypass (Scsi_Cmnd *SCpnt); -static int nsp_reselected (Scsi_Cmnd *SCpnt); +static int nsphw_start_selection(struct scsi_cmnd *SCpnt); +static void nsp_start_timer (struct scsi_cmnd *SCpnt, int time); +static int nsp_fifo_count (struct scsi_cmnd *SCpnt); +static void nsp_pio_read (struct scsi_cmnd *SCpnt); +static void nsp_pio_write (struct scsi_cmnd *SCpnt); +static int nsp_nexus (struct scsi_cmnd *SCpnt); +static void nsp_scsi_done (struct scsi_cmnd *SCpnt); +static int nsp_analyze_sdtr (struct scsi_cmnd *SCpnt); +static int nsp_negate_signal (struct scsi_cmnd *SCpnt, + unsigned char mask, char *str); +static int nsp_expect_signal (struct scsi_cmnd *SCpnt, + unsigned char current_phase, + unsigned char mask); +static int nsp_xfer (struct scsi_cmnd *SCpnt, int phase); +static int nsp_dataphase_bypass (struct scsi_cmnd *SCpnt); +static int nsp_reselected (struct scsi_cmnd *SCpnt); static struct Scsi_Host *nsp_detect(struct scsi_host_template *sht); /* Interrupt handler */ @@ -355,8 +359,8 @@ static void __exit nsp_cs_exit(void); /* Debug */ #ifdef NSP_DEBUG -static void show_command (Scsi_Cmnd *SCpnt); -static void show_phase (Scsi_Cmnd *SCpnt); +static void show_command (struct scsi_cmnd *SCpnt); +static void show_phase (struct scsi_cmnd *SCpnt); static void show_busphase(unsigned char stat); static void show_message (nsp_hw_data *data); #else diff --git a/drivers/scsi/pcmcia/nsp_debug.c b/drivers/scsi/pcmcia/nsp_debug.c index 62e5c60..2f75fe6 100644 --- a/drivers/scsi/pcmcia/nsp_debug.c +++ b/drivers/scsi/pcmcia/nsp_debug.c @@ -138,12 +138,12 @@ static void print_commandk (unsigned char *command) printk("\n"); } -static void show_command(Scsi_Cmnd *SCpnt) +static void show_command(struct scsi_cmnd *SCpnt) { print_commandk(SCpnt->cmnd); } -static void show_phase(Scsi_Cmnd *SCpnt) +static void show_phase(struct scsi_cmnd *SCpnt) { int i = SCpnt->SCp.phase; diff --git a/drivers/scsi/pcmcia/nsp_message.c b/drivers/scsi/pcmcia/nsp_message.c index d705773..ef593b7 100644 --- a/drivers/scsi/pcmcia/nsp_message.c +++ b/drivers/scsi/pcmcia/nsp_message.c @@ -8,7 +8,7 @@ /* $Id: nsp_message.c,v 1.6 2003/07/26 14:21:09 elca Exp $ */ -static void nsp_message_in(Scsi_Cmnd *SCpnt) +static void nsp_message_in(struct scsi_cmnd *SCpnt) { unsigned int base = SCpnt->device->host->io_port; nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; @@ -50,7 +50,7 @@ static void nsp_message_in(Scsi_Cmnd *SCpnt) } -static void nsp_message_out(Scsi_Cmnd *SCpnt) +static void nsp_message_out(struct scsi_cmnd *SCpnt) { nsp_hw_data *data = (nsp_hw_data *)SCpnt->device->host->hostdata; int ret = 1; -- cgit v0.10.2 From 9531c330f14c02d9f4eff7345071f485dc62dab1 Mon Sep 17 00:00:00 2001 From: Henrik Kretzschmar Date: Tue, 10 Oct 2006 14:41:42 -0700 Subject: [SCSI] fc4: Conversion to struct scsi_cmnd in fc4 Changes the obsolete Scsi_Cmnd to struct scsi_cmnd in the Fibre Channel driver (fc4). Signed-off-by: Henrik Kretzschmar Signed-off-by: Andrew Morton Signed-off-by: James Bottomley diff --git a/drivers/fc4/fc.c b/drivers/fc4/fc.c index 22d1747..ca4e67a 100644 --- a/drivers/fc4/fc.c +++ b/drivers/fc4/fc.c @@ -70,9 +70,9 @@ #define FCP_CMND(SCpnt) ((fcp_cmnd *)&(SCpnt->SCp)) #define FC_SCMND(SCpnt) ((fc_channel *)(SCpnt->device->host->hostdata[0])) -#define SC_FCMND(fcmnd) ((Scsi_Cmnd *)((long)fcmnd - (long)&(((Scsi_Cmnd *)0)->SCp))) +#define SC_FCMND(fcmnd) ((struct scsi_cmnd *)((long)fcmnd - (long)&(((struct scsi_cmnd *)0)->SCp))) -static int fcp_scsi_queue_it(fc_channel *, Scsi_Cmnd *, fcp_cmnd *, int); +static int fcp_scsi_queue_it(fc_channel *, struct scsi_cmnd *, fcp_cmnd *, int); void fcp_queue_empty(fc_channel *); static void fcp_scsi_insert_queue (fc_channel *fc, fcp_cmnd *fcmd) @@ -378,14 +378,14 @@ void fcp_register(fc_channel *fc, u8 type, int unregister) printk ("FC: %segistering unknown type %02x\n", unregister ? "Unr" : "R", type); } -static void fcp_scsi_done(Scsi_Cmnd *SCpnt); +static void fcp_scsi_done(struct scsi_cmnd *SCpnt); static inline void fcp_scsi_receive(fc_channel *fc, int token, int status, fc_hdr *fch) { fcp_cmnd *fcmd; fcp_rsp *rsp; int host_status; - Scsi_Cmnd *SCpnt; + struct scsi_cmnd *SCpnt; int sense_len; int rsp_status; @@ -757,13 +757,14 @@ void fcp_release(fc_channel *fcchain, int count) /* count must > 0 */ } -static void fcp_scsi_done (Scsi_Cmnd *SCpnt) +static void fcp_scsi_done(struct scsi_cmnd *SCpnt) { if (FCP_CMND(SCpnt)->done) FCP_CMND(SCpnt)->done(SCpnt); } -static int fcp_scsi_queue_it(fc_channel *fc, Scsi_Cmnd *SCpnt, fcp_cmnd *fcmd, int prepare) +static int fcp_scsi_queue_it(fc_channel *fc, struct scsi_cmnd *SCpnt, + fcp_cmnd *fcmd, int prepare) { long i; fcp_cmd *cmd; @@ -837,7 +838,8 @@ static int fcp_scsi_queue_it(fc_channel *fc, Scsi_Cmnd *SCpnt, fcp_cmnd *fcmd, i return 0; } -int fcp_scsi_queuecommand(Scsi_Cmnd *SCpnt, void (* done)(Scsi_Cmnd *)) +int fcp_scsi_queuecommand(struct scsi_cmnd *SCpnt, + void (* done)(struct scsi_cmnd *)) { fcp_cmnd *fcmd = FCP_CMND(SCpnt); fc_channel *fc = FC_SCMND(SCpnt); @@ -873,7 +875,7 @@ void fcp_queue_empty(fc_channel *fc) } } -int fcp_scsi_abort(Scsi_Cmnd *SCpnt) +int fcp_scsi_abort(struct scsi_cmnd *SCpnt) { /* Internal bookkeeping only. Lose 1 cmd_slots slot. */ fcp_cmnd *fcmd = FCP_CMND(SCpnt); @@ -910,7 +912,7 @@ int fcp_scsi_abort(Scsi_Cmnd *SCpnt) } #if 0 -void fcp_scsi_reset_done(Scsi_Cmnd *SCpnt) +void fcp_scsi_reset_done(struct scsi_cmnd *SCpnt) { fc_channel *fc = FC_SCMND(SCpnt); @@ -921,7 +923,7 @@ void fcp_scsi_reset_done(Scsi_Cmnd *SCpnt) #define FCP_RESET_TIMEOUT (2*HZ) -int fcp_scsi_dev_reset(Scsi_Cmnd *SCpnt) +int fcp_scsi_dev_reset(struct scsi_cmnd *SCpnt) { #if 0 /* broken junk, but if davem wants to compile this driver, let him.. */ unsigned long flags; @@ -931,7 +933,7 @@ int fcp_scsi_dev_reset(Scsi_Cmnd *SCpnt) DECLARE_MUTEX_LOCKED(sem); if (!fc->rst_pkt) { - fc->rst_pkt = (Scsi_Cmnd *) kmalloc(sizeof(SCpnt), GFP_KERNEL); + fc->rst_pkt = (struct scsi_cmnd *) kmalloc(sizeof(SCpnt), GFP_KERNEL); if (!fc->rst_pkt) return FAILED; fcmd = FCP_CMND(fc->rst_pkt); @@ -999,7 +1001,7 @@ int fcp_scsi_dev_reset(Scsi_Cmnd *SCpnt) return SUCCESS; } -static int __fcp_scsi_host_reset(Scsi_Cmnd *SCpnt) +static int __fcp_scsi_host_reset(struct scsi_cmnd *SCpnt) { fc_channel *fc = FC_SCMND(SCpnt); fcp_cmnd *fcmd = FCP_CMND(SCpnt); @@ -1020,7 +1022,7 @@ static int __fcp_scsi_host_reset(Scsi_Cmnd *SCpnt) else return FAILED; } -int fcp_scsi_host_reset(Scsi_Cmnd *SCpnt) +int fcp_scsi_host_reset(struct scsi_cmnd *SCpnt) { unsigned long flags; int rc; diff --git a/drivers/fc4/fcp_impl.h b/drivers/fc4/fcp_impl.h index c397c84..1ac6133 100644 --- a/drivers/fc4/fcp_impl.h +++ b/drivers/fc4/fcp_impl.h @@ -39,7 +39,7 @@ struct _fc_channel; typedef struct fcp_cmnd { struct fcp_cmnd *next; struct fcp_cmnd *prev; - void (*done)(Scsi_Cmnd *); + void (*done)(struct scsi_cmnd *); unsigned short proto; unsigned short token; unsigned int did; @@ -94,14 +94,14 @@ typedef struct _fc_channel { long *scsi_bitmap; long scsi_bitmap_end; int scsi_free; - int (*encode_addr)(Scsi_Cmnd *, u16 *, struct _fc_channel *, fcp_cmnd *); + int (*encode_addr)(struct scsi_cmnd *, u16 *, struct _fc_channel *, fcp_cmnd *); fcp_cmnd *scsi_que; char scsi_name[4]; fcp_cmnd **cmd_slots; int channels; int targets; long *ages; - Scsi_Cmnd *rst_pkt; + struct scsi_cmnd *rst_pkt; fcp_posmap *posmap; /* LOGIN stuff */ fcp_cmnd *login; @@ -155,9 +155,10 @@ int fc_do_prli(fc_channel *, unsigned char); for_each_fc_channel(fc) \ if (fc->state == FC_STATE_ONLINE) -int fcp_scsi_queuecommand(Scsi_Cmnd *, void (* done)(Scsi_Cmnd *)); -int fcp_scsi_abort(Scsi_Cmnd *); -int fcp_scsi_dev_reset(Scsi_Cmnd *); -int fcp_scsi_host_reset(Scsi_Cmnd *); +int fcp_scsi_queuecommand(struct scsi_cmnd *, + void (* done) (struct scsi_cmnd *)); +int fcp_scsi_abort(struct scsi_cmnd *); +int fcp_scsi_dev_reset(struct scsi_cmnd *); +int fcp_scsi_host_reset(struct scsi_cmnd *); #endif /* !(_FCP_SCSI_H) */ -- cgit v0.10.2 From 8d1a006049ff1c084d57fbea1106ecad3455bd27 Mon Sep 17 00:00:00 2001 From: Swen Schillig Date: Thu, 12 Oct 2006 11:43:44 +0200 Subject: [SCSI] zfcp: initialize scsi_host_template.max_sectors with appropriate value Define ZFCP_MAX_SECTORS and initialize scsi_host_template.max_sectors with appropriate value. Signed-off-by: Swen Schillig Signed-off-by: James Bottomley diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 8f88269..74c0eac 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -107,6 +107,10 @@ zfcp_address_to_sg(void *address, struct scatterlist *list) (ZFCP_MAX_SBALS_PER_REQ * ZFCP_MAX_SBALES_PER_SBAL - 2) /* request ID + QTCB in SBALE 0 + 1 of first SBAL in chain */ +#define ZFCP_MAX_SECTORS (ZFCP_MAX_SBALES_PER_REQ * 8) + /* max. number of (data buffer) SBALEs in largest SBAL chain + multiplied with number of sectors per 4k block */ + /* FIXME(tune): free space should be one max. SBAL chain plus what? */ #define ZFCP_QDIO_PCI_INTERVAL (QDIO_MAX_BUFFERS_PER_Q \ - (ZFCP_MAX_SBALS_PER_REQ + 4)) diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 4d2bc79..452d96f 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -58,6 +58,7 @@ struct zfcp_data zfcp_data = { .cmd_per_lun = 1, .use_clustering = 1, .sdev_attrs = zfcp_sysfs_sdev_attrs, + .max_sectors = ZFCP_MAX_SECTORS, }, .driver_version = ZFCP_VERSION, }; -- cgit v0.10.2 From f1663ad5dbb801e03c4c99c24d698ad5dba9aaff Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Fri, 13 Oct 2006 09:33:37 -0700 Subject: [SCSI] qla2xxx: Check return value of sysfs_create_bin_file() usage. Signed-off-by: Andrew Vasquez Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index ee75a71..285c8e8 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -379,21 +379,37 @@ static struct bin_attribute sysfs_sfp_attr = { .read = qla2x00_sysfs_read_sfp, }; +static struct sysfs_entry { + char *name; + struct bin_attribute *attr; + int is4GBp_only; +} bin_file_entries[] = { + { "fw_dump", &sysfs_fw_dump_attr, }, + { "nvram", &sysfs_nvram_attr, }, + { "optrom", &sysfs_optrom_attr, }, + { "optrom_ctl", &sysfs_optrom_ctl_attr, }, + { "vpd", &sysfs_vpd_attr, 1 }, + { "sfp", &sysfs_sfp_attr, 1 }, + { 0 }, +}; + void qla2x00_alloc_sysfs_attr(scsi_qla_host_t *ha) { struct Scsi_Host *host = ha->host; + struct sysfs_entry *iter; + int ret; - sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_fw_dump_attr); - sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_nvram_attr); - sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_optrom_attr); - sysfs_create_bin_file(&host->shost_gendev.kobj, - &sysfs_optrom_ctl_attr); - if (IS_QLA24XX(ha) || IS_QLA54XX(ha)) { - sysfs_create_bin_file(&host->shost_gendev.kobj, - &sysfs_vpd_attr); - sysfs_create_bin_file(&host->shost_gendev.kobj, - &sysfs_sfp_attr); + for (iter = bin_file_entries; iter->name; iter++) { + if (iter->is4GBp_only && (!IS_QLA24XX(ha) && !IS_QLA54XX(ha))) + continue; + + ret = sysfs_create_bin_file(&host->shost_gendev.kobj, + iter->attr); + if (ret) + qla_printk(KERN_INFO, ha, + "Unable to create sysfs %s binary attribute " + "(%d).\n", iter->name, ret); } } @@ -401,17 +417,14 @@ void qla2x00_free_sysfs_attr(scsi_qla_host_t *ha) { struct Scsi_Host *host = ha->host; + struct sysfs_entry *iter; + + for (iter = bin_file_entries; iter->name; iter++) { + if (iter->is4GBp_only && (!IS_QLA24XX(ha) && !IS_QLA54XX(ha))) + continue; - sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_fw_dump_attr); - sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_nvram_attr); - sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_optrom_attr); - sysfs_remove_bin_file(&host->shost_gendev.kobj, - &sysfs_optrom_ctl_attr); - if (IS_QLA24XX(ha) || IS_QLA54XX(ha)) { - sysfs_remove_bin_file(&host->shost_gendev.kobj, - &sysfs_vpd_attr); sysfs_remove_bin_file(&host->shost_gendev.kobj, - &sysfs_sfp_attr); + iter->attr); } if (ha->beacon_blink_led == 1) -- cgit v0.10.2 From 18c6c12759813c988bb05796d1b3352e98ae77de Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Fri, 13 Oct 2006 09:33:38 -0700 Subject: [SCSI] qla2xxx: Workaround D3 power-management issues. Early ISP2432 parts have a known hardware issue when coming out of a D3 hot state. This issue can result in a hung PCIe link. Recent firmwares contain a workaround whereby the stop-firmware mailbox command prevents the ISP from entering the D3 hot state. In order to ensure that the workaround succeeded the driver must verify that the stop-firmware mailbox command completes successfully. In the event of a failure, the driver attempts a shutdown-retry after resetting the ISP and re-executing firmware. Signed-off-by: Andrew Vasquez Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h index b51ce8f..ca2f660 100644 --- a/drivers/scsi/qla2xxx/qla_gbl.h +++ b/drivers/scsi/qla2xxx/qla_gbl.h @@ -48,6 +48,7 @@ extern void qla2x00_update_fcport(scsi_qla_host_t *, fc_port_t *); extern void qla2x00_reg_remote_port(scsi_qla_host_t *, fc_port_t *); extern void qla2x00_alloc_fw_dump(scsi_qla_host_t *); +extern void qla2x00_try_to_stop_firmware(scsi_qla_host_t *); /* * Global Data in qla_os.c source file. diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index d5e0a12..08cb5e3 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -3948,3 +3948,24 @@ qla24xx_load_risc(scsi_qla_host_t *ha, uint32_t *srisc_addr) fail_fw_integrity: return QLA_FUNCTION_FAILED; } + +void +qla2x00_try_to_stop_firmware(scsi_qla_host_t *ha) +{ + int ret, retries; + + if (!IS_QLA24XX(ha) && !IS_QLA54XX(ha)) + return; + + ret = qla2x00_stop_firmware(ha); + for (retries = 5; ret != QLA_SUCCESS && retries ; retries--) { + qla2x00_reset_chip(ha); + if (qla2x00_chip_diag(ha) != QLA_SUCCESS) + continue; + if (qla2x00_setup_chip(ha) != QLA_SUCCESS) + continue; + qla_printk(KERN_INFO, ha, + "Attempting retry of stop-firmware command...\n"); + ret = qla2x00_stop_firmware(ha); + } +} diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 34b6eb7..54f561d 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -1712,8 +1712,10 @@ qla2x00_free_device(scsi_qla_host_t *ha) if (ha->eft) qla2x00_trace_control(ha, TC_DISABLE, 0, 0); + ha->flags.online = 0; + /* Stop currently executing firmware. */ - qla2x00_stop_firmware(ha); + qla2x00_try_to_stop_firmware(ha); /* turn-off interrupts on the card */ if (ha->interrupts_on) @@ -1721,8 +1723,6 @@ qla2x00_free_device(scsi_qla_host_t *ha) qla2x00_mem_free(ha); - ha->flags.online = 0; - /* Detach interrupts */ if (ha->host->irq) free_irq(ha->host->irq, ha); -- cgit v0.10.2 From df7baa506c2db1c2d12abd6f05c43f911da55a2e Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Fri, 13 Oct 2006 09:33:39 -0700 Subject: [SCSI] qla2xxx: Correct QUEUE_FULL handling. - Drop queue-depths across all luns for a given fcport during TASK_SET_FULL statuses. - Ramp-up I/Os after throttling. - Consolidate completion-status handling of CS_QUEUE_FULL with CS_COMPLETE as ISP24xx firmware no longer reports CS_QUEUE_FULL. Signed-off-by: Andrew Vasquez Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index bab33f6..c4fc40f 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -1545,6 +1545,9 @@ typedef struct fc_port { spinlock_t rport_lock; struct fc_rport *rport, *drport; u32 supported_classes; + + unsigned long last_queue_full; + unsigned long last_ramp_up; } fc_port_t; /* @@ -2255,6 +2258,7 @@ typedef struct scsi_qla_host { uint16_t mgmt_svr_loop_id; uint32_t login_retry_count; + int max_q_depth; /* Fibre Channel Device List. */ struct list_head fcports; diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h index ca2f660..32ebeec 100644 --- a/drivers/scsi/qla2xxx/qla_gbl.h +++ b/drivers/scsi/qla2xxx/qla_gbl.h @@ -62,6 +62,7 @@ extern int ql2xloginretrycount; extern int ql2xfdmienable; extern int ql2xallocfwdump; extern int ql2xextended_error_logging; +extern int ql2xqfullrampup; extern void qla2x00_sp_compl(scsi_qla_host_t *, srb_t *); diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 626c717..d3b6df4 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -6,6 +6,8 @@ */ #include "qla_def.h" +#include + static void qla2x00_mbx_completion(scsi_qla_host_t *, uint16_t); static void qla2x00_async_event(scsi_qla_host_t *, uint16_t *); static void qla2x00_process_completed_request(struct scsi_qla_host *, uint32_t); @@ -593,6 +595,67 @@ qla2x00_async_event(scsi_qla_host_t *ha, uint16_t *mb) } } +static void +qla2x00_adjust_sdev_qdepth_up(struct scsi_device *sdev, void *data) +{ + fc_port_t *fcport = data; + + if (fcport->ha->max_q_depth <= sdev->queue_depth) + return; + + if (sdev->ordered_tags) + scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, + sdev->queue_depth + 1); + else + scsi_adjust_queue_depth(sdev, MSG_SIMPLE_TAG, + sdev->queue_depth + 1); + + fcport->last_ramp_up = jiffies; + + DEBUG2(qla_printk(KERN_INFO, fcport->ha, + "scsi(%ld:%d:%d:%d): Queue depth adjusted-up to %d.\n", + fcport->ha->host_no, sdev->channel, sdev->id, sdev->lun, + sdev->queue_depth)); +} + +static void +qla2x00_adjust_sdev_qdepth_down(struct scsi_device *sdev, void *data) +{ + fc_port_t *fcport = data; + + if (!scsi_track_queue_full(sdev, sdev->queue_depth - 1)) + return; + + DEBUG2(qla_printk(KERN_INFO, fcport->ha, + "scsi(%ld:%d:%d:%d): Queue depth adjusted-down to %d.\n", + fcport->ha->host_no, sdev->channel, sdev->id, sdev->lun, + sdev->queue_depth)); +} + +static inline void +qla2x00_ramp_up_queue_depth(scsi_qla_host_t *ha, srb_t *sp) +{ + fc_port_t *fcport; + struct scsi_device *sdev; + + sdev = sp->cmd->device; + if (sdev->queue_depth >= ha->max_q_depth) + return; + + fcport = sp->fcport; + if (time_before(jiffies, + fcport->last_ramp_up + ql2xqfullrampup * HZ)) + return; + if (time_before(jiffies, + fcport->last_queue_full + ql2xqfullrampup * HZ)) + return; + + spin_unlock_irq(&ha->hardware_lock); + starget_for_each_device(sdev->sdev_target, fcport, + qla2x00_adjust_sdev_qdepth_up); + spin_lock_irq(&ha->hardware_lock); +} + /** * qla2x00_process_completed_request() - Process a Fast Post response. * @ha: SCSI driver HA context @@ -624,6 +687,8 @@ qla2x00_process_completed_request(struct scsi_qla_host *ha, uint32_t index) /* Save ISP completion status */ sp->cmd->result = DID_OK << 16; + + qla2x00_ramp_up_queue_depth(ha, sp); qla2x00_sp_compl(ha, sp); } else { DEBUG2(printk("scsi(%ld): Invalid ISP SCSI completion handle\n", @@ -823,6 +888,7 @@ qla2x00_status_entry(scsi_qla_host_t *ha, void *pkt) */ switch (comp_status) { case CS_COMPLETE: + case CS_QUEUE_FULL: if (scsi_status == 0) { cp->result = DID_OK << 16; break; @@ -849,6 +915,20 @@ qla2x00_status_entry(scsi_qla_host_t *ha, void *pkt) } cp->result = DID_OK << 16 | lscsi_status; + if (lscsi_status == SAM_STAT_TASK_SET_FULL) { + DEBUG2(printk(KERN_INFO + "scsi(%ld): QUEUE FULL status detected " + "0x%x-0x%x.\n", ha->host_no, comp_status, + scsi_status)); + + /* Adjust queue depth for all luns on the port. */ + fcport->last_queue_full = jiffies; + spin_unlock_irq(&ha->hardware_lock); + starget_for_each_device(cp->device->sdev_target, + fcport, qla2x00_adjust_sdev_qdepth_down); + spin_lock_irq(&ha->hardware_lock); + break; + } if (lscsi_status != SS_CHECK_CONDITION) break; @@ -1066,17 +1146,6 @@ qla2x00_status_entry(scsi_qla_host_t *ha, void *pkt) qla2x00_mark_device_lost(ha, fcport, 1, 1); break; - case CS_QUEUE_FULL: - DEBUG2(printk(KERN_INFO - "scsi(%ld): QUEUE FULL status detected 0x%x-0x%x.\n", - ha->host_no, comp_status, scsi_status)); - - /* SCSI Mid-Layer handles device queue full */ - - cp->result = DID_OK << 16 | lscsi_status; - - break; - default: DEBUG3(printk("scsi(%ld): Error detected (unknown status) " "0x%x-0x%x.\n", ha->host_no, comp_status, scsi_status)); diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 54f561d..208607b 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -77,6 +77,19 @@ MODULE_PARM_DESC(ql2xfdmienable, "Enables FDMI registratons " "Default is 0 - no FDMI. 1 - perfom FDMI."); +#define MAX_Q_DEPTH 32 +static int ql2xmaxqdepth = MAX_Q_DEPTH; +module_param(ql2xmaxqdepth, int, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(ql2xmaxqdepth, + "Maximum queue depth to report for target devices."); + +int ql2xqfullrampup = 120; +module_param(ql2xqfullrampup, int, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(ql2xqfullrampup, + "Number of seconds to wait to begin to ramp-up the queue " + "depth for a device after a queue-full condition has been " + "detected. Default is 120 seconds."); + /* * SCSI host template entry points */ @@ -1104,9 +1117,9 @@ qla2xxx_slave_configure(struct scsi_device *sdev) struct fc_rport *rport = starget_to_rport(sdev->sdev_target); if (sdev->tagged_supported) - scsi_activate_tcq(sdev, 32); + scsi_activate_tcq(sdev, ha->max_q_depth); else - scsi_deactivate_tcq(sdev, 32); + scsi_deactivate_tcq(sdev, ha->max_q_depth); rport->dev_loss_tmo = ha->port_down_retry_count + 5; @@ -1413,6 +1426,10 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) ha->link_data_rate = PORT_SPEED_UNKNOWN; ha->optrom_size = OPTROM_SIZE_2300; + ha->max_q_depth = MAX_Q_DEPTH; + if (ql2xmaxqdepth != 0 && ql2xmaxqdepth <= 0xffffU) + ha->max_q_depth = ql2xmaxqdepth; + /* Assign ISP specific operations. */ ha->isp_ops.pci_config = qla2100_pci_config; ha->isp_ops.reset_chip = qla2x00_reset_chip; -- cgit v0.10.2 From 7c83a3ceb6d06e7cb908f5102687c9661e7d4d0a Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Fri, 13 Oct 2006 09:33:40 -0700 Subject: [SCSI] qla2xxx: Update version number to 8.01.07-k3. Signed-off-by: Andrew Vasquez Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h index e57bf45..1fa0bce 100644 --- a/drivers/scsi/qla2xxx/qla_version.h +++ b/drivers/scsi/qla2xxx/qla_version.h @@ -7,7 +7,7 @@ /* * Driver version */ -#define QLA2XXX_VERSION "8.01.07-k2" +#define QLA2XXX_VERSION "8.01.07-k3" #define QLA_DRIVER_MAJOR_VER 8 #define QLA_DRIVER_MINOR_VER 1 -- cgit v0.10.2 From 9c3121feef7e1fba86f74b2677e6f54e7153d149 Mon Sep 17 00:00:00 2001 From: Santiago Leon Date: Fri, 13 Oct 2006 10:22:50 -0500 Subject: [SCSI] ibmvscsi: correctly reenable CRQ The "ibmvscsi: treat busy and error conditions separately" patch submitted by Dave Boutcher back in June incorrectly reenables the CRQ. The broken logic causes the adapter to get disabled if the CRQ connection happens to close temporarily. This patch "fixes that obviously wrong logic check" (Dave's words). Signed-off-by: Santiago Leon Signed-off-by: David Boutcher Signed-off-by: James Bottomley diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c index 669ea4f..fbc1d5c 100644 --- a/drivers/scsi/ibmvscsi/ibmvscsi.c +++ b/drivers/scsi/ibmvscsi/ibmvscsi.c @@ -1213,7 +1213,7 @@ void ibmvscsi_handle_crq(struct viosrp_crq *crq, "ibmvscsi: Re-enabling adapter!\n"); purge_requests(hostdata, DID_REQUEUE); if ((ibmvscsi_reenable_crq_queue(&hostdata->queue, - hostdata) == 0) || + hostdata)) || (ibmvscsi_send_crq(hostdata, 0xC001000000000000LL, 0))) { atomic_set(&hostdata->request_limit, -- cgit v0.10.2 From 3fc2aef5227dda464560a3fdafc9f4c7ce10210f Mon Sep 17 00:00:00 2001 From: Sergey Kononenko Date: Sun, 15 Oct 2006 03:12:08 +0300 Subject: [SCSI] aic94xx: Supermicro motherboards support Add PCI id. Plus correct for possibly missing resistor that can cause FLASHEX to have the wrong value. Signed-off-by: Sergey Kononenko Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.h b/drivers/scsi/aic94xx/aic94xx_hwi.h index 14319d1..7b6aca0 100644 --- a/drivers/scsi/aic94xx/aic94xx_hwi.h +++ b/drivers/scsi/aic94xx/aic94xx_hwi.h @@ -46,6 +46,7 @@ #define PCI_DEVICE_ID_ADAPTEC2_RAZOR10 0x410 #define PCI_DEVICE_ID_ADAPTEC2_RAZOR12 0x412 #define PCI_DEVICE_ID_ADAPTEC2_RAZOR1E 0x41E +#define PCI_DEVICE_ID_ADAPTEC2_RAZOR1F 0x41F #define PCI_DEVICE_ID_ADAPTEC2_RAZOR30 0x430 #define PCI_DEVICE_ID_ADAPTEC2_RAZOR32 0x432 #define PCI_DEVICE_ID_ADAPTEC2_RAZOR3E 0x43E diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c index 99743ca..a4cc432 100644 --- a/drivers/scsi/aic94xx/aic94xx_init.c +++ b/drivers/scsi/aic94xx/aic94xx_init.c @@ -814,6 +814,8 @@ static const struct pci_device_id aic94xx_pci_table[] __devinitdata = { 0, 0, 1}, {PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR1E), 0, 0, 1}, + {PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR1F), + 0, 0, 1}, {PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR30), 0, 0, 2}, {PCI_DEVICE(PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_RAZOR32), diff --git a/drivers/scsi/aic94xx/aic94xx_sds.c b/drivers/scsi/aic94xx/aic94xx_sds.c index 83574b5..de7c04d 100644 --- a/drivers/scsi/aic94xx/aic94xx_sds.c +++ b/drivers/scsi/aic94xx/aic94xx_sds.c @@ -630,10 +630,6 @@ static int asd_flash_getid(struct asd_ha_struct *asd_ha) reg = asd_read_reg_dword(asd_ha, EXSICNFGR); - if (!(reg & FLASHEX)) { - ASD_DPRINTK("flash doesn't exist\n"); - return -ENOENT; - } if (pci_read_config_dword(asd_ha->pcidev, PCI_CONF_FLSH_BAR, &asd_ha->hw_prof.flash.bar)) { asd_printk("couldn't read PCI_CONF_FLSH_BAR of %s\n", -- cgit v0.10.2 From 031280f6e73d9177c93333c96e37eb05f2522faa Mon Sep 17 00:00:00 2001 From: Andrey Mirkin <(amirkin@sw.ru)> Date: Mon, 16 Oct 2006 12:08:43 +0400 Subject: [SCSI] megaraid_{mm,mbox}: 64-bit DMA capability fix It is known that 2 LSI Logic MegaRAID SATA RAID Controllers (150-4 and 150-6) don't support 64-bit DMA. Unfortunately currently this check is wrong and driver sets 64-bit DMA mode for these devices. Signed-off-by: Andrey Mirkin Acked-by: "Ju, Seokmann" Signed-off-by: James Bottomley diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c index c0edb66..7bac86d 100644 --- a/drivers/scsi/megaraid/megaraid_mbox.c +++ b/drivers/scsi/megaraid/megaraid_mbox.c @@ -884,7 +884,7 @@ megaraid_init_mbox(adapter_t *adapter) if (((magic64 == HBA_SIGNATURE_64_BIT) && ((adapter->pdev->subsystem_device != - PCI_SUBSYS_ID_MEGARAID_SATA_150_6) || + PCI_SUBSYS_ID_MEGARAID_SATA_150_6) && (adapter->pdev->subsystem_device != PCI_SUBSYS_ID_MEGARAID_SATA_150_4))) || (adapter->pdev->vendor == PCI_VENDOR_ID_LSI_LOGIC && -- cgit v0.10.2 From ed632da84c51a39fd9c982991e0f26120a035761 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Mon, 16 Oct 2006 10:06:27 -0500 Subject: [SCSI] add can_queue to host parameters Debugging TCQ issues has shown me this is a very useful parameter to be able to view. Add it to he host class parameters. Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index e7fe565..e1a9166 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -192,6 +192,7 @@ static CLASS_DEVICE_ATTR(state, S_IRUGO | S_IWUSR, show_shost_state, store_shost shost_rd_attr(unique_id, "%u\n"); shost_rd_attr(host_busy, "%hu\n"); shost_rd_attr(cmd_per_lun, "%hd\n"); +shost_rd_attr(can_queue, "%hd\n"); shost_rd_attr(sg_tablesize, "%hu\n"); shost_rd_attr(unchecked_isa_dma, "%d\n"); shost_rd_attr2(proc_name, hostt->proc_name, "%s\n"); @@ -200,6 +201,7 @@ static struct class_device_attribute *scsi_sysfs_shost_attrs[] = { &class_device_attr_unique_id, &class_device_attr_host_busy, &class_device_attr_cmd_per_lun, + &class_device_attr_can_queue, &class_device_attr_sg_tablesize, &class_device_attr_unchecked_isa_dma, &class_device_attr_proc_name, -- cgit v0.10.2 From 47bcd3546d5141e54f15e40a20dc01d7c5f5a473 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:55:46 +0100 Subject: [SCSI] Switch fdomain to the pci_get API Doesn't make the hardware hot pluggable but does ensure the driver won't crash when another device is hot-unplugged at the wrong moment. Soon I propose to deprecate pci_find_device() and some of its friends. Signed-off-by: Alan Cox Signed-off-by: James Bottomley diff --git a/drivers/scsi/fdomain.c b/drivers/scsi/fdomain.c index 41b05fc4..2a96424 100644 --- a/drivers/scsi/fdomain.c +++ b/drivers/scsi/fdomain.c @@ -387,6 +387,7 @@ static void __iomem * bios_mem; static int bios_major; static int bios_minor; static int PCI_bus; +static struct pci_dev *PCI_dev; static int Quantum; /* Quantum board variant */ static int interrupt_level; static volatile int in_command; @@ -812,9 +813,10 @@ static int fdomain_pci_bios_detect( int *irq, int *iobase, struct pci_dev **ret_ PCI_DEVICE_ID_FD_36C70 ); #endif - if ((pdev = pci_find_device(PCI_VENDOR_ID_FD, PCI_DEVICE_ID_FD_36C70, pdev)) == NULL) + if ((pdev = pci_get_device(PCI_VENDOR_ID_FD, PCI_DEVICE_ID_FD_36C70, pdev)) == NULL) return 0; - if (pci_enable_device(pdev)) return 0; + if (pci_enable_device(pdev)) + goto fail; #if DEBUG_DETECT printk( "scsi: TMC-3260 detect:" @@ -831,7 +833,7 @@ static int fdomain_pci_bios_detect( int *irq, int *iobase, struct pci_dev **ret_ pci_irq = pdev->irq; if (!request_region( pci_base, 0x10, "fdomain" )) - return 0; + goto fail; /* Now we have the I/O base address and interrupt from the PCI configuration registers. */ @@ -848,17 +850,22 @@ static int fdomain_pci_bios_detect( int *irq, int *iobase, struct pci_dev **ret_ if (!fdomain_is_valid_port(pci_base)) { printk(KERN_ERR "scsi: PCI card detected, but driver not loaded (invalid port)\n" ); release_region(pci_base, 0x10); - return 0; + goto fail; } /* Fill in a few global variables. Ugh. */ bios_major = bios_minor = -1; PCI_bus = 1; + PCI_dev = pdev; Quantum = 0; bios_base = 0; return 1; +fail: + pci_dev_put(pdev); + return 0; } + #endif struct Scsi_Host *__fdomain_16x0_detect(struct scsi_host_template *tpnt ) @@ -909,8 +916,7 @@ struct Scsi_Host *__fdomain_16x0_detect(struct scsi_host_template *tpnt ) if (setup_called) { printk(KERN_ERR "scsi: Bad LILO/INSMOD parameters?\n"); } - release_region(port_base, 0x10); - return NULL; + goto fail; } if (this_id) { @@ -942,8 +948,7 @@ struct Scsi_Host *__fdomain_16x0_detect(struct scsi_host_template *tpnt ) /* Log IRQ with kernel */ if (!interrupt_level) { printk(KERN_ERR "scsi: Card Detected, but driver not loaded (no IRQ)\n" ); - release_region(port_base, 0x10); - return NULL; + goto fail; } else { /* Register the IRQ with the kernel */ @@ -964,11 +969,14 @@ struct Scsi_Host *__fdomain_16x0_detect(struct scsi_host_template *tpnt ) printk(KERN_ERR " Send mail to faith@acm.org\n" ); } printk(KERN_ERR "scsi: Detected, but driver not loaded (IRQ)\n" ); - release_region(port_base, 0x10); - return NULL; + goto fail; } } return shpnt; +fail: + pci_dev_put(pdev); + release_region(port_base, 0x10); + return NULL; } static int fdomain_16x0_detect(struct scsi_host_template *tpnt) @@ -1714,6 +1722,8 @@ static int fdomain_16x0_release(struct Scsi_Host *shpnt) free_irq(shpnt->irq, shpnt); if (shpnt->io_port && shpnt->n_io_port) release_region(shpnt->io_port, shpnt->n_io_port); + if (PCI_bus) + pci_dev_put(PCI_dev); return 0; } -- cgit v0.10.2 From 43a145a3440c5c5f24ff2888801e40e2242187e6 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 16 Oct 2006 18:09:38 -0400 Subject: [SCSI] iscsi class: fix slab corruption during restart The transport class recv mempools are causing slab corruption. We could hack around netlink's lack of mempool support like dm, but it is just too ulgy (dm's hack is ugly enough :) when you need to support broadcast. This patch removes the recv pools. We have not used them even when we were allocting 20 MB per session and the system only had 64 MBs. And we have no pools on the send side and have been ok there. When Peter's work gets merged we can use that since the network guys are in favor of that approach and are not going to add mempools everywhere. Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 7b0019c..2d3baa9 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -21,7 +21,6 @@ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ #include -#include #include #include #include @@ -149,30 +148,6 @@ static DECLARE_TRANSPORT_CLASS(iscsi_connection_class, static struct sock *nls; static DEFINE_MUTEX(rx_queue_mutex); -struct mempool_zone { - mempool_t *pool; - atomic_t allocated; - int size; - int hiwat; - struct list_head freequeue; - spinlock_t freelock; -}; - -static struct mempool_zone *z_reply; - -/* - * Z_MAX_* - actual mempool size allocated at the mempool_zone_init() time - * Z_HIWAT_* - zone's high watermark when if_error bit will be set to -ENOMEM - * so daemon will notice OOM on NETLINK tranposrt level and will - * be able to predict or change operational behavior - */ -#define Z_MAX_REPLY 8 -#define Z_HIWAT_REPLY 6 -#define Z_MAX_PDU 8 -#define Z_HIWAT_PDU 6 -#define Z_MAX_ERROR 16 -#define Z_HIWAT_ERROR 12 - static LIST_HEAD(sesslist); static DEFINE_SPINLOCK(sesslock); static LIST_HEAD(connlist); @@ -414,59 +389,11 @@ int iscsi_destroy_session(struct iscsi_cls_session *session) } EXPORT_SYMBOL_GPL(iscsi_destroy_session); -static void mempool_zone_destroy(struct mempool_zone *zp) -{ - mempool_destroy(zp->pool); - kfree(zp); -} - -static void* -mempool_zone_alloc_skb(gfp_t gfp_mask, void *pool_data) -{ - struct mempool_zone *zone = pool_data; - - return alloc_skb(zone->size, gfp_mask); -} - -static void -mempool_zone_free_skb(void *element, void *pool_data) -{ - kfree_skb(element); -} - -static struct mempool_zone * -mempool_zone_init(unsigned max, unsigned size, unsigned hiwat) -{ - struct mempool_zone *zp; - - zp = kzalloc(sizeof(*zp), GFP_KERNEL); - if (!zp) - return NULL; - - zp->size = size; - zp->hiwat = hiwat; - INIT_LIST_HEAD(&zp->freequeue); - spin_lock_init(&zp->freelock); - atomic_set(&zp->allocated, 0); - - zp->pool = mempool_create(max, mempool_zone_alloc_skb, - mempool_zone_free_skb, zp); - if (!zp->pool) { - kfree(zp); - return NULL; - } - - return zp; -} - static void iscsi_conn_release(struct device *dev) { struct iscsi_cls_conn *conn = iscsi_dev_to_conn(dev); struct device *parent = conn->dev.parent; - mempool_zone_destroy(conn->z_pdu); - mempool_zone_destroy(conn->z_error); - kfree(conn); put_device(parent); } @@ -476,31 +403,6 @@ static int iscsi_is_conn_dev(const struct device *dev) return dev->release == iscsi_conn_release; } -static int iscsi_create_event_pools(struct iscsi_cls_conn *conn) -{ - conn->z_pdu = mempool_zone_init(Z_MAX_PDU, - NLMSG_SPACE(sizeof(struct iscsi_uevent) + - sizeof(struct iscsi_hdr) + - DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH), - Z_HIWAT_PDU); - if (!conn->z_pdu) { - dev_printk(KERN_ERR, &conn->dev, "iscsi: can not allocate " - "pdu zone for new conn\n"); - return -ENOMEM; - } - - conn->z_error = mempool_zone_init(Z_MAX_ERROR, - NLMSG_SPACE(sizeof(struct iscsi_uevent)), - Z_HIWAT_ERROR); - if (!conn->z_error) { - dev_printk(KERN_ERR, &conn->dev, "iscsi: can not allocate " - "error zone for new conn\n"); - mempool_zone_destroy(conn->z_pdu); - return -ENOMEM; - } - return 0; -} - /** * iscsi_create_conn - create iscsi class connection * @session: iscsi cls session @@ -533,12 +435,9 @@ iscsi_create_conn(struct iscsi_cls_session *session, uint32_t cid) conn->transport = transport; conn->cid = cid; - if (iscsi_create_event_pools(conn)) - goto free_conn; - /* this is released in the dev's release function */ if (!get_device(&session->dev)) - goto free_conn_pools; + goto free_conn; snprintf(conn->dev.bus_id, BUS_ID_SIZE, "connection%d:%u", session->sid, cid); @@ -555,8 +454,6 @@ iscsi_create_conn(struct iscsi_cls_session *session, uint32_t cid) release_parent_ref: put_device(&session->dev); -free_conn_pools: - free_conn: kfree(conn); return NULL; @@ -599,81 +496,31 @@ iscsi_if_transport_lookup(struct iscsi_transport *tt) return NULL; } -static inline struct list_head *skb_to_lh(struct sk_buff *skb) -{ - return (struct list_head *)&skb->cb; -} - -static void -mempool_zone_complete(struct mempool_zone *zone) -{ - unsigned long flags; - struct list_head *lh, *n; - - spin_lock_irqsave(&zone->freelock, flags); - list_for_each_safe(lh, n, &zone->freequeue) { - struct sk_buff *skb = (struct sk_buff *)((char *)lh - - offsetof(struct sk_buff, cb)); - if (!skb_shared(skb)) { - list_del(skb_to_lh(skb)); - mempool_free(skb, zone->pool); - atomic_dec(&zone->allocated); - } - } - spin_unlock_irqrestore(&zone->freelock, flags); -} - -static struct sk_buff* -mempool_zone_get_skb(struct mempool_zone *zone) -{ - struct sk_buff *skb; - - skb = mempool_alloc(zone->pool, GFP_ATOMIC); - if (skb) - atomic_inc(&zone->allocated); - return skb; -} - static int -iscsi_broadcast_skb(struct mempool_zone *zone, struct sk_buff *skb, gfp_t gfp) +iscsi_broadcast_skb(struct sk_buff *skb, gfp_t gfp) { - unsigned long flags; int rc; - skb_get(skb); rc = netlink_broadcast(nls, skb, 0, 1, gfp); if (rc < 0) { - mempool_free(skb, zone->pool); printk(KERN_ERR "iscsi: can not broadcast skb (%d)\n", rc); return rc; } - spin_lock_irqsave(&zone->freelock, flags); - INIT_LIST_HEAD(skb_to_lh(skb)); - list_add(skb_to_lh(skb), &zone->freequeue); - spin_unlock_irqrestore(&zone->freelock, flags); return 0; } static int -iscsi_unicast_skb(struct mempool_zone *zone, struct sk_buff *skb, int pid) +iscsi_unicast_skb(struct sk_buff *skb, int pid) { - unsigned long flags; int rc; - skb_get(skb); rc = netlink_unicast(nls, skb, pid, MSG_DONTWAIT); if (rc < 0) { - mempool_free(skb, zone->pool); printk(KERN_ERR "iscsi: can not unicast skb (%d)\n", rc); return rc; } - spin_lock_irqsave(&zone->freelock, flags); - INIT_LIST_HEAD(skb_to_lh(skb)); - list_add(skb_to_lh(skb), &zone->freequeue); - spin_unlock_irqrestore(&zone->freelock, flags); - return 0; } @@ -692,9 +539,7 @@ int iscsi_recv_pdu(struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr, if (!priv) return -EINVAL; - mempool_zone_complete(conn->z_pdu); - - skb = mempool_zone_get_skb(conn->z_pdu); + skb = alloc_skb(len, GFP_ATOMIC); if (!skb) { iscsi_conn_error(conn, ISCSI_ERR_CONN_FAILED); dev_printk(KERN_ERR, &conn->dev, "iscsi: can not deliver " @@ -707,15 +552,13 @@ int iscsi_recv_pdu(struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr, memset(ev, 0, sizeof(*ev)); ev->transport_handle = iscsi_handle(conn->transport); ev->type = ISCSI_KEVENT_RECV_PDU; - if (atomic_read(&conn->z_pdu->allocated) >= conn->z_pdu->hiwat) - ev->iferror = -ENOMEM; ev->r.recv_req.cid = conn->cid; ev->r.recv_req.sid = iscsi_conn_get_sid(conn); pdu = (char*)ev + sizeof(*ev); memcpy(pdu, hdr, sizeof(struct iscsi_hdr)); memcpy(pdu + sizeof(struct iscsi_hdr), data, data_size); - return iscsi_unicast_skb(conn->z_pdu, skb, priv->daemon_pid); + return iscsi_unicast_skb(skb, priv->daemon_pid); } EXPORT_SYMBOL_GPL(iscsi_recv_pdu); @@ -731,9 +574,7 @@ void iscsi_conn_error(struct iscsi_cls_conn *conn, enum iscsi_err error) if (!priv) return; - mempool_zone_complete(conn->z_error); - - skb = mempool_zone_get_skb(conn->z_error); + skb = alloc_skb(len, GFP_ATOMIC); if (!skb) { dev_printk(KERN_ERR, &conn->dev, "iscsi: gracefully ignored " "conn error (%d)\n", error); @@ -744,13 +585,11 @@ void iscsi_conn_error(struct iscsi_cls_conn *conn, enum iscsi_err error) ev = NLMSG_DATA(nlh); ev->transport_handle = iscsi_handle(conn->transport); ev->type = ISCSI_KEVENT_CONN_ERROR; - if (atomic_read(&conn->z_error->allocated) >= conn->z_error->hiwat) - ev->iferror = -ENOMEM; ev->r.connerror.error = error; ev->r.connerror.cid = conn->cid; ev->r.connerror.sid = iscsi_conn_get_sid(conn); - iscsi_broadcast_skb(conn->z_error, skb, GFP_ATOMIC); + iscsi_broadcast_skb(skb, GFP_ATOMIC); dev_printk(KERN_INFO, &conn->dev, "iscsi: detected conn error (%d)\n", error); @@ -767,9 +606,7 @@ iscsi_if_send_reply(int pid, int seq, int type, int done, int multi, int flags = multi ? NLM_F_MULTI : 0; int t = done ? NLMSG_DONE : type; - mempool_zone_complete(z_reply); - - skb = mempool_zone_get_skb(z_reply); + skb = alloc_skb(len, GFP_ATOMIC); /* * FIXME: * user is supposed to react on iferror == -ENOMEM; @@ -780,7 +617,7 @@ iscsi_if_send_reply(int pid, int seq, int type, int done, int multi, nlh = __nlmsg_put(skb, pid, seq, t, (len - sizeof(*nlh)), 0); nlh->nlmsg_flags = flags; memcpy(NLMSG_DATA(nlh), payload, size); - return iscsi_unicast_skb(z_reply, skb, pid); + return iscsi_unicast_skb(skb, pid); } static int @@ -810,9 +647,7 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh) do { int actual_size; - mempool_zone_complete(conn->z_pdu); - - skbstat = mempool_zone_get_skb(conn->z_pdu); + skbstat = alloc_skb(len, GFP_ATOMIC); if (!skbstat) { dev_printk(KERN_ERR, &conn->dev, "iscsi: can not " "deliver stats: OOM\n"); @@ -825,8 +660,6 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh) memset(evstat, 0, sizeof(*evstat)); evstat->transport_handle = iscsi_handle(conn->transport); evstat->type = nlh->nlmsg_type; - if (atomic_read(&conn->z_pdu->allocated) >= conn->z_pdu->hiwat) - evstat->iferror = -ENOMEM; evstat->u.get_stats.cid = ev->u.get_stats.cid; evstat->u.get_stats.sid = @@ -845,7 +678,7 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh) skb_trim(skbstat, NLMSG_ALIGN(actual_size)); nlhstat->nlmsg_len = actual_size; - err = iscsi_unicast_skb(conn->z_pdu, skbstat, priv->daemon_pid); + err = iscsi_unicast_skb(skbstat, priv->daemon_pid); } while (err < 0 && err != -ECONNREFUSED); return err; @@ -876,9 +709,7 @@ int iscsi_if_destroy_session_done(struct iscsi_cls_conn *conn) session = iscsi_dev_to_session(conn->dev.parent); shost = iscsi_session_to_shost(session); - mempool_zone_complete(conn->z_pdu); - - skb = mempool_zone_get_skb(conn->z_pdu); + skb = alloc_skb(len, GFP_KERNEL); if (!skb) { dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of " "session creation event\n"); @@ -896,7 +727,7 @@ int iscsi_if_destroy_session_done(struct iscsi_cls_conn *conn) * this will occur if the daemon is not up, so we just warn * the user and when the daemon is restarted it will handle it */ - rc = iscsi_broadcast_skb(conn->z_pdu, skb, GFP_KERNEL); + rc = iscsi_broadcast_skb(skb, GFP_KERNEL); if (rc < 0) dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of " "session destruction event. Check iscsi daemon\n"); @@ -939,9 +770,7 @@ int iscsi_if_create_session_done(struct iscsi_cls_conn *conn) session = iscsi_dev_to_session(conn->dev.parent); shost = iscsi_session_to_shost(session); - mempool_zone_complete(conn->z_pdu); - - skb = mempool_zone_get_skb(conn->z_pdu); + skb = alloc_skb(len, GFP_KERNEL); if (!skb) { dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of " "session creation event\n"); @@ -959,7 +788,7 @@ int iscsi_if_create_session_done(struct iscsi_cls_conn *conn) * this will occur if the daemon is not up, so we just warn * the user and when the daemon is restarted it will handle it */ - rc = iscsi_broadcast_skb(conn->z_pdu, skb, GFP_KERNEL); + rc = iscsi_broadcast_skb(skb, GFP_KERNEL); if (rc < 0) dev_printk(KERN_ERR, &conn->dev, "Cannot notify userspace of " "session creation event. Check iscsi daemon\n"); @@ -1278,9 +1107,6 @@ iscsi_if_rx(struct sock *sk, int len) err = iscsi_if_send_reply( NETLINK_CREDS(skb)->pid, nlh->nlmsg_seq, nlh->nlmsg_type, 0, 0, ev, sizeof(*ev)); - if (atomic_read(&z_reply->allocated) >= - z_reply->hiwat) - ev->iferror = -ENOMEM; } while (err < 0 && err != -ECONNREFUSED); skb_pull(skb, rlen); } @@ -1584,32 +1410,6 @@ int iscsi_unregister_transport(struct iscsi_transport *tt) } EXPORT_SYMBOL_GPL(iscsi_unregister_transport); -static int -iscsi_rcv_nl_event(struct notifier_block *this, unsigned long event, void *ptr) -{ - struct netlink_notify *n = ptr; - - if (event == NETLINK_URELEASE && - n->protocol == NETLINK_ISCSI && n->pid) { - struct iscsi_cls_conn *conn; - unsigned long flags; - - mempool_zone_complete(z_reply); - spin_lock_irqsave(&connlock, flags); - list_for_each_entry(conn, &connlist, conn_list) { - mempool_zone_complete(conn->z_error); - mempool_zone_complete(conn->z_pdu); - } - spin_unlock_irqrestore(&connlock, flags); - } - - return NOTIFY_DONE; -} - -static struct notifier_block iscsi_nl_notifier = { - .notifier_call = iscsi_rcv_nl_event, -}; - static __init int iscsi_transport_init(void) { int err; @@ -1633,25 +1433,15 @@ static __init int iscsi_transport_init(void) if (err) goto unregister_conn_class; - err = netlink_register_notifier(&iscsi_nl_notifier); - if (err) - goto unregister_session_class; - nls = netlink_kernel_create(NETLINK_ISCSI, 1, iscsi_if_rx, THIS_MODULE); if (!nls) { err = -ENOBUFS; - goto unregister_notifier; + goto unregister_session_class; } - z_reply = mempool_zone_init(Z_MAX_REPLY, - NLMSG_SPACE(sizeof(struct iscsi_uevent)), Z_HIWAT_REPLY); - if (z_reply) - return 0; + return 0; - sock_release(nls->sk_socket); -unregister_notifier: - netlink_unregister_notifier(&iscsi_nl_notifier); unregister_session_class: transport_class_unregister(&iscsi_session_class); unregister_conn_class: @@ -1665,9 +1455,7 @@ unregister_transport_class: static void __exit iscsi_transport_exit(void) { - mempool_zone_destroy(z_reply); sock_release(nls->sk_socket); - netlink_unregister_notifier(&iscsi_nl_notifier); transport_class_unregister(&iscsi_connection_class); transport_class_unregister(&iscsi_session_class); transport_class_unregister(&iscsi_host_class); diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index 39e8332..4b95c89 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -29,7 +29,6 @@ struct scsi_transport_template; struct iscsi_transport; struct Scsi_Host; -struct mempool_zone; struct iscsi_cls_conn; struct iscsi_conn; struct iscsi_cmd_task; @@ -157,9 +156,6 @@ struct iscsi_cls_conn { int active; /* must be accessed with the connlock */ struct device dev; /* sysfs transport/container device */ - struct mempool_zone *z_error; - struct mempool_zone *z_pdu; - struct list_head freequeue; }; #define iscsi_dev_to_conn(_dev) \ -- cgit v0.10.2 From 98644047916c24258fb47c3dab2bed8a44f53b83 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 16 Oct 2006 18:09:39 -0400 Subject: [SCSI] libiscsi: fix oops in connection create failure path If connection creation fails we end up calling list_del on a invalid struct. This then causes an oops. We are not acutally using the lists (old MCS code we thought might be useful elsewhere) so this patch just removes that code. Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index c542d0e..1000fe9 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -778,6 +778,10 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) } conn = session->leadconn; + if (!conn) { + reason = FAILURE_SESSION_FREED; + goto fault; + } if (!__kfifo_get(session->cmdpool.queue, (void*)&ctask, sizeof(void*))) { @@ -1377,7 +1381,6 @@ iscsi_session_setup(struct iscsi_transport *iscsit, } spin_lock_init(&session->lock); - INIT_LIST_HEAD(&session->connections); /* initialize immediate command pool */ if (iscsi_pool_init(&session->mgmtpool, session->mgmtpool_max, @@ -1580,16 +1583,11 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn) kfree(conn->persistent_address); __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, sizeof(void*)); - list_del(&conn->item); - if (list_empty(&session->connections)) + if (session->leadconn == conn) { session->leadconn = NULL; - if (session->leadconn && session->leadconn == conn) - session->leadconn = container_of(session->connections.next, - struct iscsi_conn, item); - - if (session->leadconn == NULL) /* no connections exits.. reset sequencing */ session->cmdsn = session->max_cmdsn = session->exp_cmdsn = 1; + } spin_unlock_bh(&session->lock); kfifo_free(conn->immqueue); @@ -1777,32 +1775,12 @@ int iscsi_conn_bind(struct iscsi_cls_session *cls_session, struct iscsi_cls_conn *cls_conn, int is_leading) { struct iscsi_session *session = class_to_transport_session(cls_session); - struct iscsi_conn *tmp = ERR_PTR(-EEXIST), *conn = cls_conn->dd_data; + struct iscsi_conn *conn = cls_conn->dd_data; - /* lookup for existing connection */ spin_lock_bh(&session->lock); - list_for_each_entry(tmp, &session->connections, item) { - if (tmp == conn) { - if (conn->c_stage != ISCSI_CONN_STOPPED || - conn->stop_stage == STOP_CONN_TERM) { - printk(KERN_ERR "iscsi: can't bind " - "non-stopped connection (%d:%d)\n", - conn->c_stage, conn->stop_stage); - spin_unlock_bh(&session->lock); - return -EIO; - } - break; - } - } - if (tmp != conn) { - /* bind new iSCSI connection to session */ - conn->session = session; - list_add(&conn->item, &session->connections); - } - spin_unlock_bh(&session->lock); - if (is_leading) session->leadconn = conn; + spin_unlock_bh(&session->lock); /* * Unblock xmitworker(), Login Phase will pass through. diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h index 401192e..61eebec 100644 --- a/include/scsi/libiscsi.h +++ b/include/scsi/libiscsi.h @@ -136,7 +136,6 @@ struct iscsi_conn { /* control data */ int id; /* CID */ - struct list_head item; /* maintains list of conns */ int c_stage; /* connection state */ /* * Preallocated buffer for pdus that have data but do not @@ -235,10 +234,8 @@ struct iscsi_session { * - mgmtpool, * * - r2tpool */ int state; /* session state */ - struct list_head item; int age; /* counts session re-opens */ - struct list_head connections; /* list of connections */ int cmds_max; /* size of cmds array */ struct iscsi_cmd_task **cmds; /* Original Cmds arr */ struct iscsi_queue cmdpool; /* PDU's pool */ -- cgit v0.10.2 From cd529a46e17b43976d05c1e2ece2676ec7941cc8 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 16 Oct 2006 18:09:40 -0400 Subject: [SCSI] libiscsi: fix missed iscsi_task_put in xmit error path from bhalevy@gmail.com: It looks like change 652 to libiscsi.c added some dead code around line 670 if (rc) { spin_unlock_bh(&conn->session->lock); goto again; } since 5 lines above we goto again if (rc). It looks like the previous if (rc) should go away if we want to put the ctask before breaking out of the while loop with "goto again" (see following patch). Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 1000fe9..e3a2ec2 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -661,8 +661,6 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) spin_unlock_bh(&conn->session->lock); rc = tt->xmit_cmd_task(conn, conn->ctask); - if (rc) - goto again; spin_lock_bh(&conn->session->lock); __iscsi_put_ctask(conn->ctask); -- cgit v0.10.2 From 5831c737f724aa6a655a908d202221f079f30036 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 16 Oct 2006 18:09:41 -0400 Subject: [SCSI] libiscsi: fix aen support We have been dropping the pdu. We should just send it to userspace and let it handle it. Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index e3a2ec2..f5a9560 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -481,8 +481,8 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, break; case ISCSI_OP_ASYNC_EVENT: conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - /* we need sth like iscsi_async_event_rsp() */ - rc = ISCSI_ERR_BAD_OPCODE; + if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen)) + rc = ISCSI_ERR_CONN_FAILED; break; default: rc = ISCSI_ERR_BAD_OPCODE; -- cgit v0.10.2 From b5072ea0910e5c8c79b8313e0ef70ca763983dbf Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 16 Oct 2006 18:09:42 -0400 Subject: [SCSI] libiscsi: fix logout pdu processing According to the iscsi RFC, we cannot send other requests if we have sent a logout pdu. This patch enforces this requirement by blocking the session and suspending the send thread. Userspace decides if we restart the connection or if we just free everything. Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index f5a9560..2865ebd 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -578,6 +578,27 @@ void iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err) } EXPORT_SYMBOL_GPL(iscsi_conn_failure); +static int iscsi_xmit_imm_task(struct iscsi_conn *conn) +{ + struct iscsi_hdr *hdr = conn->mtask->hdr; + int rc, was_logout = 0; + + if ((hdr->opcode & ISCSI_OPCODE_MASK) == ISCSI_OP_LOGOUT) { + conn->session->state = ISCSI_STATE_IN_RECOVERY; + iscsi_block_session(session_to_cls(conn->session)); + was_logout = 1; + } + rc = conn->session->tt->xmit_mgmt_task(conn, conn->mtask); + if (rc) + return rc; + + if (was_logout) { + set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); + return -ENODATA; + } + return 0; +} + /** * iscsi_data_xmit - xmit any command into the scheduled connection * @conn: iscsi connection @@ -623,7 +644,7 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) conn->ctask = NULL; } if (conn->mtask) { - rc = tt->xmit_mgmt_task(conn, conn->mtask); + rc = iscsi_xmit_imm_task(conn); if (rc) goto again; /* done with this in-progress mtask */ @@ -638,7 +659,7 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) list_add_tail(&conn->mtask->running, &conn->mgmt_run_list); spin_unlock_bh(&conn->session->lock); - rc = tt->xmit_mgmt_task(conn, conn->mtask); + rc = iscsi_xmit_imm_task(conn); if (rc) goto again; } -- cgit v0.10.2 From cd00b7f5d814ba87b36371f122ce36ba4a88ba69 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Tue, 17 Oct 2006 09:32:06 -0400 Subject: [SCSI] qla1280 bus reset typo Fix typo in check of return value of qla1280_bus_reset() which would result in an adapter reset in addition to the bus reset. Signed-off-by: Jes Sorensen Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla1280.c b/drivers/scsi/qla1280.c index 2521d54..16af5b7 100644 --- a/drivers/scsi/qla1280.c +++ b/drivers/scsi/qla1280.c @@ -931,11 +931,10 @@ qla1280_error_action(struct scsi_cmnd *cmd, enum action action) case BUS_RESET: if (qla1280_verbose) - printk(KERN_INFO "qla1280(%ld:%d): Issuing BUS " - "DEVICE RESET\n", ha->host_no, bus); - if (qla1280_bus_reset(ha, bus == 0)) + printk(KERN_INFO "qla1280(%ld:%d): Issued bus " + "reset.\n", ha->host_no, bus); + if (qla1280_bus_reset(ha, bus) == 0) result = SUCCESS; - break; case ADAPTER_RESET: -- cgit v0.10.2 From 5a09e39810ae0465016c380962e12dd115779b87 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Fri, 20 Oct 2006 09:58:47 +0200 Subject: [SCSI] scsi_debug: support REPORT TARGET PORT GROUPS This patch adds support for REPORT TARGET PORT GROUPS. This is used eg for the multipathing priority callout to determine the path priority. With this patch multipath-tools can use the existing mpath_prio_alua callout to exercise the path priority grouping. Signed-off-by: Hannes Reinecke Signed-off-by: Douglas Gilbert Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c index 9c0f358..30ee3d7 100644 --- a/drivers/scsi/scsi_debug.c +++ b/drivers/scsi/scsi_debug.c @@ -52,7 +52,7 @@ #include "scsi_debug.h" #define SCSI_DEBUG_VERSION "1.80" -static const char * scsi_debug_version_date = "20060914"; +static const char * scsi_debug_version_date = "20061018"; /* Additional Sense Code (ASC) used */ #define NO_ADDITIONAL_SENSE 0x0 @@ -254,6 +254,8 @@ static int resp_requests(struct scsi_cmnd * SCpnt, struct sdebug_dev_info * devip); static int resp_start_stop(struct scsi_cmnd * scp, struct sdebug_dev_info * devip); +static int resp_report_tgtpgs(struct scsi_cmnd * scp, + struct sdebug_dev_info * devip); static int resp_readcap(struct scsi_cmnd * SCpnt, struct sdebug_dev_info * devip); static int resp_readcap16(struct scsi_cmnd * SCpnt, @@ -287,9 +289,9 @@ static void __init sdebug_build_parts(unsigned char * ramp); static void __init init_all_queued(void); static void stop_all_queued(void); static int stop_queued_cmnd(struct scsi_cmnd * cmnd); -static int inquiry_evpd_83(unsigned char * arr, int target_dev_id, - int dev_id_num, const char * dev_id_str, - int dev_id_str_len); +static int inquiry_evpd_83(unsigned char * arr, int port_group_id, + int target_dev_id, int dev_id_num, + const char * dev_id_str, int dev_id_str_len); static int inquiry_evpd_88(unsigned char * arr, int target_dev_id); static int do_create_driverfs_files(void); static void do_remove_driverfs_files(void); @@ -422,6 +424,15 @@ int scsi_debug_queuecommand(struct scsi_cmnd * SCpnt, done_funct_t done) } errsts = resp_readcap16(SCpnt, devip); break; + case MAINTENANCE_IN: + if (MI_REPORT_TARGET_PGS != cmd[1]) { + mk_sense_buffer(devip, ILLEGAL_REQUEST, + INVALID_OPCODE, 0); + errsts = check_condition_result; + break; + } + errsts = resp_report_tgtpgs(SCpnt, devip); + break; case READ_16: case READ_12: case READ_10: @@ -665,8 +676,9 @@ static const char * inq_vendor_id = "Linux "; static const char * inq_product_id = "scsi_debug "; static const char * inq_product_rev = "0004"; -static int inquiry_evpd_83(unsigned char * arr, int target_dev_id, - int dev_id_num, const char * dev_id_str, +static int inquiry_evpd_83(unsigned char * arr, int port_group_id, + int target_dev_id, int dev_id_num, + const char * dev_id_str, int dev_id_str_len) { int num, port_a; @@ -720,6 +732,15 @@ static int inquiry_evpd_83(unsigned char * arr, int target_dev_id, arr[num++] = (port_a >> 16) & 0xff; arr[num++] = (port_a >> 8) & 0xff; arr[num++] = port_a & 0xff; + /* NAA-5, Target port group identifier */ + arr[num++] = 0x61; /* proto=sas, binary */ + arr[num++] = 0x95; /* piv=1, target port group id */ + arr[num++] = 0x0; + arr[num++] = 0x4; + arr[num++] = 0; + arr[num++] = 0; + arr[num++] = (port_group_id >> 8) & 0xff; + arr[num++] = port_group_id & 0xff; /* NAA-5, Target device identifier */ arr[num++] = 0x61; /* proto=sas, binary */ arr[num++] = 0xa3; /* piv=1, target device, naa */ @@ -928,12 +949,12 @@ static int resp_inquiry(struct scsi_cmnd * scp, int target, struct sdebug_dev_info * devip) { unsigned char pq_pdt; - unsigned char arr[SDEBUG_MAX_INQ_ARR_SZ]; + unsigned char * arr; unsigned char *cmd = (unsigned char *)scp->cmnd; - int alloc_len, n; + int alloc_len, n, ret; alloc_len = (cmd[3] << 8) + cmd[4]; - memset(arr, 0, SDEBUG_MAX_INQ_ARR_SZ); + arr = kzalloc(SDEBUG_MAX_INQ_ARR_SZ, GFP_KERNEL); if (devip->wlun) pq_pdt = 0x1e; /* present, wlun */ else if (scsi_debug_no_lun_0 && (0 == devip->lun)) @@ -944,12 +965,15 @@ static int resp_inquiry(struct scsi_cmnd * scp, int target, if (0x2 & cmd[1]) { /* CMDDT bit set */ mk_sense_buffer(devip, ILLEGAL_REQUEST, INVALID_FIELD_IN_CDB, 0); + kfree(arr); return check_condition_result; } else if (0x1 & cmd[1]) { /* EVPD bit set */ - int lu_id_num, target_dev_id, len; + int lu_id_num, port_group_id, target_dev_id, len; char lu_id_str[6]; int host_no = devip->sdbg_host->shost->host_no; + port_group_id = (((host_no + 1) & 0x7f) << 8) + + (devip->channel & 0x7f); if (0 == scsi_debug_vpd_use_hostno) host_no = 0; lu_id_num = devip->wlun ? -1 : (((host_no + 1) * 2000) + @@ -977,8 +1001,9 @@ static int resp_inquiry(struct scsi_cmnd * scp, int target, memcpy(&arr[4], lu_id_str, len); } else if (0x83 == cmd[2]) { /* device identification */ arr[1] = cmd[2]; /*sanity */ - arr[3] = inquiry_evpd_83(&arr[4], target_dev_id, - lu_id_num, lu_id_str, len); + arr[3] = inquiry_evpd_83(&arr[4], port_group_id, + target_dev_id, lu_id_num, + lu_id_str, len); } else if (0x84 == cmd[2]) { /* Software interface ident. */ arr[1] = cmd[2]; /*sanity */ arr[3] = inquiry_evpd_84(&arr[4]); @@ -1012,17 +1037,22 @@ static int resp_inquiry(struct scsi_cmnd * scp, int target, /* Illegal request, invalid field in cdb */ mk_sense_buffer(devip, ILLEGAL_REQUEST, INVALID_FIELD_IN_CDB, 0); + kfree(arr); return check_condition_result; } len = min(((arr[2] << 8) + arr[3]) + 4, alloc_len); - return fill_from_dev_buffer(scp, arr, + ret = fill_from_dev_buffer(scp, arr, min(len, SDEBUG_MAX_INQ_ARR_SZ)); + kfree(arr); + return ret; } /* drops through here for a standard inquiry */ arr[1] = DEV_REMOVEABLE(target) ? 0x80 : 0; /* Removable disk */ arr[2] = scsi_debug_scsi_level; arr[3] = 2; /* response_data_format==2 */ arr[4] = SDEBUG_LONG_INQ_SZ - 5; + if (0 == scsi_debug_vpd_use_hostno) + arr[5] = 0x10; /* claim: implicit TGPS */ arr[6] = 0x10; /* claim: MultiP */ /* arr[6] |= 0x40; ... claim: EncServ (enclosure services) */ arr[7] = 0xa; /* claim: LINKED + CMDQUE */ @@ -1039,8 +1069,10 @@ static int resp_inquiry(struct scsi_cmnd * scp, int target, arr[n++] = 0x3; arr[n++] = 0x60; /* SSC-2 no version */ } arr[n++] = 0xc; arr[n++] = 0xf; /* SAS-1.1 rev 10 */ - return fill_from_dev_buffer(scp, arr, + ret = fill_from_dev_buffer(scp, arr, min(alloc_len, SDEBUG_LONG_INQ_SZ)); + kfree(arr); + return ret; } static int resp_requests(struct scsi_cmnd * scp, @@ -1171,6 +1203,87 @@ static int resp_readcap16(struct scsi_cmnd * scp, min(alloc_len, SDEBUG_READCAP16_ARR_SZ)); } +#define SDEBUG_MAX_TGTPGS_ARR_SZ 1412 + +static int resp_report_tgtpgs(struct scsi_cmnd * scp, + struct sdebug_dev_info * devip) +{ + unsigned char *cmd = (unsigned char *)scp->cmnd; + unsigned char * arr; + int host_no = devip->sdbg_host->shost->host_no; + int n, ret, alen, rlen; + int port_group_a, port_group_b, port_a, port_b; + + alen = ((cmd[6] << 24) + (cmd[7] << 16) + (cmd[8] << 8) + + cmd[9]); + + arr = kzalloc(SDEBUG_MAX_TGTPGS_ARR_SZ, GFP_KERNEL); + /* + * EVPD page 0x88 states we have two ports, one + * real and a fake port with no device connected. + * So we create two port groups with one port each + * and set the group with port B to unavailable. + */ + port_a = 0x1; /* relative port A */ + port_b = 0x2; /* relative port B */ + port_group_a = (((host_no + 1) & 0x7f) << 8) + + (devip->channel & 0x7f); + port_group_b = (((host_no + 1) & 0x7f) << 8) + + (devip->channel & 0x7f) + 0x80; + + /* + * The asymmetric access state is cycled according to the host_id. + */ + n = 4; + if (0 == scsi_debug_vpd_use_hostno) { + arr[n++] = host_no % 3; /* Asymm access state */ + arr[n++] = 0x0F; /* claim: all states are supported */ + } else { + arr[n++] = 0x0; /* Active/Optimized path */ + arr[n++] = 0x01; /* claim: only support active/optimized paths */ + } + arr[n++] = (port_group_a >> 8) & 0xff; + arr[n++] = port_group_a & 0xff; + arr[n++] = 0; /* Reserved */ + arr[n++] = 0; /* Status code */ + arr[n++] = 0; /* Vendor unique */ + arr[n++] = 0x1; /* One port per group */ + arr[n++] = 0; /* Reserved */ + arr[n++] = 0; /* Reserved */ + arr[n++] = (port_a >> 8) & 0xff; + arr[n++] = port_a & 0xff; + arr[n++] = 3; /* Port unavailable */ + arr[n++] = 0x08; /* claim: only unavailalbe paths are supported */ + arr[n++] = (port_group_b >> 8) & 0xff; + arr[n++] = port_group_b & 0xff; + arr[n++] = 0; /* Reserved */ + arr[n++] = 0; /* Status code */ + arr[n++] = 0; /* Vendor unique */ + arr[n++] = 0x1; /* One port per group */ + arr[n++] = 0; /* Reserved */ + arr[n++] = 0; /* Reserved */ + arr[n++] = (port_b >> 8) & 0xff; + arr[n++] = port_b & 0xff; + + rlen = n - 4; + arr[0] = (rlen >> 24) & 0xff; + arr[1] = (rlen >> 16) & 0xff; + arr[2] = (rlen >> 8) & 0xff; + arr[3] = rlen & 0xff; + + /* + * Return the smallest value of either + * - The allocated length + * - The constructed command length + * - The maximum array size + */ + rlen = min(alen,n); + ret = fill_from_dev_buffer(scp, arr, + min(rlen, SDEBUG_MAX_TGTPGS_ARR_SZ)); + kfree(arr); + return ret; +} + /* <> */ static int resp_err_recov_pg(unsigned char * p, int pcontrol, int target) diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h index 84a6d5f..8a3f0bd 100644 --- a/include/scsi/scsi.h +++ b/include/scsi/scsi.h @@ -97,6 +97,7 @@ extern const unsigned char scsi_command_size[8]; #define PERSISTENT_RESERVE_IN 0x5e #define PERSISTENT_RESERVE_OUT 0x5f #define REPORT_LUNS 0xa0 +#define MAINTENANCE_IN 0xa3 #define MOVE_MEDIUM 0xa5 #define EXCHANGE_MEDIUM 0xa6 #define READ_12 0xa8 @@ -114,6 +115,8 @@ extern const unsigned char scsi_command_size[8]; #define SERVICE_ACTION_IN 0x9e /* values for service action in */ #define SAI_READ_CAPACITY_16 0x10 +/* values for maintenance in */ +#define MI_REPORT_TARGET_PGS 0x0a /* Values for T10/04-262r7 */ #define ATA_16 0x85 /* 16-byte pass-thru */ -- cgit v0.10.2 From 6d07cb71fdacc710fd9816cddb5c2df0f7bd96b4 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Fri, 20 Oct 2006 14:48:40 -0700 Subject: [SCSI] drivers/scsi: Handcrafted MIN/MAX macro removal Cleanups done to use min/max macros from kernel.h. Handcrafted MIN/MAX macros are changed to use macros in kernel.h [akpm@osdl.org: fix warning] Signed-off-by: Amol Lad Signed-off-by: Andrew Morton Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx.h b/drivers/scsi/aic7xxx/aic79xx.h index df3346b..c58ac6a 100644 --- a/drivers/scsi/aic7xxx/aic79xx.h +++ b/drivers/scsi/aic7xxx/aic79xx.h @@ -53,14 +53,6 @@ struct ahd_platform_data; struct scb_platform_data; /****************************** Useful Macros *********************************/ -#ifndef MAX -#define MAX(a,b) (((a) > (b)) ? (a) : (b)) -#endif - -#ifndef MIN -#define MIN(a,b) (((a) < (b)) ? (a) : (b)) -#endif - #ifndef TRUE #define TRUE 1 #endif diff --git a/drivers/scsi/aic7xxx/aic79xx_core.c b/drivers/scsi/aic7xxx/aic79xx_core.c index 653818d..6adcf79 100644 --- a/drivers/scsi/aic7xxx/aic79xx_core.c +++ b/drivers/scsi/aic7xxx/aic79xx_core.c @@ -2850,14 +2850,14 @@ ahd_devlimited_syncrate(struct ahd_softc *ahd, transinfo = &tinfo->goal; *ppr_options &= (transinfo->ppr_options|MSG_EXT_PPR_PCOMP_EN); if (transinfo->width == MSG_EXT_WDTR_BUS_8_BIT) { - maxsync = MAX(maxsync, AHD_SYNCRATE_ULTRA2); + maxsync = max(maxsync, (u_int)AHD_SYNCRATE_ULTRA2); *ppr_options &= ~MSG_EXT_PPR_DT_REQ; } if (transinfo->period == 0) { *period = 0; *ppr_options = 0; } else { - *period = MAX(*period, transinfo->period); + *period = max(*period, (u_int)transinfo->period); ahd_find_syncrate(ahd, period, ppr_options, maxsync); } } @@ -2924,12 +2924,12 @@ ahd_validate_offset(struct ahd_softc *ahd, maxoffset = MAX_OFFSET_PACED; } else maxoffset = MAX_OFFSET_NON_PACED; - *offset = MIN(*offset, maxoffset); + *offset = min(*offset, maxoffset); if (tinfo != NULL) { if (role == ROLE_TARGET) - *offset = MIN(*offset, tinfo->user.offset); + *offset = min(*offset, (u_int)tinfo->user.offset); else - *offset = MIN(*offset, tinfo->goal.offset); + *offset = min(*offset, (u_int)tinfo->goal.offset); } } @@ -2955,9 +2955,9 @@ ahd_validate_width(struct ahd_softc *ahd, struct ahd_initiator_tinfo *tinfo, } if (tinfo != NULL) { if (role == ROLE_TARGET) - *bus_width = MIN(tinfo->user.width, *bus_width); + *bus_width = min((u_int)tinfo->user.width, *bus_width); else - *bus_width = MIN(tinfo->goal.width, *bus_width); + *bus_width = min((u_int)tinfo->goal.width, *bus_width); } } @@ -6057,9 +6057,9 @@ ahd_alloc_scbs(struct ahd_softc *ahd) #endif } - newcount = MIN(scb_data->sense_left, scb_data->scbs_left); - newcount = MIN(newcount, scb_data->sgs_left); - newcount = MIN(newcount, (AHD_SCB_MAX_ALLOC - scb_data->numscbs)); + newcount = min(scb_data->sense_left, scb_data->scbs_left); + newcount = min(newcount, scb_data->sgs_left); + newcount = min(newcount, (AHD_SCB_MAX_ALLOC - scb_data->numscbs)); for (i = 0; i < newcount; i++) { struct scb_platform_data *pdata; u_int col_tag; @@ -8668,7 +8668,7 @@ ahd_resolve_seqaddr(struct ahd_softc *ahd, u_int address) if (skip_addr > i) { int end_addr; - end_addr = MIN(address, skip_addr); + end_addr = min(address, skip_addr); address_offset += end_addr - i; i = skip_addr; } else { diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.c b/drivers/scsi/aic7xxx/aic79xx_osm.c index d8d6687..77ef4d9 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm.c +++ b/drivers/scsi/aic7xxx/aic79xx_osm.c @@ -1814,9 +1814,9 @@ ahd_linux_handle_scsi_status(struct ahd_softc *ahd, u_int sense_offset; if (scb->flags & SCB_SENSE) { - sense_size = MIN(sizeof(struct scsi_sense_data) + sense_size = min(sizeof(struct scsi_sense_data) - ahd_get_sense_residual(scb), - sizeof(cmd->sense_buffer)); + (u_long)sizeof(cmd->sense_buffer)); sense_offset = 0; } else { /* @@ -1825,7 +1825,8 @@ ahd_linux_handle_scsi_status(struct ahd_softc *ahd, */ siu = (struct scsi_status_iu_header *) scb->sense_data; - sense_size = MIN(scsi_4btoul(siu->sense_length), + sense_size = min_t(size_t, + scsi_4btoul(siu->sense_length), sizeof(cmd->sense_buffer)); sense_offset = SIU_SENSE_OFFSET(siu); } diff --git a/drivers/scsi/aic7xxx/aic7xxx.h b/drivers/scsi/aic7xxx/aic7xxx.h index 62ff8c3..4850820 100644 --- a/drivers/scsi/aic7xxx/aic7xxx.h +++ b/drivers/scsi/aic7xxx/aic7xxx.h @@ -54,14 +54,6 @@ struct scb_platform_data; struct seeprom_descriptor; /****************************** Useful Macros *********************************/ -#ifndef MAX -#define MAX(a,b) (((a) > (b)) ? (a) : (b)) -#endif - -#ifndef MIN -#define MIN(a,b) (((a) < (b)) ? (a) : (b)) -#endif - #ifndef TRUE #define TRUE 1 #endif diff --git a/drivers/scsi/aic7xxx/aic7xxx_core.c b/drivers/scsi/aic7xxx/aic7xxx_core.c index 93e4e409..46bd7bc 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_core.c +++ b/drivers/scsi/aic7xxx/aic7xxx_core.c @@ -1671,7 +1671,7 @@ ahc_devlimited_syncrate(struct ahc_softc *ahc, transinfo = &tinfo->goal; *ppr_options &= transinfo->ppr_options; if (transinfo->width == MSG_EXT_WDTR_BUS_8_BIT) { - maxsync = MAX(maxsync, AHC_SYNCRATE_ULTRA2); + maxsync = max(maxsync, (u_int)AHC_SYNCRATE_ULTRA2); *ppr_options &= ~MSG_EXT_PPR_DT_REQ; } if (transinfo->period == 0) { @@ -1679,7 +1679,7 @@ ahc_devlimited_syncrate(struct ahc_softc *ahc, *ppr_options = 0; return (NULL); } - *period = MAX(*period, transinfo->period); + *period = max(*period, (u_int)transinfo->period); return (ahc_find_syncrate(ahc, period, ppr_options, maxsync)); } @@ -1804,12 +1804,12 @@ ahc_validate_offset(struct ahc_softc *ahc, else maxoffset = MAX_OFFSET_8BIT; } - *offset = MIN(*offset, maxoffset); + *offset = min(*offset, maxoffset); if (tinfo != NULL) { if (role == ROLE_TARGET) - *offset = MIN(*offset, tinfo->user.offset); + *offset = min(*offset, (u_int)tinfo->user.offset); else - *offset = MIN(*offset, tinfo->goal.offset); + *offset = min(*offset, (u_int)tinfo->goal.offset); } } @@ -1835,9 +1835,9 @@ ahc_validate_width(struct ahc_softc *ahc, struct ahc_initiator_tinfo *tinfo, } if (tinfo != NULL) { if (role == ROLE_TARGET) - *bus_width = MIN(tinfo->user.width, *bus_width); + *bus_width = min((u_int)tinfo->user.width, *bus_width); else - *bus_width = MIN(tinfo->goal.width, *bus_width); + *bus_width = min((u_int)tinfo->goal.width, *bus_width); } } @@ -4406,7 +4406,7 @@ ahc_alloc_scbs(struct ahc_softc *ahc) physaddr = sg_map->sg_physaddr; newcount = (PAGE_SIZE / (AHC_NSEG * sizeof(struct ahc_dma_seg))); - newcount = MIN(newcount, (AHC_SCB_MAX_ALLOC - scb_data->numscbs)); + newcount = min(newcount, (AHC_SCB_MAX_ALLOC - scb_data->numscbs)); for (i = 0; i < newcount; i++) { struct scb_platform_data *pdata; #ifndef __linux__ @@ -6442,7 +6442,7 @@ ahc_download_instr(struct ahc_softc *ahc, u_int instrptr, uint8_t *dconsts) if (skip_addr > i) { int end_addr; - end_addr = MIN(address, skip_addr); + end_addr = min(address, skip_addr); address_offset += end_addr - i; i = skip_addr; } else { diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.c b/drivers/scsi/aic7xxx/aic7xxx_osm.c index ad8578e..8eb1211 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.c @@ -1876,9 +1876,9 @@ ahc_linux_handle_scsi_status(struct ahc_softc *ahc, if (scb->flags & SCB_SENSE) { u_int sense_size; - sense_size = MIN(sizeof(struct scsi_sense_data) + sense_size = min(sizeof(struct scsi_sense_data) - ahc_get_sense_residual(scb), - sizeof(cmd->sense_buffer)); + (u_long)sizeof(cmd->sense_buffer)); memcpy(cmd->sense_buffer, ahc_get_sense_buf(ahc, scb), sense_size); if (sense_size < sizeof(cmd->sense_buffer)) -- cgit v0.10.2 From 289fe5b1f99c5e61ed32796cbed0a1ecc3589041 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 20 Oct 2006 14:47:57 -0700 Subject: [SCSI] aic7xxx: cleanups - make needlessly global code static - #if 0 the following unused global functions: - aic79xx_core.c: ahd_print_scb - aic79xx_core.c: ahd_suspend - aic79xx_core.c: ahd_resume - aic79xx_core.c: ahd_dump_scbs - aic79xx_osm.c: ahd_softc_comp Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Acked-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx.h b/drivers/scsi/aic7xxx/aic79xx.h index c58ac6a..b19a07c 100644 --- a/drivers/scsi/aic7xxx/aic79xx.h +++ b/drivers/scsi/aic7xxx/aic79xx.h @@ -964,8 +964,6 @@ int ahd_read_seeprom(struct ahd_softc *ahd, uint16_t *buf, int ahd_write_seeprom(struct ahd_softc *ahd, uint16_t *buf, u_int start_addr, u_int count); -int ahd_wait_seeprom(struct ahd_softc *ahd); -int ahd_verify_vpd_cksum(struct vpd_config *vpd); int ahd_verify_cksum(struct seeprom_config *sc); int ahd_acquire_seeprom(struct ahd_softc *ahd); void ahd_release_seeprom(struct ahd_softc *ahd); @@ -1312,8 +1310,6 @@ struct ahd_pci_identity { char *name; ahd_device_setup_t *setup; }; -extern struct ahd_pci_identity ahd_pci_ident_table []; -extern const u_int ahd_num_pci_devs; /***************************** VL/EISA Declarations ***************************/ struct aic7770_identity { @@ -1331,15 +1327,6 @@ extern const int ahd_num_aic7770_devs; /*************************** Function Declarations ****************************/ /******************************************************************************/ void ahd_reset_cmds_pending(struct ahd_softc *ahd); -u_int ahd_find_busy_tcl(struct ahd_softc *ahd, u_int tcl); -void ahd_busy_tcl(struct ahd_softc *ahd, - u_int tcl, u_int busyid); -static __inline void ahd_unbusy_tcl(struct ahd_softc *ahd, u_int tcl); -static __inline void -ahd_unbusy_tcl(struct ahd_softc *ahd, u_int tcl) -{ - ahd_busy_tcl(ahd, tcl, SCB_LIST_NULL); -} /***************************** PCI Front End *********************************/ struct ahd_pci_identity *ahd_find_pci_device(ahd_dev_softc_t); @@ -1348,7 +1335,6 @@ int ahd_pci_config(struct ahd_softc *, int ahd_pci_test_register_access(struct ahd_softc *); /************************** SCB and SCB queue management **********************/ -int ahd_probe_scbs(struct ahd_softc *); void ahd_qinfifo_requeue_tail(struct ahd_softc *ahd, struct scb *scb); int ahd_match_scb(struct ahd_softc *ahd, struct scb *scb, @@ -1366,33 +1352,20 @@ int ahd_parse_vpddata(struct ahd_softc *ahd, int ahd_parse_cfgdata(struct ahd_softc *ahd, struct seeprom_config *sc); void ahd_intr_enable(struct ahd_softc *ahd, int enable); -void ahd_update_coalescing_values(struct ahd_softc *ahd, - u_int timer, - u_int maxcmds, - u_int mincmds); -void ahd_enable_coalescing(struct ahd_softc *ahd, - int enable); void ahd_pause_and_flushwork(struct ahd_softc *ahd); int ahd_suspend(struct ahd_softc *ahd); -int ahd_resume(struct ahd_softc *ahd); void ahd_set_unit(struct ahd_softc *, int); void ahd_set_name(struct ahd_softc *, char *); struct scb *ahd_get_scb(struct ahd_softc *ahd, u_int col_idx); void ahd_free_scb(struct ahd_softc *ahd, struct scb *scb); -void ahd_alloc_scbs(struct ahd_softc *ahd); void ahd_free(struct ahd_softc *ahd); int ahd_reset(struct ahd_softc *ahd, int reinit); -void ahd_shutdown(void *arg); int ahd_write_flexport(struct ahd_softc *ahd, u_int addr, u_int value); int ahd_read_flexport(struct ahd_softc *ahd, u_int addr, uint8_t *value); -int ahd_wait_flexport(struct ahd_softc *ahd); /*************************** Interrupt Services *******************************/ -void ahd_pci_intr(struct ahd_softc *ahd); -void ahd_clear_intstat(struct ahd_softc *ahd); -void ahd_flush_qoutfifo(struct ahd_softc *ahd); void ahd_run_qoutfifo(struct ahd_softc *ahd); #ifdef AHD_TARGET_MODE void ahd_run_tqinfifo(struct ahd_softc *ahd, int paused); @@ -1401,7 +1374,6 @@ void ahd_handle_hwerrint(struct ahd_softc *ahd); void ahd_handle_seqint(struct ahd_softc *ahd, u_int intstat); void ahd_handle_scsiint(struct ahd_softc *ahd, u_int intstat); -void ahd_clear_critical_section(struct ahd_softc *ahd); /***************************** Error Recovery *********************************/ typedef enum { @@ -1418,23 +1390,9 @@ int ahd_search_disc_list(struct ahd_softc *ahd, int target, char channel, int lun, u_int tag, int stop_on_first, int remove, int save_state); -void ahd_freeze_devq(struct ahd_softc *ahd, struct scb *scb); int ahd_reset_channel(struct ahd_softc *ahd, char channel, int initiate_reset); -int ahd_abort_scbs(struct ahd_softc *ahd, int target, - char channel, int lun, u_int tag, - role_t role, uint32_t status); -void ahd_restart(struct ahd_softc *ahd); -void ahd_clear_fifo(struct ahd_softc *ahd, u_int fifo); -void ahd_handle_scb_status(struct ahd_softc *ahd, - struct scb *scb); -void ahd_handle_scsi_status(struct ahd_softc *ahd, - struct scb *scb); -void ahd_calc_residual(struct ahd_softc *ahd, - struct scb *scb); /*************************** Utility Functions ********************************/ -struct ahd_phase_table_entry* - ahd_lookup_phase_entry(int phase); void ahd_compile_devinfo(struct ahd_devinfo *devinfo, u_int our_id, u_int target, u_int lun, char channel, @@ -1442,14 +1400,6 @@ void ahd_compile_devinfo(struct ahd_devinfo *devinfo, /************************** Transfer Negotiation ******************************/ void ahd_find_syncrate(struct ahd_softc *ahd, u_int *period, u_int *ppr_options, u_int maxsync); -void ahd_validate_offset(struct ahd_softc *ahd, - struct ahd_initiator_tinfo *tinfo, - u_int period, u_int *offset, - int wide, role_t role); -void ahd_validate_width(struct ahd_softc *ahd, - struct ahd_initiator_tinfo *tinfo, - u_int *bus_width, - role_t role); /* * Negotiation types. These are used to qualify if we should renegotiate * even if our goal and current transport parameters are identical. @@ -1520,10 +1470,8 @@ extern uint32_t ahd_debug; #define AHD_SHOW_INT_COALESCING 0x10000 #define AHD_DEBUG_SEQUENCER 0x20000 #endif -void ahd_print_scb(struct scb *scb); void ahd_print_devinfo(struct ahd_softc *ahd, struct ahd_devinfo *devinfo); -void ahd_dump_sglist(struct scb *scb); void ahd_dump_card_state(struct ahd_softc *ahd); int ahd_print_register(ahd_reg_parse_entry_t *table, u_int num_entries, @@ -1532,5 +1480,4 @@ int ahd_print_register(ahd_reg_parse_entry_t *table, u_int value, u_int *cur_column, u_int wrap_point); -void ahd_dump_scbs(struct ahd_softc *ahd); #endif /* _AIC79XX_H_ */ diff --git a/drivers/scsi/aic7xxx/aic79xx_core.c b/drivers/scsi/aic7xxx/aic79xx_core.c index 6adcf79..7d53c64 100644 --- a/drivers/scsi/aic7xxx/aic79xx_core.c +++ b/drivers/scsi/aic7xxx/aic79xx_core.c @@ -52,7 +52,7 @@ /***************************** Lookup Tables **********************************/ -char *ahd_chip_names[] = +static char *ahd_chip_names[] = { "NONE", "aic7901", @@ -237,10 +237,33 @@ static int ahd_handle_target_cmd(struct ahd_softc *ahd, struct target_cmd *cmd); #endif +static int ahd_abort_scbs(struct ahd_softc *ahd, int target, + char channel, int lun, u_int tag, + role_t role, uint32_t status); +static void ahd_alloc_scbs(struct ahd_softc *ahd); +static void ahd_busy_tcl(struct ahd_softc *ahd, u_int tcl, + u_int scbid); +static void ahd_calc_residual(struct ahd_softc *ahd, + struct scb *scb); +static void ahd_clear_critical_section(struct ahd_softc *ahd); +static void ahd_clear_intstat(struct ahd_softc *ahd); +static void ahd_enable_coalescing(struct ahd_softc *ahd, + int enable); +static u_int ahd_find_busy_tcl(struct ahd_softc *ahd, u_int tcl); +static void ahd_freeze_devq(struct ahd_softc *ahd, + struct scb *scb); +static void ahd_handle_scb_status(struct ahd_softc *ahd, + struct scb *scb); +static struct ahd_phase_table_entry* ahd_lookup_phase_entry(int phase); +static void ahd_shutdown(void *arg); +static void ahd_update_coalescing_values(struct ahd_softc *ahd, + u_int timer, + u_int maxcmds, + u_int mincmds); +static int ahd_verify_vpd_cksum(struct vpd_config *vpd); +static int ahd_wait_seeprom(struct ahd_softc *ahd); + /******************************** Private Inlines *****************************/ -static __inline void ahd_assert_atn(struct ahd_softc *ahd); -static __inline int ahd_currently_packetized(struct ahd_softc *ahd); -static __inline int ahd_set_active_fifo(struct ahd_softc *ahd); static __inline void ahd_assert_atn(struct ahd_softc *ahd) @@ -294,11 +317,44 @@ ahd_set_active_fifo(struct ahd_softc *ahd) } } +static __inline void +ahd_unbusy_tcl(struct ahd_softc *ahd, u_int tcl) +{ + ahd_busy_tcl(ahd, tcl, SCB_LIST_NULL); +} + +/* + * Determine whether the sequencer reported a residual + * for this SCB/transaction. + */ +static __inline void +ahd_update_residual(struct ahd_softc *ahd, struct scb *scb) +{ + uint32_t sgptr; + + sgptr = ahd_le32toh(scb->hscb->sgptr); + if ((sgptr & SG_STATUS_VALID) != 0) + ahd_calc_residual(ahd, scb); +} + +static __inline void +ahd_complete_scb(struct ahd_softc *ahd, struct scb *scb) +{ + uint32_t sgptr; + + sgptr = ahd_le32toh(scb->hscb->sgptr); + if ((sgptr & SG_STATUS_VALID) != 0) + ahd_handle_scb_status(ahd, scb); + else + ahd_done(ahd, scb); +} + + /************************* Sequencer Execution Control ************************/ /* * Restart the sequencer program from address zero */ -void +static void ahd_restart(struct ahd_softc *ahd) { @@ -342,7 +398,7 @@ ahd_restart(struct ahd_softc *ahd) ahd_unpause(ahd); } -void +static void ahd_clear_fifo(struct ahd_softc *ahd, u_int fifo) { ahd_mode_state saved_modes; @@ -366,7 +422,7 @@ ahd_clear_fifo(struct ahd_softc *ahd, u_int fifo) * Flush and completed commands that are sitting in the command * complete queues down on the chip but have yet to be dma'ed back up. */ -void +static void ahd_flush_qoutfifo(struct ahd_softc *ahd) { struct scb *scb; @@ -905,6 +961,51 @@ ahd_handle_hwerrint(struct ahd_softc *ahd) ahd_free(ahd); } +#ifdef AHD_DEBUG +static void +ahd_dump_sglist(struct scb *scb) +{ + int i; + + if (scb->sg_count > 0) { + if ((scb->ahd_softc->flags & AHD_64BIT_ADDRESSING) != 0) { + struct ahd_dma64_seg *sg_list; + + sg_list = (struct ahd_dma64_seg*)scb->sg_list; + for (i = 0; i < scb->sg_count; i++) { + uint64_t addr; + uint32_t len; + + addr = ahd_le64toh(sg_list[i].addr); + len = ahd_le32toh(sg_list[i].len); + printf("sg[%d] - Addr 0x%x%x : Length %d%s\n", + i, + (uint32_t)((addr >> 32) & 0xFFFFFFFF), + (uint32_t)(addr & 0xFFFFFFFF), + sg_list[i].len & AHD_SG_LEN_MASK, + (sg_list[i].len & AHD_DMA_LAST_SEG) + ? " Last" : ""); + } + } else { + struct ahd_dma_seg *sg_list; + + sg_list = (struct ahd_dma_seg*)scb->sg_list; + for (i = 0; i < scb->sg_count; i++) { + uint32_t len; + + len = ahd_le32toh(sg_list[i].len); + printf("sg[%d] - Addr 0x%x%x : Length %d%s\n", + i, + (len & AHD_SG_HIGH_ADDR_MASK) >> 24, + ahd_le32toh(sg_list[i].addr), + len & AHD_SG_LEN_MASK, + len & AHD_DMA_LAST_SEG ? " Last" : ""); + } + } + } +} +#endif /* AHD_DEBUG */ + void ahd_handle_seqint(struct ahd_softc *ahd, u_int intstat) { @@ -2523,7 +2624,7 @@ ahd_force_renegotiation(struct ahd_softc *ahd, struct ahd_devinfo *devinfo) } #define AHD_MAX_STEPS 2000 -void +static void ahd_clear_critical_section(struct ahd_softc *ahd) { ahd_mode_state saved_modes; @@ -2646,7 +2747,7 @@ ahd_clear_critical_section(struct ahd_softc *ahd) /* * Clear any pending interrupt status. */ -void +static void ahd_clear_intstat(struct ahd_softc *ahd) { AHD_ASSERT_MODES(ahd, ~(AHD_MODE_UNKNOWN_MSK|AHD_MODE_CFG_MSK), @@ -2677,6 +2778,8 @@ ahd_clear_intstat(struct ahd_softc *ahd) #ifdef AHD_DEBUG uint32_t ahd_debug = AHD_DEBUG_OPTS; #endif + +#if 0 void ahd_print_scb(struct scb *scb) { @@ -2701,49 +2804,7 @@ ahd_print_scb(struct scb *scb) SCB_GET_TAG(scb)); ahd_dump_sglist(scb); } - -void -ahd_dump_sglist(struct scb *scb) -{ - int i; - - if (scb->sg_count > 0) { - if ((scb->ahd_softc->flags & AHD_64BIT_ADDRESSING) != 0) { - struct ahd_dma64_seg *sg_list; - - sg_list = (struct ahd_dma64_seg*)scb->sg_list; - for (i = 0; i < scb->sg_count; i++) { - uint64_t addr; - uint32_t len; - - addr = ahd_le64toh(sg_list[i].addr); - len = ahd_le32toh(sg_list[i].len); - printf("sg[%d] - Addr 0x%x%x : Length %d%s\n", - i, - (uint32_t)((addr >> 32) & 0xFFFFFFFF), - (uint32_t)(addr & 0xFFFFFFFF), - sg_list[i].len & AHD_SG_LEN_MASK, - (sg_list[i].len & AHD_DMA_LAST_SEG) - ? " Last" : ""); - } - } else { - struct ahd_dma_seg *sg_list; - - sg_list = (struct ahd_dma_seg*)scb->sg_list; - for (i = 0; i < scb->sg_count; i++) { - uint32_t len; - - len = ahd_le32toh(sg_list[i].len); - printf("sg[%d] - Addr 0x%x%x : Length %d%s\n", - i, - (len & AHD_SG_HIGH_ADDR_MASK) >> 24, - ahd_le32toh(sg_list[i].addr), - len & AHD_SG_LEN_MASK, - len & AHD_DMA_LAST_SEG ? " Last" : ""); - } - } - } -} +#endif /* 0 */ /************************* Transfer Negotiation *******************************/ /* @@ -2906,7 +2967,7 @@ ahd_find_syncrate(struct ahd_softc *ahd, u_int *period, * Truncate the given synchronous offset to a value the * current adapter type and syncrate are capable of. */ -void +static void ahd_validate_offset(struct ahd_softc *ahd, struct ahd_initiator_tinfo *tinfo, u_int period, u_int *offset, int wide, @@ -2937,7 +2998,7 @@ ahd_validate_offset(struct ahd_softc *ahd, * Truncate the given transfer width parameter to a value the * current adapter type is capable of. */ -void +static void ahd_validate_width(struct ahd_softc *ahd, struct ahd_initiator_tinfo *tinfo, u_int *bus_width, role_t role) { @@ -3466,7 +3527,7 @@ ahd_print_devinfo(struct ahd_softc *ahd, struct ahd_devinfo *devinfo) devinfo->target, devinfo->lun); } -struct ahd_phase_table_entry* +static struct ahd_phase_table_entry* ahd_lookup_phase_entry(int phase) { struct ahd_phase_table_entry *entry; @@ -5351,7 +5412,7 @@ ahd_free(struct ahd_softc *ahd) return; } -void +static void ahd_shutdown(void *arg) { struct ahd_softc *ahd; @@ -5480,7 +5541,7 @@ ahd_reset(struct ahd_softc *ahd, int reinit) /* * Determine the number of SCBs available on the controller */ -int +static int ahd_probe_scbs(struct ahd_softc *ahd) { int i; @@ -5929,7 +5990,7 @@ ahd_free_scb(struct ahd_softc *ahd, struct scb *scb) ahd_platform_scb_free(ahd, scb); } -void +static void ahd_alloc_scbs(struct ahd_softc *ahd) { struct scb_data *scb_data; @@ -6982,7 +7043,7 @@ ahd_intr_enable(struct ahd_softc *ahd, int enable) ahd_outb(ahd, HCNTRL, hcntrl); } -void +static void ahd_update_coalescing_values(struct ahd_softc *ahd, u_int timer, u_int maxcmds, u_int mincmds) { @@ -7000,7 +7061,7 @@ ahd_update_coalescing_values(struct ahd_softc *ahd, u_int timer, u_int maxcmds, ahd_outb(ahd, INT_COALESCING_MINCMDS, -mincmds); } -void +static void ahd_enable_coalescing(struct ahd_softc *ahd, int enable) { @@ -7070,6 +7131,7 @@ ahd_pause_and_flushwork(struct ahd_softc *ahd) ahd->flags &= ~AHD_ALL_INTERRUPTS; } +#if 0 int ahd_suspend(struct ahd_softc *ahd) { @@ -7083,7 +7145,9 @@ ahd_suspend(struct ahd_softc *ahd) ahd_shutdown(ahd); return (0); } +#endif /* 0 */ +#if 0 int ahd_resume(struct ahd_softc *ahd) { @@ -7093,6 +7157,7 @@ ahd_resume(struct ahd_softc *ahd) ahd_restart(ahd); return (0); } +#endif /* 0 */ /************************** Busy Target Table *********************************/ /* @@ -7125,7 +7190,7 @@ ahd_index_busy_tcl(struct ahd_softc *ahd, u_int *saved_scbid, u_int tcl) /* * Return the untagged transaction id for a given target/channel lun. */ -u_int +static u_int ahd_find_busy_tcl(struct ahd_softc *ahd, u_int tcl) { u_int scbid; @@ -7138,7 +7203,7 @@ ahd_find_busy_tcl(struct ahd_softc *ahd, u_int tcl) return (scbid); } -void +static void ahd_busy_tcl(struct ahd_softc *ahd, u_int tcl, u_int scbid) { u_int scb_offset; @@ -7186,7 +7251,7 @@ ahd_match_scb(struct ahd_softc *ahd, struct scb *scb, int target, return match; } -void +static void ahd_freeze_devq(struct ahd_softc *ahd, struct scb *scb) { int target; @@ -7690,7 +7755,7 @@ ahd_add_scb_to_free_list(struct ahd_softc *ahd, u_int scbid) * been modified from CAM_REQ_INPROG. This routine assumes that the sequencer * is paused before it is called. */ -int +static int ahd_abort_scbs(struct ahd_softc *ahd, int target, char channel, int lun, u_int tag, role_t role, uint32_t status) { @@ -8019,18 +8084,8 @@ ahd_stat_timer(void *arg) } /****************************** Status Processing *****************************/ -void -ahd_handle_scb_status(struct ahd_softc *ahd, struct scb *scb) -{ - if (scb->hscb->shared_data.istatus.scsi_status != 0) { - ahd_handle_scsi_status(ahd, scb); - } else { - ahd_calc_residual(ahd, scb); - ahd_done(ahd, scb); - } -} -void +static void ahd_handle_scsi_status(struct ahd_softc *ahd, struct scb *scb) { struct hardware_scb *hscb; @@ -8238,10 +8293,21 @@ ahd_handle_scsi_status(struct ahd_softc *ahd, struct scb *scb) } } +static void +ahd_handle_scb_status(struct ahd_softc *ahd, struct scb *scb) +{ + if (scb->hscb->shared_data.istatus.scsi_status != 0) { + ahd_handle_scsi_status(ahd, scb); + } else { + ahd_calc_residual(ahd, scb); + ahd_done(ahd, scb); + } +} + /* * Calculate the residual for a just completed SCB. */ -void +static void ahd_calc_residual(struct ahd_softc *ahd, struct scb *scb) { struct hardware_scb *hscb; @@ -9092,6 +9158,7 @@ ahd_dump_card_state(struct ahd_softc *ahd) ahd_unpause(ahd); } +#if 0 void ahd_dump_scbs(struct ahd_softc *ahd) { @@ -9117,6 +9184,7 @@ ahd_dump_scbs(struct ahd_softc *ahd) ahd_set_scbptr(ahd, saved_scb_index); ahd_restore_modes(ahd, saved_modes); } +#endif /* 0 */ /**************************** Flexport Logic **********************************/ /* @@ -9219,7 +9287,7 @@ ahd_write_seeprom(struct ahd_softc *ahd, uint16_t *buf, /* * Wait ~100us for the serial eeprom to satisfy our request. */ -int +static int ahd_wait_seeprom(struct ahd_softc *ahd) { int cnt; @@ -9237,7 +9305,7 @@ ahd_wait_seeprom(struct ahd_softc *ahd) * Validate the two checksums in the per_channel * vital product data struct. */ -int +static int ahd_verify_vpd_cksum(struct vpd_config *vpd) { int i; @@ -9316,6 +9384,24 @@ ahd_release_seeprom(struct ahd_softc *ahd) /* Currently a no-op */ } +/* + * Wait at most 2 seconds for flexport arbitration to succeed. + */ +static int +ahd_wait_flexport(struct ahd_softc *ahd) +{ + int cnt; + + AHD_ASSERT_MODES(ahd, AHD_MODE_SCSI_MSK, AHD_MODE_SCSI_MSK); + cnt = 1000000 * 2 / 5; + while ((ahd_inb(ahd, BRDCTL) & FLXARBACK) == 0 && --cnt) + ahd_delay(5); + + if (cnt == 0) + return (ETIMEDOUT); + return (0); +} + int ahd_write_flexport(struct ahd_softc *ahd, u_int addr, u_int value) { @@ -9357,24 +9443,6 @@ ahd_read_flexport(struct ahd_softc *ahd, u_int addr, uint8_t *value) return (0); } -/* - * Wait at most 2 seconds for flexport arbitration to succeed. - */ -int -ahd_wait_flexport(struct ahd_softc *ahd) -{ - int cnt; - - AHD_ASSERT_MODES(ahd, AHD_MODE_SCSI_MSK, AHD_MODE_SCSI_MSK); - cnt = 1000000 * 2 / 5; - while ((ahd_inb(ahd, BRDCTL) & FLXARBACK) == 0 && --cnt) - ahd_delay(5); - - if (cnt == 0) - return (ETIMEDOUT); - return (0); -} - /************************* Target Mode ****************************************/ #ifdef AHD_TARGET_MODE cam_status diff --git a/drivers/scsi/aic7xxx/aic79xx_inline.h b/drivers/scsi/aic7xxx/aic79xx_inline.h index a3266e0..2ceb67f 100644 --- a/drivers/scsi/aic7xxx/aic79xx_inline.h +++ b/drivers/scsi/aic7xxx/aic79xx_inline.h @@ -418,10 +418,6 @@ ahd_targetcmd_offset(struct ahd_softc *ahd, u_int index) } /*********************** Miscelaneous Support Functions ***********************/ -static __inline void ahd_complete_scb(struct ahd_softc *ahd, - struct scb *scb); -static __inline void ahd_update_residual(struct ahd_softc *ahd, - struct scb *scb); static __inline struct ahd_initiator_tinfo * ahd_fetch_transinfo(struct ahd_softc *ahd, char channel, u_int our_id, @@ -467,32 +463,6 @@ static __inline uint32_t ahd_get_sense_bufaddr(struct ahd_softc *ahd, struct scb *scb); -static __inline void -ahd_complete_scb(struct ahd_softc *ahd, struct scb *scb) -{ - uint32_t sgptr; - - sgptr = ahd_le32toh(scb->hscb->sgptr); - if ((sgptr & SG_STATUS_VALID) != 0) - ahd_handle_scb_status(ahd, scb); - else - ahd_done(ahd, scb); -} - -/* - * Determine whether the sequencer reported a residual - * for this SCB/transaction. - */ -static __inline void -ahd_update_residual(struct ahd_softc *ahd, struct scb *scb) -{ - uint32_t sgptr; - - sgptr = ahd_le32toh(scb->hscb->sgptr); - if ((sgptr & SG_STATUS_VALID) != 0) - ahd_calc_residual(ahd, scb); -} - /* * Return pointers to the transfer negotiation information * for the specified our_id/remote_id pair. diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.c b/drivers/scsi/aic7xxx/aic79xx_osm.c index 77ef4d9..5e13046 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm.c +++ b/drivers/scsi/aic7xxx/aic79xx_osm.c @@ -293,7 +293,7 @@ static uint32_t aic79xx_seltime; * force all outstanding transactions to be serviced prior to a new * transaction. */ -uint32_t aic79xx_periodic_otag; +static uint32_t aic79xx_periodic_otag; /* Some storage boxes are using an LSI chip which has a bug making it * impossible to use aic79xx Rev B chip in 320 speeds. The following diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.h b/drivers/scsi/aic7xxx/aic79xx_osm.h index fb3d4dd..3a67fc5 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm.h +++ b/drivers/scsi/aic7xxx/aic79xx_osm.h @@ -506,9 +506,6 @@ struct info_str { int pos; }; -void ahd_format_transinfo(struct info_str *info, - struct ahd_transinfo *tinfo); - /******************************** Locking *************************************/ static __inline void ahd_lockinit(struct ahd_softc *ahd) @@ -582,8 +579,6 @@ ahd_unlock(struct ahd_softc *ahd, unsigned long *flags) #define PCIXM_STATUS_MAXCRDS 0x1C00 /* Maximum Cumulative Read Size */ #define PCIXM_STATUS_RCVDSCEM 0x2000 /* Received a Split Comp w/Error msg */ -extern struct pci_driver aic79xx_pci_driver; - typedef enum { AHD_POWER_STATE_D0, diff --git a/drivers/scsi/aic7xxx/aic79xx_osm_pci.c b/drivers/scsi/aic7xxx/aic79xx_osm_pci.c index 4b53542..2001fe8 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm_pci.c +++ b/drivers/scsi/aic7xxx/aic79xx_osm_pci.c @@ -82,7 +82,7 @@ static struct pci_device_id ahd_linux_pci_id_table[] = { MODULE_DEVICE_TABLE(pci, ahd_linux_pci_id_table); -struct pci_driver aic79xx_pci_driver = { +static struct pci_driver aic79xx_pci_driver = { .name = "aic79xx", .probe = ahd_linux_pci_dev_probe, .remove = ahd_linux_pci_dev_remove, diff --git a/drivers/scsi/aic7xxx/aic79xx_pci.c b/drivers/scsi/aic7xxx/aic79xx_pci.c index 14850f3..c077358 100644 --- a/drivers/scsi/aic7xxx/aic79xx_pci.c +++ b/drivers/scsi/aic7xxx/aic79xx_pci.c @@ -97,7 +97,7 @@ static ahd_device_setup_t ahd_aic7901A_setup; static ahd_device_setup_t ahd_aic7902_setup; static ahd_device_setup_t ahd_aic790X_setup; -struct ahd_pci_identity ahd_pci_ident_table [] = +static struct ahd_pci_identity ahd_pci_ident_table [] = { /* aic7901 based controllers */ { @@ -201,7 +201,7 @@ struct ahd_pci_identity ahd_pci_ident_table [] = } }; -const u_int ahd_num_pci_devs = ARRAY_SIZE(ahd_pci_ident_table); +static const u_int ahd_num_pci_devs = ARRAY_SIZE(ahd_pci_ident_table); #define DEVCONFIG 0x40 #define PCIXINITPAT 0x0000E000ul @@ -245,6 +245,7 @@ static int ahd_check_extport(struct ahd_softc *ahd); static void ahd_configure_termination(struct ahd_softc *ahd, u_int adapter_control); static void ahd_pci_split_intr(struct ahd_softc *ahd, u_int intstat); +static void ahd_pci_intr(struct ahd_softc *ahd); struct ahd_pci_identity * ahd_find_pci_device(ahd_dev_softc_t pci) @@ -757,7 +758,7 @@ static const char *pci_status_strings[] = "%s: Address or Write Phase Parity Error Detected in %s.\n" }; -void +static void ahd_pci_intr(struct ahd_softc *ahd) { uint8_t pci_status[8]; diff --git a/drivers/scsi/aic7xxx/aic79xx_proc.c b/drivers/scsi/aic7xxx/aic79xx_proc.c index c5f0ee5..6b28beb 100644 --- a/drivers/scsi/aic7xxx/aic79xx_proc.c +++ b/drivers/scsi/aic7xxx/aic79xx_proc.c @@ -136,7 +136,7 @@ copy_info(struct info_str *info, char *fmt, ...) return (len); } -void +static void ahd_format_transinfo(struct info_str *info, struct ahd_transinfo *tinfo) { u_int speed; diff --git a/drivers/scsi/aic7xxx/aic7xxx.h b/drivers/scsi/aic7xxx/aic7xxx.h index 4850820..5802f33 100644 --- a/drivers/scsi/aic7xxx/aic7xxx.h +++ b/drivers/scsi/aic7xxx/aic7xxx.h @@ -1127,8 +1127,6 @@ struct ahc_pci_identity { char *name; ahc_device_setup_t *setup; }; -extern struct ahc_pci_identity ahc_pci_ident_table[]; -extern const u_int ahc_num_pci_devs; /***************************** VL/EISA Declarations ***************************/ struct aic7770_identity { diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.c b/drivers/scsi/aic7xxx/aic7xxx_osm.c index 8eb1211..ed85057 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.c @@ -328,7 +328,7 @@ static uint32_t aic7xxx_seltime; * force all outstanding transactions to be serviced prior to a new * transaction. */ -uint32_t aic7xxx_periodic_otag; +static uint32_t aic7xxx_periodic_otag; /* * Module information and settable options. diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.h b/drivers/scsi/aic7xxx/aic7xxx_osm.h index a87a4ce..a36b338 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.h +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.h @@ -533,8 +533,6 @@ ahc_unlock(struct ahc_softc *ahc, unsigned long *flags) #define PCIR_SUBVEND_0 0x2c #define PCIR_SUBDEV_0 0x2e -extern struct pci_driver aic7xxx_pci_driver; - typedef enum { AHC_POWER_STATE_D0, diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c b/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c index d20ca51..ea5687d 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c @@ -130,7 +130,7 @@ static struct pci_device_id ahc_linux_pci_id_table[] = { MODULE_DEVICE_TABLE(pci, ahc_linux_pci_id_table); -struct pci_driver aic7xxx_pci_driver = { +static struct pci_driver aic7xxx_pci_driver = { .name = "aic7xxx", .probe = ahc_linux_pci_dev_probe, .remove = ahc_linux_pci_dev_remove, diff --git a/drivers/scsi/aic7xxx/aic7xxx_pci.c b/drivers/scsi/aic7xxx/aic7xxx_pci.c index 63cab2d..09c8172 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_pci.c +++ b/drivers/scsi/aic7xxx/aic7xxx_pci.c @@ -168,7 +168,7 @@ static ahc_device_setup_t ahc_aha394XX_setup; static ahc_device_setup_t ahc_aha494XX_setup; static ahc_device_setup_t ahc_aha398XX_setup; -struct ahc_pci_identity ahc_pci_ident_table [] = +static struct ahc_pci_identity ahc_pci_ident_table [] = { /* aic7850 based controllers */ { @@ -559,7 +559,7 @@ struct ahc_pci_identity ahc_pci_ident_table [] = } }; -const u_int ahc_num_pci_devs = ARRAY_SIZE(ahc_pci_ident_table); +static const u_int ahc_num_pci_devs = ARRAY_SIZE(ahc_pci_ident_table); #define AHC_394X_SLOT_CHANNEL_A 4 #define AHC_394X_SLOT_CHANNEL_B 5 -- cgit v0.10.2 From ca3c3323931ef925497a9ffcb61c5eebe55f8e2b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 20 Oct 2006 14:48:07 -0700 Subject: [SCSI] aic79xx: make ahd_set_tags() static Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Acked-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx.h b/drivers/scsi/aic7xxx/aic79xx.h index b19a07c..170a434 100644 --- a/drivers/scsi/aic7xxx/aic79xx.h +++ b/drivers/scsi/aic7xxx/aic79xx.h @@ -1428,11 +1428,6 @@ typedef enum { AHD_QUEUE_TAGGED } ahd_queue_alg; -void ahd_set_tags(struct ahd_softc *ahd, - struct scsi_cmnd *cmd, - struct ahd_devinfo *devinfo, - ahd_queue_alg alg); - /**************************** Target Mode *************************************/ #ifdef AHD_TARGET_MODE void ahd_send_lstate_events(struct ahd_softc *, diff --git a/drivers/scsi/aic7xxx/aic79xx_core.c b/drivers/scsi/aic7xxx/aic79xx_core.c index 7d53c64..4c2b5a81 100644 --- a/drivers/scsi/aic7xxx/aic79xx_core.c +++ b/drivers/scsi/aic7xxx/aic79xx_core.c @@ -3271,7 +3271,7 @@ ahd_set_width(struct ahd_softc *ahd, struct ahd_devinfo *devinfo, /* * Update the current state of tagged queuing for a given target. */ -void +static void ahd_set_tags(struct ahd_softc *ahd, struct scsi_cmnd *cmd, struct ahd_devinfo *devinfo, ahd_queue_alg alg) { -- cgit v0.10.2 From afc071e6281e4f2af4748b5ddc594334726a37cf Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 23 Oct 2006 21:47:13 -0700 Subject: [SCSI] lpfc: fix printk format warning Fix printk format warning: drivers/scsi/lpfc/lpfc_attr.c:597: warning: long long unsigned int format, uint64_t arg (arg 4) Signed-off-by: Randy Dunlap Acked-by: James Smart Signed-off-by: James Bottomley diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c index 9496e87..2a4e02e 100644 --- a/drivers/scsi/lpfc/lpfc_attr.c +++ b/drivers/scsi/lpfc/lpfc_attr.c @@ -594,7 +594,8 @@ lpfc_soft_wwpn_show(struct class_device *cdev, char *buf) { struct Scsi_Host *host = class_to_shost(cdev); struct lpfc_hba *phba = (struct lpfc_hba*)host->hostdata; - return snprintf(buf, PAGE_SIZE, "0x%llx\n", phba->cfg_soft_wwpn); + return snprintf(buf, PAGE_SIZE, "0x%llx\n", + (unsigned long long)phba->cfg_soft_wwpn); } -- cgit v0.10.2 From 4a531e8c79fe459e922347461ccc0f0c13de20d5 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Fri, 20 Oct 2006 09:08:18 +0900 Subject: [SCSI] replace u8 and u32 with __u8 and __u32 in scsi.h for user space Signed-off-by: FUJITA Tomonori Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h index 8a3f0bd..5c0e979 100644 --- a/include/scsi/scsi.h +++ b/include/scsi/scsi.h @@ -433,7 +433,7 @@ struct scsi_lun { #define SCSI_IOCTL_GET_PCI 0x5387 /* Pull a u32 out of a SCSI message (using BE SCSI conventions) */ -static inline u32 scsi_to_u32(u8 *ptr) +static inline __u32 scsi_to_u32(__u8 *ptr) { return (ptr[0]<<24) + (ptr[1]<<16) + (ptr[2]<<8) + ptr[3]; } -- cgit v0.10.2 From 8883c1f182fa88d2b8e0adb6ae90a42f67e5353e Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Mon, 23 Oct 2006 15:22:37 +0200 Subject: [SCSI] aic79xx: Fixup external device reset Whenever an external device is resetted we really have to take care to keep the channel in sync. Just notifying SCSI-ML and retry is not enough as we have to make sure the SCSI bus is not getting confused, either. So whenever we detect an external reset we rewrite the command to TUR, disable packetized command and notify the internal engine that an abort has happened. This way we trigger a proper bus reset sequence and all devices will be renegotiated properly. Kudos to Justin Gibbs and Luben Tuikov for this idea. Signed-off-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx_core.c b/drivers/scsi/aic7xxx/aic79xx_core.c index 4c2b5a81..07a86a3 100644 --- a/drivers/scsi/aic7xxx/aic79xx_core.c +++ b/drivers/scsi/aic7xxx/aic79xx_core.c @@ -1154,10 +1154,12 @@ ahd_handle_seqint(struct ahd_softc *ahd, u_int intstat) * If a target takes us into the command phase * assume that it has been externally reset and * has thus lost our previous packetized negotiation - * agreement. - * Revert to async/narrow transfers until we - * can renegotiate with the device and notify - * the OSM about the reset. + * agreement. Since we have not sent an identify + * message and may not have fully qualified the + * connection, we change our command to TUR, assert + * ATN and ABORT the task when we go to message in + * phase. The OSM will see the REQUEUE_REQUEST + * status and retry the command. */ scbid = ahd_get_scbptr(ahd); scb = ahd_lookup_scb(ahd, scbid); @@ -1184,7 +1186,28 @@ ahd_handle_seqint(struct ahd_softc *ahd, u_int intstat) ahd_set_syncrate(ahd, &devinfo, /*period*/0, /*offset*/0, /*ppr_options*/0, AHD_TRANS_ACTIVE, /*paused*/TRUE); - scb->flags |= SCB_EXTERNAL_RESET; + /* Hand-craft TUR command */ + ahd_outb(ahd, SCB_CDB_STORE, 0); + ahd_outb(ahd, SCB_CDB_STORE+1, 0); + ahd_outb(ahd, SCB_CDB_STORE+2, 0); + ahd_outb(ahd, SCB_CDB_STORE+3, 0); + ahd_outb(ahd, SCB_CDB_STORE+4, 0); + ahd_outb(ahd, SCB_CDB_STORE+5, 0); + ahd_outb(ahd, SCB_CDB_LEN, 6); + scb->hscb->control &= ~(TAG_ENB|SCB_TAG_TYPE); + scb->hscb->control |= MK_MESSAGE; + ahd_outb(ahd, SCB_CONTROL, scb->hscb->control); + ahd_outb(ahd, MSG_OUT, HOST_MSG); + ahd_outb(ahd, SAVED_SCSIID, scb->hscb->scsiid); + /* + * The lun is 0, regardless of the SCB's lun + * as we have not sent an identify message. + */ + ahd_outb(ahd, SAVED_LUN, 0); + ahd_outb(ahd, SEQ_FLAGS, 0); + ahd_assert_atn(ahd); + scb->flags &= ~SCB_PACKETIZED; + scb->flags |= SCB_ABORT|SCB_EXTERNAL_RESET; ahd_freeze_devq(ahd, scb); ahd_set_transaction_status(scb, CAM_REQUEUE_REQ); ahd_freeze_scb(scb); @@ -1620,8 +1643,10 @@ ahd_handle_scsiint(struct ahd_softc *ahd, u_int intstat) /* * Ignore external resets after a bus reset. */ - if (((status & SCSIRSTI) != 0) && (ahd->flags & AHD_BUS_RESET_ACTIVE)) + if (((status & SCSIRSTI) != 0) && (ahd->flags & AHD_BUS_RESET_ACTIVE)) { + ahd_outb(ahd, CLRSINT1, CLRSCSIRSTI); return; + } /* * Clear bus reset flag @@ -2301,6 +2326,22 @@ ahd_handle_nonpkt_busfree(struct ahd_softc *ahd) if (sent_msg == MSG_ABORT_TAG) tag = SCB_GET_TAG(scb); + if ((scb->flags & SCB_EXTERNAL_RESET) != 0) { + /* + * This abort is in response to an + * unexpected switch to command phase + * for a packetized connection. Since + * the identify message was never sent, + * "saved lun" is 0. We really want to + * abort only the SCB that encountered + * this error, which could have a different + * lun. The SCB will be retried so the OS + * will see the UA after renegotiating to + * packetized. + */ + tag = SCB_GET_TAG(scb); + saved_lun = scb->hscb->lun; + } found = ahd_abort_scbs(ahd, target, 'A', saved_lun, tag, ROLE_INITIATOR, CAM_REQ_ABORTED); @@ -7985,6 +8026,11 @@ ahd_reset_channel(struct ahd_softc *ahd, char channel, int initiate_reset) ahd_clear_fifo(ahd, 1); /* + * Clear SCSI interrupt status + */ + ahd_outb(ahd, CLRSINT1, CLRSCSIRSTI); + + /* * Reenable selections */ ahd_outb(ahd, SIMODE1, ahd_inb(ahd, SIMODE1) | ENSCSIRST); @@ -8017,10 +8063,6 @@ ahd_reset_channel(struct ahd_softc *ahd, char channel, int initiate_reset) } } #endif - /* Notify the XPT that a bus reset occurred */ - ahd_send_async(ahd, devinfo.channel, CAM_TARGET_WILDCARD, - CAM_LUN_WILDCARD, AC_BUS_RESET); - /* * Revert to async/narrow transfers until we renegotiate. */ @@ -8042,6 +8084,10 @@ ahd_reset_channel(struct ahd_softc *ahd, char channel, int initiate_reset) } } + /* Notify the XPT that a bus reset occurred */ + ahd_send_async(ahd, devinfo.channel, CAM_TARGET_WILDCARD, + CAM_LUN_WILDCARD, AC_BUS_RESET); + ahd_restart(ahd); return (found); -- cgit v0.10.2 From 843822ad631889596d67716e6edbcde608aeba81 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Mon, 23 Oct 2006 15:24:23 +0200 Subject: [SCSI] aic79xx: set precompensation aic79xx has a special 'iocell' chip which handles the precompensation. If it's set via DV we should make sure to set the chip correctly, too. Signed-off-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.c b/drivers/scsi/aic7xxx/aic79xx_osm.c index 5e13046..e7a32f8 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm.c +++ b/drivers/scsi/aic7xxx/aic79xx_osm.c @@ -2636,8 +2636,22 @@ static void ahd_linux_set_pcomp_en(struct scsi_target *starget, int pcomp) pcomp ? "Enable" : "Disable"); #endif - if (pcomp) + if (pcomp) { + uint8_t precomp; + + if (ahd->unit < ARRAY_SIZE(aic79xx_iocell_info)) { + struct ahd_linux_iocell_opts *iocell_opts; + + iocell_opts = &aic79xx_iocell_info[ahd->unit]; + precomp = iocell_opts->precomp; + } else { + precomp = AIC79XX_DEFAULT_PRECOMP; + } ppr_options |= MSG_EXT_PPR_PCOMP_EN; + AHD_SET_PRECOMP(ahd, precomp); + } else { + AHD_SET_PRECOMP(ahd, 0); + } ahd_compile_devinfo(&devinfo, shost->this_id, starget->id, 0, starget->channel + 'A', ROLE_INITIATOR); -- cgit v0.10.2 From 9080063f523b09af63234a21816c825133d48c44 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Mon, 23 Oct 2006 15:25:36 +0200 Subject: [SCSI] aic7xxx: Remove slave_destroy This is a cross-port from aic79xx; we still hit the occasional BUG_ON in slave_destroy. And again we don't really need the slave_destroy callback nor the ahc_linux_target structure at all. Signed-off-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic7xxx.h b/drivers/scsi/aic7xxx/aic7xxx.h index 5802f33..954c7c2 100644 --- a/drivers/scsi/aic7xxx/aic7xxx.h +++ b/drivers/scsi/aic7xxx/aic7xxx.h @@ -1279,6 +1279,7 @@ typedef enum { } ahc_queue_alg; void ahc_set_tags(struct ahc_softc *ahc, + struct scsi_cmnd *cmd, struct ahc_devinfo *devinfo, ahc_queue_alg alg); diff --git a/drivers/scsi/aic7xxx/aic7xxx_core.c b/drivers/scsi/aic7xxx/aic7xxx_core.c index 46bd7bc..50ef785 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_core.c +++ b/drivers/scsi/aic7xxx/aic7xxx_core.c @@ -1986,7 +1986,7 @@ ahc_set_syncrate(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, tinfo->curr.ppr_options = ppr_options; ahc_send_async(ahc, devinfo->channel, devinfo->target, - CAM_LUN_WILDCARD, AC_TRANSFER_NEG, NULL); + CAM_LUN_WILDCARD, AC_TRANSFER_NEG); if (bootverbose) { if (offset != 0) { printf("%s: target %d synchronous at %sMHz%s, " @@ -2056,7 +2056,7 @@ ahc_set_width(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, tinfo->curr.width = width; ahc_send_async(ahc, devinfo->channel, devinfo->target, - CAM_LUN_WILDCARD, AC_TRANSFER_NEG, NULL); + CAM_LUN_WILDCARD, AC_TRANSFER_NEG); if (bootverbose) { printf("%s: target %d using %dbit transfers\n", ahc_name(ahc), devinfo->target, @@ -2074,12 +2074,14 @@ ahc_set_width(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, * Update the current state of tagged queuing for a given target. */ void -ahc_set_tags(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, - ahc_queue_alg alg) +ahc_set_tags(struct ahc_softc *ahc, struct scsi_cmnd *cmd, + struct ahc_devinfo *devinfo, ahc_queue_alg alg) { - ahc_platform_set_tags(ahc, devinfo, alg); + struct scsi_device *sdev = cmd->device; + + ahc_platform_set_tags(ahc, sdev, devinfo, alg); ahc_send_async(ahc, devinfo->channel, devinfo->target, - devinfo->lun, AC_TRANSFER_NEG, &alg); + devinfo->lun, AC_TRANSFER_NEG); } /* @@ -3489,7 +3491,7 @@ ahc_handle_msg_reject(struct ahc_softc *ahc, struct ahc_devinfo *devinfo) printf("(%s:%c:%d:%d): refuses tagged commands. " "Performing non-tagged I/O\n", ahc_name(ahc), devinfo->channel, devinfo->target, devinfo->lun); - ahc_set_tags(ahc, devinfo, AHC_QUEUE_NONE); + ahc_set_tags(ahc, scb->io_ctx, devinfo, AHC_QUEUE_NONE); mask = ~0x23; } else { printf("(%s:%c:%d:%d): refuses %s tagged commands. " @@ -3497,7 +3499,7 @@ ahc_handle_msg_reject(struct ahc_softc *ahc, struct ahc_devinfo *devinfo) ahc_name(ahc), devinfo->channel, devinfo->target, devinfo->lun, tag_type == MSG_ORDERED_TASK ? "ordered" : "head of queue"); - ahc_set_tags(ahc, devinfo, AHC_QUEUE_BASIC); + ahc_set_tags(ahc, scb->io_ctx, devinfo, AHC_QUEUE_BASIC); mask = ~0x03; } @@ -3763,7 +3765,7 @@ ahc_handle_devreset(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, if (status != CAM_SEL_TIMEOUT) ahc_send_async(ahc, devinfo->channel, devinfo->target, - CAM_LUN_WILDCARD, AC_SENT_BDR, NULL); + CAM_LUN_WILDCARD, AC_SENT_BDR); if (message != NULL && (verbose_level <= bootverbose)) @@ -6018,7 +6020,7 @@ ahc_reset_channel(struct ahc_softc *ahc, char channel, int initiate_reset) #endif /* Notify the XPT that a bus reset occurred */ ahc_send_async(ahc, devinfo.channel, CAM_TARGET_WILDCARD, - CAM_LUN_WILDCARD, AC_BUS_RESET, NULL); + CAM_LUN_WILDCARD, AC_BUS_RESET); /* * Revert to async/narrow transfers until we renegotiate. diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.c b/drivers/scsi/aic7xxx/aic7xxx_osm.c index ed85057..660f26e 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.c +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.c @@ -512,7 +512,6 @@ ahc_linux_target_alloc(struct scsi_target *starget) struct seeprom_config *sc = ahc->seep_config; unsigned long flags; struct scsi_target **ahc_targp = ahc_linux_target_in_softc(starget); - struct ahc_linux_target *targ = scsi_transport_target_data(starget); unsigned short scsirate; struct ahc_devinfo devinfo; struct ahc_initiator_tinfo *tinfo; @@ -533,7 +532,6 @@ ahc_linux_target_alloc(struct scsi_target *starget) BUG_ON(*ahc_targp != NULL); *ahc_targp = starget; - memset(targ, 0, sizeof(*targ)); if (sc) { int maxsync = AHC_SYNCRATE_DT; @@ -594,14 +592,11 @@ ahc_linux_slave_alloc(struct scsi_device *sdev) struct ahc_softc *ahc = *((struct ahc_softc **)sdev->host->hostdata); struct scsi_target *starget = sdev->sdev_target; - struct ahc_linux_target *targ = scsi_transport_target_data(starget); struct ahc_linux_device *dev; if (bootverbose) printf("%s: Slave Alloc %d\n", ahc_name(ahc), sdev->id); - BUG_ON(targ->sdev[sdev->lun] != NULL); - dev = scsi_transport_device_data(sdev); memset(dev, 0, sizeof(*dev)); @@ -618,8 +613,6 @@ ahc_linux_slave_alloc(struct scsi_device *sdev) */ dev->maxtags = 0; - targ->sdev[sdev->lun] = sdev; - spi_period(starget) = 0; return 0; @@ -644,22 +637,6 @@ ahc_linux_slave_configure(struct scsi_device *sdev) return 0; } -static void -ahc_linux_slave_destroy(struct scsi_device *sdev) -{ - struct ahc_softc *ahc; - struct ahc_linux_device *dev = scsi_transport_device_data(sdev); - struct ahc_linux_target *targ = scsi_transport_target_data(sdev->sdev_target); - - ahc = *((struct ahc_softc **)sdev->host->hostdata); - if (bootverbose) - printf("%s: Slave Destroy %d\n", ahc_name(ahc), sdev->id); - - BUG_ON(dev->active); - - targ->sdev[sdev->lun] = NULL; -} - #if defined(__i386__) /* * Return the disk geometry for the given SCSI device. @@ -782,7 +759,6 @@ struct scsi_host_template aic7xxx_driver_template = { .use_clustering = ENABLE_CLUSTERING, .slave_alloc = ahc_linux_slave_alloc, .slave_configure = ahc_linux_slave_configure, - .slave_destroy = ahc_linux_slave_destroy, .target_alloc = ahc_linux_target_alloc, .target_destroy = ahc_linux_target_destroy, }; @@ -1204,21 +1180,13 @@ void ahc_platform_free(struct ahc_softc *ahc) { struct scsi_target *starget; - int i, j; + int i; if (ahc->platform_data != NULL) { /* destroy all of the device and target objects */ for (i = 0; i < AHC_NUM_TARGETS; i++) { starget = ahc->platform_data->starget[i]; if (starget != NULL) { - for (j = 0; j < AHC_NUM_LUNS; j++) { - struct ahc_linux_target *targ = - scsi_transport_target_data(starget); - - if (targ->sdev[j] == NULL) - continue; - targ->sdev[j] = NULL; - } ahc->platform_data->starget[i] = NULL; } } @@ -1252,24 +1220,13 @@ ahc_platform_freeze_devq(struct ahc_softc *ahc, struct scb *scb) } void -ahc_platform_set_tags(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, - ahc_queue_alg alg) +ahc_platform_set_tags(struct ahc_softc *ahc, struct scsi_device *sdev, + struct ahc_devinfo *devinfo, ahc_queue_alg alg) { - struct scsi_target *starget; - struct ahc_linux_target *targ; struct ahc_linux_device *dev; - struct scsi_device *sdev; - u_int target_offset; int was_queuing; int now_queuing; - target_offset = devinfo->target; - if (devinfo->channel != 'A') - target_offset += 8; - starget = ahc->platform_data->starget[target_offset]; - targ = scsi_transport_target_data(starget); - BUG_ON(targ == NULL); - sdev = targ->sdev[devinfo->lun]; if (sdev == NULL) return; dev = scsi_transport_device_data(sdev); @@ -1402,11 +1359,15 @@ ahc_linux_device_queue_depth(struct scsi_device *sdev) tags = ahc_linux_user_tagdepth(ahc, &devinfo); if (tags != 0 && sdev->tagged_supported != 0) { - ahc_set_tags(ahc, &devinfo, AHC_QUEUE_TAGGED); + ahc_platform_set_tags(ahc, sdev, &devinfo, AHC_QUEUE_TAGGED); + ahc_send_async(ahc, devinfo.channel, devinfo.target, + devinfo.lun, AC_TRANSFER_NEG); ahc_print_devinfo(ahc, &devinfo); printf("Tagged Queuing enabled. Depth %d\n", tags); } else { - ahc_set_tags(ahc, &devinfo, AHC_QUEUE_NONE); + ahc_platform_set_tags(ahc, sdev, &devinfo, AHC_QUEUE_NONE); + ahc_send_async(ahc, devinfo.channel, devinfo.target, + devinfo.lun, AC_TRANSFER_NEG); } } @@ -1630,7 +1591,7 @@ ahc_platform_flushwork(struct ahc_softc *ahc) void ahc_send_async(struct ahc_softc *ahc, char channel, - u_int target, u_int lun, ac_code code, void *arg) + u_int target, u_int lun, ac_code code) { switch (code) { case AC_TRANSFER_NEG: @@ -1947,7 +1908,7 @@ ahc_linux_handle_scsi_status(struct ahc_softc *ahc, } ahc_set_transaction_status(scb, CAM_REQUEUE_REQ); ahc_set_scsi_status(scb, SCSI_STATUS_OK); - ahc_platform_set_tags(ahc, &devinfo, + ahc_platform_set_tags(ahc, sdev, &devinfo, (dev->flags & AHC_DEV_Q_BASIC) ? AHC_QUEUE_BASIC : AHC_QUEUE_TAGGED); break; @@ -1958,7 +1919,7 @@ ahc_linux_handle_scsi_status(struct ahc_softc *ahc, */ dev->openings = 1; ahc_set_scsi_status(scb, SCSI_STATUS_BUSY); - ahc_platform_set_tags(ahc, &devinfo, + ahc_platform_set_tags(ahc, sdev, &devinfo, (dev->flags & AHC_DEV_Q_BASIC) ? AHC_QUEUE_BASIC : AHC_QUEUE_TAGGED); break; @@ -2600,8 +2561,6 @@ ahc_linux_init(void) if (!ahc_linux_transport_template) return -ENODEV; - scsi_transport_reserve_target(ahc_linux_transport_template, - sizeof(struct ahc_linux_target)); scsi_transport_reserve_device(ahc_linux_transport_template, sizeof(struct ahc_linux_device)); diff --git a/drivers/scsi/aic7xxx/aic7xxx_osm.h b/drivers/scsi/aic7xxx/aic7xxx_osm.h index a36b338..85ae5d8 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_osm.h +++ b/drivers/scsi/aic7xxx/aic7xxx_osm.h @@ -256,7 +256,6 @@ typedef enum { AHC_DEV_PERIODIC_OTAG = 0x40, /* Send OTAG to prevent starvation */ } ahc_linux_dev_flags; -struct ahc_linux_target; struct ahc_linux_device { /* * The number of transactions currently @@ -329,12 +328,6 @@ struct ahc_linux_device { #define AHC_OTAG_THRESH 500 }; -struct ahc_linux_target { - struct scsi_device *sdev[AHC_NUM_LUNS]; - struct ahc_transinfo last_tinfo; - struct ahc_softc *ahc; -}; - /********************* Definitions Required by the Core ***********************/ /* * Number of SG segments we require. So long as the S/G segments for @@ -822,7 +815,7 @@ ahc_freeze_scb(struct scb *scb) } } -void ahc_platform_set_tags(struct ahc_softc *ahc, +void ahc_platform_set_tags(struct ahc_softc *ahc, struct scsi_device *sdev, struct ahc_devinfo *devinfo, ahc_queue_alg); int ahc_platform_abort_scbs(struct ahc_softc *ahc, int target, char channel, int lun, u_int tag, @@ -832,7 +825,7 @@ irqreturn_t void ahc_platform_flushwork(struct ahc_softc *ahc); void ahc_done(struct ahc_softc*, struct scb*); void ahc_send_async(struct ahc_softc *, char channel, - u_int target, u_int lun, ac_code, void *); + u_int target, u_int lun, ac_code); void ahc_print_path(struct ahc_softc *, struct scb *); void ahc_platform_dump_card_state(struct ahc_softc *ahc); diff --git a/drivers/scsi/aic7xxx/aic7xxx_proc.c b/drivers/scsi/aic7xxx/aic7xxx_proc.c index 5914b4a..99e5443 100644 --- a/drivers/scsi/aic7xxx/aic7xxx_proc.c +++ b/drivers/scsi/aic7xxx/aic7xxx_proc.c @@ -182,7 +182,6 @@ ahc_dump_target_state(struct ahc_softc *ahc, struct info_str *info, u_int our_id, char channel, u_int target_id, u_int target_offset) { - struct ahc_linux_target *targ; struct scsi_target *starget; struct ahc_initiator_tinfo *tinfo; struct ahc_tmode_tstate *tstate; @@ -198,7 +197,6 @@ ahc_dump_target_state(struct ahc_softc *ahc, struct info_str *info, starget = ahc->platform_data->starget[target_offset]; if (!starget) return; - targ = scsi_transport_target_data(starget); copy_info(info, "\tGoal: "); ahc_format_transinfo(info, &tinfo->goal); @@ -208,7 +206,7 @@ ahc_dump_target_state(struct ahc_softc *ahc, struct info_str *info, for (lun = 0; lun < AHC_NUM_LUNS; lun++) { struct scsi_device *sdev; - sdev = targ->sdev[lun]; + sdev = scsi_device_lookup_by_target(starget, lun); if (sdev == NULL) continue; @@ -383,11 +381,11 @@ ahc_linux_proc_info(struct Scsi_Host *shost, char *buffer, char **start, } copy_info(&info, "\n"); - max_targ = 15; + max_targ = 16; if ((ahc->features & (AHC_WIDE|AHC_TWIN)) == 0) - max_targ = 7; + max_targ = 8; - for (i = 0; i <= max_targ; i++) { + for (i = 0; i < max_targ; i++) { u_int our_id; u_int target_id; char channel; -- cgit v0.10.2 From d6b9ccbbeb625674891f797119f06512d27fc905 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Mon, 23 Oct 2006 15:26:37 +0200 Subject: [SCSI] aic79xx: Print out signalling This is a cross-port of a similar patch for aic7xxx; only it's a bit simpler here as we don't support HVD and all controller actually implement this register. I hope. Signed-off-by: Hannes Reinecke Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic7xxx/aic79xx_osm.c b/drivers/scsi/aic7xxx/aic79xx_osm.c index e7a32f8..9bfcca5 100644 --- a/drivers/scsi/aic7xxx/aic79xx_osm.c +++ b/drivers/scsi/aic7xxx/aic79xx_osm.c @@ -2694,7 +2694,25 @@ static void ahd_linux_set_hold_mcs(struct scsi_target *starget, int hold) ahd_unlock(ahd, &flags); } +static void ahd_linux_get_signalling(struct Scsi_Host *shost) +{ + struct ahd_softc *ahd = *(struct ahd_softc **)shost->hostdata; + unsigned long flags; + u8 mode; + ahd_lock(ahd, &flags); + ahd_pause(ahd); + mode = ahd_inb(ahd, SBLKCTL); + ahd_unpause(ahd); + ahd_unlock(ahd, &flags); + + if (mode & ENAB40) + spi_signalling(shost) = SPI_SIGNAL_LVD; + else if (mode & ENAB20) + spi_signalling(shost) = SPI_SIGNAL_SE; + else + spi_signalling(shost) = SPI_SIGNAL_UNKNOWN; +} static struct spi_function_template ahd_linux_transport_functions = { .set_offset = ahd_linux_set_offset, @@ -2719,6 +2737,7 @@ static struct spi_function_template ahd_linux_transport_functions = { .show_pcomp_en = 1, .set_hold_mcs = ahd_linux_set_hold_mcs, .show_hold_mcs = 1, + .get_signalling = ahd_linux_get_signalling, }; static int __init -- cgit v0.10.2 From 190a24f5605d95b786c92280bf7a51ebef43f29f Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Wed, 25 Oct 2006 17:32:40 -0500 Subject: [POWERPC] Make sure __cpu_preinit_ppc970 gets called on 970GX processors Add check for 970GX for __cpu_preinit_ppc970. Signed-off-by: Olof Johansson Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index f12e3c5..291e362 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -1641,6 +1641,8 @@ _GLOBAL(__start_initialization_multiplatform) cmpwi r0,0x3c /* 970FX */ beq 1f cmpwi r0,0x44 /* 970MP */ + beq 1f + cmpwi r0,0x45 /* 970GX */ bne 2f 1: bl .__cpu_preinit_ppc970 2: -- cgit v0.10.2 From 1fbece150a230d0ab447cfb2fc4df10fb89f0d8c Mon Sep 17 00:00:00 2001 From: David Brownell Date: Sat, 1 Jul 2006 13:39:55 -0700 Subject: [PATCH] pcmcia: at91_cf update More correct AT91 CF wakeup logic ... only enable/disable the IRQ wakeup capability, not the IRQ itself. That way the we know that the IRQ will be disabled correctly, in suspend/resume logic instead of ARM IRQ code. Most of the pin multiplexing setup has moved to the devices.c setup code. Signed-off-by: David Brownell Signed-off-by: Andrew Victor Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/at91_cf.c b/drivers/pcmcia/at91_cf.c index 7f5df9a..f8db6e3 100644 --- a/drivers/pcmcia/at91_cf.c +++ b/drivers/pcmcia/at91_cf.c @@ -241,12 +241,6 @@ static int __init at91_cf_probe(struct platform_device *pdev) csa = at91_sys_read(AT91_EBI_CSA); at91_sys_write(AT91_EBI_CSA, csa | AT91_EBI_CS4A_SMC_COMPACTFLASH); - /* force poweron defaults for these pins ... */ - (void) at91_set_A_periph(AT91_PIN_PC9, 0); /* A25/CFRNW */ - (void) at91_set_A_periph(AT91_PIN_PC10, 0); /* NCS4/CFCS */ - (void) at91_set_A_periph(AT91_PIN_PC11, 0); /* NCS5/CFCE1 */ - (void) at91_set_A_periph(AT91_PIN_PC12, 0); /* NCS6/CFCE2 */ - /* nWAIT is _not_ a default setting */ (void) at91_set_A_periph(AT91_PIN_PC6, 1); /* nWAIT */ @@ -322,6 +316,7 @@ fail1: if (board->irq_pin) free_irq(board->irq_pin, cf); fail0a: + device_init_wakeup(&pdev->dev, 0); free_irq(board->det_pin, cf); device_init_wakeup(&pdev->dev, 0); fail0: @@ -360,26 +355,20 @@ static int at91_cf_suspend(struct platform_device *pdev, pm_message_t mesg) struct at91_cf_data *board = cf->board; pcmcia_socket_dev_suspend(&pdev->dev, mesg); - if (device_may_wakeup(&pdev->dev)) + if (device_may_wakeup(&pdev->dev)) { enable_irq_wake(board->det_pin); - else { + if (board->irq_pin) + enable_irq_wake(board->irq_pin); + } else { disable_irq_wake(board->det_pin); - disable_irq(board->det_pin); + if (board->irq_pin) + disable_irq_wake(board->irq_pin); } - if (board->irq_pin) - disable_irq(board->irq_pin); return 0; } static int at91_cf_resume(struct platform_device *pdev) { - struct at91_cf_socket *cf = platform_get_drvdata(pdev); - struct at91_cf_data *board = cf->board; - - if (board->irq_pin) - enable_irq(board->irq_pin); - if (!device_may_wakeup(&pdev->dev)) - enable_irq(board->det_pin); pcmcia_socket_dev_resume(&pdev->dev); return 0; } -- cgit v0.10.2 From 01918d16c837485ceba92d48fb734cf520e61144 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Sun, 2 Jul 2006 21:21:51 +0200 Subject: [PATCH] pcmcia: add more IDs to hostap_cs.c As a replacement for the broad manufactor/card ID match we commented out because of conflicts with pcnet_cs, add two product ID matches. Signed-off-by: Dominik Brodowski diff --git a/drivers/net/wireless/hostap/hostap_cs.c b/drivers/net/wireless/hostap/hostap_cs.c index 686d895..f63909e 100644 --- a/drivers/net/wireless/hostap/hostap_cs.c +++ b/drivers/net/wireless/hostap/hostap_cs.c @@ -887,6 +887,13 @@ static struct pcmcia_device_id hostap_cs_ids[] = { PCMCIA_DEVICE_PROD_ID123( "U.S. Robotics", "IEEE 802.11b PC-CARD", "Version 01.02", 0xc7b8df9d, 0x1700d087, 0x4b74baa0), + PCMCIA_DEVICE_PROD_ID123( + "Allied Telesyn", "AT-WCL452 Wireless PCMCIA Radio", + "Ver. 1.00", + 0x5cd01705, 0x4271660f, 0x9d08ee12), + PCMCIA_DEVICE_PROD_ID123( + "corega", "WL PCCL-11", "ISL37300P", + 0xa21501a, 0x59868926, 0xc9049a39), PCMCIA_DEVICE_NULL }; MODULE_DEVICE_TABLE(pcmcia, hostap_cs_ids); -- cgit v0.10.2 From 4708b5faf7c53bb4128d34267bdfe4b8c74b488a Mon Sep 17 00:00:00 2001 From: Kaustav Majumdar Date: Fri, 20 Oct 2006 14:44:09 -0700 Subject: [PATCH] pcmcia: update alloc_io_space for conflict checking for multifunction PC card Some PCMCIA cards do not mention specific IO addresses in the CIS. In that case, inside the alloc_io_space function, conflicts are detected (the function returns 1) for the second function of a multifunction card unless the length of IO address range required is greater than 0x100. The following patch will remove this conflict checking for a PCMCIA function which had not mentioned any specific IO address to be mapped from. The patch is tested for Linux kernel 2.6.15.4 and works fine in the above case and is as suggested by Dave Hinds. Signed-off-by: Kaustav Majumdar Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/pcmcia_resource.c b/drivers/pcmcia/pcmcia_resource.c index 74cebd4..b9201c2 100644 --- a/drivers/pcmcia/pcmcia_resource.c +++ b/drivers/pcmcia/pcmcia_resource.c @@ -95,7 +95,7 @@ static int alloc_io_space(struct pcmcia_socket *s, u_int attr, ioaddr_t *base, * potential conflicts, just the most obvious ones. */ for (i = 0; i < MAX_IO_WIN; i++) - if ((s->io[i].res) && + if ((s->io[i].res) && *base && ((s->io[i].res->start & (align-1)) == *base)) return 1; for (i = 0; i < MAX_IO_WIN; i++) { -- cgit v0.10.2 From ace7d4772cf056d9b13b51bd496a8be968774592 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 20 Oct 2006 14:44:12 -0700 Subject: [PATCH] pcmcia/ds: driver layer error checking Check driver layer return values in pcmcia/ds.c Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index 74b3124..af392bf 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -1292,10 +1292,22 @@ struct bus_type pcmcia_bus_type = { static int __init init_pcmcia_bus(void) { + int ret; + spin_lock_init(&pcmcia_dev_list_lock); - bus_register(&pcmcia_bus_type); - class_interface_register(&pcmcia_bus_interface); + ret = bus_register(&pcmcia_bus_type); + if (ret < 0) { + printk(KERN_WARNING "pcmcia: bus_register error: %d\n", ret); + return ret; + } + ret = class_interface_register(&pcmcia_bus_interface); + if (ret < 0) { + printk(KERN_WARNING + "pcmcia: class_interface_register error: %d\n", ret); + bus_unregister(&pcmcia_bus_type); + return ret; + } pcmcia_setup_ioctl(); -- cgit v0.10.2 From f237de58b13bf65ba2f7fab896daacb92ae7ddef Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 20 Oct 2006 14:44:13 -0700 Subject: [PATCH] CONFIG_PM=n slim: drivers/pcmcia/* Remove some code which is unneeded if CONFIG_PM=n. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/i82092.c b/drivers/pcmcia/i82092.c index 82715f4..d316d95 100644 --- a/drivers/pcmcia/i82092.c +++ b/drivers/pcmcia/i82092.c @@ -41,6 +41,7 @@ static struct pci_device_id i82092aa_pci_ids[] = { }; MODULE_DEVICE_TABLE(pci, i82092aa_pci_ids); +#ifdef CONFIG_PM static int i82092aa_socket_suspend (struct pci_dev *dev, pm_message_t state) { return pcmcia_socket_dev_suspend(&dev->dev, state); @@ -50,14 +51,17 @@ static int i82092aa_socket_resume (struct pci_dev *dev) { return pcmcia_socket_dev_resume(&dev->dev); } +#endif static struct pci_driver i82092aa_pci_drv = { .name = "i82092aa", .id_table = i82092aa_pci_ids, .probe = i82092aa_pci_probe, .remove = __devexit_p(i82092aa_pci_remove), +#ifdef CONFIG_PM .suspend = i82092aa_socket_suspend, .resume = i82092aa_socket_resume, +#endif }; diff --git a/drivers/pcmcia/pd6729.c b/drivers/pcmcia/pd6729.c index c83a0a6..a70f97f 100644 --- a/drivers/pcmcia/pd6729.c +++ b/drivers/pcmcia/pd6729.c @@ -755,6 +755,7 @@ static void __devexit pd6729_pci_remove(struct pci_dev *dev) kfree(socket); } +#ifdef CONFIG_PM static int pd6729_socket_suspend(struct pci_dev *dev, pm_message_t state) { return pcmcia_socket_dev_suspend(&dev->dev, state); @@ -764,6 +765,7 @@ static int pd6729_socket_resume(struct pci_dev *dev) { return pcmcia_socket_dev_resume(&dev->dev); } +#endif static struct pci_device_id pd6729_pci_ids[] = { { @@ -781,8 +783,10 @@ static struct pci_driver pd6729_pci_drv = { .id_table = pd6729_pci_ids, .probe = pd6729_pci_probe, .remove = __devexit_p(pd6729_pci_remove), +#ifdef CONFIG_PM .suspend = pd6729_socket_suspend, .resume = pd6729_socket_resume, +#endif }; static int pd6729_module_init(void) diff --git a/drivers/pcmcia/yenta_socket.c b/drivers/pcmcia/yenta_socket.c index 26229d9..9ced52a 100644 --- a/drivers/pcmcia/yenta_socket.c +++ b/drivers/pcmcia/yenta_socket.c @@ -1213,7 +1213,7 @@ static int __devinit yenta_probe (struct pci_dev *dev, const struct pci_device_i return ret; } - +#ifdef CONFIG_PM static int yenta_dev_suspend (struct pci_dev *dev, pm_message_t state) { struct yenta_socket *socket = pci_get_drvdata(dev); @@ -1262,7 +1262,7 @@ static int yenta_dev_resume (struct pci_dev *dev) return pcmcia_socket_dev_resume(&dev->dev); } - +#endif #define CB_ID(vend,dev,type) \ { \ @@ -1359,8 +1359,10 @@ static struct pci_driver yenta_cardbus_driver = { .id_table = yenta_table, .probe = yenta_probe, .remove = __devexit_p(yenta_close), +#ifdef CONFIG_PM .suspend = yenta_dev_suspend, .resume = yenta_dev_resume, +#endif }; -- cgit v0.10.2 From f465ce176fb2f1778a04fc3fcb2b8aa564901419 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 20 Oct 2006 14:44:13 -0700 Subject: [PATCH] i82092: wire up errors from pci_register_driver() debugging goo removed to not leave assymetry in it after possible "leave" removal. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/i82092.c b/drivers/pcmcia/i82092.c index d316d95..c2ea07a 100644 --- a/drivers/pcmcia/i82092.c +++ b/drivers/pcmcia/i82092.c @@ -709,10 +709,7 @@ static int i82092aa_set_mem_map(struct pcmcia_socket *socket, struct pccard_mem_ static int i82092aa_module_init(void) { - enter("i82092aa_module_init"); - pci_register_driver(&i82092aa_pci_drv); - leave("i82092aa_module_init"); - return 0; + return pci_register_driver(&i82092aa_pci_drv); } static void i82092aa_module_exit(void) -- cgit v0.10.2 From a230a6785dd5af84b8b043a64d8df8adc81f3724 Mon Sep 17 00:00:00 2001 From: Om Narasimhan Date: Fri, 20 Oct 2006 14:44:15 -0700 Subject: [PATCH] pcmcia: au1000_generic fix The previous code did something like, if (error) goto out_err; .... do { struct au1000_pcmcia_socket *skt = PCMCIA_SOCKET(i); del_timer_sync(&skt->poll_timer); pcmcia_unregister_socket(&skt->socket); out_err: flush_scheduled_work(); ops->hw_shutdown(skt); i--; } while (i > 0) ..... - On the error path, skt would not contain a valid value for the first iteration (skt is masked by uninitialized automatic skt) - Does not do hw_shutdown() for 0th element of PCMCIA_SOCKET Signed-off-by: Om Narasimhan Cc: "Yoichi Yuasa" Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/au1000_generic.c b/drivers/pcmcia/au1000_generic.c index d5dd0ce..5387de6 100644 --- a/drivers/pcmcia/au1000_generic.c +++ b/drivers/pcmcia/au1000_generic.c @@ -351,6 +351,7 @@ struct skt_dev_info { int au1x00_pcmcia_socket_probe(struct device *dev, struct pcmcia_low_level *ops, int first, int nr) { struct skt_dev_info *sinfo; + struct au1000_pcmcia_socket *skt; int ret, i; sinfo = kzalloc(sizeof(struct skt_dev_info), GFP_KERNEL); @@ -365,7 +366,7 @@ int au1x00_pcmcia_socket_probe(struct device *dev, struct pcmcia_low_level *ops, * Initialise the per-socket structure. */ for (i = 0; i < nr; i++) { - struct au1000_pcmcia_socket *skt = PCMCIA_SOCKET(i); + skt = PCMCIA_SOCKET(i); memset(skt, 0, sizeof(*skt)); skt->socket.resource_ops = &pccard_static_ops; @@ -438,17 +439,19 @@ int au1x00_pcmcia_socket_probe(struct device *dev, struct pcmcia_low_level *ops, dev_set_drvdata(dev, sinfo); return 0; - do { - struct au1000_pcmcia_socket *skt = PCMCIA_SOCKET(i); + +out_err: + flush_scheduled_work(); + ops->hw_shutdown(skt); + while (i-- > 0) { + skt = PCMCIA_SOCKET(i); del_timer_sync(&skt->poll_timer); pcmcia_unregister_socket(&skt->socket); -out_err: flush_scheduled_work(); ops->hw_shutdown(skt); - i--; - } while (i > 0); + } kfree(sinfo); out: return ret; -- cgit v0.10.2 From 3efa9970bd0ac731302224ab9243693e91bc4bea Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Fri, 20 Oct 2006 14:44:18 -0700 Subject: [PATCH] ioremap balanced with iounmap for drivers/pcmcia ioremap must be balanced by an iounmap and failing to do so can result in a memory leak. Signed-off-by: Amol Lad Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/at91_cf.c b/drivers/pcmcia/at91_cf.c index f8db6e3..3bcb7dc 100644 --- a/drivers/pcmcia/at91_cf.c +++ b/drivers/pcmcia/at91_cf.c @@ -310,9 +310,10 @@ static int __init at91_cf_probe(struct platform_device *pdev) return 0; fail2: - iounmap((void __iomem *) cf->socket.io_offset); release_mem_region(io->start, io->end + 1 - io->start); fail1: + if (cf->socket.io_offset) + iounmap((void __iomem *) cf->socket.io_offset); if (board->irq_pin) free_irq(board->irq_pin, cf); fail0a: diff --git a/drivers/pcmcia/au1000_generic.c b/drivers/pcmcia/au1000_generic.c index 5387de6..551bde5 100644 --- a/drivers/pcmcia/au1000_generic.c +++ b/drivers/pcmcia/au1000_generic.c @@ -449,6 +449,16 @@ out_err: del_timer_sync(&skt->poll_timer); pcmcia_unregister_socket(&skt->socket); flush_scheduled_work(); + if (i == 0) { + iounmap(skt->virt_io + (u32)mips_io_port_base); + skt->virt_io = NULL; + } +#ifndef CONFIG_MIPS_XXS1500 + else { + iounmap(skt->virt_io + (u32)mips_io_port_base); + skt->virt_io = NULL; + } +#endif ops->hw_shutdown(skt); } diff --git a/drivers/pcmcia/m8xx_pcmcia.c b/drivers/pcmcia/m8xx_pcmcia.c index e070a28..3b72be8 100644 --- a/drivers/pcmcia/m8xx_pcmcia.c +++ b/drivers/pcmcia/m8xx_pcmcia.c @@ -427,7 +427,7 @@ static int voltage_set(int slot, int vcc, int vpp) reg |= BCSR1_PCCVCC1; break; default: - return 1; + goto out_unmap; } switch(vpp) { @@ -438,15 +438,15 @@ static int voltage_set(int slot, int vcc, int vpp) if(vcc == vpp) reg |= BCSR1_PCCVPP1; else - return 1; + goto out_unmap; break; case 120: if ((vcc == 33) || (vcc == 50)) reg |= BCSR1_PCCVPP0; else - return 1; + goto out_unmap; default: - return 1; + goto out_unmap; } /* first, turn off all power */ @@ -457,6 +457,10 @@ static int voltage_set(int slot, int vcc, int vpp) iounmap(bcsr_io); return 0; + +out_unmap: + iounmap(bcsr_io); + return 1; } #define socket_get(_slot_) PCMCIA_SOCKET_KEY_5V diff --git a/drivers/pcmcia/omap_cf.c b/drivers/pcmcia/omap_cf.c index c8e838c..06bf7f4 100644 --- a/drivers/pcmcia/omap_cf.c +++ b/drivers/pcmcia/omap_cf.c @@ -309,9 +309,10 @@ static int __devinit omap_cf_probe(struct device *dev) return 0; fail2: - iounmap((void __iomem *) cf->socket.io_offset); release_mem_region(cf->phys_cf, SZ_8K); fail1: + if (cf->socket.io_offset) + iounmap((void __iomem *) cf->socket.io_offset); free_irq(irq, cf); fail0: kfree(cf); -- cgit v0.10.2 From 26aaa3c202fb3bec8d6c6619122442d476f55658 Mon Sep 17 00:00:00 2001 From: Jonathan McDowell Date: Fri, 20 Oct 2006 14:44:19 -0700 Subject: [PATCH] Export soc_common_drv_pcmcia_remove to allow modular PCMCIA. Allow a modular sa1100_cs. Signed-off-by: Jonathan McDowell Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/soc_common.c b/drivers/pcmcia/soc_common.c index 3627e52..e433704 100644 --- a/drivers/pcmcia/soc_common.c +++ b/drivers/pcmcia/soc_common.c @@ -824,3 +824,4 @@ int soc_common_drv_pcmcia_remove(struct device *dev) return 0; } +EXPORT_SYMBOL(soc_common_drv_pcmcia_remove); -- cgit v0.10.2 From 4deb7c1ed2b622b565c5330b475adc5a6cea30da Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Fri, 20 Oct 2006 14:44:23 -0700 Subject: [PATCH] PCMCIA: handle sysfs, PCI errors Handle sysfs and PCI errors correctly. Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/pcmcia_ioctl.c b/drivers/pcmcia/pcmcia_ioctl.c index 9ad18e6..310ede5 100644 --- a/drivers/pcmcia/pcmcia_ioctl.c +++ b/drivers/pcmcia/pcmcia_ioctl.c @@ -128,9 +128,12 @@ static int proc_read_drivers(char *buf, char **start, off_t pos, int count, int *eof, void *data) { char *p = buf; + int rc; - bus_for_each_drv(&pcmcia_bus_type, NULL, - (void *) &p, proc_read_drivers_callback); + rc = bus_for_each_drv(&pcmcia_bus_type, NULL, + (void *) &p, proc_read_drivers_callback); + if (rc < 0) + return rc; return (p - buf); } @@ -269,8 +272,10 @@ rescan: * Prevent this racing with a card insertion. */ mutex_lock(&s->skt_mutex); - bus_rescan_devices(&pcmcia_bus_type); + ret = bus_rescan_devices(&pcmcia_bus_type); mutex_unlock(&s->skt_mutex); + if (ret) + goto err_put_module; /* check whether the driver indeed matched. I don't care if this * is racy or not, because it can only happen on cardmgr access diff --git a/drivers/pcmcia/yenta_socket.c b/drivers/pcmcia/yenta_socket.c index 9ced52a..da471bd 100644 --- a/drivers/pcmcia/yenta_socket.c +++ b/drivers/pcmcia/yenta_socket.c @@ -1197,8 +1197,12 @@ static int __devinit yenta_probe (struct pci_dev *dev, const struct pci_device_i ret = pcmcia_register_socket(&socket->socket); if (ret == 0) { /* Add the yenta register attributes */ - device_create_file(&dev->dev, &dev_attr_yenta_registers); - goto out; + ret = device_create_file(&dev->dev, &dev_attr_yenta_registers); + if (ret == 0) + goto out; + + /* error path... */ + pcmcia_unregister_socket(&socket->socket); } unmap: @@ -1248,12 +1252,18 @@ static int yenta_dev_resume (struct pci_dev *dev) struct yenta_socket *socket = pci_get_drvdata(dev); if (socket) { + int rc; + pci_set_power_state(dev, 0); /* FIXME: pci_restore_state needs to have a better interface */ pci_restore_state(dev); pci_write_config_dword(dev, 16*4, socket->saved_state[0]); pci_write_config_dword(dev, 17*4, socket->saved_state[1]); - pci_enable_device(dev); + + rc = pci_enable_device(dev); + if (rc) + return rc; + pci_set_master(dev); if (socket->type && socket->type->restore_state) -- cgit v0.10.2 From f901b8c46fa9748b9d6836e9b158cf7be89447f1 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Wed, 25 Oct 2006 19:56:55 -0400 Subject: [PATCH] PCMCIA: fix __must_check warnings Fix the remaining __must_check warnings in the PCMCIA core. Signed-off-by: Dominik Brodowski diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index af392bf..0f70192 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -717,6 +717,7 @@ static int pcmcia_requery(struct device *dev, void * _data) static void pcmcia_bus_rescan(struct pcmcia_socket *skt) { int no_devices=0; + int ret = 0; unsigned long flags; /* must be called with skt_mutex held */ @@ -729,7 +730,7 @@ static void pcmcia_bus_rescan(struct pcmcia_socket *skt) * missing resource information or other trouble, we need to * do this now. */ if (no_devices) { - int ret = pcmcia_card_add(skt); + ret = pcmcia_card_add(skt); if (ret) return; } @@ -741,7 +742,9 @@ static void pcmcia_bus_rescan(struct pcmcia_socket *skt) /* we re-scan all devices, not just the ones connected to this * socket. This does not matter, though. */ - bus_rescan_devices(&pcmcia_bus_type); + ret = bus_rescan_devices(&pcmcia_bus_type); + if (ret) + printk(KERN_INFO "pcmcia: bus_rescan_devices failed\n"); } static inline int pcmcia_devmatch(struct pcmcia_device *dev, @@ -1001,6 +1004,7 @@ static ssize_t pcmcia_store_allow_func_id_match(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct pcmcia_device *p_dev = to_pcmcia_dev(dev); + int ret; if (!count) return -EINVAL; @@ -1009,7 +1013,10 @@ static ssize_t pcmcia_store_allow_func_id_match(struct device *dev, p_dev->allow_func_id_match = 1; mutex_unlock(&p_dev->socket->skt_mutex); - bus_rescan_devices(&pcmcia_bus_type); + ret = bus_rescan_devices(&pcmcia_bus_type); + if (ret) + printk(KERN_INFO "pcmcia: bus_rescan_devices failed after " + "allowing func_id matches\n"); return count; } -- cgit v0.10.2 From 0430fb20a2a9f06262cb5ccc98869d54cdaa3963 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:39 +0200 Subject: [PATCH] AVR32: Minor Makefile cleanup Don't generate listing by default, remove unused LIBGCC variable and rename generated disassembly and listing files to vmlinux.{s,lst}. Also make sure that files generated during the build are actually removed with make clean. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/avr32/Makefile b/arch/avr32/Makefile index cefc95a..7b842e9 100644 --- a/arch/avr32/Makefile +++ b/arch/avr32/Makefile @@ -7,7 +7,7 @@ # Default target when executing plain make .PHONY: all -all: uImage vmlinux.elf linux.lst +all: uImage vmlinux.elf KBUILD_DEFCONFIG := atstk1002_defconfig @@ -21,9 +21,7 @@ cpuflags-$(CONFIG_CPU_AP7000) += -mcpu=ap7000 CFLAGS += $(cpuflags-y) AFLAGS += $(cpuflags-y) -CHECKFLAGS += -D__avr32__ - -LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name) +CHECKFLAGS += -D__avr32__ -D__BIG_ENDIAN head-$(CONFIG_LOADER_U_BOOT) += arch/avr32/boot/u-boot/head.o head-y += arch/avr32/kernel/head.o @@ -32,7 +30,7 @@ core-$(CONFIG_BOARD_ATSTK1000) += arch/avr32/boards/atstk1000/ core-$(CONFIG_LOADER_U_BOOT) += arch/avr32/boot/u-boot/ core-y += arch/avr32/kernel/ core-y += arch/avr32/mm/ -libs-y += arch/avr32/lib/ #$(LIBGCC) +libs-y += arch/avr32/lib/ archincdir-$(CONFIG_PLATFORM_AT32AP) := arch-at32ap @@ -48,6 +46,8 @@ endif archprepare: include/asm-avr32/.arch +CLEAN_FILES += include/asm-avr32/.arch include/asm-avr32/arch + BOOT_TARGETS := vmlinux.elf vmlinux.bin uImage uImage.srec .PHONY: $(BOOT_TARGETS) install @@ -71,14 +71,19 @@ vmlinux.elf vmlinux.bin uImage.srec uImage vmlinux.cso: vmlinux install: vmlinux $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@ -linux.s: vmlinux +vmlinux.s: vmlinux $(call if_changed,disasm) -linux.lst: vmlinux +vmlinux.lst: vmlinux $(call if_changed,listing) +CLEAN_FILES += vmlinux.s vmlinux.lst + +archclean: + $(Q)$(MAKE) $(clean)=$(boot) + define archhelp @echo '* vmlinux.elf - ELF image with load address 0' @echo ' vmlinux.cso - PathFinder CSO image' - @echo ' uImage - Create a bootable image for U-Boot' + @echo '* uImage - Create a bootable image for U-Boot' endef diff --git a/arch/avr32/boot/images/Makefile b/arch/avr32/boot/images/Makefile index ccd74eee..219720a 100644 --- a/arch/avr32/boot/images/Makefile +++ b/arch/avr32/boot/images/Makefile @@ -37,14 +37,12 @@ OBJCOPYFLAGS_vmlinux.elf := --change-section-lma .text-0x80000000 \ --change-section-lma .data-0x80000000 \ --change-section-lma .init-0x80000000 \ --change-section-lma .bss-0x80000000 \ - --change-section-lma .initrd-0x80000000 \ --change-section-lma __param-0x80000000 \ --change-section-lma __ksymtab-0x80000000 \ --change-section-lma __ksymtab_gpl-0x80000000 \ --change-section-lma __kcrctab-0x80000000 \ --change-section-lma __kcrctab_gpl-0x80000000 \ --change-section-lma __ksymtab_strings-0x80000000 \ - --change-section-lma .got-0x80000000 \ --set-start 0xa0000000 $(obj)/vmlinux.elf: vmlinux FORCE $(call if_changed,objcopy) @@ -59,4 +57,4 @@ install: $(BOOTIMAGE) sh $(srctree)/install-kernel.sh $< # Generated files to be removed upon make clean -clean-files := vmlinux* uImage uImage.srec +clean-files := vmlinux.elf vmlinux.bin vmlinux.gz uImage uImage.srec -- cgit v0.10.2 From 6ea850b5eb17f6d21ac1b3d99406d213a10c64e9 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:40 +0200 Subject: [PATCH] AVR32: Silence some compile warnings Silence a few compile warnings which are basically harmless, but easy to fix. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/avr32/kernel/kprobes.c b/arch/avr32/kernel/kprobes.c index 6caf9e8..ca41fc1 100644 --- a/arch/avr32/kernel/kprobes.c +++ b/arch/avr32/kernel/kprobes.c @@ -109,7 +109,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) void *addr = (void *)regs->pc; int ret = 0; - pr_debug("kprobe_handler: kprobe_running=%d\n", + pr_debug("kprobe_handler: kprobe_running=%p\n", kprobe_running()); /* diff --git a/arch/avr32/kernel/module.c b/arch/avr32/kernel/module.c index dfc32f2..b599eae 100644 --- a/arch/avr32/kernel/module.c +++ b/arch/avr32/kernel/module.c @@ -263,7 +263,7 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab, * value of PC. Just subtract the value of * GOT, and we're done. */ - pr_debug("GOTPC: PC=0x%lx, got_offset=0x%lx, core=0x%p\n", + pr_debug("GOTPC: PC=0x%x, got_offset=0x%lx, core=0x%p\n", relocation, module->arch.got_offset, module->module_core); relocation -= ((unsigned long)module->module_core @@ -282,7 +282,7 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab, && (relocation & 0xffff0000) != 0xffff0000) return reloc_overflow(module, "R_AVR32_GOT16S", relocation); - pr_debug("GOT reloc @ 0x%lx -> %lu\n", + pr_debug("GOT reloc @ 0x%x -> %u\n", rel->r_offset, relocation); value = *location; value = ((value & 0xffff0000) diff --git a/arch/avr32/kernel/ptrace.c b/arch/avr32/kernel/ptrace.c index 3c89e59..f2e81cd 100644 --- a/arch/avr32/kernel/ptrace.c +++ b/arch/avr32/kernel/ptrace.c @@ -157,7 +157,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) unsigned long tmp; int ret; - pr_debug("arch_ptrace(%ld, %ld, %#lx, %#lx)\n", + pr_debug("arch_ptrace(%ld, %d, %#lx, %#lx)\n", request, child->pid, addr, data); pr_debug("ptrace: Enabling monitor mode...\n"); diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c index 3e6c410..70da689 100644 --- a/arch/avr32/mm/init.c +++ b/arch/avr32/mm/init.c @@ -206,7 +206,7 @@ void __init setup_bootmem(void) if (mem_ramdisk) { #ifdef CONFIG_BLK_DEV_INITRD - initrd_start = __va(mem_ramdisk->addr); + initrd_start = (unsigned long)__va(mem_ramdisk->addr); initrd_end = initrd_start + mem_ramdisk->size; print_memory_map("RAMDISK images", mem_ramdisk); -- cgit v0.10.2 From bee8ce809fb1c877388be032b468574a1cfff9ef Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:41 +0200 Subject: [PATCH] AVR32: Don't try to iounmap P2 segment addresses While ioremap() will happily map a physical address through the P2 (uncached) segment when appropriate, iounmap() doesn't know how to handle those mappings. This patch makes iounmap() do the right thing, i.e. nothing, for such mappings. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/avr32/mm/ioremap.c b/arch/avr32/mm/ioremap.c index 8cfec65..3437c82 100644 --- a/arch/avr32/mm/ioremap.c +++ b/arch/avr32/mm/ioremap.c @@ -77,6 +77,8 @@ void __iounmap(void __iomem *addr) if ((unsigned long)addr >= P4SEG) return; + if (PXSEG(addr) == P2SEG) + return; p = remove_vm_area((void *)(PAGE_MASK & (unsigned long __force)addr)); if (unlikely(!p)) { -- cgit v0.10.2 From fa3522407f01ead1ec14bdd6b785ea08d17d500d Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:42 +0200 Subject: [PATCH] AVR32: Fix oversize immediates in atomic.h When calling e.g. atomic_sub_return with a large constant, the compiler may output an immediate that is too large for the sub instruction in the middle of the loop. Fix this by explicitly specifying the number of bits allowed in the constraint. Also stop atomic_add_return() and friends from falling back to their respective "sub" variants if the constant is too large to fit in an immediate. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/include/asm-avr32/atomic.h b/include/asm-avr32/atomic.h index e0b9c44..c40b603 100644 --- a/include/asm-avr32/atomic.h +++ b/include/asm-avr32/atomic.h @@ -41,7 +41,7 @@ static inline int atomic_sub_return(int i, atomic_t *v) " stcond %1, %0\n" " brne 1b" : "=&r"(result), "=o"(v->counter) - : "m"(v->counter), "ir"(i) + : "m"(v->counter), "rKs21"(i) : "cc"); return result; @@ -58,7 +58,7 @@ static inline int atomic_add_return(int i, atomic_t *v) { int result; - if (__builtin_constant_p(i)) + if (__builtin_constant_p(i) && (i >= -1048575) && (i <= 1048576)) result = atomic_sub_return(-i, v); else asm volatile( @@ -101,7 +101,7 @@ static inline int atomic_sub_unless(atomic_t *v, int a, int u) " mov %1, 1\n" "1:" : "=&r"(tmp), "=&r"(result), "=o"(v->counter) - : "m"(v->counter), "ir"(a), "ir"(u) + : "m"(v->counter), "rKs21"(a), "rKs21"(u) : "cc", "memory"); return result; @@ -121,7 +121,7 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u) { int tmp, result; - if (__builtin_constant_p(a)) + if (__builtin_constant_p(a) && (a >= -1048575) && (a <= 1048576)) result = atomic_sub_unless(v, -a, u); else { result = 0; -- cgit v0.10.2 From 065834ab3988fece5608088e83724891c8190a2f Mon Sep 17 00:00:00 2001 From: Ben Nizette Date: Tue, 24 Oct 2006 10:12:43 +0200 Subject: [PATCH] AVR32: add io{read,write}{8,16,32}{be,} support A number of new drivers require io{read,write}{8,16,32}{be,} family of io operations. These are provided for the AVR32 by this patch in the form of a series of macros. Access to the (memory mapped) io space through these macros is defined to be little endian only as little endian devices (such as PCI) are the main consumer of IO access. If high speed access is required, io{read,write}{16,32}be macros are supplied to perform native big endian access to this io space. Signed-off-by: Ben Nizette Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/include/asm-avr32/io.h b/include/asm-avr32/io.h index 2fc8f11..eec4750 100644 --- a/include/asm-avr32/io.h +++ b/include/asm-avr32/io.h @@ -76,6 +76,39 @@ static inline unsigned int readl(const volatile void __iomem *addr) #define readsw(p, d, l) __raw_readsw((unsigned int)p, d, l) #define readsl(p, d, l) __raw_readsl((unsigned int)p, d, l) + +/* + * io{read,write}{8,16,32} macros in both le (for PCI style consumers) and native be + */ +#ifndef ioread8 + +#define ioread8(p) ({ unsigned int __v = __raw_readb(p); __v; }) + +#define ioread16(p) ({ unsigned int __v = le16_to_cpu(__raw_readw(p)); __v; }) +#define ioread16be(p) ({ unsigned int __v = be16_to_cpu(__raw_readw(p)); __v; }) + +#define ioread32(p) ({ unsigned int __v = le32_to_cpu(__raw_readl(p)); __v; }) +#define ioread32be(p) ({ unsigned int __v = be32_to_cpu(__raw_readl(p)); __v; }) + +#define iowrite8(v,p) __raw_writeb(v, p) + +#define iowrite16(v,p) __raw_writew(cpu_to_le16(v), p) +#define iowrite16be(v,p) __raw_writew(cpu_to_be16(v), p) + +#define iowrite32(v,p) __raw_writel(cpu_to_le32(v), p) +#define iowrite32be(v,p) __raw_writel(cpu_to_be32(v), p) + +#define ioread8_rep(p,d,c) __raw_readsb(p,d,c) +#define ioread16_rep(p,d,c) __raw_readsw(p,d,c) +#define ioread32_rep(p,d,c) __raw_readsl(p,d,c) + +#define iowrite8_rep(p,s,c) __raw_writesb(p,s,c) +#define iowrite16_rep(p,s,c) __raw_writesw(p,s,c) +#define iowrite32_rep(p,s,c) __raw_writesl(p,s,c) + +#endif + + /* * These two are only here because ALSA _thinks_ it needs them... */ -- cgit v0.10.2 From d68041cc9fde550fe6b6a6de1d7a110daff3cb60 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:44 +0200 Subject: [PATCH] AVR32: Implement and export __raw_{read,write}s[bwl] Implement __raw_readsb and __raw_writesb. Export __raw_reads[bwl] and __raw_writes[bwl] for use by modules. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c index 04f767a..372e3f8 100644 --- a/arch/avr32/kernel/avr32_ksyms.c +++ b/arch/avr32/kernel/avr32_ksyms.c @@ -7,6 +7,7 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ +#include #include #include @@ -53,3 +54,11 @@ EXPORT_SYMBOL(find_next_zero_bit); EXPORT_SYMBOL(find_first_bit); EXPORT_SYMBOL(find_next_bit); EXPORT_SYMBOL(generic_find_next_zero_le_bit); + +/* I/O primitives (lib/io-*.S) */ +EXPORT_SYMBOL(__raw_readsb); +EXPORT_SYMBOL(__raw_readsw); +EXPORT_SYMBOL(__raw_readsl); +EXPORT_SYMBOL(__raw_writesb); +EXPORT_SYMBOL(__raw_writesw); +EXPORT_SYMBOL(__raw_writesl); diff --git a/arch/avr32/lib/Makefile b/arch/avr32/lib/Makefile index 09ac43e..084d95b 100644 --- a/arch/avr32/lib/Makefile +++ b/arch/avr32/lib/Makefile @@ -7,4 +7,5 @@ lib-y += strncpy_from_user.o strnlen_user.o lib-y += delay.o memset.o memcpy.o findbit.o lib-y += csum_partial.o csum_partial_copy_generic.o lib-y += io-readsw.o io-readsl.o io-writesw.o io-writesl.o +lib-y += io-readsb.o io-writesb.o lib-y += __avr32_lsl64.o __avr32_lsr64.o __avr32_asr64.o diff --git a/arch/avr32/lib/io-readsb.S b/arch/avr32/lib/io-readsb.S new file mode 100644 index 0000000..b319d5e --- /dev/null +++ b/arch/avr32/lib/io-readsb.S @@ -0,0 +1,47 @@ +/* + * Copyright (C) 2004-2006 Atmel Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + + .text +.Lnot_word_aligned: +1: ld.ub r8, r12[0] + sub r10, 1 + st.b r11++, r8 + reteq r12 + tst r11, r9 + brne 1b + + /* fall through */ + + .global __raw_readsb + .type __raw_readsb,@function +__raw_readsb: + cp.w r10, 0 + mov r9, 3 + reteq r12 + + tst r11, r9 + brne .Lnot_word_aligned + + sub r10, 4 + brlt 2f + +1: ldins.b r8:t, r12[0] + ldins.b r8:u, r12[0] + ldins.b r8:l, r12[0] + ldins.b r8:b, r12[0] + st.w r11++, r8 + sub r10, 4 + brge 1b + +2: sub r10, -4 + reteq r12 + +3: ld.uh r8, r12[0] + sub r10, 1 + st.b r11++, r8 + brne 3b diff --git a/arch/avr32/lib/io-writesb.S b/arch/avr32/lib/io-writesb.S new file mode 100644 index 0000000..b4ebaac --- /dev/null +++ b/arch/avr32/lib/io-writesb.S @@ -0,0 +1,52 @@ +/* + * Copyright (C) 2004-2006 Atmel Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + + .text +.Lnot_word_aligned: +1: ld.ub r8, r11++ + sub r10, 1 + st.b r12[0], r8 + reteq r12 + tst r11, r9 + brne 1b + + /* fall through */ + + .global __raw_writesb + .type __raw_writesb,@function +__raw_writesb: + cp.w r10, 0 + mov r9, 3 + reteq r12 + + tst r11, r9 + brne .Lnot_word_aligned + + sub r10, 4 + brlt 2f + +1: ld.w r8, r11++ + bfextu r9, r8, 24, 8 + st.b r12[0], r9 + bfextu r9, r8, 16, 8 + st.b r12[0], r9 + bfextu r9, r8, 8, 8 + st.b r12[0], r9 + st.b r12[0], r8 + sub r10, 4 + brge 1b + +2: sub r10, -4 + reteq r12 + +3: ld.ub r8, r11++ + sub r10, 1 + st.b r12[0], r8 + brne 3b + + retal r12 -- cgit v0.10.2 From 452976b29864f9adcdd6d4ea81001aa5483592c2 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:45 +0200 Subject: [PATCH] AVR32: Use __raw MMIO access for internal peripherals The read[bwl] and write[bwl] functions are meant for accessing PCI devices. How this is achieved on AVR32 is unknown, as there are no systems with a PCI bridge available yet. On-chip peripheral access, however, should not depend on how we end up implementing PCI access, so using __raw_read[bwl]/__raw_write[bwl] is the right thing to do for on-chip peripherals. This patch converts the drivers for the static memory controller, interrupt controller, PIO controller and system manager to use __raw MMIO access. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/avr32/mach-at32ap/hsmc.h b/arch/avr32/mach-at32ap/hsmc.h index 5681276..d1d48e2 100644 --- a/arch/avr32/mach-at32ap/hsmc.h +++ b/arch/avr32/mach-at32ap/hsmc.h @@ -120,8 +120,8 @@ /* Register access macros */ #define hsmc_readl(port,reg) \ - readl((port)->regs + HSMC_##reg) + __raw_readl((port)->regs + HSMC_##reg) #define hsmc_writel(port,reg,value) \ - writel((value), (port)->regs + HSMC_##reg) + __raw_writel((value), (port)->regs + HSMC_##reg) #endif /* __ASM_AVR32_HSMC_H__ */ diff --git a/arch/avr32/mach-at32ap/intc.h b/arch/avr32/mach-at32ap/intc.h index d289ca2..4d3664e 100644 --- a/arch/avr32/mach-at32ap/intc.h +++ b/arch/avr32/mach-at32ap/intc.h @@ -321,7 +321,9 @@ #define INTC_MKBF(name, value) (((value) & ((1 << INTC_##name##_SIZE) - 1)) << INTC_##name##_OFFSET) #define INTC_GETBF(name, value) (((value) >> INTC_##name##_OFFSET) & ((1 << INTC_##name##_SIZE) - 1)) -#define intc_readl(port,reg) readl((port)->regs + INTC_##reg) -#define intc_writel(port,reg,value) writel((value), (port)->regs + INTC_##reg) +#define intc_readl(port,reg) \ + __raw_readl((port)->regs + INTC_##reg) +#define intc_writel(port,reg,value) \ + __raw_writel((value), (port)->regs + INTC_##reg) #endif /* __ASM_AVR32_PERIHP_INTC_H__ */ diff --git a/arch/avr32/mach-at32ap/pio.h b/arch/avr32/mach-at32ap/pio.h index cfea123..50fa3ac 100644 --- a/arch/avr32/mach-at32ap/pio.h +++ b/arch/avr32/mach-at32ap/pio.h @@ -170,8 +170,10 @@ #define PIO_BFINS(name,value,old) (((old) & ~(((1 << PIO_##name##_SIZE) - 1) << PIO_##name##_OFFSET)) | PIO_BF(name,value)) /* Register access macros */ -#define pio_readl(port,reg) readl((port)->regs + PIO_##reg) -#define pio_writel(port,reg,value) writel((value), (port)->regs + PIO_##reg) +#define pio_readl(port,reg) \ + __raw_readl((port)->regs + PIO_##reg) +#define pio_writel(port,reg,value) \ + __raw_writel((value), (port)->regs + PIO_##reg) void at32_init_pio(struct platform_device *pdev); diff --git a/arch/avr32/mach-at32ap/sm.h b/arch/avr32/mach-at32ap/sm.h index 2756582..cad02b5 100644 --- a/arch/avr32/mach-at32ap/sm.h +++ b/arch/avr32/mach-at32ap/sm.h @@ -234,7 +234,9 @@ #define SM_BFINS(name,value,old) (((old) & ~(((1 << SM_##name##_SIZE) - 1) << SM_##name##_OFFSET)) | SM_BF(name,value)) /* Register access macros */ -#define sm_readl(port,reg) readl((port)->regs + SM_##reg) -#define sm_writel(port,reg,value) writel((value), (port)->regs + SM_##reg) +#define sm_readl(port,reg) \ + __raw_readl((port)->regs + SM_##reg) +#define sm_writel(port,reg,value) \ + __raw_writel((value), (port)->regs + SM_##reg) #endif /* __ASM_AVR32_SM_H__ */ -- cgit v0.10.2 From 291b58d663862c3d42d2e8092f8b0dd3f15a94f8 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 24 Oct 2006 10:12:46 +0200 Subject: [PATCH] AVR32: Update defconfig Sync atstk1002_defconfig with latest git, turn off non-existent drivers and enable a few more userspace-visible options like SysV IPC and inotify support. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/avr32/configs/atstk1002_defconfig b/arch/avr32/configs/atstk1002_defconfig index 6c2c5e0..ae92a14 100644 --- a/arch/avr32/configs/atstk1002_defconfig +++ b/arch/avr32/configs/atstk1002_defconfig @@ -1,13 +1,14 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18-rc1 -# Tue Jul 11 12:41:36 2006 +# Linux kernel version: 2.6.19-rc2 +# Fri Oct 20 11:52:37 2006 # CONFIG_AVR32=y CONFIG_GENERIC_HARDIRQS=y CONFIG_HARDIRQS_SW_RESEND=y CONFIG_GENERIC_IRQ_PROBE=y CONFIG_RWSEM_GENERIC_SPINLOCK=y +CONFIG_GENERIC_TIME=y CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" @@ -25,16 +26,23 @@ CONFIG_INIT_ENV_ARG_LIMIT=32 CONFIG_LOCALVERSION="" # CONFIG_LOCALVERSION_AUTO is not set CONFIG_SWAP=y -# CONFIG_SYSVIPC is not set -# CONFIG_POSIX_MQUEUE is not set -# CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y -# CONFIG_AUDIT is not set +CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set +CONFIG_POSIX_MQUEUE=y +CONFIG_BSD_PROCESS_ACCT=y +CONFIG_BSD_PROCESS_ACCT_V3=y +CONFIG_TASKSTATS=y +CONFIG_TASK_DELAY_ACCT=y +# CONFIG_UTS_NS is not set +CONFIG_AUDIT=y # CONFIG_IKCONFIG is not set -# CONFIG_RELAY is not set +CONFIG_RELAY=y CONFIG_INITRAMFS_SOURCE="" CONFIG_CC_OPTIMIZE_FOR_SIZE=y +# CONFIG_TASK_XACCT is not set +CONFIG_SYSCTL=y CONFIG_EMBEDDED=y +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -43,14 +51,15 @@ CONFIG_PRINTK=y CONFIG_BUG=y CONFIG_ELF_CORE=y # CONFIG_BASE_FULL is not set -# CONFIG_FUTEX is not set -# CONFIG_EPOLL is not set +CONFIG_FUTEX=y +CONFIG_EPOLL=y CONFIG_SHMEM=y -# CONFIG_SLAB is not set -# CONFIG_VM_EVENT_COUNTERS is not set +CONFIG_SLAB=y +CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=1 -CONFIG_SLOB=y +# CONFIG_SLOB is not set # # Loadable module support @@ -65,6 +74,7 @@ CONFIG_MODULE_UNLOAD=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_BLK_DEV_IO_TRACE is not set # @@ -166,10 +176,12 @@ CONFIG_IP_PNP_DHCP=y # CONFIG_INET_TUNNEL is not set # CONFIG_INET_XFRM_MODE_TRANSPORT is not set # CONFIG_INET_XFRM_MODE_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_BEET is not set CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set @@ -199,7 +211,6 @@ CONFIG_TCP_CONG_BIC=y # CONFIG_ATALK is not set # CONFIG_X25 is not set # CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set @@ -212,7 +223,6 @@ CONFIG_TCP_CONG_BIC=y # Network testing # # CONFIG_NET_PKTGEN is not set -# CONFIG_NET_TCPPROBE is not set # CONFIG_HAMRADIO is not set # CONFIG_IRDA is not set # CONFIG_BT is not set @@ -239,7 +249,84 @@ CONFIG_STANDALONE=y # # Memory Technology Devices (MTD) # -# CONFIG_MTD is not set +CONFIG_MTD=y +# CONFIG_MTD_DEBUG is not set +# CONFIG_MTD_CONCAT is not set +CONFIG_MTD_PARTITIONS=y +# CONFIG_MTD_REDBOOT_PARTS is not set +CONFIG_MTD_CMDLINE_PARTS=y + +# +# User Modules And Translation Layers +# +CONFIG_MTD_CHAR=y +CONFIG_MTD_BLOCK=y +# CONFIG_FTL is not set +# CONFIG_NFTL is not set +# CONFIG_INFTL is not set +# CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set + +# +# RAM/ROM/Flash chip drivers +# +CONFIG_MTD_CFI=y +# CONFIG_MTD_JEDECPROBE is not set +CONFIG_MTD_GEN_PROBE=y +# CONFIG_MTD_CFI_ADV_OPTIONS is not set +CONFIG_MTD_MAP_BANK_WIDTH_1=y +CONFIG_MTD_MAP_BANK_WIDTH_2=y +CONFIG_MTD_MAP_BANK_WIDTH_4=y +# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set +CONFIG_MTD_CFI_I1=y +CONFIG_MTD_CFI_I2=y +# CONFIG_MTD_CFI_I4 is not set +# CONFIG_MTD_CFI_I8 is not set +# CONFIG_MTD_CFI_INTELEXT is not set +CONFIG_MTD_CFI_AMDSTD=y +# CONFIG_MTD_CFI_STAA is not set +CONFIG_MTD_CFI_UTIL=y +# CONFIG_MTD_RAM is not set +# CONFIG_MTD_ROM is not set +# CONFIG_MTD_ABSENT is not set +# CONFIG_MTD_OBSOLETE_CHIPS is not set + +# +# Mapping drivers for chip access +# +# CONFIG_MTD_COMPLEX_MAPPINGS is not set +CONFIG_MTD_PHYSMAP=y +CONFIG_MTD_PHYSMAP_START=0x8000000 +CONFIG_MTD_PHYSMAP_LEN=0x0 +CONFIG_MTD_PHYSMAP_BANKWIDTH=2 +# CONFIG_MTD_PLATRAM is not set + +# +# Self-contained MTD device drivers +# +# CONFIG_MTD_SLRAM is not set +# CONFIG_MTD_PHRAM is not set +# CONFIG_MTD_MTDRAM is not set +# CONFIG_MTD_BLOCK2MTD is not set + +# +# Disk-On-Chip Device Drivers +# +# CONFIG_MTD_DOC2000 is not set +# CONFIG_MTD_DOC2001 is not set +# CONFIG_MTD_DOC2001PLUS is not set + +# +# NAND Flash Device Drivers +# +# CONFIG_MTD_NAND is not set + +# +# OneNAND Flash Device Drivers +# +# CONFIG_MTD_ONENAND is not set # # Parallel port support @@ -260,11 +347,18 @@ CONFIG_BLK_DEV_NBD=m CONFIG_BLK_DEV_RAM=m CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set # +# Misc devices +# +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # # CONFIG_IDE is not set @@ -274,6 +368,12 @@ CONFIG_BLK_DEV_INITRD=y # # CONFIG_RAID_ATTRS is not set # CONFIG_SCSI is not set +# CONFIG_SCSI_NETLINK is not set + +# +# Serial ATA (prod) and Parallel ATA (experimental) drivers +# +# CONFIG_ATA is not set # # Multi-device support (RAID and LVM) @@ -305,14 +405,11 @@ CONFIG_TUN=m # # PHY device support # -# CONFIG_PHYLIB is not set # # Ethernet (10 or 100Mbit) # -CONFIG_NET_ETHERNET=y -CONFIG_MII=y -CONFIG_MACB=y +# CONFIG_NET_ETHERNET is not set # # Ethernet (1000 Mbit) @@ -341,10 +438,11 @@ CONFIG_PPP=m CONFIG_PPP_ASYNC=m # CONFIG_PPP_SYNC_TTY is not set CONFIG_PPP_DEFLATE=m -# CONFIG_PPP_BSDCOMP is not set +CONFIG_PPP_BSDCOMP=m # CONFIG_PPP_MPPE is not set # CONFIG_PPPOE is not set # CONFIG_SLIP is not set +CONFIG_SLHC=m # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set # CONFIG_NETPOLL is not set @@ -417,7 +515,6 @@ CONFIG_UNIX98_PTYS=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -427,23 +524,13 @@ CONFIG_UNIX98_PTYS=y # # SPI support # -CONFIG_SPI=y -# CONFIG_SPI_DEBUG is not set -CONFIG_SPI_MASTER=y - -# -# SPI Master Controller Drivers -# -CONFIG_SPI_ATMEL=m -# CONFIG_SPI_BITBANG is not set - -# -# SPI Protocol Masters -# +# CONFIG_SPI is not set +# CONFIG_SPI_MASTER is not set # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -452,14 +539,9 @@ CONFIG_SPI_ATMEL=m # CONFIG_HWMON_VID is not set # -# Misc devices -# - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -470,28 +552,8 @@ CONFIG_VIDEO_V4L2=y # Graphics support # # CONFIG_FIRMWARE_EDID is not set -CONFIG_FB=m -CONFIG_FB_CFB_FILLRECT=m -CONFIG_FB_CFB_COPYAREA=m -CONFIG_FB_CFB_IMAGEBLIT=m -# CONFIG_FB_MACMODES is not set -# CONFIG_FB_BACKLIGHT is not set -# CONFIG_FB_MODE_HELPERS is not set -# CONFIG_FB_TILEBLITTING is not set -CONFIG_FB_SIDSA=m -CONFIG_FB_SIDSA_DEFAULT_BPP=24 -# CONFIG_FB_S1D13XXX is not set -# CONFIG_FB_VIRTUAL is not set - -# -# Logo configuration -# -# CONFIG_LOGO is not set -CONFIG_BACKLIGHT_LCD_SUPPORT=y -# CONFIG_BACKLIGHT_CLASS_DEVICE is not set -CONFIG_LCD_CLASS_DEVICE=m -CONFIG_LCD_DEVICE=y -CONFIG_LCD_LTV350QV=m +# CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set # # Sound @@ -561,18 +623,21 @@ CONFIG_LCD_LTV350QV=m # # File systems # -CONFIG_EXT2_FS=y +CONFIG_EXT2_FS=m # CONFIG_EXT2_FS_XATTR is not set # CONFIG_EXT2_FS_XIP is not set # CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set # CONFIG_REISERFS_FS is not set # CONFIG_JFS_FS is not set # CONFIG_FS_POSIX_ACL is not set # CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set CONFIG_MINIX_FS=m -CONFIG_ROMFS_FS=m -# CONFIG_INOTIFY is not set +# CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y +CONFIG_INOTIFY_USER=y # CONFIG_QUOTA is not set # CONFIG_DNOTIFY is not set # CONFIG_AUTOFS_FS is not set @@ -600,8 +665,10 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" # CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y CONFIG_CONFIGFS_FS=m @@ -616,6 +683,16 @@ CONFIG_CONFIGFS_FS=m # CONFIG_BEFS_FS is not set # CONFIG_BFS_FS is not set # CONFIG_EFS_FS is not set +# CONFIG_JFFS_FS is not set +CONFIG_JFFS2_FS=y +CONFIG_JFFS2_FS_DEBUG=0 +CONFIG_JFFS2_FS_WRITEBUFFER=y +# CONFIG_JFFS2_SUMMARY is not set +# CONFIG_JFFS2_FS_XATTR is not set +# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set +CONFIG_JFFS2_ZLIB=y +CONFIG_JFFS2_RTIME=y +# CONFIG_JFFS2_RUBIN is not set # CONFIG_CRAMFS is not set # CONFIG_VXFS_FS is not set # CONFIG_HPFS_FS is not set @@ -626,26 +703,10 @@ CONFIG_CONFIGFS_FS=m # # Network File Systems # -CONFIG_NFS_FS=y -CONFIG_NFS_V3=y -# CONFIG_NFS_V3_ACL is not set -# CONFIG_NFS_V4 is not set -# CONFIG_NFS_DIRECTIO is not set +# CONFIG_NFS_FS is not set # CONFIG_NFSD is not set -CONFIG_ROOT_NFS=y -CONFIG_LOCKD=y -CONFIG_LOCKD_V4=y -CONFIG_NFS_COMMON=y -CONFIG_SUNRPC=y -# CONFIG_RPCSEC_GSS_KRB5 is not set -# CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set -CONFIG_CIFS=m -# CONFIG_CIFS_STATS is not set -# CONFIG_CIFS_WEAK_PW_HASH is not set -# CONFIG_CIFS_XATTR is not set -# CONFIG_CIFS_DEBUG2 is not set -# CONFIG_CIFS_EXPERIMENTAL is not set +# CONFIG_CIFS is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -665,7 +726,7 @@ CONFIG_NLS_DEFAULT="iso8859-1" CONFIG_NLS_CODEPAGE_437=m # CONFIG_NLS_CODEPAGE_737 is not set # CONFIG_NLS_CODEPAGE_775 is not set -CONFIG_NLS_CODEPAGE_850=m +# CONFIG_NLS_CODEPAGE_850 is not set # CONFIG_NLS_CODEPAGE_852 is not set # CONFIG_NLS_CODEPAGE_855 is not set # CONFIG_NLS_CODEPAGE_857 is not set @@ -705,13 +766,17 @@ CONFIG_NLS_UTF8=m # Kernel hacking # CONFIG_TRACE_IRQFLAGS_SUPPORT=y -CONFIG_PRINTK_TIME=y +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=14 CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_SCHEDSTATS is not set +# CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_RT_MUTEXES is not set +# CONFIG_RT_MUTEX_TESTER is not set # CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_MUTEXES is not set # CONFIG_DEBUG_RWSEMS is not set @@ -722,11 +787,13 @@ CONFIG_DEBUG_BUGVERBOSE=y # CONFIG_DEBUG_INFO is not set CONFIG_DEBUG_FS=y # CONFIG_DEBUG_VM is not set +# CONFIG_DEBUG_LIST is not set CONFIG_FRAME_POINTER=y # CONFIG_UNWIND_INFO is not set CONFIG_FORCED_INLINING=y +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set -CONFIG_KPROBES=y +# CONFIG_KPROBES is not set # # Security options @@ -740,15 +807,13 @@ CONFIG_KPROBES=y # CONFIG_CRYPTO is not set # -# Hardware crypto devices -# - -# # Library routines # CONFIG_CRC_CCITT=m # CONFIG_CRC16 is not set -CONFIG_CRC32=m +CONFIG_CRC32=y # CONFIG_LIBCRC32C is not set -CONFIG_ZLIB_INFLATE=m -CONFIG_ZLIB_DEFLATE=m +CONFIG_AUDIT_GENERIC=y +CONFIG_ZLIB_INFLATE=y +CONFIG_ZLIB_DEFLATE=y +CONFIG_PLIST=y -- cgit v0.10.2 From 4130a4b206e7c628482aa12ec30949382c8cdc5e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 25 Oct 2006 22:31:06 -0700 Subject: [SPARC64]: Fix central/FHC bus handling on Ex000 systems. 1) probe_other_fhcs() wants to see only non-central FHC busses, so skip FHCs that don't sit off the root 2) Like SBUS, FHC can lack the appropriate address and size cell count properties, so add an of_busses[] entry and handlers for that. 3) Central FHC irq translator probing was buggy. We were trying to use dp->child in irq_trans_init but that linkage is not setup at this point. So instead, pass in the parent of "dp" and look for the child "fhc" with parent "central". Thanks to the tireless assistence of Ben Collins in tracking down these problems and testing out these fixes. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/central.c b/arch/sparc64/kernel/central.c index b66336d..e724c54 100644 --- a/arch/sparc64/kernel/central.c +++ b/arch/sparc64/kernel/central.c @@ -126,6 +126,10 @@ static void probe_other_fhcs(void) int board; u32 tmp; + if (dp->parent && + dp->parent->parent != NULL) + continue; + fhc = (struct linux_fhc *) central_alloc_bootmem(sizeof(struct linux_fhc)); if (fhc == NULL) diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c index d3dfb2a..983ca5f 100644 --- a/arch/sparc64/kernel/of_device.c +++ b/arch/sparc64/kernel/of_device.c @@ -402,16 +402,22 @@ static void of_bus_sbus_count_cells(struct device_node *child, *sizec = 1; } -static int of_bus_sbus_map(u32 *addr, const u32 *range, int na, int ns, int pna) -{ - return of_bus_default_map(addr, range, na, ns, pna); -} - -static unsigned int of_bus_sbus_get_flags(u32 *addr) +/* + * FHC/Central bus specific translator. + * + * This is just needed to hard-code the address and size cell + * counts. 'fhc' and 'central' nodes lack the #address-cells and + * #size-cells properties, and if you walk to the root on such + * Enterprise boxes all you'll get is a #size-cells of 2 which is + * not what we want to use. + */ +static int of_bus_fhc_match(struct device_node *np) { - return IORESOURCE_MEM; + return !strcmp(np->name, "fhc") || + !strcmp(np->name, "central"); } +#define of_bus_fhc_count_cells of_bus_sbus_count_cells /* * Array of bus specific translators @@ -433,8 +439,17 @@ static struct of_bus of_busses[] = { .addr_prop_name = "reg", .match = of_bus_sbus_match, .count_cells = of_bus_sbus_count_cells, - .map = of_bus_sbus_map, - .get_flags = of_bus_sbus_get_flags, + .map = of_bus_default_map, + .get_flags = of_bus_default_get_flags, + }, + /* FHC */ + { + .name = "fhc", + .addr_prop_name = "reg", + .match = of_bus_fhc_match, + .count_cells = of_bus_fhc_count_cells, + .map = of_bus_default_map, + .get_flags = of_bus_default_get_flags, }, /* Default */ { diff --git a/arch/sparc64/kernel/prom.c b/arch/sparc64/kernel/prom.c index e21cd6a..c60efb3 100644 --- a/arch/sparc64/kernel/prom.c +++ b/arch/sparc64/kernel/prom.c @@ -1079,23 +1079,22 @@ static void sun4v_vdev_irq_trans_init(struct device_node *dp) static void irq_trans_init(struct device_node *dp) { - const char *model; #ifdef CONFIG_PCI + const char *model; int i; #endif +#ifdef CONFIG_PCI model = of_get_property(dp, "model", NULL); if (!model) model = of_get_property(dp, "compatible", NULL); - if (!model) - return; - -#ifdef CONFIG_PCI - for (i = 0; i < ARRAY_SIZE(pci_irq_trans_table); i++) { - struct irq_trans *t = &pci_irq_trans_table[i]; + if (model) { + for (i = 0; i < ARRAY_SIZE(pci_irq_trans_table); i++) { + struct irq_trans *t = &pci_irq_trans_table[i]; - if (!strcmp(model, t->name)) - return t->init(dp); + if (!strcmp(model, t->name)) + return t->init(dp); + } } #endif #ifdef CONFIG_SBUS @@ -1103,8 +1102,9 @@ static void irq_trans_init(struct device_node *dp) !strcmp(dp->name, "sbi")) return sbus_irq_trans_init(dp); #endif - if (!strcmp(dp->name, "central")) - return central_irq_trans_init(dp->child); + if (!strcmp(dp->name, "fhc") && + !strcmp(dp->parent->name, "central")) + return central_irq_trans_init(dp); if (!strcmp(dp->name, "virtual-devices")) return sun4v_vdev_irq_trans_init(dp); } @@ -1516,7 +1516,7 @@ static char * __init get_one_property(phandle node, const char *name) return buf; } -static struct device_node * __init create_node(phandle node) +static struct device_node * __init create_node(phandle node, struct device_node *parent) { struct device_node *dp; @@ -1525,6 +1525,7 @@ static struct device_node * __init create_node(phandle node) dp = prom_early_alloc(sizeof(*dp)); dp->unique_id = unique_id++; + dp->parent = parent; kref_init(&dp->kref); @@ -1543,12 +1544,11 @@ static struct device_node * __init build_tree(struct device_node *parent, phandl { struct device_node *dp; - dp = create_node(node); + dp = create_node(node, parent); if (dp) { *(*nextp) = dp; *nextp = &dp->allnext; - dp->parent = parent; dp->path_component_name = build_path_component(dp); dp->full_name = build_full_name(dp); @@ -1564,7 +1564,7 @@ void __init prom_build_devicetree(void) { struct device_node **nextp; - allnodes = create_node(prom_root_node); + allnodes = create_node(prom_root_node, NULL); allnodes->path_component_name = ""; allnodes->full_name = "/"; -- cgit v0.10.2 From 012d64ff68f304df1c35ce5902f5023dc14b643f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 25 Oct 2006 22:33:07 -0700 Subject: [SPARC64]: Fix memory corruption in pci_4u_free_consistent(). The second argument to free_npages() was being incorrectly calculated, which would thus access far past the end of the arena->map[] bitmap. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/pci_iommu.c b/arch/sparc64/kernel/pci_iommu.c index 82e5455..2e7f142 100644 --- a/arch/sparc64/kernel/pci_iommu.c +++ b/arch/sparc64/kernel/pci_iommu.c @@ -281,7 +281,7 @@ static void pci_4u_free_consistent(struct pci_dev *pdev, size_t size, void *cpu, spin_lock_irqsave(&iommu->lock, flags); - free_npages(iommu, dvma, npages); + free_npages(iommu, dvma - iommu->page_table_map_base, npages); spin_unlock_irqrestore(&iommu->lock, flags); -- cgit v0.10.2 From 22119240b1c8f64eebb6ffb368c927b2f8be4136 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 25 Oct 2006 23:04:12 -0700 Subject: [TCP] cubic: scaling error Doug Leith observed a discrepancy between the version of CUBIC described in the papers and the version in 2.6.18. A math error related to scaling causes Cubic to grow too slowly. Patch is from "Sangtae Ha" . I validated that it does fix the problems. See the following to show behavior over 500ms 100 Mbit link. Sender (2.6.19-rc3) --- Bridge (2.6.18-rt7) ------- Receiver (2.6.19-rc3) 1G [netem] 100M http://developer.osdl.org/shemminger/tcp/2.6.19-rc3/cubic-orig.png http://developer.osdl.org/shemminger/tcp/2.6.19-rc3/cubic-fix.png Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c index a60ef38..6ad1848 100644 --- a/net/ipv4/tcp_cubic.c +++ b/net/ipv4/tcp_cubic.c @@ -190,7 +190,7 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd) */ /* change the unit from HZ to bictcp_HZ */ - t = ((tcp_time_stamp + ca->delay_min - ca->epoch_start) + t = ((tcp_time_stamp + (ca->delay_min>>3) - ca->epoch_start) << BICTCP_HZ) / HZ; if (t < ca->bic_K) /* t - K */ @@ -259,7 +259,7 @@ static inline void measure_delay(struct sock *sk) (s32)(tcp_time_stamp - ca->epoch_start) < HZ) return; - delay = tcp_time_stamp - tp->rx_opt.rcv_tsecr; + delay = (tcp_time_stamp - tp->rx_opt.rcv_tsecr)<<3; if (delay == 0) delay = 1; @@ -366,7 +366,7 @@ static int __init cubictcp_register(void) beta_scale = 8*(BICTCP_BETA_SCALE+beta)/ 3 / (BICTCP_BETA_SCALE - beta); - cube_rtt_scale = (bic_scale << 3) / 10; /* 1024*c/rtt */ + cube_rtt_scale = (bic_scale * 10); /* 1024*c/rtt */ /* calculate the "K" for (wmax-cwnd) = c/rtt * K^3 * so K = cubic_root( (wmax-cwnd)*rtt/c ) -- cgit v0.10.2 From 2a272f98619f188efe22119b0415aac6bc34a13f Mon Sep 17 00:00:00 2001 From: Gavin McCullagh Date: Wed, 25 Oct 2006 23:05:52 -0700 Subject: [TCP] H-TCP: fix integer overflow When using H-TCP with a single flow on a 500Mbit connection (or less actually), alpha can exceed 65000, so alpha needs to be a u32. Signed-off-by: Gavin McCullagh Signed-off-by: Doug Leith Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c index 682e7d5..283be3c 100644 --- a/net/ipv4/tcp_htcp.c +++ b/net/ipv4/tcp_htcp.c @@ -23,7 +23,7 @@ module_param(use_bandwidth_switch, int, 0644); MODULE_PARM_DESC(use_bandwidth_switch, "turn on/off bandwidth switcher"); struct htcp { - u16 alpha; /* Fixed point arith, << 7 */ + u32 alpha; /* Fixed point arith, << 7 */ u8 beta; /* Fixed point arith, << 7 */ u8 modeswitch; /* Delay modeswitch until we had at least one congestion event */ u32 last_cong; /* Time since last congestion event end */ -- cgit v0.10.2 From 1842c4bef61f985fbec6df7150041b85d8b52b1a Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 25 Oct 2006 23:07:37 -0700 Subject: [BRIDGE]: correct print message typo Correct message typo/spello. Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller diff --git a/net/bridge/br_sysfs_br.c b/net/bridge/br_sysfs_br.c index 96bcb2f..de9d1a9 100644 --- a/net/bridge/br_sysfs_br.c +++ b/net/bridge/br_sysfs_br.c @@ -376,7 +376,7 @@ int br_sysfs_addbr(struct net_device *dev) err = sysfs_create_bin_file(brobj, &bridge_forward); if (err) { - pr_info("%s: can't create attribue file %s/%s\n", + pr_info("%s: can't create attribute file %s/%s\n", __FUNCTION__, dev->name, bridge_forward.attr.name); goto out2; } -- cgit v0.10.2 From f2454a1a4b2aca38d3b7887619f43291d773c1ee Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Wed, 25 Oct 2006 05:49:53 -0400 Subject: [IA64] don't double >> PAGE_SHIFT pointer for /dev/kmem access Don't PAGE_SHIFT pointer before handing it to virt_to_page() in xlate_dev_kmem_ptr() as it results in a double shift. Spotted by Bob Montgomery. Signed-off-by: Jes Sorensen Signed-off-by: Tony Luck diff --git a/include/asm-ia64/uaccess.h b/include/asm-ia64/uaccess.h index 9adb512..449c8c0 100644 --- a/include/asm-ia64/uaccess.h +++ b/include/asm-ia64/uaccess.h @@ -389,7 +389,7 @@ xlate_dev_kmem_ptr (char * p) struct page *page; char * ptr; - page = virt_to_page((unsigned long)p >> PAGE_SHIFT); + page = virt_to_page((unsigned long)p); if (PageUncached(page)) ptr = (char *)__pa(p) + __IA64_UNCACHED_OFFSET; else -- cgit v0.10.2 From f5ef9d11fd255b30b455d18f8d721bc44cd1296b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 27 Oct 2006 01:03:31 -0700 Subject: [SPARC]: Fix bus_id[] string overflow. dp->path_component_name can be larger than ->bus_id[] so use a different naming scheme for this stuff. Noticed by Jurij Smakov. Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/ebus.c b/arch/sparc/kernel/ebus.c index 75ac24d..ba58c3a 100644 --- a/arch/sparc/kernel/ebus.c +++ b/arch/sparc/kernel/ebus.c @@ -237,12 +237,12 @@ void __init fill_ebus_device(struct device_node *dp, struct linux_ebus_device *d dev->ofdev.node = dp; dev->ofdev.dev.parent = &dev->bus->ofdev.dev; dev->ofdev.dev.bus = &ebus_bus_type; - strcpy(dev->ofdev.dev.bus_id, dp->path_component_name); + sprintf(dev->ofdev.dev.bus_id, "ebus[%08x]", dp->node); /* Register with core */ if (of_device_register(&dev->ofdev) != 0) printk(KERN_DEBUG "ebus: device registration error for %s!\n", - dev->ofdev.dev.bus_id); + dp->path_component_name); if ((dp = dp->child) != NULL) { dev->children = (struct linux_ebus_child *) @@ -332,12 +332,12 @@ void __init ebus_init(void) ebus->ofdev.node = dp; ebus->ofdev.dev.parent = &pdev->dev; ebus->ofdev.dev.bus = &ebus_bus_type; - strcpy(ebus->ofdev.dev.bus_id, dp->path_component_name); + sprintf(ebus->ofdev.dev.bus_id, "ebus%d", num_ebus); /* Register with core */ if (of_device_register(&ebus->ofdev) != 0) printk(KERN_DEBUG "ebus: device registration error for %s!\n", - ebus->ofdev.dev.bus_id); + dp->path_component_name); nd = dp->child; diff --git a/arch/sparc/kernel/of_device.c b/arch/sparc/kernel/of_device.c index 74bef2a..46200c4 100644 --- a/arch/sparc/kernel/of_device.c +++ b/arch/sparc/kernel/of_device.c @@ -651,7 +651,7 @@ build_resources: if (!parent) strcpy(op->dev.bus_id, "root"); else - strcpy(op->dev.bus_id, dp->path_component_name); + sprintf(op->dev.bus_id, "%08x", dp->node); if (of_device_register(op)) { printk("%s: Could not register of device.\n", diff --git a/arch/sparc64/kernel/ebus.c b/arch/sparc64/kernel/ebus.c index 2df25c2..35bf895 100644 --- a/arch/sparc64/kernel/ebus.c +++ b/arch/sparc64/kernel/ebus.c @@ -389,12 +389,12 @@ static void __init fill_ebus_device(struct device_node *dp, struct linux_ebus_de dev->ofdev.node = dp; dev->ofdev.dev.parent = &dev->bus->ofdev.dev; dev->ofdev.dev.bus = &ebus_bus_type; - strcpy(dev->ofdev.dev.bus_id, dp->path_component_name); + sprintf(dev->ofdev.dev.bus_id, "ebus[%08x]", dp->node); /* Register with core */ if (of_device_register(&dev->ofdev) != 0) printk(KERN_DEBUG "ebus: device registration error for %s!\n", - dev->ofdev.dev.bus_id); + dp->path_component_name); dp = dp->child; if (dp) { @@ -494,12 +494,12 @@ void __init ebus_init(void) ebus->ofdev.node = dp; ebus->ofdev.dev.parent = &pdev->dev; ebus->ofdev.dev.bus = &ebus_bus_type; - strcpy(ebus->ofdev.dev.bus_id, dp->path_component_name); + sprintf(ebus->ofdev.dev.bus_id, "ebus%d", num_ebus); /* Register with core */ if (of_device_register(&ebus->ofdev) != 0) printk(KERN_DEBUG "ebus: device registration error for %s!\n", - ebus->ofdev.dev.bus_id); + dp->path_component_name); child = dp->child; diff --git a/arch/sparc64/kernel/isa.c b/arch/sparc64/kernel/isa.c index 0f3aec7..f028e68 100644 --- a/arch/sparc64/kernel/isa.c +++ b/arch/sparc64/kernel/isa.c @@ -115,12 +115,12 @@ static void __init isa_fill_devices(struct sparc_isa_bridge *isa_br) isa_dev->ofdev.node = dp; isa_dev->ofdev.dev.parent = &isa_br->ofdev.dev; isa_dev->ofdev.dev.bus = &isa_bus_type; - strcpy(isa_dev->ofdev.dev.bus_id, dp->path_component_name); + sprintf(isa_dev->ofdev.dev.bus_id, "isa[%08x]", dp->node); /* Register with core */ if (of_device_register(&isa_dev->ofdev) != 0) { printk(KERN_DEBUG "isa: device registration error for %s!\n", - isa_dev->ofdev.dev.bus_id); + dp->path_component_name); kfree(isa_dev); goto next_sibling; } @@ -191,12 +191,12 @@ void __init isa_init(void) isa_br->ofdev.node = dp; isa_br->ofdev.dev.parent = &pdev->dev; isa_br->ofdev.dev.bus = &isa_bus_type; - strcpy(isa_br->ofdev.dev.bus_id, dp->path_component_name); + sprintf(isa_br->ofdev.dev.bus_id, "isa%d", index); /* Register with core */ if (of_device_register(&isa_br->ofdev) != 0) { printk(KERN_DEBUG "isa: device registration error for %s!\n", - isa_br->ofdev.dev.bus_id); + dp->path_component_name); kfree(isa_br); return; } diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c index 983ca5f..8cc14fc 100644 --- a/arch/sparc64/kernel/of_device.c +++ b/arch/sparc64/kernel/of_device.c @@ -861,7 +861,7 @@ static struct of_device * __init scan_one_device(struct device_node *dp, if (!parent) strcpy(op->dev.bus_id, "root"); else - sprintf(op->dev.bus_id, "%s@%08x", dp->name, dp->node); + sprintf(op->dev.bus_id, "%08x", dp->node); if (of_device_register(op)) { printk("%s: Could not register of device.\n", diff --git a/drivers/sbus/sbus.c b/drivers/sbus/sbus.c index 935952e..98fcbb3 100644 --- a/drivers/sbus/sbus.c +++ b/drivers/sbus/sbus.c @@ -61,11 +61,11 @@ static void __init fill_sbus_device(struct device_node *dp, struct sbus_dev *sde else sdev->ofdev.dev.parent = &sdev->bus->ofdev.dev; sdev->ofdev.dev.bus = &sbus_bus_type; - strcpy(sdev->ofdev.dev.bus_id, dp->path_component_name); + sprintf(sdev->ofdev.dev.bus_id, "sbus[%08x]", dp->node); if (of_device_register(&sdev->ofdev) != 0) printk(KERN_DEBUG "sbus: device registration error for %s!\n", - sdev->ofdev.dev.bus_id); + dp->path_component_name); } static void __init sbus_bus_ranges_init(struct device_node *dp, struct sbus_bus *sbus) -- cgit v0.10.2 From f4880391860f2e38bb6ff400a0bd6c85443c5b9e Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 27 Oct 2006 12:39:10 +0200 Subject: [S390] sys_getcpu compat wrapper. Looking at the new syscall additions, I noticed that sys_getcpu_wrapper wraps in to sys_tee, in what appears to be a copy and paste error. Switch it to point to sys_getcpu.. Signed-off-by: Paul Mundt Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index cb0efae..71e54ef 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1664,4 +1664,4 @@ sys_getcpu_wrapper: llgtr %r2,%r2 # unsigned * llgtr %r3,%r3 # unsigned * llgtr %r4,%r4 # struct getcpu_cache * - jg sys_tee + jg sys_getcpu -- cgit v0.10.2 From 95425f19beed99cb9b0a79f69731eda907d9db99 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Fri, 27 Oct 2006 12:39:13 +0200 Subject: [S390] Initialize interval value to 0. sscanf() could leave the interval value unchanged in which case it would be used uninitialized. Signed-off-by: Gerald Schaefer Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 45c9fa7..af1e8fc 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -310,6 +310,7 @@ appldata_interval_handler(ctl_table *ctl, int write, struct file *filp, if (copy_from_user(buf, buffer, len > sizeof(buf) ? sizeof(buf) : len)) { return -EFAULT; } + interval = 0; sscanf(buf, "%i", &interval); if (interval <= 0) { P_ERROR("Timer CPU interval has to be > 0!\n"); -- cgit v0.10.2 From c2b1449bd1fd73103ed5ff1a28d8f7cbc8a01b52 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 27 Oct 2006 12:39:17 +0200 Subject: [S390] cio: css_probe_device() must be called enabled. Move css_probe_device() behind giving up the lock for the old subchannel in css_evaluate_known_subchannel() so we aren't disabled when we call it. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c index a2dee5b..ad7f7e1 100644 --- a/drivers/s390/cio/css.c +++ b/drivers/s390/cio/css.c @@ -271,10 +271,6 @@ static int css_evaluate_known_subchannel(struct subchannel *sch, int slow) /* Reset intparm to zeroes. */ sch->schib.pmcw.intparm = 0; cio_modify(sch); - - /* Probe if necessary. */ - if (action == UNREGISTER_PROBE) - ret = css_probe_device(sch->schid); break; case REPROBE: device_trigger_reprobe(sch); @@ -283,6 +279,9 @@ static int css_evaluate_known_subchannel(struct subchannel *sch, int slow) break; } spin_unlock_irqrestore(&sch->lock, flags); + /* Probe if necessary. */ + if (action == UNREGISTER_PROBE) + ret = css_probe_device(sch->schid); return ret; } -- cgit v0.10.2 From 12bae23507129a7337378e6433bff6f8696bdb45 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 27 Oct 2006 12:39:22 +0200 Subject: [S390] uaccess error handling. Consider return values for all user space access function and return -EFAULT on error. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 2001767..5b33f82 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -757,7 +757,9 @@ asmlinkage long sys32_sysctl(struct __sysctl_args32 __user *args) put_user(oldlen, (u32 __user *)compat_ptr(tmp.oldlenp))) error = -EFAULT; } - copy_to_user(args->__unused, tmp.__unused, sizeof(tmp.__unused)); + if (copy_to_user(args->__unused, tmp.__unused, + sizeof(tmp.__unused))) + error = -EFAULT; } return error; } diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c index d49b876..861888a 100644 --- a/arch/s390/kernel/compat_signal.c +++ b/arch/s390/kernel/compat_signal.c @@ -169,12 +169,12 @@ sys32_sigaction(int sig, const struct old_sigaction32 __user *act, compat_old_sigset_t mask; if (!access_ok(VERIFY_READ, act, sizeof(*act)) || __get_user(sa_handler, &act->sa_handler) || - __get_user(sa_restorer, &act->sa_restorer)) + __get_user(sa_restorer, &act->sa_restorer) || + __get_user(new_ka.sa.sa_flags, &act->sa_flags) || + __get_user(mask, &act->sa_mask)) return -EFAULT; new_ka.sa.sa_handler = (__sighandler_t) sa_handler; new_ka.sa.sa_restorer = (void (*)(void)) sa_restorer; - __get_user(new_ka.sa.sa_flags, &act->sa_flags); - __get_user(mask, &act->sa_mask); siginitset(&new_ka.sa.sa_mask, mask); } @@ -185,10 +185,10 @@ sys32_sigaction(int sig, const struct old_sigaction32 __user *act, sa_restorer = (unsigned long) old_ka.sa.sa_restorer; if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || __put_user(sa_handler, &oact->sa_handler) || - __put_user(sa_restorer, &oact->sa_restorer)) + __put_user(sa_restorer, &oact->sa_restorer) || + __put_user(old_ka.sa.sa_flags, &oact->sa_flags) || + __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask)) return -EFAULT; - __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); } return ret; diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index 4392a77..4c8a795 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -80,10 +80,10 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, old_sigset_t mask; if (!access_ok(VERIFY_READ, act, sizeof(*act)) || __get_user(new_ka.sa.sa_handler, &act->sa_handler) || - __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) + __get_user(new_ka.sa.sa_restorer, &act->sa_restorer) || + __get_user(new_ka.sa.sa_flags, &act->sa_flags) || + __get_user(mask, &act->sa_mask)) return -EFAULT; - __get_user(new_ka.sa.sa_flags, &act->sa_flags); - __get_user(mask, &act->sa_mask); siginitset(&new_ka.sa.sa_mask, mask); } @@ -92,10 +92,10 @@ sys_sigaction(int sig, const struct old_sigaction __user *act, if (!ret && oact) { if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || - __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) + __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer) || + __put_user(old_ka.sa.sa_flags, &oact->sa_flags) || + __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask)) return -EFAULT; - __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); } return ret; diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index 66375a5..92ecffb 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -462,7 +462,8 @@ asmlinkage void illegal_op(struct pt_regs * regs, long interruption_code) local_irq_enable(); if (regs->psw.mask & PSW_MASK_PSTATE) { - get_user(*((__u16 *) opcode), (__u16 __user *) location); + if (get_user(*((__u16 *) opcode), (__u16 __user *) location)) + return; if (*((__u16 *) opcode) == S390_BREAKPOINT_U16) { if (current->ptrace & PT_PTRACED) force_sig(SIGTRAP, current); @@ -470,20 +471,25 @@ asmlinkage void illegal_op(struct pt_regs * regs, long interruption_code) signal = SIGILL; #ifdef CONFIG_MATHEMU } else if (opcode[0] == 0xb3) { - get_user(*((__u16 *) (opcode+2)), location+1); + if (get_user(*((__u16 *) (opcode+2)), location+1)) + return; signal = math_emu_b3(opcode, regs); } else if (opcode[0] == 0xed) { - get_user(*((__u32 *) (opcode+2)), - (__u32 __user *)(location+1)); + if (get_user(*((__u32 *) (opcode+2)), + (__u32 __user *)(location+1))) + return; signal = math_emu_ed(opcode, regs); } else if (*((__u16 *) opcode) == 0xb299) { - get_user(*((__u16 *) (opcode+2)), location+1); + if (get_user(*((__u16 *) (opcode+2)), location+1)) + return; signal = math_emu_srnm(opcode, regs); } else if (*((__u16 *) opcode) == 0xb29c) { - get_user(*((__u16 *) (opcode+2)), location+1); + if (get_user(*((__u16 *) (opcode+2)), location+1)) + return; signal = math_emu_stfpc(opcode, regs); } else if (*((__u16 *) opcode) == 0xb29d) { - get_user(*((__u16 *) (opcode+2)), location+1); + if (get_user(*((__u16 *) (opcode+2)), location+1)) + return; signal = math_emu_lfpc(opcode, regs); #endif } else -- cgit v0.10.2 From f3b017d8c9915cbaa8bab178dde1bd9dbbf5012c Mon Sep 17 00:00:00 2001 From: Ralph Wuerthner Date: Fri, 27 Oct 2006 12:39:26 +0200 Subject: [S390] Improve AP bus device removal. Added a call to device_unregister() in ap_scan_bus() to actively remove unavailable AP bus devices with every bus scan. Previously devices were only removed in ap_queue_message() or __ap_poll_all(). Signed-off-by: Ralph Wuerthner Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c index c5ccd20..79d89c3 100644 --- a/drivers/s390/crypto/ap_bus.c +++ b/drivers/s390/crypto/ap_bus.c @@ -739,11 +739,16 @@ static void ap_scan_bus(void *data) dev = bus_find_device(&ap_bus_type, NULL, (void *)(unsigned long)qid, __ap_scan_bus); + rc = ap_query_queue(qid, &queue_depth, &device_type); + if (dev && rc) { + put_device(dev); + device_unregister(dev); + continue; + } if (dev) { put_device(dev); continue; } - rc = ap_query_queue(qid, &queue_depth, &device_type); if (rc) continue; rc = ap_init_queue(qid); -- cgit v0.10.2 From 3c9da7ba049d11caccc219576a3a23404aa2fc50 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Fri, 27 Oct 2006 12:39:33 +0200 Subject: [S390] cio: Make ccw_device_register() static. ccw_device_register() is only called from io_subchannel_register() and io_subchannel_probe() and will never be called for possible non-io subchannels. Signed-off-by: Cornelia Huck Signed-off-by: Martin Schwidefsky diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index 94bdd4d..39c98f9 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -532,8 +532,7 @@ device_remove_files(struct device *dev) /* this is a simple abstraction for device_register that sets the * correct bus type and adds the bus specific files */ -int -ccw_device_register(struct ccw_device *cdev) +static int ccw_device_register(struct ccw_device *cdev) { struct device *dev = &cdev->dev; int ret; diff --git a/drivers/s390/cio/device.h b/drivers/s390/cio/device.h index c6140cc..9233b5c 100644 --- a/drivers/s390/cio/device.h +++ b/drivers/s390/cio/device.h @@ -78,7 +78,6 @@ void io_subchannel_recog_done(struct ccw_device *cdev); int ccw_device_cancel_halt_clear(struct ccw_device *); -int ccw_device_register(struct ccw_device *); void ccw_device_do_unreg_rereg(void *); void ccw_device_call_sch_unregister(void *); -- cgit v0.10.2 From 35ae61a0f43ebbabc3cb4345136ca529fc4d6700 Mon Sep 17 00:00:00 2001 From: MUNEDA Takahiro Date: Wed, 25 Oct 2006 11:44:57 -0700 Subject: acpiphp: fix latch status pci_hotplug.h says: * @latch_status: if the latch (if any) is open or closed (1/0) However, acpiphp returns opposite value. This patch fixes this issue. I tested this patch on my ia64 machine that has some apciphp based hotplug slots. Signed-off-by: MUNEDA Takahiro Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index c44311a..16167b0 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -1807,8 +1807,8 @@ u8 acpiphp_get_power_status(struct acpiphp_slot *slot) /* - * latch closed: 1 - * latch open: 0 + * latch open: 1 + * latch closed: 0 */ u8 acpiphp_get_latch_status(struct acpiphp_slot *slot) { @@ -1816,7 +1816,7 @@ u8 acpiphp_get_latch_status(struct acpiphp_slot *slot) sta = get_slot_status(slot); - return (sta & ACPI_STA_SHOW_IN_UI) ? 1 : 0; + return (sta & ACPI_STA_SHOW_IN_UI) ? 0 : 1; } -- cgit v0.10.2 From 6b5c76b8e2ff204fa8d7201acce461188873bf2b Mon Sep 17 00:00:00 2001 From: Eiichiro Oiwa Date: Mon, 23 Oct 2006 15:14:07 +0900 Subject: PCI: fix pci_fixup_video as it blows up on sparc64 This reverts much of the original pci_fixup_video change and makes it work for all arches that need it. fixed, and tested on x86, x86_64 and IA64 dig. Signed-off-by: Eiichiro Oiwa Acked-by: David Miller Signed-off-by: Greg Kroah-Hartman diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c index 908b410..c1949ff 100644 --- a/arch/i386/pci/fixup.c +++ b/arch/i386/pci/fixup.c @@ -343,6 +343,61 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC, pcie_ro DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_MCH_PC1, pcie_rootport_aspm_quirk ); /* + * Fixup to mark boot BIOS video selected by BIOS before it changes + * + * From information provided by "Jon Smirl" + * + * The standard boot ROM sequence for an x86 machine uses the BIOS + * to select an initial video card for boot display. This boot video + * card will have it's BIOS copied to C0000 in system RAM. + * IORESOURCE_ROM_SHADOW is used to associate the boot video + * card with this copy. On laptops this copy has to be used since + * the main ROM may be compressed or combined with another image. + * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW + * is marked here since the boot video device will be the only enabled + * video device at this point. + */ + +static void __devinit pci_fixup_video(struct pci_dev *pdev) +{ + struct pci_dev *bridge; + struct pci_bus *bus; + u16 config; + + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) + return; + + /* Is VGA routed to us? */ + bus = pdev->bus; + while (bus) { + bridge = bus->self; + + /* + * From information provided by + * "David Miller" + * The bridge control register is valid for PCI header + * type BRIDGE, or CARDBUS. Host to PCI controllers use + * PCI header type NORMAL. + */ + if (bridge + &&((bridge->hdr_type == PCI_HEADER_TYPE_BRIDGE) + ||(bridge->hdr_type == PCI_HEADER_TYPE_CARDBUS))) { + pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, + &config); + if (!(config & PCI_BRIDGE_CTL_VGA)) + return; + } + bus = bus->parent; + } + pci_read_config_word(pdev, PCI_COMMAND, &config); + if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { + pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; + printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_video); + +/* * Some Toshiba laptops need extra code to enable their TI TSB43AB22/A. * * We pretend to bring them out of full D3 state, and restore the proper diff --git a/arch/ia64/pci/Makefile b/arch/ia64/pci/Makefile index e66889e..fb14dc5 100644 --- a/arch/ia64/pci/Makefile +++ b/arch/ia64/pci/Makefile @@ -1,4 +1,4 @@ # # Makefile for the ia64-specific parts of the pci bus # -obj-y := pci.o +obj-y := pci.o fixup.o diff --git a/arch/ia64/pci/fixup.c b/arch/ia64/pci/fixup.c new file mode 100644 index 0000000..245dc1f --- /dev/null +++ b/arch/ia64/pci/fixup.c @@ -0,0 +1,69 @@ +/* + * Exceptions for specific devices. Usually work-arounds for fatal design flaws. + * Derived from fixup.c of i386 tree. + */ + +#include +#include + +#include + +/* + * Fixup to mark boot BIOS video selected by BIOS before it changes + * + * From information provided by "Jon Smirl" + * + * The standard boot ROM sequence for an x86 machine uses the BIOS + * to select an initial video card for boot display. This boot video + * card will have it's BIOS copied to C0000 in system RAM. + * IORESOURCE_ROM_SHADOW is used to associate the boot video + * card with this copy. On laptops this copy has to be used since + * the main ROM may be compressed or combined with another image. + * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW + * is marked here since the boot video device will be the only enabled + * video device at this point. + */ + +static void __devinit pci_fixup_video(struct pci_dev *pdev) +{ + struct pci_dev *bridge; + struct pci_bus *bus; + u16 config; + + if ((strcmp(platform_name, "dig") != 0) + && (strcmp(platform_name, "hpzx1") != 0)) + return; + /* Maybe, this machine supports legacy memory map. */ + + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) + return; + + /* Is VGA routed to us? */ + bus = pdev->bus; + while (bus) { + bridge = bus->self; + + /* + * From information provided by + * "David Miller" + * The bridge control register is valid for PCI header + * type BRIDGE, or CARDBUS. Host to PCI controllers use + * PCI header type NORMAL. + */ + if (bridge + &&((bridge->hdr_type == PCI_HEADER_TYPE_BRIDGE) + ||(bridge->hdr_type == PCI_HEADER_TYPE_CARDBUS))) { + pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, + &config); + if (!(config & PCI_BRIDGE_CTL_VGA)) + return; + } + bus = bus->parent; + } + pci_read_config_word(pdev, PCI_COMMAND, &config); + if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { + pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; + printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_video); diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index e8a7f1b..687ab4a 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1619,52 +1619,6 @@ static void __devinit fixup_rev1_53c810(struct pci_dev* dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1_53c810); -/* - * Fixup to mark boot BIOS video selected by BIOS before it changes - * - * From information provided by "Jon Smirl" - * - * The standard boot ROM sequence for an x86 machine uses the BIOS - * to select an initial video card for boot display. This boot video - * card will have it's BIOS copied to C0000 in system RAM. - * IORESOURCE_ROM_SHADOW is used to associate the boot video - * card with this copy. On laptops this copy has to be used since - * the main ROM may be compressed or combined with another image. - * See pci_map_rom() for use of this flag. IORESOURCE_ROM_SHADOW - * is marked here since the boot video device will be the only enabled - * video device at this point. - */ - -static void __devinit fixup_video(struct pci_dev *pdev) -{ - struct pci_dev *bridge; - struct pci_bus *bus; - u16 config; - - if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA) - return; - - /* Is VGA routed to us? */ - bus = pdev->bus; - while (bus) { - bridge = bus->self; - if (bridge) { - pci_read_config_word(bridge, PCI_BRIDGE_CONTROL, - &config); - if (!(config & PCI_BRIDGE_CTL_VGA)) - return; - } - bus = bus->parent; - } - pci_read_config_word(pdev, PCI_COMMAND, &config); - if (config & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) { - pdev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_SHADOW; - printk(KERN_DEBUG "Boot video device is %s\n", pci_name(pdev)); - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, fixup_video); - - static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { while (f < end) { diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index 43e4a49..e1dcefc 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -72,8 +72,9 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) int last_image; /* - * IORESOURCE_ROM_SHADOW set if the VGA enable bit of the Bridge Control - * register is set for embedded VGA. + * IORESOURCE_ROM_SHADOW set on x86, x86_64 and IA64 supports legacy + * memory map if the VGA enable bit of the Bridge Control register is + * set for embedded VGA. */ if (res->flags & IORESOURCE_ROM_SHADOW) { /* primary video rom always starts here */ -- cgit v0.10.2 From 3095fc0c9772b4afb3c81f76664f341ef716d380 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Fri, 20 Oct 2006 14:45:33 -0700 Subject: PCI: x86-64: mmconfig missing printk levels Signed-off-by: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index 7732f42..e61093b 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -220,7 +220,7 @@ void __init pci_mmcfg_init(int type) pci_mmcfg_virt = kmalloc(sizeof(*pci_mmcfg_virt) * pci_mmcfg_config_num, GFP_KERNEL); if (pci_mmcfg_virt == NULL) { - printk("PCI: Can not allocate memory for mmconfig structures\n"); + printk(KERN_ERR "PCI: Can not allocate memory for mmconfig structures\n"); return; } for (i = 0; i < pci_mmcfg_config_num; ++i) { @@ -228,7 +228,8 @@ void __init pci_mmcfg_init(int type) pci_mmcfg_virt[i].virt = ioremap_nocache(pci_mmcfg_config[i].base_address, MMCONFIG_APER_MAX); if (!pci_mmcfg_virt[i].virt) { - printk("PCI: Cannot map mmconfig aperture for segment %d\n", + printk(KERN_ERR "PCI: Cannot map mmconfig aperture for " + "segment %d\n", pci_mmcfg_config[i].pci_segment_group_number); return; } -- cgit v0.10.2 From 2449e06a5696b7af1c8a369b04c97f3b139cf3bb Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Fri, 20 Oct 2006 14:45:32 -0700 Subject: PCI: reset pci device state to unknown state for resume Considering below scenario: 1.Unload a PCI device's driver, the device ->current remains in PCI_D0. 2.Do suspend/resume circle. After that, BIOS puts the device to D3. 3.Reload the device driver. The calling pci_set_power_state in the driver can't change the state to D0, as set_power_state thinks the device is already in D0. A bug is reported at http://bugzilla.kernel.org/show_bug.cgi?id=6024 Pat attached a patch at http://marc.theaimsgroup.com/?l=linux-pci&m=114049761428561&w=2 for this issue, but it's lost. As pci_set_power_state can handle D3 -> D0 correctly (restore config space), I simplified Patrick's patch. Signed-off-by: Shaohua Li Cc: Patrick Mochel Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index b1c0c70..194f1d2 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -265,6 +265,13 @@ static int pci_device_remove(struct device * dev) } /* + * If the device is still on, set the power state as "unknown", + * since it might change by the next time we load the driver. + */ + if (pci_dev->current_state == PCI_D0) + pci_dev->current_state = PCI_UNKNOWN; + + /* * We would love to complain here if pci_dev->is_enabled is set, that * the driver should have called pci_disable_device(), but the * unfortunate fact is there are too many odd BIOS and bridge setups @@ -288,6 +295,12 @@ static int pci_device_suspend(struct device * dev, pm_message_t state) suspend_report_result(drv->suspend, i); } else { pci_save_state(pci_dev); + /* + * mark its power state as "unknown", since we don't know if + * e.g. the BIOS will change its device state when we suspend. + */ + if (pci_dev->current_state == PCI_D0) + pci_dev->current_state = PCI_UNKNOWN; } return i; } -- cgit v0.10.2 From 3560cc5ec3488b20d927f7160a21a0df1d1fda20 Mon Sep 17 00:00:00 2001 From: Karsten Wiese Date: Fri, 20 Oct 2006 14:45:36 -0700 Subject: PCI: Remove quirk_via_abnormal_poweroff My K8T800 mobo resumes fine from suspend to ram with and without patch applied against 2.6.18. quirk_via_abnormal_poweroff makes some boards not boot 2.6.18, so IMO patch should go to head, 2.6.18.2 and everywhere "ACPI: ACPICA 20060623" has been applied. Remove quirk_via_abnormal_poweroff Obsoleted by "ACPI: ACPICA 20060623": Implemented support for "ignored" bits in the ACPI registers. According to the ACPI specification, these bits should be preserved when writing the registers via a read/modify/write cycle. There are 3 bits preserved in this manner: PM1_CONTROL[0] (SCI_EN), PM1_CONTROL[9], and PM1_STATUS[11]. http://bugzilla.kernel.org/show_bug.cgi?id=3691 Signed-off-by: Karsten Wiese Cc: Bob Moore Cc: "Brown, Len" Acked-by: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 687ab4a..204b1c8 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -714,33 +714,6 @@ static void __devinit quirk_vt82c598_id(struct pci_dev *dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C597_0, quirk_vt82c598_id ); -#ifdef CONFIG_ACPI_SLEEP - -/* - * Some VIA systems boot with the abnormal status flag set. This can cause - * the BIOS to re-POST the system on resume rather than passing control - * back to the OS. Clear the flag on boot - */ -static void __devinit quirk_via_abnormal_poweroff(struct pci_dev *dev) -{ - u32 reg; - - acpi_hw_register_read(ACPI_MTX_DO_NOT_LOCK, ACPI_REGISTER_PM1_STATUS, - ®); - - if (reg & 0x800) { - printk("Clearing abnormal poweroff flag\n"); - acpi_hw_register_write(ACPI_MTX_DO_NOT_LOCK, - ACPI_REGISTER_PM1_STATUS, - (u16)0x800); - } -} - -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235, quirk_via_abnormal_poweroff); -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237, quirk_via_abnormal_poweroff); - -#endif - /* * CardBus controllers have a legacy base address that enables them * to respond as i82365 pcmcia controllers. We don't want them to -- cgit v0.10.2 From 61ce1efe6e40233663d27ab8ac9ba9710eebcaad Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 27 Oct 2006 11:41:44 -0700 Subject: [PATCH] vmlinux.lds: consolidate initcall sections Add a vmlinux.lds.h helper macro for defining the eight-level initcall table, teach all the architectures to use it. This is a prerequisite for a patch which performs initcall synchronisation for multithreaded-probing. Cc: Greg KH Signed-off-by: Andrew Morton [ Added AVR32 as well ] Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds diff --git a/arch/alpha/kernel/vmlinux.lds.S b/arch/alpha/kernel/vmlinux.lds.S index 71470e9..76bf071 100644 --- a/arch/alpha/kernel/vmlinux.lds.S +++ b/arch/alpha/kernel/vmlinux.lds.S @@ -48,13 +48,7 @@ SECTIONS . = ALIGN(8); __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 3ca574e..a8fa75e 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -45,13 +45,7 @@ SECTIONS *(.early_param.init) __early_end = .; __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; __con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/avr32/kernel/vmlinux.lds.c b/arch/avr32/kernel/vmlinux.lds.c index cdd627c..5c4424e 100644 --- a/arch/avr32/kernel/vmlinux.lds.c +++ b/arch/avr32/kernel/vmlinux.lds.c @@ -38,13 +38,7 @@ SECTIONS __setup_end = .; . = ALIGN(4); __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; __con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/frv/kernel/vmlinux.lds.S b/arch/frv/kernel/vmlinux.lds.S index f474534..9c1fb12 100644 --- a/arch/frv/kernel/vmlinux.lds.S +++ b/arch/frv/kernel/vmlinux.lds.S @@ -44,13 +44,7 @@ SECTIONS __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/h8300/kernel/vmlinux.lds.S b/arch/h8300/kernel/vmlinux.lds.S index 6406c38..756325d 100644 --- a/arch/h8300/kernel/vmlinux.lds.S +++ b/arch/h8300/kernel/vmlinux.lds.S @@ -118,13 +118,7 @@ SECTIONS . = ALIGN(0x4) ; ___setup_end = .; ___initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS ___initcall_end = .; ___con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index 1e7ac1c..adc1f23 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -126,13 +126,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index b3b2e38..d6083a0 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -128,13 +128,7 @@ SECTIONS .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; } diff --git a/arch/m32r/kernel/vmlinux.lds.S b/arch/m32r/kernel/vmlinux.lds.S index 13c7bb6..358b9ce 100644 --- a/arch/m32r/kernel/vmlinux.lds.S +++ b/arch/m32r/kernel/vmlinux.lds.S @@ -83,13 +83,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S index ccd2ceb..58afa8b 100644 --- a/arch/m68knommu/kernel/vmlinux.lds.S +++ b/arch/m68knommu/kernel/vmlinux.lds.S @@ -140,13 +140,7 @@ SECTIONS { *(.init.setup) __setup_end = .; __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; __con_initcall_start = .; *(.con_initcall.init) diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S index 0bb9cd8..25ed333 100644 --- a/arch/mips/kernel/vmlinux.lds.S +++ b/arch/mips/kernel/vmlinux.lds.S @@ -91,13 +91,7 @@ SECTIONS __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/arch/parisc/kernel/vmlinux.lds.S b/arch/parisc/kernel/vmlinux.lds.S index b3677fc..7b943b4 100644 --- a/arch/parisc/kernel/vmlinux.lds.S +++ b/arch/parisc/kernel/vmlinux.lds.S @@ -153,13 +153,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index cb0e8d4..e8342d8 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -108,13 +108,7 @@ SECTIONS .initcall.init : { __initcall_start = .; - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS __initcall_end = .; } diff --git a/arch/ppc/kernel/vmlinux.lds.S b/arch/ppc/kernel/vmlinux.lds.S index 095fd33..16e8661 100644 --- a/arch/ppc/kernel/vmlinux.lds.S +++ b/arch/ppc/kernel/vmlinux.lds.S @@ -115,13 +115,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index af9e69a0..fe0f2e9 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -83,13 +83,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S index 5eb9309..77b4026 100644 --- a/arch/sh/kernel/vmlinux.lds.S +++ b/arch/sh/kernel/vmlinux.lds.S @@ -76,13 +76,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sh64/kernel/vmlinux.lds.S b/arch/sh64/kernel/vmlinux.lds.S index a8fcc3a..95c4d75 100644 --- a/arch/sh64/kernel/vmlinux.lds.S +++ b/arch/sh64/kernel/vmlinux.lds.S @@ -108,13 +108,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : C_PHYS(.initcall.init) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 1dd78c8..5cc5ff7 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -49,13 +49,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/sparc64/kernel/vmlinux.lds.S b/arch/sparc64/kernel/vmlinux.lds.S index b097379..bd9de8c 100644 --- a/arch/sparc64/kernel/vmlinux.lds.S +++ b/arch/sparc64/kernel/vmlinux.lds.S @@ -57,13 +57,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/v850/kernel/vmlinux.lds.S b/arch/v850/kernel/vmlinux.lds.S index 6339921..88d087f 100644 --- a/arch/v850/kernel/vmlinux.lds.S +++ b/arch/v850/kernel/vmlinux.lds.S @@ -140,13 +140,7 @@ ___setup_end = . ; \ ___initcall_start = . ; \ *(.initcall.init) \ - *(.initcall1.init) \ - *(.initcall2.init) \ - *(.initcall3.init) \ - *(.initcall4.init) \ - *(.initcall5.init) \ - *(.initcall6.init) \ - *(.initcall7.init) \ + INITCALLS \ . = ALIGN (4) ; \ ___initcall_end = . ; \ ___con_initcall_start = .; \ diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 1283614..edb24aa 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -175,13 +175,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/xtensa/kernel/vmlinux.lds.S b/arch/xtensa/kernel/vmlinux.lds.S index ab6cdbd..cfe75f5 100644 --- a/arch/xtensa/kernel/vmlinux.lds.S +++ b/arch/xtensa/kernel/vmlinux.lds.S @@ -184,13 +184,7 @@ SECTIONS __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 9d0d11c..e3e83bc 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -213,3 +213,13 @@ #define NOTES \ .notes : { *(.note.*) } :note + +#define INITCALLS \ + *(.initcall1.init) \ + *(.initcall2.init) \ + *(.initcall3.init) \ + *(.initcall4.init) \ + *(.initcall5.init) \ + *(.initcall6.init) \ + *(.initcall7.init) + -- cgit v0.10.2 From 735a7ffb739b6efeaeb1e720306ba308eaaeb20e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 27 Oct 2006 11:42:37 -0700 Subject: [PATCH] drivers: wait for threaded probes between initcall levels The multithreaded-probing code has a problem: after one initcall level (eg, core_initcall) has been processed, we will then start processing the next level (postcore_initcall) while the kernel threads which are handling core_initcall are still executing. This breaks the guarantees which the layered initcalls previously gave us. IOW, we want to be multithreaded _within_ an initcall level, but not between different levels. Fix that up by causing the probing code to wait for all outstanding probes at one level to complete before we start processing the next level. Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/base/dd.c b/drivers/base/dd.c index db01b95..c5d6bb4 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "base.h" #include "power/power.h" @@ -70,6 +71,8 @@ struct stupid_thread_structure { }; static atomic_t probe_count = ATOMIC_INIT(0); +static DECLARE_WAIT_QUEUE_HEAD(probe_waitqueue); + static int really_probe(void *void_data) { struct stupid_thread_structure *data = void_data; @@ -121,6 +124,7 @@ probe_failed: done: kfree(data); atomic_dec(&probe_count); + wake_up(&probe_waitqueue); return ret; } @@ -337,6 +341,32 @@ void driver_detach(struct device_driver * drv) } } +#ifdef CONFIG_PCI_MULTITHREAD_PROBE +static int __init wait_for_probes(void) +{ + DEFINE_WAIT(wait); + + printk(KERN_INFO "%s: waiting for %d threads\n", __FUNCTION__, + atomic_read(&probe_count)); + if (!atomic_read(&probe_count)) + return 0; + while (atomic_read(&probe_count)) { + prepare_to_wait(&probe_waitqueue, &wait, TASK_UNINTERRUPTIBLE); + if (atomic_read(&probe_count)) + schedule(); + } + finish_wait(&probe_waitqueue, &wait); + return 0; +} + +core_initcall_sync(wait_for_probes); +postcore_initcall_sync(wait_for_probes); +arch_initcall_sync(wait_for_probes); +subsys_initcall_sync(wait_for_probes); +fs_initcall_sync(wait_for_probes); +device_initcall_sync(wait_for_probes); +late_initcall_sync(wait_for_probes); +#endif EXPORT_SYMBOL_GPL(device_bind_driver); EXPORT_SYMBOL_GPL(device_release_driver); diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index e3e83bc..9d87316 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -216,10 +216,17 @@ #define INITCALLS \ *(.initcall1.init) \ + *(.initcall1s.init) \ *(.initcall2.init) \ + *(.initcall2s.init) \ *(.initcall3.init) \ + *(.initcall3s.init) \ *(.initcall4.init) \ + *(.initcall4s.init) \ *(.initcall5.init) \ + *(.initcall5s.init) \ *(.initcall6.init) \ - *(.initcall7.init) + *(.initcall6s.init) \ + *(.initcall7.init) \ + *(.initcall7s.init) diff --git a/include/linux/init.h b/include/linux/init.h index e92b145..ff40ea1 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -84,19 +84,29 @@ extern void setup_arch(char **); * by link order. * For backwards compatibility, initcall() puts the call in * the device init subsection. + * + * The `id' arg to __define_initcall() is needed so that multiple initcalls + * can point at the same handler without causing duplicate-symbol build errors. */ -#define __define_initcall(level,fn) \ - static initcall_t __initcall_##fn __attribute_used__ \ +#define __define_initcall(level,fn,id) \ + static initcall_t __initcall_##fn##id __attribute_used__ \ __attribute__((__section__(".initcall" level ".init"))) = fn -#define core_initcall(fn) __define_initcall("1",fn) -#define postcore_initcall(fn) __define_initcall("2",fn) -#define arch_initcall(fn) __define_initcall("3",fn) -#define subsys_initcall(fn) __define_initcall("4",fn) -#define fs_initcall(fn) __define_initcall("5",fn) -#define device_initcall(fn) __define_initcall("6",fn) -#define late_initcall(fn) __define_initcall("7",fn) +#define core_initcall(fn) __define_initcall("1",fn,1) +#define core_initcall_sync(fn) __define_initcall("1s",fn,1s) +#define postcore_initcall(fn) __define_initcall("2",fn,2) +#define postcore_initcall_sync(fn) __define_initcall("2s",fn,2s) +#define arch_initcall(fn) __define_initcall("3",fn,3) +#define arch_initcall_sync(fn) __define_initcall("3s",fn,3s) +#define subsys_initcall(fn) __define_initcall("4",fn,4) +#define subsys_initcall_sync(fn) __define_initcall("4s",fn,4s) +#define fs_initcall(fn) __define_initcall("5",fn,5) +#define fs_initcall_sync(fn) __define_initcall("5s",fn,5s) +#define device_initcall(fn) __define_initcall("6",fn,6) +#define device_initcall_sync(fn) __define_initcall("6s",fn,6s) +#define late_initcall(fn) __define_initcall("7",fn,7) +#define late_initcall_sync(fn) __define_initcall("7s",fn,7s) #define __initcall(fn) device_initcall(fn) -- cgit v0.10.2 From efbfe96c5d839c367249bf1cd53249716450c0a2 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Fri, 27 Oct 2006 23:24:47 +0200 Subject: [PATCH] silence 'make xmldocs' warning by adding missing description of 'raw' in nand_base.c:1485 Add description of 'raw' in comments for drivers/mtd/nand/nand_base.c::nand_write_page_syndrome() so 'make xmldocs' will not spew a warning at us. Signed-off-by: Jesper Juhl Acked-by: Thomas Gleixner Signed-off-by: Linus Torvalds diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c index baece61..41bfcae 100644 --- a/drivers/mtd/nand/nand_base.c +++ b/drivers/mtd/nand/nand_base.c @@ -1479,6 +1479,7 @@ static void nand_write_page_syndrome(struct mtd_info *mtd, * @buf: the data to write * @page: page number to write * @cached: cached programming + * @raw: use _raw version of write_page */ static int nand_write_page(struct mtd_info *mtd, struct nand_chip *chip, const uint8_t *buf, int page, int cached, int raw) -- cgit v0.10.2 From c97d4869a23c439d2bc23cb26c1147c099f9ff78 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 25 Oct 2006 13:59:16 +0100 Subject: [ARM] Fix SMP irqflags support The IRQ changes a while back broke the build for SMP machines. Fix up the SMP code to use set_irq_regs/get_irq_regs as appropriate. Also, fix a warning in arch/arm/kernel/time.c where 'regs' becomes unused for SMP builds. Signed-off-by: Russell King diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 421329f..a07d202 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -7,6 +7,7 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ +#include #include #include #include @@ -19,6 +20,7 @@ #include #include #include +#include #include #include @@ -474,25 +476,26 @@ void show_local_irqs(struct seq_file *p) seq_putc(p, '\n'); } -static void ipi_timer(struct pt_regs *regs) +static void ipi_timer(void) { - int user = user_mode(regs); - irq_enter(); - profile_tick(CPU_PROFILING, regs); - update_process_times(user); + profile_tick(CPU_PROFILING); + update_process_times(user_mode(get_irq_regs())); irq_exit(); } #ifdef CONFIG_LOCAL_TIMERS asmlinkage void do_local_timer(struct pt_regs *regs) { + struct pt_regs *old_regs = set_irq_regs(regs); int cpu = smp_processor_id(); if (local_timer_ack()) { irq_stat[cpu].local_timer_irqs++; - ipi_timer(regs); + ipi_timer(); } + + set_irq_regs(old_regs); } #endif @@ -551,6 +554,7 @@ asmlinkage void do_IPI(struct pt_regs *regs) { unsigned int cpu = smp_processor_id(); struct ipi_data *ipi = &per_cpu(ipi_data, cpu); + struct pt_regs *old_regs = set_irq_regs(regs); ipi->ipi_count++; @@ -574,7 +578,7 @@ asmlinkage void do_IPI(struct pt_regs *regs) switch (nextmsg) { case IPI_TIMER: - ipi_timer(regs); + ipi_timer(); break; case IPI_RESCHEDULE: @@ -599,6 +603,8 @@ asmlinkage void do_IPI(struct pt_regs *regs) } } while (msgs); } + + set_irq_regs(old_regs); } void smp_send_reschedule(int cpu) diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c index c03cab5..0c5a609 100644 --- a/arch/arm/kernel/time.c +++ b/arch/arm/kernel/time.c @@ -327,13 +327,12 @@ EXPORT_SYMBOL(restore_time_delta); */ void timer_tick(void) { - struct pt_regs *regs = get_irq_regs(); profile_tick(CPU_PROFILING); do_leds(); do_set_rtc(); do_timer(1); #ifndef CONFIG_SMP - update_process_times(user_mode(regs)); + update_process_times(user_mode(get_irq_regs())); #endif } -- cgit v0.10.2 From a233bf9ee819c726c581af48010e0c0f1cdde245 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 25 Oct 2006 14:25:52 +0100 Subject: [ARM] Add realview SMP default configuration Signed-off-by: Russell King diff --git a/arch/arm/configs/realview-smp_defconfig b/arch/arm/configs/realview-smp_defconfig new file mode 100644 index 0000000..ffd905f --- /dev/null +++ b/arch/arm/configs/realview-smp_defconfig @@ -0,0 +1,994 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.19-rc3 +# Wed Oct 25 14:12:00 2006 +# +CONFIG_ARM=y +# CONFIG_GENERIC_TIME is not set +CONFIG_MMU=y +CONFIG_GENERIC_HARDIRQS=y +CONFIG_TRACE_IRQFLAGS_SUPPORT=y +CONFIG_HARDIRQS_SW_RESEND=y +CONFIG_GENERIC_IRQ_PROBE=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +CONFIG_GENERIC_HWEIGHT=y +CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_VECTORS_BASE=0xffff0000 +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" + +# +# Code maturity level options +# +CONFIG_EXPERIMENTAL=y +CONFIG_LOCK_KERNEL=y +CONFIG_INIT_ENV_ARG_LIMIT=32 + +# +# General setup +# +CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y +# CONFIG_SWAP is not set +CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set +# CONFIG_POSIX_MQUEUE is not set +# CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set +# CONFIG_UTS_NS is not set +# CONFIG_AUDIT is not set +# CONFIG_IKCONFIG is not set +# CONFIG_CPUSETS is not set +# CONFIG_RELAY is not set +CONFIG_INITRAMFS_SOURCE="" +CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_SYSCTL=y +# CONFIG_EMBEDDED is not set +CONFIG_UID16=y +# CONFIG_SYSCTL_SYSCALL is not set +CONFIG_KALLSYMS=y +CONFIG_KALLSYMS_ALL=y +# CONFIG_KALLSYMS_EXTRA_PASS is not set +CONFIG_HOTPLUG=y +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_ELF_CORE=y +CONFIG_BASE_FULL=y +CONFIG_FUTEX=y +CONFIG_EPOLL=y +CONFIG_SHMEM=y +CONFIG_SLAB=y +CONFIG_VM_EVENT_COUNTERS=y +CONFIG_RT_MUTEXES=y +# CONFIG_TINY_SHMEM is not set +CONFIG_BASE_SMALL=0 +# CONFIG_SLOB is not set + +# +# Loadable module support +# +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +# CONFIG_MODVERSIONS is not set +# CONFIG_MODULE_SRCVERSION_ALL is not set +# CONFIG_KMOD is not set +CONFIG_STOP_MACHINE=y + +# +# Block layer +# +CONFIG_BLOCK=y +# CONFIG_BLK_DEV_IO_TRACE is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +# CONFIG_IOSCHED_AS is not set +CONFIG_IOSCHED_DEADLINE=y +# CONFIG_IOSCHED_CFQ is not set +# CONFIG_DEFAULT_AS is not set +CONFIG_DEFAULT_DEADLINE=y +# CONFIG_DEFAULT_CFQ is not set +# CONFIG_DEFAULT_NOOP is not set +CONFIG_DEFAULT_IOSCHED="deadline" + +# +# System Type +# +# CONFIG_ARCH_AAEC2000 is not set +# CONFIG_ARCH_INTEGRATOR is not set +CONFIG_ARCH_REALVIEW=y +# CONFIG_ARCH_VERSATILE is not set +# CONFIG_ARCH_AT91 is not set +# CONFIG_ARCH_CLPS7500 is not set +# CONFIG_ARCH_CLPS711X is not set +# CONFIG_ARCH_CO285 is not set +# CONFIG_ARCH_EBSA110 is not set +# CONFIG_ARCH_EP93XX is not set +# CONFIG_ARCH_FOOTBRIDGE is not set +# CONFIG_ARCH_NETX is not set +# CONFIG_ARCH_H720X is not set +# CONFIG_ARCH_IMX is not set +# CONFIG_ARCH_IOP32X is not set +# CONFIG_ARCH_IOP33X is not set +# CONFIG_ARCH_IXP4XX is not set +# CONFIG_ARCH_IXP2000 is not set +# CONFIG_ARCH_IXP23XX is not set +# CONFIG_ARCH_L7200 is not set +# CONFIG_ARCH_PNX4008 is not set +# CONFIG_ARCH_PXA is not set +# CONFIG_ARCH_RPC is not set +# CONFIG_ARCH_SA1100 is not set +# CONFIG_ARCH_S3C2410 is not set +# CONFIG_ARCH_SHARK is not set +# CONFIG_ARCH_LH7A40X is not set +# CONFIG_ARCH_OMAP is not set + +# +# RealView platform type +# +CONFIG_MACH_REALVIEW_EB=y +CONFIG_REALVIEW_MPCORE=y + +# +# Processor Type +# +CONFIG_CPU_32=y +# CONFIG_CPU_ARM926T is not set +CONFIG_CPU_V6=y +CONFIG_CPU_32v6K=y +CONFIG_CPU_32v6=y +CONFIG_CPU_ABRT_EV6=y +CONFIG_CPU_CACHE_V6=y +CONFIG_CPU_CACHE_VIPT=y +CONFIG_CPU_COPY_V6=y +CONFIG_CPU_TLB_V6=y +CONFIG_CPU_CP15=y +CONFIG_CPU_CP15_MMU=y + +# +# Processor Features +# +CONFIG_ARM_THUMB=y +# CONFIG_CPU_ICACHE_DISABLE is not set +# CONFIG_CPU_DCACHE_DISABLE is not set +# CONFIG_CPU_DCACHE_WRITETHROUGH is not set +# CONFIG_CPU_BPREDICT_DISABLE is not set +CONFIG_HAS_TLS_REG=y +CONFIG_ARM_GIC=y +CONFIG_ICST307=y + +# +# Bus support +# +CONFIG_ARM_AMBA=y + +# +# PCCARD (PCMCIA/CardBus) support +# +# CONFIG_PCCARD is not set + +# +# Kernel Features +# +CONFIG_SMP=y +CONFIG_NR_CPUS=4 +CONFIG_HOTPLUG_CPU=y +CONFIG_LOCAL_TIMERS=y +# CONFIG_PREEMPT is not set +# CONFIG_NO_IDLE_HZ is not set +CONFIG_HZ=100 +# CONFIG_AEABI is not set +# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set +CONFIG_ALIGNMENT_TRAP=y + +# +# Boot options +# +CONFIG_ZBOOT_ROM_TEXT=0x0 +CONFIG_ZBOOT_ROM_BSS=0x0 +CONFIG_CMDLINE="root=/dev/nfs nfsroot=10.1.69.3:/work/nfsroot ip=dhcp console=ttyAMA0 mem=128M" +# CONFIG_XIP_KERNEL is not set + +# +# Floating point emulation +# + +# +# At least one emulation must be selected +# +CONFIG_FPE_NWFPE=y +# CONFIG_FPE_NWFPE_XP is not set +# CONFIG_FPE_FASTFPE is not set +CONFIG_VFP=y + +# +# Userspace binary formats +# +CONFIG_BINFMT_ELF=y +# CONFIG_BINFMT_AOUT is not set +# CONFIG_BINFMT_MISC is not set +# CONFIG_ARTHUR is not set + +# +# Power management options +# +# CONFIG_PM is not set +# CONFIG_APM is not set + +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +# CONFIG_NETDEBUG is not set +CONFIG_PACKET=y +# CONFIG_PACKET_MMAP is not set +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +CONFIG_IP_PNP=y +CONFIG_IP_PNP_DHCP=y +CONFIG_IP_PNP_BOOTP=y +# CONFIG_IP_PNP_RARP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_XFRM_TUNNEL is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_TRANSPORT is not set +# CONFIG_INET_XFRM_MODE_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_BEET is not set +CONFIG_INET_DIAG=y +CONFIG_INET_TCP_DIAG=y +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_IPV6 is not set +# CONFIG_INET6_XFRM_TUNNEL is not set +# CONFIG_INET6_TUNNEL is not set +# CONFIG_NETWORK_SECMARK is not set +# CONFIG_NETFILTER is not set + +# +# DCCP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_DCCP is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set + +# +# TIPC Configuration (EXPERIMENTAL) +# +# CONFIG_TIPC is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set + +# +# QoS and/or fair queueing +# +# CONFIG_NET_SCHED is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set +# CONFIG_IEEE80211 is not set + +# +# Device Drivers +# + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_FW_LOADER is not set +# CONFIG_DEBUG_DRIVER is not set +# CONFIG_SYS_HYPERVISOR is not set + +# +# Connector - unified userspace <-> kernelspace linker +# +# CONFIG_CONNECTOR is not set + +# +# Memory Technology Devices (MTD) +# +CONFIG_MTD=y +# CONFIG_MTD_DEBUG is not set +# CONFIG_MTD_CONCAT is not set +CONFIG_MTD_PARTITIONS=y +# CONFIG_MTD_REDBOOT_PARTS is not set +CONFIG_MTD_CMDLINE_PARTS=y +# CONFIG_MTD_AFS_PARTS is not set + +# +# User Modules And Translation Layers +# +CONFIG_MTD_CHAR=y +CONFIG_MTD_BLOCK=y +# CONFIG_FTL is not set +# CONFIG_NFTL is not set +# CONFIG_INFTL is not set +# CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set + +# +# RAM/ROM/Flash chip drivers +# +CONFIG_MTD_CFI=y +# CONFIG_MTD_JEDECPROBE is not set +CONFIG_MTD_GEN_PROBE=y +# CONFIG_MTD_CFI_ADV_OPTIONS is not set +CONFIG_MTD_MAP_BANK_WIDTH_1=y +CONFIG_MTD_MAP_BANK_WIDTH_2=y +CONFIG_MTD_MAP_BANK_WIDTH_4=y +# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set +CONFIG_MTD_CFI_I1=y +CONFIG_MTD_CFI_I2=y +# CONFIG_MTD_CFI_I4 is not set +# CONFIG_MTD_CFI_I8 is not set +CONFIG_MTD_CFI_INTELEXT=y +CONFIG_MTD_CFI_AMDSTD=y +# CONFIG_MTD_CFI_STAA is not set +CONFIG_MTD_CFI_UTIL=y +# CONFIG_MTD_RAM is not set +# CONFIG_MTD_ROM is not set +# CONFIG_MTD_ABSENT is not set +# CONFIG_MTD_OBSOLETE_CHIPS is not set + +# +# Mapping drivers for chip access +# +# CONFIG_MTD_COMPLEX_MAPPINGS is not set +# CONFIG_MTD_PHYSMAP is not set +CONFIG_MTD_ARM_INTEGRATOR=y +# CONFIG_MTD_PLATRAM is not set + +# +# Self-contained MTD device drivers +# +# CONFIG_MTD_SLRAM is not set +# CONFIG_MTD_PHRAM is not set +# CONFIG_MTD_MTDRAM is not set +# CONFIG_MTD_BLOCK2MTD is not set + +# +# Disk-On-Chip Device Drivers +# +# CONFIG_MTD_DOC2000 is not set +# CONFIG_MTD_DOC2001 is not set +# CONFIG_MTD_DOC2001PLUS is not set + +# +# NAND Flash Device Drivers +# +# CONFIG_MTD_NAND is not set + +# +# OneNAND Flash Device Drivers +# +# CONFIG_MTD_ONENAND is not set + +# +# Parallel port support +# +# CONFIG_PARPORT is not set + +# +# Plug and Play support +# + +# +# Block devices +# +# CONFIG_BLK_DEV_COW_COMMON is not set +# CONFIG_BLK_DEV_LOOP is not set +# CONFIG_BLK_DEV_NBD is not set +# CONFIG_BLK_DEV_RAM is not set +CONFIG_BLK_DEV_INITRD=y +# CONFIG_CDROM_PKTCDVD is not set +# CONFIG_ATA_OVER_ETH is not set + +# +# SCSI device support +# +# CONFIG_RAID_ATTRS is not set +# CONFIG_SCSI is not set +# CONFIG_SCSI_NETLINK is not set + +# +# Multi-device support (RAID and LVM) +# +# CONFIG_MD is not set + +# +# Fusion MPT device support +# +# CONFIG_FUSION is not set + +# +# IEEE 1394 (FireWire) support +# + +# +# I2O device support +# + +# +# Network device support +# +CONFIG_NETDEVICES=y +# CONFIG_DUMMY is not set +# CONFIG_BONDING is not set +# CONFIG_EQUALIZER is not set +# CONFIG_TUN is not set + +# +# PHY device support +# +# CONFIG_PHYLIB is not set + +# +# Ethernet (10 or 100Mbit) +# +CONFIG_NET_ETHERNET=y +CONFIG_MII=y +CONFIG_SMC91X=y +# CONFIG_DM9000 is not set + +# +# Ethernet (1000 Mbit) +# + +# +# Ethernet (10000 Mbit) +# + +# +# Token Ring devices +# + +# +# Wireless LAN (non-hamradio) +# +# CONFIG_NET_RADIO is not set + +# +# Wan interfaces +# +# CONFIG_WAN is not set +# CONFIG_PPP is not set +# CONFIG_SLIP is not set +# CONFIG_SHAPER is not set +# CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set + +# +# ISDN subsystem +# +# CONFIG_ISDN is not set + +# +# Input device support +# +CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set + +# +# Userland interfaces +# +CONFIG_INPUT_MOUSEDEV=y +CONFIG_INPUT_MOUSEDEV_PSAUX=y +CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024 +CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768 +# CONFIG_INPUT_JOYDEV is not set +# CONFIG_INPUT_TSDEV is not set +# CONFIG_INPUT_EVDEV is not set +# CONFIG_INPUT_EVBUG is not set + +# +# Input Device Drivers +# +CONFIG_INPUT_KEYBOARD=y +CONFIG_KEYBOARD_ATKBD=y +# CONFIG_KEYBOARD_SUNKBD is not set +# CONFIG_KEYBOARD_LKKBD is not set +# CONFIG_KEYBOARD_XTKBD is not set +# CONFIG_KEYBOARD_NEWTON is not set +# CONFIG_KEYBOARD_STOWAWAY is not set +CONFIG_INPUT_MOUSE=y +CONFIG_MOUSE_PS2=y +# CONFIG_MOUSE_SERIAL is not set +# CONFIG_MOUSE_VSXXXAA is not set +# CONFIG_INPUT_JOYSTICK is not set +# CONFIG_INPUT_TOUCHSCREEN is not set +# CONFIG_INPUT_MISC is not set + +# +# Hardware I/O ports +# +CONFIG_SERIO=y +# CONFIG_SERIO_SERPORT is not set +CONFIG_SERIO_AMBAKMI=y +CONFIG_SERIO_LIBPS2=y +# CONFIG_SERIO_RAW is not set +# CONFIG_GAMEPORT is not set + +# +# Character devices +# +CONFIG_VT=y +CONFIG_VT_CONSOLE=y +CONFIG_HW_CONSOLE=y +# CONFIG_VT_HW_CONSOLE_BINDING is not set +# CONFIG_SERIAL_NONSTANDARD is not set + +# +# Serial drivers +# +# CONFIG_SERIAL_8250 is not set + +# +# Non-8250 serial port support +# +# CONFIG_SERIAL_AMBA_PL010 is not set +CONFIG_SERIAL_AMBA_PL011=y +CONFIG_SERIAL_AMBA_PL011_CONSOLE=y +CONFIG_SERIAL_CORE=y +CONFIG_SERIAL_CORE_CONSOLE=y +CONFIG_UNIX98_PTYS=y +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=16 + +# +# IPMI +# +# CONFIG_IPMI_HANDLER is not set + +# +# Watchdog Cards +# +# CONFIG_WATCHDOG is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_NVRAM is not set +# CONFIG_DTLK is not set +# CONFIG_R3964 is not set + +# +# Ftape, the floppy tape device driver +# +# CONFIG_RAW_DRIVER is not set + +# +# TPM devices +# +# CONFIG_TCG_TPM is not set + +# +# I2C support +# +# CONFIG_I2C is not set + +# +# SPI support +# +# CONFIG_SPI is not set +# CONFIG_SPI_MASTER is not set + +# +# Dallas's 1-wire bus +# +# CONFIG_W1 is not set + +# +# Hardware Monitoring support +# +# CONFIG_HWMON is not set +# CONFIG_HWMON_VID is not set + +# +# Misc devices +# +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# +# LED devices +# +# CONFIG_NEW_LEDS is not set + +# +# LED drivers +# + +# +# LED Triggers +# + +# +# Multimedia devices +# +# CONFIG_VIDEO_DEV is not set + +# +# Digital Video Broadcasting Devices +# +# CONFIG_DVB is not set + +# +# Graphics support +# +# CONFIG_FIRMWARE_EDID is not set +CONFIG_FB=y +CONFIG_FB_CFB_FILLRECT=y +CONFIG_FB_CFB_COPYAREA=y +CONFIG_FB_CFB_IMAGEBLIT=y +# CONFIG_FB_MACMODES is not set +# CONFIG_FB_BACKLIGHT is not set +# CONFIG_FB_MODE_HELPERS is not set +# CONFIG_FB_TILEBLITTING is not set +CONFIG_FB_ARMCLCD=y +# CONFIG_FB_S1D13XXX is not set +# CONFIG_FB_VIRTUAL is not set + +# +# Console display driver support +# +# CONFIG_VGA_CONSOLE is not set +CONFIG_DUMMY_CONSOLE=y +CONFIG_FRAMEBUFFER_CONSOLE=y +# CONFIG_FRAMEBUFFER_CONSOLE_ROTATION is not set +# CONFIG_FONTS is not set +CONFIG_FONT_8x8=y +CONFIG_FONT_8x16=y + +# +# Logo configuration +# +CONFIG_LOGO=y +# CONFIG_LOGO_LINUX_MONO is not set +# CONFIG_LOGO_LINUX_VGA16 is not set +CONFIG_LOGO_LINUX_CLUT224=y +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set + +# +# Sound +# +CONFIG_SOUND=y + +# +# Advanced Linux Sound Architecture +# +CONFIG_SND=y +CONFIG_SND_TIMER=y +CONFIG_SND_PCM=y +# CONFIG_SND_SEQUENCER is not set +CONFIG_SND_OSSEMUL=y +CONFIG_SND_MIXER_OSS=y +CONFIG_SND_PCM_OSS=y +CONFIG_SND_PCM_OSS_PLUGINS=y +# CONFIG_SND_DYNAMIC_MINORS is not set +CONFIG_SND_SUPPORT_OLD_API=y +CONFIG_SND_VERBOSE_PROCFS=y +# CONFIG_SND_VERBOSE_PRINTK is not set +# CONFIG_SND_DEBUG is not set + +# +# Generic devices +# +CONFIG_SND_AC97_CODEC=m +CONFIG_SND_AC97_BUS=m +# CONFIG_SND_DUMMY is not set +# CONFIG_SND_MTPAV is not set +# CONFIG_SND_SERIAL_U16550 is not set +# CONFIG_SND_MPU401 is not set + +# +# ALSA ARM devices +# +CONFIG_SND_ARMAACI=m + +# +# Open Sound System +# +# CONFIG_SOUND_PRIME is not set + +# +# USB support +# +CONFIG_USB_ARCH_HAS_HCD=y +# CONFIG_USB_ARCH_HAS_OHCI is not set +# CONFIG_USB_ARCH_HAS_EHCI is not set +# CONFIG_USB is not set + +# +# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' +# + +# +# USB Gadget Support +# +# CONFIG_USB_GADGET is not set + +# +# MMC/SD Card support +# +CONFIG_MMC=y +# CONFIG_MMC_DEBUG is not set +CONFIG_MMC_BLOCK=y +CONFIG_MMC_ARMMMCI=y +# CONFIG_MMC_TIFM_SD is not set + +# +# Real Time Clock +# +CONFIG_RTC_LIB=y +CONFIG_RTC_CLASS=y +CONFIG_RTC_HCTOSYS=y +CONFIG_RTC_HCTOSYS_DEVICE="rtc0" +# CONFIG_RTC_DEBUG is not set + +# +# RTC interfaces +# +CONFIG_RTC_INTF_SYSFS=y +CONFIG_RTC_INTF_PROC=y +CONFIG_RTC_INTF_DEV=y +CONFIG_RTC_INTF_DEV_UIE_EMUL=y + +# +# RTC drivers +# +# CONFIG_RTC_DRV_DS1553 is not set +# CONFIG_RTC_DRV_DS1742 is not set +# CONFIG_RTC_DRV_M48T86 is not set +CONFIG_RTC_DRV_PL031=y +# CONFIG_RTC_DRV_TEST is not set +# CONFIG_RTC_DRV_V3020 is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT2_FS_XIP is not set +# CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set +# CONFIG_REISERFS_FS is not set +# CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set +# CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set +# CONFIG_OCFS2_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y +# CONFIG_INOTIFY_USER is not set +# CONFIG_QUOTA is not set +CONFIG_DNOTIFY=y +# CONFIG_AUTOFS_FS is not set +# CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set + +# +# CD-ROM/DVD Filesystems +# +# CONFIG_ISO9660_FS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +CONFIG_FAT_FS=y +# CONFIG_MSDOS_FS is not set +CONFIG_VFAT_FS=y +CONFIG_FAT_DEFAULT_CODEPAGE=437 +CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y +CONFIG_SYSFS=y +CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set +# CONFIG_HUGETLB_PAGE is not set +CONFIG_RAMFS=y +# CONFIG_CONFIGFS_FS is not set + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_JFFS_FS is not set +# CONFIG_JFFS2_FS is not set +CONFIG_CRAMFS=y +# CONFIG_VXFS_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set + +# +# Network File Systems +# +CONFIG_NFS_FS=y +CONFIG_NFS_V3=y +# CONFIG_NFS_V3_ACL is not set +# CONFIG_NFS_V4 is not set +# CONFIG_NFS_DIRECTIO is not set +# CONFIG_NFSD is not set +CONFIG_ROOT_NFS=y +CONFIG_LOCKD=y +CONFIG_LOCKD_V4=y +CONFIG_NFS_COMMON=y +CONFIG_SUNRPC=y +# CONFIG_RPCSEC_GSS_KRB5 is not set +# CONFIG_RPCSEC_GSS_SPKM3 is not set +# CONFIG_SMB_FS is not set +# CONFIG_CIFS is not set +# CONFIG_NCP_FS is not set +# CONFIG_CODA_FS is not set +# CONFIG_AFS_FS is not set +# CONFIG_9P_FS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y + +# +# Native Language Support +# +CONFIG_NLS=y +CONFIG_NLS_DEFAULT="iso8859-1" +CONFIG_NLS_CODEPAGE_437=y +# CONFIG_NLS_CODEPAGE_737 is not set +# CONFIG_NLS_CODEPAGE_775 is not set +# CONFIG_NLS_CODEPAGE_850 is not set +# CONFIG_NLS_CODEPAGE_852 is not set +# CONFIG_NLS_CODEPAGE_855 is not set +# CONFIG_NLS_CODEPAGE_857 is not set +# CONFIG_NLS_CODEPAGE_860 is not set +# CONFIG_NLS_CODEPAGE_861 is not set +# CONFIG_NLS_CODEPAGE_862 is not set +# CONFIG_NLS_CODEPAGE_863 is not set +# CONFIG_NLS_CODEPAGE_864 is not set +# CONFIG_NLS_CODEPAGE_865 is not set +# CONFIG_NLS_CODEPAGE_866 is not set +# CONFIG_NLS_CODEPAGE_869 is not set +# CONFIG_NLS_CODEPAGE_936 is not set +# CONFIG_NLS_CODEPAGE_950 is not set +# CONFIG_NLS_CODEPAGE_932 is not set +# CONFIG_NLS_CODEPAGE_949 is not set +# CONFIG_NLS_CODEPAGE_874 is not set +# CONFIG_NLS_ISO8859_8 is not set +# CONFIG_NLS_CODEPAGE_1250 is not set +# CONFIG_NLS_CODEPAGE_1251 is not set +# CONFIG_NLS_ASCII is not set +CONFIG_NLS_ISO8859_1=y +# CONFIG_NLS_ISO8859_2 is not set +# CONFIG_NLS_ISO8859_3 is not set +# CONFIG_NLS_ISO8859_4 is not set +# CONFIG_NLS_ISO8859_5 is not set +# CONFIG_NLS_ISO8859_6 is not set +# CONFIG_NLS_ISO8859_7 is not set +# CONFIG_NLS_ISO8859_9 is not set +# CONFIG_NLS_ISO8859_13 is not set +# CONFIG_NLS_ISO8859_14 is not set +# CONFIG_NLS_ISO8859_15 is not set +# CONFIG_NLS_KOI8_R is not set +# CONFIG_NLS_KOI8_U is not set +# CONFIG_NLS_UTF8 is not set + +# +# Profiling support +# +# CONFIG_PROFILING is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +# CONFIG_ENABLE_MUST_CHECK is not set +CONFIG_MAGIC_SYSRQ=y +# CONFIG_UNUSED_SYMBOLS is not set +CONFIG_DEBUG_KERNEL=y +CONFIG_LOG_BUF_SHIFT=14 +CONFIG_DETECT_SOFTLOCKUP=y +# CONFIG_SCHEDSTATS is not set +# CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_RT_MUTEXES is not set +# CONFIG_RT_MUTEX_TESTER is not set +CONFIG_DEBUG_SPINLOCK=y +CONFIG_DEBUG_MUTEXES=y +CONFIG_DEBUG_RWSEMS=y +# CONFIG_DEBUG_SPINLOCK_SLEEP is not set +# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set +# CONFIG_DEBUG_KOBJECT is not set +CONFIG_DEBUG_BUGVERBOSE=y +# CONFIG_DEBUG_INFO is not set +# CONFIG_DEBUG_FS is not set +# CONFIG_DEBUG_VM is not set +# CONFIG_DEBUG_LIST is not set +CONFIG_FRAME_POINTER=y +# CONFIG_UNWIND_INFO is not set +CONFIG_FORCED_INLINING=y +# CONFIG_HEADERS_CHECK is not set +# CONFIG_RCU_TORTURE_TEST is not set +CONFIG_DEBUG_USER=y +# CONFIG_DEBUG_WAITQ is not set +CONFIG_DEBUG_ERRORS=y +CONFIG_DEBUG_LL=y +# CONFIG_DEBUG_ICEDCC is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY is not set + +# +# Cryptographic options +# +# CONFIG_CRYPTO is not set + +# +# Library routines +# +# CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set +CONFIG_CRC32=y +# CONFIG_LIBCRC32C is not set +CONFIG_ZLIB_INFLATE=y +CONFIG_PLIST=y -- cgit v0.10.2 From 9957329800b8b554b1af669bcc6878282338c34e Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 26 Oct 2006 10:27:42 +0100 Subject: [ARM] Add __must_check to uaccess functions Signed-off-by: Russell King diff --git a/include/asm-arm/uaccess.h b/include/asm-arm/uaccess.h index 09ad0ca..5f420a0 100644 --- a/include/asm-arm/uaccess.h +++ b/include/asm-arm/uaccess.h @@ -383,19 +383,19 @@ do { \ #ifdef CONFIG_MMU -extern unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n); -extern unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n); -extern unsigned long __clear_user(void __user *addr, unsigned long n); +extern unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n); +extern unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n); +extern unsigned long __must_check __clear_user(void __user *addr, unsigned long n); #else #define __copy_from_user(to,from,n) (memcpy(to, (void __force *)from, n), 0) #define __copy_to_user(to,from,n) (memcpy((void __force *)to, from, n), 0) #define __clear_user(addr,n) (memset((void __force *)addr, 0, n), 0) #endif -extern unsigned long __strncpy_from_user(char *to, const char __user *from, unsigned long count); -extern unsigned long __strnlen_user(const char __user *s, long n); +extern unsigned long __must_check __strncpy_from_user(char *to, const char __user *from, unsigned long count); +extern unsigned long __must_check __strnlen_user(const char __user *s, long n); -static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) +static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) { if (access_ok(VERIFY_READ, from, n)) n = __copy_from_user(to, from, n); @@ -404,7 +404,7 @@ static inline unsigned long copy_from_user(void *to, const void __user *from, un return n; } -static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) +static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) { if (access_ok(VERIFY_WRITE, to, n)) n = __copy_to_user(to, from, n); @@ -414,14 +414,14 @@ static inline unsigned long copy_to_user(void __user *to, const void *from, unsi #define __copy_to_user_inatomic __copy_to_user #define __copy_from_user_inatomic __copy_from_user -static inline unsigned long clear_user(void __user *to, unsigned long n) +static inline unsigned long __must_check clear_user(void __user *to, unsigned long n) { if (access_ok(VERIFY_WRITE, to, n)) n = __clear_user(to, n); return n; } -static inline long strncpy_from_user(char *dst, const char __user *src, long count) +static inline long __must_check strncpy_from_user(char *dst, const char __user *src, long count) { long res = -EFAULT; if (access_ok(VERIFY_READ, src, 1)) @@ -431,7 +431,7 @@ static inline long strncpy_from_user(char *dst, const char __user *src, long cou #define strlen_user(s) strnlen_user(s, ~0UL >> 1) -static inline long strnlen_user(const char __user *s, long n) +static inline long __must_check strnlen_user(const char __user *s, long n) { unsigned long res = 0; -- cgit v0.10.2 From 75e31aaaf43b53517fd2b36cedc08bd4e4af14d5 Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Wed, 25 Oct 2006 23:07:50 +0100 Subject: [ARM] 3909/1: Disable UWIND_INFO for ARM (again) According to Daniel Jacobowitz, UNWIND_INFO is not useful on ARM, and in fact doesn't even compile. This patch disables the option for ARM. Signed-off-by: Kevin Hilman Signed-off-by: Russell King diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 77491e3..d367910 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -341,7 +341,7 @@ config FRAME_POINTER config UNWIND_INFO bool "Compile the kernel with frame unwind information" - depends on !IA64 && !PARISC + depends on !IA64 && !PARISC && !ARM depends on !MODULES || !(MIPS || PPC || SUPERH || V850) help If you say Y here the resulting kernel image will be slightly larger -- cgit v0.10.2 From e816d71a50a714b532e3965364f3f53c23a53d42 Mon Sep 17 00:00:00 2001 From: Takashi Ohmasa Date: Mon, 23 Oct 2006 08:30:35 +0100 Subject: [ARM] 3899/1: Fix the normalization of the denormal double precision number. The significand should be shifted until the value of bit [62] is 1 to normalize the denormal double number. Signed-off-by: Takashi Ohmasa Signed-off-by: Russell King diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c index 4fc05ee..e44b9ed 100644 --- a/arch/arm/vfp/vfpdouble.c +++ b/arch/arm/vfp/vfpdouble.c @@ -56,7 +56,7 @@ static void vfp_double_normalise_denormal(struct vfp_double *vd) { int bits = 31 - fls(vd->significand >> 32); if (bits == 31) - bits = 62 - fls(vd->significand); + bits = 63 - fls(vd->significand); vfp_double_dump("normalise_denormal: in", vd); -- cgit v0.10.2 From e0f205d9c656da9dad6340f75e33a96014b7d23f Mon Sep 17 00:00:00 2001 From: Takashi Ohmasa Date: Mon, 23 Oct 2006 11:19:40 +0100 Subject: [ARM] 3900/1: Fix VFP Division by Zero exception handling. The SIGFPE signal should be generated if Division by Zero exception is detected. Signed-off-by: Takashi Ohmasa Signed-off-by: Russell King diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c index a657a28..f08eafb 100644 --- a/arch/arm/vfp/vfpmodule.c +++ b/arch/arm/vfp/vfpmodule.c @@ -148,6 +148,7 @@ static void vfp_raise_exceptions(u32 exceptions, u32 inst, u32 fpscr, struct pt_ /* * These are arranged in priority order, least to highest. */ + RAISE(FPSCR_DZC, FPSCR_DZE, FPE_FLTDIV); RAISE(FPSCR_IXC, FPSCR_IXE, FPE_FLTRES); RAISE(FPSCR_UFC, FPSCR_UFE, FPE_FLTUND); RAISE(FPSCR_OFC, FPSCR_OFE, FPE_FLTOVF); -- cgit v0.10.2 From 2ae88149a27cadf2840e0ab8155bef13be285c03 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 28 Oct 2006 10:38:23 -0700 Subject: [PATCH] mm: clean up pagecache allocation - Consolidate page_cache_alloc - Fix splice: only the pagecache pages and filesystem data need to use mapping_gfp_mask. - Fix grab_cache_page_nowait: same as splice, also honour NUMA placement. Signed-off-by: Nick Piggin Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/splice.c b/fs/splice.c index 49fb9f1..8d70595 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -74,7 +74,7 @@ static int page_cache_pipe_buf_steal(struct pipe_inode_info *pipe, wait_on_page_writeback(page); if (PagePrivate(page)) - try_to_release_page(page, mapping_gfp_mask(mapping)); + try_to_release_page(page, GFP_KERNEL); /* * If we succeeded in removing the mapping, set LRU flag @@ -333,7 +333,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, break; error = add_to_page_cache_lru(page, mapping, index, - mapping_gfp_mask(mapping)); + GFP_KERNEL); if (unlikely(error)) { page_cache_release(page); if (error == -EEXIST) @@ -557,7 +557,6 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, { struct file *file = sd->file; struct address_space *mapping = file->f_mapping; - gfp_t gfp_mask = mapping_gfp_mask(mapping); unsigned int offset, this_len; struct page *page; pgoff_t index; @@ -591,7 +590,7 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, goto find_page; page = buf->page; - if (add_to_page_cache(page, mapping, index, gfp_mask)) { + if (add_to_page_cache(page, mapping, index, GFP_KERNEL)) { unlock_page(page); goto find_page; } @@ -613,7 +612,7 @@ find_page: * This will also lock the page */ ret = add_to_page_cache_lru(page, mapping, index, - gfp_mask); + GFP_KERNEL); if (unlikely(ret)) goto out; } diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 64f9509..c3e255b 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -52,19 +52,23 @@ static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask) void release_pages(struct page **pages, int nr, int cold); #ifdef CONFIG_NUMA -extern struct page *page_cache_alloc(struct address_space *x); -extern struct page *page_cache_alloc_cold(struct address_space *x); +extern struct page *__page_cache_alloc(gfp_t gfp); #else +static inline struct page *__page_cache_alloc(gfp_t gfp) +{ + return alloc_pages(gfp, 0); +} +#endif + static inline struct page *page_cache_alloc(struct address_space *x) { - return alloc_pages(mapping_gfp_mask(x), 0); + return __page_cache_alloc(mapping_gfp_mask(x)); } static inline struct page *page_cache_alloc_cold(struct address_space *x) { - return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD, 0); + return __page_cache_alloc(mapping_gfp_mask(x)|__GFP_COLD); } -#endif typedef int filler_t(void *, struct page *); diff --git a/mm/filemap.c b/mm/filemap.c index cb26e33..7b84dc8 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -467,25 +467,15 @@ int add_to_page_cache_lru(struct page *page, struct address_space *mapping, } #ifdef CONFIG_NUMA -struct page *page_cache_alloc(struct address_space *x) +struct page *__page_cache_alloc(gfp_t gfp) { if (cpuset_do_page_mem_spread()) { int n = cpuset_mem_spread_node(); - return alloc_pages_node(n, mapping_gfp_mask(x), 0); + return alloc_pages_node(n, gfp, 0); } - return alloc_pages(mapping_gfp_mask(x), 0); + return alloc_pages(gfp, 0); } -EXPORT_SYMBOL(page_cache_alloc); - -struct page *page_cache_alloc_cold(struct address_space *x) -{ - if (cpuset_do_page_mem_spread()) { - int n = cpuset_mem_spread_node(); - return alloc_pages_node(n, mapping_gfp_mask(x)|__GFP_COLD, 0); - } - return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD, 0); -} -EXPORT_SYMBOL(page_cache_alloc_cold); +EXPORT_SYMBOL(__page_cache_alloc); #endif static int __sleep_on_page_lock(void *word) @@ -826,7 +816,6 @@ struct page * grab_cache_page_nowait(struct address_space *mapping, unsigned long index) { struct page *page = find_get_page(mapping, index); - gfp_t gfp_mask; if (page) { if (!TestSetPageLocked(page)) @@ -834,9 +823,8 @@ grab_cache_page_nowait(struct address_space *mapping, unsigned long index) page_cache_release(page); return NULL; } - gfp_mask = mapping_gfp_mask(mapping) & ~__GFP_FS; - page = alloc_pages(gfp_mask, 0); - if (page && add_to_page_cache_lru(page, mapping, index, gfp_mask)) { + page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); + if (page && add_to_page_cache_lru(page, mapping, index, GFP_KERNEL)) { page_cache_release(page); page = NULL; } -- cgit v0.10.2 From 3bb1a852ab6c9cdf211a2f4a2f502340c8c38eca Mon Sep 17 00:00:00 2001 From: Martin Bligh Date: Sat, 28 Oct 2006 10:38:24 -0700 Subject: [PATCH] vmscan: Fix temp_priority race The temp_priority field in zone is racy, as we can walk through a reclaim path, and just before we copy it into prev_priority, it can be overwritten (say with DEF_PRIORITY) by another reclaimer. The same bug is contained in both try_to_free_pages and balance_pgdat, but it is fixed slightly differently. In balance_pgdat, we keep a separate priority record per zone in a local array. In try_to_free_pages there is no need to do this, as the priority level is the same for all zones that we reclaim from. Impact of this bug is that temp_priority is copied into prev_priority, and setting this artificially high causes reclaimers to set distress artificially low. They then fail to reclaim mapped pages, when they are, in fact, under severe memory pressure (their priority may be as low as 0). This causes the OOM killer to fire incorrectly. From: Andrew Morton __zone_reclaim() isn't modifying zone->prev_priority. But zone->prev_priority is used in the decision whether or not to bring mapped pages onto the inactive list. Hence there's a risk here that __zone_reclaim() will fail because zone->prev_priority ir large (ie: low urgency) and lots of mapped pages end up stuck on the active list. Fix that up by decreasing (ie making more urgent) zone->prev_priority as __zone_reclaim() scans the zone's pages. This bug perhaps explains why ZONE_RECLAIM_PRIORITY was created. It should be possible to remove that now, and to just start out at DEF_PRIORITY? Cc: Nick Piggin Cc: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index ed0762b..e06683e 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -218,13 +218,9 @@ struct zone { * under - it drives the swappiness decision: whether to unmap mapped * pages. * - * temp_priority is used to remember the scanning priority at which - * this zone was successfully refilled to free_pages == pages_high. - * - * Access to both these fields is quite racy even on uniprocessor. But + * Access to both this field is quite racy even on uniprocessor. But * it is expected to average out OK. */ - int temp_priority; int prev_priority; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index f5fc454..ecf853b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2407,7 +2407,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat, zone->zone_pgdat = pgdat; zone->free_pages = 0; - zone->temp_priority = zone->prev_priority = DEF_PRIORITY; + zone->prev_priority = DEF_PRIORITY; zone_pcp_init(zone); INIT_LIST_HEAD(&zone->active_list); diff --git a/mm/vmscan.c b/mm/vmscan.c index f05527b..b32560e 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -723,6 +723,20 @@ done: return nr_reclaimed; } +/* + * We are about to scan this zone at a certain priority level. If that priority + * level is smaller (ie: more urgent) than the previous priority, then note + * that priority level within the zone. This is done so that when the next + * process comes in to scan this zone, it will immediately start out at this + * priority level rather than having to build up its own scanning priority. + * Here, this priority affects only the reclaim-mapped threshold. + */ +static inline void note_zone_scanning_priority(struct zone *zone, int priority) +{ + if (priority < zone->prev_priority) + zone->prev_priority = priority; +} + static inline int zone_is_near_oom(struct zone *zone) { return zone->pages_scanned >= (zone->nr_active + zone->nr_inactive)*3; @@ -972,9 +986,7 @@ static unsigned long shrink_zones(int priority, struct zone **zones, if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; - zone->temp_priority = priority; - if (zone->prev_priority > priority) - zone->prev_priority = priority; + note_zone_scanning_priority(zone, priority); if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; /* Let kswapd poll it */ @@ -1024,7 +1036,6 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; - zone->temp_priority = DEF_PRIORITY; lru_pages += zone->nr_active + zone->nr_inactive; } @@ -1065,13 +1076,22 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) if (!sc.all_unreclaimable) ret = 1; out: + /* + * Now that we've scanned all the zones at this priority level, note + * that level within the zone so that the next thread which performs + * scanning of this zone will immediately start out at this priority + * level. This affects only the decision whether or not to bring + * mapped pages onto the inactive list. + */ + if (priority < 0) + priority = 0; for (i = 0; zones[i] != 0; i++) { struct zone *zone = zones[i]; if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; - zone->prev_priority = zone->temp_priority; + zone->prev_priority = priority; } return ret; } @@ -1111,6 +1131,11 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order) .swap_cluster_max = SWAP_CLUSTER_MAX, .swappiness = vm_swappiness, }; + /* + * temp_priority is used to remember the scanning priority at which + * this zone was successfully refilled to free_pages == pages_high. + */ + int temp_priority[MAX_NR_ZONES]; loop_again: total_scanned = 0; @@ -1118,11 +1143,8 @@ loop_again: sc.may_writepage = !laptop_mode; count_vm_event(PAGEOUTRUN); - for (i = 0; i < pgdat->nr_zones; i++) { - struct zone *zone = pgdat->node_zones + i; - - zone->temp_priority = DEF_PRIORITY; - } + for (i = 0; i < pgdat->nr_zones; i++) + temp_priority[i] = DEF_PRIORITY; for (priority = DEF_PRIORITY; priority >= 0; priority--) { int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */ @@ -1183,10 +1205,9 @@ scan: if (!zone_watermark_ok(zone, order, zone->pages_high, end_zone, 0)) all_zones_ok = 0; - zone->temp_priority = priority; - if (zone->prev_priority > priority) - zone->prev_priority = priority; + temp_priority[i] = priority; sc.nr_scanned = 0; + note_zone_scanning_priority(zone, priority); nr_reclaimed += shrink_zone(priority, zone, &sc); reclaim_state->reclaimed_slab = 0; nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL, @@ -1226,10 +1247,15 @@ scan: break; } out: + /* + * Note within each zone the priority level at which this zone was + * brought into a happy state. So that the next thread which scans this + * zone will start out at that priority level. + */ for (i = 0; i < pgdat->nr_zones; i++) { struct zone *zone = pgdat->node_zones + i; - zone->prev_priority = zone->temp_priority; + zone->prev_priority = temp_priority[i]; } if (!all_zones_ok) { cond_resched(); @@ -1614,6 +1640,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) */ priority = ZONE_RECLAIM_PRIORITY; do { + note_zone_scanning_priority(zone, priority); nr_reclaimed += shrink_zone(priority, zone, &sc); priority--; } while (priority >= 0 && nr_reclaimed < nr_pages); diff --git a/mm/vmstat.c b/mm/vmstat.c index 45b124e..8614e8f 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -587,11 +587,9 @@ static int zoneinfo_show(struct seq_file *m, void *arg) seq_printf(m, "\n all_unreclaimable: %u" "\n prev_priority: %i" - "\n temp_priority: %i" "\n start_pfn: %lu", zone->all_unreclaimable, zone->prev_priority, - zone->temp_priority, zone->zone_start_pfn); spin_unlock_irqrestore(&zone->lock, flags); seq_putc(m, '\n'); -- cgit v0.10.2 From bbdb396a60b2ebf7de3b717991e5d3e28c8b7bbd Mon Sep 17 00:00:00 2001 From: Martin Bligh Date: Sat, 28 Oct 2006 10:38:25 -0700 Subject: [PATCH] Use min of two prio settings in calculating distress for reclaim If try_to_free_pages / balance_pgdat are called with a gfp_mask specifying GFP_IO and/or GFP_FS, they will reclaim the requisite number of pages, and the reset prev_priority to DEF_PRIORITY (or to some other high (ie: unurgent) value). However, another reclaimer without those gfp_mask flags set (say, GFP_NOIO) may still be struggling to reclaim pages. The concurrent overwrite of zone->prev_priority will cause this GFP_NOIO thread to unexpectedly cease deactivating mapped pages, thus causing reclaim difficulties. Fix this is to key the distress calculation not off zone->prev_priority, but also take into account the local caller's priority by using min(zone->prev_priority, sc->priority) Signed-off-by: Martin J. Bligh Cc: Nick Piggin Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/vmscan.c b/mm/vmscan.c index b32560e..518540a 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -760,7 +760,7 @@ static inline int zone_is_near_oom(struct zone *zone) * But we had to alter page->flags anyway. */ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, - struct scan_control *sc) + struct scan_control *sc, int priority) { unsigned long pgmoved; int pgdeactivate = 0; @@ -784,7 +784,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, * `distress' is a measure of how much trouble we're having * reclaiming pages. 0 -> no problems. 100 -> great trouble. */ - distress = 100 >> zone->prev_priority; + distress = 100 >> min(zone->prev_priority, priority); /* * The point of this algorithm is to decide when to start @@ -936,7 +936,7 @@ static unsigned long shrink_zone(int priority, struct zone *zone, nr_to_scan = min(nr_active, (unsigned long)sc->swap_cluster_max); nr_active -= nr_to_scan; - shrink_active_list(nr_to_scan, zone, sc); + shrink_active_list(nr_to_scan, zone, sc, priority); } if (nr_inactive) { @@ -1384,7 +1384,7 @@ static unsigned long shrink_all_zones(unsigned long nr_pages, int pass, if (zone->nr_scan_active >= nr_pages || pass > 3) { zone->nr_scan_active = 0; nr_to_scan = min(nr_pages, zone->nr_active); - shrink_active_list(nr_to_scan, zone, sc); + shrink_active_list(nr_to_scan, zone, sc, prio); } } -- cgit v0.10.2 From 1939e49a0cb9d73785857bf312f4f65661b4b513 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 28 Oct 2006 10:38:26 -0700 Subject: [PATCH] ext4: fix printk format warnings fs/ext4/resize.c:72: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:76: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:81: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:85: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:89: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:89: warning: long long unsigned int format, __u64 arg (arg 5) fs/ext4/resize.c:93: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:93: warning: long long unsigned int format, __u64 arg (arg 5) fs/ext4/resize.c:98: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:103: warning: long long unsigned int format, __u64 arg (arg 4) fs/ext4/resize.c:109: warning: long long unsigned int format, __u64 arg (arg 4) Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 1e95780..4fe49c3 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -69,44 +69,49 @@ static int verify_group_input(struct super_block *sb, else if (outside(input->block_bitmap, start, end)) ext4_warning(sb, __FUNCTION__, "Block bitmap not in group (block %llu)", - input->block_bitmap); + (unsigned long long)input->block_bitmap); else if (outside(input->inode_bitmap, start, end)) ext4_warning(sb, __FUNCTION__, "Inode bitmap not in group (block %llu)", - input->inode_bitmap); + (unsigned long long)input->inode_bitmap); else if (outside(input->inode_table, start, end) || outside(itend - 1, start, end)) ext4_warning(sb, __FUNCTION__, "Inode table not in group (blocks %llu-%llu)", - input->inode_table, itend - 1); + (unsigned long long)input->inode_table, itend - 1); else if (input->inode_bitmap == input->block_bitmap) ext4_warning(sb, __FUNCTION__, "Block bitmap same as inode bitmap (%llu)", - input->block_bitmap); + (unsigned long long)input->block_bitmap); else if (inside(input->block_bitmap, input->inode_table, itend)) ext4_warning(sb, __FUNCTION__, "Block bitmap (%llu) in inode table (%llu-%llu)", - input->block_bitmap, input->inode_table, itend-1); + (unsigned long long)input->block_bitmap, + (unsigned long long)input->inode_table, itend - 1); else if (inside(input->inode_bitmap, input->inode_table, itend)) ext4_warning(sb, __FUNCTION__, "Inode bitmap (%llu) in inode table (%llu-%llu)", - input->inode_bitmap, input->inode_table, itend-1); + (unsigned long long)input->inode_bitmap, + (unsigned long long)input->inode_table, itend - 1); else if (inside(input->block_bitmap, start, metaend)) ext4_warning(sb, __FUNCTION__, "Block bitmap (%llu) in GDT table" " (%llu-%llu)", - input->block_bitmap, start, metaend - 1); + (unsigned long long)input->block_bitmap, + start, metaend - 1); else if (inside(input->inode_bitmap, start, metaend)) ext4_warning(sb, __FUNCTION__, "Inode bitmap (%llu) in GDT table" " (%llu-%llu)", - input->inode_bitmap, start, metaend - 1); + (unsigned long long)input->inode_bitmap, + start, metaend - 1); else if (inside(input->inode_table, start, metaend) || inside(itend - 1, start, metaend)) ext4_warning(sb, __FUNCTION__, "Inode table (%llu-%llu) overlaps" "GDT table (%llu-%llu)", - input->inode_table, itend - 1, start, metaend - 1); + (unsigned long long)input->inode_table, + itend - 1, start, metaend - 1); else err = 0; brelse(bh); -- cgit v0.10.2 From f58a74dca88d48b0669609b4957f3dd757bdc898 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Sat, 28 Oct 2006 10:38:27 -0700 Subject: [PATCH] jbd: journal_dirty_data re-check for unmapped buffers When running several fsx's and other filesystem stress tests, we found cases where an unmapped buffer was still being sent to submit_bh by the ext3 dirty data journaling code. I saw this happen in two ways, both related to another thread doing a truncate which would unmap the buffer in question. Either we would get into journal_dirty_data with a bh which was already unmapped (although journal_dirty_data_fn had checked for this earlier, the state was not locked at that point), or it would get unmapped in the middle of journal_dirty_data when we dropped locks to call sync_dirty_buffer. By re-checking for mapped state after we've acquired the bh state lock, we should avoid these races. If we find a buffer which is no longer mapped, we essentially ignore it, because journal_unmap_buffer has already decided that this buffer can go away. I've also added tracepoints in these two cases, and made a couple other tracepoint changes that I found useful in debugging this. Signed-off-by: Eric Sandeen Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index d5c6304..4f82bcd6 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -967,6 +967,13 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) */ jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); + + /* Now that we have bh_state locked, are we really still mapped? */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "unmapped buffer, bailing out"); + goto no_journal; + } + if (jh->b_transaction) { JBUFFER_TRACE(jh, "has transaction"); if (jh->b_transaction != handle->h_transaction) { @@ -1028,6 +1035,11 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) sync_dirty_buffer(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); + /* Since we dropped the lock... */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "buffer got unmapped"); + goto no_journal; + } /* The buffer may become locked again at any time if it is redirtied */ } @@ -1824,6 +1836,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) } } } else if (transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "on committing transaction"); if (jh->b_jlist == BJ_Locked) { /* * The buffer is on the committing transaction's locked @@ -1838,7 +1851,6 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * can remove it's next_transaction pointer from the * running transaction if that is set, but nothing * else. */ - JBUFFER_TRACE(jh, "on committing transaction"); set_buffer_freed(bh); if (jh->b_next_transaction) { J_ASSERT(jh->b_next_transaction == @@ -1858,6 +1870,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * i_size already for this truncate so recovery will not * expose the disk blocks we are discarding here.) */ J_ASSERT_JH(jh, transaction == journal->j_running_transaction); + JBUFFER_TRACE(jh, "on running transaction"); may_free = __dispose_buffer(jh, transaction); } -- cgit v0.10.2 From 9b57988db9b2c81794546cb792133f0cfd064ea8 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Sat, 28 Oct 2006 10:38:28 -0700 Subject: [PATCH] jbd2: journal_dirty_data re-check for unmapped buffers When running several fsx's and other filesystem stress tests, we found cases where an unmapped buffer was still being sent to submit_bh by the ext3 dirty data journaling code. I saw this happen in two ways, both related to another thread doing a truncate which would unmap the buffer in question. Either we would get into journal_dirty_data with a bh which was already unmapped (although journal_dirty_data_fn had checked for this earlier, the state was not locked at that point), or it would get unmapped in the middle of journal_dirty_data when we dropped locks to call sync_dirty_buffer. By re-checking for mapped state after we've acquired the bh state lock, we should avoid these races. If we find a buffer which is no longer mapped, we essentially ignore it, because journal_unmap_buffer has already decided that this buffer can go away. I've also added tracepoints in these two cases, and made a couple other tracepoint changes that I found useful in debugging this. Signed-off-by: Eric Sandeen Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index b6cf2be..c051a94 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -967,6 +967,13 @@ int jbd2_journal_dirty_data(handle_t *handle, struct buffer_head *bh) */ jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); + + /* Now that we have bh_state locked, are we really still mapped? */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "unmapped buffer, bailing out"); + goto no_journal; + } + if (jh->b_transaction) { JBUFFER_TRACE(jh, "has transaction"); if (jh->b_transaction != handle->h_transaction) { @@ -1028,6 +1035,11 @@ int jbd2_journal_dirty_data(handle_t *handle, struct buffer_head *bh) sync_dirty_buffer(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); + /* Since we dropped the lock... */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "buffer got unmapped"); + goto no_journal; + } /* The buffer may become locked again at any time if it is redirtied */ } @@ -1824,6 +1836,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) } } } else if (transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "on committing transaction"); if (jh->b_jlist == BJ_Locked) { /* * The buffer is on the committing transaction's locked @@ -1838,7 +1851,6 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * can remove it's next_transaction pointer from the * running transaction if that is set, but nothing * else. */ - JBUFFER_TRACE(jh, "on committing transaction"); set_buffer_freed(bh); if (jh->b_next_transaction) { J_ASSERT(jh->b_next_transaction == @@ -1858,6 +1870,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * i_size already for this truncate so recovery will not * expose the disk blocks we are discarding here.) */ J_ASSERT_JH(jh, transaction == journal->j_running_transaction); + JBUFFER_TRACE(jh, "on running transaction"); may_free = __dispose_buffer(jh, transaction); } -- cgit v0.10.2 From ae74589cb381cc2838107f92d4e0e1f178c6825d Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Sat, 28 Oct 2006 10:38:29 -0700 Subject: [PATCH] fix efi_memory_present_wrapper() efi_memory_present_wrapper() parameter start/end is physical address, but function memory_present parameter is PFN, this patch converts physical address to PFN. Signed-off-by: bibo, mao Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c index 519e63c..141041d 100644 --- a/arch/i386/kernel/setup.c +++ b/arch/i386/kernel/setup.c @@ -846,7 +846,7 @@ efi_find_max_pfn(unsigned long start, unsigned long end, void *arg) static int __init efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg) { - memory_present(0, start, end); + memory_present(0, PFN_UP(start), PFN_DOWN(end)); return 0; } -- cgit v0.10.2 From 2b6e845986347ef86729e8651908af3e8a8441f5 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 28 Oct 2006 10:38:30 -0700 Subject: [PATCH] md: fix bug where spares don't always get rebuilt properly when they become live If save_raid_disk is >= 0, then the device could be a device that is already in sync that is being re-added. So we need to default this value to -1. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index 7daa7b1..c40ce9f 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2003,6 +2003,7 @@ static mdk_rdev_t *md_import_device(dev_t newdev, int super_format, int super_mi kobject_init(&rdev->kobj); rdev->desc_nr = -1; + rdev->saved_raid_disk = -1; rdev->flags = 0; rdev->data_offset = 0; rdev->sb_events = 0; -- cgit v0.10.2 From 01ab5662f573fe3a6bcefa200f15ab3069cec8a3 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 28 Oct 2006 10:38:30 -0700 Subject: [PATCH] md: simplify checking of available size when resizing an array When "mdadm --grow --size=xxx" is used to resize an array (use more or less of each device), we check the new siza against the available space in each device. We already have that number recorded in rdev->size, so calculating it is pointless (and wrong in one obscure case). Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index c40ce9f..50ab4a9 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -4045,11 +4045,8 @@ static int update_size(mddev_t *mddev, unsigned long size) return -EBUSY; ITERATE_RDEV(mddev,rdev,tmp) { sector_t avail; - if (rdev->sb_offset > rdev->data_offset) - avail = (rdev->sb_offset*2) - rdev->data_offset; - else - avail = get_capacity(rdev->bdev->bd_disk) - - rdev->data_offset; + avail = rdev->size * 2; + if (fit && (size == 0 || size > avail/2)) size = avail/2; if (avail < ((sector_t)size << 1)) -- cgit v0.10.2 From 750a8f3e8f64654a584e54038c2c8db380813c79 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 28 Oct 2006 10:38:31 -0700 Subject: [PATCH] md: fix up maintenance of ->degraded in multipath A recent fix which made sure ->degraded was initialised properly exposed a second bug - ->degraded wasn't been updated when drives failed or were hot-added. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index a6260f0..14da37f 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -277,6 +277,7 @@ static void multipath_error (mddev_t *mddev, mdk_rdev_t *rdev) set_bit(Faulty, &rdev->flags); set_bit(MD_CHANGE_DEVS, &mddev->flags); conf->working_disks--; + mddev->degraded++; printk(KERN_ALERT "multipath: IO failure on %s," " disabling IO path. \n Operation continuing" " on %d IO paths.\n", @@ -336,6 +337,7 @@ static int multipath_add_disk(mddev_t *mddev, mdk_rdev_t *rdev) blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); conf->working_disks++; + mddev->degraded--; rdev->raid_disk = path; set_bit(In_sync, &rdev->flags); rcu_assign_pointer(p->rdev, rdev); -- cgit v0.10.2 From 969b755aadf7bcf3df5991a127a103acd0145a52 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 28 Oct 2006 10:38:32 -0700 Subject: [PATCH] md: fix printk format warnings, seen on powerpc64: drivers/md/raid1.c:1479: warning: long long unsigned int format, long unsigned int arg (arg 4) drivers/md/raid10.c:1475: warning: long long unsigned int format, long unsigned int arg (arg 4) Signed-off-by: Randy Dunlap Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index dc9d2de..656fae9 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1474,8 +1474,8 @@ static void fix_read_error(conf_t *conf, int read_disk, "raid1:%s: read error corrected " "(%d sectors at %llu on %s)\n", mdname(mddev), s, - (unsigned long long)sect + - rdev->data_offset, + (unsigned long long)(sect + + rdev->data_offset), bdevname(rdev->bdev, b)); } } diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 74f17a9..7492d60 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -1470,8 +1470,8 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio) "raid10:%s: read error corrected" " (%d sectors at %llu on %s)\n", mdname(mddev), s, - (unsigned long long)sect+ - rdev->data_offset, + (unsigned long long)(sect+ + rdev->data_offset), bdevname(rdev->bdev, b)); rdev_dec_pending(rdev, mddev); -- cgit v0.10.2 From f2d0aa5bf8d4f7ae4cb1a7feebf5b1afddd0b9b0 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Sat, 28 Oct 2006 10:38:32 -0700 Subject: [PATCH] memory hotplug: __GFP_NOWARN is better for __kmalloc_section_memmap() Add __GFP_NOWARN flag to calling of __alloc_pages() in __kmalloc_section_memmap(). It can reduce noisy failure message. In ia64, section size is 1 GB, this means that order 8 pages are necessary for each section's memmap. It is often very hard requirement under heavy memory pressure as you know. So, __alloc_pages() gives up allocation and shows many noisy stack traces which means no page for each sections. (Current my environment shows 32 times of stack trace....) But, __kmalloc_section_memmap() calls vmalloc() after failure of it, and it can succeed allocation of memmap. So, its stack trace warning becomes just noisy. I suppose it shouldn't be shown. Signed-off-by: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/sparse.c b/mm/sparse.c index 86c52ab..b3c82ba 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -211,7 +211,7 @@ static struct page *__kmalloc_section_memmap(unsigned long nr_pages) struct page *page, *ret; unsigned long memmap_size = sizeof(struct page) * nr_pages; - page = alloc_pages(GFP_KERNEL, get_order(memmap_size)); + page = alloc_pages(GFP_KERNEL|__GFP_NOWARN, get_order(memmap_size)); if (page) goto got_map_page; -- cgit v0.10.2 From 6a2aae06cc1e87e9712a26a639f6a2f3442e2027 Mon Sep 17 00:00:00 2001 From: Pavel Emelianov Date: Sat, 28 Oct 2006 10:38:33 -0700 Subject: [PATCH] Fix potential OOPs in blkdev_open() blkdev_open() calls bc_acquire() to get a struct block_device. Since bc_acquire() may return NULL when system is out of memory an appropriate check is required. Signed-off-by: Pavel Emelianov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/block_dev.c b/fs/block_dev.c index bc8f27c..702b88c 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1131,6 +1131,8 @@ static int blkdev_open(struct inode * inode, struct file * filp) filp->f_flags |= O_LARGEFILE; bdev = bd_acquire(inode); + if (bdev == NULL) + return -ENOMEM; res = do_open(bdev, filp, BD_MUTEX_NORMAL); if (res) -- cgit v0.10.2 From 52fd24ca1db3a741f144bbc229beefe044202cac Mon Sep 17 00:00:00 2001 From: Giridhar Pemmasani Date: Sat, 28 Oct 2006 10:38:34 -0700 Subject: [PATCH] __vmalloc with GFP_ATOMIC causes 'sleeping from invalid context' If __vmalloc is called to allocate memory with GFP_ATOMIC in atomic context, the chain of calls results in __get_vm_area_node allocating memory for vm_struct with GFP_KERNEL, causing the 'sleeping from invalid context' warning. This patch fixes it by passing the gfp flags along so __get_vm_area_node allocates memory for vm_struct with the same flags. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index ce5f148..dc9a29d 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -60,7 +60,8 @@ extern struct vm_struct *get_vm_area(unsigned long size, unsigned long flags); extern struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, unsigned long start, unsigned long end); extern struct vm_struct *get_vm_area_node(unsigned long size, - unsigned long flags, int node); + unsigned long flags, int node, + gfp_t gfp_mask); extern struct vm_struct *remove_vm_area(void *addr); extern int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages); diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 1133dd3..6d381df 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -160,13 +160,15 @@ int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages) return err; } -struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, - unsigned long start, unsigned long end, int node) +static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, + unsigned long start, unsigned long end, + int node, gfp_t gfp_mask) { struct vm_struct **p, *tmp, *area; unsigned long align = 1; unsigned long addr; + BUG_ON(in_interrupt()); if (flags & VM_IOREMAP) { int bit = fls(size); @@ -180,7 +182,7 @@ struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, addr = ALIGN(start, align); size = PAGE_ALIGN(size); - area = kmalloc_node(sizeof(*area), GFP_KERNEL, node); + area = kmalloc_node(sizeof(*area), gfp_mask, node); if (unlikely(!area)) return NULL; @@ -236,7 +238,7 @@ out: struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, unsigned long start, unsigned long end) { - return __get_vm_area_node(size, flags, start, end, -1); + return __get_vm_area_node(size, flags, start, end, -1, GFP_KERNEL); } /** @@ -253,9 +255,11 @@ struct vm_struct *get_vm_area(unsigned long size, unsigned long flags) return __get_vm_area(size, flags, VMALLOC_START, VMALLOC_END); } -struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags, int node) +struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags, + int node, gfp_t gfp_mask) { - return __get_vm_area_node(size, flags, VMALLOC_START, VMALLOC_END, node); + return __get_vm_area_node(size, flags, VMALLOC_START, VMALLOC_END, node, + gfp_mask); } /* Caller must hold vmlist_lock */ @@ -487,7 +491,7 @@ static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot, if (!size || (size >> PAGE_SHIFT) > num_physpages) return NULL; - area = get_vm_area_node(size, VM_ALLOC, node); + area = get_vm_area_node(size, VM_ALLOC, node, gfp_mask); if (!area) return NULL; -- cgit v0.10.2 From 08d892f11aae7125fe078cf93ec5cf6af288c5e7 Mon Sep 17 00:00:00 2001 From: Andrey Panin Date: Sat, 28 Oct 2006 10:38:35 -0700 Subject: [PATCH] visws build fix Fix this: > Subject : CONFIG_X86_VISWS=3Dy, CONFIG_SMP=3Dn compile error > References : http://lkml.org/lkml/2006/10/7/51 > Submitter : Jesper Juhl > Caused-By : David Howells > commit 7d12e780e003f93433d49ce78cfedf4b4c52adc5 > Status : unknown Via undescribed means. Signed-off-by: Andrey Panin Cc: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/mach-visws/visws_apic.c b/arch/i386/mach-visws/visws_apic.c index 07097ed..38c2b13 100644 --- a/arch/i386/mach-visws/visws_apic.c +++ b/arch/i386/mach-visws/visws_apic.c @@ -122,7 +122,7 @@ static void end_cobalt_irq(unsigned int irq) spin_unlock_irqrestore(&cobalt_lock, flags); } -static struct hw_interrupt_type cobalt_irq_type = { +static struct irq_chip cobalt_irq_type = { .typename = "Cobalt-APIC", .startup = startup_cobalt_irq, .shutdown = disable_cobalt_irq, @@ -159,7 +159,7 @@ static void end_piix4_master_irq(unsigned int irq) spin_unlock_irqrestore(&cobalt_lock, flags); } -static struct hw_interrupt_type piix4_master_irq_type = { +static struct irq_chip piix4_master_irq_type = { .typename = "PIIX4-master", .startup = startup_piix4_master_irq, .ack = ack_cobalt_irq, @@ -167,9 +167,8 @@ static struct hw_interrupt_type piix4_master_irq_type = { }; -static struct hw_interrupt_type piix4_virtual_irq_type = { +static struct irq_chip piix4_virtual_irq_type = { .typename = "PIIX4-virtual", - .startup = startup_8259A_irq, .shutdown = disable_8259A_irq, .enable = enable_8259A_irq, .disable = disable_8259A_irq, diff --git a/include/asm-i386/mach-visws/do_timer.h b/include/asm-i386/mach-visws/do_timer.h deleted file mode 100644 index 21cd696..0000000 --- a/include/asm-i386/mach-visws/do_timer.h +++ /dev/null @@ -1,53 +0,0 @@ -/* defines for inline arch setup functions */ - -#include -#include -#include "cobalt.h" - -static inline void do_timer_interrupt_hook(void) -{ - /* Clear the interrupt */ - co_cpu_write(CO_CPU_STAT,co_cpu_read(CO_CPU_STAT) & ~CO_STAT_TIMEINTR); - - do_timer(1); -#ifndef CONFIG_SMP - update_process_times(user_mode_vm(irq_regs)); -#endif -/* - * In the SMP case we use the local APIC timer interrupt to do the - * profiling, except when we simulate SMP mode on a uniprocessor - * system, in that case we have to call the local interrupt handler. - */ -#ifndef CONFIG_X86_LOCAL_APIC - profile_tick(CPU_PROFILING); -#else - if (!using_apic_timer) - smp_local_timer_interrupt(); -#endif -} - -static inline int do_timer_overflow(int count) -{ - int i; - - spin_lock(&i8259A_lock); - /* - * This is tricky when I/O APICs are used; - * see do_timer_interrupt(). - */ - i = inb(0x20); - spin_unlock(&i8259A_lock); - - /* assumption about timer being IRQ0 */ - if (i & 0x01) { - /* - * We cannot detect lost timer interrupts ... - * well, that's why we call them lost, don't we? :) - * [hmm, on the Pentium and Alpha we can ... sort of] - */ - count -= LATCH; - } else { - printk("do_slow_gettimeoffset(): hardware timer problem?\n"); - } - return count; -} diff --git a/include/asm-i386/mach-visws/mach_apic.h b/include/asm-i386/mach-visws/mach_apic.h index de438c71..18afe6b 100644 --- a/include/asm-i386/mach-visws/mach_apic.h +++ b/include/asm-i386/mach-visws/mach_apic.h @@ -51,6 +51,11 @@ static inline void clustered_apic_check(void) { } +static inline int apicid_to_node(int logical_apicid) +{ + return 0; +} + /* Mapping from cpu number to logical apicid */ static inline int cpu_to_logical_apicid(int cpu) { -- cgit v0.10.2 From 1d4d262769cd1894a0306b9c57e72f005cd9e75a Mon Sep 17 00:00:00 2001 From: Jan Dittmer Date: Sat, 28 Oct 2006 10:38:38 -0700 Subject: [PATCH] Add missing space in module.c for taintskernel Obvious fix. Signed-off-by: Jan Dittmer Acked-by: Florin Malita Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/module.c b/kernel/module.c index 67009bd..5072a94 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1342,7 +1342,7 @@ static void set_license(struct module *mod, const char *license) if (!license_is_gpl_compatible(license)) { if (!(tainted & TAINT_PROPRIETARY_MODULE)) - printk(KERN_WARNING "%s: module license '%s' taints" + printk(KERN_WARNING "%s: module license '%s' taints " "kernel.\n", mod->name, license); add_taint_module(mod, TAINT_PROPRIETARY_MODULE); } -- cgit v0.10.2 From 760fe9ad1692361770bb56fa5c69cf6b3354858c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 28 Oct 2006 10:38:39 -0700 Subject: [PATCH] ioc4: fix printk format warning Fix printk format warning: drivers/misc/ioc4.c:213: warning: long long int format, u64 arg (arg 3) Signed-off-by: Randy Dunlap Acked-by: Brent Casavant Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/misc/ioc4.c b/drivers/misc/ioc4.c index 79354bb..b995a15 100644 --- a/drivers/misc/ioc4.c +++ b/drivers/misc/ioc4.c @@ -210,8 +210,8 @@ ioc4_clock_calibrate(struct ioc4_driver_data *idd) do_div(ns, IOC4_EXTINT_COUNT_DIVISOR); printk(KERN_DEBUG - "IOC4 %s: PCI clock is %lld ns.\n", - pci_name(idd->idd_pdev), ns); + "IOC4 %s: PCI clock is %llu ns.\n", + pci_name(idd->idd_pdev), (unsigned long long)ns); } /* Remember results. We store the extint clock period rather -- cgit v0.10.2 From 7b92aadfdae85ef837db343be38d4172115be72b Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 28 Oct 2006 10:38:40 -0700 Subject: [PATCH] cciss: fix printk format warning Fix printk format warnings: drivers/block/cciss.c:2000: warning: long long int format, long unsigned int arg (arg 2) drivers/block/cciss.c:2035: warning: long long int format, long unsigned int arg (arg 2) Signed-off-by: Randy Dunlap Acked-by: Mike Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index bc66026..6ffe2b2 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -1992,8 +1992,8 @@ cciss_read_capacity(int ctlr, int logvol, int withirq, sector_t *total_size, *block_size = BLOCK_SIZE; } if (*total_size != (__u32) 0) - printk(KERN_INFO " blocks= %lld block_size= %d\n", - *total_size, *block_size); + printk(KERN_INFO " blocks= %llu block_size= %d\n", + (unsigned long long)*total_size, *block_size); kfree(buf); return; } @@ -2027,8 +2027,8 @@ cciss_read_capacity_16(int ctlr, int logvol, int withirq, sector_t *total_size, *total_size = 0; *block_size = BLOCK_SIZE; } - printk(KERN_INFO " blocks= %lld block_size= %d\n", - *total_size, *block_size); + printk(KERN_INFO " blocks= %llu block_size= %d\n", + (unsigned long long)*total_size, *block_size); kfree(buf); return; } -- cgit v0.10.2 From b9d7e6ae82da124dc9c579fe1061264ef2a69407 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Sat, 28 Oct 2006 10:38:41 -0700 Subject: [PATCH] hugetlb: fix size=4G parsing On 32-bit machines, mount -t hugetlbfs -o size=4G gave a 0GB filesystem, size=5G gave a 1GB filesystem etc: there's no point in masking size with HPAGE_MASK just before shifting its lower bits away, and since HPAGE_MASK is a UL, that removed all the higher bits of the unsigned long long size. Signed-off-by: Hugh Dickins Cc: Adam Litke Cc: David Gibson Cc: "Chen, Kenneth W" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 4ee3f00..0b23b96 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -624,7 +624,6 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) do_div(size, 100); rest++; } - size &= HPAGE_MASK; pconfig->nr_blocks = (size >> HPAGE_SHIFT); value = rest; } else if (!strcmp(opt,"nr_inodes")) { -- cgit v0.10.2 From 856fc29505556cf263f3dcda2533cf3766c14ab6 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Sat, 28 Oct 2006 10:38:43 -0700 Subject: [PATCH] hugetlb: fix prio_tree unit hugetlb_vmtruncate_list was misconverted to prio_tree: its prio_tree is in units of PAGE_SIZE (PAGE_CACHE_SIZE) like any other, not HPAGE_SIZE (whereas its radix_tree is kept in units of HPAGE_SIZE, otherwise slots would be absurdly sparse). At first I thought the error benign, just calling __unmap_hugepage_range on more vmas than necessary; but on 32-bit machines, when the prio_tree is searched correctly, it happens to ensure the v_offset calculation won't overflow. As it stood, when truncating at or beyond 4GB, it was liable to discard pages COWed from lower offsets; or even to clear pmd entries of preceding vmas, triggering exit_mmap's BUG_ON(nr_ptes). Signed-off-by: Hugh Dickins Cc: Adam Litke Cc: David Gibson Cc: "Chen, Kenneth W" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 0b23b96..0bea6a6 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -271,26 +271,24 @@ static void hugetlbfs_drop_inode(struct inode *inode) hugetlbfs_forget_inode(inode); } -/* - * h_pgoff is in HPAGE_SIZE units. - * vma->vm_pgoff is in PAGE_SIZE units. - */ static inline void -hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) +hugetlb_vmtruncate_list(struct prio_tree_root *root, pgoff_t pgoff) { struct vm_area_struct *vma; struct prio_tree_iter iter; - vma_prio_tree_foreach(vma, &iter, root, h_pgoff, ULONG_MAX) { - unsigned long h_vm_pgoff; + vma_prio_tree_foreach(vma, &iter, root, pgoff, ULONG_MAX) { unsigned long v_offset; - h_vm_pgoff = vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT); - v_offset = (h_pgoff - h_vm_pgoff) << HPAGE_SHIFT; /* - * Is this VMA fully outside the truncation point? + * Can the expression below overflow on 32-bit arches? + * No, because the prio_tree returns us only those vmas + * which overlap the truncated area starting at pgoff, + * and no vma on a 32-bit arch can span beyond the 4GB. */ - if (h_vm_pgoff >= h_pgoff) + if (vma->vm_pgoff < pgoff) + v_offset = (pgoff - vma->vm_pgoff) << PAGE_SHIFT; + else v_offset = 0; __unmap_hugepage_range(vma, @@ -303,14 +301,14 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) */ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) { - unsigned long pgoff; + pgoff_t pgoff; struct address_space *mapping = inode->i_mapping; if (offset > inode->i_size) return -EINVAL; BUG_ON(offset & ~HPAGE_MASK); - pgoff = offset >> HPAGE_SHIFT; + pgoff = offset >> PAGE_SHIFT; inode->i_size = offset; spin_lock(&mapping->i_mmap_lock); -- cgit v0.10.2 From ebed4bfc8da8df5b6b0bc4a5064a949f04683509 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Sat, 28 Oct 2006 10:38:43 -0700 Subject: [PATCH] hugetlb: fix absurd HugePages_Rsvd If you truncated an mmap'ed hugetlbfs file, then faulted on the truncated area, /proc/meminfo's HugePages_Rsvd wrapped hugely "negative". Reinstate my preliminary i_size check before attempting to allocate the page (though this only fixes the most obvious case: more work will be needed here). Signed-off-by: Hugh Dickins Cc: Adam Litke Cc: David Gibson Cc: "Chen, Kenneth W" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 2dbec90..a088f59 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -478,6 +478,9 @@ int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, retry: page = find_lock_page(mapping, idx); if (!page) { + size = i_size_read(mapping->host) >> HPAGE_SHIFT; + if (idx >= size) + goto out; if (hugetlb_get_quota(mapping)) goto out; page = alloc_huge_page(vma, address); -- cgit v0.10.2 From 6eac3f93f5e6b7256fb20b7608d62ec192da12be Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Sat, 28 Oct 2006 10:38:44 -0700 Subject: [PATCH] missing unused dentry in prune_dcache()? On the the following patch: http://linux.bkbits.net:8080/linux-2.6/gnupatch@449b144ecSF1rYskg3q-SeR2vf88zg # ChangeSet # 2006/06/22 15:05:57-07:00 neilb@suse.de # [PATCH] Fix dcache race during umount # If prune_dcache finds a dentry that it cannot free, it leaves it where it # is (at the tail of the list) and exits, on the assumption that some other # thread will be removing that dentry soon. However as far as I see this comment is not correct: when we cannot take s_umount rw_semaphore (for example because it was taken in do_remount) this dentry is already extracted from dentry_unused list and we do not add it into the list again. Therefore dentry will not be found by prune_dcache() and shrink_dcache_sb() and will leave in memory very long time until the partition will be unmounted. The patch adds this dentry into tail of the dentry_unused list. Signed-off-by: Vasily Averin Cc: Neil Brown Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/dcache.c b/fs/dcache.c index a1ff91e..a5b76b6 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -478,11 +478,12 @@ static void prune_dcache(int count, struct super_block *sb) up_read(s_umount); } spin_unlock(&dentry->d_lock); - /* Cannot remove the first dentry, and it isn't appropriate - * to move it to the head of the list, so give up, and try - * later + /* + * Insert dentry at the head of the list as inserting at the + * tail leads to a cycle. */ - break; + list_add(&dentry->d_lru, &dentry_unused); + dentry_stat.nr_unused++; } spin_unlock(&dcache_lock); } -- cgit v0.10.2 From f87135762de4328c6f17897e803e6909bc056feb Mon Sep 17 00:00:00 2001 From: David Howells Date: Sat, 28 Oct 2006 10:38:46 -0700 Subject: [PATCH] VFS: Fix an error in unused dentry counting With Vasily Averin Fix an error in unused dentry counting in shrink_dcache_for_umount_subtree() in which the count is modified without the dcache_lock held. Signed-off-by: David Howells Cc: Vasily Averin Cc: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/dcache.c b/fs/dcache.c index a5b76b6..fd4a428 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -557,6 +557,7 @@ repeat: static void shrink_dcache_for_umount_subtree(struct dentry *dentry) { struct dentry *parent; + unsigned detached = 0; BUG_ON(!IS_ROOT(dentry)); @@ -621,7 +622,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) atomic_dec(&parent->d_count); list_del(&dentry->d_u.d_child); - dentry_stat.nr_dentry--; /* For d_free, below */ + detached++; inode = dentry->d_inode; if (inode) { @@ -639,7 +640,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) * otherwise we ascend to the parent and move to the * next sibling if there is one */ if (!parent) - return; + goto out; dentry = parent; @@ -648,6 +649,11 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) dentry = list_entry(dentry->d_subdirs.next, struct dentry, d_u.d_child); } +out: + /* several dentries were freed, need to correct nr_dentry */ + spin_lock(&dcache_lock); + dentry_stat.nr_dentry -= detached; + spin_unlock(&dcache_lock); } /* -- cgit v0.10.2 From 5fa3839a64203b2ab727dcb37da9b2d7079fca28 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sat, 28 Oct 2006 10:38:46 -0700 Subject: [PATCH] Constify compat_get_bitmap argument This means we can call it when the bitmap we want to fetch is declared const. Signed-off-by: Stephen Rothwell Cc: Christoph Lameter Cc: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/compat.h b/include/linux/compat.h index f4ebf96..f155319 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -196,7 +196,7 @@ asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp, #define BITS_TO_COMPAT_LONGS(bits) \ (((bits)+BITS_PER_COMPAT_LONG-1)/BITS_PER_COMPAT_LONG) -long compat_get_bitmap(unsigned long *mask, compat_ulong_t __user *umask, +long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, unsigned long bitmap_size); long compat_put_bitmap(compat_ulong_t __user *umask, unsigned long *mask, unsigned long bitmap_size); diff --git a/kernel/compat.c b/kernel/compat.c index 75573e5..d4898aa 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -678,7 +678,7 @@ int get_compat_sigevent(struct sigevent *event, ? -EFAULT : 0; } -long compat_get_bitmap(unsigned long *mask, compat_ulong_t __user *umask, +long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, unsigned long bitmap_size) { int i, j; -- cgit v0.10.2 From 6e6d9fa6f95e382bb2d5725dda18b9e811418e79 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sat, 28 Oct 2006 10:38:47 -0700 Subject: [PATCH] strstrip remove last blank fix strstrip() does not remove the last blank from strings which only consist of blanks. Example: char string[] = " "; strstrip(string); results in " ", but should produce an empty string! The following patch solves this problem: Acked-by: Martin Schwidefsky Signed-off-by: Michael Holzheu Acked-by: Pekka Enberg Acked-by Joern Engel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/string.c b/lib/string.c index 6307726..a485d75 100644 --- a/lib/string.c +++ b/lib/string.c @@ -320,7 +320,7 @@ char *strstrip(char *s) return s; end = s + size - 1; - while (end != s && isspace(*end)) + while (end >= s && isspace(*end)) end--; *(end + 1) = '\0'; -- cgit v0.10.2 From fca178c0c6e8d52a1875be36b070f30884ebfae9 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:49 -0700 Subject: [PATCH] fill_tgid: fix task_struct leak and possible oops 1. fill_tgid() forgets to do put_task_struct(first). 2. release_task(first) can happen after fill_tgid() drops tasklist_lock, it is unsafe to dereference first->signal. This is a temporary fix, imho the locking should be reworked. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 5d6a8c5..9aeee51 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -237,14 +237,17 @@ static int fill_tgid(pid_t tgid, struct task_struct *tgidtsk, } else get_task_struct(first); - /* Start with stats from dead tasks */ - spin_lock_irqsave(&first->signal->stats_lock, flags); - if (first->signal->stats) - memcpy(stats, first->signal->stats, sizeof(*stats)); - spin_unlock_irqrestore(&first->signal->stats_lock, flags); tsk = first; read_lock(&tasklist_lock); + /* Start with stats from dead tasks */ + if (first->signal) { + spin_lock_irqsave(&first->signal->stats_lock, flags); + if (first->signal->stats) + memcpy(stats, first->signal->stats, sizeof(*stats)); + spin_unlock_irqrestore(&first->signal->stats_lock, flags); + } + do { if (tsk->exit_state == EXIT_ZOMBIE && thread_group_leader(tsk)) continue; @@ -264,7 +267,7 @@ static int fill_tgid(pid_t tgid, struct task_struct *tgidtsk, * Accounting subsytems can also add calls here to modify * fields of taskstats. */ - + put_task_struct(first); return 0; } -- cgit v0.10.2 From 05d5bcd60e8202e5c7b28cf61186043a4d612623 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:50 -0700 Subject: [PATCH] bacct_add_tsk: fix unsafe and wrong parent/group_leader dereference 1. ts = timespec_sub(uptime, current->group_leader->start_time); It is possible that current != tsk. Probably it was supposed to be 'tsk->group_leader->start_time. But why we are reading group_leader's start_time ? This accounting is per thread, not per procees, I changed this to 'tsk->start_time. Please corect me. 2. stats->ac_ppid = (tsk->parent) ? tsk->parent->pid : 0; tsk->parent never == NULL, and it is unsafe to dereference it. Both the task and it's parent may exit after the caller unlocks tasklist_lock, the memory could be unmapped (DEBUG_SLAB). (And we should use ->real_parent->tgid in fact). Q: I don't understand the 'if (thread_group_leader(tsk))' check. Why it is needed ? Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Acked-by: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/tsacct.c b/kernel/tsacct.c index db44322..65a5036 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -36,7 +36,7 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) /* calculate task elapsed time in timespec */ do_posix_clock_monotonic_gettime(&uptime); - ts = timespec_sub(uptime, current->group_leader->start_time); + ts = timespec_sub(uptime, tsk->start_time); /* rebase elapsed time to usec */ ac_etime = timespec_to_ns(&ts); do_div(ac_etime, NSEC_PER_USEC); @@ -58,7 +58,10 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) stats->ac_uid = tsk->uid; stats->ac_gid = tsk->gid; stats->ac_pid = tsk->pid; - stats->ac_ppid = (tsk->parent) ? tsk->parent->pid : 0; + rcu_read_lock(); + stats->ac_ppid = pid_alive(tsk) ? + rcu_dereference(tsk->real_parent)->tgid : 0; + rcu_read_unlock(); stats->ac_utime = cputime_to_msecs(tsk->utime) * USEC_PER_MSEC; stats->ac_stime = cputime_to_msecs(tsk->stime) * USEC_PER_MSEC; stats->ac_minflt = tsk->min_flt; -- cgit v0.10.2 From 093a8e8aecd77b2799934996a55a6838e1e2b8f3 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:51 -0700 Subject: [PATCH] taskstats_tgid_free: fix usage taskstats_tgid_free() is called on copy_process's error path. This is wrong. IF (clone_flags & CLONE_THREAD) We should not clear ->signal->taskstats, current uses it, it probably has a valid accumulated info. ELSE taskstats_tgid_init() set ->signal->taskstats = NULL, there is nothing to free. Move the callsite to __exit_signal(). We don't need any locking, entire thread group is exiting, nobody should have a reference to soon to be released ->signal. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index 16894b7..a437ca0 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -49,17 +49,8 @@ static inline void taskstats_tgid_alloc(struct signal_struct *sig) static inline void taskstats_tgid_free(struct signal_struct *sig) { - struct taskstats *stats = NULL; - unsigned long flags; - - spin_lock_irqsave(&sig->stats_lock, flags); - if (sig->stats) { - stats = sig->stats; - sig->stats = NULL; - } - spin_unlock_irqrestore(&sig->stats_lock, flags); - if (stats) - kmem_cache_free(taskstats_cache, stats); + if (sig->stats) + kmem_cache_free(taskstats_cache, sig->stats); } extern void taskstats_exit_alloc(struct taskstats **, unsigned int *); diff --git a/kernel/exit.c b/kernel/exit.c index f250a5e..06de6c4 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -128,6 +128,7 @@ static void __exit_signal(struct task_struct *tsk) flush_sigqueue(&tsk->pending); if (sig) { flush_sigqueue(&sig->shared_pending); + taskstats_tgid_free(sig); __cleanup_signal(sig); } } diff --git a/kernel/fork.c b/kernel/fork.c index 29ebb30..2133266 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -897,7 +897,6 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts void __cleanup_signal(struct signal_struct *sig) { exit_thread_group_keys(sig); - taskstats_tgid_free(sig); kmem_cache_free(signal_cachep, sig); } -- cgit v0.10.2 From 17b02695b254aa2ef0e53df9c8e6548f86e66a9d Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:52 -0700 Subject: [PATCH] taskstats_tgid_alloc: optimization Every subthread (except first) does unneeded kmem_cache_alloc/kmem_cache_free. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index a437ca0..6642240 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -32,6 +32,9 @@ static inline void taskstats_tgid_alloc(struct signal_struct *sig) struct taskstats *stats; unsigned long flags; + if (sig->stats != NULL) + return; + stats = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); if (!stats) return; -- cgit v0.10.2 From b8534d7bd89df0cd41cd47bcd6733a05ea9a691a Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:53 -0700 Subject: [PATCH] taskstats: kill ->taskstats_lock in favor of ->siglock signal_struct is (mostly) protected by ->sighand->siglock, I think we don't need ->taskstats_lock to protect ->stats. This also allows us to simplify the locking in fill_tgid(). Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/sched.h b/include/linux/sched.h index 6735c1c..eafe4a7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -466,7 +466,6 @@ struct signal_struct { struct pacct_struct pacct; /* per-process accounting information */ #endif #ifdef CONFIG_TASKSTATS - spinlock_t stats_lock; struct taskstats *stats; #endif }; diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index 6642240..6562a20 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -23,28 +23,26 @@ static inline void taskstats_exit_free(struct taskstats *tidstats) static inline void taskstats_tgid_init(struct signal_struct *sig) { - spin_lock_init(&sig->stats_lock); sig->stats = NULL; } -static inline void taskstats_tgid_alloc(struct signal_struct *sig) +static inline void taskstats_tgid_alloc(struct task_struct *tsk) { + struct signal_struct *sig = tsk->signal; struct taskstats *stats; - unsigned long flags; if (sig->stats != NULL) return; + /* No problem if kmem_cache_zalloc() fails */ stats = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); - if (!stats) - return; - spin_lock_irqsave(&sig->stats_lock, flags); + spin_lock_irq(&tsk->sighand->siglock); if (!sig->stats) { sig->stats = stats; stats = NULL; } - spin_unlock_irqrestore(&sig->stats_lock, flags); + spin_unlock_irq(&tsk->sighand->siglock); if (stats) kmem_cache_free(taskstats_cache, stats); @@ -59,7 +57,6 @@ static inline void taskstats_tgid_free(struct signal_struct *sig) extern void taskstats_exit_alloc(struct taskstats **, unsigned int *); extern void taskstats_exit_send(struct task_struct *, struct taskstats *, int, unsigned int); extern void taskstats_init_early(void); -extern void taskstats_tgid_alloc(struct signal_struct *); #else static inline void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) {} @@ -71,7 +68,7 @@ static inline void taskstats_exit_send(struct task_struct *tsk, {} static inline void taskstats_tgid_init(struct signal_struct *sig) {} -static inline void taskstats_tgid_alloc(struct signal_struct *sig) +static inline void taskstats_tgid_alloc(struct task_struct *tsk) {} static inline void taskstats_tgid_free(struct signal_struct *sig) {} diff --git a/kernel/fork.c b/kernel/fork.c index 2133266..3da978e 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -830,7 +830,7 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts if (clone_flags & CLONE_THREAD) { atomic_inc(¤t->signal->count); atomic_inc(¤t->signal->live); - taskstats_tgid_alloc(current->signal); + taskstats_tgid_alloc(current); return 0; } sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 9aeee51..b2efda9 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -241,11 +241,11 @@ static int fill_tgid(pid_t tgid, struct task_struct *tgidtsk, tsk = first; read_lock(&tasklist_lock); /* Start with stats from dead tasks */ - if (first->signal) { - spin_lock_irqsave(&first->signal->stats_lock, flags); + if (first->sighand) { + spin_lock_irqsave(&first->sighand->siglock, flags); if (first->signal->stats) memcpy(stats, first->signal->stats, sizeof(*stats)); - spin_unlock_irqrestore(&first->signal->stats_lock, flags); + spin_unlock_irqrestore(&first->sighand->siglock, flags); } do { @@ -276,7 +276,7 @@ static void fill_tgid_exit(struct task_struct *tsk) { unsigned long flags; - spin_lock_irqsave(&tsk->signal->stats_lock, flags); + spin_lock_irqsave(&tsk->sighand->siglock, flags); if (!tsk->signal->stats) goto ret; @@ -288,7 +288,7 @@ static void fill_tgid_exit(struct task_struct *tsk) */ delayacct_add_tsk(tsk->signal->stats, tsk); ret: - spin_unlock_irqrestore(&tsk->signal->stats_lock, flags); + spin_unlock_irqrestore(&tsk->sighand->siglock, flags); return; } @@ -464,15 +464,10 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, size_t size; int is_thread_group; struct nlattr *na; - unsigned long flags; if (!family_registered || !tidstats) return; - spin_lock_irqsave(&tsk->signal->stats_lock, flags); - is_thread_group = tsk->signal->stats ? 1 : 0; - spin_unlock_irqrestore(&tsk->signal->stats_lock, flags); - rc = 0; /* * Size includes space for nested attributes @@ -480,6 +475,7 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); + is_thread_group = (tsk->signal->stats != NULL); if (is_thread_group) size = 2 * size; /* PID + STATS + TGID + STATS */ -- cgit v0.10.2 From a98b6094261c0112e9c455c96995972181bff049 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:54 -0700 Subject: [PATCH] taskstats: don't use tasklist_lock Remove tasklist_lock from taskstats.c. find_task_by_pid() is rcu-safe. ->siglock allows us to traverse subthread without tasklist. Q: delay accounting looks wrong to me. If sub-thread has already called taskstats_exit_send() but didn't call release_task(self) yet it will be accounted twice. The window is big. No? Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index b2efda9..b724aee 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -174,21 +174,19 @@ static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) up_write(&listeners->sem); } -static int fill_pid(pid_t pid, struct task_struct *pidtsk, +static int fill_pid(pid_t pid, struct task_struct *tsk, struct taskstats *stats) { int rc = 0; - struct task_struct *tsk = pidtsk; - if (!pidtsk) { - read_lock(&tasklist_lock); + if (!tsk) { + rcu_read_lock(); tsk = find_task_by_pid(pid); - if (!tsk) { - read_unlock(&tasklist_lock); + if (tsk) + get_task_struct(tsk); + rcu_read_unlock(); + if (!tsk) return -ESRCH; - } - get_task_struct(tsk); - read_unlock(&tasklist_lock); } else get_task_struct(tsk); @@ -214,40 +212,28 @@ static int fill_pid(pid_t pid, struct task_struct *pidtsk, } -static int fill_tgid(pid_t tgid, struct task_struct *tgidtsk, +static int fill_tgid(pid_t tgid, struct task_struct *first, struct taskstats *stats) { - struct task_struct *tsk, *first; + struct task_struct *tsk; unsigned long flags; + int rc = -ESRCH; /* * Add additional stats from live tasks except zombie thread group * leaders who are already counted with the dead tasks */ - first = tgidtsk; - if (!first) { - read_lock(&tasklist_lock); + rcu_read_lock(); + if (!first) first = find_task_by_pid(tgid); - if (!first) { - read_unlock(&tasklist_lock); - return -ESRCH; - } - get_task_struct(first); - read_unlock(&tasklist_lock); - } else - get_task_struct(first); + if (!first || !lock_task_sighand(first, &flags)) + goto out; - tsk = first; - read_lock(&tasklist_lock); - /* Start with stats from dead tasks */ - if (first->sighand) { - spin_lock_irqsave(&first->sighand->siglock, flags); - if (first->signal->stats) - memcpy(stats, first->signal->stats, sizeof(*stats)); - spin_unlock_irqrestore(&first->sighand->siglock, flags); - } + if (first->signal->stats) + memcpy(stats, first->signal->stats, sizeof(*stats)); + tsk = first; do { if (tsk->exit_state == EXIT_ZOMBIE && thread_group_leader(tsk)) continue; @@ -260,15 +246,18 @@ static int fill_tgid(pid_t tgid, struct task_struct *tgidtsk, delayacct_add_tsk(stats, tsk); } while_each_thread(first, tsk); - read_unlock(&tasklist_lock); - stats->version = TASKSTATS_VERSION; + unlock_task_sighand(first, &flags); + rc = 0; +out: + rcu_read_unlock(); + + stats->version = TASKSTATS_VERSION; /* * Accounting subsytems can also add calls here to modify * fields of taskstats. */ - put_task_struct(first); - return 0; + return rc; } -- cgit v0.10.2 From d7c3f5f231c60d7e6ada5770b536df2b3ec1bd08 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:54 -0700 Subject: [PATCH] fill_tgid: cleanup delays accounting fill_tgid() should skip not only an already exited group leader. If the task has ->exit_state != 0 it already did exit_notify(), so it also did fill_tgid_exit()->delayacct_add_tsk(->signal->stats) and we should skip it to avoid a double accounting. This patch doesn't close the race completely, but it cleanups the code. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index b724aee..8adfb80 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -235,7 +235,7 @@ static int fill_tgid(pid_t tgid, struct task_struct *first, tsk = first; do { - if (tsk->exit_state == EXIT_ZOMBIE && thread_group_leader(tsk)) + if (tsk->exit_state) continue; /* * Accounting subsystem can call its functions here to -- cgit v0.10.2 From eba6cd671427df295c10b54ee69cd5de419d38fe Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 28 Oct 2006 10:38:55 -0700 Subject: [PATCH] move SYS_HYPERVISOR inside the Generic Driver menu Put SYS_HYPERVISOR inside the Generic Driver Config menu where it should be. Otherwise xconfig displays it as a dangling (lost) menu item under Device Drivers, all by itself (when all options are displayed). Signed-off-by: Randy Dunlap Cc: Cc: Martin Schwidefsky Cc: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 0b4e224..1429f3a2 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -37,8 +37,8 @@ config DEBUG_DRIVER If you are unsure about this, say N here. -endmenu - config SYS_HYPERVISOR bool default n + +endmenu -- cgit v0.10.2 From bb1d860551c4307b1a7ee9a21b120319075e987e Mon Sep 17 00:00:00 2001 From: Jim Houston Date: Sat, 28 Oct 2006 10:38:56 -0700 Subject: [PATCH] time_adjust cleared before use I notice that the code which implements adjtime clears the time_adjust value before using it. The attached patch makes the obvious fix. Acked-by: Roman Zippel Signed-off-by: Jim Houston Cc: John Stultz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 47195fa..3afeaa3 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -161,9 +161,9 @@ void second_overflow(void) time_adjust += MAX_TICKADJ; tick_length -= MAX_TICKADJ_SCALED; } else { - time_adjust = 0; tick_length += (s64)(time_adjust * NSEC_PER_USEC / HZ) << TICK_LENGTH_SHIFT; + time_adjust = 0; } } } -- cgit v0.10.2 From 8fa1d7d3b2c51594c0f3aa151983dd51f605e07d Mon Sep 17 00:00:00 2001 From: Satoru Takeuchi Date: Sat, 28 Oct 2006 10:38:57 -0700 Subject: [PATCH] cpu-hotplug: release `workqueue_mutex' properly on CPU hot-remove _cpu_down() acquires `workqueue_mutex' on its process, but doen't release it if __cpu_disable() fails. Signed-off-by: Satoru Takeuchi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/cpu.c b/kernel/cpu.c index 27dd3ee..663c920 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -150,18 +150,18 @@ static int _cpu_down(unsigned int cpu) p = __stop_machine_run(take_cpu_down, NULL, cpu); mutex_unlock(&cpu_bitmask_lock); - if (IS_ERR(p)) { + if (IS_ERR(p) || cpu_online(cpu)) { /* CPU didn't die: tell everyone. Can't complain. */ if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, (void *)(long)cpu) == NOTIFY_BAD) BUG(); - err = PTR_ERR(p); - goto out_allowed; - } - - if (cpu_online(cpu)) + if (IS_ERR(p)) { + err = PTR_ERR(p); + goto out_allowed; + } goto out_thread; + } /* Wait for it to sleep (leaving idle task). */ while (!idle_cpu(cpu)) -- cgit v0.10.2 From c333526f489044be2b93085720eb898f0037b346 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Sat, 28 Oct 2006 10:38:57 -0700 Subject: [PATCH] JMB 368 PATA detection The Jmicron JMB368 is PATA only so has the PATA on function zero. Don't therefore skip function zero on this device when probing Signed-off-by: Alan Cox Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c index ad418ce..e72ab36 100644 --- a/drivers/ide/pci/generic.c +++ b/drivers/ide/pci/generic.c @@ -247,8 +247,10 @@ static int __devinit generic_init_one(struct pci_dev *dev, const struct pci_devi (!(PCI_FUNC(dev->devfn) & 1))) goto out; - if (dev->vendor == PCI_VENDOR_ID_JMICRON && PCI_FUNC(dev->devfn) != 1) - goto out; + if (dev->vendor == PCI_VENDOR_ID_JMICRON) { + if (dev->device != PCI_DEVICE_ID_JMICRON_JMB368 && PCI_FUNC(dev->devfn) != 1) + goto out; + } if (dev->vendor != PCI_VENDOR_ID_JMICRON) { pci_read_config_word(dev, PCI_COMMAND, &command); -- cgit v0.10.2 From 057647fc47b3a5fbcfa997041db3f483d506603c Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sat, 28 Oct 2006 10:38:58 -0700 Subject: [PATCH] workqueue: update kerneldoc This patch (as812) changes the kerneldoc comments explaining the return values from queue_work(), queue_delayed_work(), and queue_delayed_work_on(). The updated comments explain more accurately the meaning of the return code and avoid suggesting that a 0 value means the routine was unsuccessful. Signed-off-by: Alan Stern Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 3df9bfc..17c2f03 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -99,7 +99,7 @@ static void __queue_work(struct cpu_workqueue_struct *cwq, * @wq: workqueue to use * @work: work to queue * - * Returns non-zero if it was successfully added. + * Returns 0 if @work was already on a queue, non-zero otherwise. * * We queue the work to the CPU it was submitted, but there is no * guarantee that it will be processed by that CPU. @@ -138,7 +138,7 @@ static void delayed_work_timer_fn(unsigned long __data) * @work: work to queue * @delay: number of jiffies to wait before queueing * - * Returns non-zero if it was successfully added. + * Returns 0 if @work was already on a queue, non-zero otherwise. */ int fastcall queue_delayed_work(struct workqueue_struct *wq, struct work_struct *work, unsigned long delay) @@ -169,7 +169,7 @@ EXPORT_SYMBOL_GPL(queue_delayed_work); * @work: work to queue * @delay: number of jiffies to wait before queueing * - * Returns non-zero if it was successfully added. + * Returns 0 if @work was already on a queue, non-zero otherwise. */ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, struct work_struct *work, unsigned long delay) -- cgit v0.10.2 From 0c6cb974636dd29681b03f8eb0ae227decab01fb Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Sat, 28 Oct 2006 10:38:59 -0700 Subject: [PATCH] Calculation fix for memory holes beyong the end of physical memory absent_pages_in_range() made the assumption that users of the arch-independent zone-sizing API would not care about holes beyound the end of physical memory. This was not the case and was "fixed" in a patch called "Account for holes that are outside the range of physical memory". However, when given a range that started before a hole in "real" memory and ended beyond the end of memory, it would get the result wrong. The bug is in mainline but a patch is below. It has been tested successfully on a number of machines and architectures. Additional credit to Keith Mannthey for discovering the problem, helping identify the correct fix and confirming it Worked For Him. Signed-off-by: Mel Gorman Cc: keith mannthey Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ecf853b..b55bb35 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2261,7 +2261,7 @@ unsigned long __init __absent_pages_in_range(int nid, /* Account for ranges past physical memory on this node */ if (range_end_pfn > prev_end_pfn) - hole_pages = range_end_pfn - + hole_pages += range_end_pfn - max(range_start_pfn, prev_end_pfn); return hole_pages; -- cgit v0.10.2 From 84b5abe69ff600a559e1a1fa29f1edad707d4e2f Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 28 Oct 2006 22:30:17 +0100 Subject: [ARM] Fix i2c-pxa slave mode support i2c-pxa times out when trying to enable slave mode due to an incorrect test. Also, check that i2c->slave is non-NULL before dereferencing it. Signed-off-by: Russell King diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c index 81050d3..c95a6c1 100644 --- a/drivers/i2c/busses/i2c-pxa.c +++ b/drivers/i2c/busses/i2c-pxa.c @@ -272,7 +272,8 @@ static int i2c_pxa_wait_slave(struct pxa_i2c *i2c) dev_dbg(&i2c->adap.dev, "%s: %ld: ISR=%08x, ICR=%08x, IBMR=%02x\n", __func__, (long)jiffies, ISR, ICR, IBMR); - if ((ISR & (ISR_UB|ISR_IBB|ISR_SAD)) == ISR_SAD || + if ((ISR & (ISR_UB|ISR_IBB)) == 0 || + (ISR & ISR_SAD) != 0 || (ICR & ICR_SCLE) == 0) { if (i2c_debug > 1) dev_dbg(&i2c->adap.dev, "%s: done\n", __func__); @@ -492,7 +493,10 @@ static void i2c_pxa_slave_txempty(struct pxa_i2c *i2c, u32 isr) if (isr & ISR_BED) { /* what should we do here? */ } else { - int ret = i2c->slave->read(i2c->slave->data); + int ret = 0; + + if (i2c->slave != NULL) + ret = i2c->slave->read(i2c->slave->data); IDBR = ret; ICR |= ICR_TB; /* allow next byte */ -- cgit v0.10.2 From 9468613b2bb0a386af563953b613efc6c77bd8c1 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 28 Oct 2006 22:42:56 +0100 Subject: [ARM] Fix suspend oops caused by PXA2xx PCMCIA driver The PXA2xx PCMCIA driver was registering a device_driver with the platform_bus_type. Unfortunately, this causes data outside the device_driver structure to be dereferenced as if it were a platform_driver structure, causing an oops. Convert the PXA2xx core driver to use the proper platform_driver structure. Signed-off-by: Russell King diff --git a/drivers/pcmcia/pxa2xx_base.c b/drivers/pcmcia/pxa2xx_base.c index b351813..dca9f85 100644 --- a/drivers/pcmcia/pxa2xx_base.c +++ b/drivers/pcmcia/pxa2xx_base.c @@ -166,7 +166,7 @@ pxa2xx_pcmcia_frequency_change(struct soc_pcmcia_socket *skt, } #endif -int pxa2xx_drv_pcmcia_probe(struct device *dev) +int __pxa2xx_drv_pcmcia_probe(struct device *dev) { int ret; struct pcmcia_low_level *ops; @@ -203,35 +203,52 @@ int pxa2xx_drv_pcmcia_probe(struct device *dev) return ret; } -EXPORT_SYMBOL(pxa2xx_drv_pcmcia_probe); +EXPORT_SYMBOL(__pxa2xx_drv_pcmcia_probe); -static int pxa2xx_drv_pcmcia_resume(struct device *dev) + +static int pxa2xx_drv_pcmcia_probe(struct platform_device *dev) +{ + return __pxa2xx_drv_pcmcia_probe(&dev->dev); +} + +static int pxa2xx_drv_pcmcia_remove(struct platform_device *dev) +{ + return soc_common_drv_pcmcia_remove(&dev->dev); +} + +static int pxa2xx_drv_pcmcia_suspend(struct platform_device *dev, pm_message_t state) +{ + return pcmcia_socket_dev_suspend(&dev->dev, state); +} + +static int pxa2xx_drv_pcmcia_resume(struct platform_device *dev) { - struct pcmcia_low_level *ops = dev->platform_data; + struct pcmcia_low_level *ops = dev->dev.platform_data; int nr = ops ? ops->nr : 0; MECR = nr > 1 ? MECR_CIT | MECR_NOS : (nr > 0 ? MECR_CIT : 0); - return pcmcia_socket_dev_resume(dev); + return pcmcia_socket_dev_resume(&dev->dev); } -static struct device_driver pxa2xx_pcmcia_driver = { +static struct platform_driver pxa2xx_pcmcia_driver = { .probe = pxa2xx_drv_pcmcia_probe, - .remove = soc_common_drv_pcmcia_remove, - .suspend = pcmcia_socket_dev_suspend, + .remove = pxa2xx_drv_pcmcia_remove, + .suspend = pxa2xx_drv_pcmcia_suspend, .resume = pxa2xx_drv_pcmcia_resume, - .name = "pxa2xx-pcmcia", - .bus = &platform_bus_type, + .driver = { + .name = "pxa2xx-pcmcia", + }, }; static int __init pxa2xx_pcmcia_init(void) { - return driver_register(&pxa2xx_pcmcia_driver); + return platform_driver_register(&pxa2xx_pcmcia_driver); } static void __exit pxa2xx_pcmcia_exit(void) { - driver_unregister(&pxa2xx_pcmcia_driver); + platform_driver_unregister(&pxa2xx_pcmcia_driver); } fs_initcall(pxa2xx_pcmcia_init); diff --git a/drivers/pcmcia/pxa2xx_base.h b/drivers/pcmcia/pxa2xx_base.h index e46cff3..235d681 100644 --- a/drivers/pcmcia/pxa2xx_base.h +++ b/drivers/pcmcia/pxa2xx_base.h @@ -1,3 +1,3 @@ /* temporary measure */ -extern int pxa2xx_drv_pcmcia_probe(struct device *); +extern int __pxa2xx_drv_pcmcia_probe(struct device *); diff --git a/drivers/pcmcia/pxa2xx_lubbock.c b/drivers/pcmcia/pxa2xx_lubbock.c index fd1f691..a92f111 100644 --- a/drivers/pcmcia/pxa2xx_lubbock.c +++ b/drivers/pcmcia/pxa2xx_lubbock.c @@ -260,7 +260,7 @@ int __init pcmcia_lubbock_init(struct sa1111_dev *sadev) lubbock_set_misc_wr((1 << 15) | (1 << 14), 0); sadev->dev.platform_data = &lubbock_pcmcia_ops; - ret = pxa2xx_drv_pcmcia_probe(&sadev->dev); + ret = __pxa2xx_drv_pcmcia_probe(&sadev->dev); } return ret; -- cgit v0.10.2 From 51342d7126342e1f1a40cd96a5798a3cfd6136eb Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 29 Oct 2006 12:51:05 +0000 Subject: [ARM] Add KBUILD_IMAGE target support Add support for KBUILD_IMAGE on ARM. This takes the usual target specifiers (zImage/Image/etc) in the same way that powerpc does (iow, without the arch/arm/boot prefix). Signed-off-by: Russell King diff --git a/arch/arm/Makefile b/arch/arm/Makefile index 2a0b2c8..6f4f8bf 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -174,11 +174,13 @@ libs-y := arch/arm/lib/ $(libs-y) # Default target when executing plain make ifeq ($(CONFIG_XIP_KERNEL),y) -all: xipImage +KBUILD_IMAGE := xipImage else -all: zImage +KBUILD_IMAGE := zImage endif +all: $(KBUILD_IMAGE) + boot := arch/arm/boot # Update machine arch and proc symlinks if something which affects -- cgit v0.10.2 From 5211e6e6c671f0d4b1e1a1023384d20227d8ee65 Mon Sep 17 00:00:00 2001 From: Giridhar Pemmasani Date: Sun, 29 Oct 2006 04:46:55 -0800 Subject: [PATCH] Fix GFP_HIGHMEM slab panic As reported by Martin J. Bligh , we let through some non-slab bits to slab allocation through __get_vm_area_node when doing a vmalloc. I haven't been able to reproduce this, although I understand why it happens: vmalloc allocates memory with GFP_KERNEL | __GFP_HIGHMEM and commit 52fd24ca1db3a741f144bbc229beefe044202cac resulted in the same flags are passed down to cache_alloc_refill, causing the BUG. The following patch fixes it. Note that when calling kmalloc_node, I am masking off __GFP_HIGHMEM with GFP_LEVEL_MASK, whereas __vmalloc_area_node does the same with ~(__GFP_HIGHMEM | __GFP_ZERO). IMHO, using GFP_LEVEL_MASK is preferable, but either should fix this problem. Signed-off-by: Giridhar Pemmasani (pgiri@yahoo.com) Cc: Martin J. Bligh Cc: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 6d381df..46606c1 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -182,7 +182,7 @@ static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long fl addr = ALIGN(start, align); size = PAGE_ALIGN(size); - area = kmalloc_node(sizeof(*area), gfp_mask, node); + area = kmalloc_node(sizeof(*area), gfp_mask & GFP_LEVEL_MASK, node); if (unlikely(!area)) return NULL; -- cgit v0.10.2 From 811c9a4b618ffc7de097490f8dc61b43b8464b9c Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Sun, 29 Oct 2006 14:15:10 +0100 Subject: [ARM] 3913/1: n2100: fix IRQ routing for second ethernet port The second ethernet port on the Thecus n2100 was incorrectly assigned to XINT1 instead of the correct XINT3 (PCI INTB instead of INTD), which caused that port to be non-functional. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King diff --git a/arch/arm/mach-iop32x/n2100.c b/arch/arm/mach-iop32x/n2100.c index a2c94a4..2499a77 100644 --- a/arch/arm/mach-iop32x/n2100.c +++ b/arch/arm/mach-iop32x/n2100.c @@ -85,7 +85,7 @@ n2100_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin) irq = IRQ_IOP32X_XINT0; } else if (PCI_SLOT(dev->devfn) == 2) { /* RTL8110SB #2 */ - irq = IRQ_IOP32X_XINT1; + irq = IRQ_IOP32X_XINT3; } else if (PCI_SLOT(dev->devfn) == 3) { /* Sil3512 */ irq = IRQ_IOP32X_XINT2; -- cgit v0.10.2 From 346f5c7ee7fa4ebee0e4c96415a7e59716bfa1d0 Mon Sep 17 00:00:00 2001 From: Stefan Richter Date: Sun, 29 Oct 2006 19:52:49 +0100 Subject: ieee1394: ohci1394: revert fail on error in suspend Some errors during preparation for suspended state can be skipped with a warning instead of a failure of the whole suspend transition, notably an error in pci_set_power_state. Signed-off-by: Stefan Richter diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c index dea1352..6e8ea91 100644 --- a/drivers/ieee1394/ohci1394.c +++ b/drivers/ieee1394/ohci1394.c @@ -3552,12 +3552,21 @@ static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state) { int err; + printk(KERN_INFO "%s does not fully support suspend and resume yet\n", + OHCI1394_DRIVER_NAME); + err = pci_save_state(pdev); - if (err) - goto out; + if (err) { + printk(KERN_ERR "%s: pci_save_state failed with %d\n", + OHCI1394_DRIVER_NAME, err); + return err; + } err = pci_set_power_state(pdev, pci_choose_state(pdev, state)); +#ifdef OHCI1394_DEBUG if (err) - goto out; + printk(KERN_DEBUG "%s: pci_set_power_state failed with %d\n", + OHCI1394_DRIVER_NAME, err); +#endif /* OHCI1394_DEBUG */ /* PowerMac suspend code comes last */ #ifdef CONFIG_PPC_PMAC @@ -3570,8 +3579,8 @@ static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state) pmac_call_feature(PMAC_FTR_1394_ENABLE, of_node, 0, 0); } #endif /* CONFIG_PPC_PMAC */ -out: - return err; + + return 0; } #endif /* CONFIG_PM */ -- cgit v0.10.2 From d46a3d0d07ba539aea5b0e1ad30e568f0cb03576 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Oct 2006 16:45:58 +0300 Subject: [PATCH] taskstats: fix sk_buff leak 'return genlmsg_cancel()' in taskstats_user_cmd/taskstats_exit_send potentially leaks a skb. Unless we pass 'rep_skb' to the netlink layer we own sk_buff. This means we should always do kfree_skb() on failure. [ Thomas acked and pointed out missing return value in original version ] Signed-off-by: Oleg Nesterov Acked-by: Thomas Graf Cc: Andrew Morton Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 8adfb80..f3c3e9d 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -411,7 +411,7 @@ static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) return send_reply(rep_skb, info->snd_pid); nla_put_failure: - return genlmsg_cancel(rep_skb, reply); + rc = genlmsg_cancel(rep_skb, reply); err: nlmsg_free(rep_skb); return rc; @@ -507,7 +507,6 @@ send: nla_put_failure: genlmsg_cancel(rep_skb, reply); - goto ret; err_skb: nlmsg_free(rep_skb); ret: -- cgit v0.10.2 From 3d8334def5cf831d2ed438aae021696a2faa4ddd Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Oct 2006 18:57:16 +0300 Subject: [PATCH] taskstats: fix sk_buff size calculation prepare_reply() adds GENL_HDRLEN to the payload (genlmsg_total_size()), but then it does genlmsg_put()->nlmsg_put(). This means we forget to reserve a room for 'struct nlmsghdr'. Signed-off-by: Oleg Nesterov Cc: Thomas Graf Cc: Andrew Morton Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index f3c3e9d..2039585 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -77,7 +77,8 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, /* * If new attributes are added, please revisit this allocation */ - skb = nlmsg_new(genlmsg_total_size(size), GFP_KERNEL); + size = nlmsg_total_size(genlmsg_total_size(size)); + skb = nlmsg_new(size, GFP_KERNEL); if (!skb) return -ENOMEM; -- cgit v0.10.2 From 5fdb51a10f3ab4320c3170b79d66ce34a6e65a67 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 29 Oct 2006 11:21:51 +0100 Subject: [PATCH] m68k: consolidate initcall sections Commit 61ce1efe6e40233663d27ab8ac9ba9710eebcaad missed the m68k initcall sections. Signed-off-by: Geert Uytterhoeven Signed-off-by: Linus Torvalds diff --git a/arch/m68k/kernel/vmlinux-std.lds b/arch/m68k/kernel/vmlinux-std.lds index 69d1d3d..d279445 100644 --- a/arch/m68k/kernel/vmlinux-std.lds +++ b/arch/m68k/kernel/vmlinux-std.lds @@ -54,13 +54,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; diff --git a/arch/m68k/kernel/vmlinux-sun3.lds b/arch/m68k/kernel/vmlinux-sun3.lds index 65cc39c..2550b4a 100644 --- a/arch/m68k/kernel/vmlinux-sun3.lds +++ b/arch/m68k/kernel/vmlinux-sun3.lds @@ -48,13 +48,7 @@ __init_begin = .; __setup_end = .; __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; -- cgit v0.10.2 From 150ed8ed63b96d7f93ef7e6081797aa0df2b1abd Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 29 Oct 2006 03:43:19 +0900 Subject: [WATCHDOG] sc1200wdt.c pnp unregister fix. If no devices found or invalid parameter is specified, scl200wdt_pnp_driver is left unregistered. It breaks global list of pnp drivers. Signed-off-by: Akinobu Mita Signed-off-by: Wim Van Sebroeck Signed-off-by: Andrew Morton diff --git a/drivers/char/watchdog/sc1200wdt.c b/drivers/char/watchdog/sc1200wdt.c index d8d0f28..e323983 100644 --- a/drivers/char/watchdog/sc1200wdt.c +++ b/drivers/char/watchdog/sc1200wdt.c @@ -392,7 +392,7 @@ static int __init sc1200wdt_init(void) if (io == -1) { printk(KERN_ERR PFX "io parameter must be specified\n"); ret = -EINVAL; - goto out_clean; + goto out_pnp; } #if defined CONFIG_PNP @@ -405,7 +405,7 @@ static int __init sc1200wdt_init(void) if (!request_region(io, io_len, SC1200_MODULE_NAME)) { printk(KERN_ERR PFX "Unable to register IO port %#x\n", io); ret = -EBUSY; - goto out_clean; + goto out_pnp; } ret = sc1200wdt_probe(); @@ -435,6 +435,11 @@ out_rbt: out_io: release_region(io, io_len); +out_pnp: +#if defined CONFIG_PNP + if (isapnp) + pnp_unregister_driver(&scl200wdt_pnp_driver); +#endif goto out_clean; } -- cgit v0.10.2 From 9f0f9313cec8c76c89bc8a68653f928fa12fab96 Mon Sep 17 00:00:00 2001 From: Kristoffer Ericson Date: Sun, 29 Oct 2006 22:38:08 +0100 Subject: [ARM] 3914/1: [Jornada7xx] - Typo Fix in cpu-sa1110.c (b != B) "K4S281632b-1H" should read "K4S281632B-1H" (As it does everywhere else). No more coffe! Signed-off-by: Kristoffer Ericson Signed-off-by: Russell King diff --git a/arch/arm/mach-sa1100/cpu-sa1110.c b/arch/arm/mach-sa1100/cpu-sa1110.c index 90a4130..78f4c13 100644 --- a/arch/arm/mach-sa1100/cpu-sa1110.c +++ b/arch/arm/mach-sa1100/cpu-sa1110.c @@ -83,7 +83,7 @@ static struct sdram_params sdram_tbl[] __initdata = { .refresh = 64000, .cas_latency = 3, }, { /* Samsung K4S281632B-1H */ - .name = "K4S281632b-1H", + .name = "K4S281632B-1H", .rows = 12, .tck = 10, .trp = 20, -- cgit v0.10.2 From 209ad53bc19667a128d9c51beba873a5c62bff6e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 29 Oct 2006 17:31:49 -0800 Subject: Revert "r8169: mac address change support" This reverts commit a2b98a697fa4e7564f78905b83db122824916cf9. As per Guennadi Liakhovetski, the mac address change support code breaks some normal uses (_without_ any address changes), and until it's all sorted out, we're better off without it. Says Francois: "Go revert it. Despite what I claimed, I can not find a third-party confirmation by email that it works elsewhere. It would probably be enough to remove the call to __rtl8169_set_mac_addr() in rtl8169_hw_start() though." See also http://bugzilla.kernel.org/show_bug.cgi?id=6032 Cc: Guennadi Liakhovetski Acked-by: Francois Romieu Signed-off-by: Linus Torvalds diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index d132fe7..27f90b2 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -1397,41 +1397,6 @@ static void rtl8169_netpoll(struct net_device *dev) } #endif -static void __rtl8169_set_mac_addr(struct net_device *dev, void __iomem *ioaddr) -{ - unsigned int i, j; - - RTL_W8(Cfg9346, Cfg9346_Unlock); - for (i = 0; i < 2; i++) { - __le32 l = 0; - - for (j = 0; j < 4; j++) { - l <<= 8; - l |= dev->dev_addr[4*i + j]; - } - RTL_W32(MAC0 + 4*i, cpu_to_be32(l)); - } - RTL_W8(Cfg9346, Cfg9346_Lock); -} - -static int rtl8169_set_mac_addr(struct net_device *dev, void *p) -{ - struct rtl8169_private *tp = netdev_priv(dev); - struct sockaddr *addr = p; - - if (!is_valid_ether_addr(addr->sa_data)) - return -EINVAL; - - memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); - - if (netif_running(dev)) { - spin_lock_irq(&tp->lock); - __rtl8169_set_mac_addr(dev, tp->mmio_addr); - spin_unlock_irq(&tp->lock); - } - return 0; -} - static void rtl8169_release_board(struct pci_dev *pdev, struct net_device *dev, void __iomem *ioaddr) { @@ -1681,7 +1646,6 @@ rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) dev->stop = rtl8169_close; dev->tx_timeout = rtl8169_tx_timeout; dev->set_multicast_list = rtl8169_set_rx_mode; - dev->set_mac_address = rtl8169_set_mac_addr; dev->watchdog_timeo = RTL8169_TX_TIMEOUT; dev->irq = pdev->irq; dev->base_addr = (unsigned long) ioaddr; @@ -1929,8 +1893,6 @@ rtl8169_hw_start(struct net_device *dev) /* Enable all known interrupts by setting the interrupt mask. */ RTL_W16(IntrMask, rtl8169_intr_mask); - __rtl8169_set_mac_addr(dev, ioaddr); - netif_start_queue(dev); } -- cgit v0.10.2 From 89f68225876db7df638de2884b561facb1870239 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Mon, 30 Oct 2006 11:47:02 -0600 Subject: jfs: Add splice support This allows the splice() and tee() syscalls to be used with JFS. Signed-off-by: Daniel Drake Signed-off-by: Dave Kleikamp diff --git a/fs/jfs/file.c b/fs/jfs/file.c index 34181b8..aa9132d 100644 --- a/fs/jfs/file.c +++ b/fs/jfs/file.c @@ -109,6 +109,8 @@ const struct file_operations jfs_file_operations = { .aio_write = generic_file_aio_write, .mmap = generic_file_mmap, .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, .fsync = jfs_fsync, .release = jfs_release, .ioctl = jfs_ioctl, -- cgit v0.10.2 From 0261d6886eb5822867a5310dc1e4479b940a1942 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 30 Oct 2006 19:07:48 +0100 Subject: [PATCH] CFQ: use irq safe locking in cfq_cic_link() If cfq_set_request() is called for a new process AND a non-fs io request (so that __GFP_WAIT may not be set), cfq_cic_link() may use spin_lock_irq() and spin_unlock_irq() with interrupts already disabled. Fix is to always use irq safe locking in cfq_cic_link() Acked-By: Arjan van de Ven Acked-by: Ingo Molnar Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index d3d7613..5c3da89 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1362,6 +1362,7 @@ cfq_cic_link(struct cfq_data *cfqd, struct io_context *ioc, struct rb_node **p; struct rb_node *parent; struct cfq_io_context *__cic; + unsigned long flags; void *k; cic->ioc = ioc; @@ -1391,9 +1392,9 @@ restart: rb_link_node(&cic->rb_node, parent, p); rb_insert_color(&cic->rb_node, &ioc->cic_root); - spin_lock_irq(cfqd->queue->queue_lock); + spin_lock_irqsave(cfqd->queue->queue_lock, flags); list_add(&cic->queue_list, &cfqd->cic_list); - spin_unlock_irq(cfqd->queue->queue_lock); + spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); } /* -- cgit v0.10.2 From c1b707d253fe918b92882cff1dbd926b47e14fd2 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 30 Oct 2006 19:54:23 +0100 Subject: [PATCH] CFQ: bad locking in changed_ioprio() When the ioprio code recently got juggled a bit, a bug was introduced. changed_ioprio() is no longer called with interrupts disabled, so using plain spin_lock() on the queue_lock is a bug. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 5c3da89..25c4e7e 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1215,11 +1215,12 @@ static inline void changed_ioprio(struct cfq_io_context *cic) { struct cfq_data *cfqd = cic->key; struct cfq_queue *cfqq; + unsigned long flags; if (unlikely(!cfqd)) return; - spin_lock(cfqd->queue->queue_lock); + spin_lock_irqsave(cfqd->queue->queue_lock, flags); cfqq = cic->cfqq[ASYNC]; if (cfqq) { @@ -1236,7 +1237,7 @@ static inline void changed_ioprio(struct cfq_io_context *cic) if (cfqq) cfq_mark_cfqq_prio_changed(cfqq); - spin_unlock(cfqd->queue->queue_lock); + spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); } static void cfq_ioc_set_ioprio(struct io_context *ioc) -- cgit v0.10.2 From 37af6560f7978c60791b5f3df17ce8b3e97f2d6e Mon Sep 17 00:00:00 2001 From: Christophe Saout Date: Mon, 30 Oct 2006 20:39:08 +0100 Subject: [PATCH] Fix dmsetup table output change Fix dm-crypt after the block cipher API changes to correctly return the backwards compatible cipher-chainmode[-ivmode] format for "dmsetup table". Signed-off-by: Christophe Saout Cc: Alasdair G Kergon Cc: Herbert Xu Signed-off-by: Linus Torvalds diff linux-2.6.19-rc3.orig/drivers/md/dm-crypt.c linux-2.6.19-rc3/drivers/md/dm-crypt.c diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index a625576..08a40f4 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -915,8 +915,6 @@ static int crypt_status(struct dm_target *ti, status_type_t type, char *result, unsigned int maxlen) { struct crypt_config *cc = (struct crypt_config *) ti->private; - const char *cipher; - const char *chainmode = NULL; unsigned int sz = 0; switch (type) { @@ -925,14 +923,11 @@ static int crypt_status(struct dm_target *ti, status_type_t type, break; case STATUSTYPE_TABLE: - cipher = crypto_blkcipher_name(cc->tfm); - - chainmode = cc->chainmode; - if (cc->iv_mode) - DMEMIT("%s-%s-%s ", cipher, chainmode, cc->iv_mode); + DMEMIT("%s-%s-%s ", cc->cipher, cc->chainmode, + cc->iv_mode); else - DMEMIT("%s-%s ", cipher, chainmode); + DMEMIT("%s-%s ", cc->cipher, cc->chainmode); if (cc->key_size > 0) { if ((maxlen - sz) < ((cc->key_size << 1) + 1)) -- cgit v0.10.2 From a5a89bae0449634fdb7aa7cdb1c5ba154e4a789b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 29 Oct 2006 22:46:33 -0800 Subject: [PATCH] ioc4_serial: irq flags fix Use the correct type for the CPU flags. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index ff4fa25..711bd15 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -921,7 +921,7 @@ static void handle_dma_error_intr(void *arg, uint32_t other_ir) { struct ioc4_port *port = (struct ioc4_port *)arg; struct hooks *hooks = port->ip_hooks; - unsigned int flags; + unsigned long flags; spin_lock_irqsave(&port->ip_lock, flags); @@ -1834,7 +1834,7 @@ static void handle_intr(void *arg, uint32_t sio_ir) struct ioc4_port *port = (struct ioc4_port *)arg; struct hooks *hooks = port->ip_hooks; unsigned int rx_high_rd_aborted = 0; - unsigned int flags; + unsigned long flags; struct uart_port *the_port; int loop_counter; -- cgit v0.10.2 From 0e2d57fc6e7dabdbfdd4f26c861e7e6c75d5bdcf Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 29 Oct 2006 22:46:34 -0800 Subject: [PATCH] ndiswrapper: don't set the module->taints flags For ndiswrapper, don't set the module->taints flags, just set the kernel global tainted flag. This should allow ndiswrapper to continue to use GPL symbols. Signed-off-by: Randy Dunlap Cc: Florin Malita Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/module.c b/kernel/module.c index 5072a94..f016656 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1718,7 +1718,7 @@ static struct module *load_module(void __user *umod, set_license(mod, get_modinfo(sechdrs, infoindex, "license")); if (strcmp(mod->name, "ndiswrapper") == 0) - add_taint_module(mod, TAINT_PROPRIETARY_MODULE); + add_taint(TAINT_PROPRIETARY_MODULE); if (strcmp(mod->name, "driverloader") == 0) add_taint_module(mod, TAINT_PROPRIETARY_MODULE); -- cgit v0.10.2 From 70812522b847bdb8fabee963191734f5fa3143f3 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 29 Oct 2006 22:46:35 -0800 Subject: [PATCH] isdn/gigaset: avoid cs->dev null pointer dereference When gigaset_initbcs() is called, cs->dev is not initialized yet. If dev_alloc_skb() failed in this function, NULL poinster dereference will happen at dev_warn(). Cc: Kai Germaschewski Cc: Hansjoerg Lipp Cc: Tilman Schmidt Acked-by: Karsten Keil Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index aca165d..d8d256d 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -616,7 +616,7 @@ static struct bc_state *gigaset_initbcs(struct bc_state *bcs, } else if ((bcs->skb = dev_alloc_skb(SBUFSIZE + HW_HDR_LEN)) != NULL) skb_reserve(bcs->skb, HW_HDR_LEN); else { - dev_warn(cs->dev, "could not allocate skb\n"); + gig_dbg(DEBUG_INIT, "could not allocate skb\n"); bcs->inputstate |= INS_skip_frame; } -- cgit v0.10.2 From 7259f0d05d595b73ef312a082e628627c6414969 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sun, 29 Oct 2006 22:46:36 -0800 Subject: [PATCH] lockdep: annotate DECLARE_WAIT_QUEUE_HEAD kernel: INFO: trying to register non-static key. kernel: the code is fine but needs lockdep annotation. kernel: turning off the locking correctness validator. kernel: [] show_trace_log_lvl+0x58/0x16a kernel: [] show_trace+0xd/0x10 kernel: [] dump_stack+0x19/0x1b kernel: [] __lock_acquire+0xf0/0x90d kernel: [] lock_acquire+0x4b/0x6b kernel: [] _spin_lock_irqsave+0x22/0x32 kernel: [] prepare_to_wait+0x17/0x4b kernel: [] lpfc_do_work+0xdd/0xcc2 [lpfc] kernel: [] kthread+0xc3/0xf2 kernel: [] kernel_thread_helper+0x5/0xb Another case of non-static lockdep keys; duplicate the paradigm set by DECLARE_COMPLETION_ONSTACK and introduce DECLARE_WAIT_QUEUE_HEAD_ONSTACK. Signed-off-by: Peter Zijlstra Cc: Greg KH Cc: Markus Lidel Acked-by: Ingo Molnar Cc: Arjan van de Ven Cc: James Bottomley Cc: Marcel Holtmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/bluetooth/bluecard_cs.c b/drivers/bluetooth/bluecard_cs.c index 845b868..cbc0725 100644 --- a/drivers/bluetooth/bluecard_cs.c +++ b/drivers/bluetooth/bluecard_cs.c @@ -282,7 +282,7 @@ static void bluecard_write_wakeup(bluecard_info_t *info) clear_bit(ready_bit, &(info->tx_state)); if (bt_cb(skb)->pkt_type & 0x80) { - DECLARE_WAIT_QUEUE_HEAD(wq); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); DEFINE_WAIT(wait); unsigned char baud_reg; diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 01a5a70..a235064 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -124,7 +124,7 @@ static void i2o_exec_wait_free(struct i2o_exec_wait *wait) int i2o_msg_post_wait_mem(struct i2o_controller *c, struct i2o_message *msg, unsigned long timeout, struct i2o_dma *dma) { - DECLARE_WAIT_QUEUE_HEAD(wq); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); struct i2o_exec_wait *wait; static u32 tcntxt = 0x80000000; unsigned long flags; diff --git a/drivers/scsi/dpt/dpti_i2o.h b/drivers/scsi/dpt/dpti_i2o.h index b3fa7ed..5a49216 100644 --- a/drivers/scsi/dpt/dpti_i2o.h +++ b/drivers/scsi/dpt/dpti_i2o.h @@ -49,7 +49,7 @@ #include typedef wait_queue_head_t adpt_wait_queue_head_t; -#define ADPT_DECLARE_WAIT_QUEUE_HEAD(wait) DECLARE_WAIT_QUEUE_HEAD(wait) +#define ADPT_DECLARE_WAIT_QUEUE_HEAD(wait) DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wait) typedef wait_queue_t adpt_wait_queue_t; /* diff --git a/drivers/scsi/imm.c b/drivers/scsi/imm.c index 2d95ac9..e31f612 100644 --- a/drivers/scsi/imm.c +++ b/drivers/scsi/imm.c @@ -1153,7 +1153,7 @@ static int __imm_attach(struct parport *pb) { struct Scsi_Host *host; imm_struct *dev; - DECLARE_WAIT_QUEUE_HEAD(waiting); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waiting); DEFINE_WAIT(wait); int ports; int modes, ppb; diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index d586c3d..19c79a0 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -305,7 +305,7 @@ lpfc_do_work(void *p) { struct lpfc_hba *phba = p; int rc; - DECLARE_WAIT_QUEUE_HEAD(work_waitq); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(work_waitq); set_user_nice(current, -20); phba->work_wait = &work_waitq; diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 24a1779..582f5ea 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -2983,7 +2983,7 @@ lpfc_sli_issue_iocb_wait(struct lpfc_hba * phba, struct lpfc_iocbq * prspiocbq, uint32_t timeout) { - DECLARE_WAIT_QUEUE_HEAD(done_q); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(done_q); long timeleft, timeout_req = 0; int retval = IOCB_SUCCESS; uint32_t creg_val; @@ -3061,7 +3061,7 @@ int lpfc_sli_issue_mbox_wait(struct lpfc_hba * phba, LPFC_MBOXQ_t * pmboxq, uint32_t timeout) { - DECLARE_WAIT_QUEUE_HEAD(done_q); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(done_q); DECLARE_WAITQUEUE(wq_entry, current); uint32_t timeleft = 0; int retval; diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c index b0eba39..89a2a9f 100644 --- a/drivers/scsi/ppa.c +++ b/drivers/scsi/ppa.c @@ -1012,7 +1012,7 @@ static LIST_HEAD(ppa_hosts); static int __ppa_attach(struct parport *pb) { struct Scsi_Host *host; - DECLARE_WAIT_QUEUE_HEAD(waiting); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waiting); DEFINE_WAIT(wait); ppa_struct *dev; int ports; diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index cf3d20e..4087363 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -554,7 +554,7 @@ static int usbnet_stop (struct net_device *net) { struct usbnet *dev = netdev_priv(net); int temp; - DECLARE_WAIT_QUEUE_HEAD (unlink_wakeup); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK (unlink_wakeup); DECLARE_WAITQUEUE (wait, current); netif_stop_queue (net); diff --git a/include/linux/wait.h b/include/linux/wait.h index b3b9048..e820d00 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -79,6 +79,15 @@ struct task_struct; extern void init_waitqueue_head(wait_queue_head_t *q); +#ifdef CONFIG_LOCKDEP +# define __WAIT_QUEUE_HEAD_INIT_ONSTACK(name) \ + ({ init_waitqueue_head(&name); name; }) +# define DECLARE_WAIT_QUEUE_HEAD_ONSTACK(name) \ + wait_queue_head_t name = __WAIT_QUEUE_HEAD_INIT_ONSTACK(name) +#else +# define DECLARE_WAIT_QUEUE_HEAD_ONSTACK(name) DECLARE_WAIT_QUEUE_HEAD(name) +#endif + static inline void init_waitqueue_entry(wait_queue_t *q, struct task_struct *p) { q->flags = 0; -- cgit v0.10.2 From f8da1f746588ba997734274b4aadc5ae501fbd88 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sun, 29 Oct 2006 22:46:39 -0800 Subject: [PATCH] cryptocop: double spin_lock_irqsave() Signed-off-by: Alexey Dobriyan Cc: Mikael Starvik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/cris/arch-v32/drivers/cryptocop.c b/arch/cris/arch-v32/drivers/cryptocop.c index ba096eb..2449637 100644 --- a/arch/cris/arch-v32/drivers/cryptocop.c +++ b/arch/cris/arch-v32/drivers/cryptocop.c @@ -2051,7 +2051,6 @@ static void cryptocop_job_queue_close(void) spin_lock_irqsave(&cryptocop_process_lock, process_flags); /* Empty the job queue. */ - spin_lock_irqsave(&cryptocop_process_lock, process_flags); for (i = 0; i < cryptocop_prio_no_prios; i++){ if (!list_empty(&(cryptocop_job_queues[i].jobs))){ list_for_each_safe(node, tmp, &(cryptocop_job_queues[i].jobs)) { -- cgit v0.10.2 From 351edd240d0ba8620789ca9e24f5a38b62157f23 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 29 Oct 2006 22:46:40 -0800 Subject: [PATCH] MTD: fix last kernel-doc warning Fix the last current kernel-doc warning: Warning(/var/linsrc/linux-2619-rc3g5//include/linux/mtd/nand.h:416): No description found for parameter 'write_page' Signed-off-by: Randy Dunlap Cc: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index 70420bb..8b3ef41 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -355,7 +355,7 @@ struct nand_buffers { * @priv: [OPTIONAL] pointer to private chip date * @errstat: [OPTIONAL] hardware specific function to perform additional error status checks * (determine if errors are correctable) - * @write_page [REPLACEABLE] High-level page write function + * @write_page: [REPLACEABLE] High-level page write function */ struct nand_chip { -- cgit v0.10.2 From bbb5bbb037812ba36d638014a93134148a568684 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 29 Oct 2006 22:46:40 -0800 Subject: [PATCH] docbook: make a filesystems book Make a filesystems DocBook book/file by moving all filesystems info from kernel-api.tmpl. Will also merge journal-api.tmpl into it soon (with permission from Roger Gammans). Localizes filesystem info and reduces size of the huge (produced) kernel-api output files. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 66e1cf7..3bf5086 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -9,7 +9,7 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ kernel-hacking.xml kernel-locking.xml deviceiobook.xml \ procfs-guide.xml writing_usb_driver.xml \ - kernel-api.xml journal-api.xml lsm.xml usb.xml \ + kernel-api.xml filesystems.xml journal-api.xml lsm.xml usb.xml \ gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \ genericirq.xml diff --git a/Documentation/DocBook/filesystems.tmpl b/Documentation/DocBook/filesystems.tmpl new file mode 100644 index 0000000..4785032 --- /dev/null +++ b/Documentation/DocBook/filesystems.tmpl @@ -0,0 +1,101 @@ + + + + + + Linux Filesystems API + + + + This documentation is free software; you can redistribute + it and/or modify it under the terms of the GNU General Public + License as published by the Free Software Foundation; either + version 2 of the License, or (at your option) any later + version. + + + + This program is distributed in the hope that it will be + useful, but WITHOUT ANY WARRANTY; without even the implied + warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + See the GNU General Public License for more details. + + + + You should have received a copy of the GNU General Public + License along with this program; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, + MA 02111-1307 USA + + + + For more details see the file COPYING in the source + distribution of Linux. + + + + + + + + The Linux VFS + The Filesystem types +!Iinclude/linux/fs.h + + The Directory Cache +!Efs/dcache.c +!Iinclude/linux/dcache.h + + Inode Handling +!Efs/inode.c +!Efs/bad_inode.c + + Registration and Superblocks +!Efs/super.c + + File Locks +!Efs/locks.c +!Ifs/locks.c + + Other Functions +!Efs/mpage.c +!Efs/namei.c +!Efs/buffer.c +!Efs/bio.c +!Efs/seq_file.c +!Efs/filesystems.c +!Efs/fs-writeback.c +!Efs/block_dev.c + + + + + The proc filesystem + + sysctl interface +!Ekernel/sysctl.c + + + proc filesystem interface +!Ifs/proc/base.c + + + + + The Filesystem for Exporting Kernel Objects +!Efs/sysfs/file.c +!Efs/sysfs/symlink.c +!Efs/sysfs/bin.c + + + + The debugfs filesystem + + debugfs interface +!Efs/debugfs/inode.c +!Efs/debugfs/file.c + + + + diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index 2b5ac60..a166675 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -182,66 +182,6 @@ X!Ilib/string.c - - The Linux VFS - The Filesystem types -!Iinclude/linux/fs.h - - The Directory Cache -!Efs/dcache.c -!Iinclude/linux/dcache.h - - Inode Handling -!Efs/inode.c -!Efs/bad_inode.c - - Registration and Superblocks -!Efs/super.c - - File Locks -!Efs/locks.c -!Ifs/locks.c - - Other Functions -!Efs/mpage.c -!Efs/namei.c -!Efs/buffer.c -!Efs/bio.c -!Efs/seq_file.c -!Efs/filesystems.c -!Efs/fs-writeback.c -!Efs/block_dev.c - - - - - The proc filesystem - - sysctl interface -!Ekernel/sysctl.c - - - proc filesystem interface -!Ifs/proc/base.c - - - - - The Filesystem for Exporting Kernel Objects -!Efs/sysfs/file.c -!Efs/sysfs/symlink.c -!Efs/sysfs/bin.c - - - - The debugfs filesystem - - debugfs interface -!Efs/debugfs/inode.c -!Efs/debugfs/file.c - - - relay interface support -- cgit v0.10.2 From b428b51ed9a4ff8445ea50769961f948480c1d36 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 29 Oct 2006 22:46:41 -0800 Subject: [PATCH] Fix "Remove the use of _syscallX macros in UML" Fix commit 5f4c6bc1f369f20807a8e753c2308d1629478c61: it spits out warnings about missing syscall prototype (it is in ) and it does not recognize that two uses of _syscallX are to be resolved against kernel headers in the source tree, not against _syscallX; they in fact do not compile and would not work anyway. If _syscallX macros will be removed from the kernel tree altogether, the only reasonable solution for that piece of code is switching to open-coded inline assembly (it's remapping the whole executable from memory, except the page containing this code). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/os-Linux/sys-i386/tls.c b/arch/um/os-Linux/sys-i386/tls.c index 6e945ab..2565320 100644 --- a/arch/um/os-Linux/sys-i386/tls.c +++ b/arch/um/os-Linux/sys-i386/tls.c @@ -1,6 +1,9 @@ #include #include + #include +#include + #include "sysdep/tls.h" #include "user_util.h" diff --git a/arch/um/os-Linux/tls.c b/arch/um/os-Linux/tls.c index a2de258..9f7999f 100644 --- a/arch/um/os-Linux/tls.c +++ b/arch/um/os-Linux/tls.c @@ -1,6 +1,7 @@ #include #include #include +#include #include #include "sysdep/tls.h" #include "uml-config.h" diff --git a/arch/um/sys-i386/unmap.c b/arch/um/sys-i386/unmap.c index 8e55cd5..1b0ad0e4 100644 --- a/arch/um/sys-i386/unmap.c +++ b/arch/um/sys-i386/unmap.c @@ -5,17 +5,20 @@ #include #include -#include +static int errno; + +static inline _syscall2(int,munmap,void *,start,size_t,len) +static inline _syscall6(void *,mmap2,void *,addr,size_t,len,int,prot,int,flags,int,fd,off_t,offset) int switcheroo(int fd, int prot, void *from, void *to, int size) { - if (syscall(__NR_munmap, to, size) < 0){ + if(munmap(to, size) < 0){ return(-1); } - if (syscall(__NR_mmap2, to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) == (void*) -1 ){ + if(mmap2(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) == (void*) -1 ){ return(-1); } - if (syscall(__NR_munmap, from, size) < 0){ + if(munmap(from, size) < 0){ return(-1); } return(0); diff --git a/arch/um/sys-x86_64/unmap.c b/arch/um/sys-x86_64/unmap.c index 57c9286..f4a4bff 100644 --- a/arch/um/sys-x86_64/unmap.c +++ b/arch/um/sys-x86_64/unmap.c @@ -5,17 +5,20 @@ #include #include -#include +static int errno; + +static inline _syscall2(int,munmap,void *,start,size_t,len) +static inline _syscall6(void *,mmap,void *,addr,size_t,len,int,prot,int,flags,int,fd,off_t,offset) int switcheroo(int fd, int prot, void *from, void *to, int size) { - if (syscall(__NR_munmap, to, size) < 0){ + if(munmap(to, size) < 0){ return(-1); } - if (syscall(__NR_mmap, to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) == (void*) -1){ + if(mmap(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) == (void*) -1){ return(-1); } - if (syscall(__NR_munmap, from, size) < 0){ + if(munmap(from, size) < 0){ return(-1); } return(0); -- cgit v0.10.2 From d45e44d4be60ef508579001792f33753b5cb6d36 Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 29 Oct 2006 22:46:42 -0800 Subject: [PATCH] uml: fix compilation options for USER_OBJS From: Jeff Dike , Paolo Giarrusso Make sure that when compiling USER_OBJS the correct compilation options are passed; since they are compiled with USER_CFLAGS which is derived from CFLAGS, make sure it is a recursively evaluated variable, so that changes to CFLAGS done afterwards the inclusion of arch/$(ARCH)/Makefile are reflected in USER_CFLAGS. For instance, without this patch userspace objects are never compiled with debug info active. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Makefile b/arch/um/Makefile index c8016a9..5d5ed72 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -64,9 +64,14 @@ CFLAGS += $(CFLAGS-y) -D__arch_um__ -DSUBARCH=\"$(SUBARCH)\" \ AFLAGS += $(ARCH_INCLUDE) -USER_CFLAGS := $(patsubst -I%,,$(CFLAGS)) -USER_CFLAGS := $(patsubst -D__KERNEL__,,$(USER_CFLAGS)) $(ARCH_INCLUDE) \ - $(MODE_INCLUDE) -D_FILE_OFFSET_BITS=64 +USER_CFLAGS = $(patsubst $(KERNEL_DEFINES),,$(patsubst -D__KERNEL__,,\ + $(patsubst -I%,,$(CFLAGS)))) $(ARCH_INCLUDE) $(MODE_INCLUDE) \ + -D_FILE_OFFSET_BITS=64 + +include $(srctree)/$(ARCH_DIR)/Makefile-$(SUBARCH) + +#This will adjust *FLAGS accordingly to the platform. +include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS) # -Derrno=kernel_errno - This turns all kernel references to errno into # kernel_errno to separate them from the libc errno. This allows -fno-common @@ -74,15 +79,11 @@ USER_CFLAGS := $(patsubst -D__KERNEL__,,$(USER_CFLAGS)) $(ARCH_INCLUDE) \ # errnos. # These apply to kernelspace only. -CFLAGS += -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \ - -Dmktime=kernel_mktime +KERNEL_DEFINES = -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \ + -Dmktime=kernel_mktime $(ARCH_KERNEL_DEFINES) +CFLAGS += $(KERNEL_DEFINES) CFLAGS += $(call cc-option,-fno-unit-at-a-time,) -include $(srctree)/$(ARCH_DIR)/Makefile-$(SUBARCH) - -#This will adjust *FLAGS accordingly to the platform. -include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS) - # These are needed for clean and mrproper, since in that case .config is not # included; the values here are meaningless diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386 index b65ca11..c9f1c5b 100644 --- a/arch/um/Makefile-i386 +++ b/arch/um/Makefile-i386 @@ -16,7 +16,6 @@ OBJCOPYFLAGS := -O binary -R .note -R .comment -S ifeq ("$(origin SUBARCH)", "command line") ifneq ("$(shell uname -m | sed -e s/i.86/i386/)", "$(SUBARCH)") CFLAGS += $(call cc-option,-m32) -USER_CFLAGS += $(call cc-option,-m32) AFLAGS += $(call cc-option,-m32) LINK-y += $(call cc-option,-m32) UML_OBJCOPYFLAGS += -F $(ELF_FORMAT) @@ -25,7 +24,7 @@ export LDFLAGS HOSTCFLAGS HOSTLDFLAGS UML_OBJCOPYFLAGS endif endif -CFLAGS += -U__$(SUBARCH)__ -U$(SUBARCH) +ARCH_KERNEL_DEFINES += -U__$(SUBARCH)__ -U$(SUBARCH) # First of all, tune CFLAGS for the specific CPU. This actually sets cflags-y. include $(srctree)/arch/i386/Makefile.cpu @@ -38,4 +37,3 @@ cflags-y += $(call cc-option,-mpreferred-stack-boundary=2) cflags-y += -ffreestanding CFLAGS += $(cflags-y) -USER_CFLAGS += $(cflags-y) diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64 index d278682..69ecea6 100644 --- a/arch/um/Makefile-x86_64 +++ b/arch/um/Makefile-x86_64 @@ -8,8 +8,8 @@ _extra_flags_ = -fno-builtin -m64 #We #undef __x86_64__ for kernelspace, not for userspace where #it's needed for headers to work! -CFLAGS += -U__$(SUBARCH)__ $(_extra_flags_) -USER_CFLAGS += $(_extra_flags_) +ARCH_KERNEL_DEFINES = -U__$(SUBARCH)__ +CFLAGS += $(_extra_flags_) CHECKFLAGS += -m64 AFLAGS += -m64 -- cgit v0.10.2 From f0ec1aaf54caddd21c259aea8b2ecfbde4ee4fb9 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Oct 2006 22:46:43 -0800 Subject: [PATCH] xacct_add_tsk: fix pure theoretical ->mm use-after-free Paranoid fix. The task can free its ->mm after the 'if (p->mm)' check. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 65a5036..96f7701 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -80,13 +80,17 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) */ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) { + struct mm_struct *mm; + /* convert pages-jiffies to Mbyte-usec */ stats->coremem = jiffies_to_usecs(p->acct_rss_mem1) * PAGE_SIZE / MB; stats->virtmem = jiffies_to_usecs(p->acct_vm_mem1) * PAGE_SIZE / MB; - if (p->mm) { + mm = get_task_mm(p); + if (mm) { /* adjust to KB unit */ - stats->hiwater_rss = p->mm->hiwater_rss * PAGE_SIZE / KB; - stats->hiwater_vm = p->mm->hiwater_vm * PAGE_SIZE / KB; + stats->hiwater_rss = mm->hiwater_rss * PAGE_SIZE / KB; + stats->hiwater_vm = mm->hiwater_vm * PAGE_SIZE / KB; + mmput(mm); } stats->read_char = p->rchar; stats->write_char = p->wchar; -- cgit v0.10.2 From 2b52c9590d5ad2fb67b720ec12018dd2cf061480 Mon Sep 17 00:00:00 2001 From: Sergey Vlasov Date: Sun, 29 Oct 2006 22:46:44 -0800 Subject: [PATCH] drivers/ide/pci/generic.c: add missing newline to the all-generic-ide message Signed-off-by: Sergey Vlasov Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c index e72ab36..9f30688 100644 --- a/drivers/ide/pci/generic.c +++ b/drivers/ide/pci/generic.c @@ -48,7 +48,7 @@ static int ide_generic_all; /* Set to claim all devices */ static int __init ide_generic_all_on(char *unused) { ide_generic_all = 1; - printk(KERN_INFO "IDE generic will claim all unknown PCI IDE storage controllers."); + printk(KERN_INFO "IDE generic will claim all unknown PCI IDE storage controllers.\n"); return 1; } __setup("all-generic-ide", ide_generic_all_on); -- cgit v0.10.2 From d6740df98e12a8e49ef3a699dcc1e2913f22c51b Mon Sep 17 00:00:00 2001 From: Neil Brown Date: Sun, 29 Oct 2006 22:46:45 -0800 Subject: [PATCH] sunrpc: fix refcounting problems in rpc servers A recent patch fixed a problem which would occur when the refcount on an auth_domain reached zero. This problem has not been reported in practice despite existing in two major kernel releases because the refcount can never reach zero. This patch fixes the problems that stop the refcount reaching zero. 1/ We were adding to the refcount when inserting in the hash table, but only removing from the hashtable when the refcount reached zero. Obviously it never would. So don't count the implied reference of being in the hash table. 2/ There are two paths on which a socket can be destroyed. One called svcauth_unix_info_release(). The other didn't. So when the other was taken, we can lose a reference to an ip_map which in-turn holds a reference to an auth_domain So unify the exit paths into svc_sock_put. This highlights the fact that svc_delete_socket has slightly odd semantics - it does not drop a reference but probably should. Fixing this need a bit more thought and testing. Signed-off-by: Neil Brown Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index 8f2320a..0004c1f 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -147,10 +147,8 @@ auth_domain_lookup(char *name, struct auth_domain *new) return hp; } } - if (new) { + if (new) hlist_add_head(&new->hash, head); - kref_get(&new->ref); - } spin_unlock(&auth_domain_lock); return new; } diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 96521f1..db0d104 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -300,8 +300,13 @@ static inline void svc_sock_put(struct svc_sock *svsk) { if (atomic_dec_and_test(&svsk->sk_inuse) && test_bit(SK_DEAD, &svsk->sk_flags)) { - dprintk("svc: releasing dead socket\n"); - sock_release(svsk->sk_sock); + printk("svc: releasing dead socket\n"); + if (svsk->sk_sock->file) + sockfd_put(svsk->sk_sock); + else + sock_release(svsk->sk_sock); + if (svsk->sk_info_authunix != NULL) + svcauth_unix_info_release(svsk->sk_info_authunix); kfree(svsk); } } @@ -1604,20 +1609,13 @@ svc_delete_socket(struct svc_sock *svsk) if (test_bit(SK_TEMP, &svsk->sk_flags)) serv->sv_tmpcnt--; - if (!atomic_read(&svsk->sk_inuse)) { - spin_unlock_bh(&serv->sv_lock); - if (svsk->sk_sock->file) - sockfd_put(svsk->sk_sock); - else - sock_release(svsk->sk_sock); - if (svsk->sk_info_authunix != NULL) - svcauth_unix_info_release(svsk->sk_info_authunix); - kfree(svsk); - } else { - spin_unlock_bh(&serv->sv_lock); - dprintk(KERN_NOTICE "svc: server socket destroy delayed\n"); - /* svsk->sk_server = NULL; */ - } + /* This atomic_inc should be needed - svc_delete_socket + * should have the semantic of dropping a reference. + * But it doesn't yet.... + */ + atomic_inc(&svsk->sk_inuse); + spin_unlock_bh(&serv->sv_lock); + svc_sock_put(svsk); } /* -- cgit v0.10.2 From 3f4b23e9833f0816b57d07401eac5e184b627fd7 Mon Sep 17 00:00:00 2001 From: Kristian Mueller Date: Sun, 29 Oct 2006 22:46:46 -0800 Subject: [PATCH] APM: URL of APM 1.2 specs has changed APM BIOS Interface Secification can now be found at http://www.microsoft.com/whdc/archive/amp_12.mspx Signed-off-by: Kristian Mueller Acked-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index 2af6585..a60358f 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -198,7 +198,7 @@ * (APM) BIOS Interface Specification, Revision 1.2, February 1996. * * [This document is available from Microsoft at: - * http://www.microsoft.com/hwdev/busbios/amp_12.htm] + * http://www.microsoft.com/whdc/archive/amp_12.mspx] */ #include -- cgit v0.10.2 From 202dd450246cd658e298a86dddca5592fcb898cd Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 29 Oct 2006 22:57:57 -0800 Subject: [PATCH] fix "sunrpc: fix refcounting problems in rpc servers" - printk should remain dprintk - fix coding-style. Cc: Neil Brown Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index db0d104..64ca1f6 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -299,8 +299,9 @@ void svc_reserve(struct svc_rqst *rqstp, int space) static inline void svc_sock_put(struct svc_sock *svsk) { - if (atomic_dec_and_test(&svsk->sk_inuse) && test_bit(SK_DEAD, &svsk->sk_flags)) { - printk("svc: releasing dead socket\n"); + if (atomic_dec_and_test(&svsk->sk_inuse) && + test_bit(SK_DEAD, &svsk->sk_flags)) { + dprintk("svc: releasing dead socket\n"); if (svsk->sk_sock->file) sockfd_put(svsk->sk_sock); else -- cgit v0.10.2 From a7aacdf9ea45bf6139cfd750e558a3dcbc6f16c3 Mon Sep 17 00:00:00 2001 From: Albert Cahalan Date: Sun, 29 Oct 2006 22:26:17 -0500 Subject: [PATCH] fix i386 regparm=3 RT signal handlers on x86_64 The recent change to make x86_64 support i386 binaries compiled with -mregparm=3 only covered signal handlers without SA_SIGINFO. (the 3-arg "real-time" ones) This is useful for klibc at least. Signed-off-by: Albert Cahalan Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c index a6ba995..0e0a266 100644 --- a/arch/x86_64/ia32/ia32_signal.c +++ b/arch/x86_64/ia32/ia32_signal.c @@ -579,6 +579,11 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->rsp = (unsigned long) frame; regs->rip = (unsigned long) ka->sa.sa_handler; + /* Make -mregparm=3 work */ + regs->rax = sig; + regs->rdx = (unsigned long) &frame->info; + regs->rcx = (unsigned long) &frame->uc; + asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); -- cgit v0.10.2 From be609f3546d3dd96afd7df8856a91fa2b8825fbc Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 23 Oct 2006 13:22:06 +0100 Subject: [MIPS] Oprofile: fix on non-VSMP / non-SMTC SMP configurations. Signed-off-by: Ralf Baechle diff --git a/arch/mips/oprofile/op_model_mipsxx.c b/arch/mips/oprofile/op_model_mipsxx.c index dd0aec9..ebce715 100644 --- a/arch/mips/oprofile/op_model_mipsxx.c +++ b/arch/mips/oprofile/op_model_mipsxx.c @@ -31,16 +31,18 @@ #define M_COUNTER_OVERFLOW (1UL << 31) #ifdef CONFIG_MIPS_MT_SMP -#define WHAT (M_TC_EN_VPE | M_PERFCTL_VPEID(smp_processor_id())) +#define WHAT (M_TC_EN_VPE | M_PERFCTL_VPEID(smp_processor_id())) +#define vpe_id() smp_processor_id() #else -#define WHAT 0 +#define WHAT 0 +#define vpe_id() smp_processor_id() #endif #define __define_perf_accessors(r, n, np) \ \ static inline unsigned int r_c0_ ## r ## n(void) \ { \ - unsigned int cpu = smp_processor_id(); \ + unsigned int cpu = vpe_id(); \ \ switch (cpu) { \ case 0: \ @@ -55,7 +57,7 @@ static inline unsigned int r_c0_ ## r ## n(void) \ \ static inline void w_c0_ ## r ## n(unsigned int value) \ { \ - unsigned int cpu = smp_processor_id(); \ + unsigned int cpu = vpe_id(); \ \ switch (cpu) { \ case 0: \ -- cgit v0.10.2 From 53c1b192ece077e1c3892e6afb453ab6f009af81 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 3 Sep 2006 22:17:10 +0400 Subject: [MIPS] Au1xx0 code sets incorrect mips_hpt_frequency Alchemy CPU counter ticks at the full CPU clock speed. Signed-off-by: Sergei Shtylyov Signed-off-by: Ralf Baechle diff --git a/arch/mips/au1000/common/time.c b/arch/mips/au1000/common/time.c index 94f0919..7313ef2 100644 --- a/arch/mips/au1000/common/time.c +++ b/arch/mips/au1000/common/time.c @@ -231,7 +231,6 @@ wakeup_counter0_set(int ticks) */ unsigned long cal_r4koff(void) { - unsigned long count; unsigned long cpu_speed; unsigned long flags; unsigned long counter; @@ -258,7 +257,7 @@ unsigned long cal_r4koff(void) #if defined(CONFIG_AU1000_USE32K) { - unsigned long start, end; + unsigned long start, end, count; start = au_readl(SYS_RTCREAD); start += 2; @@ -282,7 +281,6 @@ unsigned long cal_r4koff(void) #else cpu_speed = (au_readl(SYS_CPUPLL) & 0x0000003f) * AU1000_SRC_CLK; - count = cpu_speed / 2; #endif } else { @@ -291,10 +289,9 @@ unsigned long cal_r4koff(void) * NOTE: some old silicon doesn't allow reading the PLL. */ cpu_speed = (au_readl(SYS_CPUPLL) & 0x0000003f) * AU1000_SRC_CLK; - count = cpu_speed / 2; no_au1xxx_32khz = 1; } - mips_hpt_frequency = count; + mips_hpt_frequency = cpu_speed; // Equation: Baudrate = CPU / (SD * 2 * CLKDIV * 16) set_au1x00_uart_baud_base(cpu_speed / (2 * ((int)(au_readl(SYS_POWERCTRL)&0x03) + 2) * 16)); spin_unlock_irqrestore(&time_lock, flags); -- cgit v0.10.2 From ea3df4ac7d4adad915e2d0431a9407b10617114c Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 23 Oct 2006 23:21:21 +0100 Subject: [MIPS] Oprofile: Fix MIPSxx counter number detection. Signed-off-by: Ralf Baechle diff --git a/arch/mips/oprofile/op_model_mipsxx.c b/arch/mips/oprofile/op_model_mipsxx.c index ebce715..1fb240c 100644 --- a/arch/mips/oprofile/op_model_mipsxx.c +++ b/arch/mips/oprofile/op_model_mipsxx.c @@ -220,7 +220,7 @@ static inline int n_counters(void) { int counters = __n_counters(); -#ifndef CONFIG_SMP +#ifdef CONFIG_MIPS_MT_SMP if (current_cpu_data.cputype == CPU_34K) return counters >> 1; #endif -- cgit v0.10.2 From 8cde4a3162fdcccaf0ecbf3c21cdb13df422936d Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 26 Oct 2006 14:29:01 +0100 Subject: [MIPS] SMTC: Make 8 the default number of processors. 8 is the next larger power of two of the currently 5 supported TCs. Signed-off-by: Ralf Baechle diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 14af6cc..9c92b49 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1690,6 +1690,7 @@ config NR_CPUS depends on SMP default "64" if SGI_IP27 default "2" + default "8" if MIPS_MT_SMTC help This allows you to specify the maximum number of CPUs which this kernel will support. The maximum supported value is 32 for 32-bit -- cgit v0.10.2 From f0ec69e52989986e861a352099803cbb77ca53ba Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Sat, 28 Oct 2006 00:42:24 +0900 Subject: [MIPS] Fix warning about unused definition in c-sb1.c arch/mips/mm/c-sb1.c: In function `sb1_cache_init': arch/mips/mm/c-sb1.c:447: warning: unused variable `handle_vec2_sb1' Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/mm/c-sb1.c b/arch/mips/mm/c-sb1.c index 5537558..f2941b4 100644 --- a/arch/mips/mm/c-sb1.c +++ b/arch/mips/mm/c-sb1.c @@ -444,7 +444,6 @@ static __init void probe_cache_sizes(void) void sb1_cache_init(void) { extern char except_vec2_sb1; - extern char handle_vec2_sb1; /* Special cache error handler for SB1 */ set_uncached_handler (0x100, &except_vec2_sb1, 0x80); -- cgit v0.10.2 From 9448b8f6a014f46450ef65d81c0be2ca5a81c867 Mon Sep 17 00:00:00 2001 From: Manish Lachwani Date: Thu, 5 Oct 2006 16:30:44 -0700 Subject: [MIPS] Make SB1 cache flushes not to use on_each_cpu This fixes the start_kernel(): bug: interrupts were enabled early messages. Signed-off-by: Manish Lachwani Signed-off-by: Ralf Baechle diff --git a/arch/mips/mm/c-sb1.c b/arch/mips/mm/c-sb1.c index f2941b4..ea49a77 100644 --- a/arch/mips/mm/c-sb1.c +++ b/arch/mips/mm/c-sb1.c @@ -49,6 +49,15 @@ static unsigned short dcache_sets; static unsigned int icache_range_cutoff; static unsigned int dcache_range_cutoff; +static inline void sb1_on_each_cpu(void (*func) (void *info), void *info, + int retry, int wait) +{ + preempt_disable(); + smp_call_function(func, info, retry, wait); + func(info); + preempt_enable(); +} + /* * The dcache is fully coherent to the system, with one * big caveat: the instruction stream. In other words, @@ -226,7 +235,7 @@ static void sb1_flush_cache_page(struct vm_area_struct *vma, unsigned long addr, args.vma = vma; args.addr = addr; args.pfn = pfn; - on_each_cpu(sb1_flush_cache_page_ipi, (void *) &args, 1, 1); + sb1_on_each_cpu(sb1_flush_cache_page_ipi, (void *) &args, 1, 1); } #else void sb1_flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn) @@ -249,7 +258,7 @@ void sb1___flush_cache_all_ipi(void *ignored) static void sb1___flush_cache_all(void) { - on_each_cpu(sb1___flush_cache_all_ipi, 0, 1, 1); + sb1_on_each_cpu(sb1___flush_cache_all_ipi, 0, 1, 1); } #else void sb1___flush_cache_all(void) @@ -299,7 +308,7 @@ void sb1_flush_icache_range(unsigned long start, unsigned long end) args.start = start; args.end = end; - on_each_cpu(sb1_flush_icache_range_ipi, &args, 1, 1); + sb1_on_each_cpu(sb1_flush_icache_range_ipi, &args, 1, 1); } #else void sb1_flush_icache_range(unsigned long start, unsigned long end) @@ -326,7 +335,7 @@ static void sb1_flush_cache_sigtramp_ipi(void *info) static void sb1_flush_cache_sigtramp(unsigned long addr) { - on_each_cpu(sb1_flush_cache_sigtramp_ipi, (void *) addr, 1, 1); + sb1_on_each_cpu(sb1_flush_cache_sigtramp_ipi, (void *) addr, 1, 1); } #else void sb1_flush_cache_sigtramp(unsigned long addr) -- cgit v0.10.2 From 991ea26dcbc2524a054f37911ea375e631cb8891 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 29 Oct 2006 21:07:40 +0000 Subject: [MIPS] Wire up getcpu(2) and epoll_wait(2) syscalls. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index 720fac3..a95f37d 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -654,6 +654,8 @@ einval: li v0, -EINVAL sys sys_set_robust_list 2 sys sys_get_robust_list 3 /* 4310 */ sys sys_ni_syscall 0 + sys sys_getcpu 3 + sys sys_epoll_pwait 6 .endm /* We pre-compute the number of _instruction_ bytes needed to diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 3a34f62..8fb0f60 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -469,3 +469,5 @@ sys_call_table: PTR sys_set_robust_list PTR sys_get_robust_list PTR sys_ni_syscall /* 5270 */ + PTR sys_getcpu + PTR sys_epoll_pwait diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 67b92a1..0da5ca2 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -395,3 +395,5 @@ EXPORT(sysn32_call_table) PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list PTR sys_ni_syscall + PTR sys_getcpu + PTR sys_epoll_pwait diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 2875c4a..b9d00ca 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -517,4 +517,6 @@ sys_call_table: PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list /* 4310 */ PTR sys_ni_syscall + PTR sys_getcpu + PTR sys_epoll_pwait .size sys_call_table,.-sys_call_table diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h index 30240a4..f1ef98c 100644 --- a/include/asm-mips/unistd.h +++ b/include/asm-mips/unistd.h @@ -332,16 +332,18 @@ #define __NR_set_robust_list (__NR_Linux + 309) #define __NR_get_robust_list (__NR_Linux + 310) #define __NR_kexec_load (__NR_Linux + 311) +#define __NR_getcpu (__NR_Linux + 312) +#define __NR_epoll_pwait (__NR_Linux + 313) /* * Offset of the last Linux o32 flavoured syscall */ -#define __NR_Linux_syscalls 311 +#define __NR_Linux_syscalls 313 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */ #define __NR_O32_Linux 4000 -#define __NR_O32_Linux_syscalls 311 +#define __NR_O32_Linux_syscalls 313 #if _MIPS_SIM == _MIPS_SIM_ABI64 @@ -620,16 +622,18 @@ #define __NR_set_robust_list (__NR_Linux + 268) #define __NR_get_robust_list (__NR_Linux + 269) #define __NR_kexec_load (__NR_Linux + 270) +#define __NR_getcpu (__NR_Linux + 271) +#define __NR_epoll_pwait (__NR_Linux + 272) /* * Offset of the last Linux 64-bit flavoured syscall */ -#define __NR_Linux_syscalls 270 +#define __NR_Linux_syscalls 272 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */ #define __NR_64_Linux 5000 -#define __NR_64_Linux_syscalls 270 +#define __NR_64_Linux_syscalls 272 #if _MIPS_SIM == _MIPS_SIM_NABI32 @@ -912,16 +916,18 @@ #define __NR_set_robust_list (__NR_Linux + 272) #define __NR_get_robust_list (__NR_Linux + 273) #define __NR_kexec_load (__NR_Linux + 274) +#define __NR_getcpu (__NR_Linux + 275) +#define __NR_epoll_pwait (__NR_Linux + 276) /* * Offset of the last N32 flavoured syscall */ -#define __NR_Linux_syscalls 274 +#define __NR_Linux_syscalls 276 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */ #define __NR_N32_Linux 6000 -#define __NR_N32_Linux_syscalls 274 +#define __NR_N32_Linux_syscalls 276 #ifdef __KERNEL__ -- cgit v0.10.2 From a597a473b422f5a69b1f567f12e04db39fa01f39 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Sun, 29 Oct 2006 23:37:40 +0900 Subject: [MIPS] Au1000: Fix warning about unused variable. arch/mips/au1000/common/time.c: In function `mips_timer_interrupt': arch/mips/au1000/common/time.c:82: warning: unused variable `count' Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/au1000/common/time.c b/arch/mips/au1000/common/time.c index 7313ef2..6768638 100644 --- a/arch/mips/au1000/common/time.c +++ b/arch/mips/au1000/common/time.c @@ -82,7 +82,6 @@ unsigned long wtimer; void mips_timer_interrupt(void) { int irq = 63; - unsigned long count; irq_enter(); kstat_this_cpu.irqs[irq]++; -- cgit v0.10.2 From c39c30da2dad1aa3fb61862039634e9480a16fde Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Sun, 29 Oct 2006 23:37:55 +0900 Subject: [MIPS] Fix return value of TXX9 SPI interrupt handler Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/tx4938/toshiba_rbtx4938/spi_txx9.c b/arch/mips/tx4938/toshiba_rbtx4938/spi_txx9.c index b926e6a..08b20cd 100644 --- a/arch/mips/tx4938/toshiba_rbtx4938/spi_txx9.c +++ b/arch/mips/tx4938/toshiba_rbtx4938/spi_txx9.c @@ -36,14 +36,18 @@ void __init txx9_spi_init(unsigned long base, int (*cs_func)(int chipid, int on) static DECLARE_WAIT_QUEUE_HEAD(txx9_spi_wait); -static void txx9_spi_interrupt(int irq, void *dev_id) +static irqreturn_t txx9_spi_interrupt(int irq, void *dev_id) { /* disable rx intr */ tx4938_spiptr->cr0 &= ~TXx9_SPCR0_RBSIE; wake_up(&txx9_spi_wait); + + return IRQ_HANDLED; } + static struct irqaction txx9_spi_action = { - txx9_spi_interrupt, 0, 0, "spi", NULL, NULL, + .handler = txx9_spi_interrupt, + .name = "spi", }; void __init txx9_spi_irqinit(int irc_irq) -- cgit v0.10.2 From 6ceb6d3ab2d402cea326320a4143db90a66fd216 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 12:52:16 +0000 Subject: [MIPS] Ocelot G: Fix build error and numerous warnings. The cause of the build errors was a 64-bit kernel being configured in ocelot_g_defconfig without the code being 64-bit proof. Fixed for now by limiting 64-bit selection to SYS_SUPPORTS_64BIT_KERNEL if BROKEN. Signed-off-by: Ralf Baechle diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 9c92b49..74ba763 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -408,7 +408,7 @@ config MOMENCO_OCELOT_C select SWAP_IO_SPACE select SYS_HAS_CPU_RM7000 select SYS_SUPPORTS_32BIT_KERNEL - select SYS_SUPPORTS_64BIT_KERNEL + select SYS_SUPPORTS_64BIT_KERNEL if BROKEN select SYS_SUPPORTS_BIG_ENDIAN help The Ocelot is a MIPS-based Single Board Computer (SBC) made by diff --git a/arch/mips/momentum/ocelot_g/ocelot_pld.h b/arch/mips/momentum/ocelot_g/ocelot_pld.h index fcb8275..95e0534 100644 --- a/arch/mips/momentum/ocelot_g/ocelot_pld.h +++ b/arch/mips/momentum/ocelot_g/ocelot_pld.h @@ -23,8 +23,8 @@ #define OCELOT_REG_INTSET (12) #define OCELOT_REG_INTCLR (13) -#define OCELOT_PLD_WRITE(x, y) writeb(x, OCELOT_CS0_ADDR + OCELOT_REG_##y) -#define OCELOT_PLD_READ(x) readb(OCELOT_CS0_ADDR + OCELOT_REG_##x) - +#define __PLD_REG_TO_ADDR(reg) ((void *) OCELOT_CS0_ADDR + OCELOT_REG_##reg) +#define OCELOT_PLD_WRITE(x, reg) writeb(x, __PLD_REG_TO_ADDR(reg)) +#define OCELOT_PLD_READ(reg) readb(__PLD_REG_TO_ADDR(reg)) #endif /* __MOMENCO_OCELOT_PLD_H__ */ diff --git a/arch/mips/momentum/ocelot_g/setup.c b/arch/mips/momentum/ocelot_g/setup.c index 56ec470..d288f7b 100644 --- a/arch/mips/momentum/ocelot_g/setup.c +++ b/arch/mips/momentum/ocelot_g/setup.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include #include @@ -160,6 +161,10 @@ static void __init setup_l3cache(unsigned long size) printk("Done\n"); } +void __init plat_timer_setup(struct irqaction *irq) +{ +} + void __init plat_mem_setup(void) { void (*l3func)(unsigned long) = (void *) KSEG1ADDR(setup_l3cache); -- cgit v0.10.2 From 408d3258f99458f2dabcb1aa33918250e4864f00 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 14:19:37 +0000 Subject: [MIPS] EMMA 2 / Markeins: Fix build wreckage due to genirq wreckage. I wonder if the original contributor still cares ... Signed-off-by: Ralf Baechle diff --git a/arch/mips/emma2rh/common/irq_emma2rh.c b/arch/mips/emma2rh/common/irq_emma2rh.c index 7c93086..197ed4c 100644 --- a/arch/mips/emma2rh/common/irq_emma2rh.c +++ b/arch/mips/emma2rh/common/irq_emma2rh.c @@ -97,7 +97,7 @@ void emma2rh_irq_init(u32 irq_base) irq_desc[i].status = IRQ_DISABLED; irq_desc[i].action = NULL; irq_desc[i].depth = 1; - irq_desc[i].handler = &emma2rh_irq_controller; + irq_desc[i].chip = &emma2rh_irq_controller; } emma2rh_irq_base = irq_base; diff --git a/arch/mips/emma2rh/markeins/irq_markeins.c b/arch/mips/emma2rh/markeins/irq_markeins.c index f23ae9f..0b36eb0 100644 --- a/arch/mips/emma2rh/markeins/irq_markeins.c +++ b/arch/mips/emma2rh/markeins/irq_markeins.c @@ -86,7 +86,7 @@ void emma2rh_sw_irq_init(u32 irq_base) irq_desc[i].status = IRQ_DISABLED; irq_desc[i].action = NULL; irq_desc[i].depth = 2; - irq_desc[i].handler = &emma2rh_sw_irq_controller; + irq_desc[i].chip = &emma2rh_sw_irq_controller; } emma2rh_sw_irq_base = irq_base; @@ -166,7 +166,7 @@ void emma2rh_gpio_irq_init(u32 irq_base) irq_desc[i].status = IRQ_DISABLED; irq_desc[i].action = NULL; irq_desc[i].depth = 2; - irq_desc[i].handler = &emma2rh_gpio_irq_controller; + irq_desc[i].chip = &emma2rh_gpio_irq_controller; } emma2rh_gpio_irq_base = irq_base; -- cgit v0.10.2 From e30e66becaa237d1753d148703cf8f1301ab27f0 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 14:30:07 +0000 Subject: [MIPS] EMMA 2 / Markeins: Formitting fixes split from actual address fixes. Signed-off-by: Ralf Baechle diff --git a/arch/mips/emma2rh/markeins/platform.c b/arch/mips/emma2rh/markeins/platform.c index 15cc61d..0b884a1 100644 --- a/arch/mips/emma2rh/markeins/platform.c +++ b/arch/mips/emma2rh/markeins/platform.c @@ -83,32 +83,29 @@ struct platform_device i2c_emma_devices[] = { #define EMMA2RH_SERIAL_FLAGS UPF_BOOT_AUTOCONF | UPF_SKIP_TEST static struct plat_serial8250_port platform_serial_ports[] = { - [0] = { - .membase = (void __iomem*)KSEG1ADDR(EMMA2RH_PFUR0_BASE + 3), - .irq = EMMA2RH_IRQ_PFUR0, - .uartclk = EMMA2RH_SERIAL_CLOCK, - .regshift = 4, - .iotype = UPIO_MEM, - .flags = EMMA2RH_SERIAL_FLAGS, - }, - [1] = { - .membase = (void __iomem*)KSEG1ADDR(EMMA2RH_PFUR1_BASE + 3), - .irq = EMMA2RH_IRQ_PFUR1, - .uartclk = EMMA2RH_SERIAL_CLOCK, - .regshift = 4, - .iotype = UPIO_MEM, - .flags = EMMA2RH_SERIAL_FLAGS, - }, - [2] = { - .membase = (void __iomem*)KSEG1ADDR(EMMA2RH_PFUR2_BASE + 3), - .irq = EMMA2RH_IRQ_PFUR2, - .uartclk = EMMA2RH_SERIAL_CLOCK, - .regshift = 4, - .iotype = UPIO_MEM, - .flags = EMMA2RH_SERIAL_FLAGS, - }, - [3] = { - .flags = 0, + [0] = { + .membase= (void __iomem*)KSEG1ADDR(EMMA2RH_PFUR0_BASE + 3), + .irq = EMMA2RH_IRQ_PFUR0, + .uartclk = EMMA2RH_SERIAL_CLOCK, + .regshift = 4, + .iotype = UPIO_MEM, + .flags = EMMA2RH_SERIAL_FLAGS, + }, [1] = { + .membase = (void __iomem*)KSEG1ADDR(EMMA2RH_PFUR1_BASE + 3), + .irq = EMMA2RH_IRQ_PFUR1, + .uartclk = EMMA2RH_SERIAL_CLOCK, + .regshift = 4, + .iotype = UPIO_MEM, + .flags = EMMA2RH_SERIAL_FLAGS, + }, [2] = { + .membase = (void __iomem*)KSEG1ADDR(EMMA2RH_PFUR2_BASE + 3), + .irq = EMMA2RH_IRQ_PFUR2, + .uartclk = EMMA2RH_SERIAL_CLOCK, + .regshift = 4, + .iotype = UPIO_MEM, + .flags = EMMA2RH_SERIAL_FLAGS, + }, [3] = { + .flags = 0, }, }; -- cgit v0.10.2 From 4aad7b726996a2d453d210cd5691730aca087b83 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 14:47:21 +0000 Subject: [MIPS] EMMA 2 / Markeins: Convert to name struct resource initialization. This fixes the wreckage caused by shuffeling the order of struct resource members. Signed-off-by: Ralf Baechle diff --git a/arch/mips/emma2rh/markeins/platform.c b/arch/mips/emma2rh/markeins/platform.c index 0b884a1..be13cec 100644 --- a/arch/mips/emma2rh/markeins/platform.c +++ b/arch/mips/emma2rh/markeins/platform.c @@ -44,18 +44,45 @@ #define I2C_EMMA2RH "emma2rh-iic" /* must be in sync with IIC driver */ static struct resource i2c_emma_resources_0[] = { - { NULL, EMMA2RH_IRQ_PIIC0, EMMA2RH_IRQ_PIIC0, IORESOURCE_IRQ }, - { NULL, KSEG1ADDR(EMMA2RH_PIIC0_BASE), KSEG1ADDR(EMMA2RH_PIIC0_BASE + 0x1000), 0 }, + { + .name = NULL, + .start = EMMA2RH_IRQ_PIIC0, + .end = EMMA2RH_IRQ_PIIC0, + .flags = IORESOURCE_IRQ + }, { + .name = NULL, + .start = KSEG1ADDR(EMMA2RH_PIIC0_BASE), + .end = KSEG1ADDR(EMMA2RH_PIIC0_BASE + 0x1000), + .flags = 0 + }, }; struct resource i2c_emma_resources_1[] = { - { NULL, EMMA2RH_IRQ_PIIC1, EMMA2RH_IRQ_PIIC1, IORESOURCE_IRQ }, - { NULL, KSEG1ADDR(EMMA2RH_PIIC1_BASE), KSEG1ADDR(EMMA2RH_PIIC1_BASE + 0x1000), 0 }, + { + .name = NULL, + .start = EMMA2RH_IRQ_PIIC1, + .end = EMMA2RH_IRQ_PIIC1, + .flags = IORESOURCE_IRQ + }, { + .name = NULL, + .start = KSEG1ADDR(EMMA2RH_PIIC1_BASE), + .end = KSEG1ADDR(EMMA2RH_PIIC1_BASE + 0x1000), + .flags = 0 + }, }; struct resource i2c_emma_resources_2[] = { - { NULL, EMMA2RH_IRQ_PIIC2, EMMA2RH_IRQ_PIIC2, IORESOURCE_IRQ }, - { NULL, KSEG1ADDR(EMMA2RH_PIIC2_BASE), KSEG1ADDR(EMMA2RH_PIIC2_BASE + 0x1000), 0 }, + { + .name = NULL, + .start = EMMA2RH_IRQ_PIIC2, + .end = EMMA2RH_IRQ_PIIC2, + .flags = IORESOURCE_IRQ + }, { + .name = NULL, + .start = KSEG1ADDR(EMMA2RH_PIIC2_BASE), + .end = KSEG1ADDR(EMMA2RH_PIIC2_BASE + 0x1000), + .flags = 0 + }, }; struct platform_device i2c_emma_devices[] = { -- cgit v0.10.2 From 77aec99906367276afedf6f5583671de2ff609be Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 14:59:54 +0000 Subject: [MIPS] EMMA 2 / Markeins: struct resource takes physical addresses. Signed-off-by: Ralf Baechle diff --git a/arch/mips/emma2rh/markeins/platform.c b/arch/mips/emma2rh/markeins/platform.c index be13cec..1156770 100644 --- a/arch/mips/emma2rh/markeins/platform.c +++ b/arch/mips/emma2rh/markeins/platform.c @@ -51,8 +51,8 @@ static struct resource i2c_emma_resources_0[] = { .flags = IORESOURCE_IRQ }, { .name = NULL, - .start = KSEG1ADDR(EMMA2RH_PIIC0_BASE), - .end = KSEG1ADDR(EMMA2RH_PIIC0_BASE + 0x1000), + .start = EMMA2RH_PIIC0_BASE, + .end = EMMA2RH_PIIC0_BASE + 0x1000, .flags = 0 }, }; @@ -65,8 +65,8 @@ struct resource i2c_emma_resources_1[] = { .flags = IORESOURCE_IRQ }, { .name = NULL, - .start = KSEG1ADDR(EMMA2RH_PIIC1_BASE), - .end = KSEG1ADDR(EMMA2RH_PIIC1_BASE + 0x1000), + .start = EMMA2RH_PIIC1_BASE, + .end = EMMA2RH_PIIC1_BASE + 0x1000, .flags = 0 }, }; @@ -79,8 +79,8 @@ struct resource i2c_emma_resources_2[] = { .flags = IORESOURCE_IRQ }, { .name = NULL, - .start = KSEG1ADDR(EMMA2RH_PIIC2_BASE), - .end = KSEG1ADDR(EMMA2RH_PIIC2_BASE + 0x1000), + .start = EMMA2RH_PIIC2_BASE, + .end = EMMA2RH_PIIC2_BASE + 0x1000, .flags = 0 }, }; -- cgit v0.10.2 From e52331845784daeefb78ab3b13efce51af3255bc Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 15:32:27 +0000 Subject: [MIPS] JMR3927: Fixup another victim of the irq pt_regs cleanup. Signed-off-by: Ralf Baechle diff --git a/arch/mips/jmr3927/rbhma3100/irq.c b/arch/mips/jmr3927/rbhma3100/irq.c index 39a0243..de4a238 100644 --- a/arch/mips/jmr3927/rbhma3100/irq.c +++ b/arch/mips/jmr3927/rbhma3100/irq.c @@ -288,6 +288,8 @@ static void tx_branch_likely_bug_fixup(void) static void jmr3927_spurious(void) { + struct pt_regs * regs = get_irq_regs(); + #ifdef CONFIG_TX_BRANCH_LIKELY_BUG_WORKAROUND tx_branch_likely_bug_fixup(); #endif @@ -297,6 +299,7 @@ static void jmr3927_spurious(void) asmlinkage void plat_irq_dispatch(void) { + struct pt_regs * regs = get_irq_regs(); int irq; #ifdef CONFIG_TX_BRANCH_LIKELY_BUG_WORKAROUND -- cgit v0.10.2 From 21e9ac7b2dd96dfca997313bae6d9a8f642635c7 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 21:38:22 +0000 Subject: [MIPS] MIPS doesn't need compat_sys_getdents. Signed-off-by: Ralf Baechle diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h index f1ef98c..ec56aa5 100644 --- a/include/asm-mips/unistd.h +++ b/include/asm-mips/unistd.h @@ -1195,6 +1195,7 @@ type name (atype a,btype b,ctype c,dtype d,etype e,ftype f) \ #endif /* (_MIPS_SIM == _MIPS_SIM_NABI32) || (_MIPS_SIM == _MIPS_SIM_ABI64) */ +#define __ARCH_OMIT_COMPAT_SYS_GETDENTS64 #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_SYS_ALARM -- cgit v0.10.2 From 7ca85ba752e521f1b5ead1f3b91c562cc3910c7b Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 30 Oct 2006 21:42:57 +0000 Subject: [CIFS] Fix readdir breakage when blocksize set too small Do not treat filldir running out of space as an error that needs to be returned. Fixes Redhat bugzilla bug # 211070 Signed-off-by: Steve French diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index b5b0a2a..ed18c39 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -896,6 +896,10 @@ static int cifs_filldir(char *pfindEntry, struct file *file, tmp_inode->i_ino,obj_type); if(rc) { cFYI(1,("filldir rc = %d",rc)); + /* we can not return filldir errors to the caller + since they are "normal" when the stat blocksize + is too small - we return remapped error instead */ + rc = -EOVERFLOW; } dput(tmp_dentry); @@ -1074,6 +1078,11 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) we want to check for that here? */ rc = cifs_filldir(current_entry, file, filldir, direntry, tmp_buf, max_len); + if(rc == -EOVERFLOW) { + rc = 0; + break; + } + file->f_pos++; if(file->f_pos == cifsFile->srch_inf.index_of_last_entry) { -- cgit v0.10.2 From 4b952a9b0877dbe8f0f69b2747abe79e3bbd2865 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 30 Oct 2006 21:46:13 +0000 Subject: [CIFS] Allow null user connections Some servers are configured to only allow null user mounts for guest access. Allow nul user (anonymous) mounts e.g. mount -t cifs //server/share /mnt -o username= Signed-off-by: Steve French diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 1eb9a2e..50afab8 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,6 +1,9 @@ Version 1.46 ------------ Support deep tree mounts. Better support OS/2, Win9x (DOS) time stamps. +Allow null user to be specified on mount ("username="). Do not return +EINVAL on readdir when filldir fails due to overwritten blocksize +(fixes FC problem) Version 1.45 ------------ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 4093d53..71f7791 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -822,10 +822,13 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) } else if (strnicmp(data, "nouser_xattr",12) == 0) { vol->no_xattr = 1; } else if (strnicmp(data, "user", 4) == 0) { - if (!value || !*value) { + if (!value) { printk(KERN_WARNING "CIFS: invalid or missing username\n"); return 1; /* needs_arg; */ + } else if(!*value) { + /* null user, ie anonymous, authentication */ + vol->nullauth = 1; } if (strnlen(value, 200) < 200) { vol->username = value; @@ -1642,6 +1645,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, /* BB fixme parse for domain name here */ cFYI(1, ("Username: %s ", volume_info.username)); + } else if (volume_info.nullauth) { + cFYI(1,("null user")); } else { cifserror("No username specified"); /* In userspace mount helper we can get user name from alternate -- cgit v0.10.2 From d458fd82c9bb536e4a582955e88554a02a92bf78 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 26 Oct 2006 17:15:20 -0700 Subject: [NET] sealevel: uses arp_broken_ops MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On Wed, 25 Oct 2006 18:03:13 +0200 Toralf Förster wrote: > WARNING: "arp_broken_ops" [drivers/net/wan/sealevel.ko] undefined! > make[1]: *** [__modpost] Error 1 > make: *** [modules] Error 2 > > Here's the config: ... > # CONFIG_INET is not set > CONFIG_SEALEVEL_4021=m Sealevel uses arp_broken_ops so it needs to depend on INET. Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig index 58b7efb..b5d0d7f 100644 --- a/drivers/net/wan/Kconfig +++ b/drivers/net/wan/Kconfig @@ -127,7 +127,7 @@ config LANMEDIA # There is no way to detect a Sealevel board. Force it modular config SEALEVEL_4021 tristate "Sealevel Systems 4021 support" - depends on WAN && ISA && m && ISA_DMA_API + depends on WAN && ISA && m && ISA_DMA_API && INET help This is a driver for the Sealevel Systems ACB 56 serial I/O adapter. -- cgit v0.10.2 From 201a95afaa324b23188eeec268f6bb0b4b70b710 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 27 Oct 2006 15:26:21 -0700 Subject: [APPLETALK]: Fix potential OOPS in atalk_sendmsg(). atrtr_find() can return NULL, so do not blindly dereference rt->dev before we check for rt being NULL. Signed-off-by: David S. Miller diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 708e2e0..485e35c 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -1584,7 +1584,6 @@ static int atalk_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr if (usat->sat_addr.s_net || usat->sat_addr.s_node == ATADDR_ANYNODE) { rt = atrtr_find(&usat->sat_addr); - dev = rt->dev; } else { struct atalk_addr at_hint; @@ -1592,7 +1591,6 @@ static int atalk_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr at_hint.s_net = at->src_net; rt = atrtr_find(&at_hint); - dev = rt->dev; } if (!rt) return -ENETUNREACH; -- cgit v0.10.2 From 54489c14c058822f7019648b3718bd3820dee802 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 27 Oct 2006 15:29:47 -0700 Subject: [XFRM] xfrm_user: Fix unaligned accesses. Use memcpy() to move xfrm_address_t objects in and out of netlink messages. The vast majority of xfrm_user was doing this properly, except for copy_from_user_state() and copy_to_user_state(). Signed-off-by: David S. Miller diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 2b2e59d..b43e764 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -323,7 +323,7 @@ static void copy_from_user_state(struct xfrm_state *x, struct xfrm_usersa_info * x->props.replay_window = p->replay_window; x->props.reqid = p->reqid; x->props.family = p->family; - x->props.saddr = p->saddr; + memcpy(&x->props.saddr, &p->saddr, sizeof(x->props.saddr)); x->props.flags = p->flags; } @@ -545,7 +545,7 @@ static void copy_to_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p) memcpy(&p->lft, &x->lft, sizeof(p->lft)); memcpy(&p->curlft, &x->curlft, sizeof(p->curlft)); memcpy(&p->stats, &x->stats, sizeof(p->stats)); - p->saddr = x->props.saddr; + memcpy(&p->saddr, &x->props.saddr, sizeof(p->saddr)); p->mode = x->props.mode; p->replay_window = x->props.replay_window; p->reqid = x->props.reqid; -- cgit v0.10.2 From c8884edd078748905552d667857259e5358e1232 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sun, 29 Oct 2006 15:59:41 -0800 Subject: [NET]: Fix segmentation of linear packets skb_segment fails to segment linear packets correctly because it tries to write all linear parts of the original skb into each segment. This will always panic as each segment only contains enough space for one MSS. This was not detected earlier because linear packets should be rare for GSO. In fact it still remains to be seen what exactly created the linear packets that triggered this bug. Basically the only time this should happen is if someone enables GSO emulation on an interface that does not support SG. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3c23760..f735455 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1946,7 +1946,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) do { struct sk_buff *nskb; skb_frag_t *frag; - int hsize, nsize; + int hsize; int k; int size; @@ -1957,11 +1957,10 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) hsize = skb_headlen(skb) - offset; if (hsize < 0) hsize = 0; - nsize = hsize + doffset; - if (nsize > len + doffset || !sg) - nsize = len + doffset; + if (hsize > len || !sg) + hsize = len; - nskb = alloc_skb(nsize + headroom, GFP_ATOMIC); + nskb = alloc_skb(hsize + doffset + headroom, GFP_ATOMIC); if (unlikely(!nskb)) goto err; -- cgit v0.10.2 From 234af4840135342ab295b4e1219fd35c27fdd439 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 29 Oct 2006 16:03:30 -0800 Subject: [DCCP]: fix printk format warnings Fix printk format warnings: build2.out:net/dccp/ccids/ccid2.c:355: warning: long long unsigned int format, u64 arg (arg 3) build2.out:net/dccp/ccids/ccid2.c:360: warning: long long unsigned int format, u64 arg (arg 3) build2.out:net/dccp/ccids/ccid2.c:482: warning: long long unsigned int format, u64 arg (arg 5) build2.out:net/dccp/ccids/ccid2.c:639: warning: long long unsigned int format, u64 arg (arg 3) build2.out:net/dccp/ccids/ccid2.c:639: warning: long long unsigned int format, u64 arg (arg 4) build2.out:net/dccp/ccids/ccid2.c:674: warning: long long unsigned int format, u64 arg (arg 3) build2.out:net/dccp/ccids/ccid2.c:720: warning: long long unsigned int format, u64 arg (arg 3) Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 2fbb84b..162032b 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -352,14 +352,14 @@ static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, int len) #ifdef CONFIG_IP_DCCP_CCID2_DEBUG ccid2_pr_debug("pipe=%d\n", hctx->ccid2hctx_pipe); - ccid2_pr_debug("Sent: seq=%llu\n", seq); + ccid2_pr_debug("Sent: seq=%llu\n", (unsigned long long)seq); do { struct ccid2_seq *seqp = hctx->ccid2hctx_seqt; while (seqp != hctx->ccid2hctx_seqh) { ccid2_pr_debug("out seq=%llu acked=%d time=%lu\n", - seqp->ccid2s_seq, seqp->ccid2s_acked, - seqp->ccid2s_sent); + (unsigned long long)seqp->ccid2s_seq, + seqp->ccid2s_acked, seqp->ccid2s_sent); seqp = seqp->ccid2s_next; } } while (0); @@ -480,7 +480,8 @@ static inline void ccid2_new_ack(struct sock *sk, /* first measurement */ if (hctx->ccid2hctx_srtt == -1) { ccid2_pr_debug("R: %lu Time=%lu seq=%llu\n", - r, jiffies, seqp->ccid2s_seq); + r, jiffies, + (unsigned long long)seqp->ccid2s_seq); ccid2_change_srtt(hctx, r); hctx->ccid2hctx_rttvar = r >> 1; } else { @@ -636,8 +637,9 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) u64 ackno_end_rl; dccp_set_seqno(&ackno_end_rl, ackno - rl); - ccid2_pr_debug("ackvec start:%llu end:%llu\n", ackno, - ackno_end_rl); + ccid2_pr_debug("ackvec start:%llu end:%llu\n", + (unsigned long long)ackno, + (unsigned long long)ackno_end_rl); /* if the seqno we are analyzing is larger than the * current ackno, then move towards the tail of our * seqnos. @@ -672,7 +674,7 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) seqp->ccid2s_acked = 1; ccid2_pr_debug("Got ack for %llu\n", - seqp->ccid2s_seq); + (unsigned long long)seqp->ccid2s_seq); ccid2_hc_tx_dec_pipe(sk); } if (seqp == hctx->ccid2hctx_seqt) { @@ -718,7 +720,7 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) while (1) { if (!seqp->ccid2s_acked) { ccid2_pr_debug("Packet lost: %llu\n", - seqp->ccid2s_seq); + (unsigned long long)seqp->ccid2s_seq); /* XXX need to traverse from tail -> head in * order to detect multiple congestion events in * one ack vector. -- cgit v0.10.2 From c20e3945c761502b9d5d73ef0ff5f1a84b3a717e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 29 Oct 2006 16:14:55 -0800 Subject: [ETH1394]: Fix unaligned accesses. Several u64 objects are derefernced in situations where the pointer is not guarenteed to be aligned correctly. Use get_unaligned() as needed. Thanks to Will Simoneau for lots of testing and debugging help. Signed-off-by: David S. Miller Acked-by: Stefan Richter diff --git a/drivers/ieee1394/eth1394.c b/drivers/ieee1394/eth1394.c index 8a7b8fa..31e5cc4 100644 --- a/drivers/ieee1394/eth1394.c +++ b/drivers/ieee1394/eth1394.c @@ -64,6 +64,7 @@ #include #include #include +#include #include #include "config_roms.h" @@ -491,7 +492,7 @@ static void ether1394_reset_priv (struct net_device *dev, int set_mtu) int i; struct eth1394_priv *priv = netdev_priv(dev); struct hpsb_host *host = priv->host; - u64 guid = *((u64*)&(host->csr.rom->bus_info_data[3])); + u64 guid = get_unaligned((u64*)&(host->csr.rom->bus_info_data[3])); u16 maxpayload = 1 << (host->csr.max_rec + 1); int max_speed = IEEE1394_SPEED_MAX; @@ -514,8 +515,8 @@ static void ether1394_reset_priv (struct net_device *dev, int set_mtu) ETHER1394_GASP_OVERHEAD))); /* Set our hardware address while we're at it */ - *(u64*)dev->dev_addr = guid; - *(u64*)dev->broadcast = ~0x0ULL; + memcpy(dev->dev_addr, &guid, sizeof(u64)); + memset(dev->broadcast, 0xff, sizeof(u64)); } spin_unlock_irqrestore (&priv->lock, flags); @@ -894,6 +895,7 @@ static inline u16 ether1394_parse_encap(struct sk_buff *skb, u16 maxpayload; struct eth1394_node_ref *node; struct eth1394_node_info *node_info; + __be64 guid; /* Sanity check. MacOSX seems to be sending us 131 in this * field (atleast on my Panther G5). Not sure why. */ @@ -902,8 +904,9 @@ static inline u16 ether1394_parse_encap(struct sk_buff *skb, maxpayload = min(eth1394_speedto_maxpayload[sspd], (u16)(1 << (max_rec + 1))); + guid = get_unaligned(&arp1394->s_uniq_id); node = eth1394_find_node_guid(&priv->ip_node_list, - be64_to_cpu(arp1394->s_uniq_id)); + be64_to_cpu(guid)); if (!node) { return 0; } @@ -931,10 +934,9 @@ static inline u16 ether1394_parse_encap(struct sk_buff *skb, arp_ptr += arp->ar_pln; /* skip over sender IP addr */ if (arp->ar_op == htons(ARPOP_REQUEST)) - /* just set ARP req target unique ID to 0 */ - *((u64*)arp_ptr) = 0; + memset(arp_ptr, 0, sizeof(u64)); else - *((u64*)arp_ptr) = *((u64*)dev->dev_addr); + memcpy(arp_ptr, dev->dev_addr, sizeof(u64)); } /* Now add the ethernet header. */ @@ -1675,8 +1677,10 @@ static int ether1394_tx (struct sk_buff *skb, struct net_device *dev) if (max_payload < dg_size + hdr_type_len[ETH1394_HDR_LF_UF]) priv->bc_dgl++; } else { + __be64 guid = get_unaligned((u64 *)eth->h_dest); + node = eth1394_find_node_guid(&priv->ip_node_list, - be64_to_cpu(*(u64*)eth->h_dest)); + be64_to_cpu(guid)); if (!node) { ret = -EAGAIN; goto fail; -- cgit v0.10.2 From 28cd7752734563d5b0967b96a6bade7a1dc89c7f Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sun, 29 Oct 2006 23:46:42 -0800 Subject: [SCTP]: Always linearise packet on input I was looking at a RHEL5 bug report involving Xen and SCTP (https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=212550). It turns out that SCTP wasn't written to handle skb fragments at all. The absence of any calls to skb_may_pull is testament to that. It just so happens that Xen creates fragmented packets more often than other scenarios (header & data split when going from domU to dom0). That's what caused this bug to show up. Until someone has the time sits down and audits the entire net/sctp directory, here is a conservative and safe solution that simply linearises all packets on input. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/net/sctp/input.c b/net/sctp/input.c index 64f6301..99c0501 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -135,6 +135,9 @@ int sctp_rcv(struct sk_buff *skb) SCTP_INC_STATS_BH(SCTP_MIB_INSCTPPACKS); + if (skb_linearize(skb)) + goto discard_it; + sh = (struct sctphdr *) skb->h.raw; /* Pull up the IP and SCTP headers. */ -- cgit v0.10.2 From a27b58fed90cc5654e2daf1d292cc5bc61be4dd7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 30 Oct 2006 15:06:12 -0800 Subject: [NET]: fix uaccess handling Signed-off-by: Heiko Carstens Signed-off-by: David S. Miller diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index b430cf2..5c31dea 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -329,7 +329,7 @@ error: return err; } -static void raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) +static int raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) { struct iovec *iov; u8 __user *type = NULL; @@ -338,7 +338,7 @@ static void raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) unsigned int i; if (!msg->msg_iov) - return; + return 0; for (i = 0; i < msg->msg_iovlen; i++) { iov = &msg->msg_iov[i]; @@ -360,8 +360,9 @@ static void raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) code = iov->iov_base; if (type && code) { - get_user(fl->fl_icmp_type, type); - get_user(fl->fl_icmp_code, code); + if (get_user(fl->fl_icmp_type, type) || + get_user(fl->fl_icmp_code, code)) + return -EFAULT; probed = 1; } break; @@ -372,6 +373,7 @@ static void raw_probe_proto_opt(struct flowi *fl, struct msghdr *msg) if (probed) break; } + return 0; } static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, @@ -480,8 +482,11 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, .proto = inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol, }; - if (!inet->hdrincl) - raw_probe_proto_opt(&fl, msg); + if (!inet->hdrincl) { + err = raw_probe_proto_opt(&fl, msg); + if (err) + goto done; + } security_sk_classify_flow(sk, &fl); err = ip_route_output_flow(&rt, &fl, sk, !(msg->msg_flags&MSG_DONTWAIT)); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index d09329c..d6dedc4 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -604,7 +604,7 @@ error: return err; } -static void rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) +static int rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) { struct iovec *iov; u8 __user *type = NULL; @@ -616,7 +616,7 @@ static void rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) int i; if (!msg->msg_iov) - return; + return 0; for (i = 0; i < msg->msg_iovlen; i++) { iov = &msg->msg_iov[i]; @@ -638,8 +638,9 @@ static void rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) code = iov->iov_base; if (type && code) { - get_user(fl->fl_icmp_type, type); - get_user(fl->fl_icmp_code, code); + if (get_user(fl->fl_icmp_type, type) || + get_user(fl->fl_icmp_code, code)) + return -EFAULT; probed = 1; } break; @@ -650,7 +651,8 @@ static void rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) /* check if type field is readable or not. */ if (iov->iov_len > 2 - len) { u8 __user *p = iov->iov_base; - get_user(fl->fl_mh_type, &p[2 - len]); + if (get_user(fl->fl_mh_type, &p[2 - len])) + return -EFAULT; probed = 1; } else len += iov->iov_len; @@ -664,6 +666,7 @@ static void rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg) if (probed) break; } + return 0; } static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, @@ -787,7 +790,9 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, opt = ipv6_fixup_options(&opt_space, opt); fl.proto = proto; - rawv6_probe_proto_opt(&fl, msg); + err = rawv6_probe_proto_opt(&fl, msg); + if (err) + goto out; ipv6_addr_copy(&fl.fl6_dst, daddr); if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr)) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index d56e0d2..d527c89 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1075,8 +1075,9 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname, return -EINVAL; len = sizeof(int); val = nlk->flags & NETLINK_RECV_PKTINFO ? 1 : 0; - put_user(len, optlen); - put_user(val, optval); + if (put_user(len, optlen) || + put_user(val, optval)) + return -EFAULT; err = 0; break; default: -- cgit v0.10.2 From bcd620757d3a4ae78ef0ca41adb5d9e400ed92b6 Mon Sep 17 00:00:00 2001 From: James Morris Date: Mon, 30 Oct 2006 15:08:42 -0800 Subject: [IPV6]: fix lockup via /proc/net/ip6_flowlabel There's a bug in the seqfile handling for /proc/net/ip6_flowlabel, where, after finding a flowlabel, the code will loop forever not finding any further flowlabels, first traversing the rest of the hash bucket then just looping. This patch fixes the problem by breaking after the hash bucket has been traversed. Note that this bug can cause lockups and oopses, and is trivially invoked by an unpriveleged user. Signed-off-by: James Morris Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 1d672b0..062e526 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -587,6 +587,8 @@ static struct ip6_flowlabel *ip6fl_get_next(struct seq_file *seq, struct ip6_flo while (!fl) { if (++state->bucket <= FL_HASH_MASK) fl = fl_ht[state->bucket]; + else + break; } return fl; } -- cgit v0.10.2 From 844dc7c88046ecd2e52596730d7cc400d6c3ad67 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 30 Oct 2006 15:12:16 -0800 Subject: [NETFILTER]: remove masq/NAT from ip6tables Kconfig help Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 4bc4e5b..d7c45a9c 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -40,7 +40,7 @@ config IP6_NF_QUEUE To compile it as a module, choose M here. If unsure, say N. config IP6_NF_IPTABLES - tristate "IP6 tables support (required for filtering/masq/NAT)" + tristate "IP6 tables support (required for filtering)" depends on NETFILTER_XTABLES help ip6tables is a general, extensible packet identification framework. -- cgit v0.10.2 From 590bdf7fd2292b47c428111cb1360e312eff207e Mon Sep 17 00:00:00 2001 From: Dmitry Mishin Date: Mon, 30 Oct 2006 15:12:55 -0800 Subject: [NETFILTER]: Missed and reordered checks in {arp,ip,ip6}_tables There is a number of issues in parsing user-provided table in translate_table(). Malicious user with CAP_NET_ADMIN may crash system by passing special-crafted table to the *_tables. The first issue is that mark_source_chains() function is called before entry content checks. In case of standard target, mark_source_chains() function uses t->verdict field in order to determine new position. But the check, that this field leads no further, than the table end, is in check_entry(), which is called later, than mark_source_chains(). The second issue, that there is no check that target_offset points inside entry. If so, *_ITERATE_MATCH macro will follow further, than the entry ends. As a result, we'll have oops or memory disclosure. And the third issue, that there is no check that the target is completely inside entry. Results are the same, as in previous issue. Signed-off-by: Dmitry Mishin Acked-by: Kirill Korotaev Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 0849f1c..413c2d0a 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -466,7 +466,13 @@ static inline int check_entry(struct arpt_entry *e, const char *name, unsigned i return -EINVAL; } + if (e->target_offset + sizeof(struct arpt_entry_target) > e->next_offset) + return -EINVAL; + t = arpt_get_target(e); + if (e->target_offset + t->u.target_size > e->next_offset) + return -EINVAL; + target = try_then_request_module(xt_find_target(NF_ARP, t->u.user.name, t->u.user.revision), "arpt_%s", t->u.user.name); @@ -621,20 +627,18 @@ static int translate_table(const char *name, } } - if (!mark_source_chains(newinfo, valid_hooks, entry0)) { - duprintf("Looping hook\n"); - return -ELOOP; - } - /* Finally, each sanity check must pass */ i = 0; ret = ARPT_ENTRY_ITERATE(entry0, newinfo->size, check_entry, name, size, &i); - if (ret != 0) { - ARPT_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, &i); - return ret; + if (ret != 0) + goto cleanup; + + ret = -ELOOP; + if (!mark_source_chains(newinfo, valid_hooks, entry0)) { + duprintf("Looping hook\n"); + goto cleanup; } /* And one copy for every other CPU */ @@ -643,6 +647,9 @@ static int translate_table(const char *name, memcpy(newinfo->entries[i], entry0, newinfo->size); } + return 0; +cleanup: + ARPT_ENTRY_ITERATE(entry0, newinfo->size, cleanup_entry, &i); return ret; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 4b90927..e2c7f6e 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -547,12 +547,18 @@ check_entry(struct ipt_entry *e, const char *name, unsigned int size, return -EINVAL; } + if (e->target_offset + sizeof(struct ipt_entry_target) > e->next_offset) + return -EINVAL; + j = 0; ret = IPT_MATCH_ITERATE(e, check_match, name, &e->ip, e->comefrom, &j); if (ret != 0) goto cleanup_matches; t = ipt_get_target(e); + ret = -EINVAL; + if (e->target_offset + t->u.target_size > e->next_offset) + goto cleanup_matches; target = try_then_request_module(xt_find_target(AF_INET, t->u.user.name, t->u.user.revision), @@ -712,19 +718,17 @@ translate_table(const char *name, } } - if (!mark_source_chains(newinfo, valid_hooks, entry0)) - return -ELOOP; - /* Finally, each sanity check must pass */ i = 0; ret = IPT_ENTRY_ITERATE(entry0, newinfo->size, check_entry, name, size, &i); - if (ret != 0) { - IPT_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, &i); - return ret; - } + if (ret != 0) + goto cleanup; + + ret = -ELOOP; + if (!mark_source_chains(newinfo, valid_hooks, entry0)) + goto cleanup; /* And one copy for every other CPU */ for_each_possible_cpu(i) { @@ -732,6 +736,9 @@ translate_table(const char *name, memcpy(newinfo->entries[i], entry0, newinfo->size); } + return 0; +cleanup: + IPT_ENTRY_ITERATE(entry0, newinfo->size, cleanup_entry, &i); return ret; } @@ -1463,6 +1470,10 @@ check_compat_entry_size_and_hooks(struct ipt_entry *e, return -EINVAL; } + if (e->target_offset + sizeof(struct compat_xt_entry_target) > + e->next_offset) + return -EINVAL; + off = 0; entry_offset = (void *)e - (void *)base; j = 0; @@ -1472,6 +1483,9 @@ check_compat_entry_size_and_hooks(struct ipt_entry *e, goto cleanup_matches; t = ipt_get_target(e); + ret = -EINVAL; + if (e->target_offset + t->u.target_size > e->next_offset) + goto cleanup_matches; target = try_then_request_module(xt_find_target(AF_INET, t->u.user.name, t->u.user.revision), diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 53bf977..167c2ea 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -586,12 +586,19 @@ check_entry(struct ip6t_entry *e, const char *name, unsigned int size, return -EINVAL; } + if (e->target_offset + sizeof(struct ip6t_entry_target) > + e->next_offset) + return -EINVAL; + j = 0; ret = IP6T_MATCH_ITERATE(e, check_match, name, &e->ipv6, e->comefrom, &j); if (ret != 0) goto cleanup_matches; t = ip6t_get_target(e); + ret = -EINVAL; + if (e->target_offset + t->u.target_size > e->next_offset) + goto cleanup_matches; target = try_then_request_module(xt_find_target(AF_INET6, t->u.user.name, t->u.user.revision), @@ -751,19 +758,17 @@ translate_table(const char *name, } } - if (!mark_source_chains(newinfo, valid_hooks, entry0)) - return -ELOOP; - /* Finally, each sanity check must pass */ i = 0; ret = IP6T_ENTRY_ITERATE(entry0, newinfo->size, check_entry, name, size, &i); - if (ret != 0) { - IP6T_ENTRY_ITERATE(entry0, newinfo->size, - cleanup_entry, &i); - return ret; - } + if (ret != 0) + goto cleanup; + + ret = -ELOOP; + if (!mark_source_chains(newinfo, valid_hooks, entry0)) + goto cleanup; /* And one copy for every other CPU */ for_each_possible_cpu(i) { @@ -771,6 +776,9 @@ translate_table(const char *name, memcpy(newinfo->entries[i], entry0, newinfo->size); } + return 0; +cleanup: + IP6T_ENTRY_ITERATE(entry0, newinfo->size, cleanup_entry, &i); return ret; } -- cgit v0.10.2 From ef4512e76679b4f4997f60f93f8a576a0d20c26b Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Mon, 30 Oct 2006 15:13:28 -0800 Subject: [NETFILTER]: ip_tables: compat error way cleanup This patch adds forgotten compat_flush_offset() call to error way of translate_compat_table(). May lead to table corruption on the next compat_do_replace(). Signed-off-by: Vasily Averin Acked-by: Dmitry Mishin Acked-by: Kirill Korotaev Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index e2c7f6e..0f4835c 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1725,6 +1725,7 @@ free_newinfo: out: return ret; out_unlock: + compat_flush_offsets(); xt_compat_unlock(AF_INET); goto out; } -- cgit v0.10.2 From c073e3fa8b7f9841aa6451885f135656d455f511 Mon Sep 17 00:00:00 2001 From: Martin Josefsson Date: Mon, 30 Oct 2006 15:13:58 -0800 Subject: [NETFILTER]: nf_conntrack: add missing unlock in get_next_corpse() Add missing unlock in get_next_corpse() in nf_conntrack. It was missed during the removal of listhelp.h . Also remove an unneeded use of nf_ct_tuplehash_to_ctrack() in the same function. Should be applied before 2.6.19 is released. Signed-off-by: Martin Josefsson Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 093b3dd..836541e5 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1520,9 +1520,10 @@ get_next_corpse(int (*iter)(struct nf_conn *i, void *data), if (iter(ct, data)) goto found; } + write_unlock_bh(&nf_conntrack_lock); return NULL; found: - atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use); + atomic_inc(&ct->ct_general.use); write_unlock_bh(&nf_conntrack_lock); return ct; } -- cgit v0.10.2 From 920b868ae1dfdac77c5e8c97e7067b23680f043e Mon Sep 17 00:00:00 2001 From: Dmitry Mishin Date: Mon, 30 Oct 2006 15:14:27 -0800 Subject: [NETFILTER]: ip_tables: compat code module refcounting fix This patch fixes bug in iptables modules refcounting on compat error way. As we are getting modules in check_compat_entry_size_and_hooks(), in case of later error, we should put them all in translate_compat_table(), not in the compat_copy_entry_from_user() or compat_copy_match_from_user(), as it is now. Signed-off-by: Dmitry Mishin Acked-by: Vasily Averin Acked-by: Kirill Korotaev Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 0f4835c..8a45543 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1527,7 +1527,7 @@ cleanup_matches: static inline int compat_copy_match_from_user(struct ipt_entry_match *m, void **dstptr, compat_uint_t *size, const char *name, - const struct ipt_ip *ip, unsigned int hookmask, int *i) + const struct ipt_ip *ip, unsigned int hookmask) { struct ipt_entry_match *dm; struct ipt_match *match; @@ -1540,22 +1540,13 @@ static inline int compat_copy_match_from_user(struct ipt_entry_match *m, ret = xt_check_match(match, AF_INET, dm->u.match_size - sizeof(*dm), name, hookmask, ip->proto, ip->invflags & IPT_INV_PROTO); - if (ret) - goto err; - - if (m->u.kernel.match->checkentry + if (!ret && m->u.kernel.match->checkentry && !m->u.kernel.match->checkentry(name, ip, match, dm->data, hookmask)) { duprintf("ip_tables: check failed for `%s'.\n", m->u.kernel.match->name); ret = -EINVAL; - goto err; } - (*i)++; - return 0; - -err: - module_put(m->u.kernel.match->me); return ret; } @@ -1567,19 +1558,18 @@ static int compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr, struct ipt_target *target; struct ipt_entry *de; unsigned int origsize; - int ret, h, j; + int ret, h; ret = 0; origsize = *size; de = (struct ipt_entry *)*dstptr; memcpy(de, e, sizeof(struct ipt_entry)); - j = 0; *dstptr += sizeof(struct compat_ipt_entry); ret = IPT_MATCH_ITERATE(e, compat_copy_match_from_user, dstptr, size, - name, &de->ip, de->comefrom, &j); + name, &de->ip, de->comefrom); if (ret) - goto cleanup_matches; + goto err; de->target_offset = e->target_offset - (origsize - *size); t = ipt_get_target(e); target = t->u.kernel.target; @@ -1613,12 +1603,7 @@ static int compat_copy_entry_from_user(struct ipt_entry *e, void **dstptr, goto err; } ret = 0; - return ret; - err: - module_put(t->u.kernel.target->me); -cleanup_matches: - IPT_MATCH_ITERATE(e, cleanup_match, &j); return ret; } @@ -1632,7 +1617,7 @@ translate_compat_table(const char *name, unsigned int *hook_entries, unsigned int *underflows) { - unsigned int i; + unsigned int i, j; struct xt_table_info *newinfo, *info; void *pos, *entry0, *entry1; unsigned int size; @@ -1650,21 +1635,21 @@ translate_compat_table(const char *name, } duprintf("translate_compat_table: size %u\n", info->size); - i = 0; + j = 0; xt_compat_lock(AF_INET); /* Walk through entries, checking offsets. */ ret = IPT_ENTRY_ITERATE(entry0, total_size, check_compat_entry_size_and_hooks, info, &size, entry0, entry0 + total_size, - hook_entries, underflows, &i, name); + hook_entries, underflows, &j, name); if (ret != 0) goto out_unlock; ret = -EINVAL; - if (i != number) { + if (j != number) { duprintf("translate_compat_table: %u not %u entries\n", - i, number); + j, number); goto out_unlock; } @@ -1723,6 +1708,7 @@ translate_compat_table(const char *name, free_newinfo: xt_free_table_info(newinfo); out: + IPT_ENTRY_ITERATE(entry0, total_size, cleanup_entry, &j); return ret; out_unlock: compat_flush_offsets(); -- cgit v0.10.2 From f8687afefcc821fc47c75775eec87731fe3de360 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Mon, 30 Oct 2006 15:22:15 -0800 Subject: [NetLabel]: protect the CIPSOv4 socket option from setsockopt() This patch makes two changes to protect applications from either removing or tampering with the CIPSOv4 IP option on a socket. The first is the requirement that applications have the CAP_NET_RAW capability to set an IPOPT_CIPSO option on a socket; this prevents untrusted applications from setting their own CIPSOv4 security attributes on the packets they send. The second change is to SELinux and it prevents applications from setting any IPv4 options when there is an IPOPT_CIPSO option already present on the socket; this prevents applications from removing CIPSOv4 security attributes from the packets they send. Signed-off-by: Paul Moore Signed-off-by: James Morris Signed-off-by: David S. Miller diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index e2077a3..6460233 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -1307,7 +1307,8 @@ int cipso_v4_socket_setattr(const struct socket *sock, /* We can't use ip_options_get() directly because it makes a call to * ip_options_get_alloc() which allocates memory with GFP_KERNEL and - * we can't block here. */ + * we won't always have CAP_NET_RAW even though we _always_ want to + * set the IPOPT_CIPSO option. */ opt_len = (buf_len + 3) & ~3; opt = kzalloc(sizeof(*opt) + opt_len, GFP_ATOMIC); if (opt == NULL) { @@ -1317,11 +1318,9 @@ int cipso_v4_socket_setattr(const struct socket *sock, memcpy(opt->__data, buf, buf_len); opt->optlen = opt_len; opt->is_data = 1; + opt->cipso = sizeof(struct iphdr); kfree(buf); buf = NULL; - ret_val = ip_options_compile(opt, NULL); - if (ret_val != 0) - goto socket_setattr_failure; sk_inet = inet_sk(sk); if (sk_inet->is_icsk) { diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 8dabbfc..9f02917 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -443,7 +443,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb) opt->router_alert = optptr - iph; break; case IPOPT_CIPSO: - if (opt->cipso) { + if ((!skb && !capable(CAP_NET_RAW)) || opt->cipso) { pp_ptr = optptr; goto error; } diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index e9969a2..8ab5679 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3313,7 +3313,13 @@ static int selinux_socket_getpeername(struct socket *sock) static int selinux_socket_setsockopt(struct socket *sock,int level,int optname) { - return socket_has_perm(current, sock, SOCKET__SETOPT); + int err; + + err = socket_has_perm(current, sock, SOCKET__SETOPT); + if (err) + return err; + + return selinux_netlbl_socket_setsockopt(sock, level, optname); } static int selinux_socket_getsockopt(struct socket *sock, int level, diff --git a/security/selinux/include/selinux_netlabel.h b/security/selinux/include/selinux_netlabel.h index ecab4bd..9de10cc2 100644 --- a/security/selinux/include/selinux_netlabel.h +++ b/security/selinux/include/selinux_netlabel.h @@ -53,6 +53,9 @@ void selinux_netlbl_sk_security_init(struct sk_security_struct *ssec, void selinux_netlbl_sk_clone_security(struct sk_security_struct *ssec, struct sk_security_struct *newssec); int selinux_netlbl_inode_permission(struct inode *inode, int mask); +int selinux_netlbl_socket_setsockopt(struct socket *sock, + int level, + int optname); #else static inline void selinux_netlbl_cache_invalidate(void) { @@ -114,6 +117,13 @@ static inline int selinux_netlbl_inode_permission(struct inode *inode, { return 0; } + +static inline int selinux_netlbl_socket_setsockopt(struct socket *sock, + int level, + int optname) +{ + return 0; +} #endif /* CONFIG_NETLABEL */ #endif diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index b1f6fb3..bfe1227 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2682,4 +2682,41 @@ u32 selinux_netlbl_socket_getpeersec_dgram(struct sk_buff *skb) return peer_sid; } + +/** + * selinux_netlbl_socket_setsockopt - Do not allow users to remove a NetLabel + * @sock: the socket + * @level: the socket level or protocol + * @optname: the socket option name + * + * Description: + * Check the setsockopt() call and if the user is trying to replace the IP + * options on a socket and a NetLabel is in place for the socket deny the + * access; otherwise allow the access. Returns zero when the access is + * allowed, -EACCES when denied, and other negative values on error. + * + */ +int selinux_netlbl_socket_setsockopt(struct socket *sock, + int level, + int optname) +{ + int rc = 0; + struct inode *inode = SOCK_INODE(sock); + struct sk_security_struct *sksec = sock->sk->sk_security; + struct inode_security_struct *isec = inode->i_security; + struct netlbl_lsm_secattr secattr; + + mutex_lock(&isec->lock); + if (level == IPPROTO_IP && optname == IP_OPTIONS && + sksec->nlbl_state == NLBL_LABELED) { + netlbl_secattr_init(&secattr); + rc = netlbl_socket_getattr(sock, &secattr); + if (rc == 0 && (secattr.cache || secattr.mls_lvl_vld)) + rc = -EACCES; + netlbl_secattr_destroy(&secattr); + } + mutex_unlock(&isec->lock); + + return rc; +} #endif /* CONFIG_NETLABEL */ -- cgit v0.10.2 From 4f4443088b763ca4ac7521e9b4a881b52c294dec Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 30 Oct 2006 18:54:32 -0800 Subject: [SCTP]: Correctly set IP id for SCTP traffic Make SCTP 1-1 style and peeled-off associations behave like TCP when setting IP id. In both cases, we set the inet_sk(sk)->daddr and initialize inet_sk(sk)->id to a random value. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index fac7674..5b4f82f 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -591,7 +591,7 @@ static struct sock *sctp_v4_create_accept_sk(struct sock *sk, newinet->dport = htons(asoc->peer.port); newinet->daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; newinet->pmtudisc = inet->pmtudisc; - newinet->id = 0; + newinet->id = asoc->next_tsn ^ jiffies; newinet->uc_ttl = -1; newinet->mc_loop = 1; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 9f34dec..935bc91 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3372,6 +3372,7 @@ SCTP_STATIC int sctp_do_peeloff(struct sctp_association *asoc, { struct sock *sk = asoc->base.sk; struct socket *sock; + struct inet_sock *inetsk; int err = 0; /* An association cannot be branched off from an already peeled-off @@ -3389,6 +3390,14 @@ SCTP_STATIC int sctp_do_peeloff(struct sctp_association *asoc, * asoc to the newsk. */ sctp_sock_migrate(sk, sock->sk, asoc, SCTP_SOCKET_UDP_HIGH_BANDWIDTH); + + /* Make peeled-off sockets more like 1-1 accepted sockets. + * Set the daddr and initialize id to something more random + */ + inetsk = inet_sk(sock->sk); + inetsk->daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; + inetsk->id = asoc->next_tsn ^ jiffies; + *sockp = sock; return err; -- cgit v0.10.2 From de76e695a5ce19c121ba7e246b45f258be678a75 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 30 Oct 2006 18:55:11 -0800 Subject: [SCTP]: Remove temporary associations from backlog and hash. Every time SCTP creates a temporary association, the stack hashes it, puts it on a list of endpoint associations and increments the backlog. However, the lifetime of a temporary association is the processing time of a current packet and it's destroyed after that. In fact, we don't really want anyone else finding this association. There is no reason to do this extra work. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 27329ce..ed0445f 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -346,11 +346,18 @@ void sctp_association_free(struct sctp_association *asoc) struct list_head *pos, *temp; int i; - list_del(&asoc->asocs); + /* Only real associations count against the endpoint, so + * don't bother for if this is a temporary association. + */ + if (!asoc->temp) { + list_del(&asoc->asocs); - /* Decrement the backlog value for a TCP-style listening socket. */ - if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) - sk->sk_ack_backlog--; + /* Decrement the backlog value for a TCP-style listening + * socket. + */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) + sk->sk_ack_backlog--; + } /* Mark as dead, so other users can know this structure is * going away. diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 35c49ff..9b6b394 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -144,6 +144,13 @@ void sctp_endpoint_add_asoc(struct sctp_endpoint *ep, { struct sock *sk = ep->base.sk; + /* If this is a temporary association, don't bother + * since we'll be removing it shortly and don't + * want anyone to find it anyway. + */ + if (asoc->temp) + return; + /* Now just add it to our list of asocs */ list_add_tail(&asoc->asocs, &ep->asocs); diff --git a/net/sctp/input.c b/net/sctp/input.c index 99c0501..6d82f40 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -771,6 +771,9 @@ static void __sctp_hash_established(struct sctp_association *asoc) /* Add an association to the hash. Local BH-safe. */ void sctp_hash_established(struct sctp_association *asoc) { + if (asoc->temp) + return; + sctp_local_bh_disable(); __sctp_hash_established(asoc); sctp_local_bh_enable(); @@ -804,6 +807,9 @@ static void __sctp_unhash_established(struct sctp_association *asoc) /* Remove association from the hash table. Local BH-safe. */ void sctp_unhash_established(struct sctp_association *asoc) { + if (asoc->temp) + return; + sctp_local_bh_disable(); __sctp_unhash_established(asoc); sctp_local_bh_enable(); -- cgit v0.10.2 From c6817e4c32d8c4118405d2dec30ac1c264349085 Mon Sep 17 00:00:00 2001 From: James Morris Date: Mon, 30 Oct 2006 18:56:06 -0800 Subject: [IPV6]: return EINVAL for invalid address with flowlabel lease request Currently, when an application requests a lease for a flowlabel via the IPV6_FLOWLABEL_MGR socket option, no error is returned if an invalid type of destination address is supplied as part of the request, leading to a silent failure. This patch ensures that EINVAL is returned to the application in this case. Signed-off-by: James Morris Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 062e526..2b45f2d6 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -330,8 +330,10 @@ fl_create(struct in6_flowlabel_req *freq, char __user *optval, int optlen, int * fl->share = freq->flr_share; addr_type = ipv6_addr_type(&freq->flr_dst); if ((addr_type&IPV6_ADDR_MAPPED) - || addr_type == IPV6_ADDR_ANY) + || addr_type == IPV6_ADDR_ANY) { + err = -EINVAL; goto done; + } ipv6_addr_copy(&fl->dst, &freq->flr_dst); atomic_set(&fl->users, 1); switch (fl->share) { -- cgit v0.10.2 From bcb55165d3d1ae3ec95807d118fd6d5956cd127b Mon Sep 17 00:00:00 2001 From: Jun'ichi Nomura Date: Mon, 30 Oct 2006 16:23:45 -0500 Subject: [PATCH] fix bd_claim_by_kobject error handling This fixes bd_claim_by_kobject to release bdev correctly in case that bd_claim succeeds but following add_bd_holder fails. Signed-off-by: Jun'ichi Nomura Signed-off-by: Linus Torvalds diff --git a/fs/block_dev.c b/fs/block_dev.c index 702b88c..b54b0a1 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -751,8 +751,11 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); res = bd_claim(bdev, holder); - if (res == 0) + if (res == 0) { res = add_bd_holder(bdev, bo); + if (res) + bd_release(bdev); + } if (res) free_bd_holder(bo); mutex_unlock(&bdev->bd_mutex); -- cgit v0.10.2 From df6c0cd9a872ebf2298f5d66d8c789f62dbe35fc Mon Sep 17 00:00:00 2001 From: Jun'ichi Nomura Date: Mon, 30 Oct 2006 16:23:56 -0500 Subject: [PATCH] clean up add_bd_holder() add_bd_holder() is called from bd_claim_by_kobject to put a given struct bd_holder in the list if there is no matching entry. There are 3 possible results of add_bd_holder(): 1. there is no matching entry and add the given one to the list 2. there is matching entry, so just increment reference count of the existing one 3. something failed during its course 1 and 2 are successful cases. But for case 2, someone has to free the unused struct bd_holder. The current code frees it inside of add_bd_holder and returns same value 0 for both cases 1 and 2. However, it's natural and less error-prone if caller frees it since it's allocated by the caller. Signed-off-by: Jun'ichi Nomura Signed-off-by: Linus Torvalds diff --git a/fs/block_dev.c b/fs/block_dev.c index b54b0a1..aaa8301 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -642,16 +642,38 @@ static void free_bd_holder(struct bd_holder *bo) } /** + * find_bd_holder - find matching struct bd_holder from the block device + * + * @bdev: struct block device to be searched + * @bo: target struct bd_holder + * + * Returns matching entry with @bo in @bdev->bd_holder_list. + * If found, increment the reference count and return the pointer. + * If not found, returns NULL. + */ +static int find_bd_holder(struct block_device *bdev, struct bd_holder *bo) +{ + struct bd_holder *tmp; + + list_for_each_entry(tmp, &bdev->bd_holder_list, list) + if (tmp->sdir == bo->sdir) { + tmp->count++; + return tmp; + } + + return NULL; +} + +/** * add_bd_holder - create sysfs symlinks for bd_claim() relationship * * @bdev: block device to be bd_claimed * @bo: preallocated and initialized by alloc_bd_holder() * - * If there is no matching entry with @bo in @bdev->bd_holder_list, - * add @bo to the list, create symlinks. + * Add @bo to @bdev->bd_holder_list, create symlinks. * - * Returns 0 if symlinks are created or already there. - * Returns -ve if something fails and @bo can be freed. + * Returns 0 if symlinks are created. + * Returns -ve if something fails. */ static int add_bd_holder(struct block_device *bdev, struct bd_holder *bo) { @@ -661,15 +683,6 @@ static int add_bd_holder(struct block_device *bdev, struct bd_holder *bo) if (!bo) return -EINVAL; - list_for_each_entry(tmp, &bdev->bd_holder_list, list) { - if (tmp->sdir == bo->sdir) { - tmp->count++; - /* We've already done what we need to do here. */ - free_bd_holder(bo); - return 0; - } - } - if (!bd_holder_grab_dirs(bdev, bo)) return -EBUSY; @@ -740,7 +753,7 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, struct kobject *kobj) { int res; - struct bd_holder *bo; + struct bd_holder *bo, *found; if (!kobj) return -EINVAL; @@ -752,11 +765,15 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); res = bd_claim(bdev, holder); if (res == 0) { - res = add_bd_holder(bdev, bo); - if (res) - bd_release(bdev); + found = find_bd_holder(bdev, bo); + if (found == NULL) { + res = add_bd_holder(bdev, bo); + if (res) + bd_release(bdev); + } } - if (res) + + if (res || found) free_bd_holder(bo); mutex_unlock(&bdev->bd_mutex); -- cgit v0.10.2 From 9001f2850ff92b52d7654379e7b7feb72f78f161 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 29 Oct 2006 16:32:31 -0800 Subject: [SPARC64]: Fix Tomatillo/Schizo IRQ handling. The code in schizo_irq_trans_init() should set irq_data->sync_reg to the location of the SYNC register if this is Tomatillo, and set it to zero otherwise. But that is not what it is doing. As a result, non-Tomatillo systems were trying to access a non-existent register resulting in bus errors at the first PCI interrupt. Thanks to Roland Stigge for the bug report. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/prom.c b/arch/sparc64/kernel/prom.c index c60efb3..0917c24 100644 --- a/arch/sparc64/kernel/prom.c +++ b/arch/sparc64/kernel/prom.c @@ -793,7 +793,7 @@ static unsigned int schizo_irq_build(struct device_node *dp, return virt_irq; } -static void schizo_irq_trans_init(struct device_node *dp) +static void __schizo_irq_trans_init(struct device_node *dp, int is_tomatillo) { struct linux_prom64_registers *regs; struct schizo_irq_data *irq_data; @@ -807,11 +807,24 @@ static void schizo_irq_trans_init(struct device_node *dp) dp->irq_trans->data = irq_data; irq_data->pbm_regs = regs[0].phys_addr; - irq_data->sync_reg = regs[3].phys_addr + 0x1a18UL; + if (is_tomatillo) + irq_data->sync_reg = regs[3].phys_addr + 0x1a18UL; + else + irq_data->sync_reg = 0UL; irq_data->portid = of_getintprop_default(dp, "portid", 0); irq_data->chip_version = of_getintprop_default(dp, "version#", 0); } +static void schizo_irq_trans_init(struct device_node *dp) +{ + __schizo_irq_trans_init(dp, 0); +} + +static void tomatillo_irq_trans_init(struct device_node *dp) +{ + __schizo_irq_trans_init(dp, 1); +} + static unsigned int pci_sun4v_irq_build(struct device_node *dp, unsigned int devino, void *_data) @@ -1050,8 +1063,8 @@ static struct irq_trans pci_irq_trans_table[] = { { "pci108e,8001", schizo_irq_trans_init }, { "SUNW,schizo+", schizo_irq_trans_init }, { "pci108e,8002", schizo_irq_trans_init }, - { "SUNW,tomatillo", schizo_irq_trans_init }, - { "pci108e,a801", schizo_irq_trans_init }, + { "SUNW,tomatillo", tomatillo_irq_trans_init }, + { "pci108e,a801", tomatillo_irq_trans_init }, { "SUNW,sun4v-pci", pci_sun4v_irq_trans_init }, }; #endif -- cgit v0.10.2 From 5af47db796cb7e06e9bafb0d75ad98693b55f8b6 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 30 Oct 2006 01:10:20 -0800 Subject: [SPARC64]: Add some missing print_symbol() calls. Signed-off-by: David S. Miller diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c index 68420e2..fe1796c 100644 --- a/arch/sparc64/kernel/traps.c +++ b/arch/sparc64/kernel/traps.c @@ -87,6 +87,7 @@ static void dump_tl1_traplog(struct tl1_traplog *p) i + 1, p->trapstack[i].tstate, p->trapstack[i].tpc, p->trapstack[i].tnpc, p->trapstack[i].tt); + print_symbol("TRAPLOG: TPC<%s>\n", p->trapstack[i].tpc); } } @@ -1134,6 +1135,9 @@ static void cheetah_log_errors(struct pt_regs *regs, struct cheetah_err_info *in printk("%s" "ERROR(%d): TPC[%lx] TNPC[%lx] O7[%lx] TSTATE[%lx]\n", (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id(), regs->tpc, regs->tnpc, regs->u_regs[UREG_I7], regs->tstate); + printk("%s" "ERROR(%d): ", + (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id()); + print_symbol("TPC<%s>\n", regs->tpc); printk("%s" "ERROR(%d): M_SYND(%lx), E_SYND(%lx)%s%s\n", (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id(), (afsr & CHAFSR_M_SYNDROME) >> CHAFSR_M_SYNDROME_SHIFT, @@ -1741,6 +1745,7 @@ void cheetah_plus_parity_error(int type, struct pt_regs *regs) smp_processor_id(), (type & 0x1) ? 'I' : 'D', regs->tpc); + print_symbol(KERN_EMERG "TPC<%s>\n", regs->tpc); panic("Irrecoverable Cheetah+ parity error."); } @@ -1748,6 +1753,7 @@ void cheetah_plus_parity_error(int type, struct pt_regs *regs) smp_processor_id(), (type & 0x1) ? 'I' : 'D', regs->tpc); + print_symbol(KERN_WARNING "TPC<%s>\n", regs->tpc); } struct sun4v_error_entry { @@ -1946,6 +1952,7 @@ void sun4v_itlb_error_report(struct pt_regs *regs, int tl) printk(KERN_EMERG "SUN4V-ITLB: Error at TPC[%lx], tl %d\n", regs->tpc, tl); + print_symbol(KERN_EMERG "SUN4V-ITLB: TPC<%s>\n", regs->tpc); printk(KERN_EMERG "SUN4V-ITLB: vaddr[%lx] ctx[%lx] " "pte[%lx] error[%lx]\n", sun4v_err_itlb_vaddr, sun4v_err_itlb_ctx, @@ -1966,6 +1973,7 @@ void sun4v_dtlb_error_report(struct pt_regs *regs, int tl) printk(KERN_EMERG "SUN4V-DTLB: Error at TPC[%lx], tl %d\n", regs->tpc, tl); + print_symbol(KERN_EMERG "SUN4V-DTLB: TPC<%s>\n", regs->tpc); printk(KERN_EMERG "SUN4V-DTLB: vaddr[%lx] ctx[%lx] " "pte[%lx] error[%lx]\n", sun4v_err_dtlb_vaddr, sun4v_err_dtlb_ctx, -- cgit v0.10.2 From ae99a78af33f00565a05dbbc6ca9b247fed002c5 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 30 Oct 2006 19:37:36 -0800 Subject: Linux 2.6.19-rc4 Not halloween. Not scary. Just a regular -rc release. diff --git a/Makefile b/Makefile index 25b3599..9557619 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 19 -EXTRAVERSION =-rc3 +EXTRAVERSION =-rc4 NAME=Avast! A bilge rat! # *DOCUMENTATION* -- cgit v0.10.2 From 6887d83d6a537b5002edff7efa1a7c600af0ce26 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Tue, 31 Oct 2006 11:44:25 +0900 Subject: sh: Wire up new syscalls. This wires up sys_move_pages, sys_getcpu, and sys_epoll_pwait. Signed-off-by: Paul Mundt diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S index 768334e..ca81976 100644 --- a/arch/sh/kernel/syscalls.S +++ b/arch/sh/kernel/syscalls.S @@ -351,3 +351,6 @@ ENTRY(sys_call_table) .long sys_sync_file_range .long sys_tee /* 315 */ .long sys_vmsplice + .long sys_move_pages + .long sys_getcpu + .long sys_epoll_pwait diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h index f1a0cbc..1c2abde 100644 --- a/include/asm-sh/unistd.h +++ b/include/asm-sh/unistd.h @@ -324,8 +324,11 @@ #define __NR_sync_file_range 314 #define __NR_tee 315 #define __NR_vmsplice 316 +#define __NR_move_pages 317 +#define __NR_getcpu 318 +#define __NR_epoll_pwait 319 -#define NR_syscalls 317 +#define NR_syscalls 320 #ifdef __KERNEL__ -- cgit v0.10.2 From 1aea7e00f6b83d32d359aeb8d670f1f7aaa88d55 Mon Sep 17 00:00:00 2001 From: Kristoffer Ericson Date: Tue, 31 Oct 2006 11:47:27 +0900 Subject: video: Fix include in hp680_bl. The hp6xx.h header moved location, causing the build to fail, fix it up. Signed-off-by: Kristoffer Ericson Signed-off-by: Paul Mundt diff --git a/drivers/video/backlight/hp680_bl.c b/drivers/video/backlight/hp680_bl.c index fe14883..e399321 100644 --- a/drivers/video/backlight/hp680_bl.c +++ b/drivers/video/backlight/hp680_bl.c @@ -19,7 +19,7 @@ #include #include -#include +#include #include #define HP680_MAX_INTENSITY 255 -- cgit v0.10.2 From 1f6c526c409ed7ecdd02469c46ab4b4a50ebec45 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Tue, 31 Oct 2006 12:33:30 +0900 Subject: sh: Update r7780rp_defconfig. Small defconfig updates for R7780RP, enabling SH-RTC. Signed-off-by: Paul Mundt diff --git a/arch/sh/configs/r7780rp_defconfig b/arch/sh/configs/r7780rp_defconfig index 2470364..34e2046 100644 --- a/arch/sh/configs/r7780rp_defconfig +++ b/arch/sh/configs/r7780rp_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18 -# Tue Oct 3 11:32:47 2006 +# Linux kernel version: 2.6.19-rc3 +# Tue Oct 31 12:32:06 2006 # CONFIG_SUPERH=y CONFIG_RWSEM_GENERIC_SPINLOCK=y @@ -10,6 +10,7 @@ CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_HARDIRQS=y CONFIG_GENERIC_IRQ_PROBE=y CONFIG_GENERIC_CALIBRATE_DELAY=y +# CONFIG_GENERIC_TIME is not set CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # @@ -178,7 +179,7 @@ CONFIG_MMU=y CONFIG_PAGE_OFFSET=0x80000000 CONFIG_MEMORY_START=0x08000000 CONFIG_MEMORY_SIZE=0x08000000 -CONFIG_32BIT=y +# CONFIG_32BIT is not set CONFIG_VSYSCALL=y CONFIG_HUGETLB_PAGE_SIZE_64K=y # CONFIG_HUGETLB_PAGE_SIZE_1MB is not set @@ -229,9 +230,7 @@ CONFIG_SH_PCLK_FREQ=32000000 # # DMA support # -CONFIG_SH_DMA=y -CONFIG_NR_ONCHIP_DMA_CHANNELS=6 -# CONFIG_NR_DMA_CHANNELS_BOOL is not set +# CONFIG_SH_DMA is not set # # Companion Chips @@ -259,7 +258,7 @@ CONFIG_ZERO_PAGE_OFFSET=0x00001000 CONFIG_BOOT_LINK_OFFSET=0x00800000 # CONFIG_UBC_WAKEUP is not set CONFIG_CMDLINE_BOOL=y -CONFIG_CMDLINE="mem=128M console=ttySC0,115200 root=/dev/hda1" +CONFIG_CMDLINE="mem=128M console=ttySC0,115200 root=/dev/sda1" # # Bus options @@ -336,6 +335,7 @@ CONFIG_IP_PNP_DHCP=y # CONFIG_INET_TUNNEL is not set CONFIG_INET_XFRM_MODE_TRANSPORT=y CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set @@ -441,76 +441,28 @@ CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024 # CONFIG_ATA_OVER_ETH is not set # +# Misc devices +# +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # -CONFIG_IDE=m -CONFIG_IDE_MAX_HWIFS=4 -CONFIG_BLK_DEV_IDE=m - -# -# Please see Documentation/ide.txt for help/info on IDE drives -# -CONFIG_BLK_DEV_IDE_SATA=y -CONFIG_BLK_DEV_IDEDISK=m -CONFIG_IDEDISK_MULTI_MODE=y -# CONFIG_BLK_DEV_IDECD is not set -# CONFIG_BLK_DEV_IDETAPE is not set -# CONFIG_BLK_DEV_IDEFLOPPY is not set -CONFIG_BLK_DEV_IDESCSI=m -# CONFIG_IDE_TASK_IOCTL is not set - -# -# IDE chipset support/bugfixes -# -CONFIG_IDE_GENERIC=m -CONFIG_BLK_DEV_IDEPCI=y -CONFIG_IDEPCI_SHARE_IRQ=y -# CONFIG_BLK_DEV_OFFBOARD is not set -CONFIG_BLK_DEV_GENERIC=m -# CONFIG_BLK_DEV_OPTI621 is not set -CONFIG_BLK_DEV_IDEDMA_PCI=y -# CONFIG_BLK_DEV_IDEDMA_FORCED is not set -CONFIG_IDEDMA_PCI_AUTO=y -# CONFIG_IDEDMA_ONLYDISK is not set -CONFIG_BLK_DEV_AEC62XX=m -# CONFIG_BLK_DEV_ALI15X3 is not set -# CONFIG_BLK_DEV_AMD74XX is not set -# CONFIG_BLK_DEV_CMD64X is not set -# CONFIG_BLK_DEV_TRIFLEX is not set -# CONFIG_BLK_DEV_CY82C693 is not set -# CONFIG_BLK_DEV_CS5520 is not set -# CONFIG_BLK_DEV_CS5530 is not set -# CONFIG_BLK_DEV_HPT34X is not set -# CONFIG_BLK_DEV_HPT366 is not set -# CONFIG_BLK_DEV_SC1200 is not set -# CONFIG_BLK_DEV_PIIX is not set -# CONFIG_BLK_DEV_IT821X is not set -# CONFIG_BLK_DEV_NS87415 is not set -# CONFIG_BLK_DEV_PDC202XX_OLD is not set -CONFIG_BLK_DEV_PDC202XX_NEW=m -# CONFIG_BLK_DEV_SVWKS is not set -CONFIG_BLK_DEV_SIIMAGE=m -# CONFIG_BLK_DEV_SLC90E66 is not set -# CONFIG_BLK_DEV_TRM290 is not set -# CONFIG_BLK_DEV_VIA82CXXX is not set -# CONFIG_IDE_ARM is not set -CONFIG_BLK_DEV_IDEDMA=y -# CONFIG_IDEDMA_IVB is not set -CONFIG_IDEDMA_AUTO=y -# CONFIG_BLK_DEV_HD is not set +# CONFIG_IDE is not set # # SCSI device support # # CONFIG_RAID_ATTRS is not set -CONFIG_SCSI=m +CONFIG_SCSI=y # CONFIG_SCSI_NETLINK is not set CONFIG_SCSI_PROC_FS=y # # SCSI support type (disk, tape, CD-ROM) # -CONFIG_BLK_DEV_SD=m +CONFIG_BLK_DEV_SD=y # CONFIG_CHR_DEV_ST is not set # CONFIG_CHR_DEV_OSST is not set # CONFIG_BLK_DEV_SR is not set @@ -561,6 +513,7 @@ CONFIG_CHR_DEV_SG=m # CONFIG_SCSI_IPR is not set # CONFIG_SCSI_QLOGIC_1280 is not set # CONFIG_SCSI_QLA_FC is not set +# CONFIG_SCSI_QLA_ISCSI is not set # CONFIG_SCSI_LPFC is not set # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set @@ -570,7 +523,55 @@ CONFIG_CHR_DEV_SG=m # # Serial ATA (prod) and Parallel ATA (experimental) drivers # -# CONFIG_ATA is not set +CONFIG_ATA=y +# CONFIG_SATA_AHCI is not set +# CONFIG_SATA_SVW is not set +# CONFIG_ATA_PIIX is not set +# CONFIG_SATA_MV is not set +# CONFIG_SATA_NV is not set +# CONFIG_PDC_ADMA is not set +# CONFIG_SATA_QSTOR is not set +# CONFIG_SATA_PROMISE is not set +# CONFIG_SATA_SX4 is not set +CONFIG_SATA_SIL=y +# CONFIG_SATA_SIL24 is not set +# CONFIG_SATA_SIS is not set +# CONFIG_SATA_ULI is not set +# CONFIG_SATA_VIA is not set +# CONFIG_SATA_VITESSE is not set +# CONFIG_PATA_ALI is not set +# CONFIG_PATA_AMD is not set +# CONFIG_PATA_ARTOP is not set +# CONFIG_PATA_ATIIXP is not set +# CONFIG_PATA_CMD64X is not set +# CONFIG_PATA_CS5520 is not set +# CONFIG_PATA_CS5530 is not set +# CONFIG_PATA_CYPRESS is not set +# CONFIG_PATA_EFAR is not set +# CONFIG_ATA_GENERIC is not set +# CONFIG_PATA_HPT366 is not set +# CONFIG_PATA_HPT37X is not set +# CONFIG_PATA_HPT3X2N is not set +# CONFIG_PATA_HPT3X3 is not set +# CONFIG_PATA_IT821X is not set +# CONFIG_PATA_JMICRON is not set +# CONFIG_PATA_TRIFLEX is not set +# CONFIG_PATA_MPIIX is not set +# CONFIG_PATA_OLDPIIX is not set +# CONFIG_PATA_NETCELL is not set +# CONFIG_PATA_NS87410 is not set +# CONFIG_PATA_OPTI is not set +# CONFIG_PATA_OPTIDMA is not set +# CONFIG_PATA_PDC_OLD is not set +# CONFIG_PATA_RADISYS is not set +# CONFIG_PATA_RZ1000 is not set +# CONFIG_PATA_SC1200 is not set +# CONFIG_PATA_SERVERWORKS is not set +# CONFIG_PATA_PDC2027X is not set +# CONFIG_PATA_SIL680 is not set +# CONFIG_PATA_SIS is not set +# CONFIG_PATA_VIA is not set +# CONFIG_PATA_WINBOND is not set # # Multi-device support (RAID and LVM) @@ -840,7 +841,6 @@ CONFIG_HW_RANDOM=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -856,6 +856,7 @@ CONFIG_HW_RANDOM=y # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -868,14 +869,9 @@ CONFIG_HWMON=y # CONFIG_HWMON_DEBUG_CHIP is not set # -# Misc devices -# - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -959,7 +955,29 @@ CONFIG_USB_ARCH_HAS_EHCI=y # # Real Time Clock # -# CONFIG_RTC_CLASS is not set +CONFIG_RTC_LIB=y +CONFIG_RTC_CLASS=y +CONFIG_RTC_HCTOSYS=y +CONFIG_RTC_HCTOSYS_DEVICE="rtc0" +# CONFIG_RTC_DEBUG is not set + +# +# RTC interfaces +# +CONFIG_RTC_INTF_SYSFS=y +CONFIG_RTC_INTF_PROC=y +CONFIG_RTC_INTF_DEV=y +# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set + +# +# RTC drivers +# +# CONFIG_RTC_DRV_DS1553 is not set +# CONFIG_RTC_DRV_DS1742 is not set +# CONFIG_RTC_DRV_M48T86 is not set +CONFIG_RTC_DRV_SH=y +# CONFIG_RTC_DRV_TEST is not set +# CONFIG_RTC_DRV_V3020 is not set # # DMA Engine support @@ -984,6 +1002,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -991,6 +1010,7 @@ CONFIG_FS_MBCACHE=y # CONFIG_JFS_FS is not set CONFIG_FS_POSIX_ACL=y # CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set CONFIG_MINIX_FS=y # CONFIG_ROMFS_FS is not set @@ -1027,7 +1047,8 @@ CONFIG_PROC_FS=y CONFIG_PROC_KCORE=y CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y -# CONFIG_TMPFS is not set +CONFIG_TMPFS=y +# CONFIG_TMPFS_POSIX_ACL is not set CONFIG_HUGETLBFS=y CONFIG_HUGETLB_PAGE=y CONFIG_RAMFS=y @@ -1159,6 +1180,7 @@ CONFIG_DEBUG_FS=y # CONFIG_DEBUG_LIST is not set CONFIG_FRAME_POINTER=y CONFIG_FORCED_INLINING=y +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_SH_STANDARD_BIOS is not set # CONFIG_EARLY_SCIF_CONSOLE is not set @@ -1178,9 +1200,9 @@ CONFIG_FORCED_INLINING=y # CONFIG_CRYPTO=y CONFIG_CRYPTO_ALGAPI=y -CONFIG_CRYPTO_BLKCIPHER=m +CONFIG_CRYPTO_BLKCIPHER=y CONFIG_CRYPTO_HASH=y -CONFIG_CRYPTO_MANAGER=m +CONFIG_CRYPTO_MANAGER=y CONFIG_CRYPTO_HMAC=y # CONFIG_CRYPTO_NULL is not set # CONFIG_CRYPTO_MD4 is not set @@ -1191,7 +1213,7 @@ CONFIG_CRYPTO_MD5=y # CONFIG_CRYPTO_WP512 is not set # CONFIG_CRYPTO_TGR192 is not set CONFIG_CRYPTO_ECB=m -CONFIG_CRYPTO_CBC=m +CONFIG_CRYPTO_CBC=y CONFIG_CRYPTO_DES=y # CONFIG_CRYPTO_BLOWFISH is not set # CONFIG_CRYPTO_TWOFISH is not set -- cgit v0.10.2 From bd71ab88deab3358241f22ed6c035c427aacc4e7 Mon Sep 17 00:00:00 2001 From: Jamie Lenehan Date: Tue, 31 Oct 2006 12:35:02 +0900 Subject: sh: Fix IPR-IRQ's for IRQ-chip change breakage. The conversion from IPR-IRQ to IRQ-chip resulted in the ipr data being allocated in a local variable in make_ipr_irq - breaking anything using IPR interrupts. This changes all of the callers of make_ipr_irq to allocate a static structure containing the IPR data which is then passed to make_ipr_irq. This removes the need for make_ipr_irq to allocate any additional space for the IPR information. Signed-off-by: Jamie Lenehan Signed-off-by: Paul Mundt diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c index 1d997ff..f7d0e30 100644 --- a/arch/sh/boards/renesas/hs7751rvoip/setup.c +++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c @@ -15,12 +15,16 @@ #include #include -static void __init hs7751rvoip_init_irq(void) -{ +static struct ipr_data hs77501rvoip_ipr_map[] = { #if defined(CONFIG_HS7751RVOIP_CODEC) - make_ipr_irq(DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); - make_ipr_irq(DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); + { DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, #endif +}; + +static void __init hs7751rvoip_init_irq(void) +{ + make_ipr_irq(hs77501rvoip_ipr_map, ARRAY_SIZE(hs77501rvoip_ipr_map)); init_hs7751rvoip_IRQ(); } diff --git a/arch/sh/boards/renesas/sh7710voipgw/setup.c b/arch/sh/boards/renesas/sh7710voipgw/setup.c index e57e7af..180810b 100644 --- a/arch/sh/boards/renesas/sh7710voipgw/setup.c +++ b/arch/sh/boards/renesas/sh7710voipgw/setup.c @@ -13,6 +13,51 @@ #include #include +static struct ipr_data sh7710voipgw_ipr_map[] = { + { TIMER2_IRQ, TIMER2_IPR_ADDR, TIMER2_IPR_POS, TIMER2_PRIORITY }, + { WDT_IRQ, WDT_IPR_ADDR, WDT_IPR_POS, WDT_PRIORITY }, + + /* SCIF0 */ + { SCIF0_ERI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, SCIF0_PRIORITY }, + { SCIF0_RXI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, SCIF0_PRIORITY }, + { SCIF0_BRI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, SCIF0_PRIORITY }, + { SCIF0_TXI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, SCIF0_PRIORITY }, + + /* DMAC-1 */ + { DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE2_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE3_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + + /* DMAC-2 */ + { DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY }, + { DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY }, + + /* IPSEC */ + { IPSEC_IRQ, IPSEC_IPR_ADDR, IPSEC_IPR_POS, IPSEC_PRIORITY }, + + /* EDMAC */ + { EDMAC0_IRQ, EDMAC0_IPR_ADDR, EDMAC0_IPR_POS, EDMAC0_PRIORITY }, + { EDMAC1_IRQ, EDMAC1_IPR_ADDR, EDMAC1_IPR_POS, EDMAC1_PRIORITY }, + { EDMAC2_IRQ, EDMAC2_IPR_ADDR, EDMAC2_IPR_POS, EDMAC2_PRIORITY }, + + /* SIOF0 */ + { SIOF0_ERI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + { SIOF0_TXI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + { SIOF0_RXI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + { SIOF0_CCI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + + /* SIOF1 */ + { SIOF1_ERI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, SIOF1_PRIORITY }, + { SIOF1_TXI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, SIOF1_PRIORITY }, + { SIOF1_RXI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, SIOF1_PRIORITY }, + { SIOF1_CCI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, SIOF1_PRIORITY }, + + /* SLIC IRQ's */ + { IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, IRQ1_PRIORITY }, + { IRQ2_IRQ, IRQ2_IPR_ADDR, IRQ2_IPR_POS, IRQ2_PRIORITY }, +}; + /* * Initialize IRQ setting */ @@ -37,65 +82,7 @@ static void __init sh7710voipgw_init_irq(void) */ ctrl_outw(0x2aa, INTC_ICR1); - /* Now make IPR interrupts */ - make_ipr_irq(TIMER2_IRQ, TIMER2_IPR_ADDR, - TIMER2_IPR_POS, TIMER2_PRIORITY); - make_ipr_irq(WDT_IRQ, WDT_IPR_ADDR, WDT_IPR_POS, WDT_PRIORITY); - - /* SCIF0 */ - make_ipr_irq(SCIF0_ERI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, - SCIF0_PRIORITY); - make_ipr_irq(SCIF0_RXI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, - SCIF0_PRIORITY); - make_ipr_irq(SCIF0_BRI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, - SCIF0_PRIORITY); - make_ipr_irq(SCIF0_TXI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, - SCIF0_PRIORITY); - - /* DMAC-1 */ - make_ipr_irq(DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); - make_ipr_irq(DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); - make_ipr_irq(DMTE2_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); - make_ipr_irq(DMTE3_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); - - /* DMAC-2 */ - make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY); - make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY); - - /* IPSEC */ - make_ipr_irq(IPSEC_IRQ, IPSEC_IPR_ADDR, IPSEC_IPR_POS, IPSEC_PRIORITY); - - /* EDMAC */ - make_ipr_irq(EDMAC0_IRQ, EDMAC0_IPR_ADDR, EDMAC0_IPR_POS, - EDMAC0_PRIORITY); - make_ipr_irq(EDMAC1_IRQ, EDMAC1_IPR_ADDR, EDMAC1_IPR_POS, - EDMAC1_PRIORITY); - make_ipr_irq(EDMAC2_IRQ, EDMAC2_IPR_ADDR, EDMAC2_IPR_POS, - EDMAC2_PRIORITY); - - /* SIOF0 */ - make_ipr_irq(SIOF0_ERI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, - SIOF0_PRIORITY); - make_ipr_irq(SIOF0_TXI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, - SIOF0_PRIORITY); - make_ipr_irq(SIOF0_RXI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, - SIOF0_PRIORITY); - make_ipr_irq(SIOF0_CCI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, - SIOF0_PRIORITY); - - /* SIOF1 */ - make_ipr_irq(SIOF1_ERI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, - SIOF1_PRIORITY); - make_ipr_irq(SIOF1_TXI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, - SIOF1_PRIORITY); - make_ipr_irq(SIOF1_RXI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, - SIOF1_PRIORITY); - make_ipr_irq(SIOF1_CCI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS, - SIOF1_PRIORITY); - - /* SLIC IRQ's */ - make_ipr_irq(IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, IRQ1_PRIORITY); - make_ipr_irq(IRQ2_IRQ, IRQ2_IPR_ADDR, IRQ2_IPR_POS, IRQ2_PRIORITY); + make_ipr_irq(sh7710voipgw_ipr_map, ARRAY_SIZE(sh7710voipgw_ipr_map)); } /* diff --git a/arch/sh/boards/se/7300/irq.c b/arch/sh/boards/se/7300/irq.c index ad1034f..1279d77 100644 --- a/arch/sh/boards/se/7300/irq.c +++ b/arch/sh/boards/se/7300/irq.c @@ -13,6 +13,17 @@ #include #include +static struct ipr_data se7300_ipr_map[] = { + /* PC_IRQ[0-3] -> IRQ0 (32) */ + { IRQ0_IRQ, IRQ0_IPR_ADDR, IRQ0_IPR_POS, 0x0f - IRQ0_IRQ }, + /* A_IRQ[0-3] -> IRQ1 (33) */ + { IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, 0x0f - IRQ1_IRQ }, + { SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + { DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { VIO_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, +}; + /* * Initialize IRQ setting */ @@ -23,14 +34,7 @@ init_7300se_IRQ(void) ctrl_outw(0xa000, INTC_ICR1); /* IRQ mode; IRQ0,1 enable. */ ctrl_outw(0x0000, PORT_PFCR); /* use F for IRQ[3:0] and SIU. */ - /* PC_IRQ[0-3] -> IRQ0 (32) */ - make_ipr_irq(IRQ0_IRQ, IRQ0_IPR_ADDR, IRQ0_IPR_POS, 0x0f - IRQ0_IRQ); - /* A_IRQ[0-3] -> IRQ1 (33) */ - make_ipr_irq(IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, 0x0f - IRQ1_IRQ); - make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY); - make_ipr_irq(DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(VIO_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); + make_ipr_irq(se7300_ipr_map, ARRAY_SIZE(se7300_ipr_map)); ctrl_outw(0x2000, PA_MRSHPC + 0x0c); /* mrshpc irq enable */ } diff --git a/arch/sh/boards/se/73180/irq.c b/arch/sh/boards/se/73180/irq.c index 2c62b8e..e7200c5 100644 --- a/arch/sh/boards/se/73180/irq.c +++ b/arch/sh/boards/se/73180/irq.c @@ -87,13 +87,38 @@ shmse_irq_demux(int irq) return irq; } +static struct ipr_data se73180_siof0_ipr_map[] = { + { SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, +}; +static struct ipr_data se73180_vpu_ipr_map[] = { + { VPU_IRQ, VPU_IPR_ADDR, VPU_IPR_POS, 8 }, +}; +static struct ipr_data se73180_other_ipr_map[] = { + { DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY }, + { IIC0_ALI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { IIC0_TACKI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { IIC0_WAITI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { IIC0_DTEI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + { SIU_IRQ, SIU_IPR_ADDR, SIU_IPR_POS, SIU_PRIORITY }, + + /* VIO interrupt */ + { CEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, + { BEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, + { VEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, + + { LCDC_IRQ, LCDC_IPR_ADDR, LCDC_IPR_POS, LCDC_PRIORITY }, +}; + /* * Initialize IRQ setting */ void __init init_73180se_IRQ(void) { - make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY); + make_ipr_irq(se73180_siof0_ipr_map, ARRAY_SIZE(se73180_siof0_ipr_map)); ctrl_outw(0x2000, 0xb03fffec); /* mrshpc irq enable */ ctrl_outw(0x2000, 0xb07fffec); /* mrshpc irq enable */ @@ -101,27 +126,11 @@ init_73180se_IRQ(void) ctrl_outw(2 << ((7 - 5) * 2), INTC_ICR1); /* low-level irq */ make_intreq_irq(10); - make_ipr_irq(VPU_IRQ, VPU_IPR_ADDR, VPU_IPR_POS, 8); + make_ipr_irq(se73180_vpu_ipr_map, ARRAY_SIZE(se73180_vpu_ipr_map)); ctrl_outb(0x0f, INTC_IMCR5); /* enable SCIF IRQ */ - make_ipr_irq(DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY); - make_ipr_irq(IIC0_ALI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY); - make_ipr_irq(IIC0_TACKI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, - IIC0_PRIORITY); - make_ipr_irq(IIC0_WAITI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, - IIC0_PRIORITY); - make_ipr_irq(IIC0_DTEI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY); - make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY); - make_ipr_irq(SIU_IRQ, SIU_IPR_ADDR, SIU_IPR_POS, SIU_PRIORITY); - - /* VIO interrupt */ - make_ipr_irq(CEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); - make_ipr_irq(BEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); - make_ipr_irq(VEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); + make_ipr_irq(se73180_other_ipr_map, ARRAY_SIZE(se73180_other_ipr_map)); - make_ipr_irq(LCDC_IRQ, LCDC_IPR_ADDR, LCDC_IPR_POS, LCDC_PRIORITY); ctrl_outw(0x2000, PA_MRSHPC + 0x0c); /* mrshpc irq enable */ } diff --git a/arch/sh/boards/se/7343/irq.c b/arch/sh/boards/se/7343/irq.c index 288b62f..360153e 100644 --- a/arch/sh/boards/se/7343/irq.c +++ b/arch/sh/boards/se/7343/irq.c @@ -102,6 +102,51 @@ shmse_irq_demux(int irq) static struct irqaction irq5 = { no_action, 0, CPU_MASK_NONE, "IRQ5-cascade", NULL, NULL}; +static struct ipr_data se7343_irq5_ipr_map[] = { + { IRQ5_IRQ, IRQ5_IPR_ADDR+2, IRQ5_IPR_POS, IRQ5_PRIORITY }, +}; +static struct ipr_data se7343_siof0_vpu_ipr_map[] = { + { SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + { VPU_IRQ, VPU_IPR_ADDR, VPU_IPR_POS, 8 }, +}; +static struct ipr_data se7343_other_ipr_map[] = { + { DMTE0_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE1_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY }, + { DMTE5_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY }, + + /* I2C block */ + { IIC0_ALI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { IIC0_TACKI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { IIC0_WAITI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + { IIC0_DTEI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY }, + + { IIC1_ALI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY }, + { IIC1_TACKI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY }, + { IIC1_WAITI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY }, + { IIC1_DTEI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY }, + + /* SIOF */ + { SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY }, + + /* SIU */ + { SIU_IRQ, SIU_IPR_ADDR, SIU_IPR_POS, SIU_PRIORITY }, + + /* VIO interrupt */ + { CEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, + { BEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, + { VEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, + + /*MFI interrupt*/ + + { MFI_IRQ, MFI_IPR_ADDR, MFI_IPR_POS, MFI_PRIORITY }, + + /* LCD controller */ + { LCDC_IRQ, LCDC_IPR_ADDR, LCDC_IPR_POS, LCDC_PRIORITY }, +}; + /* * Initialize IRQ setting */ @@ -138,54 +183,17 @@ init_7343se_IRQ(void) /* Setup all external interrupts to be active low */ ctrl_outw(0xaaaa, INTC_ICR1); - make_ipr_irq(IRQ5_IRQ, IRQ5_IPR_ADDR+2, IRQ5_IPR_POS, IRQ5_PRIORITY); + make_ipr_irq(se7343_irq5_ipr_map, ARRAY_SIZE(se7343_irq5_ipr_map)); + setup_irq(IRQ5_IRQ, &irq5); /* Set port control to use IRQ5 */ *(u16 *)0xA4050108 &= ~0xc; - make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY); - make_ipr_irq(VPU_IRQ, VPU_IPR_ADDR, VPU_IPR_POS, 8); + make_ipr_irq(se7343_siof0_vpu_ipr_map, ARRAY_SIZE(se7343_siof0_vpu_ipr_map)); ctrl_outb(0x0f, INTC_IMCR5); /* enable SCIF IRQ */ - make_ipr_irq(DMTE0_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE1_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY); - make_ipr_irq(DMTE5_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY); - - /* I2C block */ - make_ipr_irq(IIC0_ALI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY); - make_ipr_irq(IIC0_TACKI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, - IIC0_PRIORITY); - make_ipr_irq(IIC0_WAITI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, - IIC0_PRIORITY); - make_ipr_irq(IIC0_DTEI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY); - - make_ipr_irq(IIC1_ALI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY); - make_ipr_irq(IIC1_TACKI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, - IIC1_PRIORITY); - make_ipr_irq(IIC1_WAITI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, - IIC1_PRIORITY); - make_ipr_irq(IIC1_DTEI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY); - - /* SIOF */ - make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY); + make_ipr_irq(se7343_other_ipr_map, ARRAY_SIZE(se7343_other_ipr_map)); - /* SIU */ - make_ipr_irq(SIU_IRQ, SIU_IPR_ADDR, SIU_IPR_POS, SIU_PRIORITY); - - /* VIO interrupt */ - make_ipr_irq(CEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); - make_ipr_irq(BEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); - make_ipr_irq(VEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); - - /*MFI interrupt*/ - - make_ipr_irq(MFI_IRQ, MFI_IPR_ADDR, MFI_IPR_POS, MFI_PRIORITY); - - /* LCD controller */ - make_ipr_irq(LCDC_IRQ, LCDC_IPR_ADDR, LCDC_IPR_POS, LCDC_PRIORITY); ctrl_outw(0x2000, PA_MRSHPC + 0x0c); /* mrshpc irq enable */ } diff --git a/arch/sh/boards/se/770x/irq.c b/arch/sh/boards/se/770x/irq.c index cff6700..fcd7cd7 100644 --- a/arch/sh/boards/se/770x/irq.c +++ b/arch/sh/boards/se/770x/irq.c @@ -13,6 +13,48 @@ #include #include +static struct ipr_data se770x_ipr_map[] = { +#if defined(CONFIG_CPU_SUBTYPE_SH7705) + /* This is default value */ + { 0xf-0x2, BCR_ILCRA, 2, 0x2 }, + { 0xf-0xa, BCR_ILCRA, 1, 0xa }, + { 0xf-0x5, BCR_ILCRB, 0, 0x5 }, + { 0xf-0x8, BCR_ILCRC, 1, 0x8 }, + { 0xf-0xc, BCR_ILCRC, 0, 0xc }, + { 0xf-0xe, BCR_ILCRD, 3, 0xe }, + { 0xf-0x3, BCR_ILCRD, 1, 0x3 }, /* LAN */ + { 0xf-0xd, BCR_ILCRE, 2, 0xd }, + { 0xf-0x9, BCR_ILCRE, 1, 0x9 }, + { 0xf-0x1, BCR_ILCRE, 0, 0x1 }, + { 0xf-0xf, BCR_ILCRF, 3, 0xf }, + { 0xf-0xb, BCR_ILCRF, 1, 0xb }, + { 0xf-0x7, BCR_ILCRG, 3, 0x7 }, + { 0xf-0x6, BCR_ILCRG, 2, 0x6 }, + { 0xf-0x4, BCR_ILCRG, 1, 0x4 }, +#else + { 14, BCR_ILCRA, 2, 0x0f-14 }, + { 12, BCR_ILCRA, 1, 0x0f-12 }, + { 8, BCR_ILCRB, 1, 0x0f- 8 }, + { 6, BCR_ILCRC, 3, 0x0f- 6 }, + { 5, BCR_ILCRC, 2, 0x0f- 5 }, + { 4, BCR_ILCRC, 1, 0x0f- 4 }, + { 3, BCR_ILCRC, 0, 0x0f- 3 }, + { 1, BCR_ILCRD, 3, 0x0f- 1 }, + + { 10, BCR_ILCRD, 1, 0x0f-10 }, /* LAN */ + + { 0, BCR_ILCRE, 3, 0x0f- 0 }, /* PCIRQ3 */ + { 11, BCR_ILCRE, 2, 0x0f-11 }, /* PCIRQ2 */ + { 9, BCR_ILCRE, 1, 0x0f- 9 }, /* PCIRQ1 */ + { 7, BCR_ILCRE, 0, 0x0f- 7 }, /* PCIRQ0 */ + + /* #2, #13 are allocated for SLOT IRQ #1 and #2 (for now) */ + /* NOTE: #2 and #13 are not used on PC */ + { 13, BCR_ILCRG, 1, 0x0f-13 }, /* SLOTIRQ2 */ + { 2, BCR_ILCRG, 0, 0x0f- 2 }, /* SLOTIRQ1 */ +#endif +}; + /* * Initialize IRQ setting */ @@ -38,42 +80,6 @@ void __init init_se_IRQ(void) ctrl_outw(0, BCR_ILCRE); ctrl_outw(0, BCR_ILCRF); ctrl_outw(0, BCR_ILCRG); - /* This is default value */ - make_ipr_irq(0xf-0x2, BCR_ILCRA, 2, 0x2); - make_ipr_irq(0xf-0xa, BCR_ILCRA, 1, 0xa); - make_ipr_irq(0xf-0x5, BCR_ILCRB, 0, 0x5); - make_ipr_irq(0xf-0x8, BCR_ILCRC, 1, 0x8); - make_ipr_irq(0xf-0xc, BCR_ILCRC, 0, 0xc); - make_ipr_irq(0xf-0xe, BCR_ILCRD, 3, 0xe); - make_ipr_irq(0xf-0x3, BCR_ILCRD, 1, 0x3); /* LAN */ - make_ipr_irq(0xf-0xd, BCR_ILCRE, 2, 0xd); - make_ipr_irq(0xf-0x9, BCR_ILCRE, 1, 0x9); - make_ipr_irq(0xf-0x1, BCR_ILCRE, 0, 0x1); - make_ipr_irq(0xf-0xf, BCR_ILCRF, 3, 0xf); - make_ipr_irq(0xf-0xb, BCR_ILCRF, 1, 0xb); - make_ipr_irq(0xf-0x7, BCR_ILCRG, 3, 0x7); - make_ipr_irq(0xf-0x6, BCR_ILCRG, 2, 0x6); - make_ipr_irq(0xf-0x4, BCR_ILCRG, 1, 0x4); -#else - make_ipr_irq(14, BCR_ILCRA, 2, 0x0f-14); - make_ipr_irq(12, BCR_ILCRA, 1, 0x0f-12); - make_ipr_irq( 8, BCR_ILCRB, 1, 0x0f- 8); - make_ipr_irq( 6, BCR_ILCRC, 3, 0x0f- 6); - make_ipr_irq( 5, BCR_ILCRC, 2, 0x0f- 5); - make_ipr_irq( 4, BCR_ILCRC, 1, 0x0f- 4); - make_ipr_irq( 3, BCR_ILCRC, 0, 0x0f- 3); - make_ipr_irq( 1, BCR_ILCRD, 3, 0x0f- 1); - - make_ipr_irq(10, BCR_ILCRD, 1, 0x0f-10); /* LAN */ - - make_ipr_irq( 0, BCR_ILCRE, 3, 0x0f- 0); /* PCIRQ3 */ - make_ipr_irq(11, BCR_ILCRE, 2, 0x0f-11); /* PCIRQ2 */ - make_ipr_irq( 9, BCR_ILCRE, 1, 0x0f- 9); /* PCIRQ1 */ - make_ipr_irq( 7, BCR_ILCRE, 0, 0x0f- 7); /* PCIRQ0 */ - - /* #2, #13 are allocated for SLOT IRQ #1 and #2 (for now) */ - /* NOTE: #2 and #13 are not used on PC */ - make_ipr_irq(13, BCR_ILCRG, 1, 0x0f-13); /* SLOTIRQ2 */ - make_ipr_irq( 2, BCR_ILCRG, 0, 0x0f- 2); /* SLOTIRQ1 */ #endif + make_ipr_irq(se770x_ipr_map, ARRAY_SIZE(se770x_ipr_map)); } diff --git a/arch/sh/boards/se/7751/irq.c b/arch/sh/boards/se/7751/irq.c index c607b0a..e4c63a4 100644 --- a/arch/sh/boards/se/7751/irq.c +++ b/arch/sh/boards/se/7751/irq.c @@ -14,53 +14,50 @@ #include #include -/* - * Initialize IRQ setting - */ -void __init init_7751se_IRQ(void) -{ - +static struct ipr_data se7751_ipr_map[] = { /* Leave old Solution Engine code in for reference. */ #if defined(CONFIG_SH_SOLUTION_ENGINE) - /* - * Super I/O (Just mimic PC): - * 1: keyboard - * 3: serial 0 - * 4: serial 1 - * 5: printer - * 6: floppy - * 8: rtc - * 12: mouse - * 14: ide0 - */ - make_ipr_irq(14, BCR_ILCRA, 2, 0x0f-14); - make_ipr_irq(12, BCR_ILCRA, 1, 0x0f-12); - make_ipr_irq( 8, BCR_ILCRB, 1, 0x0f- 8); - make_ipr_irq( 6, BCR_ILCRC, 3, 0x0f- 6); - make_ipr_irq( 5, BCR_ILCRC, 2, 0x0f- 5); - make_ipr_irq( 4, BCR_ILCRC, 1, 0x0f- 4); - make_ipr_irq( 3, BCR_ILCRC, 0, 0x0f- 3); - make_ipr_irq( 1, BCR_ILCRD, 3, 0x0f- 1); - - make_ipr_irq(10, BCR_ILCRD, 1, 0x0f-10); /* LAN */ - - make_ipr_irq( 0, BCR_ILCRE, 3, 0x0f- 0); /* PCIRQ3 */ - make_ipr_irq(11, BCR_ILCRE, 2, 0x0f-11); /* PCIRQ2 */ - make_ipr_irq( 9, BCR_ILCRE, 1, 0x0f- 9); /* PCIRQ1 */ - make_ipr_irq( 7, BCR_ILCRE, 0, 0x0f- 7); /* PCIRQ0 */ - - /* #2, #13 are allocated for SLOT IRQ #1 and #2 (for now) */ - /* NOTE: #2 and #13 are not used on PC */ - make_ipr_irq(13, BCR_ILCRG, 1, 0x0f-13); /* SLOTIRQ2 */ - make_ipr_irq( 2, BCR_ILCRG, 0, 0x0f- 2); /* SLOTIRQ1 */ - + /* + * Super I/O (Just mimic PC): + * 1: keyboard + * 3: serial 0 + * 4: serial 1 + * 5: printer + * 6: floppy + * 8: rtc + * 12: mouse + * 14: ide0 + */ + { 14, BCR_ILCRA, 2, 0x0f-14 }, + { 12, BCR_ILCRA, 1, 0x0f-12 }, + { 8, BCR_ILCRB, 1, 0x0f- 8 }, + { 6, BCR_ILCRC, 3, 0x0f- 6 }, + { 5, BCR_ILCRC, 2, 0x0f- 5 }, + { 4, BCR_ILCRC, 1, 0x0f- 4 }, + { 3, BCR_ILCRC, 0, 0x0f- 3 }, + { 1, BCR_ILCRD, 3, 0x0f- 1 }, + + { 10, BCR_ILCRD, 1, 0x0f-10 }, /* LAN */ + + { 0, BCR_ILCRE, 3, 0x0f- 0 }, /* PCIRQ3 */ + { 11, BCR_ILCRE, 2, 0x0f-11 }, /* PCIRQ2 */ + { 9, BCR_ILCRE, 1, 0x0f- 9 }, /* PCIRQ1 */ + { 7, BCR_ILCRE, 0, 0x0f- 7 }, /* PCIRQ0 */ + + /* #2, #13 are allocated for SLOT IRQ #1 and #2 (for now) */ + /* NOTE: #2 and #13 are not used on PC */ + { 13, BCR_ILCRG, 1, 0x0f-13 }, /* SLOTIRQ2 */ + { 2, BCR_ILCRG, 0, 0x0f- 2 }, /* SLOTIRQ1 */ #elif defined(CONFIG_SH_7751_SOLUTION_ENGINE) - - make_ipr_irq(13, BCR_ILCRD, 3, 2); - - /* Add additional calls to make_ipr_irq() as drivers are added - * and tested. - */ + { 13, BCR_ILCRD, 3, 2 }, + /* Add additional entries here as drivers are added and tested. */ #endif +}; +/* + * Initialize IRQ setting + */ +void __init init_7751se_IRQ(void) +{ + make_ipr_irq(se7751_ipr_map, ARRAY_SIZE(se7751_ipr_map)); } diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c index 137e2ba..5ad1e19 100644 --- a/arch/sh/boards/sh03/setup.c +++ b/arch/sh/boards/sh03/setup.c @@ -14,14 +14,17 @@ #include #include +static struct ipr_data sh03_ipr_map[] = { + { IRL0_IRQ, IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY }, + { IRL1_IRQ, IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY }, + { IRL2_IRQ, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY }, + { IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY }, +}; + static void __init init_sh03_IRQ(void) { ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR); - - make_ipr_irq(IRL0_IRQ, IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY); - make_ipr_irq(IRL1_IRQ, IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY); - make_ipr_irq(IRL2_IRQ, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY); - make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); + make_ipr_irq(sh03_ipr_map, ARRAY_SIZE(sh03_ipr_map)); } extern void *cf_io_base; diff --git a/arch/sh/boards/snapgear/setup.c b/arch/sh/boards/snapgear/setup.c index 540d0bf..650fb36 100644 --- a/arch/sh/boards/snapgear/setup.c +++ b/arch/sh/boards/snapgear/setup.c @@ -68,6 +68,13 @@ module_init(eraseconfig_init); * IRL3 = crypto */ +static struct ipr_data snapgear_ipr_map[] = { + make_ipr_irq(IRL0_IRQ, IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY); + make_ipr_irq(IRL1_IRQ, IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY); + make_ipr_irq(IRL2_IRQ, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY); + make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); +}; + static void __init init_snapgear_IRQ(void) { /* enable individual interrupt mode for externals */ @@ -75,10 +82,7 @@ static void __init init_snapgear_IRQ(void) printk("Setup SnapGear IRQ/IPR ...\n"); - make_ipr_irq(IRL0_IRQ, IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY); - make_ipr_irq(IRL1_IRQ, IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY); - make_ipr_irq(IRL2_IRQ, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY); - make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); + make_ipr_irq(snapgear_ipr_map, ARRAY_SIZE(snapgear_ipr_map)); } /* diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c index 52b66d8..a6046d9 100644 --- a/arch/sh/boards/titan/setup.c +++ b/arch/sh/boards/titan/setup.c @@ -9,15 +9,19 @@ extern void __init pcibios_init_platform(void); +static struct ipr_data titan_ipr_map[] = { + { TITAN_IRQ_WAN, IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY }, + { TITAN_IRQ_LAN, IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY }, + { TITAN_IRQ_MPCIA, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY }, + { TITAN_IRQ_USB, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY }, +}; + static void __init init_titan_irq(void) { /* enable individual interrupt mode for externals */ ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR); - make_ipr_irq( TITAN_IRQ_WAN, IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY); /* PCIRQ0 */ - make_ipr_irq( TITAN_IRQ_LAN, IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY); /* PCIRQ1 */ - make_ipr_irq( TITAN_IRQ_MPCIA, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY); /* PCIRQ2 */ - make_ipr_irq( TITAN_IRQ_USB, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); /* PCIRQ3 */ + make_ipr_irq(titan_ipr_map, ARRAY_SIZE(titan_ipr_map)); } struct sh_machine_vector mv_titan __initmv = { diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c index d8ece20..6607860 100644 --- a/arch/sh/drivers/dma/dma-sh.c +++ b/arch/sh/drivers/dma/dma-sh.c @@ -19,23 +19,34 @@ #include #include "dma-sh.h" -static inline unsigned int get_dmte_irq(unsigned int chan) -{ - unsigned int irq = 0; + +#ifdef CONFIG_CPU_SH4 +static struct ipr_data dmae_ipr_map[] = { + { DMAE_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, +}; +#endif +static struct ipr_data dmte_ipr_map[] = { /* * Normally we could just do DMTE0_IRQ + chan outright, though in the * case of the 7751R, the DMTE IRQs for channels > 4 start right above * the SCIF */ - if (chan < 4) { - irq = DMTE0_IRQ + chan; - } else { -#ifdef DMTE4_IRQ - irq = DMTE4_IRQ + chan - 4; -#endif - } + { DMTE0_IRQ + 0, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE0_IRQ + 1, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE0_IRQ + 2, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE0_IRQ + 3, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE4_IRQ + 0, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE4_IRQ + 1, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE4_IRQ + 2, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, + { DMTE4_IRQ + 3, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY }, +}; +static inline unsigned int get_dmte_irq(unsigned int chan) +{ + unsigned int irq = 0; + if (chan < ARRAY_SIZE(dmte_ipr_map)) + irq = dmte_ipr_map[chan].irq; return irq; } @@ -258,17 +269,16 @@ static int __init sh_dmac_init(void) int i; #ifdef CONFIG_CPU_SH4 - make_ipr_irq(DMAE_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); + make_ipr_irq(dmae_ipr_map, ARRAY_SIZE(dmae_ipr_map)); i = request_irq(DMAE_IRQ, dma_err, IRQF_DISABLED, "DMAC Address Error", 0); if (unlikely(i < 0)) return i; #endif - for (i = 0; i < info->nr_channels; i++) { - int irq = get_dmte_irq(i); - - make_ipr_irq(irq, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY); - } + i = info->nr_channels; + if (i > ARRAY_SIZE(dmte_ipr_map)) + i = ARRAY_SIZE(dmte_ipr_map); + make_ipr_irq(dmte_ipr_map, i); /* * Initialize DMAOR, and clean up any error flags that may have diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c index f799731..a008956 100644 --- a/arch/sh/kernel/cpu/irq/ipr.c +++ b/arch/sh/kernel/cpu/irq/ipr.c @@ -23,24 +23,21 @@ #include #include -struct ipr_data { - unsigned int addr; /* Address of Interrupt Priority Register */ - int shift; /* Shifts of the 16-bit data */ - int priority; /* The priority */ -}; static void disable_ipr_irq(unsigned int irq) { struct ipr_data *p = get_irq_chip_data(irq); + int shift = p->shift*4; /* Set the priority in IPR to 0 */ - ctrl_outw(ctrl_inw(p->addr) & (0xffff ^ (0xf << p->shift)), p->addr); + ctrl_outw(ctrl_inw(p->addr) & (0xffff ^ (0xf << shift)), p->addr); } static void enable_ipr_irq(unsigned int irq) { struct ipr_data *p = get_irq_chip_data(irq); + int shift = p->shift*4; /* Set priority in IPR back to original value */ - ctrl_outw(ctrl_inw(p->addr) | (p->priority << p->shift), p->addr); + ctrl_outw(ctrl_inw(p->addr) | (p->priority << shift), p->addr); } static struct irq_chip ipr_irq_chip = { @@ -50,67 +47,57 @@ static struct irq_chip ipr_irq_chip = { .mask_ack = disable_ipr_irq, }; -void make_ipr_irq(unsigned int irq, unsigned int addr, int pos, int priority) +void make_ipr_irq(struct ipr_data *table, unsigned int nr_irqs) { - struct ipr_data ipr_data; - - disable_irq_nosync(irq); - - ipr_data.addr = addr; - ipr_data.shift = pos*4; /* POSition (0-3) x 4 means shift */ - ipr_data.priority = priority; + int i; - set_irq_chip_and_handler_name(irq, &ipr_irq_chip, + for (i = 0; i < nr_irqs; i++) { + unsigned int irq = table[i].irq; + disable_irq_nosync(irq); + set_irq_chip_and_handler_name(irq, &ipr_irq_chip, handle_level_irq, "level"); - set_irq_chip_data(irq, &ipr_data); - - enable_ipr_irq(irq); + set_irq_chip_data(irq, &table[i]); + enable_ipr_irq(irq); + } } +EXPORT_SYMBOL(make_ipr_irq); -/* XXX: This needs to die a horrible death.. */ -void __init init_IRQ(void) -{ +static struct ipr_data sys_ipr_map[] = { #ifndef CONFIG_CPU_SUBTYPE_SH7780 - make_ipr_irq(TIMER_IRQ, TIMER_IPR_ADDR, TIMER_IPR_POS, TIMER_PRIORITY); - make_ipr_irq(TIMER1_IRQ, TIMER1_IPR_ADDR, TIMER1_IPR_POS, TIMER1_PRIORITY); + { TIMER_IRQ, TIMER_IPR_ADDR, TIMER_IPR_POS, TIMER_PRIORITY }, + { TIMER1_IRQ, TIMER1_IPR_ADDR, TIMER1_IPR_POS, TIMER1_PRIORITY }, #ifdef RTC_IRQ - make_ipr_irq(RTC_IRQ, RTC_IPR_ADDR, RTC_IPR_POS, RTC_PRIORITY); + { RTC_IRQ, RTC_IPR_ADDR, RTC_IPR_POS, RTC_PRIORITY }, #endif - #ifdef SCI_ERI_IRQ - make_ipr_irq(SCI_ERI_IRQ, SCI_IPR_ADDR, SCI_IPR_POS, SCI_PRIORITY); - make_ipr_irq(SCI_RXI_IRQ, SCI_IPR_ADDR, SCI_IPR_POS, SCI_PRIORITY); - make_ipr_irq(SCI_TXI_IRQ, SCI_IPR_ADDR, SCI_IPR_POS, SCI_PRIORITY); + { SCI_ERI_IRQ, SCI_IPR_ADDR, SCI_IPR_POS, SCI_PRIORITY }, + { SCI_RXI_IRQ, SCI_IPR_ADDR, SCI_IPR_POS, SCI_PRIORITY }, + { SCI_TXI_IRQ, SCI_IPR_ADDR, SCI_IPR_POS, SCI_PRIORITY }, #endif - #ifdef SCIF1_ERI_IRQ - make_ipr_irq(SCIF1_ERI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY); - make_ipr_irq(SCIF1_RXI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY); - make_ipr_irq(SCIF1_BRI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY); - make_ipr_irq(SCIF1_TXI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY); + { SCIF1_ERI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY }, + { SCIF1_RXI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY }, + { SCIF1_BRI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY }, + { SCIF1_TXI_IRQ, SCIF1_IPR_ADDR, SCIF1_IPR_POS, SCIF1_PRIORITY }, #endif - #if defined(CONFIG_CPU_SUBTYPE_SH7300) - make_ipr_irq(SCIF0_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, SCIF0_PRIORITY); - make_ipr_irq(DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY); - make_ipr_irq(VIO_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY); + { SCIF0_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS, SCIF0_PRIORITY }, + { DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY }, + { VIO_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY }, #endif - #ifdef SCIF_ERI_IRQ - make_ipr_irq(SCIF_ERI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY); - make_ipr_irq(SCIF_RXI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY); - make_ipr_irq(SCIF_BRI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY); - make_ipr_irq(SCIF_TXI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY); + { SCIF_ERI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY }, + { SCIF_RXI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY }, + { SCIF_BRI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY }, + { SCIF_TXI_IRQ, SCIF_IPR_ADDR, SCIF_IPR_POS, SCIF_PRIORITY }, #endif - #ifdef IRDA_ERI_IRQ - make_ipr_irq(IRDA_ERI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY); - make_ipr_irq(IRDA_RXI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY); - make_ipr_irq(IRDA_BRI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY); - make_ipr_irq(IRDA_TXI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY); + { IRDA_ERI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY }, + { IRDA_RXI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY }, + { IRDA_BRI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY }, + { IRDA_TXI_IRQ, IRDA_IPR_ADDR, IRDA_IPR_POS, IRDA_PRIORITY }, #endif - #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \ defined(CONFIG_CPU_SUBTYPE_SH7706) || \ defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705) @@ -124,14 +111,19 @@ void __init init_IRQ(void) * You should set corresponding bits of PFC to "00" * to enable these interrupts. */ - make_ipr_irq(IRQ0_IRQ, IRQ0_IPR_ADDR, IRQ0_IPR_POS, IRQ0_PRIORITY); - make_ipr_irq(IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, IRQ1_PRIORITY); - make_ipr_irq(IRQ2_IRQ, IRQ2_IPR_ADDR, IRQ2_IPR_POS, IRQ2_PRIORITY); - make_ipr_irq(IRQ3_IRQ, IRQ3_IPR_ADDR, IRQ3_IPR_POS, IRQ3_PRIORITY); - make_ipr_irq(IRQ4_IRQ, IRQ4_IPR_ADDR, IRQ4_IPR_POS, IRQ4_PRIORITY); - make_ipr_irq(IRQ5_IRQ, IRQ5_IPR_ADDR, IRQ5_IPR_POS, IRQ5_PRIORITY); + { IRQ0_IRQ, IRQ0_IPR_ADDR, IRQ0_IPR_POS, IRQ0_PRIORITY }, + { IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, IRQ1_PRIORITY }, + { IRQ2_IRQ, IRQ2_IPR_ADDR, IRQ2_IPR_POS, IRQ2_PRIORITY }, + { IRQ3_IRQ, IRQ3_IPR_ADDR, IRQ3_IPR_POS, IRQ3_PRIORITY }, + { IRQ4_IRQ, IRQ4_IPR_ADDR, IRQ4_IPR_POS, IRQ4_PRIORITY }, + { IRQ5_IRQ, IRQ5_IPR_ADDR, IRQ5_IPR_POS, IRQ5_PRIORITY }, #endif #endif +}; + +void __init init_IRQ(void) +{ + make_ipr_irq(sys_ipr_map, ARRAY_SIZE(sys_ipr_map)); #ifdef CONFIG_CPU_HAS_PINT_IRQ init_IRQ_pint(); @@ -153,5 +145,3 @@ int ipr_irq_demux(int irq) return irq; } #endif - -EXPORT_SYMBOL(make_ipr_irq); diff --git a/arch/sh/kernel/cpu/irq/pint.c b/arch/sh/kernel/cpu/irq/pint.c index 17f47b3..f600077 100644 --- a/arch/sh/kernel/cpu/irq/pint.c +++ b/arch/sh/kernel/cpu/irq/pint.c @@ -84,12 +84,16 @@ void make_pint_irq(unsigned int irq) disable_pint_irq(irq); } +static struct ipr_data pint_ipr_map[] = { + { PINT0_IRQ, PINT0_IPR_ADDR, PINT0_IPR_POS, PINT0_PRIORITY }, + { PINT8_IRQ, PINT8_IPR_ADDR, PINT8_IPR_POS, PINT8_PRIORITY }, +}; + void __init init_IRQ_pint(void) { int i; - make_ipr_irq(PINT0_IRQ, PINT0_IPR_ADDR, PINT0_IPR_POS, PINT0_PRIORITY); - make_ipr_irq(PINT8_IRQ, PINT8_IPR_ADDR, PINT8_IPR_POS, PINT8_PRIORITY); + make_ipr_irq(pint_ipr_map, ARRAY_SIZE(pint_ipr_map)); enable_irq(PINT0_IRQ); enable_irq(PINT8_IRQ); diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h index 7596ab8..6cd3e9e 100644 --- a/include/asm-sh/irq.h +++ b/include/asm-sh/irq.h @@ -327,11 +327,17 @@ extern unsigned short *irq_mask_register; */ void init_IRQ_pint(void); +struct ipr_data { + unsigned int irq; + unsigned int addr; /* Address of Interrupt Priority Register */ + int shift; /* Shifts of the 16-bit data */ + int priority; /* The priority */ +}; + /* * Function for "on chip support modules". */ -extern void make_ipr_irq(unsigned int irq, unsigned int addr, - int pos, int priority); +extern void make_ipr_irq(struct ipr_data *table, unsigned int nr_irqs); extern void make_imask_irq(unsigned int irq); #if defined(CONFIG_CPU_SUBTYPE_SH7300) -- cgit v0.10.2 From 4731f2dfd5049b7a2b3b5a7131525f6151855f0d Mon Sep 17 00:00:00 2001 From: Jamie Lenehan Date: Tue, 31 Oct 2006 12:36:35 +0900 Subject: sh: Titan defconfig update. Small defconfig update for titan for 2.6.19-rc3, adding SH-RTC. Signed-off-by: Jamie Lenehan Signed-off-by: Paul Mundt diff --git a/arch/sh/configs/titan_defconfig b/arch/sh/configs/titan_defconfig index 5e81754..41049cf 100644 --- a/arch/sh/configs/titan_defconfig +++ b/arch/sh/configs/titan_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18 -# Tue Oct 3 12:59:14 2006 +# Linux kernel version: 2.6.19-rc3 +# Mon Oct 30 18:04:49 2006 # CONFIG_SUPERH=y CONFIG_RWSEM_GENERIC_SPINLOCK=y @@ -10,6 +10,7 @@ CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_HARDIRQS=y CONFIG_GENERIC_IRQ_PROBE=y CONFIG_GENERIC_CALIBRATE_DELAY=y +# CONFIG_GENERIC_TIME is not set CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # @@ -23,7 +24,7 @@ CONFIG_INIT_ENV_ARG_LIMIT=32 # General setup # CONFIG_LOCALVERSION="" -CONFIG_LOCALVERSION_AUTO=y +# CONFIG_LOCALVERSION_AUTO is not set CONFIG_SWAP=y CONFIG_SYSVIPC=y # CONFIG_IPC_NS is not set @@ -236,8 +237,8 @@ CONFIG_HZ_250=y CONFIG_HZ=250 # CONFIG_KEXEC is not set # CONFIG_SMP is not set -CONFIG_PREEMPT_NONE=y -# CONFIG_PREEMPT_VOLUNTARY is not set +# CONFIG_PREEMPT_NONE is not set +CONFIG_PREEMPT_VOLUNTARY=y # CONFIG_PREEMPT is not set # @@ -247,7 +248,7 @@ CONFIG_ZERO_PAGE_OFFSET=0x00001000 CONFIG_BOOT_LINK_OFFSET=0x009e0000 # CONFIG_UBC_WAKEUP is not set CONFIG_CMDLINE_BOOL=y -CONFIG_CMDLINE="console=ttySC1,38400N81 root=/dev/nfs ip=:::::eth1:autoconf" +CONFIG_CMDLINE="console=ttySC1,38400N81 root=/dev/nfs ip=:::::eth1:autoconf rw" # # Bus options @@ -334,6 +335,7 @@ CONFIG_INET_XFRM_TUNNEL=y CONFIG_INET_TUNNEL=y CONFIG_INET_XFRM_MODE_TRANSPORT=y CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=m CONFIG_INET_TCP_DIAG=m # CONFIG_TCP_CONG_ADVANCED is not set @@ -355,9 +357,10 @@ CONFIG_INET6_XFRM_TUNNEL=y CONFIG_INET6_TUNNEL=y CONFIG_INET6_XFRM_MODE_TRANSPORT=y CONFIG_INET6_XFRM_MODE_TUNNEL=y +CONFIG_INET6_XFRM_MODE_BEET=y # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=m CONFIG_IPV6_TUNNEL=y -# CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set # CONFIG_NETWORK_SECMARK is not set CONFIG_NETFILTER=y @@ -714,6 +717,12 @@ CONFIG_BLK_DEV_INITRD=y CONFIG_ATA_OVER_ETH=m # +# Misc devices +# +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + +# # ATA/ATAPI/MFM/RLL support # # CONFIG_IDE is not set @@ -778,9 +787,9 @@ CONFIG_CHR_DEV_SG=m # CONFIG_SCSI_INIA100 is not set # CONFIG_SCSI_STEX is not set # CONFIG_SCSI_SYM53C8XX_2 is not set -# CONFIG_SCSI_IPR is not set # CONFIG_SCSI_QLOGIC_1280 is not set # CONFIG_SCSI_QLA_FC is not set +# CONFIG_SCSI_QLA_ISCSI is not set # CONFIG_SCSI_LPFC is not set # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set @@ -1095,7 +1104,6 @@ CONFIG_HW_RANDOM=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -1124,14 +1132,9 @@ CONFIG_HWMON=y # CONFIG_HWMON_DEBUG_CHIP is not set # -# Misc devices -# - -# # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -1177,9 +1180,9 @@ CONFIG_USB_DEVICEFS=y # USB Host Controller Drivers # CONFIG_USB_EHCI_HCD=y -# CONFIG_USB_EHCI_SPLIT_ISO is not set -# CONFIG_USB_EHCI_ROOT_HUB_TT is not set -# CONFIG_USB_EHCI_TT_NEWSCHED is not set +CONFIG_USB_EHCI_SPLIT_ISO=y +CONFIG_USB_EHCI_ROOT_HUB_TT=y +CONFIG_USB_EHCI_TT_NEWSCHED=y # CONFIG_USB_ISP116X_HCD is not set CONFIG_USB_OHCI_HCD=y # CONFIG_USB_OHCI_BIG_ENDIAN is not set @@ -1235,7 +1238,6 @@ CONFIG_USB_STORAGE=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set -# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -1246,11 +1248,20 @@ CONFIG_USB_STORAGE=y # # USB Network Adapters # -# CONFIG_USB_CATC is not set -# CONFIG_USB_KAWETH is not set -# CONFIG_USB_PEGASUS is not set -# CONFIG_USB_RTL8150 is not set -# CONFIG_USB_USBNET is not set +CONFIG_USB_CATC=m +CONFIG_USB_KAWETH=m +CONFIG_USB_PEGASUS=m +CONFIG_USB_RTL8150=m +CONFIG_USB_USBNET=m +CONFIG_USB_NET_AX8817X=m +CONFIG_USB_NET_CDCETHER=m +# CONFIG_USB_NET_GL620A is not set +CONFIG_USB_NET_NET1080=m +CONFIG_USB_NET_PLUSB=m +# CONFIG_USB_NET_MCS7830 is not set +# CONFIG_USB_NET_RNDIS_HOST is not set +# CONFIG_USB_NET_CDC_SUBSET is not set +CONFIG_USB_NET_ZAURUS=m CONFIG_USB_MON=y # @@ -1285,6 +1296,7 @@ CONFIG_USB_SERIAL_ARK3116=m # CONFIG_USB_SERIAL_KLSI is not set # CONFIG_USB_SERIAL_KOBIL_SCT is not set # CONFIG_USB_SERIAL_MCT_U232 is not set +# CONFIG_USB_SERIAL_MOS7720 is not set # CONFIG_USB_SERIAL_MOS7840 is not set # CONFIG_USB_SERIAL_NAVMAN is not set CONFIG_USB_SERIAL_PL2303=m @@ -1316,6 +1328,7 @@ CONFIG_USB_SERIAL_PL2303=m # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1357,7 +1370,26 @@ CONFIG_USB_SERIAL_PL2303=m # # Real Time Clock # -# CONFIG_RTC_CLASS is not set +CONFIG_RTC_LIB=m +CONFIG_RTC_CLASS=m + +# +# RTC interfaces +# +CONFIG_RTC_INTF_SYSFS=m +CONFIG_RTC_INTF_PROC=m +CONFIG_RTC_INTF_DEV=m +# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set + +# +# RTC drivers +# +# CONFIG_RTC_DRV_DS1553 is not set +# CONFIG_RTC_DRV_DS1742 is not set +# CONFIG_RTC_DRV_M48T86 is not set +CONFIG_RTC_DRV_SH=m +# CONFIG_RTC_DRV_TEST is not set +# CONFIG_RTC_DRV_V3020 is not set # # DMA Engine support @@ -1380,8 +1412,12 @@ CONFIG_EXT2_FS=y # CONFIG_EXT2_FS_XIP is not set CONFIG_EXT3_FS=y # CONFIG_EXT3_FS_XATTR is not set +CONFIG_EXT4DEV_FS=m +# CONFIG_EXT4DEV_FS_XATTR is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set +CONFIG_JBD2=m +# CONFIG_JBD2_DEBUG is not set CONFIG_REISERFS_FS=m # CONFIG_REISERFS_CHECK is not set # CONFIG_REISERFS_PROC_INFO is not set @@ -1393,9 +1429,10 @@ CONFIG_XFS_FS=m # CONFIG_XFS_SECURITY is not set # CONFIG_XFS_POSIX_ACL is not set # CONFIG_XFS_RT is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set -# CONFIG_ROMFS_FS is not set +CONFIG_ROMFS_FS=y CONFIG_INOTIFY=y CONFIG_INOTIFY_USER=y # CONFIG_QUOTA is not set @@ -1480,7 +1517,12 @@ CONFIG_SUNRPC=y # CONFIG_RPCSEC_GSS_SPKM3 is not set CONFIG_SMB_FS=m # CONFIG_SMB_NLS_DEFAULT is not set -# CONFIG_CIFS is not set +CONFIG_CIFS=m +# CONFIG_CIFS_STATS is not set +CONFIG_CIFS_WEAK_PW_HASH=y +# CONFIG_CIFS_XATTR is not set +# CONFIG_CIFS_DEBUG2 is not set +# CONFIG_CIFS_EXPERIMENTAL is not set # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set @@ -1583,9 +1625,10 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_DEBUG_LIST is not set # CONFIG_FRAME_POINTER is not set # CONFIG_FORCED_INLINING is not set +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_SH_STANDARD_BIOS is not set -CONFIG_EARLY_SCIF_CONSOLE=y +# CONFIG_EARLY_SCIF_CONSOLE is not set # CONFIG_EARLY_PRINTK is not set # CONFIG_DEBUG_STACKOVERFLOW is not set # CONFIG_DEBUG_STACK_USAGE is not set @@ -1605,7 +1648,7 @@ CONFIG_CRYPTO=y CONFIG_CRYPTO_ALGAPI=y CONFIG_CRYPTO_BLKCIPHER=y CONFIG_CRYPTO_HASH=y -CONFIG_CRYPTO_MANAGER=m +CONFIG_CRYPTO_MANAGER=y CONFIG_CRYPTO_HMAC=y CONFIG_CRYPTO_NULL=m CONFIG_CRYPTO_MD4=m @@ -1615,7 +1658,7 @@ CONFIG_CRYPTO_SHA256=m CONFIG_CRYPTO_SHA512=m CONFIG_CRYPTO_WP512=m CONFIG_CRYPTO_TGR192=m -CONFIG_CRYPTO_ECB=m +CONFIG_CRYPTO_ECB=y CONFIG_CRYPTO_CBC=y CONFIG_CRYPTO_DES=y CONFIG_CRYPTO_BLOWFISH=m -- cgit v0.10.2 From 2e7a7426282bfa2d7dff6eddc5485af8c79a68f3 Mon Sep 17 00:00:00 2001 From: Erez Zilber Date: Sun, 22 Oct 2006 10:28:38 +0200 Subject: IB/iser: Start connection after enabling iSER When a connection is started (a new connection or a recovered one), iSER should prepare its resources for full-featured mode and only then notify the iSCSI layer that it is ready to start queueing commands. Signed-off-by: Erez Zilber Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index eb6f98d..9b2041e 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -363,11 +363,11 @@ iscsi_iser_conn_start(struct iscsi_cls_conn *cls_conn) struct iscsi_conn *conn = cls_conn->dd_data; int err; - err = iscsi_conn_start(cls_conn); + err = iser_conn_set_full_featured_mode(conn); if (err) return err; - return iser_conn_set_full_featured_mode(conn); + return iscsi_conn_start(cls_conn); } static struct iscsi_transport iscsi_iser_transport; -- cgit v0.10.2 From 255d0c14b3757e8bd85add874e4dca4c3621b39e Mon Sep 17 00:00:00 2001 From: Krishna Kumar Date: Tue, 24 Oct 2006 13:22:28 -0700 Subject: RDMA/cma: rdma_bind_addr() leaks a cma_dev reference count rdma_bind_addr() leaks a cma_dev reference count in failure case. Signed-off-by: Krishna Kumar Signed-off-by: Sean Hefty diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 9ae4f3a..d8ca3c1 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -1762,22 +1762,29 @@ int rdma_bind_addr(struct rdma_cm_id *id, struct sockaddr *addr) if (!cma_any_addr(addr)) { ret = rdma_translate_ip(addr, &id->route.addr.dev_addr); - if (!ret) { - mutex_lock(&lock); - ret = cma_acquire_dev(id_priv); - mutex_unlock(&lock); - } if (ret) - goto err; + goto err1; + + mutex_lock(&lock); + ret = cma_acquire_dev(id_priv); + mutex_unlock(&lock); + if (ret) + goto err1; } memcpy(&id->route.addr.src_addr, addr, ip_addr_size(addr)); ret = cma_get_port(id_priv); if (ret) - goto err; + goto err2; return 0; -err: +err2: + if (!cma_any_addr(addr)) { + mutex_lock(&lock); + cma_detach_from_dev(id_priv); + mutex_unlock(&lock); + } +err1: cma_comp_exch(id_priv, CMA_ADDR_BOUND, CMA_IDLE); return ret; } -- cgit v0.10.2 From 04d03bc576f244bfa9692452aab83fa357ac0d57 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 25 Oct 2006 15:28:08 +1000 Subject: IB/ehca: Fix eHCA driver compilation for uniprocessor The eHCA driver does not compile for a uniprocessor configuration (CONFIG_SMP=n), due to H_SUCCESS and other symbols being undefined. This fixes it. Signed-off-by: Paul Mackerras Acked-by: Hoang-Nam Nguyen Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/ehca/ehca_tools.h b/drivers/infiniband/hw/ehca/ehca_tools.h index 809da3e..973c4b5 100644 --- a/drivers/infiniband/hw/ehca/ehca_tools.h +++ b/drivers/infiniband/hw/ehca/ehca_tools.h @@ -63,6 +63,7 @@ #include #include #include +#include extern int ehca_debug_level; -- cgit v0.10.2 From 8de94ce19dd3c6fc6e9d9658da11cf3d76841ee5 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Fri, 27 Oct 2006 17:28:35 -0500 Subject: IB/amso1100: Use dma_alloc_coherent() instead of kmalloc/dma_map_single The Ammasso driver needs to use dma_alloc_coherent() for allocating memory that will be used by the HW for dma. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/amso1100/c2_alloc.c b/drivers/infiniband/hw/amso1100/c2_alloc.c index 028a60b..0315f99 100644 --- a/drivers/infiniband/hw/amso1100/c2_alloc.c +++ b/drivers/infiniband/hw/amso1100/c2_alloc.c @@ -42,13 +42,14 @@ static int c2_alloc_mqsp_chunk(struct c2_dev *c2dev, gfp_t gfp_mask, { int i; struct sp_chunk *new_head; + dma_addr_t dma_addr; - new_head = (struct sp_chunk *) __get_free_page(gfp_mask); + new_head = dma_alloc_coherent(&c2dev->pcidev->dev, PAGE_SIZE, + &dma_addr, gfp_mask); if (new_head == NULL) return -ENOMEM; - new_head->dma_addr = dma_map_single(c2dev->ibdev.dma_device, new_head, - PAGE_SIZE, DMA_FROM_DEVICE); + new_head->dma_addr = dma_addr; pci_unmap_addr_set(new_head, mapping, new_head->dma_addr); new_head->next = NULL; @@ -80,10 +81,8 @@ void c2_free_mqsp_pool(struct c2_dev *c2dev, struct sp_chunk *root) while (root) { next = root->next; - dma_unmap_single(c2dev->ibdev.dma_device, - pci_unmap_addr(root, mapping), PAGE_SIZE, - DMA_FROM_DEVICE); - __free_page((struct page *) root); + dma_free_coherent(&c2dev->pcidev->dev, PAGE_SIZE, root, + pci_unmap_addr(root, mapping)); root = next; } } diff --git a/drivers/infiniband/hw/amso1100/c2_cq.c b/drivers/infiniband/hw/amso1100/c2_cq.c index 9d7bcc5..05c9154 100644 --- a/drivers/infiniband/hw/amso1100/c2_cq.c +++ b/drivers/infiniband/hw/amso1100/c2_cq.c @@ -246,20 +246,17 @@ int c2_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify notify) static void c2_free_cq_buf(struct c2_dev *c2dev, struct c2_mq *mq) { - - dma_unmap_single(c2dev->ibdev.dma_device, pci_unmap_addr(mq, mapping), - mq->q_size * mq->msg_size, DMA_FROM_DEVICE); - free_pages((unsigned long) mq->msg_pool.host, - get_order(mq->q_size * mq->msg_size)); + dma_free_coherent(&c2dev->pcidev->dev, mq->q_size * mq->msg_size, + mq->msg_pool.host, pci_unmap_addr(mq, mapping)); } static int c2_alloc_cq_buf(struct c2_dev *c2dev, struct c2_mq *mq, int q_size, int msg_size) { - unsigned long pool_start; + u8 *pool_start; - pool_start = __get_free_pages(GFP_KERNEL, - get_order(q_size * msg_size)); + pool_start = dma_alloc_coherent(&c2dev->pcidev->dev, q_size * msg_size, + &mq->host_dma, GFP_KERNEL); if (!pool_start) return -ENOMEM; @@ -267,13 +264,10 @@ static int c2_alloc_cq_buf(struct c2_dev *c2dev, struct c2_mq *mq, int q_size, 0, /* index (currently unknown) */ q_size, msg_size, - (u8 *) pool_start, + pool_start, NULL, /* peer (currently unknown) */ C2_MQ_HOST_TARGET); - mq->host_dma = dma_map_single(c2dev->ibdev.dma_device, - (void *)pool_start, - q_size * msg_size, DMA_FROM_DEVICE); pci_unmap_addr_set(mq, mapping, mq->host_dma); return 0; diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c index 30409e1..030238d 100644 --- a/drivers/infiniband/hw/amso1100/c2_rnic.c +++ b/drivers/infiniband/hw/amso1100/c2_rnic.c @@ -517,14 +517,12 @@ int c2_rnic_init(struct c2_dev *c2dev) /* Initialize the Verbs Reply Queue */ qsize = be32_to_cpu(readl(mmio_regs + C2_REGS_Q1_QSIZE)); msgsize = be32_to_cpu(readl(mmio_regs + C2_REGS_Q1_MSGSIZE)); - q1_pages = kmalloc(qsize * msgsize, GFP_KERNEL); + q1_pages = dma_alloc_coherent(&c2dev->pcidev->dev, qsize * msgsize, + &c2dev->rep_vq.host_dma, GFP_KERNEL); if (!q1_pages) { err = -ENOMEM; goto bail1; } - c2dev->rep_vq.host_dma = dma_map_single(c2dev->ibdev.dma_device, - (void *)q1_pages, qsize * msgsize, - DMA_FROM_DEVICE); pci_unmap_addr_set(&c2dev->rep_vq, mapping, c2dev->rep_vq.host_dma); pr_debug("%s rep_vq va %p dma %llx\n", __FUNCTION__, q1_pages, (unsigned long long) c2dev->rep_vq.host_dma); @@ -540,14 +538,12 @@ int c2_rnic_init(struct c2_dev *c2dev) /* Initialize the Asynchronus Event Queue */ qsize = be32_to_cpu(readl(mmio_regs + C2_REGS_Q2_QSIZE)); msgsize = be32_to_cpu(readl(mmio_regs + C2_REGS_Q2_MSGSIZE)); - q2_pages = kmalloc(qsize * msgsize, GFP_KERNEL); + q2_pages = dma_alloc_coherent(&c2dev->pcidev->dev, qsize * msgsize, + &c2dev->aeq.host_dma, GFP_KERNEL); if (!q2_pages) { err = -ENOMEM; goto bail2; } - c2dev->aeq.host_dma = dma_map_single(c2dev->ibdev.dma_device, - (void *)q2_pages, qsize * msgsize, - DMA_FROM_DEVICE); pci_unmap_addr_set(&c2dev->aeq, mapping, c2dev->aeq.host_dma); pr_debug("%s aeq va %p dma %llx\n", __FUNCTION__, q1_pages, (unsigned long long) c2dev->rep_vq.host_dma); @@ -597,17 +593,13 @@ int c2_rnic_init(struct c2_dev *c2dev) bail4: vq_term(c2dev); bail3: - dma_unmap_single(c2dev->ibdev.dma_device, - pci_unmap_addr(&c2dev->aeq, mapping), - c2dev->aeq.q_size * c2dev->aeq.msg_size, - DMA_FROM_DEVICE); - kfree(q2_pages); + dma_free_coherent(&c2dev->pcidev->dev, + c2dev->aeq.q_size * c2dev->aeq.msg_size, + q2_pages, pci_unmap_addr(&c2dev->aeq, mapping)); bail2: - dma_unmap_single(c2dev->ibdev.dma_device, - pci_unmap_addr(&c2dev->rep_vq, mapping), - c2dev->rep_vq.q_size * c2dev->rep_vq.msg_size, - DMA_FROM_DEVICE); - kfree(q1_pages); + dma_free_coherent(&c2dev->pcidev->dev, + c2dev->rep_vq.q_size * c2dev->rep_vq.msg_size, + q1_pages, pci_unmap_addr(&c2dev->rep_vq, mapping)); bail1: c2_free_mqsp_pool(c2dev, c2dev->kern_mqsp_pool); bail0: @@ -640,19 +632,17 @@ void c2_rnic_term(struct c2_dev *c2dev) /* Free the verbs request allocator */ vq_term(c2dev); - /* Unmap and free the asynchronus event queue */ - dma_unmap_single(c2dev->ibdev.dma_device, - pci_unmap_addr(&c2dev->aeq, mapping), - c2dev->aeq.q_size * c2dev->aeq.msg_size, - DMA_FROM_DEVICE); - kfree(c2dev->aeq.msg_pool.host); - - /* Unmap and free the verbs reply queue */ - dma_unmap_single(c2dev->ibdev.dma_device, - pci_unmap_addr(&c2dev->rep_vq, mapping), - c2dev->rep_vq.q_size * c2dev->rep_vq.msg_size, - DMA_FROM_DEVICE); - kfree(c2dev->rep_vq.msg_pool.host); + /* Free the asynchronus event queue */ + dma_free_coherent(&c2dev->pcidev->dev, + c2dev->aeq.q_size * c2dev->aeq.msg_size, + c2dev->aeq.msg_pool.host, + pci_unmap_addr(&c2dev->aeq, mapping)); + + /* Free the verbs reply queue */ + dma_free_coherent(&c2dev->pcidev->dev, + c2dev->rep_vq.q_size * c2dev->rep_vq.msg_size, + c2dev->rep_vq.msg_pool.host, + pci_unmap_addr(&c2dev->rep_vq, mapping)); /* Free the MQ shared pointer pool */ c2_free_mqsp_pool(c2dev, c2dev->kern_mqsp_pool); -- cgit v0.10.2 From d7b748d63c908a0a85099ce546594192ae0926f6 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Mon, 30 Oct 2006 20:52:53 -0800 Subject: IB/amso1100: Fix incorrect pr_debug() pr_debug() was printing the wrong stuff. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c index 030238d..21d9612 100644 --- a/drivers/infiniband/hw/amso1100/c2_rnic.c +++ b/drivers/infiniband/hw/amso1100/c2_rnic.c @@ -545,8 +545,8 @@ int c2_rnic_init(struct c2_dev *c2dev) goto bail2; } pci_unmap_addr_set(&c2dev->aeq, mapping, c2dev->aeq.host_dma); - pr_debug("%s aeq va %p dma %llx\n", __FUNCTION__, q1_pages, - (unsigned long long) c2dev->rep_vq.host_dma); + pr_debug("%s aeq va %p dma %llx\n", __FUNCTION__, q2_pages, + (unsigned long long) c2dev->aeq.host_dma); c2_mq_rep_init(&c2dev->aeq, 2, qsize, -- cgit v0.10.2 From 0b26c88f29ad8bcf91a2ea8f25a36f2028ebabea Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Wed, 25 Oct 2006 12:54:20 +0200 Subject: IB/uverbs: Return sq_draining value in query_qp response Return the sq_draining value back to user space for query_qp instead of the en_sqd_async notify value, which is valid only for modify_qp. For query_qp, the draining status should returned. Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index b72c7f6..743247e 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1214,7 +1214,7 @@ ssize_t ib_uverbs_query_qp(struct ib_uverbs_file *file, resp.qp_access_flags = attr->qp_access_flags; resp.pkey_index = attr->pkey_index; resp.alt_pkey_index = attr->alt_pkey_index; - resp.en_sqd_async_notify = attr->en_sqd_async_notify; + resp.sq_draining = attr->sq_draining; resp.max_rd_atomic = attr->max_rd_atomic; resp.max_dest_rd_atomic = attr->max_dest_rd_atomic; resp.min_rnr_timer = attr->min_rnr_timer; diff --git a/include/rdma/ib_user_verbs.h b/include/rdma/ib_user_verbs.h index db1b814..64a721f 100644 --- a/include/rdma/ib_user_verbs.h +++ b/include/rdma/ib_user_verbs.h @@ -458,7 +458,7 @@ struct ib_uverbs_query_qp_resp { __u8 cur_qp_state; __u8 path_mtu; __u8 path_mig_state; - __u8 en_sqd_async_notify; + __u8 sq_draining; __u8 max_rd_atomic; __u8 max_dest_rd_atomic; __u8 min_rnr_timer; -- cgit v0.10.2 From 1b7c2dbc07bf0663a41e3dc838992930019f08fd Mon Sep 17 00:00:00 2001 From: James Morris Date: Tue, 31 Oct 2006 00:43:44 -0800 Subject: [IPV6]: fix flowlabel seqfile handling There's a bug in the seqfile show operation for flowlabel objects, where each hash chain is traversed cumulatively for each element. The following function is called for each element of each chain: static void ip6fl_fl_seq_show(struct seq_file *seq, struct ip6_flowlabel *fl) { while(fl) { seq_printf... fl = fl->next; } } Thus, objects can appear mutliple times when reading /proc/net/ip6_flowlabel, as the above is called for each element in the chain. The solution is to remove the while() loop from the above, and traverse each chain exactly once, per the patch below. This also removes the ip6fl_fl_seq_show() function, which does nothing else. Signed-off-by: James Morris Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 2b45f2d6..6d4533b 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -627,9 +627,13 @@ static void ip6fl_seq_stop(struct seq_file *seq, void *v) read_unlock_bh(&ip6_fl_lock); } -static void ip6fl_fl_seq_show(struct seq_file *seq, struct ip6_flowlabel *fl) +static int ip6fl_seq_show(struct seq_file *seq, void *v) { - while(fl) { + if (v == SEQ_START_TOKEN) + seq_printf(seq, "%-5s %-1s %-6s %-6s %-6s %-8s %-32s %s\n", + "Label", "S", "Owner", "Users", "Linger", "Expires", "Dst", "Opt"); + else { + struct ip6_flowlabel *fl = v; seq_printf(seq, "%05X %-1d %-6d %-6d %-6ld %-8ld " NIP6_SEQFMT " %-4d\n", (unsigned)ntohl(fl->label), @@ -640,17 +644,7 @@ static void ip6fl_fl_seq_show(struct seq_file *seq, struct ip6_flowlabel *fl) (long)(fl->expires - jiffies)/HZ, NIP6(fl->dst), fl->opt ? fl->opt->opt_nflen : 0); - fl = fl->next; } -} - -static int ip6fl_seq_show(struct seq_file *seq, void *v) -{ - if (v == SEQ_START_TOKEN) - seq_printf(seq, "%-5s %-1s %-6s %-6s %-6s %-8s %-32s %s\n", - "Label", "S", "Owner", "Users", "Linger", "Expires", "Dst", "Opt"); - else - ip6fl_fl_seq_show(seq, v); return 0; } -- cgit v0.10.2 From 36a561d6a95c4b89ae4845bf91456b4f784b6eec Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Oct 2006 22:07:03 -0800 Subject: [PATCH] find_bd_holder() fix fs/block_dev.c: In function 'find_bd_holder': fs/block_dev.c:666: warning: return makes integer from pointer without a cast fs/block_dev.c:669: warning: return makes integer from pointer without a cast fs/block_dev.c: In function 'add_bd_holder': fs/block_dev.c:685: warning: unused variable 'tmp' fs/block_dev.c: In function 'bd_claim_by_kobject': fs/block_dev.c:773: warning: assignment makes pointer from integer without a cast Acked-by: Jun'ichi Nomura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/block_dev.c b/fs/block_dev.c index aaa8301..36c0e7a 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -651,7 +651,8 @@ static void free_bd_holder(struct bd_holder *bo) * If found, increment the reference count and return the pointer. * If not found, returns NULL. */ -static int find_bd_holder(struct block_device *bdev, struct bd_holder *bo) +static struct bd_holder *find_bd_holder(struct block_device *bdev, + struct bd_holder *bo) { struct bd_holder *tmp; @@ -677,7 +678,6 @@ static int find_bd_holder(struct block_device *bdev, struct bd_holder *bo) */ static int add_bd_holder(struct block_device *bdev, struct bd_holder *bo) { - struct bd_holder *tmp; int ret; if (!bo) -- cgit v0.10.2 From 97d88ac83540f9ba9536326c30db4815c5b9169b Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:03 -0800 Subject: [PATCH] uml ubd driver: allow using up to 16 UBD devices With 256 minors and 16 minors used per each UBD device, we can allow the use of up to 16 UBD devices per UML. Also chnage parse_unit and leave to the caller (which already do it) the check for excess numbers, since this is just supposed to do raw parsing. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index bc458f5..54d2473 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -117,7 +117,7 @@ static int ubd_ioctl(struct inode * inode, struct file * file, unsigned int cmd, unsigned long arg); static int ubd_getgeo(struct block_device *bdev, struct hd_geometry *geo); -#define MAX_DEV (8) +#define MAX_DEV (16) static struct block_device_operations ubd_blops = { .owner = THIS_MODULE, @@ -277,7 +277,7 @@ static int parse_unit(char **ptr) return(-1); *ptr = end; } - else if (('a' <= *str) && (*str <= 'h')) { + else if (('a' <= *str) && (*str <= 'z')) { n = *str - 'a'; str++; *ptr = str; -- cgit v0.10.2 From 2a9d32f682b2b4928dcd4680bc99e98a3d096816 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:04 -0800 Subject: [PATCH] uml ubd driver: document some struct fields Add documentation about some fields in struct ubd, whose meaning is non-obvious due to struct names (should change names altogether, I agree). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 54d2473..e104f59 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -150,8 +150,9 @@ static struct gendisk *fake_gendisk[MAX_DEV]; static struct openflags global_openflags = OPEN_FLAGS; struct cow { - /* This is the backing file, actually */ + /* backing file name */ char *file; + /* backing file fd */ int fd; unsigned long *bitmap; unsigned long bitmap_len; @@ -160,6 +161,8 @@ struct cow { }; struct ubd { + /* name (and fd, below) of the file opened for writing, either the + * backing or the cow file. */ char *file; int count; int fd; -- cgit v0.10.2 From 7d314e346d6081e8013a96206e601a48528d8f60 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:05 -0800 Subject: [PATCH] uml ubd driver: var renames Rename the ubd_dev array to ubd_devs and then call any "struct ubd" ubd_dev instead of dev, which doesn't make clear what we're treating (and no, it's not hungarian notation - not any more than calling all vm_area_struct vma or all inodes inode). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index e104f59..761e3f9 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -195,14 +195,14 @@ struct ubd { .cow = DEFAULT_COW, \ } -struct ubd ubd_dev[MAX_DEV] = { [ 0 ... MAX_DEV - 1 ] = DEFAULT_UBD }; +struct ubd ubd_devs[MAX_DEV] = { [ 0 ... MAX_DEV - 1 ] = DEFAULT_UBD }; static int ubd0_init(void) { - struct ubd *dev = &ubd_dev[0]; + struct ubd *ubd_dev = &ubd_devs[0]; - if(dev->file == NULL) - dev->file = "root_fs"; + if(ubd_dev->file == NULL) + ubd_dev->file = "root_fs"; return(0); } @@ -290,7 +290,7 @@ static int parse_unit(char **ptr) static int ubd_setup_common(char *str, int *index_out) { - struct ubd *dev; + struct ubd *ubd_dev; struct openflags flags = global_openflags; char *backing_file; int n, err, i; @@ -345,8 +345,8 @@ static int ubd_setup_common(char *str, int *index_out) err = 1; spin_lock(&ubd_lock); - dev = &ubd_dev[n]; - if(dev->file != NULL){ + ubd_dev = &ubd_devs[n]; + if(ubd_dev->file != NULL){ printk(KERN_ERR "ubd_setup : device already configured\n"); goto out; } @@ -363,10 +363,10 @@ static int ubd_setup_common(char *str, int *index_out) flags.s = 1; break; case 'd': - dev->no_cow = 1; + ubd_dev->no_cow = 1; break; case 'c': - dev->shared = 1; + ubd_dev->shared = 1; break; case '=': str++; @@ -393,7 +393,7 @@ break_loop: } if(backing_file){ - if(dev->no_cow) + if(ubd_dev->no_cow) printk(KERN_ERR "Can't specify both 'd' and a " "cow file\n"); else { @@ -401,9 +401,9 @@ break_loop: backing_file++; } } - dev->file = str; - dev->cow.file = backing_file; - dev->boot_openflags = flags; + ubd_dev->file = str; + ubd_dev->cow.file = backing_file; + ubd_dev->boot_openflags = flags; out: spin_unlock(&ubd_lock); return(err); @@ -544,83 +544,83 @@ void kill_io_thread(void) __uml_exitcall(kill_io_thread); -static int ubd_file_size(struct ubd *dev, __u64 *size_out) +static int ubd_file_size(struct ubd *ubd_dev, __u64 *size_out) { char *file; - file = dev->cow.file ? dev->cow.file : dev->file; + file = ubd_dev->cow.file ? ubd_dev->cow.file : ubd_dev->file; return(os_file_size(file, size_out)); } -static void ubd_close(struct ubd *dev) +static void ubd_close(struct ubd *ubd_dev) { - os_close_file(dev->fd); - if(dev->cow.file == NULL) + os_close_file(ubd_dev->fd); + if(ubd_dev->cow.file == NULL) return; - os_close_file(dev->cow.fd); - vfree(dev->cow.bitmap); - dev->cow.bitmap = NULL; + os_close_file(ubd_dev->cow.fd); + vfree(ubd_dev->cow.bitmap); + ubd_dev->cow.bitmap = NULL; } -static int ubd_open_dev(struct ubd *dev) +static int ubd_open_dev(struct ubd *ubd_dev) { struct openflags flags; char **back_ptr; int err, create_cow, *create_ptr; - dev->openflags = dev->boot_openflags; + ubd_dev->openflags = ubd_dev->boot_openflags; create_cow = 0; - create_ptr = (dev->cow.file != NULL) ? &create_cow : NULL; - back_ptr = dev->no_cow ? NULL : &dev->cow.file; - dev->fd = open_ubd_file(dev->file, &dev->openflags, dev->shared, - back_ptr, &dev->cow.bitmap_offset, - &dev->cow.bitmap_len, &dev->cow.data_offset, + create_ptr = (ubd_dev->cow.file != NULL) ? &create_cow : NULL; + back_ptr = ubd_dev->no_cow ? NULL : &ubd_dev->cow.file; + ubd_dev->fd = open_ubd_file(ubd_dev->file, &ubd_dev->openflags, ubd_dev->shared, + back_ptr, &ubd_dev->cow.bitmap_offset, + &ubd_dev->cow.bitmap_len, &ubd_dev->cow.data_offset, create_ptr); - if((dev->fd == -ENOENT) && create_cow){ - dev->fd = create_cow_file(dev->file, dev->cow.file, - dev->openflags, 1 << 9, PAGE_SIZE, - &dev->cow.bitmap_offset, - &dev->cow.bitmap_len, - &dev->cow.data_offset); - if(dev->fd >= 0){ + if((ubd_dev->fd == -ENOENT) && create_cow){ + ubd_dev->fd = create_cow_file(ubd_dev->file, ubd_dev->cow.file, + ubd_dev->openflags, 1 << 9, PAGE_SIZE, + &ubd_dev->cow.bitmap_offset, + &ubd_dev->cow.bitmap_len, + &ubd_dev->cow.data_offset); + if(ubd_dev->fd >= 0){ printk(KERN_INFO "Creating \"%s\" as COW file for " - "\"%s\"\n", dev->file, dev->cow.file); + "\"%s\"\n", ubd_dev->file, ubd_dev->cow.file); } } - if(dev->fd < 0){ - printk("Failed to open '%s', errno = %d\n", dev->file, - -dev->fd); - return(dev->fd); + if(ubd_dev->fd < 0){ + printk("Failed to open '%s', errno = %d\n", ubd_dev->file, + -ubd_dev->fd); + return(ubd_dev->fd); } - if(dev->cow.file != NULL){ + if(ubd_dev->cow.file != NULL){ err = -ENOMEM; - dev->cow.bitmap = (void *) vmalloc(dev->cow.bitmap_len); - if(dev->cow.bitmap == NULL){ + ubd_dev->cow.bitmap = (void *) vmalloc(ubd_dev->cow.bitmap_len); + if(ubd_dev->cow.bitmap == NULL){ printk(KERN_ERR "Failed to vmalloc COW bitmap\n"); goto error; } flush_tlb_kernel_vm(); - err = read_cow_bitmap(dev->fd, dev->cow.bitmap, - dev->cow.bitmap_offset, - dev->cow.bitmap_len); + err = read_cow_bitmap(ubd_dev->fd, ubd_dev->cow.bitmap, + ubd_dev->cow.bitmap_offset, + ubd_dev->cow.bitmap_len); if(err < 0) goto error; - flags = dev->openflags; + flags = ubd_dev->openflags; flags.w = 0; - err = open_ubd_file(dev->cow.file, &flags, dev->shared, NULL, + err = open_ubd_file(ubd_dev->cow.file, &flags, ubd_dev->shared, NULL, NULL, NULL, NULL, NULL); if(err < 0) goto error; - dev->cow.fd = err; + ubd_dev->cow.fd = err; } return(0); error: - os_close_file(dev->fd); + os_close_file(ubd_dev->fd); return(err); } @@ -645,13 +645,13 @@ static int ubd_new_disk(int major, u64 size, int unit, /* sysfs register (not for ide fake devices) */ if (major == MAJOR_NR) { - ubd_dev[unit].pdev.id = unit; - ubd_dev[unit].pdev.name = DRIVER_NAME; - platform_device_register(&ubd_dev[unit].pdev); - disk->driverfs_dev = &ubd_dev[unit].pdev.dev; + ubd_devs[unit].pdev.id = unit; + ubd_devs[unit].pdev.name = DRIVER_NAME; + platform_device_register(&ubd_devs[unit].pdev); + disk->driverfs_dev = &ubd_devs[unit].pdev.dev; } - disk->private_data = &ubd_dev[unit]; + disk->private_data = &ubd_devs[unit]; disk->queue = ubd_queue; add_disk(disk); @@ -663,25 +663,25 @@ static int ubd_new_disk(int major, u64 size, int unit, static int ubd_add(int n) { - struct ubd *dev = &ubd_dev[n]; + struct ubd *ubd_dev = &ubd_devs[n]; int err; err = -ENODEV; - if(dev->file == NULL) + if(ubd_dev->file == NULL) goto out; - err = ubd_file_size(dev, &dev->size); + err = ubd_file_size(ubd_dev, &ubd_dev->size); if(err < 0) goto out; - dev->size = ROUND_BLOCK(dev->size); + ubd_dev->size = ROUND_BLOCK(ubd_dev->size); - err = ubd_new_disk(MAJOR_NR, dev->size, n, &ubd_gendisk[n]); + err = ubd_new_disk(MAJOR_NR, ubd_dev->size, n, &ubd_gendisk[n]); if(err) goto out; if(fake_major != MAJOR_NR) - ubd_new_disk(fake_major, dev->size, n, + ubd_new_disk(fake_major, ubd_dev->size, n, &fake_gendisk[n]); /* perhaps this should also be under the "if (fake_major)" above */ @@ -713,7 +713,7 @@ static int ubd_config(char *str) spin_lock(&ubd_lock); err = ubd_add(n); if(err) - ubd_dev[n].file = NULL; + ubd_devs[n].file = NULL; spin_unlock(&ubd_lock); return(err); @@ -721,7 +721,7 @@ static int ubd_config(char *str) static int ubd_get_config(char *name, char *str, int size, char **error_out) { - struct ubd *dev; + struct ubd *ubd_dev; int n, len = 0; n = parse_unit(&name); @@ -730,19 +730,19 @@ static int ubd_get_config(char *name, char *str, int size, char **error_out) return(-1); } - dev = &ubd_dev[n]; + ubd_dev = &ubd_devs[n]; spin_lock(&ubd_lock); - if(dev->file == NULL){ + if(ubd_dev->file == NULL){ CONFIG_CHUNK(str, size, len, "", 1); goto out; } - CONFIG_CHUNK(str, size, len, dev->file, 0); + CONFIG_CHUNK(str, size, len, ubd_dev->file, 0); - if(dev->cow.file != NULL){ + if(ubd_dev->cow.file != NULL){ CONFIG_CHUNK(str, size, len, ",", 0); - CONFIG_CHUNK(str, size, len, dev->cow.file, 1); + CONFIG_CHUNK(str, size, len, ubd_dev->cow.file, 1); } else CONFIG_CHUNK(str, size, len, "", 1); @@ -763,7 +763,7 @@ static int ubd_id(char **str, int *start_out, int *end_out) static int ubd_remove(int n) { - struct ubd *dev; + struct ubd *ubd_dev; int err = -ENODEV; spin_lock(&ubd_lock); @@ -771,14 +771,14 @@ static int ubd_remove(int n) if(ubd_gendisk[n] == NULL) goto out; - dev = &ubd_dev[n]; + ubd_dev = &ubd_devs[n]; - if(dev->file == NULL) + if(ubd_dev->file == NULL) goto out; /* you cannot remove a open disk */ err = -EBUSY; - if(dev->count > 0) + if(ubd_dev->count > 0) goto out; del_gendisk(ubd_gendisk[n]); @@ -791,8 +791,8 @@ static int ubd_remove(int n) fake_gendisk[n] = NULL; } - platform_device_unregister(&dev->pdev); - *dev = ((struct ubd) DEFAULT_UBD); + platform_device_unregister(&ubd_dev->pdev); + *ubd_dev = ((struct ubd) DEFAULT_UBD); err = 0; out: spin_unlock(&ubd_lock); @@ -870,7 +870,7 @@ int ubd_driver_init(void){ return(0); } err = um_request_irq(UBD_IRQ, thread_fd, IRQ_READ, ubd_intr, - IRQF_DISABLED, "ubd", ubd_dev); + IRQF_DISABLED, "ubd", ubd_devs); if(err != 0) printk(KERN_ERR "um_request_irq failed - errno = %d\n", -err); return 0; @@ -881,24 +881,24 @@ device_initcall(ubd_driver_init); static int ubd_open(struct inode *inode, struct file *filp) { struct gendisk *disk = inode->i_bdev->bd_disk; - struct ubd *dev = disk->private_data; + struct ubd *ubd_dev = disk->private_data; int err = 0; - if(dev->count == 0){ - err = ubd_open_dev(dev); + if(ubd_dev->count == 0){ + err = ubd_open_dev(ubd_dev); if(err){ printk(KERN_ERR "%s: Can't open \"%s\": errno = %d\n", - disk->disk_name, dev->file, -err); + disk->disk_name, ubd_dev->file, -err); goto out; } } - dev->count++; - set_disk_ro(disk, !dev->openflags.w); + ubd_dev->count++; + set_disk_ro(disk, !ubd_dev->openflags.w); /* This should no more be needed. And it didn't work anyway to exclude * read-write remounting of filesystems.*/ - /*if((filp->f_mode & FMODE_WRITE) && !dev->openflags.w){ - if(--dev->count == 0) ubd_close(dev); + /*if((filp->f_mode & FMODE_WRITE) && !ubd_dev->openflags.w){ + if(--ubd_dev->count == 0) ubd_close(ubd_dev); err = -EROFS; }*/ out: @@ -908,10 +908,10 @@ static int ubd_open(struct inode *inode, struct file *filp) static int ubd_release(struct inode * inode, struct file * file) { struct gendisk *disk = inode->i_bdev->bd_disk; - struct ubd *dev = disk->private_data; + struct ubd *ubd_dev = disk->private_data; - if(--dev->count == 0) - ubd_close(dev); + if(--ubd_dev->count == 0) + ubd_close(ubd_dev); return(0); } @@ -979,12 +979,12 @@ static void cowify_req(struct io_thread_req *req, unsigned long *bitmap, static int prepare_request(struct request *req, struct io_thread_req *io_req) { struct gendisk *disk = req->rq_disk; - struct ubd *dev = disk->private_data; + struct ubd *ubd_dev = disk->private_data; __u64 offset; int len; /* This should be impossible now */ - if((rq_data_dir(req) == WRITE) && !dev->openflags.w){ + if((rq_data_dir(req) == WRITE) && !ubd_dev->openflags.w){ printk("Write attempted on readonly ubd device %s\n", disk->disk_name); end_request(req, 0); @@ -994,8 +994,8 @@ static int prepare_request(struct request *req, struct io_thread_req *io_req) offset = ((__u64) req->sector) << 9; len = req->current_nr_sectors << 9; - io_req->fds[0] = (dev->cow.file != NULL) ? dev->cow.fd : dev->fd; - io_req->fds[1] = dev->fd; + io_req->fds[0] = (ubd_dev->cow.file != NULL) ? ubd_dev->cow.fd : ubd_dev->fd; + io_req->fds[1] = ubd_dev->fd; io_req->cow_offset = -1; io_req->offset = offset; io_req->length = len; @@ -1004,13 +1004,13 @@ static int prepare_request(struct request *req, struct io_thread_req *io_req) io_req->op = (rq_data_dir(req) == READ) ? UBD_READ : UBD_WRITE; io_req->offsets[0] = 0; - io_req->offsets[1] = dev->cow.data_offset; + io_req->offsets[1] = ubd_dev->cow.data_offset; io_req->buffer = req->buffer; io_req->sectorsize = 1 << 9; - if(dev->cow.file != NULL) - cowify_req(io_req, dev->cow.bitmap, dev->cow.bitmap_offset, - dev->cow.bitmap_len); + if(ubd_dev->cow.file != NULL) + cowify_req(io_req, ubd_dev->cow.bitmap, ubd_dev->cow.bitmap_offset, + ubd_dev->cow.bitmap_len); return(0); } @@ -1048,18 +1048,18 @@ static void do_ubd_request(request_queue_t *q) static int ubd_getgeo(struct block_device *bdev, struct hd_geometry *geo) { - struct ubd *dev = bdev->bd_disk->private_data; + struct ubd *ubd_dev = bdev->bd_disk->private_data; geo->heads = 128; geo->sectors = 32; - geo->cylinders = dev->size / (128 * 32 * 512); + geo->cylinders = ubd_dev->size / (128 * 32 * 512); return 0; } static int ubd_ioctl(struct inode * inode, struct file * file, unsigned int cmd, unsigned long arg) { - struct ubd *dev = inode->i_bdev->bd_disk->private_data; + struct ubd *ubd_dev = inode->i_bdev->bd_disk->private_data; struct hd_driveid ubd_id = { .cyls = 0, .heads = 128, @@ -1069,7 +1069,7 @@ static int ubd_ioctl(struct inode * inode, struct file * file, switch (cmd) { struct cdrom_volctrl volume; case HDIO_GET_IDENTITY: - ubd_id.cyls = dev->size / (128 * 32 * 512); + ubd_id.cyls = ubd_dev->size / (128 * 32 * 512); if(copy_to_user((char __user *) arg, (char *) &ubd_id, sizeof(ubd_id))) return(-EFAULT); -- cgit v0.10.2 From 5f75a4f887a35b99878fc07ed749a90375194b63 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:06 -0800 Subject: [PATCH] uml ubd driver: give better names to some functions. To rethink locking, I needed to understand well what each function does. While doing this I renamed some: * ubd_close -> ubd_close_dev (since it pairs with ubd_open_dev) * ubd_new_disk -> ubd_disk_register (it handles registration with the block layer - one hopes this makes clearer the difference with ubd_add()) Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 761e3f9..cdd82e8 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -552,7 +552,7 @@ static int ubd_file_size(struct ubd *ubd_dev, __u64 *size_out) return(os_file_size(file, size_out)); } -static void ubd_close(struct ubd *ubd_dev) +static void ubd_close_dev(struct ubd *ubd_dev) { os_close_file(ubd_dev->fd); if(ubd_dev->cow.file == NULL) @@ -624,7 +624,7 @@ static int ubd_open_dev(struct ubd *ubd_dev) return(err); } -static int ubd_new_disk(int major, u64 size, int unit, +static int ubd_disk_register(int major, u64 size, int unit, struct gendisk **disk_out) { @@ -676,12 +676,12 @@ static int ubd_add(int n) ubd_dev->size = ROUND_BLOCK(ubd_dev->size); - err = ubd_new_disk(MAJOR_NR, ubd_dev->size, n, &ubd_gendisk[n]); + err = ubd_disk_register(MAJOR_NR, ubd_dev->size, n, &ubd_gendisk[n]); if(err) goto out; if(fake_major != MAJOR_NR) - ubd_new_disk(fake_major, ubd_dev->size, n, + ubd_disk_register(fake_major, ubd_dev->size, n, &fake_gendisk[n]); /* perhaps this should also be under the "if (fake_major)" above */ @@ -898,7 +898,7 @@ static int ubd_open(struct inode *inode, struct file *filp) /* This should no more be needed. And it didn't work anyway to exclude * read-write remounting of filesystems.*/ /*if((filp->f_mode & FMODE_WRITE) && !ubd_dev->openflags.w){ - if(--ubd_dev->count == 0) ubd_close(ubd_dev); + if(--ubd_dev->count == 0) ubd_close_dev(ubd_dev); err = -EROFS; }*/ out: @@ -911,7 +911,7 @@ static int ubd_release(struct inode * inode, struct file * file) struct ubd *ubd_dev = disk->private_data; if(--ubd_dev->count == 0) - ubd_close(ubd_dev); + ubd_close_dev(ubd_dev); return(0); } -- cgit v0.10.2 From d7fb2c3865ca0f95d92e2864c3dc9220789d83f5 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:07 -0800 Subject: [PATCH] uml ubd driver: change ubd_lock to be a mutex This lock protects ubd setup and teardown, so is only used in process context; beyond that, during such setup memory allocations must be performed and some generic functions which can sleep must be called (such as add_disk()). So the only correct solution is to make it a mutex instead of a spin_lock. No other change is done - this lock must be acquired in different places but it's done afterwards. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index cdd82e8..a3061ae 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -107,7 +107,8 @@ static inline void ubd_set_bit(__u64 bit, unsigned char *data) #define DRIVER_NAME "uml-blkdev" static DEFINE_SPINLOCK(ubd_io_lock); -static DEFINE_SPINLOCK(ubd_lock); + +static DEFINE_MUTEX(ubd_lock); static void (*do_ubd)(void); @@ -314,7 +315,7 @@ static int ubd_setup_common(char *str, int *index_out) } err = 1; - spin_lock(&ubd_lock); + mutex_lock(&ubd_lock); if(fake_major != MAJOR_NR){ printk(KERN_ERR "Can't assign a fake major twice\n"); goto out1; @@ -326,7 +327,7 @@ static int ubd_setup_common(char *str, int *index_out) major); err = 0; out1: - spin_unlock(&ubd_lock); + mutex_unlock(&ubd_lock); return(err); } @@ -343,7 +344,7 @@ static int ubd_setup_common(char *str, int *index_out) } err = 1; - spin_lock(&ubd_lock); + mutex_lock(&ubd_lock); ubd_dev = &ubd_devs[n]; if(ubd_dev->file != NULL){ @@ -405,7 +406,7 @@ break_loop: ubd_dev->cow.file = backing_file; ubd_dev->boot_openflags = flags; out: - spin_unlock(&ubd_lock); + mutex_unlock(&ubd_lock); return(err); } @@ -710,11 +711,11 @@ static int ubd_config(char *str) } if(n == -1) return(0); - spin_lock(&ubd_lock); + mutex_lock(&ubd_lock); err = ubd_add(n); if(err) ubd_devs[n].file = NULL; - spin_unlock(&ubd_lock); + mutex_unlock(&ubd_lock); return(err); } @@ -731,7 +732,7 @@ static int ubd_get_config(char *name, char *str, int size, char **error_out) } ubd_dev = &ubd_devs[n]; - spin_lock(&ubd_lock); + mutex_lock(&ubd_lock); if(ubd_dev->file == NULL){ CONFIG_CHUNK(str, size, len, "", 1); @@ -747,7 +748,7 @@ static int ubd_get_config(char *name, char *str, int size, char **error_out) else CONFIG_CHUNK(str, size, len, "", 1); out: - spin_unlock(&ubd_lock); + mutex_unlock(&ubd_lock); return(len); } @@ -766,7 +767,7 @@ static int ubd_remove(int n) struct ubd *ubd_dev; int err = -ENODEV; - spin_lock(&ubd_lock); + mutex_lock(&ubd_lock); if(ubd_gendisk[n] == NULL) goto out; @@ -795,7 +796,7 @@ static int ubd_remove(int n) *ubd_dev = ((struct ubd) DEFAULT_UBD); err = 0; out: - spin_unlock(&ubd_lock); + mutex_unlock(&ubd_lock); return err; } -- cgit v0.10.2 From 33f775eea185e8df7701c4afc2c8fcee85c83282 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:08 -0800 Subject: [PATCH] uml ubd driver: ubd_io_lock usage fixup Add some comments about requirements for ubd_io_lock and expand its use. When an irq signals that the "controller" (i.e. another thread on the host, which does the actual requests and is the only one blocked on I/O on the host) has done some work, we call again the request function ourselves (do_ubd_request). We now do that with ubd_io_lock held - that's useful to protect against concurrent calls to elv_next_request and so on. XXX: Maybe we shouldn't call at all the request function. Input needed on this. Are we supposed to plug and unplug the queue? That code "indirectly" does that by setting a flag, called do_ubd, which makes the request function return (it's a residual of 2.4 block layer interface). Meanwhile, however, merge this patch, which improves things. Cc: Jens Axboe Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index a3061ae..6cd8988 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -106,6 +106,8 @@ static inline void ubd_set_bit(__u64 bit, unsigned char *data) #define DRIVER_NAME "uml-blkdev" +/* Can be taken in interrupt context, and is passed to the block layer to lock + * the request queue. Kernel side code knows that. */ static DEFINE_SPINLOCK(ubd_io_lock); static DEFINE_MUTEX(ubd_lock); @@ -497,6 +499,8 @@ static void __ubd_finish(struct request *req, int error) end_request(req, 1); } +/* Callable only from interrupt context - otherwise you need to do + * spin_lock_irq()/spin_lock_irqsave() */ static inline void ubd_finish(struct request *req, int error) { spin_lock(&ubd_io_lock); @@ -504,7 +508,7 @@ static inline void ubd_finish(struct request *req, int error) spin_unlock(&ubd_io_lock); } -/* Called without ubd_io_lock held */ +/* Called without ubd_io_lock held, and only in interrupt context. */ static void ubd_handler(void) { struct io_thread_req req; @@ -525,7 +529,9 @@ static void ubd_handler(void) ubd_finish(rq, req.error); reactivate_fd(thread_fd, UBD_IRQ); + spin_lock(&ubd_io_lock); do_ubd_request(ubd_queue); + spin_unlock(&ubd_io_lock); } static irqreturn_t ubd_intr(int irq, void *dev) -- cgit v0.10.2 From 2fe30a34a141c2188ff2cdd670d031088d9c5d20 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:09 -0800 Subject: [PATCH] uml ubd driver: convert do_ubd to a boolean variable do_ubd is actually just a boolean variable - the way it is used currently is a leftover from the old 2.4 block layer, but it is still used; its use is suspicious, but removing it would be too intrusive for now and needs more thinking. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 6cd8988..460d669 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -112,7 +112,9 @@ static DEFINE_SPINLOCK(ubd_io_lock); static DEFINE_MUTEX(ubd_lock); -static void (*do_ubd)(void); +/* XXX - this made sense in 2.4 days, now it's only used as a boolean, and + * probably it doesn't make sense even for that. */ +static int do_ubd; static int ubd_open(struct inode * inode, struct file * filp); static int ubd_release(struct inode * inode, struct file * file); @@ -508,6 +510,7 @@ static inline void ubd_finish(struct request *req, int error) spin_unlock(&ubd_io_lock); } +/* XXX - move this inside ubd_intr. */ /* Called without ubd_io_lock held, and only in interrupt context. */ static void ubd_handler(void) { @@ -515,7 +518,7 @@ static void ubd_handler(void) struct request *rq = elv_next_request(ubd_queue); int n; - do_ubd = NULL; + do_ubd = 0; intr_count++; n = os_read_file(thread_fd, &req, sizeof(req)); if(n != sizeof(req)){ @@ -1043,7 +1046,7 @@ static void do_ubd_request(request_queue_t *q) return; err = prepare_request(req, &io_req); if(!err){ - do_ubd = ubd_handler; + do_ubd = 1; n = os_write_file(thread_fd, (char *) &io_req, sizeof(io_req)); if(n != sizeof(io_req)) -- cgit v0.10.2 From e7f6552f237498c805af9f01aba168b731e0a4ce Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:09 -0800 Subject: [PATCH] uml ubd driver: reformat ubd_config Pure whitespace and style fixes split out from subsequent patch. Some changes (err -> ret) don't make sense now, only later, but I split them out anyway since they cluttered the patch. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 460d669..eed95dc 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -706,27 +706,36 @@ out: static int ubd_config(char *str) { - int n, err; + int n, ret; str = kstrdup(str, GFP_KERNEL); - if(str == NULL){ + if (str == NULL) { printk(KERN_ERR "ubd_config failed to strdup string\n"); - return(1); + ret = 1; + goto out; } - err = ubd_setup_common(str, &n); - if(err){ - kfree(str); - return(-1); + ret = ubd_setup_common(str, &n); + if (ret) { + ret = -1; + goto err_free; + } + if (n == -1) { + ret = 0; + goto out; } - if(n == -1) return(0); mutex_lock(&ubd_lock); - err = ubd_add(n); - if(err) + ret = ubd_add(n); + if (ret) ubd_devs[n].file = NULL; mutex_unlock(&ubd_lock); - return(err); +out: + return ret; + +err_free: + kfree(str); + goto out; } static int ubd_get_config(char *name, char *str, int size, char **error_out) -- cgit v0.10.2 From 84e945e399ce9710a34035ea81eaf5719aa709af Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:10 -0800 Subject: [PATCH] uml ubd driver: use bitfields where possible Use bitfields for boolean fields in ubd data structure. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index eed95dc..3e3bb22 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -174,8 +174,8 @@ struct ubd { __u64 size; struct openflags boot_openflags; struct openflags openflags; - int shared; - int no_cow; + unsigned shared:1; + unsigned no_cow:1; struct cow cow; struct platform_device pdev; }; -- cgit v0.10.2 From 0bf16bffeef65622603af22151156807323d7dc7 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:11 -0800 Subject: [PATCH] uml ubd driver: do not store error codes as ->fd To simplify error handling, make sure fd is saved into ubd_dev->fd only when we are sure it is an fd and not an error code. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 3e3bb22..125a63f 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -578,33 +578,36 @@ static int ubd_open_dev(struct ubd *ubd_dev) struct openflags flags; char **back_ptr; int err, create_cow, *create_ptr; + int fd; ubd_dev->openflags = ubd_dev->boot_openflags; create_cow = 0; create_ptr = (ubd_dev->cow.file != NULL) ? &create_cow : NULL; back_ptr = ubd_dev->no_cow ? NULL : &ubd_dev->cow.file; - ubd_dev->fd = open_ubd_file(ubd_dev->file, &ubd_dev->openflags, ubd_dev->shared, + + fd = open_ubd_file(ubd_dev->file, &ubd_dev->openflags, ubd_dev->shared, back_ptr, &ubd_dev->cow.bitmap_offset, &ubd_dev->cow.bitmap_len, &ubd_dev->cow.data_offset, create_ptr); - if((ubd_dev->fd == -ENOENT) && create_cow){ - ubd_dev->fd = create_cow_file(ubd_dev->file, ubd_dev->cow.file, + if((fd == -ENOENT) && create_cow){ + fd = create_cow_file(ubd_dev->file, ubd_dev->cow.file, ubd_dev->openflags, 1 << 9, PAGE_SIZE, &ubd_dev->cow.bitmap_offset, &ubd_dev->cow.bitmap_len, &ubd_dev->cow.data_offset); - if(ubd_dev->fd >= 0){ + if(fd >= 0){ printk(KERN_INFO "Creating \"%s\" as COW file for " "\"%s\"\n", ubd_dev->file, ubd_dev->cow.file); } } - if(ubd_dev->fd < 0){ + if(fd < 0){ printk("Failed to open '%s', errno = %d\n", ubd_dev->file, - -ubd_dev->fd); - return(ubd_dev->fd); + -fd); + return fd; } + ubd_dev->fd = fd; if(ubd_dev->cow.file != NULL){ err = -ENOMEM; -- cgit v0.10.2 From d8d7c28ec0b50ac57ddc909ae6eca1519473f300 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Mon, 30 Oct 2006 22:07:12 -0800 Subject: [PATCH] uml ubd driver: various little changes Fix a small memory leak in ubd_config, and clearify the confusion which lead to it. Then, some little changes not affecting operations - * move init functions together, * add a comment about a potential problem in case of some evolution in the block layer, * mark all initcalls as static __init functions * mark an used once little function as inline * document that mconsole methods are all called in process context (was triggered when checking ubd mconsole methods). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 125a63f..49c047b 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -202,17 +202,6 @@ struct ubd { struct ubd ubd_devs[MAX_DEV] = { [ 0 ... MAX_DEV - 1 ] = DEFAULT_UBD }; -static int ubd0_init(void) -{ - struct ubd *ubd_dev = &ubd_devs[0]; - - if(ubd_dev->file == NULL) - ubd_dev->file = "root_fs"; - return(0); -} - -__initcall(ubd0_init); - /* Only changed by fake_ide_setup which is a setup */ static int fake_ide = 0; static struct proc_dir_entry *proc_ide_root = NULL; @@ -293,6 +282,10 @@ static int parse_unit(char **ptr) return(n); } +/* If *index_out == -1 at exit, the passed option was a general one; + * otherwise, the str pointer is used (and owned) inside ubd_devs array, so it + * should not be freed on exit. + */ static int ubd_setup_common(char *str, int *index_out) { struct ubd *ubd_dev; @@ -480,8 +473,9 @@ int thread_fd = -1; /* Changed by ubd_handler, which is serialized because interrupts only * happen on CPU 0. + * XXX: currently unused. */ -int intr_count = 0; +static int intr_count = 0; /* call ubd_finish if you need to serialize */ static void __ubd_finish(struct request *req, int error) @@ -554,7 +548,7 @@ void kill_io_thread(void) __uml_exitcall(kill_io_thread); -static int ubd_file_size(struct ubd *ubd_dev, __u64 *size_out) +static inline int ubd_file_size(struct ubd *ubd_dev, __u64 *size_out) { char *file; @@ -724,7 +718,7 @@ static int ubd_config(char *str) } if (n == -1) { ret = 0; - goto out; + goto err_free; } mutex_lock(&ubd_lock); @@ -821,6 +815,7 @@ out: return err; } +/* All these are called by mconsole in process context and without ubd-specific locks. */ static struct mc_device ubd_mc = { .name = "ubd", .config = ubd_config, @@ -829,7 +824,7 @@ static struct mc_device ubd_mc = { .remove = ubd_remove, }; -static int ubd_mc_init(void) +static int __init ubd_mc_init(void) { mconsole_register_dev(&ubd_mc); return 0; @@ -837,13 +832,24 @@ static int ubd_mc_init(void) __initcall(ubd_mc_init); +static int __init ubd0_init(void) +{ + struct ubd *ubd_dev = &ubd_devs[0]; + + if(ubd_dev->file == NULL) + ubd_dev->file = "root_fs"; + return(0); +} + +__initcall(ubd0_init); + static struct platform_driver ubd_driver = { .driver = { .name = DRIVER_NAME, }, }; -int ubd_init(void) +static int __init ubd_init(void) { int i; @@ -871,7 +877,7 @@ int ubd_init(void) late_initcall(ubd_init); -int ubd_driver_init(void){ +static int __init ubd_driver_init(void){ unsigned long stack; int err; @@ -1378,8 +1384,8 @@ void do_io(struct io_thread_req *req) */ int kernel_fd = -1; -/* Only changed by the io thread */ -int io_count = 0; +/* Only changed by the io thread. XXX: currently unused. */ +static int io_count = 0; int io_thread(void *arg) { diff --git a/arch/um/include/mconsole_kern.h b/arch/um/include/mconsole_kern.h index d0b6901..1ea6d928 100644 --- a/arch/um/include/mconsole_kern.h +++ b/arch/um/include/mconsole_kern.h @@ -14,6 +14,7 @@ struct mconsole_entry { struct mc_request request; }; +/* All these methods are called in process context. */ struct mc_device { struct list_head list; char *name; diff --git a/arch/um/kernel/tt/tracer.c b/arch/um/kernel/tt/tracer.c index 9882342..b919535 100644 --- a/arch/um/kernel/tt/tracer.c +++ b/arch/um/kernel/tt/tracer.c @@ -176,7 +176,6 @@ struct { int signal_index[32]; int nsignals = 0; int debug_trace = 0; -extern int io_nsignals, io_count, intr_count; extern void signal_usr1(int sig); -- cgit v0.10.2 From d1480c56feb2a894fd47fb399999a5b320c3e103 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Mon, 30 Oct 2006 22:07:13 -0800 Subject: [PATCH] uml: add _text definition to linker scripts kallsyms now refers to addresses as '_text + 0xADDRESS', rather than just '0xADDRESS', so we need to define _text. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/dyn.lds.S b/arch/um/kernel/dyn.lds.S index 68ed24d..e36f92b 100644 --- a/arch/um/kernel/dyn.lds.S +++ b/arch/um/kernel/dyn.lds.S @@ -14,6 +14,7 @@ SECTIONS * is remapped.*/ __binary_start = .; . = ALIGN(4096); /* Init code and data */ + _text = .; _stext = .; __init_begin = .; .init.text : { diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S index 8eca47a..f630127 100644 --- a/arch/um/kernel/uml.lds.S +++ b/arch/um/kernel/uml.lds.S @@ -25,6 +25,7 @@ SECTIONS . = ALIGN(4096); /* Init code and data */ #endif + _text = .; _stext = .; __init_begin = .; .init.text : { -- cgit v0.10.2 From 525fdb6cc929b515ad7e0be40fd023cff8660ed8 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Mon, 30 Oct 2006 22:07:14 -0800 Subject: [PATCH] uml: add INITCALLS This is the UML piece of the INITCALLS tidying. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-um/common.lds.S b/include/asm-um/common.lds.S index 1010153..f045451 100644 --- a/include/asm-um/common.lds.S +++ b/include/asm-um/common.lds.S @@ -42,13 +42,7 @@ __initcall_start = .; .initcall.init : { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; -- cgit v0.10.2 From 4a279ff1ea1cf325775ada983035123fcdc8e986 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 30 Oct 2006 22:07:15 -0800 Subject: [PATCH] taskstats: fix sub-threads accounting If there are no listeners, taskstats_exit_send() just returns because taskstats_exit_alloc() didn't allocate *tidstats. This is wrong, each sub-thread should do fill_tgid_exit() on exit, otherwise its ->delays is not recorded in ->signal->stats and lost. Q: We don't send TASKSTATS_TYPE_AGGR_TGID when single-threaded process exits. Is it good? How can the listener figure out that it was actually a process exit, not sub-thread? Signed-off-by: Oleg Nesterov Cc: Balbir Singh Acked-by: Shailabh Nagar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 2039585..f45c5e7 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -455,10 +455,9 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, int is_thread_group; struct nlattr *na; - if (!family_registered || !tidstats) + if (!family_registered) return; - rc = 0; /* * Size includes space for nested attributes */ @@ -466,8 +465,15 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); is_thread_group = (tsk->signal->stats != NULL); - if (is_thread_group) - size = 2 * size; /* PID + STATS + TGID + STATS */ + if (is_thread_group) { + /* PID + STATS + TGID + STATS */ + size = 2 * size; + /* fill the tsk->signal->stats structure */ + fill_tgid_exit(tsk); + } + + if (!tidstats) + return; rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, &reply, size); if (rc < 0) @@ -487,11 +493,8 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, goto send; /* - * tsk has/had a thread group so fill the tsk->signal->stats structure * Doesn't matter if tsk is the leader or the last group member leaving */ - - fill_tgid_exit(tsk); if (!group_dead) goto send; -- cgit v0.10.2 From e5d9cbde6ce0001e49994df5fcdcbeff8be8037b Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Mon, 30 Oct 2006 22:07:16 -0800 Subject: [PATCH] eCryptfs: Clean up crypto initialization Clean up the crypto initialization code; let the crypto API take care of the key size checks. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index ed35a9712..82e7d02 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1573,35 +1573,26 @@ out: /** * ecryptfs_process_cipher - Perform cipher initialization. - * @tfm: Crypto context set by this function * @key_tfm: Crypto context for key material, set by this function - * @cipher_name: Name of the cipher. - * @key_size: Size of the key in bytes. + * @cipher_name: Name of the cipher + * @key_size: Size of the key in bytes * * Returns zero on success. Any crypto_tfm structs allocated here * should be released by other functions, such as on a superblock put * event, regardless of whether this function succeeds for fails. */ int -ecryptfs_process_cipher(struct crypto_tfm **tfm, struct crypto_tfm **key_tfm, - char *cipher_name, size_t key_size) +ecryptfs_process_cipher(struct crypto_tfm **key_tfm, char *cipher_name, + size_t *key_size) { char dummy_key[ECRYPTFS_MAX_KEY_BYTES]; int rc; - *tfm = *key_tfm = NULL; - if (key_size > ECRYPTFS_MAX_KEY_BYTES) { + *key_tfm = NULL; + if (*key_size > ECRYPTFS_MAX_KEY_BYTES) { rc = -EINVAL; printk(KERN_ERR "Requested key size is [%Zd] bytes; maximum " - "allowable is [%d]\n", key_size, ECRYPTFS_MAX_KEY_BYTES); - goto out; - } - *tfm = crypto_alloc_tfm(cipher_name, (ECRYPTFS_DEFAULT_CHAINING_MODE - | CRYPTO_TFM_REQ_WEAK_KEY)); - if (!(*tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Unable to allocate crypto cipher with name " - "[%s]\n", cipher_name); + "allowable is [%d]\n", *key_size, ECRYPTFS_MAX_KEY_BYTES); goto out; } *key_tfm = crypto_alloc_tfm(cipher_name, CRYPTO_TFM_REQ_WEAK_KEY); @@ -1611,46 +1602,13 @@ ecryptfs_process_cipher(struct crypto_tfm **tfm, struct crypto_tfm **key_tfm, "[%s]\n", cipher_name); goto out; } - if (key_size < crypto_tfm_alg_min_keysize(*tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; minimum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*tfm)); - goto out; - } - if (key_size < crypto_tfm_alg_min_keysize(*key_tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; minimum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*key_tfm)); - goto out; - } - if (key_size > crypto_tfm_alg_max_keysize(*tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; maximum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*tfm)); - goto out; - } - if (key_size > crypto_tfm_alg_max_keysize(*key_tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; maximum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*key_tfm)); - goto out; - } - get_random_bytes(dummy_key, key_size); - rc = crypto_cipher_setkey(*tfm, dummy_key, key_size); - if (rc) { - printk(KERN_ERR "Error attempting to set key of size [%Zd] for " - "cipher [%s]; rc = [%d]\n", key_size, cipher_name, rc); - rc = -EINVAL; - goto out; - } - rc = crypto_cipher_setkey(*key_tfm, dummy_key, key_size); + if (*key_size == 0) + *key_size = crypto_tfm_alg_max_keysize(*key_tfm); + get_random_bytes(dummy_key, *key_size); + rc = crypto_cipher_setkey(*key_tfm, dummy_key, *key_size); if (rc) { printk(KERN_ERR "Error attempting to set key of size [%Zd] for " - "cipher [%s]; rc = [%d]\n", key_size, cipher_name, rc); + "cipher [%s]; rc = [%d]\n", *key_size, cipher_name, rc); rc = -EINVAL; goto out; } diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 872c995..4112df9 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -473,8 +473,8 @@ ecryptfs_parse_packet_set(struct ecryptfs_crypt_stat *crypt_stat, unsigned char *src, struct dentry *ecryptfs_dentry); int ecryptfs_truncate(struct dentry *dentry, loff_t new_length); int -ecryptfs_process_cipher(struct crypto_tfm **tfm, struct crypto_tfm **key_tfm, - char *cipher_name, size_t key_size); +ecryptfs_process_cipher(struct crypto_tfm **key_tfm, char *cipher_name, + size_t *key_size); int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode); int ecryptfs_inode_set(struct inode *inode, void *lower_inode); void ecryptfs_init_inode(struct inode *inode, struct inode *lower_inode); diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index ba45478..bc706d3 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -493,19 +493,16 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, goto out; } } - if (password_s_ptr->session_key_encryption_key_bytes - < crypto_tfm_alg_min_keysize(tfm)) { - printk(KERN_WARNING "Session key encryption key is [%d] bytes; " - "minimum keysize for selected cipher is [%d] bytes.\n", - password_s_ptr->session_key_encryption_key_bytes, - crypto_tfm_alg_min_keysize(tfm)); - rc = -EINVAL; - goto out; - } if (tfm_mutex) mutex_lock(tfm_mutex); - crypto_cipher_setkey(tfm, password_s_ptr->session_key_encryption_key, - crypt_stat->key_size); + rc = crypto_cipher_setkey(tfm, + password_s_ptr->session_key_encryption_key, + crypt_stat->key_size); + if (rc < 0) { + printk(KERN_ERR "Error setting key for crypto context\n"); + rc = -EINVAL; + goto out_free_tfm; + } /* TODO: virt_to_scatterlist */ encrypted_session_key = (char *)__get_free_page(GFP_KERNEL); if (!encrypted_session_key) { diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 5938a23..a65f486 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -208,7 +208,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) char *cipher_name_dst; char *cipher_name_src; char *cipher_key_bytes_src; - struct crypto_tfm *tmp_tfm; int cipher_name_len; if (!options) { @@ -305,20 +304,12 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) = '\0'; } if (!cipher_key_bytes_set) { - mount_crypt_stat->global_default_cipher_key_size = - ECRYPTFS_DEFAULT_KEY_BYTES; - ecryptfs_printk(KERN_DEBUG, "Cipher key size was not " - "specified. Defaulting to [%d]\n", - mount_crypt_stat-> - global_default_cipher_key_size); + mount_crypt_stat->global_default_cipher_key_size = 0; } rc = ecryptfs_process_cipher( - &tmp_tfm, &mount_crypt_stat->global_key_tfm, mount_crypt_stat->global_default_cipher_name, - mount_crypt_stat->global_default_cipher_key_size); - if (tmp_tfm) - crypto_free_tfm(tmp_tfm); + &mount_crypt_stat->global_default_cipher_key_size); if (rc) { printk(KERN_ERR "Error attempting to initialize cipher [%s] " "with key size [%Zd] bytes; rc = [%d]\n", -- cgit v0.10.2 From 565d9724b8ce49b530287de34aa17f45f21624d5 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Mon, 30 Oct 2006 22:07:17 -0800 Subject: [PATCH] eCryptfs: Hash code to new crypto API Update eCryptfs hash code to the new kernel crypto API. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 82e7d02..f14c5a3 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -94,25 +94,31 @@ static int ecryptfs_calculate_md5(char *dst, struct ecryptfs_crypt_stat *crypt_stat, char *src, int len) { - int rc = 0; struct scatterlist sg; + struct hash_desc desc = { + .tfm = crypt_stat->hash_tfm, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; + int rc = 0; - mutex_lock(&crypt_stat->cs_md5_tfm_mutex); + mutex_lock(&crypt_stat->cs_hash_tfm_mutex); sg_init_one(&sg, (u8 *)src, len); - if (!crypt_stat->md5_tfm) { - crypt_stat->md5_tfm = - crypto_alloc_tfm("md5", CRYPTO_TFM_REQ_MAY_SLEEP); - if (!crypt_stat->md5_tfm) { - rc = -ENOMEM; + if (!desc.tfm) { + desc.tfm = crypto_alloc_hash(ECRYPTFS_DEFAULT_HASH, 0, + CRYPTO_ALG_ASYNC); + if (IS_ERR(desc.tfm)) { + rc = PTR_ERR(desc.tfm); ecryptfs_printk(KERN_ERR, "Error attempting to " - "allocate crypto context\n"); + "allocate crypto context; rc = [%d]\n", + rc); goto out; } + crypt_stat->hash_tfm = desc.tfm; } - crypto_digest_init(crypt_stat->md5_tfm); - crypto_digest_update(crypt_stat->md5_tfm, &sg, 1); - crypto_digest_final(crypt_stat->md5_tfm, dst); - mutex_unlock(&crypt_stat->cs_md5_tfm_mutex); + crypto_hash_init(&desc); + crypto_hash_update(&desc, &sg, len); + crypto_hash_final(&desc, dst); + mutex_unlock(&crypt_stat->cs_hash_tfm_mutex); out: return rc; } @@ -178,7 +184,7 @@ ecryptfs_init_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) memset((void *)crypt_stat, 0, sizeof(struct ecryptfs_crypt_stat)); mutex_init(&crypt_stat->cs_mutex); mutex_init(&crypt_stat->cs_tfm_mutex); - mutex_init(&crypt_stat->cs_md5_tfm_mutex); + mutex_init(&crypt_stat->cs_hash_tfm_mutex); ECRYPTFS_SET_FLAG(crypt_stat->flags, ECRYPTFS_STRUCT_INITIALIZED); } @@ -192,8 +198,8 @@ void ecryptfs_destruct_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) { if (crypt_stat->tfm) crypto_free_tfm(crypt_stat->tfm); - if (crypt_stat->md5_tfm) - crypto_free_tfm(crypt_stat->md5_tfm); + if (crypt_stat->hash_tfm) + crypto_free_hash(crypt_stat->hash_tfm); memset(crypt_stat, 0, sizeof(struct ecryptfs_crypt_stat)); } diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 4112df9..840aa01 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -175,6 +175,7 @@ ecryptfs_get_key_payload_data(struct key *key) #define ECRYPTFS_DEFAULT_CIPHER "aes" #define ECRYPTFS_DEFAULT_KEY_BYTES 16 #define ECRYPTFS_DEFAULT_CHAINING_MODE CRYPTO_TFM_MODE_CBC +#define ECRYPTFS_DEFAULT_HASH "md5" #define ECRYPTFS_TAG_3_PACKET_TYPE 0x8C #define ECRYPTFS_TAG_11_PACKET_TYPE 0xED #define MD5_DIGEST_SIZE 16 @@ -205,14 +206,14 @@ struct ecryptfs_crypt_stat { unsigned int extent_mask; struct ecryptfs_mount_crypt_stat *mount_crypt_stat; struct crypto_tfm *tfm; - struct crypto_tfm *md5_tfm; /* Crypto context for generating - * the initialization vectors */ + struct crypto_hash *hash_tfm; /* Crypto context for generating + * the initialization vectors */ unsigned char cipher[ECRYPTFS_MAX_CIPHER_NAME_SIZE]; unsigned char key[ECRYPTFS_MAX_KEY_BYTES]; unsigned char root_iv[ECRYPTFS_MAX_IV_BYTES]; unsigned char keysigs[ECRYPTFS_MAX_NUM_KEYSIGS][ECRYPTFS_SIG_SIZE_HEX]; struct mutex cs_tfm_mutex; - struct mutex cs_md5_tfm_mutex; + struct mutex cs_hash_tfm_mutex; struct mutex cs_mutex; }; -- cgit v0.10.2 From 8bba066f4e3854755a303cee37ea37bd080a46b3 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Mon, 30 Oct 2006 22:07:18 -0800 Subject: [PATCH] eCryptfs: Cipher code to new crypto API Update cipher block encryption code to the new crypto API. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index f14c5a3..2a1b6aa 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -123,6 +123,28 @@ out: return rc; } +int ecryptfs_crypto_api_algify_cipher_name(char **algified_name, + char *cipher_name, + char *chaining_modifier) +{ + int cipher_name_len = strlen(cipher_name); + int chaining_modifier_len = strlen(chaining_modifier); + int algified_name_len; + int rc; + + algified_name_len = (chaining_modifier_len + cipher_name_len + 3); + (*algified_name) = kmalloc(algified_name_len, GFP_KERNEL); + if (!(algified_name)) { + rc = -ENOMEM; + goto out; + } + snprintf((*algified_name), algified_name_len, "%s(%s)", + chaining_modifier, cipher_name); + rc = 0; +out: + return rc; +} + /** * ecryptfs_derive_iv * @iv: destination for the derived iv vale @@ -197,7 +219,7 @@ ecryptfs_init_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) void ecryptfs_destruct_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) { if (crypt_stat->tfm) - crypto_free_tfm(crypt_stat->tfm); + crypto_free_blkcipher(crypt_stat->tfm); if (crypt_stat->hash_tfm) crypto_free_hash(crypt_stat->hash_tfm); memset(crypt_stat, 0, sizeof(struct ecryptfs_crypt_stat)); @@ -209,7 +231,7 @@ void ecryptfs_destruct_mount_crypt_stat( if (mount_crypt_stat->global_auth_tok_key) key_put(mount_crypt_stat->global_auth_tok_key); if (mount_crypt_stat->global_key_tfm) - crypto_free_tfm(mount_crypt_stat->global_key_tfm); + crypto_free_blkcipher(mount_crypt_stat->global_key_tfm); memset(mount_crypt_stat, 0, sizeof(struct ecryptfs_mount_crypt_stat)); } @@ -275,6 +297,11 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, struct scatterlist *src_sg, int size, unsigned char *iv) { + struct blkcipher_desc desc = { + .tfm = crypt_stat->tfm, + .info = iv, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; int rc = 0; BUG_ON(!crypt_stat || !crypt_stat->tfm @@ -288,8 +315,8 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, } /* Consider doing this once, when the file is opened */ mutex_lock(&crypt_stat->cs_tfm_mutex); - rc = crypto_cipher_setkey(crypt_stat->tfm, crypt_stat->key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(crypt_stat->tfm, crypt_stat->key, + crypt_stat->key_size); if (rc) { ecryptfs_printk(KERN_ERR, "Error setting key; rc = [%d]\n", rc); @@ -298,7 +325,7 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, goto out; } ecryptfs_printk(KERN_DEBUG, "Encrypting [%d] bytes.\n", size); - crypto_cipher_encrypt_iv(crypt_stat->tfm, dest_sg, src_sg, size, iv); + crypto_blkcipher_encrypt_iv(&desc, dest_sg, src_sg, size); mutex_unlock(&crypt_stat->cs_tfm_mutex); out: return rc; @@ -681,12 +708,17 @@ static int decrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, struct scatterlist *src_sg, int size, unsigned char *iv) { + struct blkcipher_desc desc = { + .tfm = crypt_stat->tfm, + .info = iv, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; int rc = 0; /* Consider doing this once, when the file is opened */ mutex_lock(&crypt_stat->cs_tfm_mutex); - rc = crypto_cipher_setkey(crypt_stat->tfm, crypt_stat->key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(crypt_stat->tfm, crypt_stat->key, + crypt_stat->key_size); if (rc) { ecryptfs_printk(KERN_ERR, "Error setting key; rc = [%d]\n", rc); @@ -695,8 +727,7 @@ static int decrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, goto out; } ecryptfs_printk(KERN_DEBUG, "Decrypting [%d] bytes.\n", size); - rc = crypto_cipher_decrypt_iv(crypt_stat->tfm, dest_sg, src_sg, size, - iv); + rc = crypto_blkcipher_decrypt_iv(&desc, dest_sg, src_sg, size); mutex_unlock(&crypt_stat->cs_tfm_mutex); if (rc) { ecryptfs_printk(KERN_ERR, "Error decrypting; rc = [%d]\n", @@ -765,6 +796,7 @@ ecryptfs_decrypt_page_offset(struct ecryptfs_crypt_stat *crypt_stat, */ int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat) { + char *full_alg_name; int rc = -EINVAL; if (!crypt_stat->cipher) { @@ -781,16 +813,24 @@ int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat) goto out; } mutex_lock(&crypt_stat->cs_tfm_mutex); - crypt_stat->tfm = crypto_alloc_tfm(crypt_stat->cipher, - ECRYPTFS_DEFAULT_CHAINING_MODE - | CRYPTO_TFM_REQ_WEAK_KEY); - mutex_unlock(&crypt_stat->cs_tfm_mutex); + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, + crypt_stat->cipher, "cbc"); + if (rc) + goto out; + crypt_stat->tfm = crypto_alloc_blkcipher(full_alg_name, 0, + CRYPTO_ALG_ASYNC); + kfree(full_alg_name); if (!crypt_stat->tfm) { ecryptfs_printk(KERN_ERR, "cryptfs: init_crypt_ctx(): " "Error initializing cipher [%s]\n", crypt_stat->cipher); + mutex_unlock(&crypt_stat->cs_tfm_mutex); goto out; } + crypto_blkcipher_set_flags(crypt_stat->tfm, + (ECRYPTFS_DEFAULT_CHAINING_MODE + | CRYPTO_TFM_REQ_WEAK_KEY)); + mutex_unlock(&crypt_stat->cs_tfm_mutex); rc = 0; out: return rc; @@ -1588,10 +1628,11 @@ out: * event, regardless of whether this function succeeds for fails. */ int -ecryptfs_process_cipher(struct crypto_tfm **key_tfm, char *cipher_name, +ecryptfs_process_cipher(struct crypto_blkcipher **key_tfm, char *cipher_name, size_t *key_size) { char dummy_key[ECRYPTFS_MAX_KEY_BYTES]; + char *full_alg_name; int rc; *key_tfm = NULL; @@ -1601,17 +1642,26 @@ ecryptfs_process_cipher(struct crypto_tfm **key_tfm, char *cipher_name, "allowable is [%d]\n", *key_size, ECRYPTFS_MAX_KEY_BYTES); goto out; } - *key_tfm = crypto_alloc_tfm(cipher_name, CRYPTO_TFM_REQ_WEAK_KEY); - if (!(*key_tfm)) { - rc = -EINVAL; + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, cipher_name, + "ecb"); + if (rc) + goto out; + *key_tfm = crypto_alloc_blkcipher(full_alg_name, 0, CRYPTO_ALG_ASYNC); + kfree(full_alg_name); + if (IS_ERR(*key_tfm)) { + rc = PTR_ERR(*key_tfm); printk(KERN_ERR "Unable to allocate crypto cipher with name " - "[%s]\n", cipher_name); + "[%s]; rc = [%d]\n", cipher_name, rc); goto out; } - if (*key_size == 0) - *key_size = crypto_tfm_alg_max_keysize(*key_tfm); + crypto_blkcipher_set_flags(*key_tfm, CRYPTO_TFM_REQ_WEAK_KEY); + if (*key_size == 0) { + struct blkcipher_alg *alg = crypto_blkcipher_alg(*key_tfm); + + *key_size = alg->max_keysize; + } get_random_bytes(dummy_key, *key_size); - rc = crypto_cipher_setkey(*key_tfm, dummy_key, *key_size); + rc = crypto_blkcipher_setkey(*key_tfm, dummy_key, *key_size); if (rc) { printk(KERN_ERR "Error attempting to set key of size [%Zd] for " "cipher [%s]; rc = [%d]\n", *key_size, cipher_name, rc); diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 840aa01..199fcda 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -205,7 +205,7 @@ struct ecryptfs_crypt_stat { size_t extent_shift; unsigned int extent_mask; struct ecryptfs_mount_crypt_stat *mount_crypt_stat; - struct crypto_tfm *tfm; + struct crypto_blkcipher *tfm; struct crypto_hash *hash_tfm; /* Crypto context for generating * the initialization vectors */ unsigned char cipher[ECRYPTFS_MAX_CIPHER_NAME_SIZE]; @@ -245,7 +245,7 @@ struct ecryptfs_mount_crypt_stat { struct ecryptfs_auth_tok *global_auth_tok; struct key *global_auth_tok_key; size_t global_default_cipher_key_size; - struct crypto_tfm *global_key_tfm; + struct crypto_blkcipher *global_key_tfm; struct mutex global_key_tfm_mutex; unsigned char global_default_cipher_name[ECRYPTFS_MAX_CIPHER_NAME_SIZE + 1]; @@ -426,6 +426,9 @@ void ecryptfs_destruct_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat); void ecryptfs_destruct_mount_crypt_stat( struct ecryptfs_mount_crypt_stat *mount_crypt_stat); int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat); +int ecryptfs_crypto_api_algify_cipher_name(char **algified_name, + char *cipher_name, + char *chaining_modifier); int ecryptfs_write_inode_size_to_header(struct file *lower_file, struct inode *lower_inode, struct inode *inode); @@ -474,7 +477,7 @@ ecryptfs_parse_packet_set(struct ecryptfs_crypt_stat *crypt_stat, unsigned char *src, struct dentry *ecryptfs_dentry); int ecryptfs_truncate(struct dentry *dentry, loff_t new_length); int -ecryptfs_process_cipher(struct crypto_tfm **key_tfm, char *cipher_name, +ecryptfs_process_cipher(struct crypto_blkcipher **key_tfm, char *cipher_name, size_t *key_size); int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode); int ecryptfs_inode_set(struct inode *inode, void *lower_inode); diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index bc706d3..c3746f5 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -458,14 +458,16 @@ out: static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, struct ecryptfs_crypt_stat *crypt_stat) { - int rc = 0; struct ecryptfs_password *password_s_ptr; - struct crypto_tfm *tfm = NULL; struct scatterlist src_sg[2], dst_sg[2]; struct mutex *tfm_mutex = NULL; /* TODO: Use virt_to_scatterlist for these */ char *encrypted_session_key; char *session_key; + struct blkcipher_desc desc = { + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; + int rc = 0; password_s_ptr = &auth_tok->token.password; if (ECRYPTFS_CHECK_FLAG(password_s_ptr->flags, @@ -482,22 +484,32 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, if (!strcmp(crypt_stat->cipher, crypt_stat->mount_crypt_stat->global_default_cipher_name) && crypt_stat->mount_crypt_stat->global_key_tfm) { - tfm = crypt_stat->mount_crypt_stat->global_key_tfm; + desc.tfm = crypt_stat->mount_crypt_stat->global_key_tfm; tfm_mutex = &crypt_stat->mount_crypt_stat->global_key_tfm_mutex; } else { - tfm = crypto_alloc_tfm(crypt_stat->cipher, - CRYPTO_TFM_REQ_WEAK_KEY); - if (!tfm) { - printk(KERN_ERR "Error allocating crypto context\n"); - rc = -ENOMEM; + char *full_alg_name; + + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, + crypt_stat->cipher, + "ecb"); + if (rc) + goto out; + desc.tfm = crypto_alloc_blkcipher(full_alg_name, 0, + CRYPTO_ALG_ASYNC); + kfree(full_alg_name); + if (IS_ERR(desc.tfm)) { + rc = PTR_ERR(desc.tfm); + printk(KERN_ERR "Error allocating crypto context; " + "rc = [%d]\n", rc); goto out; } + crypto_blkcipher_set_flags(desc.tfm, CRYPTO_TFM_REQ_WEAK_KEY); } if (tfm_mutex) mutex_lock(tfm_mutex); - rc = crypto_cipher_setkey(tfm, - password_s_ptr->session_key_encryption_key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(desc.tfm, + password_s_ptr->session_key_encryption_key, + crypt_stat->key_size); if (rc < 0) { printk(KERN_ERR "Error setting key for crypto context\n"); rc = -EINVAL; @@ -528,9 +540,12 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, auth_tok->session_key.decrypted_key_size = auth_tok->session_key.encrypted_key_size; dst_sg[0].length = auth_tok->session_key.encrypted_key_size; - /* TODO: Handle error condition */ - crypto_cipher_decrypt(tfm, dst_sg, src_sg, - auth_tok->session_key.encrypted_key_size); + rc = crypto_blkcipher_decrypt(&desc, dst_sg, src_sg, + auth_tok->session_key.encrypted_key_size); + if (rc) { + printk(KERN_ERR "Error decrypting; rc = [%d]\n", rc); + goto out_free_memory; + } auth_tok->session_key.decrypted_key_size = auth_tok->session_key.encrypted_key_size; memcpy(auth_tok->session_key.decrypted_key, session_key, @@ -543,6 +558,7 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, if (ecryptfs_verbosity > 0) ecryptfs_dump_hex(crypt_stat->key, crypt_stat->key_size); +out_free_memory: memset(encrypted_session_key, 0, PAGE_CACHE_SIZE); free_page((unsigned long)encrypted_session_key); memset(session_key, 0, PAGE_CACHE_SIZE); @@ -551,7 +567,7 @@ out_free_tfm: if (tfm_mutex) mutex_unlock(tfm_mutex); else - crypto_free_tfm(tfm); + crypto_free_blkcipher(desc.tfm); out: return rc; } @@ -800,19 +816,21 @@ write_tag_3_packet(char *dest, size_t max, struct ecryptfs_auth_tok *auth_tok, struct ecryptfs_crypt_stat *crypt_stat, struct ecryptfs_key_record *key_rec, size_t *packet_size) { - int rc = 0; - size_t i; size_t signature_is_valid = 0; size_t encrypted_session_key_valid = 0; char session_key_encryption_key[ECRYPTFS_MAX_KEY_BYTES]; struct scatterlist dest_sg[2]; struct scatterlist src_sg[2]; - struct crypto_tfm *tfm = NULL; struct mutex *tfm_mutex = NULL; size_t key_rec_size; size_t packet_size_length; size_t cipher_code; + struct blkcipher_desc desc = { + .tfm = NULL, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; + int rc = 0; (*packet_size) = 0; /* Check for a valid signature on the auth_tok */ @@ -879,33 +897,48 @@ write_tag_3_packet(char *dest, size_t max, struct ecryptfs_auth_tok *auth_tok, if (!strcmp(crypt_stat->cipher, crypt_stat->mount_crypt_stat->global_default_cipher_name) && crypt_stat->mount_crypt_stat->global_key_tfm) { - tfm = crypt_stat->mount_crypt_stat->global_key_tfm; + desc.tfm = crypt_stat->mount_crypt_stat->global_key_tfm; tfm_mutex = &crypt_stat->mount_crypt_stat->global_key_tfm_mutex; - } else - tfm = crypto_alloc_tfm(crypt_stat->cipher, 0); - if (!tfm) { - ecryptfs_printk(KERN_ERR, "Could not initialize crypto " - "context for cipher [%s]\n", - crypt_stat->cipher); - rc = -EINVAL; - goto out; + } else { + char *full_alg_name; + + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, + crypt_stat->cipher, + "ecb"); + if (rc) + goto out; + desc.tfm = crypto_alloc_blkcipher(full_alg_name, 0, + CRYPTO_ALG_ASYNC); + kfree(full_alg_name); + if (IS_ERR(desc.tfm)) { + rc = PTR_ERR(desc.tfm); + ecryptfs_printk(KERN_ERR, "Could not initialize crypto " + "context for cipher [%s]; rc = [%d]\n", + crypt_stat->cipher, rc); + goto out; + } + crypto_blkcipher_set_flags(desc.tfm, CRYPTO_TFM_REQ_WEAK_KEY); } if (tfm_mutex) mutex_lock(tfm_mutex); - rc = crypto_cipher_setkey(tfm, session_key_encryption_key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(desc.tfm, session_key_encryption_key, + crypt_stat->key_size); if (rc < 0) { if (tfm_mutex) mutex_unlock(tfm_mutex); ecryptfs_printk(KERN_ERR, "Error setting key for crypto " - "context\n"); + "context; rc = [%d]\n", rc); goto out; } rc = 0; ecryptfs_printk(KERN_DEBUG, "Encrypting [%d] bytes of the key\n", crypt_stat->key_size); - crypto_cipher_encrypt(tfm, dest_sg, src_sg, - (*key_rec).enc_key_size); + rc = crypto_blkcipher_encrypt(&desc, dest_sg, src_sg, + (*key_rec).enc_key_size); + if (rc) { + printk(KERN_ERR "Error encrypting; rc = [%d]\n", rc); + goto out; + } if (tfm_mutex) mutex_unlock(tfm_mutex); ecryptfs_printk(KERN_DEBUG, "This should be the encrypted key:\n"); @@ -968,8 +1001,8 @@ encrypted_session_key_set: (*key_rec).enc_key_size); (*packet_size) += (*key_rec).enc_key_size; out: - if (tfm && !tfm_mutex) - crypto_free_tfm(tfm); + if (desc.tfm && !tfm_mutex) + crypto_free_blkcipher(desc.tfm); if (rc) (*packet_size) = 0; return rc; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a65f486..a78d87d 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -315,6 +315,8 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) "with key size [%Zd] bytes; rc = [%d]\n", mount_crypt_stat->global_default_cipher_name, mount_crypt_stat->global_default_cipher_key_size, rc); + mount_crypt_stat->global_key_tfm = NULL; + mount_crypt_stat->global_auth_tok_key = NULL; rc = -EINVAL; goto out; } -- cgit v0.10.2 From 7ff1d74f5670329ac4b5959a675f8698ba95be20 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Mon, 30 Oct 2006 22:07:19 -0800 Subject: [PATCH] eCryptfs: Consolidate lower dentry_open's Opens on lower dentry objects happen in several places in eCryptfs, and they all involve the same steps (dget, mntget, dentry_open). This patch consolidates the lower open events into a single function call. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 2a1b6aa..f49f105 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1191,28 +1191,28 @@ int ecryptfs_cipher_code_to_string(char *str, u16 cipher_code) int ecryptfs_read_header_region(char *data, struct dentry *dentry, struct vfsmount *mnt) { - struct file *file; + struct file *lower_file; mm_segment_t oldfs; int rc; - mnt = mntget(mnt); - file = dentry_open(dentry, mnt, O_RDONLY); - if (IS_ERR(file)) { - ecryptfs_printk(KERN_DEBUG, "Error opening file to " - "read header region\n"); - mntput(mnt); - rc = PTR_ERR(file); + if ((rc = ecryptfs_open_lower_file(&lower_file, dentry, mnt, + O_RDONLY))) { + printk(KERN_ERR + "Error opening lower_file to read header region\n"); goto out; } - file->f_pos = 0; + lower_file->f_pos = 0; oldfs = get_fs(); set_fs(get_ds()); /* For releases 0.1 and 0.2, all of the header information * fits in the first data extent-sized region. */ - rc = file->f_op->read(file, (char __user *)data, - ECRYPTFS_DEFAULT_EXTENT_SIZE, &file->f_pos); + rc = lower_file->f_op->read(lower_file, (char __user *)data, + ECRYPTFS_DEFAULT_EXTENT_SIZE, &lower_file->f_pos); set_fs(oldfs); - fput(file); + if ((rc = ecryptfs_close_lower_file(lower_file))) { + printk(KERN_ERR "Error closing lower_file\n"); + goto out; + } rc = 0; out: return rc; diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 199fcda..f992533 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -482,5 +482,9 @@ ecryptfs_process_cipher(struct crypto_blkcipher **key_tfm, char *cipher_name, int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode); int ecryptfs_inode_set(struct inode *inode, void *lower_inode); void ecryptfs_init_inode(struct inode *inode, struct inode *lower_inode); +int ecryptfs_open_lower_file(struct file **lower_file, + struct dentry *lower_dentry, + struct vfsmount *lower_mnt, int flags); +int ecryptfs_close_lower_file(struct file *lower_file); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index c8550c9..a92ef05 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -198,6 +198,33 @@ retry: struct kmem_cache *ecryptfs_file_info_cache; +int ecryptfs_open_lower_file(struct file **lower_file, + struct dentry *lower_dentry, + struct vfsmount *lower_mnt, int flags) +{ + int rc = 0; + + dget(lower_dentry); + mntget(lower_mnt); + *lower_file = dentry_open(lower_dentry, lower_mnt, flags); + if (IS_ERR(*lower_file)) { + printk(KERN_ERR "Error opening lower file for lower_dentry " + "[0x%p], lower_mnt [0x%p], and flags [0x%x]\n", + lower_dentry, lower_mnt, flags); + rc = PTR_ERR(*lower_file); + *lower_file = NULL; + goto out; + } +out: + return rc; +} + +int ecryptfs_close_lower_file(struct file *lower_file) +{ + fput(lower_file); + return 0; +} + /** * ecryptfs_open * @inode: inode speciying file to open @@ -244,19 +271,15 @@ static int ecryptfs_open(struct inode *inode, struct file *file) ECRYPTFS_SET_FLAG(crypt_stat->flags, ECRYPTFS_ENCRYPTED); } mutex_unlock(&crypt_stat->cs_mutex); - /* This mntget & dget is undone via fput when the file is released */ - dget(lower_dentry); lower_flags = file->f_flags; if ((lower_flags & O_ACCMODE) == O_WRONLY) lower_flags = (lower_flags & O_ACCMODE) | O_RDWR; if (file->f_flags & O_APPEND) lower_flags &= ~O_APPEND; lower_mnt = ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry); - mntget(lower_mnt); /* Corresponding fput() in ecryptfs_release() */ - lower_file = dentry_open(lower_dentry, lower_mnt, lower_flags); - if (IS_ERR(lower_file)) { - rc = PTR_ERR(lower_file); + if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt, + lower_flags))) { ecryptfs_printk(KERN_ERR, "Error opening lower file\n"); goto out_puts; } @@ -341,11 +364,16 @@ static int ecryptfs_release(struct inode *inode, struct file *file) struct file *lower_file = ecryptfs_file_to_lower(file); struct ecryptfs_file_info *file_info = ecryptfs_file_to_private(file); struct inode *lower_inode = ecryptfs_inode_to_lower(inode); + int rc; - fput(lower_file); + if ((rc = ecryptfs_close_lower_file(lower_file))) { + printk(KERN_ERR "Error closing lower_file\n"); + goto out; + } inode->i_blocks = lower_inode->i_blocks; kmem_cache_free(ecryptfs_file_info_cache, file_info); - return 0; +out: + return rc; } static int diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index efdd2b7..2f2c6cf 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -231,7 +231,6 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) int lower_flags; struct ecryptfs_crypt_stat *crypt_stat; struct dentry *lower_dentry; - struct dentry *tlower_dentry = NULL; struct file *lower_file; struct inode *inode, *lower_inode; struct vfsmount *lower_mnt; @@ -241,30 +240,19 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) lower_dentry->d_name.name); inode = ecryptfs_dentry->d_inode; crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat; - tlower_dentry = dget(lower_dentry); - if (!tlower_dentry) { - rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Error dget'ing lower_dentry\n"); - goto out; - } lower_flags = ((O_CREAT | O_WRONLY | O_TRUNC) & O_ACCMODE) | O_RDWR; #if BITS_PER_LONG != 32 lower_flags |= O_LARGEFILE; #endif lower_mnt = ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry); - mntget(lower_mnt); /* Corresponding fput() at end of this function */ - lower_file = dentry_open(tlower_dentry, lower_mnt, lower_flags); - if (IS_ERR(lower_file)) { - rc = PTR_ERR(lower_file); + if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt, + lower_flags))) { ecryptfs_printk(KERN_ERR, "Error opening dentry; rc = [%i]\n", rc); goto out; } - /* fput(lower_file) should handle the puts if we do this */ - lower_file->f_dentry = tlower_dentry; - lower_file->f_vfsmnt = lower_mnt; - lower_inode = tlower_dentry->d_inode; + lower_inode = lower_dentry->d_inode; if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) { ecryptfs_printk(KERN_DEBUG, "This is a directory\n"); ECRYPTFS_CLEAR_FLAG(crypt_stat->flags, ECRYPTFS_ENCRYPTED); @@ -285,7 +273,8 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) } rc = grow_file(ecryptfs_dentry, lower_file, inode, lower_inode); out_fput: - fput(lower_file); + if ((rc = ecryptfs_close_lower_file(lower_file))) + printk(KERN_ERR "Error closing lower_file\n"); out: return rc; } @@ -832,12 +821,11 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) } lower_dentry = ecryptfs_dentry_to_lower(dentry); /* This dget & mntget is released through fput at out_fput: */ - dget(lower_dentry); lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry); - mntget(lower_mnt); - lower_file = dentry_open(lower_dentry, lower_mnt, O_RDWR); - if (unlikely(IS_ERR(lower_file))) { - rc = PTR_ERR(lower_file); + if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt, + O_RDWR))) { + ecryptfs_printk(KERN_ERR, + "Error opening dentry; rc = [%i]\n", rc); goto out_free; } ecryptfs_set_file_lower(&fake_ecryptfs_file, lower_file); @@ -879,7 +867,8 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) = CURRENT_TIME; mark_inode_dirty_sync(inode); out_fput: - fput(lower_file); + if ((rc = ecryptfs_close_lower_file(lower_file))) + printk(KERN_ERR "Error closing lower_file\n"); out_free: if (ecryptfs_file_to_private(&fake_ecryptfs_file)) kmem_cache_free(ecryptfs_file_info_cache, -- cgit v0.10.2 From 316bb95e8ed0ddcd767e8aa54264b6c6190f150c Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Mon, 30 Oct 2006 22:07:20 -0800 Subject: [PATCH] eCryptfs: Remove ecryptfs_umount_begin There is no point to calling the lower umount_begin when the eCryptfs umount_begin is called. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index c337c04..825757a 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -138,23 +138,6 @@ static void ecryptfs_clear_inode(struct inode *inode) } /** - * ecryptfs_umount_begin - * - * Called in do_umount(). - */ -static void ecryptfs_umount_begin(struct vfsmount *vfsmnt, int flags) -{ - struct vfsmount *lower_mnt = - ecryptfs_dentry_to_lower_mnt(vfsmnt->mnt_sb->s_root); - struct super_block *lower_sb; - - mntput(lower_mnt); - lower_sb = lower_mnt->mnt_sb; - if (lower_sb->s_op->umount_begin) - lower_sb->s_op->umount_begin(lower_mnt, flags); -} - -/** * ecryptfs_show_options * * Prints the directory we are currently mounted over. @@ -193,6 +176,5 @@ struct super_operations ecryptfs_sops = { .statfs = ecryptfs_statfs, .remount_fs = NULL, .clear_inode = ecryptfs_clear_inode, - .umount_begin = ecryptfs_umount_begin, .show_options = ecryptfs_show_options }; -- cgit v0.10.2 From 45ec4ababe999cb95f9c0cad03b2689cb0b77a2b Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Mon, 30 Oct 2006 22:07:20 -0800 Subject: [PATCH] eCryptfs: Fix handling of lower d_count Fix the use of dget/dput calls to balance out on the lower filesystem. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c index f0d2a43..0b9992a 100644 --- a/fs/ecryptfs/dentry.c +++ b/fs/ecryptfs/dentry.c @@ -24,6 +24,7 @@ #include #include +#include #include "ecryptfs_kernel.h" /** @@ -76,8 +77,13 @@ static void ecryptfs_d_release(struct dentry *dentry) if (ecryptfs_dentry_to_private(dentry)) kmem_cache_free(ecryptfs_dentry_info_cache, ecryptfs_dentry_to_private(dentry)); - if (lower_dentry) + if (lower_dentry) { + struct vfsmount *lower_mnt = + ecryptfs_dentry_to_lower_mnt(dentry); + + mntput(lower_mnt); dput(lower_dentry); + } return; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 2f2c6cf..ff4865d 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -325,7 +325,6 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, struct dentry *lower_dir_dentry; struct dentry *lower_dentry; struct vfsmount *lower_mnt; - struct dentry *tlower_dentry = NULL; char *encoded_name; unsigned int encoded_namelen; struct ecryptfs_crypt_stat *crypt_stat = NULL; @@ -336,27 +335,32 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, lower_dir_dentry = ecryptfs_dentry_to_lower(dentry->d_parent); dentry->d_op = &ecryptfs_dops; if ((dentry->d_name.len == 1 && !strcmp(dentry->d_name.name, ".")) - || (dentry->d_name.len == 2 && !strcmp(dentry->d_name.name, ".."))) - goto out_drop; + || (dentry->d_name.len == 2 + && !strcmp(dentry->d_name.name, ".."))) { + d_drop(dentry); + goto out; + } encoded_namelen = ecryptfs_encode_filename(crypt_stat, dentry->d_name.name, dentry->d_name.len, &encoded_name); if (encoded_namelen < 0) { rc = encoded_namelen; - goto out_drop; + d_drop(dentry); + goto out; } ecryptfs_printk(KERN_DEBUG, "encoded_name = [%s]; encoded_namelen " "= [%d]\n", encoded_name, encoded_namelen); lower_dentry = lookup_one_len(encoded_name, lower_dir_dentry, encoded_namelen - 1); kfree(encoded_name); - lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent)); if (IS_ERR(lower_dentry)) { ecryptfs_printk(KERN_ERR, "ERR from lower_dentry\n"); rc = PTR_ERR(lower_dentry); - goto out_drop; + d_drop(dentry); + goto out; } + lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent)); ecryptfs_printk(KERN_DEBUG, "lower_dentry = [%p]; lower_dentry->" "d_name.name = [%s]\n", lower_dentry, lower_dentry->d_name.name); @@ -397,12 +401,6 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, "as we *think* we are about to unlink\n"); goto out; } - tlower_dentry = dget(lower_dentry); - if (!tlower_dentry || IS_ERR(tlower_dentry)) { - rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Cannot dget lower_dentry\n"); - goto out_dput; - } /* Released in this function */ page_virt = (char *)kmem_cache_alloc(ecryptfs_header_cache_2, @@ -414,7 +412,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, goto out_dput; } memset(page_virt, 0, PAGE_CACHE_SIZE); - rc = ecryptfs_read_header_region(page_virt, tlower_dentry, nd->mnt); + rc = ecryptfs_read_header_region(page_virt, lower_dentry, nd->mnt); crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; if (!ECRYPTFS_CHECK_FLAG(crypt_stat->flags, ECRYPTFS_POLICY_APPLIED)) ecryptfs_set_default_sizes(crypt_stat); @@ -437,9 +435,6 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, out_dput: dput(lower_dentry); - if (tlower_dentry) - dput(tlower_dentry); -out_drop: d_drop(dentry); out: return ERR_PTR(rc); @@ -475,8 +470,8 @@ out_lock: unlock_dir(lower_dir_dentry); dput(lower_new_dentry); dput(lower_old_dentry); - if (!new_dentry->d_inode) - d_drop(new_dentry); + d_drop(new_dentry); + d_drop(old_dentry); return rc; } @@ -565,41 +560,24 @@ out: static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) { - int rc = 0; - struct dentry *tdentry = NULL; struct dentry *lower_dentry; - struct dentry *tlower_dentry = NULL; struct dentry *lower_dir_dentry; + int rc; lower_dentry = ecryptfs_dentry_to_lower(dentry); - if (!(tdentry = dget(dentry))) { - rc = -EINVAL; - ecryptfs_printk(KERN_ERR, "Error dget'ing dentry [%p]\n", - dentry); - goto out; - } + dget(dentry); lower_dir_dentry = lock_parent(lower_dentry); - if (!(tlower_dentry = dget(lower_dentry))) { - rc = -EINVAL; - ecryptfs_printk(KERN_ERR, "Error dget'ing lower_dentry " - "[%p]\n", lower_dentry); - goto out; - } + dget(lower_dentry); rc = vfs_rmdir(lower_dir_dentry->d_inode, lower_dentry); - if (!rc) { - d_delete(tlower_dentry); - tlower_dentry = NULL; - } + dput(lower_dentry); + if (!rc) + d_delete(lower_dentry); ecryptfs_copy_attr_times(dir, lower_dir_dentry->d_inode); dir->i_nlink = lower_dir_dentry->d_inode->i_nlink; unlock_dir(lower_dir_dentry); if (!rc) d_drop(dentry); -out: - if (tdentry) - dput(tdentry); - if (tlower_dentry) - dput(tlower_dentry); + dput(dentry); return rc; } -- cgit v0.10.2 From 5ddfe9691c91a244e8d1be597b6428fcefd58103 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 30 Oct 2006 22:07:21 -0800 Subject: [PATCH] md: check bio address after mapping through partitions. Partitions are not limited to live within a device. So we should range check after partition mapping. Note that 'maxsector' was being used for two different things. I have split off the second usage into 'old_sector' so that maxsector can be still be used for it's primary usage later in the function. Cc: Jens Axboe Signed-off-by: Neil Brown Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 1360665..c7b1dac 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -2999,6 +2999,7 @@ void generic_make_request(struct bio *bio) { request_queue_t *q; sector_t maxsector; + sector_t old_sector; int ret, nr_sectors = bio_sectors(bio); dev_t old_dev; @@ -3027,7 +3028,7 @@ void generic_make_request(struct bio *bio) * NOTE: we don't repeat the blk_size check for each new device. * Stacking drivers are expected to know what they are doing. */ - maxsector = -1; + old_sector = -1; old_dev = 0; do { char b[BDEVNAME_SIZE]; @@ -3061,15 +3062,30 @@ end_io: */ blk_partition_remap(bio); - if (maxsector != -1) + if (old_sector != -1) blk_add_trace_remap(q, bio, old_dev, bio->bi_sector, - maxsector); + old_sector); blk_add_trace_bio(q, bio, BLK_TA_QUEUE); - maxsector = bio->bi_sector; + old_sector = bio->bi_sector; old_dev = bio->bi_bdev->bd_dev; + maxsector = bio->bi_bdev->bd_inode->i_size >> 9; + if (maxsector) { + sector_t sector = bio->bi_sector; + + if (maxsector < nr_sectors || maxsector - nr_sectors < sector) { + /* + * This may well happen - partitions are not checked + * to make sure they are within the size of the + * whole device. + */ + handle_bad_sector(bio); + goto end_io; + } + } + ret = q->make_request_fn(q, bio); } while (ret); } -- cgit v0.10.2 From 5fccbf61be2a7f32d2002b04afca4c5009612a58 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 31 Oct 2006 14:21:55 +0100 Subject: [PATCH] CFQ: request <-> request merging rr_list fixup In very rare circumstances would we be pruning a merged request and at the same time delete the implicated cfqq from the rr_list, and not readd it when the merged request got added. This could cause io stalls until that process issued io again. Fix it up by putting the rr_list add handling into cfq_add_rq_rb(), identical to how pruning is handled in cfq_del_rq_rb(). This fixes a hang reproducible with fsx-linux. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 25c4e7e..1d9c3c7 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -456,6 +456,9 @@ static void cfq_add_rq_rb(struct request *rq) */ while ((__alias = elv_rb_add(&cfqq->sort_list, rq)) != NULL) cfq_dispatch_insert(cfqd->queue, __alias); + + if (!cfq_cfqq_on_rr(cfqq)) + cfq_add_cfqq_rr(cfqd, cfqq); } static inline void @@ -1652,9 +1655,6 @@ static void cfq_insert_request(request_queue_t *q, struct request *rq) cfq_add_rq_rb(rq); - if (!cfq_cfqq_on_rr(cfqq)) - cfq_add_cfqq_rr(cfqd, cfqq); - list_add_tail(&rq->queuelist, &cfqq->fifo); cfq_rq_enqueued(cfqd, cfqq, rq); -- cgit v0.10.2 From 80fc115d461031dc66bb7f31b8c84868e370fea6 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 30 Oct 2006 20:41:11 -0800 Subject: [PATCH] SCSI: ISCSI build failure SCSI_QLA_ISCSI needs to depend on NET to prevent build (link) failures that are caused by selecting SCSI_ISCSI_ATTRS. Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds diff --git a/drivers/scsi/qla4xxx/Kconfig b/drivers/scsi/qla4xxx/Kconfig index 08a07f0..69cbff3 100644 --- a/drivers/scsi/qla4xxx/Kconfig +++ b/drivers/scsi/qla4xxx/Kconfig @@ -1,6 +1,6 @@ config SCSI_QLA_ISCSI tristate "QLogic ISP4XXX host adapter family support" - depends on PCI && SCSI + depends on PCI && SCSI && NET select SCSI_ISCSI_ATTRS ---help--- This driver supports the QLogic 40xx (ISP4XXX) iSCSI host -- cgit v0.10.2 From 68586b67ab1a2fd618f79e29a06f10ae886f4b46 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 30 Oct 2006 16:31:52 +0200 Subject: IB/mthca: Fix MAD extended header format for MAD_IFC firmware command Several fields in an incoming MAD extended info header were passed into the MAD_IFC firmware command at incorrect offsets (mostly off by 4 bytes). As the result, the HCA will fail to generate traps in which this info is needed (e.g. traps which include the GRH of the incoming packet), in violation of the IB spec. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c index 99a94d7..768df72 100644 --- a/drivers/infiniband/hw/mthca/mthca_cmd.c +++ b/drivers/infiniband/hw/mthca/mthca_cmd.c @@ -1820,11 +1820,11 @@ int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey, #define MAD_IFC_BOX_SIZE 0x400 #define MAD_IFC_MY_QPN_OFFSET 0x100 -#define MAD_IFC_RQPN_OFFSET 0x104 -#define MAD_IFC_SL_OFFSET 0x108 -#define MAD_IFC_G_PATH_OFFSET 0x109 -#define MAD_IFC_RLID_OFFSET 0x10a -#define MAD_IFC_PKEY_OFFSET 0x10e +#define MAD_IFC_RQPN_OFFSET 0x108 +#define MAD_IFC_SL_OFFSET 0x10c +#define MAD_IFC_G_PATH_OFFSET 0x10d +#define MAD_IFC_RLID_OFFSET 0x10e +#define MAD_IFC_PKEY_OFFSET 0x112 #define MAD_IFC_GRH_OFFSET 0x140 inmailbox = mthca_alloc_mailbox(dev, GFP_KERNEL); @@ -1862,7 +1862,7 @@ int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey, val = in_wc->dlid_path_bits | (in_wc->wc_flags & IB_WC_GRH ? 0x80 : 0); - MTHCA_PUT(inbox, val, MAD_IFC_GRH_OFFSET); + MTHCA_PUT(inbox, val, MAD_IFC_G_PATH_OFFSET); MTHCA_PUT(inbox, in_wc->slid, MAD_IFC_RLID_OFFSET); MTHCA_PUT(inbox, in_wc->pkey_index, MAD_IFC_PKEY_OFFSET); @@ -1870,7 +1870,7 @@ int mthca_MAD_IFC(struct mthca_dev *dev, int ignore_mkey, int ignore_bkey, if (in_grh) memcpy(inbox + MAD_IFC_GRH_OFFSET, in_grh, 40); - op_modifier |= 0x10; + op_modifier |= 0x4; in_modifier |= in_wc->slid << 16; } -- cgit v0.10.2 From 0887fa5158fe7da4a46ae526d313ae636440deae Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 31 Oct 2006 11:52:03 +0000 Subject: [MIPS] TX4927: Remove indent error message that somehow ended in the code. Signed-off-by: Ralf Baechle diff --git a/arch/mips/tx4927/common/tx4927_setup.c b/arch/mips/tx4927/common/tx4927_setup.c index 4658b2a..941c441 100644 --- a/arch/mips/tx4927/common/tx4927_setup.c +++ b/arch/mips/tx4927/common/tx4927_setup.c @@ -112,8 +112,6 @@ void print_cp0(char *key, int num, char *name, u32 val) return; } -indent: Standard input:25: Error:Unexpected end of file - void dump_cp0(char *key) { -- cgit v0.10.2 From e8f05de54055e90e6d58e45866f84ecdb0b48acd Mon Sep 17 00:00:00 2001 From: Manish Lachwani Date: Sat, 22 Jan 2005 09:23:38 -0800 Subject: [MIPS] Add missing file for support of backplane on TX4927 based board Signed-off-by: Manish Lachwani Signed-off-by: Ralf Baechle diff --git a/arch/mips/tx4927/common/smsc_fdc37m81x.c b/arch/mips/tx4927/common/smsc_fdc37m81x.c new file mode 100644 index 0000000..33f517b --- /dev/null +++ b/arch/mips/tx4927/common/smsc_fdc37m81x.c @@ -0,0 +1,172 @@ +/* + * Interface for smsc fdc48m81x Super IO chip + * + * Author: MontaVista Software, Inc. source@mvista.com + * + * 2001-2003 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + * + * Copyright 2004 (c) MontaVista Software, Inc. + */ +#include +#include +#include +#include + +#define DEBUG + +/* Common Registers */ +#define SMSC_FDC37M81X_CONFIG_INDEX 0x00 +#define SMSC_FDC37M81X_CONFIG_DATA 0x01 +#define SMSC_FDC37M81X_CONF 0x02 +#define SMSC_FDC37M81X_INDEX 0x03 +#define SMSC_FDC37M81X_DNUM 0x07 +#define SMSC_FDC37M81X_DID 0x20 +#define SMSC_FDC37M81X_DREV 0x21 +#define SMSC_FDC37M81X_PCNT 0x22 +#define SMSC_FDC37M81X_PMGT 0x23 +#define SMSC_FDC37M81X_OSC 0x24 +#define SMSC_FDC37M81X_CONFPA0 0x26 +#define SMSC_FDC37M81X_CONFPA1 0x27 +#define SMSC_FDC37M81X_TEST4 0x2B +#define SMSC_FDC37M81X_TEST5 0x2C +#define SMSC_FDC37M81X_TEST1 0x2D +#define SMSC_FDC37M81X_TEST2 0x2E +#define SMSC_FDC37M81X_TEST3 0x2F + +/* Logical device numbers */ +#define SMSC_FDC37M81X_FDD 0x00 +#define SMSC_FDC37M81X_SERIAL1 0x04 +#define SMSC_FDC37M81X_SERIAL2 0x05 +#define SMSC_FDC37M81X_KBD 0x07 + +/* Logical device Config Registers */ +#define SMSC_FDC37M81X_ACTIVE 0x30 +#define SMSC_FDC37M81X_BASEADDR0 0x60 +#define SMSC_FDC37M81X_BASEADDR1 0x61 +#define SMSC_FDC37M81X_INT 0x70 +#define SMSC_FDC37M81X_INT2 0x72 +#define SMSC_FDC37M81X_MODE 0xF0 + +/* Chip Config Values */ +#define SMSC_FDC37M81X_CONFIG_ENTER 0x55 +#define SMSC_FDC37M81X_CONFIG_EXIT 0xaa +#define SMSC_FDC37M81X_CHIP_ID 0x4d + +static unsigned long g_smsc_fdc37m81x_base = 0; + +static inline unsigned char smsc_fdc37m81x_rd(unsigned char index) +{ + outb(index, g_smsc_fdc37m81x_base + SMSC_FDC37M81X_CONFIG_INDEX); + + return inb(g_smsc_fdc37m81x_base + SMSC_FDC37M81X_CONFIG_DATA); +} + +static inline void smsc_dc37m81x_wr(unsigned char index, unsigned char data) +{ + outb(index, g_smsc_fdc37m81x_base + SMSC_FDC37M81X_CONFIG_INDEX); + outb(data, g_smsc_fdc37m81x_base + SMSC_FDC37M81X_CONFIG_DATA); +} + +void smsc_fdc37m81x_config_beg(void) +{ + if (g_smsc_fdc37m81x_base) { + outb(SMSC_FDC37M81X_CONFIG_ENTER, + g_smsc_fdc37m81x_base + SMSC_FDC37M81X_CONFIG_INDEX); + } +} + +void smsc_fdc37m81x_config_end(void) +{ + if (g_smsc_fdc37m81x_base) + outb(SMSC_FDC37M81X_CONFIG_EXIT, + g_smsc_fdc37m81x_base + SMSC_FDC37M81X_CONFIG_INDEX); +} + +u8 smsc_fdc37m81x_config_get(u8 reg) +{ + u8 val = 0; + + if (g_smsc_fdc37m81x_base) + val = smsc_fdc37m81x_rd(reg); + + return val; +} + +void smsc_fdc37m81x_config_set(u8 reg, u8 val) +{ + if (g_smsc_fdc37m81x_base) + smsc_dc37m81x_wr(reg, val); +} + +unsigned long __init smsc_fdc37m81x_init(unsigned long port) +{ + const int field = sizeof(unsigned long) * 2; + u8 chip_id; + + if (g_smsc_fdc37m81x_base) + printk("smsc_fdc37m81x_init() stepping on old base=0x%0*lx\n", + field, g_smsc_fdc37m81x_base); + + g_smsc_fdc37m81x_base = port; + + smsc_fdc37m81x_config_beg(); + + chip_id = smsc_fdc37m81x_rd(SMSC_FDC37M81X_DID); + if (chip_id == SMSC_FDC37M81X_CHIP_ID) + smsc_fdc37m81x_config_end(); + else { + printk("smsc_fdc37m81x_init() unknow chip id 0x%02x\n", + chip_id); + g_smsc_fdc37m81x_base = 0; + } + + return g_smsc_fdc37m81x_base; +} + +#ifdef DEBUG +void smsc_fdc37m81x_config_dump_one(char *key, u8 dev, u8 reg) +{ + printk("%s: dev=0x%02x reg=0x%02x val=0x%02x\n", key, dev, reg, + smsc_fdc37m81x_rd(reg)); +} + +void smsc_fdc37m81x_config_dump(void) +{ + u8 orig; + char *fname = "smsc_fdc37m81x_config_dump()"; + + smsc_fdc37m81x_config_beg(); + + orig = smsc_fdc37m81x_rd(SMSC_FDC37M81X_DNUM); + + printk("%s: common\n", fname); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_NONE, + SMSC_FDC37M81X_DNUM); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_NONE, + SMSC_FDC37M81X_DID); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_NONE, + SMSC_FDC37M81X_DREV); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_NONE, + SMSC_FDC37M81X_PCNT); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_NONE, + SMSC_FDC37M81X_PMGT); + + printk("%s: keyboard\n", fname); + smsc_dc37m81x_wr(SMSC_FDC37M81X_DNUM, SMSC_FDC37M81X_KBD); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_KBD, + SMSC_FDC37M81X_ACTIVE); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_KBD, + SMSC_FDC37M81X_INT); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_KBD, + SMSC_FDC37M81X_INT2); + smsc_fdc37m81x_config_dump_one(fname, SMSC_FDC37M81X_KBD, + SMSC_FDC37M81X_LDCR_F0); + + smsc_dc37m81x_wr(SMSC_FDC37M81X_DNUM, orig); + + smsc_fdc37m81x_config_end(); +} +#endif -- cgit v0.10.2 From c21e6d65f70d64b359a37545592f39e28074864e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 31 Oct 2006 13:41:59 +0000 Subject: [MIPS] Sort out missuse of __init for prom_getcmdline() Signed-off-by: Ralf Baechle diff --git a/arch/mips/au1000/common/prom.c b/arch/mips/au1000/common/prom.c index b4b010a..6fce60a 100644 --- a/arch/mips/au1000/common/prom.c +++ b/arch/mips/au1000/common/prom.c @@ -47,7 +47,7 @@ extern int prom_argc; extern char **prom_argv, **prom_envp; -char * prom_getcmdline(void) +char * __init_or_module prom_getcmdline(void) { return &(arcs_cmdline[0]); } diff --git a/arch/mips/au1000/common/setup.c b/arch/mips/au1000/common/setup.c index 377ae0d..919172db 100644 --- a/arch/mips/au1000/common/setup.c +++ b/arch/mips/au1000/common/setup.c @@ -43,7 +43,7 @@ #include #include -extern char * __init prom_getcmdline(void); +extern char * prom_getcmdline(void); extern void __init board_setup(void); extern void au1000_restart(char *); extern void au1000_halt(void); diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 4873dc6..7db3c8a 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -102,7 +102,7 @@ static void enable_mac(struct net_device *, int); // externs extern int get_ethernet_addr(char *ethernet_addr); extern void str2eaddr(unsigned char *ea, unsigned char *str); -extern char * __init prom_getcmdline(void); +extern char * prom_getcmdline(void); /* * Theory of operation -- cgit v0.10.2 From 38384c8bd82474bf74ea68e4e44aaa14504deb07 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Tue, 31 Oct 2006 13:38:44 +0900 Subject: [MIPS] Yosemite: fix uninitialized variable in titan_i2c_xfer() Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/pmc-sierra/yosemite/i2c-yosemite.c b/arch/mips/pmc-sierra/yosemite/i2c-yosemite.c index 416da22..85b14c7 100644 --- a/arch/mips/pmc-sierra/yosemite/i2c-yosemite.c +++ b/arch/mips/pmc-sierra/yosemite/i2c-yosemite.c @@ -74,7 +74,7 @@ static int titan_i2c_poll(void) int titan_i2c_xfer(unsigned int slave_addr, titan_i2c_command * cmd, int size, unsigned int *addr) { - int loop = 0, bytes, i; + int loop, bytes = 0, i; unsigned int *write_data, data, *read_data; unsigned long reg_val, val; -- cgit v0.10.2 From 3ab0f40f333007eb31dc1e08f578ec224c7d71c2 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Tue, 31 Oct 2006 13:44:38 +0900 Subject: [MIPS] Fix warning of printk format in mips_srs_init() arch/mips/kernel/traps.c:1115: warning: int format, long unsigned int arg (arg 2) Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index cce8313..9fda1b8 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -1111,7 +1111,7 @@ static struct shadow_registers { static void mips_srs_init(void) { shadow_registers.sr_supported = ((read_c0_srsctl() >> 26) & 0x0f) + 1; - printk(KERN_INFO "%d MIPSR2 register sets available\n", + printk(KERN_INFO "%ld MIPSR2 register sets available\n", shadow_registers.sr_supported); shadow_registers.sr_allocated = 1; /* Set 0 used by kernel */ } -- cgit v0.10.2 From 781b0f8d4f9c90137ea32771346ab49f0e5319b3 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 31 Oct 2006 18:25:10 +0000 Subject: [MIPS] VSMP: Fix initialization ordering bug. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c index 3b5f3b6..06b29fa 100644 --- a/arch/mips/kernel/smp-mt.c +++ b/arch/mips/kernel/smp-mt.c @@ -140,15 +140,88 @@ static struct irqaction irq_call = { .name = "IPI_call" }; +static void __init smp_copy_vpe_config(void) +{ + write_vpe_c0_status( + (read_c0_status() & ~(ST0_IM | ST0_IE | ST0_KSU)) | ST0_CU0); + + /* set config to be the same as vpe0, particularly kseg0 coherency alg */ + write_vpe_c0_config( read_c0_config()); + + /* make sure there are no software interrupts pending */ + write_vpe_c0_cause(0); + + /* Propagate Config7 */ + write_vpe_c0_config7(read_c0_config7()); +} + +static unsigned int __init smp_vpe_init(unsigned int tc, unsigned int mvpconf0, + unsigned int ncpu) +{ + if (tc > ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT)) + return ncpu; + + /* Deactivate all but VPE 0 */ + if (tc != 0) { + unsigned long tmp = read_vpe_c0_vpeconf0(); + + tmp &= ~VPECONF0_VPA; + + /* master VPE */ + tmp |= VPECONF0_MVP; + write_vpe_c0_vpeconf0(tmp); + + /* Record this as available CPU */ + cpu_set(tc, phys_cpu_present_map); + __cpu_number_map[tc] = ++ncpu; + __cpu_logical_map[ncpu] = tc; + } + + /* Disable multi-threading with TC's */ + write_vpe_c0_vpecontrol(read_vpe_c0_vpecontrol() & ~VPECONTROL_TE); + + if (tc != 0) + smp_copy_vpe_config(); + + return ncpu; +} + +static void __init smp_tc_init(unsigned int tc, unsigned int mvpconf0) +{ + unsigned long tmp; + + if (!tc) + return; + + /* bind a TC to each VPE, May as well put all excess TC's + on the last VPE */ + if (tc >= (((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT)+1)) + write_tc_c0_tcbind(read_tc_c0_tcbind() | ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT)); + else { + write_tc_c0_tcbind(read_tc_c0_tcbind() | tc); + + /* and set XTC */ + write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() | (tc << VPECONF0_XTC_SHIFT)); + } + + tmp = read_tc_c0_tcstatus(); + + /* mark not allocated and not dynamically allocatable */ + tmp &= ~(TCSTATUS_A | TCSTATUS_DA); + tmp |= TCSTATUS_IXMT; /* interrupt exempt */ + write_tc_c0_tcstatus(tmp); + + write_tc_c0_tchalt(TCHALT_H); +} + /* * Common setup before any secondaries are started * Make sure all CPU's are in a sensible state before we boot any of the * secondarys */ -void plat_smp_setup(void) +void __init plat_smp_setup(void) { - unsigned long val; - int i, num; + unsigned int mvpconf0, ntc, tc, ncpu = 0; #ifdef CONFIG_MIPS_MT_FPAFF /* If we have an FPU, enroll ourselves in the FPU-full mask */ @@ -167,75 +240,16 @@ void plat_smp_setup(void) /* Put MVPE's into 'configuration state' */ set_c0_mvpcontrol(MVPCONTROL_VPC); - val = read_c0_mvpconf0(); + mvpconf0 = read_c0_mvpconf0(); + ntc = (mvpconf0 & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT; /* we'll always have more TC's than VPE's, so loop setting everything to a sensible state */ - for (i = 0, num = 0; i <= ((val & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT); i++) { - settc(i); - - /* VPE's */ - if (i <= ((val & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT)) { - - /* deactivate all but vpe0 */ - if (i != 0) { - unsigned long tmp = read_vpe_c0_vpeconf0(); - - tmp &= ~VPECONF0_VPA; - - /* master VPE */ - tmp |= VPECONF0_MVP; - write_vpe_c0_vpeconf0(tmp); - - /* Record this as available CPU */ - cpu_set(i, phys_cpu_present_map); - __cpu_number_map[i] = ++num; - __cpu_logical_map[num] = i; - } - - /* disable multi-threading with TC's */ - write_vpe_c0_vpecontrol(read_vpe_c0_vpecontrol() & ~VPECONTROL_TE); - - if (i != 0) { - write_vpe_c0_status((read_c0_status() & ~(ST0_IM | ST0_IE | ST0_KSU)) | ST0_CU0); + for (tc = 0; tc <= ntc; tc++) { + settc(tc); - /* set config to be the same as vpe0, particularly kseg0 coherency alg */ - write_vpe_c0_config( read_c0_config()); - - /* make sure there are no software interrupts pending */ - write_vpe_c0_cause(0); - - /* Propagate Config7 */ - write_vpe_c0_config7(read_c0_config7()); - } - - } - - /* TC's */ - - if (i != 0) { - unsigned long tmp; - - /* bind a TC to each VPE, May as well put all excess TC's - on the last VPE */ - if ( i >= (((val & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT)+1) ) - write_tc_c0_tcbind(read_tc_c0_tcbind() | ((val & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) ); - else { - write_tc_c0_tcbind( read_tc_c0_tcbind() | i); - - /* and set XTC */ - write_vpe_c0_vpeconf0( read_vpe_c0_vpeconf0() | (i << VPECONF0_XTC_SHIFT)); - } - - tmp = read_tc_c0_tcstatus(); - - /* mark not allocated and not dynamically allocatable */ - tmp &= ~(TCSTATUS_A | TCSTATUS_DA); - tmp |= TCSTATUS_IXMT; /* interrupt exempt */ - write_tc_c0_tcstatus(tmp); - - write_tc_c0_tchalt(TCHALT_H); - } + smp_tc_init(tc, mvpconf0); + ncpu = smp_vpe_init(tc, mvpconf0, ncpu); } /* Release config state */ @@ -243,7 +257,7 @@ void plat_smp_setup(void) /* We'll wait until starting the secondaries before starting MVPE */ - printk(KERN_INFO "Detected %i available secondary CPU(s)\n", num); + printk(KERN_INFO "Detected %i available secondary CPU(s)\n", ncpu); } void __init plat_prepare_cpus(unsigned int max_cpus) -- cgit v0.10.2 From e79f55a8c7aaae5a33e8c2b29682ec8e603b5434 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 31 Oct 2006 19:53:15 +0000 Subject: [MIPS] Flags must be unsigned long. Signed-off-by: Ralf Baechle diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c index c079e2a..6597067 100644 --- a/arch/mips/mips-boards/generic/time.c +++ b/arch/mips/mips-boards/generic/time.c @@ -208,7 +208,7 @@ static unsigned int __init estimate_cpu_frequency(void) count = 6000000; #endif #if defined(CONFIG_MIPS_ATLAS) || defined(CONFIG_MIPS_MALTA) - unsigned int flags; + unsigned long flags; local_irq_save(flags); -- cgit v0.10.2 From 70e46f48cb5933119712ee27945309a4bfc98282 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 31 Oct 2006 18:33:09 +0000 Subject: [MIPS] VSMP: Synchronize cp0 counters on bootup. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c index 06b29fa..2ac19a6c 100644 --- a/arch/mips/kernel/smp-mt.c +++ b/arch/mips/kernel/smp-mt.c @@ -153,6 +153,8 @@ static void __init smp_copy_vpe_config(void) /* Propagate Config7 */ write_vpe_c0_config7(read_c0_config7()); + + write_vpe_c0_count(read_c0_count()); } static unsigned int __init smp_vpe_init(unsigned int tc, unsigned int mvpconf0, diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c index 6597067..d817c60 100644 --- a/arch/mips/mips-boards/generic/time.c +++ b/arch/mips/mips-boards/generic/time.c @@ -209,6 +209,7 @@ static unsigned int __init estimate_cpu_frequency(void) #endif #if defined(CONFIG_MIPS_ATLAS) || defined(CONFIG_MIPS_MALTA) unsigned long flags; + unsigned int start; local_irq_save(flags); @@ -217,13 +218,13 @@ static unsigned int __init estimate_cpu_frequency(void) while (!(CMOS_READ(RTC_REG_A) & RTC_UIP)); /* Start r4k counter. */ - write_c0_count(0); + start = read_c0_count(); /* Read counter exactly on falling edge of update flag */ while (CMOS_READ(RTC_REG_A) & RTC_UIP); while (!(CMOS_READ(RTC_REG_A) & RTC_UIP)); - count = read_c0_count(); + count = read_c0_count() - start; /* restore interrupts */ local_irq_restore(flags); diff --git a/include/asm-mips/mipsmtregs.h b/include/asm-mips/mipsmtregs.h index f637ce7..3e9468f 100644 --- a/include/asm-mips/mipsmtregs.h +++ b/include/asm-mips/mipsmtregs.h @@ -352,6 +352,8 @@ do { \ #define write_vpe_c0_vpecontrol(val) mttc0(1, 1, val) #define read_vpe_c0_vpeconf0() mftc0(1, 2) #define write_vpe_c0_vpeconf0(val) mttc0(1, 2, val) +#define read_vpe_c0_count() mftc0(9, 0) +#define write_vpe_c0_count(val) mttc0(9, 0, val) #define read_vpe_c0_status() mftc0(12, 0) #define write_vpe_c0_status(val) mttc0(12, 0, val) #define read_vpe_c0_cause() mftc0(13, 0) -- cgit v0.10.2 From 16b7b2ac0148e839da86af8747b6fa4aad43a9b7 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Tue, 24 Oct 2006 00:21:27 +0900 Subject: [MIPS] Fixup migration to GENERIC_TIME Since we already moved to GENERIC_TIME, we should implement alternatives of old do_gettimeoffset routines to get sub-jiffies resolution from gettimeofday(). This patch includes: * MIPS clocksource support (based on works by Manish Lachwani). * remove unused gettimeoffset routines and related codes. * remove unised 64bit do_div64_32(). * simplify mips_hpt_init. (no argument needed, __init tag) * simplify c0_hpt_timer_init. (no need to write to c0_count) * remove some hpt_init routines. * mips_hpt_mask variable to specify bitmask of hpt value. * convert jmr3927_do_gettimeoffset to jmr3927_hpt_read. * convert ip27_do_gettimeoffset to ip27_hpt_read. * convert bcm1480_do_gettimeoffset to bcm1480_hpt_read. * simplify sb1250 hpt functions. (no need to subtract and shift) Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle diff --git a/Documentation/mips/time.README b/Documentation/mips/time.README index e1304b6..a4ce603 100644 --- a/Documentation/mips/time.README +++ b/Documentation/mips/time.README @@ -38,19 +38,14 @@ The new time code provide the following services: a) Implements functions required by Linux common code: time_init - do_gettimeofday - do_settimeofday b) provides an abstraction of RTC and null RTC implementation as default. extern unsigned long (*rtc_get_time)(void); extern int (*rtc_set_time)(unsigned long); - c) a set of gettimeoffset functions for different CPUs and different - needs. - - d) high-level and low-level timer interrupt routines where the timer - interrupt source may or may not be the CPU timer. The high-level - routine is dispatched through do_IRQ() while the low-level is + c) high-level and low-level timer interrupt routines where the timer + interrupt source may or may not be the CPU timer. The high-level + routine is dispatched through do_IRQ() while the low-level is dispatched in assemably code (usually int-handler.S) @@ -73,8 +68,7 @@ the following functions or values: c) (optional) board-specific RTC routines. d) (optional) mips_hpt_frequency - It must be definied if the board - is using CPU counter for timer interrupt or it is using fixed rate - gettimeoffset(). + is using CPU counter for timer interrupt. PORTING GUIDE @@ -89,16 +83,6 @@ Step 1: decide how you like to implement the time services. If the answer is no, you need a timer to provide the timer interrupt at 100 HZ speed. - You cannot use the fast gettimeoffset functions, i.e., - - unsigned long fixed_rate_gettimeoffset(void); - unsigned long calibrate_div32_gettimeoffset(void); - unsigned long calibrate_div64_gettimeoffset(void); - - You can use null_gettimeoffset() will gives the same time resolution as - jiffy. Or you can implement your own gettimeoffset (probably based on - some ad hoc hardware on your machine.) - c) The following sub steps assume your CPU has counter register. Do you plan to use the CPU counter register as the timer interrupt or use an exnternal timer? @@ -123,8 +107,8 @@ Step 3: implement rtc routines, board_time_init() and plat_timer_setup() board_time_init() - a) (optional) set up RTC routines, b) (optional) calibrate and set the mips_hpt_frequency - (only needed if you intended to use fixed_rate_gettimeoffset - or use cpu counter as timer interrupt source) + (only needed if you intended to use cpu counter as timer interrupt + source) plat_timer_setup() - a) (optional) over-write any choices made above by time_init(). @@ -154,8 +138,8 @@ for some of the functions in time.c. For example, you may define your own timer interrupt routine, which does some of its own processing and then calls timer_interrupt(). -You can also over-ride any of the built-in functions (gettimeoffset, -RTC routines and/or timer interrupt routine). +You can also over-ride any of the built-in functions (RTC routines +and/or timer interrupt routine). PORTING NOTES FOR SMP @@ -187,10 +171,3 @@ You need to decide on your timer interrupt sources. You can also do the low-level version of those interrupt routines, following similar dispatching routes described above. - -Note about do_gettimeoffset(): - - It is very likely the CPU counter registers are not sync'ed up in a SMP box. - Therefore you cannot really use the many of the existing routines that - are based on CPU counter. You should wirte your own gettimeoffset rouinte - if you want intra-jiffy resolution. diff --git a/arch/mips/au1000/common/time.c b/arch/mips/au1000/common/time.c index 6768638..fa1c62f 100644 --- a/arch/mips/au1000/common/time.c +++ b/arch/mips/au1000/common/time.c @@ -53,9 +53,6 @@ static unsigned long r4k_cur; /* What counter should be at next timer irq */ int no_au1xxx_32khz; extern int allow_au1k_wait; /* default off for CP0 Counter */ -/* Cycle counter value at the previous timer interrupt.. */ -static unsigned int timerhi = 0, timerlo = 0; - #ifdef CONFIG_PM #if HZ < 100 || HZ > 1000 #error "unsupported HZ value! Must be in [100,1000]" @@ -90,10 +87,6 @@ void mips_timer_interrupt(void) goto null; do { - count = read_c0_count(); - timerhi += (count < timerlo); /* Wrap around */ - timerlo = count; - kstat_this_cpu.irqs[irq]++; do_timer(1); #ifndef CONFIG_SMP @@ -297,88 +290,6 @@ unsigned long cal_r4koff(void) return (cpu_speed / HZ); } -/* This is for machines which generate the exact clock. */ -#define USECS_PER_JIFFY (1000000/HZ) -#define USECS_PER_JIFFY_FRAC (0x100000000LL*1000000/HZ&0xffffffff) - -static unsigned long -div64_32(unsigned long v1, unsigned long v2, unsigned long v3) -{ - unsigned long r0; - do_div64_32(r0, v1, v2, v3); - return r0; -} - -static unsigned long do_fast_cp0_gettimeoffset(void) -{ - u32 count; - unsigned long res, tmp; - unsigned long r0; - - /* Last jiffy when do_fast_gettimeoffset() was called. */ - static unsigned long last_jiffies=0; - unsigned long quotient; - - /* - * Cached "1/(clocks per usec)*2^32" value. - * It has to be recalculated once each jiffy. - */ - static unsigned long cached_quotient=0; - - tmp = jiffies; - - quotient = cached_quotient; - - if (tmp && last_jiffies != tmp) { - last_jiffies = tmp; - if (last_jiffies != 0) { - r0 = div64_32(timerhi, timerlo, tmp); - quotient = div64_32(USECS_PER_JIFFY, USECS_PER_JIFFY_FRAC, r0); - cached_quotient = quotient; - } - } - - /* Get last timer tick in absolute kernel time */ - count = read_c0_count(); - - /* .. relative to previous jiffy (32 bits is enough) */ - count -= timerlo; - - __asm__("multu\t%1,%2\n\t" - "mfhi\t%0" - : "=r" (res) - : "r" (count), "r" (quotient) - : "hi", "lo", GCC_REG_ACCUM); - - /* - * Due to possible jiffies inconsistencies, we need to check - * the result so that we'll get a timer that is monotonic. - */ - if (res >= USECS_PER_JIFFY) - res = USECS_PER_JIFFY-1; - - return res; -} - -#ifdef CONFIG_PM -static unsigned long do_fast_pm_gettimeoffset(void) -{ - unsigned long pc0; - unsigned long offset; - - pc0 = au_readl(SYS_TOYREAD); - au_sync(); - offset = pc0 - last_pc0; - if (offset > 2*MATCH20_INC) { - printk("huge offset %x, last_pc0 %x last_match20 %x pc0 %x\n", - (unsigned)offset, (unsigned)last_pc0, - (unsigned)last_match20, (unsigned)pc0); - } - offset = (unsigned long)((offset * 305) / 10); - return offset; -} -#endif - void __init plat_timer_setup(struct irqaction *irq) { unsigned int est_freq; @@ -416,7 +327,6 @@ void __init plat_timer_setup(struct irqaction *irq) unsigned int c0_status; printk("WARNING: no 32KHz clock found.\n"); - do_gettimeoffset = do_fast_cp0_gettimeoffset; /* Ensure we get CPO_COUNTER interrupts. */ @@ -441,19 +351,11 @@ void __init plat_timer_setup(struct irqaction *irq) while (au_readl(SYS_COUNTER_CNTRL) & SYS_CNTRL_M20); startup_match20_interrupt(counter0_irq); - do_gettimeoffset = do_fast_pm_gettimeoffset; - /* We can use the real 'wait' instruction. */ allow_au1k_wait = 1; } -#else - /* We have to do this here instead of in timer_init because - * the generic code in arch/mips/kernel/time.c will write - * over our function pointer. - */ - do_gettimeoffset = do_fast_cp0_gettimeoffset; #endif } diff --git a/arch/mips/dec/time.c b/arch/mips/dec/time.c index 4cf0c06..69e424e 100644 --- a/arch/mips/dec/time.c +++ b/arch/mips/dec/time.c @@ -160,11 +160,6 @@ static unsigned int dec_ioasic_hpt_read(void) return ioasic_read(IO_REG_FCTR); } -static void dec_ioasic_hpt_init(unsigned int count) -{ - ioasic_write(IO_REG_FCTR, ioasic_read(IO_REG_FCTR) - count); -} - void __init dec_time_init(void) { @@ -174,11 +169,9 @@ void __init dec_time_init(void) mips_timer_state = dec_timer_state; mips_timer_ack = dec_timer_ack; - if (!cpu_has_counter && IOASIC) { + if (!cpu_has_counter && IOASIC) /* For pre-R4k systems we use the I/O ASIC's counter. */ mips_hpt_read = dec_ioasic_hpt_read; - mips_hpt_init = dec_ioasic_hpt_init; - } /* Set up the rate of periodic DS1287 interrupts. */ CMOS_WRITE(RTC_REF_CLCK_32KHZ | (16 - __ffs(HZ)), RTC_REG_A); diff --git a/arch/mips/jmr3927/rbhma3100/setup.c b/arch/mips/jmr3927/rbhma3100/setup.c index 0254340..16e5dfe 100644 --- a/arch/mips/jmr3927/rbhma3100/setup.c +++ b/arch/mips/jmr3927/rbhma3100/setup.c @@ -170,12 +170,20 @@ static void jmr3927_machine_power_off(void) while (1); } +static unsigned int jmr3927_hpt_read(void) +{ + /* We assume this function is called xtime_lock held. */ + return jiffies * (JMR3927_TIMER_CLK / HZ) + jmr3927_tmrptr->trr; +} + #define USE_RTC_DS1742 #ifdef USE_RTC_DS1742 extern void rtc_ds1742_init(unsigned long base); #endif static void __init jmr3927_time_init(void) { + mips_hpt_read = jmr3927_hpt_read; + mips_hpt_frequency = JMR3927_TIMER_CLK; #ifdef USE_RTC_DS1742 if (jmr3927_have_nvram()) { rtc_ds1742_init(JMR3927_IOC_NVRAMB_ADDR); @@ -183,12 +191,8 @@ static void __init jmr3927_time_init(void) #endif } -unsigned long jmr3927_do_gettimeoffset(void); - void __init plat_timer_setup(struct irqaction *irq) { - do_gettimeoffset = jmr3927_do_gettimeoffset; - jmr3927_tmrptr->cpra = JMR3927_TIMER_CLK / HZ; jmr3927_tmrptr->itmr = TXx927_TMTITMR_TIIE | TXx927_TMTITMR_TZCE; jmr3927_tmrptr->ccdr = JMR3927_TIMER_CCD; @@ -200,34 +204,6 @@ void __init plat_timer_setup(struct irqaction *irq) #define USECS_PER_JIFFY (1000000/HZ) -unsigned long jmr3927_do_gettimeoffset(void) -{ - unsigned long count; - unsigned long res = 0; - - /* MUST read TRR before TISR. */ - count = jmr3927_tmrptr->trr; - - if (jmr3927_tmrptr->tisr & TXx927_TMTISR_TIIS) { - /* timer interrupt is pending. use Max value. */ - res = USECS_PER_JIFFY - 1; - } else { - /* convert to usec */ - /* res = count / (JMR3927_TIMER_CLK / 1000000); */ - res = (count << 7) / ((JMR3927_TIMER_CLK << 7) / 1000000); - - /* - * Due to possible jiffies inconsistencies, we need to check - * the result so that we'll get a timer that is monotonic. - */ - if (res >= USECS_PER_JIFFY) - res = USECS_PER_JIFFY-1; - } - - return res; -} - - //#undef DO_WRITE_THROUGH #define DO_WRITE_THROUGH #define DO_ENABLE_CACHE diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c index debe86c..e535f86 100644 --- a/arch/mips/kernel/time.c +++ b/arch/mips/kernel/time.c @@ -11,6 +11,7 @@ * Free Software Foundation; either version 2 of the License, or (at your * option) any later version. */ +#include #include #include #include @@ -67,15 +68,9 @@ int (*rtc_mips_set_time)(unsigned long) = null_rtc_set_time; int (*rtc_mips_set_mmss)(unsigned long); -/* usecs per counter cycle, shifted to left by 32 bits */ -static unsigned int sll32_usecs_per_cycle; - /* how many counter cycles in a jiffy */ static unsigned long cycles_per_jiffy __read_mostly; -/* Cycle counter value at the previous timer interrupt.. */ -static unsigned int timerhi, timerlo; - /* expirelo is the count value for next CPU timer interrupt */ static unsigned int expirelo; @@ -93,7 +88,7 @@ static unsigned int null_hpt_read(void) return 0; } -static void null_hpt_init(unsigned int count) +static void __init null_hpt_init(void) { /* nothing */ } @@ -128,186 +123,18 @@ static unsigned int c0_hpt_read(void) return read_c0_count(); } -/* For use solely as a high precision timer. */ -static void c0_hpt_init(unsigned int count) -{ - write_c0_count(read_c0_count() - count); -} - /* For use both as a high precision timer and an interrupt source. */ -static void c0_hpt_timer_init(unsigned int count) +static void __init c0_hpt_timer_init(void) { - count = read_c0_count() - count; - expirelo = (count / cycles_per_jiffy + 1) * cycles_per_jiffy; - write_c0_count(expirelo - cycles_per_jiffy); + expirelo = read_c0_count() + cycles_per_jiffy; write_c0_compare(expirelo); - write_c0_count(count); } int (*mips_timer_state)(void); void (*mips_timer_ack)(void); unsigned int (*mips_hpt_read)(void); -void (*mips_hpt_init)(unsigned int); - -/* - * Gettimeoffset routines. These routines returns the time duration - * since last timer interrupt in usecs. - * - * If the exact CPU counter frequency is known, use fixed_rate_gettimeoffset. - * Otherwise use calibrate_gettimeoffset() - * - * If the CPU does not have the counter register, you can either supply - * your own gettimeoffset() routine, or use null_gettimeoffset(), which - * gives the same resolution as HZ. - */ - -static unsigned long null_gettimeoffset(void) -{ - return 0; -} - - -/* The function pointer to one of the gettimeoffset funcs. */ -unsigned long (*do_gettimeoffset)(void) = null_gettimeoffset; - - -static unsigned long fixed_rate_gettimeoffset(void) -{ - u32 count; - unsigned long res; - - /* Get last timer tick in absolute kernel time */ - count = mips_hpt_read(); - - /* .. relative to previous jiffy (32 bits is enough) */ - count -= timerlo; - - __asm__("multu %1,%2" - : "=h" (res) - : "r" (count), "r" (sll32_usecs_per_cycle) - : "lo", GCC_REG_ACCUM); - - /* - * Due to possible jiffies inconsistencies, we need to check - * the result so that we'll get a timer that is monotonic. - */ - if (res >= USECS_PER_JIFFY) - res = USECS_PER_JIFFY - 1; - - return res; -} - - -/* - * Cached "1/(clocks per usec) * 2^32" value. - * It has to be recalculated once each jiffy. - */ -static unsigned long cached_quotient; - -/* Last jiffy when calibrate_divXX_gettimeoffset() was called. */ -static unsigned long last_jiffies; - -/* - * This is moved from dec/time.c:do_ioasic_gettimeoffset() by Maciej. - */ -static unsigned long calibrate_div32_gettimeoffset(void) -{ - u32 count; - unsigned long res, tmp; - unsigned long quotient; - - tmp = jiffies; - - quotient = cached_quotient; - - if (last_jiffies != tmp) { - last_jiffies = tmp; - if (last_jiffies != 0) { - unsigned long r0; - do_div64_32(r0, timerhi, timerlo, tmp); - do_div64_32(quotient, USECS_PER_JIFFY, - USECS_PER_JIFFY_FRAC, r0); - cached_quotient = quotient; - } - } - - /* Get last timer tick in absolute kernel time */ - count = mips_hpt_read(); - - /* .. relative to previous jiffy (32 bits is enough) */ - count -= timerlo; - - __asm__("multu %1,%2" - : "=h" (res) - : "r" (count), "r" (quotient) - : "lo", GCC_REG_ACCUM); - - /* - * Due to possible jiffies inconsistencies, we need to check - * the result so that we'll get a timer that is monotonic. - */ - if (res >= USECS_PER_JIFFY) - res = USECS_PER_JIFFY - 1; - - return res; -} - -static unsigned long calibrate_div64_gettimeoffset(void) -{ - u32 count; - unsigned long res, tmp; - unsigned long quotient; - - tmp = jiffies; - - quotient = cached_quotient; - - if (last_jiffies != tmp) { - last_jiffies = tmp; - if (last_jiffies) { - unsigned long r0; - __asm__(".set push\n\t" - ".set mips3\n\t" - "lwu %0,%3\n\t" - "dsll32 %1,%2,0\n\t" - "or %1,%1,%0\n\t" - "ddivu $0,%1,%4\n\t" - "mflo %1\n\t" - "dsll32 %0,%5,0\n\t" - "or %0,%0,%6\n\t" - "ddivu $0,%0,%1\n\t" - "mflo %0\n\t" - ".set pop" - : "=&r" (quotient), "=&r" (r0) - : "r" (timerhi), "m" (timerlo), - "r" (tmp), "r" (USECS_PER_JIFFY), - "r" (USECS_PER_JIFFY_FRAC) - : "hi", "lo", GCC_REG_ACCUM); - cached_quotient = quotient; - } - } - - /* Get last timer tick in absolute kernel time */ - count = mips_hpt_read(); - - /* .. relative to previous jiffy (32 bits is enough) */ - count -= timerlo; - - __asm__("multu %1,%2" - : "=h" (res) - : "r" (count), "r" (quotient) - : "lo", GCC_REG_ACCUM); - - /* - * Due to possible jiffies inconsistencies, we need to check - * the result so that we'll get a timer that is monotonic. - */ - if (res >= USECS_PER_JIFFY) - res = USECS_PER_JIFFY - 1; - - return res; -} - +void (*mips_hpt_init)(void) __initdata = null_hpt_init; +unsigned int mips_hpt_mask = 0xffffffff; /* last time when xtime and rtc are sync'ed up */ static long last_rtc_update; @@ -334,18 +161,10 @@ void local_timer_interrupt(int irq, void *dev_id) */ irqreturn_t timer_interrupt(int irq, void *dev_id) { - unsigned long j; - unsigned int count; - write_seqlock(&xtime_lock); - count = mips_hpt_read(); mips_timer_ack(); - /* Update timerhi/timerlo for intra-jiffy calibration. */ - timerhi += count < timerlo; /* Wrap around */ - timerlo = count; - /* * call the generic timer interrupt handling */ @@ -368,47 +187,6 @@ irqreturn_t timer_interrupt(int irq, void *dev_id) } } - /* - * If jiffies has overflown in this timer_interrupt, we must - * update the timer[hi]/[lo] to make fast gettimeoffset funcs - * quotient calc still valid. -arca - * - * The first timer interrupt comes late as interrupts are - * enabled long after timers are initialized. Therefore the - * high precision timer is fast, leading to wrong gettimeoffset() - * calculations. We deal with it by setting it based on the - * number of its ticks between the second and the third interrupt. - * That is still somewhat imprecise, but it's a good estimate. - * --macro - */ - j = jiffies; - if (j < 4) { - static unsigned int prev_count; - static int hpt_initialized; - - switch (j) { - case 0: - timerhi = timerlo = 0; - mips_hpt_init(count); - break; - case 2: - prev_count = count; - break; - case 3: - if (!hpt_initialized) { - unsigned int c3 = 3 * (count - prev_count); - - timerhi = 0; - timerlo = c3; - mips_hpt_init(count - c3); - hpt_initialized = 1; - } - break; - default: - break; - } - } - write_sequnlock(&xtime_lock); /* @@ -476,12 +254,11 @@ asmlinkage void ll_local_timer_interrupt(int irq) * 1) board_time_init() - * a) (optional) set up RTC routines, * b) (optional) calibrate and set the mips_hpt_frequency - * (only needed if you intended to use fixed_rate_gettimeoffset - * or use cpu counter as timer interrupt source) + * (only needed if you intended to use cpu counter as timer interrupt + * source) * 2) setup xtime based on rtc_mips_get_time(). - * 3) choose a appropriate gettimeoffset routine. - * 4) calculate a couple of cached variables for later usage - * 5) plat_timer_setup() - + * 3) calculate a couple of cached variables for later usage + * 4) plat_timer_setup() - * a) (optional) over-write any choices made above by time_init(). * b) machine specific code should setup the timer irqaction. * c) enable the timer interrupt @@ -533,13 +310,48 @@ static unsigned int __init calibrate_hpt(void) } while (--i); hpt_end = mips_hpt_read(); - hpt_count = hpt_end - hpt_start; + hpt_count = (hpt_end - hpt_start) & mips_hpt_mask; hz = HZ; frequency = (u64)hpt_count * (u64)hz; return frequency >> log_2_loops; } +static cycle_t read_mips_hpt(void) +{ + return (cycle_t)mips_hpt_read(); +} + +static struct clocksource clocksource_mips = { + .name = "MIPS", + .read = read_mips_hpt, + .is_continuous = 1, +}; + +static void __init init_mips_clocksource(void) +{ + u64 temp; + u32 shift; + + if (!mips_hpt_frequency || mips_hpt_read == null_hpt_read) + return; + + /* Calclate a somewhat reasonable rating value */ + clocksource_mips.rating = 200 + mips_hpt_frequency / 10000000; + /* Find a shift value */ + for (shift = 32; shift > 0; shift--) { + temp = (u64) NSEC_PER_SEC << shift; + do_div(temp, mips_hpt_frequency); + if ((temp >> 32) == 0) + break; + } + clocksource_mips.shift = shift; + clocksource_mips.mult = (u32)temp; + clocksource_mips.mask = mips_hpt_mask; + + clocksource_register(&clocksource_mips); +} + void __init time_init(void) { if (board_time_init) @@ -555,41 +367,21 @@ void __init time_init(void) -xtime.tv_sec, -xtime.tv_nsec); /* Choose appropriate high precision timer routines. */ - if (!cpu_has_counter && !mips_hpt_read) { + if (!cpu_has_counter && !mips_hpt_read) /* No high precision timer -- sorry. */ mips_hpt_read = null_hpt_read; - mips_hpt_init = null_hpt_init; - } else if (!mips_hpt_frequency && !mips_timer_state) { + else if (!mips_hpt_frequency && !mips_timer_state) { /* A high precision timer of unknown frequency. */ - if (!mips_hpt_read) { + if (!mips_hpt_read) /* No external high precision timer -- use R4k. */ mips_hpt_read = c0_hpt_read; - mips_hpt_init = c0_hpt_init; - } - - if (cpu_has_mips32r1 || cpu_has_mips32r2 || - (current_cpu_data.isa_level == MIPS_CPU_ISA_I) || - (current_cpu_data.isa_level == MIPS_CPU_ISA_II)) - /* - * We need to calibrate the counter but we don't have - * 64-bit division. - */ - do_gettimeoffset = calibrate_div32_gettimeoffset; - else - /* - * We need to calibrate the counter but we *do* have - * 64-bit division. - */ - do_gettimeoffset = calibrate_div64_gettimeoffset; } else { /* We know counter frequency. Or we can get it. */ if (!mips_hpt_read) { /* No external high precision timer -- use R4k. */ mips_hpt_read = c0_hpt_read; - if (mips_timer_state) - mips_hpt_init = c0_hpt_init; - else { + if (!mips_timer_state) { /* No external timer interrupt -- use R4k. */ mips_hpt_init = c0_hpt_timer_init; mips_timer_ack = c0_timer_ack; @@ -598,16 +390,9 @@ void __init time_init(void) if (!mips_hpt_frequency) mips_hpt_frequency = calibrate_hpt(); - do_gettimeoffset = fixed_rate_gettimeoffset; - /* Calculate cache parameters. */ cycles_per_jiffy = (mips_hpt_frequency + HZ / 2) / HZ; - /* sll32_usecs_per_cycle = 10^6 * 2^32 / mips_counter_freq */ - do_div64_32(sll32_usecs_per_cycle, - 1000000, mips_hpt_frequency / 2, - mips_hpt_frequency); - /* Report the high precision timer rate for a reference. */ printk("Using %u.%03u MHz high precision timer.\n", ((mips_hpt_frequency + 500) / 1000) / 1000, @@ -619,7 +404,7 @@ void __init time_init(void) mips_timer_ack = null_timer_ack; /* This sets up the high precision timer for the first interrupt. */ - mips_hpt_init(mips_hpt_read()); + mips_hpt_init(); /* * Call board specific timer interrupt setup. @@ -633,6 +418,8 @@ void __init time_init(void) * is not invoked accidentally. */ plat_timer_setup(&timer_irqaction); + + init_mips_clocksource(); } #define FEBRUARY 2 diff --git a/arch/mips/philips/pnx8550/common/time.c b/arch/mips/philips/pnx8550/common/time.c index 0af655b..65c440e 100644 --- a/arch/mips/philips/pnx8550/common/time.c +++ b/arch/mips/philips/pnx8550/common/time.c @@ -41,8 +41,8 @@ extern unsigned int mips_hpt_frequency; * 1) board_time_init() - * a) (optional) set up RTC routines, * b) (optional) calibrate and set the mips_hpt_frequency - * (only needed if you intended to use fixed_rate_gettimeoffset - * or use cpu counter as timer interrupt source) + * (only needed if you intended to use cpu counter as timer interrupt + * source) */ void pnx8550_time_init(void) diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c index 65fa3a2..3cc0436 100644 --- a/arch/mips/pmc-sierra/yosemite/smp.c +++ b/arch/mips/pmc-sierra/yosemite/smp.c @@ -3,9 +3,7 @@ #include #include - -extern unsigned int (*mips_hpt_read)(void); -extern void (*mips_hpt_init)(unsigned int); +#include #define LAUNCHSTACK_SIZE 256 @@ -101,7 +99,7 @@ void prom_cpus_done(void) */ void prom_init_secondary(void) { - mips_hpt_init(mips_hpt_read()); + mips_hpt_init(); set_c0_status(ST0_CO | ST0_IE | ST0_IM); } diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c index 4e870fc..c965705 100644 --- a/arch/mips/sgi-ip27/ip27-timer.c +++ b/arch/mips/sgi-ip27/ip27-timer.c @@ -134,13 +134,6 @@ again: irq_exit(); } -unsigned long ip27_do_gettimeoffset(void) -{ - unsigned long ct_cur1; - ct_cur1 = REMOTE_HUB_L(cputonasid(0), PI_RT_COUNT) + CYCLES_PER_JIFFY; - return (ct_cur1 - ct_cur[0]) * NSEC_PER_CYCLE / 1000; -} - /* Includes for ioc3_init(). */ #include #include @@ -248,12 +241,17 @@ void __init plat_timer_setup(struct irqaction *irq) setup_irq(irqno, &rt_irqaction); } +static unsigned int ip27_hpt_read(void) +{ + return REMOTE_HUB_L(cputonasid(0), PI_RT_COUNT); +} + void __init ip27_time_init(void) { + mips_hpt_read = ip27_hpt_read; + mips_hpt_frequency = CYCLES_PER_SEC; xtime.tv_sec = get_m48t35_time(); xtime.tv_nsec = 0; - - do_gettimeoffset = ip27_do_gettimeoffset; } void __init cpu_time_init(void) diff --git a/arch/mips/sibyte/bcm1480/time.c b/arch/mips/sibyte/bcm1480/time.c index bf12af4..e136bde 100644 --- a/arch/mips/sibyte/bcm1480/time.c +++ b/arch/mips/sibyte/bcm1480/time.c @@ -47,6 +47,12 @@ #define IMR_IP3_VAL K_BCM1480_INT_MAP_I1 #define IMR_IP4_VAL K_BCM1480_INT_MAP_I2 +#ifdef CONFIG_SIMULATION +#define BCM1480_HPT_VALUE 50000 +#else +#define BCM1480_HPT_VALUE 1000000 +#endif + extern int bcm1480_steal_irq(int irq); void bcm1480_time_init(void) @@ -59,11 +65,6 @@ void bcm1480_time_init(void) BUG(); } - if (!cpu) { - /* Use our own gettimeoffset() routine */ - do_gettimeoffset = bcm1480_gettimeoffset; - } - bcm1480_mask_irq(cpu, irq); /* Map the timer interrupt to ip[4] of this cpu */ @@ -74,11 +75,7 @@ void bcm1480_time_init(void) /* Disable the timer and set up the count */ __raw_writeq(0, IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG))); __raw_writeq( -#ifndef CONFIG_SIMULATION - 1000000/HZ -#else - 50000/HZ -#endif + BCM1480_HPT_VALUE/HZ , IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_INIT))); /* Set the timer running */ @@ -122,16 +119,16 @@ void bcm1480_timer_interrupt(void) } } -/* - * We use our own do_gettimeoffset() instead of the generic one, - * because the generic one does not work for SMP case. - * In addition, since we use general timer 0 for system time, - * we can get accurate intra-jiffy offset without calibration. - */ -unsigned long bcm1480_gettimeoffset(void) +static unsigned int bcm1480_hpt_read(void) { + /* We assume this function is called xtime_lock held. */ unsigned long count = __raw_readq(IOADDR(A_SCD_TIMER_REGISTER(0, R_SCD_TIMER_CNT))); + return (jiffies + 1) * (BCM1480_HPT_VALUE / HZ) - count; +} - return 1000000/HZ - count; +void __init bcm1480_hpt_setup(void) +{ + mips_hpt_read = bcm1480_hpt_read; + mips_hpt_frequency = BCM1480_HPT_VALUE; } diff --git a/arch/mips/sibyte/sb1250/time.c b/arch/mips/sibyte/sb1250/time.c index 0ccf179..bcb74f2 100644 --- a/arch/mips/sibyte/sb1250/time.c +++ b/arch/mips/sibyte/sb1250/time.c @@ -47,15 +47,11 @@ #define SB1250_HPT_NUM 3 #define SB1250_HPT_VALUE M_SCD_TIMER_CNT /* max value */ -#define SB1250_HPT_SHIFT ((sizeof(unsigned int)*8)-V_SCD_TIMER_WIDTH) extern int sb1250_steal_irq(int irq); static unsigned int sb1250_hpt_read(void); -static void sb1250_hpt_init(unsigned int); - -static unsigned int hpt_offset; void __init sb1250_hpt_setup(void) { @@ -69,13 +65,9 @@ void __init sb1250_hpt_setup(void) __raw_writeq(M_SCD_TIMER_ENABLE | M_SCD_TIMER_MODE_CONTINUOUS, IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM, R_SCD_TIMER_CFG))); - /* - * we need to fill 32 bits, so just use the upper 23 bits and pretend - * the timer is going 512Mhz instead of 1Mhz - */ - mips_hpt_frequency = V_SCD_TIMER_FREQ << SB1250_HPT_SHIFT; - mips_hpt_init = sb1250_hpt_init; + mips_hpt_frequency = V_SCD_TIMER_FREQ; mips_hpt_read = sb1250_hpt_read; + mips_hpt_mask = M_SCD_TIMER_INIT; } } @@ -149,11 +141,7 @@ void sb1250_timer_interrupt(void) /* * The HPT is free running from SB1250_HPT_VALUE down to 0 then starts over - * again. There's no easy way to set to a specific value so store init value - * in hpt_offset and subtract each time. - * - * Note: Timer isn't full 32bits so shift it into the upper part making - * it appear to run at a higher frequency. + * again. */ static unsigned int sb1250_hpt_read(void) { @@ -161,13 +149,5 @@ static unsigned int sb1250_hpt_read(void) count = G_SCD_TIMER_CNT(__raw_readq(IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM, R_SCD_TIMER_CNT)))); - count = (SB1250_HPT_VALUE - count) << SB1250_HPT_SHIFT; - - return count - hpt_offset; -} - -static void sb1250_hpt_init(unsigned int count) -{ - hpt_offset = count; - return; + return SB1250_HPT_VALUE - count; } diff --git a/include/asm-mips/div64.h b/include/asm-mips/div64.h index 5f7dcf5..d107832 100644 --- a/include/asm-mips/div64.h +++ b/include/asm-mips/div64.h @@ -83,27 +83,6 @@ #if (_MIPS_SZLONG == 64) /* - * Don't use this one in new code - */ -#define do_div64_32(res, high, low, base) ({ \ - unsigned int __quot, __mod; \ - unsigned long __div; \ - unsigned int __low, __high, __base; \ - \ - __high = (high); \ - __low = (low); \ - __div = __high; \ - __div = __div << 32 | __low; \ - __base = (base); \ - \ - __mod = __div % __base; \ - __div = __div / __base; \ - \ - __quot = __div; \ - (res) = __quot; \ - __mod; }) - -/* * Hey, we're already 64-bit, no * need to play games.. */ diff --git a/include/asm-mips/sibyte/sb1250.h b/include/asm-mips/sibyte/sb1250.h index b09e16c..2ba6988 100644 --- a/include/asm-mips/sibyte/sb1250.h +++ b/include/asm-mips/sibyte/sb1250.h @@ -51,8 +51,8 @@ extern void sb1250_mask_irq(int cpu, int irq); extern void sb1250_unmask_irq(int cpu, int irq); extern void sb1250_smp_finish(void); +extern void bcm1480_hpt_setup(void); extern void bcm1480_time_init(void); -extern unsigned long bcm1480_gettimeoffset(void); extern void bcm1480_mask_irq(int cpu, int irq); extern void bcm1480_unmask_irq(int cpu, int irq); extern void bcm1480_smp_finish(void); diff --git a/include/asm-mips/time.h b/include/asm-mips/time.h index 28512ba..625acd3 100644 --- a/include/asm-mips/time.h +++ b/include/asm-mips/time.h @@ -48,7 +48,8 @@ extern void (*mips_timer_ack)(void); * If mips_hpt_read is NULL, an R4k-compatible timer setup is attempted. */ extern unsigned int (*mips_hpt_read)(void); -extern void (*mips_hpt_init)(unsigned int); +extern void (*mips_hpt_init)(void); +extern unsigned int mips_hpt_mask; /* * to_tm() converts system time back to (year, mon, day, hour, min, sec). @@ -58,13 +59,6 @@ extern void (*mips_hpt_init)(unsigned int); extern void to_tm(unsigned long tim, struct rtc_time *tm); /* - * do_gettimeoffset(). By default, this func pointer points to - * do_null_gettimeoffset(), which leads to the same resolution as HZ. - * Higher resolution versions are available, which give ~1us resolution. - */ -extern unsigned long (*do_gettimeoffset)(void); - -/* * high-level timer interrupt routines. */ extern irqreturn_t timer_interrupt(int irq, void *dev_id); -- cgit v0.10.2 From 5ee7737379b1d7f0c977c0f1661fbaf01a8d4721 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Fri, 27 Oct 2006 19:49:53 +0900 Subject: [IA64] cpu-hotplug: Fixing confliction between CPU hot-add and IPI Signed-off-by: Kenji Kaneshige Acked-by: Satoru Takeuchi Acked-by: KAMEZAWA Hiroyuki Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index 657ac99..2763e79 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -328,10 +328,14 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic, int wait) { struct call_data_struct data; - int cpus = num_online_cpus()-1; + int cpus; - if (!cpus) + spin_lock(&call_lock); + cpus = num_online_cpus() - 1; + if (!cpus) { + spin_unlock(&call_lock); return 0; + } /* Can deadlock when called with interrupts disabled */ WARN_ON(irqs_disabled()); @@ -343,8 +347,6 @@ smp_call_function (void (*func) (void *info), void *info, int nonatomic, int wai if (wait) atomic_set(&data.finished, 0); - spin_lock(&call_lock); - call_data = &data; mb(); /* ensure store to call_data precedes setting of IPI_CALL_FUNC */ send_IPI_allbutself(IPI_CALL_FUNC); -- cgit v0.10.2 From 264b0f99308436deaee38bab99e586612d012fc1 Mon Sep 17 00:00:00 2001 From: Russ Anderson Date: Wed, 25 Oct 2006 17:59:47 -0500 Subject: [IA64] MCA recovery: Montecito support The information in MCA records is filled in slightly differently on Montecito than on Madison/McKinley. Usually, the cache check and bus check target identifiers have the same address. On Montecito the cache check and bus check target identifiers can be different if a corrected error (ie SBE or unconsumed poison data) was encountered and then an uncorrected error (ie DBE) was consumed. In that case, the cache check target identifier is the physical address of the DBE (that caused the MCA to surface) while the bus check target identifier is the physical address of the SBE. This patch correctly finds the target identifier that triggered the MCA. If there are multiple valid cache target identifiers in the same error record then use the one with the lowest cache level. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index a45009d..afc1403 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -435,6 +435,50 @@ is_mca_global(peidx_table_t *peidx, pal_bus_check_info_t *pbci, } /** + * get_target_identifier - Get the valid Cache or Bus check target identifier. + * @peidx: pointer of index of processor error section + * + * Return value: + * target address on Success / 0 on Failue + */ +static u64 +get_target_identifier(peidx_table_t *peidx) +{ + u64 target_address = 0; + sal_log_mod_error_info_t *smei; + pal_cache_check_info_t *pcci; + int i, level = 9; + + /* + * Look through the cache checks for a valid target identifier + * If more than one valid target identifier, return the one + * with the lowest cache level. + */ + for (i = 0; i < peidx_cache_check_num(peidx); i++) { + smei = (sal_log_mod_error_info_t *)peidx_cache_check(peidx, i); + if (smei->valid.target_identifier && smei->target_identifier) { + pcci = (pal_cache_check_info_t *)&(smei->check_info); + if (!target_address || (pcci->level < level)) { + target_address = smei->target_identifier; + level = pcci->level; + continue; + } + } + } + if (target_address) + return target_address; + + /* + * Look at the bus check for a valid target identifier + */ + smei = peidx_bus_check(peidx, 0); + if (smei && smei->valid.target_identifier) + return smei->target_identifier; + + return 0; +} + +/** * recover_from_read_error - Try to recover the errors which type are "read"s. * @slidx: pointer of index of SAL error record * @peidx: pointer of index of processor error section @@ -450,13 +494,14 @@ recover_from_read_error(slidx_table_t *slidx, peidx_table_t *peidx, pal_bus_check_info_t *pbci, struct ia64_sal_os_state *sos) { - sal_log_mod_error_info_t *smei; + u64 target_identifier; pal_min_state_area_t *pmsa; struct ia64_psr *psr1, *psr2; ia64_fptr_t *mca_hdlr_bh = (ia64_fptr_t*)mca_handler_bhhook; /* Is target address valid? */ - if (!pbci->tv) + target_identifier = get_target_identifier(peidx); + if (!target_identifier) return fatal_mca("target address not valid"); /* @@ -487,32 +532,28 @@ recover_from_read_error(slidx_table_t *slidx, pmsa = sos->pal_min_state; if (psr1->cpl != 0 || ((psr2->cpl != 0) && mca_recover_range(pmsa->pmsa_iip))) { - smei = peidx_bus_check(peidx, 0); - if (smei->valid.target_identifier) { - /* - * setup for resume to bottom half of MCA, - * "mca_handler_bhhook" - */ - /* pass to bhhook as argument (gr8, ...) */ - pmsa->pmsa_gr[8-1] = smei->target_identifier; - pmsa->pmsa_gr[9-1] = pmsa->pmsa_iip; - pmsa->pmsa_gr[10-1] = pmsa->pmsa_ipsr; - /* set interrupted return address (but no use) */ - pmsa->pmsa_br0 = pmsa->pmsa_iip; - /* change resume address to bottom half */ - pmsa->pmsa_iip = mca_hdlr_bh->fp; - pmsa->pmsa_gr[1-1] = mca_hdlr_bh->gp; - /* set cpl with kernel mode */ - psr2 = (struct ia64_psr *)&pmsa->pmsa_ipsr; - psr2->cpl = 0; - psr2->ri = 0; - psr2->bn = 1; - psr2->i = 0; - - return mca_recovered("user memory corruption. " + /* + * setup for resume to bottom half of MCA, + * "mca_handler_bhhook" + */ + /* pass to bhhook as argument (gr8, ...) */ + pmsa->pmsa_gr[8-1] = target_identifier; + pmsa->pmsa_gr[9-1] = pmsa->pmsa_iip; + pmsa->pmsa_gr[10-1] = pmsa->pmsa_ipsr; + /* set interrupted return address (but no use) */ + pmsa->pmsa_br0 = pmsa->pmsa_iip; + /* change resume address to bottom half */ + pmsa->pmsa_iip = mca_hdlr_bh->fp; + pmsa->pmsa_gr[1-1] = mca_hdlr_bh->gp; + /* set cpl with kernel mode */ + psr2 = (struct ia64_psr *)&pmsa->pmsa_ipsr; + psr2->cpl = 0; + psr2->ri = 0; + psr2->bn = 1; + psr2->i = 0; + + return mca_recovered("user memory corruption. " "kill affected process - recovered."); - } - } return fatal_mca("kernel context not recovered, iip 0x%lx\n", -- cgit v0.10.2 From fa1d19e5d9a94120f31e5783ab44758f46892d94 Mon Sep 17 00:00:00 2001 From: Troy Heber Date: Wed, 25 Oct 2006 14:46:15 -0600 Subject: [IA64] move SAL_CACHE_FLUSH check later in boot The check to see if the firmware drops interrupts during a SAL_CACHE_FLUSH is done to early in the boot. SAL_CACHE_FLUSH expects to be able to make PAL calls in virtual mode, on some cell based machines a fault occurs causing a MCA. This patch moves the check after mmu_context_init so the TLB and VHPT are properly setup. Signed-off-by Troy Heber Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/sal.c b/arch/ia64/kernel/sal.c index 642fdc7..20bad78 100644 --- a/arch/ia64/kernel/sal.c +++ b/arch/ia64/kernel/sal.c @@ -223,12 +223,13 @@ static void __init sal_desc_ap_wakeup(void *p) { } */ static int sal_cache_flush_drops_interrupts; -static void __init +void __init check_sal_cache_flush (void) { unsigned long flags; int cpu; - u64 vector; + u64 vector, cache_type = 3; + struct ia64_sal_retval isrv; cpu = get_cpu(); local_irq_save(flags); @@ -243,7 +244,10 @@ check_sal_cache_flush (void) while (!ia64_get_irr(IA64_TIMER_VECTOR)) cpu_relax(); - ia64_sal_cache_flush(3); + SAL_CALL(isrv, SAL_CACHE_FLUSH, cache_type, 0, 0, 0, 0, 0, 0); + + if (isrv.status) + printk(KERN_ERR "SAL_CAL_FLUSH failed with %ld\n", isrv.status); if (ia64_get_irr(IA64_TIMER_VECTOR)) { vector = ia64_get_ivr(); @@ -331,7 +335,6 @@ ia64_sal_init (struct ia64_sal_systab *systab) p += SAL_DESC_SIZE(*p); } - check_sal_cache_flush(); } int diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index c4caa80..d10404a 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -457,6 +457,8 @@ setup_arch (char **cmdline_p) cpu_init(); /* initialize the bootstrap CPU */ mmu_context_init(); /* initialize context_id bitmap */ + check_sal_cache_flush(); + #ifdef CONFIG_ACPI acpi_boot_init(); #endif diff --git a/include/asm-ia64/sal.h b/include/asm-ia64/sal.h index 0b210ab..d000689 100644 --- a/include/asm-ia64/sal.h +++ b/include/asm-ia64/sal.h @@ -659,6 +659,7 @@ ia64_sal_freq_base (unsigned long which, unsigned long *ticks_per_second, } extern s64 ia64_sal_cache_flush (u64 cache_type); +extern void __init check_sal_cache_flush (void); /* Initialize all the processor and platform level instruction and data caches */ static inline s64 -- cgit v0.10.2 From 024e4f2c5175a482c234cf67ed22368d770bf78f Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Wed, 18 Oct 2006 15:36:49 +1000 Subject: [IA64] Correct definition of handle_IPI The declaration of handle_IPI in arch/ia64/kernel/smp.c was changed but not the definition of this function. Remove struct pt_regs from handle_IPI(). Signed-off-by: Keith Owens Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index 2763e79..6ab95ce 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -108,7 +108,7 @@ cpu_die(void) } irqreturn_t -handle_IPI (int irq, void *dev_id, struct pt_regs *regs) +handle_IPI (int irq, void *dev_id) { int this_cpu = get_cpu(); unsigned long *pending_ipis = &__ia64_per_cpu_var(ipi_operation); -- cgit v0.10.2 From 2d38caba5fd148976f54930782e8209fa45879a0 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 30 Oct 2006 19:52:31 +0100 Subject: [PATCH] ep93xx_eth: fix RX/TXstatus ring full handling Ray Lehtiniemi reported that an incoming UDP packet flood can lock up the ep93xx ethernet driver. Herbert Valerio Riedel noted that due to the way ep93xx_eth manages the RX/TXstatus rings, it cannot distinguish a full ring from an empty one, and correctly suggested that this was likely to be causing this lockup to occur. Instead of looking at the hardware's RX/TXstatus ring write pointers to determine when to stop reading from those rings, we should just check every individual RX/TXstatus descriptor's valid bit instead, since there is no other way to distinguish an empty ring from a full ring, and if there is a descriptor waiting, we take the hit of reading the descriptor from memory anyway. Signed-off-by: Lennert Buytenhek Signed-off-by: Jeff Garzik diff --git a/drivers/net/arm/ep93xx_eth.c b/drivers/net/arm/ep93xx_eth.c index 127561c..2fc8b2a 100644 --- a/drivers/net/arm/ep93xx_eth.c +++ b/drivers/net/arm/ep93xx_eth.c @@ -193,12 +193,9 @@ static struct net_device_stats *ep93xx_get_stats(struct net_device *dev) static int ep93xx_rx(struct net_device *dev, int *budget) { struct ep93xx_priv *ep = netdev_priv(dev); - int tail_offset; int rx_done; int processed; - tail_offset = rdl(ep, REG_RXSTSQCURADD) - ep->descs_dma_addr; - rx_done = 0; processed = 0; while (*budget > 0) { @@ -211,28 +208,23 @@ static int ep93xx_rx(struct net_device *dev, int *budget) entry = ep->rx_pointer; rstat = ep->descs->rstat + entry; - if ((void *)rstat - (void *)ep->descs == tail_offset) { + + rstat0 = rstat->rstat0; + rstat1 = rstat->rstat1; + if (!(rstat0 & RSTAT0_RFP) || !(rstat1 & RSTAT1_RFP)) { rx_done = 1; break; } - rstat0 = rstat->rstat0; - rstat1 = rstat->rstat1; rstat->rstat0 = 0; rstat->rstat1 = 0; - if (!(rstat0 & RSTAT0_RFP)) - printk(KERN_CRIT "ep93xx_rx: buffer not done " - " %.8x %.8x\n", rstat0, rstat1); if (!(rstat0 & RSTAT0_EOF)) printk(KERN_CRIT "ep93xx_rx: not end-of-frame " " %.8x %.8x\n", rstat0, rstat1); if (!(rstat0 & RSTAT0_EOB)) printk(KERN_CRIT "ep93xx_rx: not end-of-buffer " " %.8x %.8x\n", rstat0, rstat1); - if (!(rstat1 & RSTAT1_RFP)) - printk(KERN_CRIT "ep93xx_rx: buffer1 not done " - " %.8x %.8x\n", rstat0, rstat1); if ((rstat1 & RSTAT1_BUFFER_INDEX) >> 16 != entry) printk(KERN_CRIT "ep93xx_rx: entry mismatch " " %.8x %.8x\n", rstat0, rstat1); @@ -301,13 +293,8 @@ err: static int ep93xx_have_more_rx(struct ep93xx_priv *ep) { - struct ep93xx_rstat *rstat; - int tail_offset; - - rstat = ep->descs->rstat + ep->rx_pointer; - tail_offset = rdl(ep, REG_RXSTSQCURADD) - ep->descs_dma_addr; - - return !((void *)rstat - (void *)ep->descs == tail_offset); + struct ep93xx_rstat *rstat = ep->descs->rstat + ep->rx_pointer; + return !!((rstat->rstat0 & RSTAT0_RFP) && (rstat->rstat1 & RSTAT1_RFP)); } static int ep93xx_poll(struct net_device *dev, int *budget) @@ -379,10 +366,8 @@ static int ep93xx_xmit(struct sk_buff *skb, struct net_device *dev) static void ep93xx_tx_complete(struct net_device *dev) { struct ep93xx_priv *ep = netdev_priv(dev); - int tail_offset; int wake; - tail_offset = rdl(ep, REG_TXSTSQCURADD) - ep->descs_dma_addr; wake = 0; spin_lock(&ep->tx_pending_lock); @@ -393,15 +378,13 @@ static void ep93xx_tx_complete(struct net_device *dev) entry = ep->tx_clean_pointer; tstat = ep->descs->tstat + entry; - if ((void *)tstat - (void *)ep->descs == tail_offset) - break; tstat0 = tstat->tstat0; + if (!(tstat0 & TSTAT0_TXFP)) + break; + tstat->tstat0 = 0; - if (!(tstat0 & TSTAT0_TXFP)) - printk(KERN_CRIT "ep93xx_tx_complete: buffer not done " - " %.8x\n", tstat0); if (tstat0 & TSTAT0_FA) printk(KERN_CRIT "ep93xx_tx_complete: frame aborted " " %.8x\n", tstat0); -- cgit v0.10.2 From 79c356f44b26da9fe357ed1a11e7faec4fd94e13 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 30 Oct 2006 19:52:54 +0100 Subject: [PATCH] ep93xx_eth: fix unlikely(x) > y test Fix unlikely(x) > y test in ep93xx_eth. Signed-off-by: Lennert Buytenhek Signed-off-by: Jeff Garzik diff --git a/drivers/net/arm/ep93xx_eth.c b/drivers/net/arm/ep93xx_eth.c index 2fc8b2a..90d77ec 100644 --- a/drivers/net/arm/ep93xx_eth.c +++ b/drivers/net/arm/ep93xx_eth.c @@ -334,7 +334,7 @@ static int ep93xx_xmit(struct sk_buff *skb, struct net_device *dev) struct ep93xx_priv *ep = netdev_priv(dev); int entry; - if (unlikely(skb->len) > MAX_PKT_SIZE) { + if (unlikely(skb->len > MAX_PKT_SIZE)) { ep->stats.tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; -- cgit v0.10.2 From 06f0015ace46ce9d313ec02d6b13c47c8e795a6c Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 30 Oct 2006 19:54:08 +0100 Subject: [PATCH] ep93xx_eth: don't report RX errors Flooding the console with error messages for every RX FIFO overrun, checksum error and framing error isn't very sensible. Each of these errors can occur during normal operation, so stop printk'ing error messages for RX errors at all. Signed-off-by: Lennert Buytenhek Signed-off-by: Jeff Garzik diff --git a/drivers/net/arm/ep93xx_eth.c b/drivers/net/arm/ep93xx_eth.c index 90d77ec..8ebd68e 100644 --- a/drivers/net/arm/ep93xx_eth.c +++ b/drivers/net/arm/ep93xx_eth.c @@ -230,9 +230,6 @@ static int ep93xx_rx(struct net_device *dev, int *budget) " %.8x %.8x\n", rstat0, rstat1); if (!(rstat0 & RSTAT0_RWE)) { - printk(KERN_NOTICE "ep93xx_rx: receive error " - " %.8x %.8x\n", rstat0, rstat1); - ep->stats.rx_errors++; if (rstat0 & RSTAT0_OE) ep->stats.rx_fifo_errors++; -- cgit v0.10.2 From 9d4df9e0fadfc84cd826e0f7e946691b4d7baee5 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 29 Oct 2006 03:52:14 +0900 Subject: [PATCH] tokenring: fix module_init error handling - Call platform_driver_unregister() before return when no cards found. (fixes data corruption when no cards found) - Check platform_device_register_simple() return value Cc: Jeff Garzik Cc: Mike Phillips Signed-off-by: Akinobu Mita drivers/net/tokenring/proteon.c | 9 +++++++-- drivers/net/tokenring/skisa.c | 9 +++++++-- 2 files changed, 14 insertions(+), 4 deletions(-) Signed-off-by: Jeff Garzik diff --git a/drivers/net/tokenring/proteon.c b/drivers/net/tokenring/proteon.c index 4f75696..cb7dbb6 100644 --- a/drivers/net/tokenring/proteon.c +++ b/drivers/net/tokenring/proteon.c @@ -370,6 +370,10 @@ static int __init proteon_init(void) dev->dma = dma[i]; pdev = platform_device_register_simple("proteon", i, NULL, 0); + if (IS_ERR(pdev)) { + free_netdev(dev); + continue; + } err = setup_card(dev, &pdev->dev); if (!err) { proteon_dev[i] = pdev; @@ -385,9 +389,10 @@ static int __init proteon_init(void) /* Probe for cards. */ if (num == 0) { printk(KERN_NOTICE "proteon.c: No cards found.\n"); - return (-ENODEV); + platform_driver_unregister(&proteon_driver); + return -ENODEV; } - return (0); + return 0; } static void __exit proteon_cleanup(void) diff --git a/drivers/net/tokenring/skisa.c b/drivers/net/tokenring/skisa.c index d6ba41c..33afea3 100644 --- a/drivers/net/tokenring/skisa.c +++ b/drivers/net/tokenring/skisa.c @@ -380,6 +380,10 @@ static int __init sk_isa_init(void) dev->dma = dma[i]; pdev = platform_device_register_simple("skisa", i, NULL, 0); + if (IS_ERR(pdev)) { + free_netdev(dev); + continue; + } err = setup_card(dev, &pdev->dev); if (!err) { sk_isa_dev[i] = pdev; @@ -395,9 +399,10 @@ static int __init sk_isa_init(void) /* Probe for cards. */ if (num == 0) { printk(KERN_NOTICE "skisa.c: No cards found.\n"); - return (-ENODEV); + platform_driver_unregister(&sk_isa_driver); + return -ENODEV; } - return (0); + return 0; } static void __exit sk_isa_cleanup(void) -- cgit v0.10.2 From 09669585b5d0cfdebe28250d442693b3baac66a2 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 29 Oct 2006 03:47:12 +0900 Subject: [PATCH] n2: fix confusing error code modprobe n2 with no parameters or no such devices will get confusing error message. # modprobe n2 ... Kernel does not have module support This patch replaces return code from -ENOSYS to -EINVAL. Cc: Jeff Garzik Cc: Krzysztof Halasa Signed-off-by: Akinobu Mita drivers/net/wan/n2.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) Signed-off-by: Jeff Garzik diff --git a/drivers/net/wan/n2.c b/drivers/net/wan/n2.c index dcf46ad..5c322df 100644 --- a/drivers/net/wan/n2.c +++ b/drivers/net/wan/n2.c @@ -500,7 +500,7 @@ static int __init n2_init(void) #ifdef MODULE printk(KERN_INFO "n2: no card initialized\n"); #endif - return -ENOSYS; /* no parameters specified, abort */ + return -EINVAL; /* no parameters specified, abort */ } printk(KERN_INFO "%s\n", version); @@ -538,11 +538,11 @@ static int __init n2_init(void) n2_run(io, irq, ram, valid[0], valid[1]); if (*hw == '\x0') - return first_card ? 0 : -ENOSYS; + return first_card ? 0 : -EINVAL; }while(*hw++ == ':'); printk(KERN_ERR "n2: invalid hardware parameters\n"); - return first_card ? 0 : -ENOSYS; + return first_card ? 0 : -EINVAL; } -- cgit v0.10.2 From f479b322a0949d540b45aea645793058b0c50be5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 27 Oct 2006 10:22:10 -0700 Subject: [PATCH] sky2: not experimental The sky2 driver is no longer in experimental state. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index e38846e..28c17d1 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2112,7 +2112,7 @@ config SKGE config SKY2 tristate "SysKonnect Yukon2 support (EXPERIMENTAL)" - depends on PCI && EXPERIMENTAL + depends on PCI select CRC32 ---help--- This driver supports Gigabit Ethernet adapters based on the @@ -2120,8 +2120,8 @@ config SKY2 Marvell 88E8021/88E8022/88E8035/88E8036/88E8038/88E8050/88E8052/ 88E8053/88E8055/88E8061/88E8062, SysKonnect SK-9E21D/SK-9S21 - This driver does not support the original Yukon chipset: a seperate - driver, skge, is provided for Yukon-based adapters. + There is companion driver for the older Marvell Yukon and + Genesis based adapters: skge. To compile this driver as a module, choose M here: the module will be called sky2. This is recommended. -- cgit v0.10.2 From 1e7bed081968c42469bd02842b4190a115008221 Mon Sep 17 00:00:00 2001 From: Brice Goglin Date: Thu, 26 Oct 2006 22:51:33 +0200 Subject: [PATCH] myri10ge: ServerWorks HT2000 PCI id is already defined in pci_ids.h No need to keep defining PCI_DEVICE_ID_SERVERWORKS_HT2000_PCIE in the driver code since it is now defined in pci_ids.h. Signed-off-by: Brice Goglin Signed-off-by: Jeff Garzik diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index fdbb0d7..806081b 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -2416,7 +2416,6 @@ static void myri10ge_enable_ecrc(struct myri10ge_priv *mgp) * firmware image, and set tx.boundary to 4KB. */ -#define PCI_DEVICE_ID_SERVERWORKS_HT2000_PCIE 0x0132 #define PCI_DEVICE_ID_INTEL_E5000_PCIE23 0x25f7 #define PCI_DEVICE_ID_INTEL_E5000_PCIE47 0x25fa -- cgit v0.10.2 From 1e1675ccf758cbb4303ab052d58405cda6c745a7 Mon Sep 17 00:00:00 2001 From: Jan-Bernd Themann Date: Wed, 25 Oct 2006 13:11:42 +0200 Subject: [PATCH] ehea: kzalloc GFP_ATOMIC fix This patch fixes kzalloc parameters (GFP_ATOMIC instead of GFP_KERNEL) Signed-off-by: Jan-Bernd Themann Signed-off-by: Jeff Garzik diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index eb7d44d..4538c99 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -586,8 +586,8 @@ int ehea_sense_port_attr(struct ehea_port *port) u64 hret; struct hcp_ehea_port_cb0 *cb0; - cb0 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); - if (!cb0) { + cb0 = kzalloc(H_CB_ALIGNMENT, GFP_ATOMIC); /* May be called via */ + if (!cb0) { /* ehea_neq_tasklet() */ ehea_error("no mem for cb0"); ret = -ENOMEM; goto out; @@ -765,8 +765,7 @@ static void ehea_parse_eqe(struct ehea_adapter *adapter, u64 eqe) if (EHEA_BMASK_GET(NEQE_PORT_UP, eqe)) { if (!netif_carrier_ok(port->netdev)) { - ret = ehea_sense_port_attr( - port); + ret = ehea_sense_port_attr(port); if (ret) { ehea_error("failed resensing port " "attributes"); @@ -1502,7 +1501,7 @@ static void ehea_promiscuous(struct net_device *dev, int enable) if ((enable && port->promisc) || (!enable && !port->promisc)) return; - cb7 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb7 = kzalloc(H_CB_ALIGNMENT, GFP_ATOMIC); if (!cb7) { ehea_error("no mem for cb7"); goto out; @@ -1606,7 +1605,7 @@ static void ehea_add_multicast_entry(struct ehea_port* port, u8* mc_mac_addr) struct ehea_mc_list *ehea_mcl_entry; u64 hret; - ehea_mcl_entry = kzalloc(sizeof(*ehea_mcl_entry), GFP_KERNEL); + ehea_mcl_entry = kzalloc(sizeof(*ehea_mcl_entry), GFP_ATOMIC); if (!ehea_mcl_entry) { ehea_error("no mem for mcl_entry"); return; -- cgit v0.10.2 From 2ceaac755423cb93c1bb2f59ebd1a06f027ac095 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 30 Oct 2006 14:19:25 -0800 Subject: [PATCH] net s2io: return on NULL dev_alloc_skb() Checks for NULL dev_alloc_skb() and returns on true to avoid subsequent dereference. Cc: Jeff Garzik Cc: Christoph Hellwig Signed-off-by: David Rientjes Signed-off-by: Jeff Garzik diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index a231ab7..33569ec 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -5985,6 +5985,11 @@ static int set_rxd_buffer_pointer(nic_t *sp, RxD_t *rxdp, buffAdd_t *ba, ((RxD3_t*)rxdp)->Buffer1_ptr = *temp1; } else { *skb = dev_alloc_skb(size); + if (!(*skb)) { + DBG_PRINT(ERR_DBG, "%s: dev_alloc_skb failed\n", + dev->name); + return -ENOMEM; + } ((RxD3_t*)rxdp)->Buffer2_ptr = *temp2 = pci_map_single(sp->pdev, (*skb)->data, dev->mtu + 4, @@ -6007,7 +6012,11 @@ static int set_rxd_buffer_pointer(nic_t *sp, RxD_t *rxdp, buffAdd_t *ba, ((RxD3_t*)rxdp)->Buffer2_ptr = *temp2; } else { *skb = dev_alloc_skb(size); - + if (!(*skb)) { + DBG_PRINT(ERR_DBG, "%s: dev_alloc_skb failed\n", + dev->name); + return -ENOMEM; + } ((RxD3_t*)rxdp)->Buffer0_ptr = *temp0 = pci_map_single(sp->pdev, ba->ba_0, BUF0_LEN, PCI_DMA_FROMDEVICE); -- cgit v0.10.2 From 798b6b19d7a4b6e1ea5340ec8b3b92811e05b81b Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sun, 22 Oct 2006 20:16:57 -0700 Subject: [PATCH] skge, sky2, et all. gplv2 only I don't want my code to downgraded to GPLv3 because of cut-n-pasted the comments. These files which I hold copyright on were started before it was clear what GPLv3 was going to be. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/irda/stir4200.c b/drivers/net/irda/stir4200.c index be8a66e..3b4c478 100644 --- a/drivers/net/irda/stir4200.c +++ b/drivers/net/irda/stir4200.c @@ -15,8 +15,7 @@ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by -* the Free Software Foundation; either version 2 of the License, or -* (at your option) any later version. +* the Free Software Foundation; either version 2 of the License. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of diff --git a/drivers/net/skge.c b/drivers/net/skge.c index e7e4149..b294903 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -11,8 +11,7 @@ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. + * the Free Software Foundation; either version 2 of the License. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 95efdb5..53171f6 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -10,8 +10,7 @@ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. + * the Free Software Foundation; either version 2 of the License. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index ef8874b..0441876 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -4,7 +4,7 @@ * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. + * 2 of the License. * * Many of the algorithms and ideas for this came from * NIST Net which is not copyrighted. -- cgit v0.10.2 From 0ca43235b34c92278fa903297acef37198ec3e26 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 18 Oct 2006 13:39:28 -0700 Subject: [PATCH] sky2: netpoll on dual port cards The sky2 driver uses a single NAPI poll routine for both ports on dual ported cards (because there is a single IRQ and status ring). Netpoll makes assumptions about the relationship between network device and NAPI that aren't correct on the second port, this will cause the port to never clear work. Most systems, just have single port, so not a big issue. The easy fix is just make the second port, not netpoll capable. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 53171f6..16616f5 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -3238,7 +3238,11 @@ static __devinit struct net_device *sky2_init_netdev(struct sky2_hw *hw, dev->poll = sky2_poll; dev->weight = NAPI_WEIGHT; #ifdef CONFIG_NET_POLL_CONTROLLER - dev->poll_controller = sky2_netpoll; + /* Network console (only works on port 0) + * because netpoll makes assumptions about NAPI + */ + if (port == 0) + dev->poll_controller = sky2_netpoll; #endif sky2 = netdev_priv(dev); -- cgit v0.10.2 From cf0e812f0e90ee496af072b136e8bd02770387e6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 27 Oct 2006 19:08:47 -0700 Subject: [PATCH] sata_sis: fix flags handling for the secondary port sis_init_one() modifies probe_ent->port_flags after allocating and initializing it using ata_pci_init_native_mode(). This makes port_flags for the secondary port (probe_ent->pinfo2->flags) go out of sync resulting in misdetection of device due to incorrectly initialized SCR access flag. This patch make probe_ent alloc/init happen after the final port flags value is determined. This is fragile but probe_ent and all the related mess are scheduled to go away soon for exactly this reason. We just need to hold everything together till then. This has been spotted and diagnosed and tested by Patrick McHardy. Signed-off-by: Tejun Heo Cc: Patric McHardy Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c index 0738f52..9d1235b 100644 --- a/drivers/ata/sata_sis.c +++ b/drivers/ata/sata_sis.c @@ -240,7 +240,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) struct ata_probe_ent *probe_ent = NULL; int rc; u32 genctl; - struct ata_port_info *ppi[2]; + struct ata_port_info pi = sis_port_info, *ppi[2] = { &pi, &pi }; int pci_dev_busy = 0; u8 pmr; u8 port2_start; @@ -265,27 +265,20 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) goto err_out_regions; - ppi[0] = ppi[1] = &sis_port_info; - probe_ent = ata_pci_init_native_mode(pdev, ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); - if (!probe_ent) { - rc = -ENOMEM; - goto err_out_regions; - } - /* check and see if the SCRs are in IO space or PCI cfg space */ pci_read_config_dword(pdev, SIS_GENCTL, &genctl); if ((genctl & GENCTL_IOMAPPED_SCR) == 0) - probe_ent->port_flags |= SIS_FLAG_CFGSCR; + pi.flags |= SIS_FLAG_CFGSCR; /* if hardware thinks SCRs are in IO space, but there are * no IO resources assigned, change to PCI cfg space. */ - if ((!(probe_ent->port_flags & SIS_FLAG_CFGSCR)) && + if ((!(pi.flags & SIS_FLAG_CFGSCR)) && ((pci_resource_start(pdev, SIS_SCR_PCI_BAR) == 0) || (pci_resource_len(pdev, SIS_SCR_PCI_BAR) < 128))) { genctl &= ~GENCTL_IOMAPPED_SCR; pci_write_config_dword(pdev, SIS_GENCTL, genctl); - probe_ent->port_flags |= SIS_FLAG_CFGSCR; + pi.flags |= SIS_FLAG_CFGSCR; } pci_read_config_byte(pdev, SIS_PMR, &pmr); @@ -306,6 +299,12 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) port2_start = 0x20; } + probe_ent = ata_pci_init_native_mode(pdev, ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); + if (!probe_ent) { + rc = -ENOMEM; + goto err_out_regions; + } + if (!(probe_ent->port_flags & SIS_FLAG_CFGSCR)) { probe_ent->port[0].scr_addr = pci_resource_start(pdev, SIS_SCR_PCI_BAR); -- cgit v0.10.2 From f833229c96c0bf53c05995e4bd58709d9e9edd67 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 31 Oct 2006 09:31:37 +0100 Subject: [PATCH] Add 0x7110 piix to ata_piix.c Hi Jeff, I tested the PATA support on my old VAIO notebook, and it failed to find my piix device: 00:07.1 Class 0101: 8086:7111 (rev 01) (prog-if 80 [Master]) Control: I/O+ Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 5250187..4fad8d2 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -169,6 +169,7 @@ static const struct pci_device_id piix_pci_tbl[] = { #ifdef ATA_ENABLE_PATA /* Intel PIIX4 for the 430TX/440BX/MX chipset: UDMA 33 */ /* Also PIIX4E (fn3 rev 2) and PIIX4M (fn3 rev 3) */ + { 0x8086, 0x7110, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 }, { 0x8086, 0x7111, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 }, { 0x8086, 0x24db, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 }, { 0x8086, 0x25a2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 }, -- cgit v0.10.2 From 6e42acc4115bc376b8523acbcba2b2b7cc27d016 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 27 Oct 2006 19:08:42 -0700 Subject: [PATCH] libata: unexport ata_dev_revalidate() ata_dev_revalidate() isn't used outside of libata core. Unexport it. Signed-off-by: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 83728a9..a8fd0c3 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -6122,7 +6122,6 @@ EXPORT_SYMBOL_GPL(ata_std_prereset); EXPORT_SYMBOL_GPL(ata_std_softreset); EXPORT_SYMBOL_GPL(sata_std_hardreset); EXPORT_SYMBOL_GPL(ata_std_postreset); -EXPORT_SYMBOL_GPL(ata_dev_revalidate); EXPORT_SYMBOL_GPL(ata_dev_classify); EXPORT_SYMBOL_GPL(ata_dev_pair); EXPORT_SYMBOL_GPL(ata_port_disable); diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index a5ecb71..0ed263b 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -53,6 +53,7 @@ extern unsigned ata_exec_internal(struct ata_device *dev, extern unsigned int ata_do_simple_cmd(struct ata_device *dev, u8 cmd); extern int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class, int post_reset, u16 *id); +extern int ata_dev_revalidate(struct ata_device *dev, int post_reset); extern int ata_dev_configure(struct ata_device *dev, int print_info); extern int sata_down_spd_limit(struct ata_port *ap); extern int sata_set_spd_needed(struct ata_port *ap); diff --git a/include/linux/libata.h b/include/linux/libata.h index b03d5a3..abd2deb 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -702,7 +702,6 @@ extern int ata_std_prereset(struct ata_port *ap); extern int ata_std_softreset(struct ata_port *ap, unsigned int *classes); extern int sata_std_hardreset(struct ata_port *ap, unsigned int *class); extern void ata_std_postreset(struct ata_port *ap, unsigned int *classes); -extern int ata_dev_revalidate(struct ata_device *dev, int post_reset); extern void ata_port_disable(struct ata_port *); extern void ata_std_ports(struct ata_ioports *ioaddr); #ifdef CONFIG_PCI -- cgit v0.10.2 From c6446a4cdadaf411bafe1565e9fa7666f3c2fe95 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Oct 2006 13:23:58 +0900 Subject: [PATCH] ata_piix: allow 01b MAP for both ICH6M and ICH7M ICH7M was separated from ICH6M to allow undocumented MAP value 01b which was spotted on an ASUS notebook. However, there is also notebooks with MAP value 01b on ICH6M. This patch re-merges ICH6M and ICH7M entries and allows MAP value 01b for both. This problem has been reported and initial patch provided by Jonathan Dieter. Signed-off-by: Tejun Heo Cc: Jonathan Dieter Cc: Tom Deblauwe Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 4fad8d2..8385387 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -126,8 +126,7 @@ enum { ich6_sata = 7, ich6_sata_ahci = 8, ich6m_sata_ahci = 9, - ich7m_sata_ahci = 10, - ich8_sata_ahci = 11, + ich8_sata_ahci = 10, /* constants for mapping table */ P0 = 0, /* port 0 */ @@ -228,7 +227,7 @@ static const struct pci_device_id piix_pci_tbl[] = { /* 82801GB/GR/GH (ICH7, identical to ICH6) */ { 0x8086, 0x27c0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich6_sata_ahci }, /* 2801GBM/GHM (ICH7M, identical to ICH6M) */ - { 0x8086, 0x27c4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich7m_sata_ahci }, + { 0x8086, 0x27c4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich6m_sata_ahci }, /* Enterprise Southbridge 2 (where's the datasheet?) */ { 0x8086, 0x2680, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich6_sata_ahci }, /* SATA Controller 1 IDE (ICH8, no datasheet yet) */ @@ -400,23 +399,10 @@ static const struct piix_map_db ich6m_map_db = { .mask = 0x3, .port_enable = 0x5, .present_shift = 4, - .map = { - /* PM PS SM SS MAP */ - { P0, P2, RV, RV }, /* 00b */ - { RV, RV, RV, RV }, - { P0, P2, IDE, IDE }, /* 10b */ - { RV, RV, RV, RV }, - }, -}; - -static const struct piix_map_db ich7m_map_db = { - .mask = 0x3, - .port_enable = 0x5, - .present_shift = 4, /* Map 01b isn't specified in the doc but some notebooks use - * it anyway. ATM, the only case spotted carries subsystem ID - * 1025:0107. This is the only difference from ich6m. + * it anyway. MAP 01b have been spotted on both ICH6M and + * ICH7M. */ .map = { /* PM PS SM SS MAP */ @@ -446,7 +432,6 @@ static const struct piix_map_db *piix_map_db_table[] = { [ich6_sata] = &ich6_map_db, [ich6_sata_ahci] = &ich6_map_db, [ich6m_sata_ahci] = &ich6m_map_db, - [ich7m_sata_ahci] = &ich7m_map_db, [ich8_sata_ahci] = &ich8_map_db, }; @@ -557,19 +542,7 @@ static struct ata_port_info piix_port_info[] = { .port_ops = &piix_sata_ops, }, - /* ich7m_sata_ahci: 10 */ - { - .sht = &piix_sht, - .flags = ATA_FLAG_SATA | - PIIX_FLAG_CHECKINTR | PIIX_FLAG_SCR | - PIIX_FLAG_AHCI, - .pio_mask = 0x1f, /* pio0-4 */ - .mwdma_mask = 0x07, /* mwdma0-2 */ - .udma_mask = 0x7f, /* udma0-6 */ - .port_ops = &piix_sata_ops, - }, - - /* ich8_sata_ahci: 11 */ + /* ich8_sata_ahci: 10 */ { .sht = &piix_sht, .flags = ATA_FLAG_SATA | -- cgit v0.10.2 From aec41a0d02342fc9e3b6bb278eae50fa29f04d1f Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Wed, 18 Oct 2006 19:34:40 +0200 Subject: [PATCH] ieee80211: don't flood log with errors The "ieee80211: Workaround malformed 802.11 frames from AP" patch (see http://kernel.org/git/?p=linux/kernel/git/linville/wireless-2.6.git;a=commit;h=f09fc44d8c25f22c4d985bb93857338ed02feac6 ) fixes the problem with some buggy APs but also converts debug message into an error one. This floods the log with errors when you are near such AP (you get a message for every beacon). This patch reverts the error message back to the debug one. Signed-off-by: Jiri Benc Signed-off-by: John W. Linville diff --git a/net/ieee80211/ieee80211_rx.c b/net/ieee80211/ieee80211_rx.c index 7707041..2759312 100644 --- a/net/ieee80211/ieee80211_rx.c +++ b/net/ieee80211/ieee80211_rx.c @@ -1078,12 +1078,12 @@ static int ieee80211_parse_info_param(struct ieee80211_info_element while (length >= sizeof(*info_element)) { if (sizeof(*info_element) + info_element->len > length) { - IEEE80211_ERROR("Info elem: parse failed: " - "info_element->len + 2 > left : " - "info_element->len+2=%zd left=%d, id=%d.\n", - info_element->len + - sizeof(*info_element), - length, info_element->id); + IEEE80211_DEBUG_MGMT("Info elem: parse failed: " + "info_element->len + 2 > left : " + "info_element->len+2=%zd left=%d, id=%d.\n", + info_element->len + + sizeof(*info_element), + length, info_element->id); /* We stop processing but don't return an error here * because some misbehaviour APs break this rule. ie. * Orinoco AP1000. */ -- cgit v0.10.2 From 115e222d538e7838bffa0f76409acd9816a0ef32 Mon Sep 17 00:00:00 2001 From: Pavel Roskin Date: Tue, 24 Oct 2006 22:41:27 -0400 Subject: [PATCH] hostap_plx: fix CIS verification The length of the manfid CIS should be at least 4, and it's normally 4. It's incorrect to require it to be at least 5. This breaks support for most (if not all) cards. The right place to ensure that we don't access beyond the CIS buffer is to strengthen another check. Make sure that the next tuple begins at least at the CIS buffer end (in which case we stop processing) or before that. Reported by ph35sm@free.fr Signed-off-by: Pavel Roskin Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/hostap/hostap_plx.c b/drivers/net/wireless/hostap/hostap_plx.c index 6dfa041..bc81b13 100644 --- a/drivers/net/wireless/hostap/hostap_plx.c +++ b/drivers/net/wireless/hostap/hostap_plx.c @@ -364,7 +364,7 @@ static int prism2_plx_check_cis(void __iomem *attr_mem, int attr_len, pos = 0; while (pos < CIS_MAX_LEN - 1 && cis[pos] != CISTPL_END) { - if (pos + cis[pos + 1] >= CIS_MAX_LEN) + if (pos + 2 + cis[pos + 1] > CIS_MAX_LEN) goto cis_error; switch (cis[pos]) { @@ -391,7 +391,7 @@ static int prism2_plx_check_cis(void __iomem *attr_mem, int attr_len, break; case CISTPL_MANFID: - if (cis[pos + 1] < 5) + if (cis[pos + 1] < 4) goto cis_error; manfid1 = cis[pos + 2] + (cis[pos + 3] << 8); manfid2 = cis[pos + 4] + (cis[pos + 5] << 8); -- cgit v0.10.2 From 81e171b95d2d06a64465a1e6ab1e2fb864ea2448 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Sat, 28 Oct 2006 17:52:34 -0500 Subject: [PATCH] bcm43xx: Fix low-traffic netdev watchdog TX timeouts This fixes a netdev watchdog timeout problem. The software needs to call netif_tx_disable before running the hardware calibration code. The problem condition can be shown by the following timegraph. |---5secs - ~10 jiffies time---|---|OOPS ^ ^ last real TX periodic work stops netif At OOPS, the following happens: The watchdog timer triggers, because the timeout of 5secs is over. The watchdog first checks for stopped TX. _Usually_ TX is only stopped from the TX handler to indicate a full TX queue. But this is different. We need to stop TX here, regardless of the TX queue state. So the watchdog recognizes the stopped device and assumes it is stopped due to full TX queues (Which is a _wrong_ assumption in this case). It then tests how far the last TX has been in the past. If it's more than 5secs (which is the case for low or no traffic), it will fire a TX timeout. Signed-off-by: Michael Buesch Signed-off-by: Larry Finger Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index a94c6d8..65edb56 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -3163,9 +3163,11 @@ static int estimate_periodic_work_badness(unsigned int state) static void bcm43xx_periodic_work_handler(void *d) { struct bcm43xx_private *bcm = d; + struct net_device *net_dev = bcm->net_dev; unsigned long flags; u32 savedirqs = 0; int badness; + unsigned long orig_trans_start = 0; mutex_lock(&bcm->mutex); badness = estimate_periodic_work_badness(bcm->periodic_state); @@ -3173,7 +3175,18 @@ static void bcm43xx_periodic_work_handler(void *d) /* Periodic work will take a long time, so we want it to * be preemtible. */ - netif_tx_disable(bcm->net_dev); + + netif_tx_lock_bh(net_dev); + /* We must fake a started transmission here, as we are going to + * disable TX. If we wouldn't fake a TX, it would be possible to + * trigger the netdev watchdog, if the last real TX is already + * some time on the past (slightly less than 5secs) + */ + orig_trans_start = net_dev->trans_start; + net_dev->trans_start = jiffies; + netif_stop_queue(net_dev); + netif_tx_unlock_bh(net_dev); + spin_lock_irqsave(&bcm->irq_lock, flags); bcm43xx_mac_suspend(bcm); if (bcm43xx_using_pio(bcm)) @@ -3198,6 +3211,7 @@ static void bcm43xx_periodic_work_handler(void *d) bcm43xx_pio_thaw_txqueues(bcm); bcm43xx_mac_enable(bcm); netif_wake_queue(bcm->net_dev); + net_dev->trans_start = orig_trans_start; } mmiowb(); spin_unlock_irqrestore(&bcm->irq_lock, flags); -- cgit v0.10.2 From df6d7c94b0c3ae6a1185c9e5fa8ee3368e4a5efb Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 17 Oct 2006 23:38:26 -0500 Subject: [PATCH] bcm43xx: fix unexpected LED control values in BCM4303 sprom The bcm43xx driver uses 4 locations in the devices sprom to determine the behavior of the leds. Certain defaults are assigned if all bits are set in those locations. On at least one BCM4303 chip, the sprom contains values other than the default, which executes an assertion placed in the default case of a following switch statement. This patch makes the leds on the above mentioned interface behave correctly. In addition, it limits the number of logged messages to 20 for the case of unexpected values in the sprom locations. Signed-off-by: Larry Finger Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_leds.c b/drivers/net/wireless/bcm43xx/bcm43xx_leds.c index 2ddbec6..7d383a2 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_leds.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_leds.c @@ -189,20 +189,24 @@ void bcm43xx_leds_update(struct bcm43xx_private *bcm, int activity) case BCM43xx_LED_INACTIVE: continue; case BCM43xx_LED_OFF: + case BCM43xx_LED_BCM4303_3: break; case BCM43xx_LED_ON: turn_on = 1; break; case BCM43xx_LED_ACTIVITY: + case BCM43xx_LED_BCM4303_0: turn_on = activity; break; case BCM43xx_LED_RADIO_ALL: turn_on = radio->enabled; break; case BCM43xx_LED_RADIO_A: + case BCM43xx_LED_BCM4303_2: turn_on = (radio->enabled && phy->type == BCM43xx_PHYTYPE_A); break; case BCM43xx_LED_RADIO_B: + case BCM43xx_LED_BCM4303_1: turn_on = (radio->enabled && (phy->type == BCM43xx_PHYTYPE_B || phy->type == BCM43xx_PHYTYPE_G)); @@ -257,7 +261,8 @@ void bcm43xx_leds_update(struct bcm43xx_private *bcm, int activity) continue; #endif /* CONFIG_BCM43XX_DEBUG */ default: - assert(0); + dprintkl(KERN_INFO PFX "Bad value in leds_update," + " led->behaviour: 0x%x\n", led->behaviour); }; if (led->activelow) diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_leds.h b/drivers/net/wireless/bcm43xx/bcm43xx_leds.h index d3716cf..811e14a 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_leds.h +++ b/drivers/net/wireless/bcm43xx/bcm43xx_leds.h @@ -46,6 +46,12 @@ enum { /* LED behaviour values */ BCM43xx_LED_TEST_BLINKSLOW, BCM43xx_LED_TEST_BLINKMEDIUM, BCM43xx_LED_TEST_BLINKFAST, + + /* Misc values for BCM4303 */ + BCM43xx_LED_BCM4303_0 = 0x2B, + BCM43xx_LED_BCM4303_1 = 0x78, + BCM43xx_LED_BCM4303_2 = 0x2E, + BCM43xx_LED_BCM4303_3 = 0x19, }; int bcm43xx_leds_init(struct bcm43xx_private *bcm); -- cgit v0.10.2 From 441cbd8dace80545db2ac43175ac1c097d96f75c Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 26 Oct 2006 15:38:10 +1000 Subject: [POWERPC] Fix various offb issues This patch fixes a few issues in offb: - A test was inverted causing the palette hack to never work (no device node was passed down to the init function) - Some cards seem to have their assigned-addresses property in a random order, thus we need to try using of_get_pci_address() first, which will fail if it's not a PCI device, and fallback to of_get_address() in that case. of_get_pci_address() properly parsees assigned-addresses to test the BAR number and thus will get it right whatever the order is. - Some cards (like GXT4500) provide a linebytes of 0xffffffff in the device-tree which does no good. This patch handles that by using the screen width when that happens. (Also fixes btext.c while at it). - Add detection of the GXT4500 in addition to the GXT2000 for the palette hacks (we use the same hack, palette is linear in register space at offset 0x6000). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/btext.c b/arch/powerpc/kernel/btext.c index 995fcef..93f21aa 100644 --- a/arch/powerpc/kernel/btext.c +++ b/arch/powerpc/kernel/btext.c @@ -182,7 +182,7 @@ int btext_initialize(struct device_node *np) prop = get_property(np, "linux,bootx-linebytes", NULL); if (prop == NULL) prop = get_property(np, "linebytes", NULL); - if (prop) + if (prop && *prop != 0xffffffffu) pitch = *prop; if (pitch == 1) pitch = 0x1000; diff --git a/drivers/video/offb.c b/drivers/video/offb.c index bad0e98..9a40bbe 100644 --- a/drivers/video/offb.c +++ b/drivers/video/offb.c @@ -157,7 +157,7 @@ static int offb_setcolreg(u_int regno, u_int red, u_int green, u_int blue, out_le32(par->cmap_adr + 0xb4, (red << 16 | green << 8 | blue)); break; case cmap_gxt2000: - out_le32((unsigned __iomem *) par->cmap_adr + regno, + out_le32(((unsigned __iomem *) par->cmap_adr) + regno, (red << 16 | green << 8 | blue)); break; } @@ -213,7 +213,7 @@ static int offb_blank(int blank, struct fb_info *info) out_le32(par->cmap_adr + 0xb4, 0); break; case cmap_gxt2000: - out_le32((unsigned __iomem *) par->cmap_adr + i, + out_le32(((unsigned __iomem *) par->cmap_adr) + i, 0); break; } @@ -226,13 +226,23 @@ static int offb_blank(int blank, struct fb_info *info) static void __iomem *offb_map_reg(struct device_node *np, int index, unsigned long offset, unsigned long size) { - struct resource r; - - if (of_address_to_resource(np, index, &r)) - return 0; - if ((r.start + offset + size) > r.end) - return 0; - return ioremap(r.start + offset, size); + const u32 *addrp; + u64 asize, taddr; + unsigned int flags; + + addrp = of_get_pci_address(np, index, &asize, &flags); + if (addrp == NULL) + addrp = of_get_address(np, index, &asize, &flags); + if (addrp == NULL) + return NULL; + if ((flags & (IORESOURCE_IO | IORESOURCE_MEM)) == 0) + return NULL; + if ((offset + size) > asize) + return NULL; + taddr = of_translate_address(np, addrp); + if (taddr == OF_BAD_ADDR) + return NULL; + return ioremap(taddr + offset, size); } static void __init offb_init_fb(const char *name, const char *full_name, @@ -289,7 +299,6 @@ static void __init offb_init_fb(const char *name, const char *full_name, par->cmap_type = cmap_unknown; if (depth == 8) { - /* Palette hacks disabled for now */ if (dp && !strncmp(name, "ATY,Rage128", 11)) { par->cmap_adr = offb_map_reg(dp, 2, 0, 0x1fff); if (par->cmap_adr) @@ -313,7 +322,8 @@ static void __init offb_init_fb(const char *name, const char *full_name, ioremap(base + 0x7ff000, 0x1000) + 0xcc0; par->cmap_data = par->cmap_adr + 1; par->cmap_type = cmap_m64; - } else if (dp && device_is_compatible(dp, "pci1014,b7")) { + } else if (dp && (device_is_compatible(dp, "pci1014,b7") || + device_is_compatible(dp, "pci1014,21c"))) { par->cmap_adr = offb_map_reg(dp, 0, 0x6000, 0x1000); if (par->cmap_adr) par->cmap_type = cmap_gxt2000; @@ -433,7 +443,7 @@ static void __init offb_init_nodriver(struct device_node *dp, int no_real_node) pp = get_property(dp, "linux,bootx-linebytes", &len); if (pp == NULL) pp = get_property(dp, "linebytes", &len); - if (pp && len == sizeof(u32)) + if (pp && len == sizeof(u32) && (*pp != 0xffffffffu)) pitch = *pp; else pitch = width * ((depth + 7) / 8); @@ -496,7 +506,7 @@ static void __init offb_init_nodriver(struct device_node *dp, int no_real_node) offb_init_fb(no_real_node ? "bootx" : dp->name, no_real_node ? "display" : dp->full_name, width, height, depth, pitch, address, - no_real_node ? dp : NULL); + no_real_node ? NULL : dp); } } -- cgit v0.10.2 From e0da0daee14862e0a5c49f2059641a8deb27eca2 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Fri, 27 Oct 2006 14:31:07 -0500 Subject: [POWERPC] Fix rmb() for e500-based machines it The e500 core generates an illegal instruction exception when it tries to execute the lwsync instruction, which we currently use for rmb(). This fixes it by using the LWSYNC macro, which turns into a plain sync on 32-bit machines. Signed-off-by: Andrew Fleming Signed-off-by: Paul Mackerras diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h index 4362759..f7b1227 100644 --- a/include/asm-powerpc/system.h +++ b/include/asm-powerpc/system.h @@ -25,8 +25,8 @@ * * We have to use the sync instructions for mb(), since lwsync doesn't * order loads with respect to previous stores. Lwsync is fine for - * rmb(), though. Note that lwsync is interpreted as sync by - * 32-bit and older 64-bit CPUs. + * rmb(), though. Note that rmb() actually uses a sync on 32-bit + * architectures. * * For wmb(), we use sync since wmb is used in drivers to order * stores to system memory with respect to writes to the device. @@ -34,7 +34,7 @@ * SMP since it is only used to order updates to system memory. */ #define mb() __asm__ __volatile__ ("sync" : : : "memory") -#define rmb() __asm__ __volatile__ ("lwsync" : : : "memory") +#define rmb() __asm__ __volatile__ (__stringify(LWSYNC) : : : "memory") #define wmb() __asm__ __volatile__ ("sync" : : : "memory") #define read_barrier_depends() do { } while(0) -- cgit v0.10.2 From dd6c89f686bdb2a5de72fab636fc839e5a0add6d Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Fri, 27 Oct 2006 15:06:32 -0500 Subject: [POWERPC] Fix oprofile support for e500 in arch/powerpc Fixed a compile error in building the 85xx support with oprofile, and in the process cleaned up some issues with the fsl_booke performance monitor code. * Reorganized FSL Book-E performance monitoring code so that the 7450 wouldn't be built if the e500 was, and cleaned it up so it was more self-contained. * Added a cpu_setup function for FSL Book-E. The original cpu_setup function prototype had no arguments, assuming that the reg_setup function would copy the required information into variables which represented the registers. This was silly for e500, since it has 1 register per counter (rather than 3 for all counters), so the code has been restructured to have cpu_setup take the current counter config array as an argument, with op_powerpc_setup() invoking op_powerpc_cpu_setup() through on_each_cpu(), and op_powerpc_cpu_setup() invoking the model-specific cpu_setup function with an argument. The argument is ignored on all other platforms at present. * Fixed a confusing line where a trinary operator only had two arguments Signed-off-by: Andrew Fleming Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index 8b133af..7af23c4 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -38,7 +38,6 @@ obj-$(CONFIG_6xx) += idle_6xx.o l2cr_6xx.o cpu_setup_6xx.o obj-$(CONFIG_TAU) += tau_6xx.o obj32-$(CONFIG_SOFTWARE_SUSPEND) += swsusp_32.o obj32-$(CONFIG_MODULES) += module_32.o -obj-$(CONFIG_E500) += perfmon_fsl_booke.o ifeq ($(CONFIG_PPC_MERGE),y) diff --git a/arch/powerpc/kernel/perfmon_fsl_booke.c b/arch/powerpc/kernel/perfmon_fsl_booke.c deleted file mode 100644 index e0dcf2b..0000000 --- a/arch/powerpc/kernel/perfmon_fsl_booke.c +++ /dev/null @@ -1,221 +0,0 @@ -/* arch/powerpc/kernel/perfmon_fsl_booke.c - * Freescale Book-E Performance Monitor code - * - * Author: Andy Fleming - * Copyright (c) 2004 Freescale Semiconductor, Inc - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include - -static inline u32 get_pmlca(int ctr); -static inline void set_pmlca(int ctr, u32 pmlca); - -static inline u32 get_pmlca(int ctr) -{ - u32 pmlca; - - switch (ctr) { - case 0: - pmlca = mfpmr(PMRN_PMLCA0); - break; - case 1: - pmlca = mfpmr(PMRN_PMLCA1); - break; - case 2: - pmlca = mfpmr(PMRN_PMLCA2); - break; - case 3: - pmlca = mfpmr(PMRN_PMLCA3); - break; - default: - panic("Bad ctr number\n"); - } - - return pmlca; -} - -static inline void set_pmlca(int ctr, u32 pmlca) -{ - switch (ctr) { - case 0: - mtpmr(PMRN_PMLCA0, pmlca); - break; - case 1: - mtpmr(PMRN_PMLCA1, pmlca); - break; - case 2: - mtpmr(PMRN_PMLCA2, pmlca); - break; - case 3: - mtpmr(PMRN_PMLCA3, pmlca); - break; - default: - panic("Bad ctr number\n"); - } -} - -void init_pmc_stop(int ctr) -{ - u32 pmlca = (PMLCA_FC | PMLCA_FCS | PMLCA_FCU | - PMLCA_FCM1 | PMLCA_FCM0); - u32 pmlcb = 0; - - switch (ctr) { - case 0: - mtpmr(PMRN_PMLCA0, pmlca); - mtpmr(PMRN_PMLCB0, pmlcb); - break; - case 1: - mtpmr(PMRN_PMLCA1, pmlca); - mtpmr(PMRN_PMLCB1, pmlcb); - break; - case 2: - mtpmr(PMRN_PMLCA2, pmlca); - mtpmr(PMRN_PMLCB2, pmlcb); - break; - case 3: - mtpmr(PMRN_PMLCA3, pmlca); - mtpmr(PMRN_PMLCB3, pmlcb); - break; - default: - panic("Bad ctr number!\n"); - } -} - -void set_pmc_event(int ctr, int event) -{ - u32 pmlca; - - pmlca = get_pmlca(ctr); - - pmlca = (pmlca & ~PMLCA_EVENT_MASK) | - ((event << PMLCA_EVENT_SHIFT) & - PMLCA_EVENT_MASK); - - set_pmlca(ctr, pmlca); -} - -void set_pmc_user_kernel(int ctr, int user, int kernel) -{ - u32 pmlca; - - pmlca = get_pmlca(ctr); - - if(user) - pmlca &= ~PMLCA_FCU; - else - pmlca |= PMLCA_FCU; - - if(kernel) - pmlca &= ~PMLCA_FCS; - else - pmlca |= PMLCA_FCS; - - set_pmlca(ctr, pmlca); -} - -void set_pmc_marked(int ctr, int mark0, int mark1) -{ - u32 pmlca = get_pmlca(ctr); - - if(mark0) - pmlca &= ~PMLCA_FCM0; - else - pmlca |= PMLCA_FCM0; - - if(mark1) - pmlca &= ~PMLCA_FCM1; - else - pmlca |= PMLCA_FCM1; - - set_pmlca(ctr, pmlca); -} - -void pmc_start_ctr(int ctr, int enable) -{ - u32 pmlca = get_pmlca(ctr); - - pmlca &= ~PMLCA_FC; - - if (enable) - pmlca |= PMLCA_CE; - else - pmlca &= ~PMLCA_CE; - - set_pmlca(ctr, pmlca); -} - -void pmc_start_ctrs(int enable) -{ - u32 pmgc0 = mfpmr(PMRN_PMGC0); - - pmgc0 &= ~PMGC0_FAC; - pmgc0 |= PMGC0_FCECE; - - if (enable) - pmgc0 |= PMGC0_PMIE; - else - pmgc0 &= ~PMGC0_PMIE; - - mtpmr(PMRN_PMGC0, pmgc0); -} - -void pmc_stop_ctrs(void) -{ - u32 pmgc0 = mfpmr(PMRN_PMGC0); - - pmgc0 |= PMGC0_FAC; - - pmgc0 &= ~(PMGC0_PMIE | PMGC0_FCECE); - - mtpmr(PMRN_PMGC0, pmgc0); -} - -void dump_pmcs(void) -{ - printk("pmgc0: %x\n", mfpmr(PMRN_PMGC0)); - printk("pmc\t\tpmlca\t\tpmlcb\n"); - printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC0), - mfpmr(PMRN_PMLCA0), mfpmr(PMRN_PMLCB0)); - printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC1), - mfpmr(PMRN_PMLCA1), mfpmr(PMRN_PMLCB1)); - printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC2), - mfpmr(PMRN_PMLCA2), mfpmr(PMRN_PMLCB2)); - printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC3), - mfpmr(PMRN_PMLCA3), mfpmr(PMRN_PMLCB3)); -} - -EXPORT_SYMBOL(init_pmc_stop); -EXPORT_SYMBOL(set_pmc_event); -EXPORT_SYMBOL(set_pmc_user_kernel); -EXPORT_SYMBOL(set_pmc_marked); -EXPORT_SYMBOL(pmc_start_ctr); -EXPORT_SYMBOL(pmc_start_ctrs); -EXPORT_SYMBOL(pmc_stop_ctrs); -EXPORT_SYMBOL(dump_pmcs); diff --git a/arch/powerpc/kernel/pmc.c b/arch/powerpc/kernel/pmc.c index a0a2efa..3d8f6f4 100644 --- a/arch/powerpc/kernel/pmc.c +++ b/arch/powerpc/kernel/pmc.c @@ -71,7 +71,7 @@ int reserve_pmc_hardware(perf_irq_t new_perf_irq) } pmc_owner_caller = __builtin_return_address(0); - perf_irq = new_perf_irq ? : dummy_perf; + perf_irq = new_perf_irq ? new_perf_irq : dummy_perf; out: spin_unlock(&pmc_owner_lock); diff --git a/arch/powerpc/oprofile/Makefile b/arch/powerpc/oprofile/Makefile index 3145d61..0b5df9c 100644 --- a/arch/powerpc/oprofile/Makefile +++ b/arch/powerpc/oprofile/Makefile @@ -13,4 +13,4 @@ DRIVER_OBJS := $(addprefix ../../../drivers/oprofile/, \ oprofile-y := $(DRIVER_OBJS) common.o backtrace.o oprofile-$(CONFIG_PPC64) += op_model_rs64.o op_model_power4.o oprofile-$(CONFIG_FSL_BOOKE) += op_model_fsl_booke.o -oprofile-$(CONFIG_PPC32) += op_model_7450.o +oprofile-$(CONFIG_6xx) += op_model_7450.o diff --git a/arch/powerpc/oprofile/common.c b/arch/powerpc/oprofile/common.c index fd0bbbe..63bbef3 100644 --- a/arch/powerpc/oprofile/common.c +++ b/arch/powerpc/oprofile/common.c @@ -34,6 +34,11 @@ static void op_handle_interrupt(struct pt_regs *regs) model->handle_interrupt(regs, ctr); } +static void op_powerpc_cpu_setup(void *dummy) +{ + model->cpu_setup(ctr); +} + static int op_powerpc_setup(void) { int err; @@ -47,7 +52,7 @@ static int op_powerpc_setup(void) model->reg_setup(ctr, &sys, model->num_counters); /* Configure the registers on all cpus. */ - on_each_cpu(model->cpu_setup, NULL, 0, 1); + on_each_cpu(op_powerpc_cpu_setup, NULL, 0, 1); return 0; } @@ -142,7 +147,8 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) case PPC_OPROFILE_POWER4: model = &op_model_power4; break; -#else +#endif +#ifdef CONFIG_6xx case PPC_OPROFILE_G4: model = &op_model_7450; break; diff --git a/arch/powerpc/oprofile/op_model_7450.c b/arch/powerpc/oprofile/op_model_7450.c index d8ee3ae..f481c0e 100644 --- a/arch/powerpc/oprofile/op_model_7450.c +++ b/arch/powerpc/oprofile/op_model_7450.c @@ -81,7 +81,7 @@ static void pmc_stop_ctrs(void) /* Configures the counters on this CPU based on the global * settings */ -static void fsl7450_cpu_setup(void *unused) +static void fsl7450_cpu_setup(struct op_counter_config *ctr) { /* freeze all counters */ pmc_stop_ctrs(); diff --git a/arch/powerpc/oprofile/op_model_fsl_booke.c b/arch/powerpc/oprofile/op_model_fsl_booke.c index e29dede..0b3c31f 100644 --- a/arch/powerpc/oprofile/op_model_fsl_booke.c +++ b/arch/powerpc/oprofile/op_model_fsl_booke.c @@ -32,42 +32,152 @@ static unsigned long reset_value[OP_MAX_COUNTER]; static int num_counters; static int oprofile_running; -static inline unsigned int ctr_read(unsigned int i) +static void init_pmc_stop(int ctr) { - switch(i) { - case 0: - return mfpmr(PMRN_PMC0); - case 1: - return mfpmr(PMRN_PMC1); - case 2: - return mfpmr(PMRN_PMC2); - case 3: - return mfpmr(PMRN_PMC3); - default: - return 0; - } -} + u32 pmlca = (PMLCA_FC | PMLCA_FCS | PMLCA_FCU | + PMLCA_FCM1 | PMLCA_FCM0); + u32 pmlcb = 0; -static inline void ctr_write(unsigned int i, unsigned int val) -{ - switch(i) { + switch (ctr) { case 0: - mtpmr(PMRN_PMC0, val); + mtpmr(PMRN_PMLCA0, pmlca); + mtpmr(PMRN_PMLCB0, pmlcb); break; case 1: - mtpmr(PMRN_PMC1, val); + mtpmr(PMRN_PMLCA1, pmlca); + mtpmr(PMRN_PMLCB1, pmlcb); break; case 2: - mtpmr(PMRN_PMC2, val); + mtpmr(PMRN_PMLCA2, pmlca); + mtpmr(PMRN_PMLCB2, pmlcb); break; case 3: - mtpmr(PMRN_PMC3, val); + mtpmr(PMRN_PMLCA3, pmlca); + mtpmr(PMRN_PMLCB3, pmlcb); break; default: - break; + panic("Bad ctr number!\n"); } } +static void set_pmc_event(int ctr, int event) +{ + u32 pmlca; + + pmlca = get_pmlca(ctr); + + pmlca = (pmlca & ~PMLCA_EVENT_MASK) | + ((event << PMLCA_EVENT_SHIFT) & + PMLCA_EVENT_MASK); + + set_pmlca(ctr, pmlca); +} + +static void set_pmc_user_kernel(int ctr, int user, int kernel) +{ + u32 pmlca; + + pmlca = get_pmlca(ctr); + + if(user) + pmlca &= ~PMLCA_FCU; + else + pmlca |= PMLCA_FCU; + + if(kernel) + pmlca &= ~PMLCA_FCS; + else + pmlca |= PMLCA_FCS; + + set_pmlca(ctr, pmlca); +} + +static void set_pmc_marked(int ctr, int mark0, int mark1) +{ + u32 pmlca = get_pmlca(ctr); + + if(mark0) + pmlca &= ~PMLCA_FCM0; + else + pmlca |= PMLCA_FCM0; + + if(mark1) + pmlca &= ~PMLCA_FCM1; + else + pmlca |= PMLCA_FCM1; + + set_pmlca(ctr, pmlca); +} + +static void pmc_start_ctr(int ctr, int enable) +{ + u32 pmlca = get_pmlca(ctr); + + pmlca &= ~PMLCA_FC; + + if (enable) + pmlca |= PMLCA_CE; + else + pmlca &= ~PMLCA_CE; + + set_pmlca(ctr, pmlca); +} + +static void pmc_start_ctrs(int enable) +{ + u32 pmgc0 = mfpmr(PMRN_PMGC0); + + pmgc0 &= ~PMGC0_FAC; + pmgc0 |= PMGC0_FCECE; + + if (enable) + pmgc0 |= PMGC0_PMIE; + else + pmgc0 &= ~PMGC0_PMIE; + + mtpmr(PMRN_PMGC0, pmgc0); +} + +static void pmc_stop_ctrs(void) +{ + u32 pmgc0 = mfpmr(PMRN_PMGC0); + + pmgc0 |= PMGC0_FAC; + + pmgc0 &= ~(PMGC0_PMIE | PMGC0_FCECE); + + mtpmr(PMRN_PMGC0, pmgc0); +} + +static void dump_pmcs(void) +{ + printk("pmgc0: %x\n", mfpmr(PMRN_PMGC0)); + printk("pmc\t\tpmlca\t\tpmlcb\n"); + printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC0), + mfpmr(PMRN_PMLCA0), mfpmr(PMRN_PMLCB0)); + printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC1), + mfpmr(PMRN_PMLCA1), mfpmr(PMRN_PMLCB1)); + printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC2), + mfpmr(PMRN_PMLCA2), mfpmr(PMRN_PMLCB2)); + printk("%8x\t%8x\t%8x\n", mfpmr(PMRN_PMC3), + mfpmr(PMRN_PMLCA3), mfpmr(PMRN_PMLCB3)); +} + +static void fsl_booke_cpu_setup(struct op_counter_config *ctr) +{ + int i; + + /* freeze all counters */ + pmc_stop_ctrs(); + + for (i = 0;i < num_counters;i++) { + init_pmc_stop(i); + + set_pmc_event(i, ctr[i].event); + + set_pmc_user_kernel(i, ctr[i].user, ctr[i].kernel); + } +} static void fsl_booke_reg_setup(struct op_counter_config *ctr, struct op_system_config *sys, @@ -77,23 +187,14 @@ static void fsl_booke_reg_setup(struct op_counter_config *ctr, num_counters = num_ctrs; - /* freeze all counters */ - pmc_stop_ctrs(); - /* Our counters count up, and "count" refers to * how much before the next interrupt, and we interrupt * on overflow. So we calculate the starting value * which will give us "count" until overflow. * Then we set the events on the enabled counters */ - for (i = 0; i < num_counters; ++i) { + for (i = 0; i < num_counters; ++i) reset_value[i] = 0x80000000UL - ctr[i].count; - init_pmc_stop(i); - - set_pmc_event(i, ctr[i].event); - - set_pmc_user_kernel(i, ctr[i].user, ctr[i].kernel); - } } static void fsl_booke_start(struct op_counter_config *ctr) @@ -105,8 +206,8 @@ static void fsl_booke_start(struct op_counter_config *ctr) for (i = 0; i < num_counters; ++i) { if (ctr[i].enabled) { ctr_write(i, reset_value[i]); - /* Set Each enabled counterd to only - * count when the Mark bit is not set */ + /* Set each enabled counter to only + * count when the Mark bit is *not* set */ set_pmc_marked(i, 1, 0); pmc_start_ctr(i, 1); } else { @@ -177,6 +278,7 @@ static void fsl_booke_handle_interrupt(struct pt_regs *regs, struct op_powerpc_model op_model_fsl_booke = { .reg_setup = fsl_booke_reg_setup, + .cpu_setup = fsl_booke_cpu_setup, .start = fsl_booke_start, .stop = fsl_booke_stop, .handle_interrupt = fsl_booke_handle_interrupt, diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c index 6a927ef..356709d 100644 --- a/arch/powerpc/oprofile/op_model_power4.c +++ b/arch/powerpc/oprofile/op_model_power4.c @@ -82,7 +82,7 @@ static inline int mmcra_must_set_sample(void) return 0; } -static void power4_cpu_setup(void *unused) +static void power4_cpu_setup(struct op_counter_config *ctr) { unsigned int mmcr0 = mmcr0_val; unsigned long mmcra = mmcra_val; diff --git a/arch/powerpc/oprofile/op_model_rs64.c b/arch/powerpc/oprofile/op_model_rs64.c index 042f8f4..19c5ee0 100644 --- a/arch/powerpc/oprofile/op_model_rs64.c +++ b/arch/powerpc/oprofile/op_model_rs64.c @@ -102,7 +102,7 @@ static void rs64_reg_setup(struct op_counter_config *ctr, /* XXX setup user and kernel profiling */ } -static void rs64_cpu_setup(void *unused) +static void rs64_cpu_setup(struct op_counter_config *ctr) { unsigned int mmcr0; diff --git a/include/asm-powerpc/oprofile_impl.h b/include/asm-powerpc/oprofile_impl.h index 5b33994..07a10e5 100644 --- a/include/asm-powerpc/oprofile_impl.h +++ b/include/asm-powerpc/oprofile_impl.h @@ -42,7 +42,7 @@ struct op_powerpc_model { void (*reg_setup) (struct op_counter_config *, struct op_system_config *, int num_counters); - void (*cpu_setup) (void *); + void (*cpu_setup) (struct op_counter_config *); void (*start) (struct op_counter_config *); void (*stop) (void); void (*handle_interrupt) (struct pt_regs *, @@ -121,7 +121,90 @@ static inline void ctr_write(unsigned int i, unsigned int val) break; } } -#endif /* !CONFIG_FSL_BOOKE */ +#else /* CONFIG_FSL_BOOKE */ +static inline u32 get_pmlca(int ctr) +{ + u32 pmlca; + + switch (ctr) { + case 0: + pmlca = mfpmr(PMRN_PMLCA0); + break; + case 1: + pmlca = mfpmr(PMRN_PMLCA1); + break; + case 2: + pmlca = mfpmr(PMRN_PMLCA2); + break; + case 3: + pmlca = mfpmr(PMRN_PMLCA3); + break; + default: + panic("Bad ctr number\n"); + } + + return pmlca; +} + +static inline void set_pmlca(int ctr, u32 pmlca) +{ + switch (ctr) { + case 0: + mtpmr(PMRN_PMLCA0, pmlca); + break; + case 1: + mtpmr(PMRN_PMLCA1, pmlca); + break; + case 2: + mtpmr(PMRN_PMLCA2, pmlca); + break; + case 3: + mtpmr(PMRN_PMLCA3, pmlca); + break; + default: + panic("Bad ctr number\n"); + } +} + +static inline unsigned int ctr_read(unsigned int i) +{ + switch(i) { + case 0: + return mfpmr(PMRN_PMC0); + case 1: + return mfpmr(PMRN_PMC1); + case 2: + return mfpmr(PMRN_PMC2); + case 3: + return mfpmr(PMRN_PMC3); + default: + return 0; + } +} + +static inline void ctr_write(unsigned int i, unsigned int val) +{ + switch(i) { + case 0: + mtpmr(PMRN_PMC0, val); + break; + case 1: + mtpmr(PMRN_PMC1, val); + break; + case 2: + mtpmr(PMRN_PMC2, val); + break; + case 3: + mtpmr(PMRN_PMC3, val); + break; + default: + break; + } +} + + +#endif /* CONFIG_FSL_BOOKE */ + extern void op_powerpc_backtrace(struct pt_regs * const regs, unsigned int depth); diff --git a/include/asm-powerpc/pmc.h b/include/asm-powerpc/pmc.h index 07d6a42..8588be6 100644 --- a/include/asm-powerpc/pmc.h +++ b/include/asm-powerpc/pmc.h @@ -32,18 +32,5 @@ void release_pmc_hardware(void); void power4_enable_pmcs(void); #endif -#ifdef CONFIG_FSL_BOOKE -void init_pmc_stop(int ctr); -void set_pmc_event(int ctr, int event); -void set_pmc_user_kernel(int ctr, int user, int kernel); -void set_pmc_marked(int ctr, int mark0, int mark1); -void pmc_start_ctr(int ctr, int enable); -void pmc_start_ctrs(int enable); -void pmc_stop_ctrs(void); -void dump_pmcs(void); - -extern struct op_powerpc_model op_model_fsl_booke; -#endif - #endif /* __KERNEL__ */ #endif /* _POWERPC_PMC_H */ -- cgit v0.10.2 From 5d2efba64b231a1733c4048d1708d77e07f26426 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Mon, 30 Oct 2006 16:15:59 +1100 Subject: [POWERPC] Use 4kB iommu pages even on 64kB-page systems The 10Gigabit ethernet device drivers appear to be able to chew up all 256MB of TCE mappings on pSeries systems, as evidenced by numerous error messages: iommu_alloc failed, tbl c0000000010d5c48 vaddr c0000000d875eff0 npages 1 Some experimentation indicates that this is essentially because one 1500 byte ethernet MTU gets mapped as a 64K DMA region when the large 64K pages are enabled. Thus, it doesn't take much to exhaust all of the available DMA mappings for a high-speed card. This patch changes the iommu allocator to work with its own unique, distinct page size. Although the patch is long, its actually quite simple: it just #defines a distinct IOMMU_PAGE_SIZE and then uses this in all the places that matter. As a side effect, it also dramatically improves network performance on platforms with H-calls on iommu translation inserts/removes (since we no longer call it 16 times for a 1500 bytes packet when the iommu HW is still 4k). In the future, we might want to make the IOMMU_PAGE_SIZE a variable in the iommu_table instance, thus allowing support for different HW page sizes in the iommu itself. Signed-off-by: Linas Vepstas Signed-off-by: Benjamin Herrenschmidt Acked-by: Olof Johansson Acked-by: Stephen Rothwell Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c index f88a2a6..ba6b725 100644 --- a/arch/powerpc/kernel/iommu.c +++ b/arch/powerpc/kernel/iommu.c @@ -47,6 +47,17 @@ static int novmerge = 0; static int novmerge = 1; #endif +static inline unsigned long iommu_num_pages(unsigned long vaddr, + unsigned long slen) +{ + unsigned long npages; + + npages = IOMMU_PAGE_ALIGN(vaddr + slen) - (vaddr & IOMMU_PAGE_MASK); + npages >>= IOMMU_PAGE_SHIFT; + + return npages; +} + static int __init setup_iommu(char *str) { if (!strcmp(str, "novmerge")) @@ -178,10 +189,10 @@ static dma_addr_t iommu_alloc(struct iommu_table *tbl, void *page, } entry += tbl->it_offset; /* Offset into real TCE table */ - ret = entry << PAGE_SHIFT; /* Set the return dma address */ + ret = entry << IOMMU_PAGE_SHIFT; /* Set the return dma address */ /* Put the TCEs in the HW table */ - ppc_md.tce_build(tbl, entry, npages, (unsigned long)page & PAGE_MASK, + ppc_md.tce_build(tbl, entry, npages, (unsigned long)page & IOMMU_PAGE_MASK, direction); @@ -203,7 +214,7 @@ static void __iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr, unsigned long entry, free_entry; unsigned long i; - entry = dma_addr >> PAGE_SHIFT; + entry = dma_addr >> IOMMU_PAGE_SHIFT; free_entry = entry - tbl->it_offset; if (((free_entry + npages) > tbl->it_size) || @@ -270,7 +281,7 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, /* Init first segment length for backout at failure */ outs->dma_length = 0; - DBG("mapping %d elements:\n", nelems); + DBG("sg mapping %d elements:\n", nelems); spin_lock_irqsave(&(tbl->it_lock), flags); @@ -285,9 +296,8 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, } /* Allocate iommu entries for that segment */ vaddr = (unsigned long)page_address(s->page) + s->offset; - npages = PAGE_ALIGN(vaddr + slen) - (vaddr & PAGE_MASK); - npages >>= PAGE_SHIFT; - entry = iommu_range_alloc(tbl, npages, &handle, mask >> PAGE_SHIFT, 0); + npages = iommu_num_pages(vaddr, slen); + entry = iommu_range_alloc(tbl, npages, &handle, mask >> IOMMU_PAGE_SHIFT, 0); DBG(" - vaddr: %lx, size: %lx\n", vaddr, slen); @@ -301,14 +311,14 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, /* Convert entry to a dma_addr_t */ entry += tbl->it_offset; - dma_addr = entry << PAGE_SHIFT; - dma_addr |= s->offset; + dma_addr = entry << IOMMU_PAGE_SHIFT; + dma_addr |= (s->offset & ~IOMMU_PAGE_MASK); - DBG(" - %lx pages, entry: %lx, dma_addr: %lx\n", + DBG(" - %lu pages, entry: %lx, dma_addr: %lx\n", npages, entry, dma_addr); /* Insert into HW table */ - ppc_md.tce_build(tbl, entry, npages, vaddr & PAGE_MASK, direction); + ppc_md.tce_build(tbl, entry, npages, vaddr & IOMMU_PAGE_MASK, direction); /* If we are in an open segment, try merging */ if (segstart != s) { @@ -323,7 +333,7 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, DBG(" can't merge, new segment.\n"); } else { outs->dma_length += s->length; - DBG(" merged, new len: %lx\n", outs->dma_length); + DBG(" merged, new len: %ux\n", outs->dma_length); } } @@ -367,9 +377,8 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, if (s->dma_length != 0) { unsigned long vaddr, npages; - vaddr = s->dma_address & PAGE_MASK; - npages = (PAGE_ALIGN(s->dma_address + s->dma_length) - vaddr) - >> PAGE_SHIFT; + vaddr = s->dma_address & IOMMU_PAGE_MASK; + npages = iommu_num_pages(s->dma_address, s->dma_length); __iommu_free(tbl, vaddr, npages); s->dma_address = DMA_ERROR_CODE; s->dma_length = 0; @@ -398,8 +407,7 @@ void iommu_unmap_sg(struct iommu_table *tbl, struct scatterlist *sglist, if (sglist->dma_length == 0) break; - npages = (PAGE_ALIGN(dma_handle + sglist->dma_length) - - (dma_handle & PAGE_MASK)) >> PAGE_SHIFT; + npages = iommu_num_pages(dma_handle,sglist->dma_length); __iommu_free(tbl, dma_handle, npages); sglist++; } @@ -532,12 +540,11 @@ dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr, BUG_ON(direction == DMA_NONE); uaddr = (unsigned long)vaddr; - npages = PAGE_ALIGN(uaddr + size) - (uaddr & PAGE_MASK); - npages >>= PAGE_SHIFT; + npages = iommu_num_pages(uaddr, size); if (tbl) { dma_handle = iommu_alloc(tbl, vaddr, npages, direction, - mask >> PAGE_SHIFT, 0); + mask >> IOMMU_PAGE_SHIFT, 0); if (dma_handle == DMA_ERROR_CODE) { if (printk_ratelimit()) { printk(KERN_INFO "iommu_alloc failed, " @@ -545,7 +552,7 @@ dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr, tbl, vaddr, npages); } } else - dma_handle |= (uaddr & ~PAGE_MASK); + dma_handle |= (uaddr & ~IOMMU_PAGE_MASK); } return dma_handle; @@ -554,11 +561,14 @@ dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr, void iommu_unmap_single(struct iommu_table *tbl, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction) { + unsigned int npages; + BUG_ON(direction == DMA_NONE); - if (tbl) - iommu_free(tbl, dma_handle, (PAGE_ALIGN(dma_handle + size) - - (dma_handle & PAGE_MASK)) >> PAGE_SHIFT); + if (tbl) { + npages = iommu_num_pages(dma_handle, size); + iommu_free(tbl, dma_handle, npages); + } } /* Allocates a contiguous real buffer and creates mappings over it. @@ -570,11 +580,11 @@ void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size, { void *ret = NULL; dma_addr_t mapping; - unsigned int npages, order; + unsigned int order; + unsigned int nio_pages, io_order; struct page *page; size = PAGE_ALIGN(size); - npages = size >> PAGE_SHIFT; order = get_order(size); /* @@ -598,8 +608,10 @@ void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size, memset(ret, 0, size); /* Set up tces to cover the allocated range */ - mapping = iommu_alloc(tbl, ret, npages, DMA_BIDIRECTIONAL, - mask >> PAGE_SHIFT, order); + nio_pages = size >> IOMMU_PAGE_SHIFT; + io_order = get_iommu_order(size); + mapping = iommu_alloc(tbl, ret, nio_pages, DMA_BIDIRECTIONAL, + mask >> IOMMU_PAGE_SHIFT, io_order); if (mapping == DMA_ERROR_CODE) { free_pages((unsigned long)ret, order); return NULL; @@ -611,12 +623,13 @@ void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size, void iommu_free_coherent(struct iommu_table *tbl, size_t size, void *vaddr, dma_addr_t dma_handle) { - unsigned int npages; - if (tbl) { + unsigned int nio_pages; + + size = PAGE_ALIGN(size); + nio_pages = size >> IOMMU_PAGE_SHIFT; + iommu_free(tbl, dma_handle, nio_pages); size = PAGE_ALIGN(size); - npages = size >> PAGE_SHIFT; - iommu_free(tbl, dma_handle, npages); free_pages((unsigned long)vaddr, get_order(size)); } } diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c index cb87e71..ed00787 100644 --- a/arch/powerpc/kernel/vio.c +++ b/arch/powerpc/kernel/vio.c @@ -92,9 +92,9 @@ static struct iommu_table *vio_build_iommu_table(struct vio_dev *dev) &tbl->it_index, &offset, &size); /* TCE table size - measured in tce entries */ - tbl->it_size = size >> PAGE_SHIFT; + tbl->it_size = size >> IOMMU_PAGE_SHIFT; /* offset for VIO should always be 0 */ - tbl->it_offset = offset >> PAGE_SHIFT; + tbl->it_offset = offset >> IOMMU_PAGE_SHIFT; tbl->it_busno = 0; tbl->it_type = TCE_VB; diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c index f4cbbcf..218817d 100644 --- a/arch/powerpc/platforms/iseries/iommu.c +++ b/arch/powerpc/platforms/iseries/iommu.c @@ -43,9 +43,6 @@ static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages, u64 rc; u64 tce, rpn; - index <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - while (npages--) { rpn = virt_to_abs(uaddr) >> TCE_SHIFT; tce = (rpn & TCE_RPN_MASK) << TCE_RPN_SHIFT; @@ -75,9 +72,6 @@ static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages) { u64 rc; - npages <<= TCE_PAGE_FACTOR; - index <<= TCE_PAGE_FACTOR; - while (npages--) { rc = HvCallXm_setTce((u64)tbl->it_index, (u64)index, 0); if (rc) @@ -136,10 +130,9 @@ void iommu_table_getparms_iSeries(unsigned long busno, panic("PCI_DMA: parms->size is zero, parms is 0x%p", parms); /* itc_size is in pages worth of table, it_size is in # of entries */ - tbl->it_size = ((parms->itc_size * TCE_PAGE_SIZE) / - TCE_ENTRY_SIZE) >> TCE_PAGE_FACTOR; + tbl->it_size = (parms->itc_size * TCE_PAGE_SIZE) / TCE_ENTRY_SIZE; tbl->it_busno = parms->itc_busno; - tbl->it_offset = parms->itc_offset >> TCE_PAGE_FACTOR; + tbl->it_offset = parms->itc_offset; tbl->it_index = parms->itc_index; tbl->it_blocksize = 1; tbl->it_type = virtbus ? TCE_VB : TCE_PCI; diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index d24ba54..556c279 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -57,9 +57,6 @@ static void tce_build_pSeries(struct iommu_table *tbl, long index, u64 *tcep; u64 rpn; - index <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - proto_tce = TCE_PCI_READ; // Read allowed if (direction != DMA_TO_DEVICE) @@ -82,9 +79,6 @@ static void tce_free_pSeries(struct iommu_table *tbl, long index, long npages) { u64 *tcep; - npages <<= TCE_PAGE_FACTOR; - index <<= TCE_PAGE_FACTOR; - tcep = ((u64 *)tbl->it_base) + index; while (npages--) @@ -95,7 +89,6 @@ static unsigned long tce_get_pseries(struct iommu_table *tbl, long index) { u64 *tcep; - index <<= TCE_PAGE_FACTOR; tcep = ((u64 *)tbl->it_base) + index; return *tcep; @@ -109,9 +102,6 @@ static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum, u64 proto_tce, tce; u64 rpn; - tcenum <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - rpn = (virt_to_abs(uaddr)) >> TCE_SHIFT; proto_tce = TCE_PCI_READ; if (direction != DMA_TO_DEVICE) @@ -146,7 +136,7 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, u64 rpn; long l, limit; - if (TCE_PAGE_FACTOR == 0 && npages == 1) + if (npages == 1) return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr, direction); @@ -164,9 +154,6 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, __get_cpu_var(tce_page) = tcep; } - tcenum <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - rpn = (virt_to_abs(uaddr)) >> TCE_SHIFT; proto_tce = TCE_PCI_READ; if (direction != DMA_TO_DEVICE) @@ -207,9 +194,6 @@ static void tce_free_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages { u64 rc; - tcenum <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - while (npages--) { rc = plpar_tce_put((u64)tbl->it_index, (u64)tcenum << 12, 0); @@ -229,9 +213,6 @@ static void tce_freemulti_pSeriesLP(struct iommu_table *tbl, long tcenum, long n { u64 rc; - tcenum <<= TCE_PAGE_FACTOR; - npages <<= TCE_PAGE_FACTOR; - rc = plpar_tce_stuff((u64)tbl->it_index, (u64)tcenum << 12, 0, npages); if (rc && printk_ratelimit()) { @@ -248,7 +229,6 @@ static unsigned long tce_get_pSeriesLP(struct iommu_table *tbl, long tcenum) u64 rc; unsigned long tce_ret; - tcenum <<= TCE_PAGE_FACTOR; rc = plpar_tce_get((u64)tbl->it_index, (u64)tcenum << 12, &tce_ret); if (rc && printk_ratelimit()) { @@ -289,7 +269,7 @@ static void iommu_table_setparms(struct pci_controller *phb, tbl->it_busno = phb->bus->number; /* Units of tce entries */ - tbl->it_offset = phb->dma_window_base_cur >> PAGE_SHIFT; + tbl->it_offset = phb->dma_window_base_cur >> IOMMU_PAGE_SHIFT; /* Test if we are going over 2GB of DMA space */ if (phb->dma_window_base_cur + phb->dma_window_size > 0x80000000ul) { @@ -300,7 +280,7 @@ static void iommu_table_setparms(struct pci_controller *phb, phb->dma_window_base_cur += phb->dma_window_size; /* Set the tce table size - measured in entries */ - tbl->it_size = phb->dma_window_size >> PAGE_SHIFT; + tbl->it_size = phb->dma_window_size >> IOMMU_PAGE_SHIFT; tbl->it_index = 0; tbl->it_blocksize = 16; @@ -325,8 +305,8 @@ static void iommu_table_setparms_lpar(struct pci_controller *phb, tbl->it_base = 0; tbl->it_blocksize = 16; tbl->it_type = TCE_PCI; - tbl->it_offset = offset >> PAGE_SHIFT; - tbl->it_size = size >> PAGE_SHIFT; + tbl->it_offset = offset >> IOMMU_PAGE_SHIFT; + tbl->it_size = size >> IOMMU_PAGE_SHIFT; } static void iommu_bus_setup_pSeries(struct pci_bus *bus) @@ -522,8 +502,6 @@ static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev) const void *dma_window = NULL; struct pci_dn *pci; - DBG("iommu_dev_setup_pSeriesLP, dev %p (%s)\n", dev, pci_name(dev)); - /* dev setup for LPAR is a little tricky, since the device tree might * contain the dma-window properties per-device and not neccesarily * for the bus. So we need to search upwards in the tree until we @@ -532,6 +510,9 @@ static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev) */ dn = pci_device_to_OF_node(dev); + DBG("iommu_dev_setup_pSeriesLP, dev %p (%s) %s\n", + dev, pci_name(dev), dn->full_name); + for (pdn = dn; pdn && PCI_DN(pdn) && !PCI_DN(pdn)->iommu_table; pdn = pdn->parent) { dma_window = get_property(pdn, "ibm,dma-window", NULL); diff --git a/arch/powerpc/sysdev/dart.h b/arch/powerpc/sysdev/dart.h index 1c8817c..ff202ed 100644 --- a/arch/powerpc/sysdev/dart.h +++ b/arch/powerpc/sysdev/dart.h @@ -72,7 +72,6 @@ #define DART_PAGE_SHIFT 12 #define DART_PAGE_SIZE (1 << DART_PAGE_SHIFT) -#define DART_PAGE_FACTOR (PAGE_SHIFT - DART_PAGE_SHIFT) #endif /* _POWERPC_SYSDEV_DART_H */ diff --git a/arch/powerpc/sysdev/dart_iommu.c b/arch/powerpc/sysdev/dart_iommu.c index 03b4477..572b784 100644 --- a/arch/powerpc/sysdev/dart_iommu.c +++ b/arch/powerpc/sysdev/dart_iommu.c @@ -156,9 +156,6 @@ static void dart_build(struct iommu_table *tbl, long index, DBG("dart: build at: %lx, %lx, addr: %x\n", index, npages, uaddr); - index <<= DART_PAGE_FACTOR; - npages <<= DART_PAGE_FACTOR; - dp = ((unsigned int*)tbl->it_base) + index; /* On U3, all memory is contigous, so we can move this @@ -199,9 +196,6 @@ static void dart_free(struct iommu_table *tbl, long index, long npages) DBG("dart: free at: %lx, %lx\n", index, npages); - index <<= DART_PAGE_FACTOR; - npages <<= DART_PAGE_FACTOR; - dp = ((unsigned int *)tbl->it_base) + index; while (npages--) @@ -281,7 +275,7 @@ static void iommu_table_dart_setup(void) iommu_table_dart.it_busno = 0; iommu_table_dart.it_offset = 0; /* it_size is in number of entries */ - iommu_table_dart.it_size = (dart_tablesize / sizeof(u32)) >> DART_PAGE_FACTOR; + iommu_table_dart.it_size = dart_tablesize / sizeof(u32); /* Initialize the common IOMMU code */ iommu_table_dart.it_base = (unsigned long)dart_vbase; diff --git a/include/asm-powerpc/iommu.h b/include/asm-powerpc/iommu.h index a5e9864..39fad68 100644 --- a/include/asm-powerpc/iommu.h +++ b/include/asm-powerpc/iommu.h @@ -22,17 +22,35 @@ #define _ASM_IOMMU_H #ifdef __KERNEL__ -#include +#include #include #include #include +#include +#include + +#define IOMMU_PAGE_SHIFT 12 +#define IOMMU_PAGE_SIZE (ASM_CONST(1) << IOMMU_PAGE_SHIFT) +#define IOMMU_PAGE_MASK (~((1 << IOMMU_PAGE_SHIFT) - 1)) +#define IOMMU_PAGE_ALIGN(addr) _ALIGN_UP(addr, IOMMU_PAGE_SIZE) + +#ifndef __ASSEMBLY__ + +/* Pure 2^n version of get_order */ +static __inline__ __attribute_const__ int get_iommu_order(unsigned long size) +{ + return __ilog2((size - 1) >> IOMMU_PAGE_SHIFT) + 1; +} + +#endif /* __ASSEMBLY__ */ + /* * IOMAP_MAX_ORDER defines the largest contiguous block * of dma space we can get. IOMAP_MAX_ORDER = 13 * allows up to 2**12 pages (4096 * 4096) = 16 MB */ -#define IOMAP_MAX_ORDER 13 +#define IOMAP_MAX_ORDER 13 struct iommu_table { unsigned long it_busno; /* Bus number this table belongs to */ diff --git a/include/asm-powerpc/tce.h b/include/asm-powerpc/tce.h index c9483ad..f663634 100644 --- a/include/asm-powerpc/tce.h +++ b/include/asm-powerpc/tce.h @@ -22,6 +22,8 @@ #define _ASM_POWERPC_TCE_H #ifdef __KERNEL__ +#include + /* * Tces come in two formats, one for the virtual bus and a different * format for PCI @@ -33,7 +35,6 @@ #define TCE_SHIFT 12 #define TCE_PAGE_SIZE (1 << TCE_SHIFT) -#define TCE_PAGE_FACTOR (PAGE_SHIFT - TCE_SHIFT) #define TCE_ENTRY_SIZE 8 /* each TCE is 64 bits */ -- cgit v0.10.2 From 302439d2167e0f1e01a6480ac40c06063f4e16a1 Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Tue, 31 Oct 2006 17:53:42 +0800 Subject: [POWERPC] qe_lib: qe_issue_cmd writes wrong value to CECDR Changed qe_issue_cmd() to write cmd_input to the CECDR unmodified. It was treating cmd_input as a virtual address and tried to convert it to a physical address. Signed-off-by: Timur Tabi Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/sysdev/qe_lib/qe.c b/arch/powerpc/sysdev/qe_lib/qe.c index 2bae632..e422322 100644 --- a/arch/powerpc/sysdev/qe_lib/qe.c +++ b/arch/powerpc/sysdev/qe_lib/qe.c @@ -122,8 +122,7 @@ int qe_issue_cmd(u32 cmd, u32 device, u8 mcn_protocol, u32 cmd_input) mcn_shift = QE_CR_MCN_NORMAL_SHIFT; } - out_be32(&qe_immr->cp.cecdr, - immrbar_virt_to_phys((void *)cmd_input)); + out_be32(&qe_immr->cp.cecdr, cmd_input); out_be32(&qe_immr->cp.cecr, (cmd | QE_CR_FLG | ((u32) device << dev_shift) | (u32) mcn_protocol << mcn_shift)); -- cgit v0.10.2 From 5fe8e8b88e68e517637e3f8287f1fee89e2d9252 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 31 Oct 2006 18:39:31 +0000 Subject: [POWERPC] Make current preempt-safe Repeated -j20 kernel builds on a G5 Quad running an SMP PREEMPT kernel would often collapse within a day, some exec failing with "Bad address". In each case examined, load_elf_binary was doing a kernel_read, but generic_file_aio_read's access_ok saw current->thread.fs.seg as USER_DS instead of KERNEL_DS. objdump of filemap.o shows gcc 4.1.0 emitting "mr r5,r13 ... ld r9,416(r5)" here for get_paca()->__current, instead of the expected and much more usual "ld r9,416(r13)"; I've seen other gcc4s do the same, but perhaps not gcc3s. So, if the task is preempted and rescheduled on a different cpu in between the mr and the ld, r5 will be looking at a different paca_struct from the one it's now on, pick up the wrong __current, and perhaps the wrong seg. Presumably much worse could happen elsewhere, though that split is rare. Other architectures appear to be safe (x86_64's read_pda is more limiting than get_paca), but ppc64 needs to force "current" into one instruction. Signed-off-by: Hugh Dickins Signed-off-by: Paul Mackerras diff --git a/include/asm-powerpc/current.h b/include/asm-powerpc/current.h index 1938d6ab..b8708ae 100644 --- a/include/asm-powerpc/current.h +++ b/include/asm-powerpc/current.h @@ -14,7 +14,17 @@ struct task_struct; #ifdef __powerpc64__ #include -#define current (get_paca()->__current) +static inline struct task_struct *get_current(void) +{ + struct task_struct *task; + + __asm__ __volatile__("ld %0,%1(13)" + : "=r" (task) + : "i" (offsetof(struct paca_struct, __current))); + + return task; +} +#define current get_current() #else -- cgit v0.10.2 From 96268889ee369b36203b7a06e8aabb197270216e Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 31 Oct 2006 18:40:39 +0000 Subject: [POWERPC] Make high hugepage areas preempt safe Checking source for other get_paca()->field preemption dangers found that open_high_hpage_areas does a structure copy into its paca while preemption is enabled: unsafe however gcc accomplishes it. Just remove that copy: it's done safely afterwards by on_each_cpu, as in open_low_hpage_areas. Signed-off-by: Hugh Dickins Acked-by: David Gibson Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 5615acc..fd68b74 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -480,9 +480,6 @@ static int open_high_hpage_areas(struct mm_struct *mm, u16 newareas) mm->context.high_htlb_areas |= newareas; - /* update the paca copy of the context struct */ - get_paca()->context = mm->context; - /* the context change must make it to memory before the flush, * so that further SLB misses do the right thing. */ mb(); -- cgit v0.10.2 From 292f86f005e3867277b2126c2399eea3e773a4fc Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 31 Oct 2006 18:41:51 +0000 Subject: [POWERPC] Make mmiowb's io_sync preempt safe If mmiowb() is always used prior to releasing spinlock as Doc suggests, then it's safe against preemption; but I'm not convinced that's always the case. If preemption occurs between sync and get_paca()->io_sync = 0, I believe there's no problem. But in the unlikely event that gcc does the store relative to another register than r13 (as it did with current), then there's a small danger of setting another cpu's io_sync to 0, after it had just set it to 1. Rewrite ppc64 mmiowb to prevent that. The remaining io_sync assignments in io.h all get_paca()->io_sync = 1, which is harmless even if preempted to the wrong cpu (the context switch itself syncs); and those in spinlock.h are while preemption is disabled. Signed-off-by: Hugh Dickins Signed-off-by: Paul Mackerras diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h index 3baff8b..c2c5f14 100644 --- a/include/asm-powerpc/io.h +++ b/include/asm-powerpc/io.h @@ -163,8 +163,11 @@ extern void _outsl_ns(volatile u32 __iomem *port, const void *buf, long count); static inline void mmiowb(void) { - __asm__ __volatile__ ("sync" : : : "memory"); - get_paca()->io_sync = 0; + unsigned long tmp; + + __asm__ __volatile__("sync; li %0,0; stb %0,%1(13)" + : "=&r" (tmp) : "i" (offsetof(struct paca_struct, io_sync)) + : "memory"); } /* -- cgit v0.10.2 From 0d69a052d4d7c4085706b9ac0d1bd28ff90c9fca Mon Sep 17 00:00:00 2001 From: "Gui,Jian" Date: Wed, 1 Nov 2006 10:50:15 +0800 Subject: [POWERPC] Disallow kprobes on emulate_step and branch_taken On powerpc, probing on emulate_step function will crash 2.6.18.1 when it is triggered. When kprobe is triggered, emulate_step() is on its kernel path and will cause recursive kprobe fault. And branch_taken() is called in emulate_step(). This disallows kprobes on both of them. Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c index 9590ba7..7e8ded0 100644 --- a/arch/powerpc/lib/sstep.c +++ b/arch/powerpc/lib/sstep.c @@ -9,6 +9,7 @@ * 2 of the License, or (at your option) any later version. */ #include +#include #include #include #include @@ -25,7 +26,7 @@ extern char system_call_common[]; /* * Determine whether a conditional branch instruction would branch. */ -static int branch_taken(unsigned int instr, struct pt_regs *regs) +static int __kprobes branch_taken(unsigned int instr, struct pt_regs *regs) { unsigned int bo = (instr >> 21) & 0x1f; unsigned int bi; @@ -51,7 +52,7 @@ static int branch_taken(unsigned int instr, struct pt_regs *regs) * or -1 if the instruction is one that should not be stepped, * such as an rfid, or a mtmsrd that would clear MSR_RI. */ -int emulate_step(struct pt_regs *regs, unsigned int instr) +int __kprobes emulate_step(struct pt_regs *regs, unsigned int instr) { unsigned int opcode, rd; unsigned long int imm; -- cgit v0.10.2 From 4393c4f6788cee65095dd838cfeca6edefbfeb52 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 1 Nov 2006 15:11:39 +1100 Subject: [POWERPC] Make alignment exception always check exception table The alignment exception used to only check the exception table for -EFAULT, not for other errors. That opens an oops window if we can coerce the kernel into getting an alignment exception for other reasons in what would normally be a user-protected accessor, which can be done via some of the futex ops. This fixes it by always checking the exception tables. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 5ed4c2c..c66b477 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -843,7 +843,7 @@ void __kprobes program_check_exception(struct pt_regs *regs) void alignment_exception(struct pt_regs *regs) { - int fixed = 0; + int sig, code, fixed = 0; /* we don't implement logging of alignment exceptions */ if (!(current->thread.align_ctl & PR_UNALIGN_SIGBUS)) @@ -857,14 +857,16 @@ void alignment_exception(struct pt_regs *regs) /* Operand address was bad */ if (fixed == -EFAULT) { - if (user_mode(regs)) - _exception(SIGSEGV, regs, SEGV_ACCERR, regs->dar); - else - /* Search exception table */ - bad_page_fault(regs, regs->dar, SIGSEGV); - return; + sig = SIGSEGV; + code = SEGV_ACCERR; + } else { + sig = SIGBUS; + code = BUS_ADRALN; } - _exception(SIGBUS, regs, BUS_ADRALN, regs->dar); + if (user_mode(regs)) + _exception(sig, regs, code, regs->dar); + else + bad_page_fault(regs, regs->dar, sig); } void StackOverflow(struct pt_regs *regs) diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c index aafc8e8..9661a91 100644 --- a/arch/ppc/kernel/traps.c +++ b/arch/ppc/kernel/traps.c @@ -708,7 +708,7 @@ void single_step_exception(struct pt_regs *regs) void alignment_exception(struct pt_regs *regs) { - int fixed; + int sig, code, fixed = 0; fixed = fix_alignment(regs); if (fixed == 1) { @@ -717,14 +717,16 @@ void alignment_exception(struct pt_regs *regs) return; } if (fixed == -EFAULT) { - /* fixed == -EFAULT means the operand address was bad */ - if (user_mode(regs)) - _exception(SIGSEGV, regs, SEGV_ACCERR, regs->dar); - else - bad_page_fault(regs, regs->dar, SIGSEGV); - return; + sig = SIGSEGV; + code = SEGV_ACCERR; + } else { + sig = SIGBUS; + code = BUS_ADRALN; } - _exception(SIGBUS, regs, BUS_ADRALN, regs->dar); + if (user_mode(regs)) + _exception(sig, regs, code, regs->dar); + else + bad_page_fault(regs, regs->dar, sig); } void StackOverflow(struct pt_regs *regs) -- cgit v0.10.2 From 1244a19cde42c268aa159d264fc2df072a3ff82f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 1 Nov 2006 17:19:18 +0900 Subject: [PATCH] ahci: fix status register check in ahci_softreset ahci_softreset() used to use ahci_tf_read() which reads D2H_REG area to check for the Status register. However, this area is zeroed on initialization and not set by initial signature FIS. Replace it with ahci_check_status(). This bug prevented CLO code from being activated whenever BSY and/or DRQ is set prior to softreset. This fix makes AHCI_FLAG_RESET_NEEDS_CLO flag redundant. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index cef2e70..988f8bb 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -736,8 +736,7 @@ static int ahci_softreset(struct ata_port *ap, unsigned int *class) } /* check BUSY/DRQ, perform Command List Override if necessary */ - ahci_tf_read(ap, &tf); - if (tf.command & (ATA_BUSY | ATA_DRQ)) { + if (ahci_check_status(ap) & (ATA_BUSY | ATA_DRQ)) { rc = ahci_clo(ap); if (rc == -EOPNOTSUPP) { -- cgit v0.10.2 From 8fc2d9cae99e47e236cb7b77015b9faf69a097cc Mon Sep 17 00:00:00 2001 From: Peer Chen Date: Wed, 1 Nov 2006 05:23:11 -0500 Subject: [libata] sata_nv: Add PCI IDs Signed-off-by: Jeff Garzik diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c index 323b607..d65ebfd 100644 --- a/drivers/ata/sata_nv.c +++ b/drivers/ata/sata_nv.c @@ -117,10 +117,14 @@ static const struct pci_device_id nv_pci_tbl[] = { { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_SATA), GENERIC }, { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_SATA2), GENERIC }, { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_SATA3), GENERIC }, - { PCI_VDEVICE(NVIDIA, 0x045c), GENERIC }, - { PCI_VDEVICE(NVIDIA, 0x045d), GENERIC }, - { PCI_VDEVICE(NVIDIA, 0x045e), GENERIC }, - { PCI_VDEVICE(NVIDIA, 0x045f), GENERIC }, + { PCI_VDEVICE(NVIDIA, 0x045c), GENERIC }, /* MCP65 */ + { PCI_VDEVICE(NVIDIA, 0x045d), GENERIC }, /* MCP65 */ + { PCI_VDEVICE(NVIDIA, 0x045e), GENERIC }, /* MCP65 */ + { PCI_VDEVICE(NVIDIA, 0x045f), GENERIC }, /* MCP65 */ + { PCI_VDEVICE(NVIDIA, 0x0550), GENERIC }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0551), GENERIC }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0552), GENERIC }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0553), GENERIC }, /* MCP67 */ { PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE<<8, 0xffff00, GENERIC }, -- cgit v0.10.2 From 130fe05dbc0114609cfef9815c0c5580b42decfa Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 1 Nov 2006 09:11:00 -0800 Subject: i386: clean up io-apic accesses This is preparation for fixing the ordering of the accesses that got broken by the commit cf4c6a2f27f5db810b69dcb1da7f194489e8ff88 when factoring out the "common" io apic routing entry accesses. Move the accessor function (that were only used by io_apic.c) out of a header file, and use proper memory-mapped accesses rather than making up our own "volatile" pointers. Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 350192d..eb10bd5 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -91,6 +91,46 @@ static struct irq_pin_list { int apic, pin, next; } irq_2_pin[PIN_MAP_SIZE]; +struct io_apic { + unsigned int index; + unsigned int unused[3]; + unsigned int data; +}; + +static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) +{ + return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) + + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); +} + +static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + return readl(&io_apic->data); +} + +static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + +/* + * Re-write a value: to be used for read-modify-write + * cycles where the read already set up the index register. + * + * Older SiS APIC requires we rewrite the index register + */ +static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) +{ + volatile struct io_apic *io_apic = io_apic_base(apic); + if (sis_apic_bug) + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + union entry_union { struct { u32 w1, w2; }; struct IO_APIC_route_entry entry; diff --git a/include/asm-i386/io_apic.h b/include/asm-i386/io_apic.h index 276ea7e..059a9ff 100644 --- a/include/asm-i386/io_apic.h +++ b/include/asm-i386/io_apic.h @@ -12,10 +12,6 @@ #ifdef CONFIG_X86_IO_APIC -#define IO_APIC_BASE(idx) \ - ((volatile int *)(__fix_to_virt(FIX_IO_APIC_BASE_0 + idx) \ - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK))) - /* * The structure of the IO-APIC: */ @@ -119,31 +115,8 @@ extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* non-0 if default (table-less) MP configuration */ extern int mpc_default_type; -static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) -{ - *IO_APIC_BASE(apic) = reg; - return *(IO_APIC_BASE(apic)+4); -} - -static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) -{ - *IO_APIC_BASE(apic) = reg; - *(IO_APIC_BASE(apic)+4) = value; -} - -/* - * Re-write a value: to be used for read-modify-write - * cycles where the read already set up the index register. - * - * Older SiS APIC requires we rewrite the index regiser - */ +/* Older SiS APIC requires we rewrite the index register */ extern int sis_apic_bug; -static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) -{ - if (sis_apic_bug) - *IO_APIC_BASE(apic) = reg; - *(IO_APIC_BASE(apic)+4) = value; -} /* 1 if "noapic" boot option passed */ extern int skip_ioapic_setup; -- cgit v0.10.2 From 242954b5aa8e5ec84f46a84637daf08ee4247c6e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 24 Oct 2006 02:29:01 +0100 Subject: [MIPS] 16K & 64K page size fixes Derived from Peter Watkins 's work. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/asm-offsets.c b/arch/mips/kernel/asm-offsets.c index e9ce5b3..ff88b06 100644 --- a/arch/mips/kernel/asm-offsets.c +++ b/arch/mips/kernel/asm-offsets.c @@ -22,7 +22,7 @@ #define offset(string, ptr, member) \ __asm__("\n@@@" string "%0" : : "i" (_offset(ptr, member))) #define constant(string, member) \ - __asm__("\n@@@" string "%x0" : : "ri" (member)) + __asm__("\n@@@" string "%X0" : : "ri" (member)) #define size(string, size) \ __asm__("\n@@@" string "%0" : : "i" (sizeof(size))) #define linefeed text("") diff --git a/arch/mips/kernel/head.S b/arch/mips/kernel/head.S index 8c6db0f..ddc1b71 100644 --- a/arch/mips/kernel/head.S +++ b/arch/mips/kernel/head.S @@ -189,7 +189,8 @@ NESTED(kernel_entry, 16, sp) # kernel entry point MTC0 zero, CP0_CONTEXT # clear context register PTR_LA $28, init_thread_union - PTR_ADDIU sp, $28, _THREAD_SIZE - 32 + PTR_LI sp, _THREAD_SIZE - 32 + PTR_ADDU sp, $28 set_saved_sp sp, t0, t1 PTR_SUBU sp, 4 * SZREG # init stack pointer diff --git a/arch/mips/kernel/r4k_switch.S b/arch/mips/kernel/r4k_switch.S index d5c8b82..cc566cf 100644 --- a/arch/mips/kernel/r4k_switch.S +++ b/arch/mips/kernel/r4k_switch.S @@ -85,7 +85,12 @@ move $28, a2 cpu_restore_nonscratch a1 +#if (_THREAD_SIZE - 32) < 0x10000 PTR_ADDIU t0, $28, _THREAD_SIZE - 32 +#else + PTR_LI t0, _THREAD_SIZE - 32 + PTR_ADDU t0, $28 +#endif set_saved_sp t0, t1, t2 #ifdef CONFIG_MIPS_MT_SMTC /* Read-modify-writes of Status must be atomic on a VPE */ diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S index 25ed333..79f0317 100644 --- a/arch/mips/kernel/vmlinux.lds.S +++ b/arch/mips/kernel/vmlinux.lds.S @@ -50,6 +50,16 @@ SECTIONS /* writeable */ .data : { /* Data */ . = . + DATAOFFSET; /* for CONFIG_MAPPED_KERNEL */ + /* + * This ALIGN is needed as a workaround for a bug a gcc bug upto 4.1 which + * limits the maximum alignment to at most 32kB and results in the following + * warning: + * + * CC arch/mips/kernel/init_task.o + * arch/mips/kernel/init_task.c:30: warning: alignment of ‘init_thread_union’ + * is greater than maximum object file alignment. Using 32768 + */ + . = ALIGN(_PAGE_SIZE); *(.data.init_task) *(.data) diff --git a/arch/mips/lib-64/dump_tlb.c b/arch/mips/lib-64/dump_tlb.c index be8261b..594df1a 100644 --- a/arch/mips/lib-64/dump_tlb.c +++ b/arch/mips/lib-64/dump_tlb.c @@ -149,7 +149,7 @@ void dump_list_process(struct task_struct *t, void *address) printk("Addr == %08lx\n", addr); printk("tasks->mm.pgd == %08lx\n", (unsigned long) t->mm->pgd); - page_dir = pgd_offset(t->mm, 0); + page_dir = pgd_offset(t->mm, 0UL); printk("page_dir == %016lx\n", (unsigned long) page_dir); pgd = pgd_offset(t->mm, addr); @@ -184,13 +184,13 @@ void dump_list_current(void *address) dump_list_process(current, address); } -unsigned int vtop(void *address) +unsigned long vtop(void *address) { pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t *pte; - unsigned int addr, paddr; + unsigned long addr, paddr; addr = (unsigned long) address; pgd = pgd_offset(current->mm, addr); diff --git a/arch/mips/mips-boards/generic/memory.c b/arch/mips/mips-boards/generic/memory.c index be80c5d..eeed944 100644 --- a/arch/mips/mips-boards/generic/memory.c +++ b/arch/mips/mips-boards/generic/memory.c @@ -176,7 +176,7 @@ unsigned long __init prom_free_prom_memory(void) if (boot_mem_map.map[i].type != BOOT_MEM_ROM_DATA) continue; - addr = boot_mem_map.map[i].addr; + addr = PAGE_ALIGN(boot_mem_map.map[i].addr); while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); diff --git a/arch/mips/mm/pg-r4k.c b/arch/mips/mm/pg-r4k.c index b7c7492..d41fc58 100644 --- a/arch/mips/mm/pg-r4k.c +++ b/arch/mips/mm/pg-r4k.c @@ -270,6 +270,20 @@ static inline void build_addiu_a2_a0(unsigned long offset) emit_instruction(mi); } +static inline void build_addiu_a2(unsigned long offset) +{ + union mips_instruction mi; + + BUG_ON(offset > 0x7fff); + + mi.i_format.opcode = cpu_has_64bit_gp_regs ? daddiu_op : addiu_op; + mi.i_format.rs = 6; /* $a2 */ + mi.i_format.rt = 6; /* $a2 */ + mi.i_format.simmediate = offset; + + emit_instruction(mi); +} + static inline void build_addiu_a1(unsigned long offset) { union mips_instruction mi; @@ -333,6 +347,7 @@ static inline void build_jr_ra(void) void __init build_clear_page(void) { unsigned int loop_start; + unsigned long off; epc = (unsigned int *) &clear_page_array; instruction_pending = 0; @@ -369,7 +384,12 @@ void __init build_clear_page(void) } } - build_addiu_a2_a0(PAGE_SIZE - (cpu_has_prefetch ? pref_offset_clear : 0)); + off = PAGE_SIZE - (cpu_has_prefetch ? pref_offset_clear : 0); + if (off > 0x7fff) { + build_addiu_a2_a0(off >> 1); + build_addiu_a2(off >> 1); + } else + build_addiu_a2_a0(off); if (R4600_V2_HIT_CACHEOP_WAR && cpu_is_r4600_v2_x()) build_insn_word(0x3c01a000); /* lui $at, 0xa000 */ @@ -420,12 +440,18 @@ dest = label(); void __init build_copy_page(void) { unsigned int loop_start; + unsigned long off; epc = (unsigned int *) ©_page_array; store_offset = load_offset = 0; instruction_pending = 0; - build_addiu_a2_a0(PAGE_SIZE - (cpu_has_prefetch ? pref_offset_copy : 0)); + off = PAGE_SIZE - (cpu_has_prefetch ? pref_offset_copy : 0); + if (off > 0x7fff) { + build_addiu_a2_a0(off >> 1); + build_addiu_a2(off >> 1); + } else + build_addiu_a2_a0(off); if (R4600_V2_HIT_CACHEOP_WAR && cpu_is_r4600_v2_x()) build_insn_word(0x3c01a000); /* lui $at, 0xa000 */ diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c index 6f8b25c..fec318a 100644 --- a/arch/mips/mm/tlbex.c +++ b/arch/mips/mm/tlbex.c @@ -102,7 +102,7 @@ enum opcode { insn_addu, insn_addiu, insn_and, insn_andi, insn_beq, insn_beql, insn_bgez, insn_bgezl, insn_bltz, insn_bltzl, insn_bne, insn_daddu, insn_daddiu, insn_dmfc0, insn_dmtc0, - insn_dsll, insn_dsll32, insn_dsra, insn_dsrl, + insn_dsll, insn_dsll32, insn_dsra, insn_dsrl, insn_dsrl32, insn_dsubu, insn_eret, insn_j, insn_jal, insn_jr, insn_ld, insn_ll, insn_lld, insn_lui, insn_lw, insn_mfc0, insn_mtc0, insn_ori, insn_rfe, insn_sc, insn_scd, insn_sd, insn_sll, @@ -145,6 +145,7 @@ static __initdata struct insn insn_table[] = { { insn_dsll32, M(spec_op,0,0,0,0,dsll32_op), RT | RD | RE }, { insn_dsra, M(spec_op,0,0,0,0,dsra_op), RT | RD | RE }, { insn_dsrl, M(spec_op,0,0,0,0,dsrl_op), RT | RD | RE }, + { insn_dsrl32, M(spec_op,0,0,0,0,dsrl32_op), RT | RD | RE }, { insn_dsubu, M(spec_op,0,0,0,0,dsubu_op), RS | RT | RD }, { insn_eret, M(cop0_op,cop_op,0,0,0,eret_op), 0 }, { insn_j, M(j_op,0,0,0,0,0), JIMM }, @@ -385,6 +386,7 @@ I_u2u1u3(_dsll); I_u2u1u3(_dsll32); I_u2u1u3(_dsra); I_u2u1u3(_dsrl); +I_u2u1u3(_dsrl32); I_u3u1u2(_dsubu); I_0(_eret); I_u1(_j); @@ -996,7 +998,12 @@ build_get_pmde64(u32 **p, struct label **l, struct reloc **r, #endif l_vmalloc_done(l, *p); - i_dsrl(p, tmp, tmp, PGDIR_SHIFT-3); /* get pgd offset in bytes */ + + if (PGDIR_SHIFT - 3 < 32) /* get pgd offset in bytes */ + i_dsrl(p, tmp, tmp, PGDIR_SHIFT-3); + else + i_dsrl32(p, tmp, tmp, PGDIR_SHIFT - 3 - 32); + i_andi(p, tmp, tmp, (PTRS_PER_PGD - 1)<<3); i_daddu(p, ptr, ptr, tmp); /* add in pgd offset */ i_dmfc0(p, tmp, C0_BADVADDR); /* get faulting address */ @@ -1073,7 +1080,7 @@ build_get_pgde32(u32 **p, unsigned int tmp, unsigned int ptr) static __init void build_adjust_context(u32 **p, unsigned int ctx) { - unsigned int shift = 4 - (PTE_T_LOG2 + 1); + unsigned int shift = 4 - (PTE_T_LOG2 + 1) + PAGE_SHIFT - 12; unsigned int mask = (PTRS_PER_PTE / 2 - 1) << (PTE_T_LOG2 + 1); switch (current_cpu_data.cputype) { diff --git a/include/asm-mips/asm.h b/include/asm-mips/asm.h index e3038a4..838eb31 100644 --- a/include/asm-mips/asm.h +++ b/include/asm-mips/asm.h @@ -344,6 +344,7 @@ symbol = value #define PTR_L lw #define PTR_S sw #define PTR_LA la +#define PTR_LI li #define PTR_SLL sll #define PTR_SLLV sllv #define PTR_SRL srl @@ -368,6 +369,7 @@ symbol = value #define PTR_L ld #define PTR_S sd #define PTR_LA dla +#define PTR_LI dli #define PTR_SLL dsll #define PTR_SLLV dsllv #define PTR_SRL dsrl diff --git a/include/asm-mips/pgalloc.h b/include/asm-mips/pgalloc.h index 582c1fe..af121c6 100644 --- a/include/asm-mips/pgalloc.h +++ b/include/asm-mips/pgalloc.h @@ -48,7 +48,7 @@ static inline pgd_t *pgd_alloc(struct mm_struct *mm) ret = (pgd_t *) __get_free_pages(GFP_KERNEL, PGD_ORDER); if (ret) { - init = pgd_offset(&init_mm, 0); + init = pgd_offset(&init_mm, 0UL); pgd_init((unsigned long)ret); memcpy(ret + USER_PTRS_PER_PGD, init + USER_PTRS_PER_PGD, (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t)); diff --git a/include/asm-mips/pgtable-64.h b/include/asm-mips/pgtable-64.h index d05fb6f..7e73203 100644 --- a/include/asm-mips/pgtable-64.h +++ b/include/asm-mips/pgtable-64.h @@ -174,7 +174,7 @@ static inline void pud_clear(pud_t *pudp) #define __pmd_offset(address) pmd_index(address) /* to find an entry in a kernel page-table-directory */ -#define pgd_offset_k(address) pgd_offset(&init_mm, 0) +#define pgd_offset_k(address) pgd_offset(&init_mm, 0UL) #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1)) #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)) -- cgit v0.10.2 From cb56837ea5f15fa5279fd490f292134c3a92e5de Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 31 Oct 2006 22:49:04 +0000 Subject: [MIPS] SMTC: Fix crash if # of TC's > # of VPE's after pt_regs irq cleanup. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/entry.S b/arch/mips/kernel/entry.S index 417c08a..f10b6a1 100644 --- a/arch/mips/kernel/entry.S +++ b/arch/mips/kernel/entry.S @@ -83,7 +83,10 @@ FEXPORT(syscall_exit) FEXPORT(restore_all) # restore full frame #ifdef CONFIG_MIPS_MT_SMTC /* Detect and execute deferred IPI "interrupts" */ + LONG_L s0, TI_REGS($28) + LONG_S sp, TI_REGS($28) jal deferred_smtc_ipi + LONG_S s0, TI_REGS($28) /* Re-arm any temporarily masked interrupts not explicitly "acked" */ mfc0 v0, CP0_TCSTATUS ori v1, v0, TCSTATUS_IXMT diff --git a/arch/mips/kernel/smtc-asm.S b/arch/mips/kernel/smtc-asm.S index 1cb9441..921207c 100644 --- a/arch/mips/kernel/smtc-asm.S +++ b/arch/mips/kernel/smtc-asm.S @@ -101,7 +101,9 @@ FEXPORT(__smtc_ipi_vector) lw t0,PT_PADSLOT5(sp) /* Argument from sender passed in stack pad slot 4 */ lw a0,PT_PADSLOT4(sp) - PTR_LA ra, _ret_from_irq + LONG_L s0, TI_REGS($28) + LONG_S sp, TI_REGS($28) + PTR_LA ra, ret_from_irq jr t0 /* @@ -119,7 +121,10 @@ LEAF(self_ipi) subu t1,sp,PT_SIZE sw ra,PT_EPC(t1) sw a0,PT_PADSLOT4(t1) + LONG_L s0, TI_REGS($28) + LONG_S sp, TI_REGS($28) la t2,ipi_decode + LONG_S s0, TI_REGS($28) sw t2,PT_PADSLOT5(t1) /* Save pre-disable value of TCStatus */ sw t0,PT_TCSTATUS(t1) -- cgit v0.10.2 From 64c590b7a62ae1272fe4afd7b915de314591f35e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 1 Nov 2006 00:22:00 +0000 Subject: [MIPS] SMTC: Synchronize cp0 counters on bootup. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index cc1f747..3b78caf 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -476,6 +476,7 @@ void mipsmt_prepare_cpus(void) write_vpe_c0_compare(0); /* Propagate Config7 */ write_vpe_c0_config7(read_c0_config7()); + write_vpe_c0_count(read_c0_count()); } /* enable multi-threading within VPE */ write_vpe_c0_vpecontrol(read_vpe_c0_vpecontrol() | VPECONTROL_TE); -- cgit v0.10.2 From 67cac1eba7eee92e2b25c1e8e4737968dc7c8522 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 1 Nov 2006 18:55:22 +0900 Subject: [MIPS] Fix warning in mips-boards generic PCI arch/mips/mips-boards/generic/pci.c: In function `mips_pcibios_init': arch/mips/mips-boards/generic/pci.c:227: warning: comparison of distinct pointer types lacks a cast arch/mips/mips-boards/generic/pci.c:228: warning: comparison of distinct pointer types lacks a cast Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle diff --git a/arch/mips/mips-boards/generic/pci.c b/arch/mips/mips-boards/generic/pci.c index 9337f6c..3192a14 100644 --- a/arch/mips/mips-boards/generic/pci.c +++ b/arch/mips/mips-boards/generic/pci.c @@ -90,7 +90,7 @@ static struct pci_controller msc_controller = { void __init mips_pcibios_init(void) { struct pci_controller *controller; - unsigned long start, end, map, start1, end1, map1, map2, map3, mask; + resource_size_t start, end, map, start1, end1, map1, map2, map3, mask; switch (mips_revision_corid) { case MIPS_REVISION_CORID_QED_RM5261: -- cgit v0.10.2 From f9dadfa71bc594df09044da61d1c72701121d802 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 1 Nov 2006 10:05:35 -0800 Subject: i386: write IO APIC irq routing entries in correct order Since the "mask" bit is in the low word, when we write a new entry, we need to write the high word first, before we potentially unmask it. The exception is when we actually want to mask the interrupt, in which case we want to write the low word first to make sure that the high word doesn't change while the interrupt routing is still active. Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index eb10bd5..507983c 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -147,12 +147,34 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) return eu.entry; } +/* + * When we write a new IO APIC routing entry, we need to write the high + * word first! If the mask bit in the low word is clear, we will enable + * the interrupt, and we need to make sure the entry is fully populated + * before that happens. + */ static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; union entry_union eu; eu.entry = e; spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0x11 + 2*pin, eu.w2); + io_apic_write(apic, 0x10 + 2*pin, eu.w1); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +/* + * When we mask an IO APIC routing entry, we need to write the low + * word first, in order to set the mask bit before we change the + * high bits! + */ +static void ioapic_mask_entry(int apic, int pin) +{ + unsigned long flags; + union entry_union eu = { .entry.mask = 1 }; + + spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x10 + 2*pin, eu.w1); io_apic_write(apic, 0x11 + 2*pin, eu.w2); spin_unlock_irqrestore(&ioapic_lock, flags); @@ -274,9 +296,7 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) /* * Disable it in the IO-APIC irq-routing table: */ - memset(&entry, 0, sizeof(entry)); - entry.mask = 1; - ioapic_write_entry(apic, pin, entry); + ioapic_mask_entry(apic, pin); } static void clear_IO_APIC (void) -- cgit v0.10.2 From 3ccfc65c5004e5fe5cfbffe43b8acc686680b53e Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 2 Nov 2006 09:44:37 +1100 Subject: [PATCH] powerpc: Eliminate "exceeds stub group size" linker warning It turns out that the linker warnings on 64-bit powerpc about "section blah exceeds stub group size" were being triggered by conditional branches in head_64.S branching to global symbols, whether in head_64.S or in other files. This eliminates the warnings by making some global symbols in head_64.S no longer global, and by rearranging some branches. Signed-off-by: Paul Mackerras [ Yee-haa. Maybe I'll notice newly introduced real warnings now - Linus ] Signed-off-by: Linus Torvalds diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 291e362..e720729 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -487,7 +487,7 @@ BEGIN_FTR_SECTION rlwimi r13,r12,16,0x20 mfcr r12 cmpwi r13,0x2c - beq .do_stab_bolted_pSeries + beq do_stab_bolted_pSeries mtcrf 0x80,r12 mfspr r12,SPRN_SPRG2 END_FTR_SECTION_IFCLR(CPU_FTR_SLB) @@ -600,7 +600,7 @@ system_call_pSeries: STD_EXCEPTION_PSERIES(., performance_monitor) .align 7 -_GLOBAL(do_stab_bolted_pSeries) +do_stab_bolted_pSeries: mtcrf 0x80,r12 mfspr r12,SPRN_SPRG2 EXCEPTION_PROLOG_PSERIES(PACA_EXSLB, .do_stab_bolted) @@ -1046,7 +1046,7 @@ slb_miss_fault: li r5,0 std r4,_DAR(r1) std r5,_DSISR(r1) - b .handle_page_fault + b handle_page_fault unrecov_user_slb: EXCEPTION_PROLOG_COMMON(0x4200, PACA_EXGEN) @@ -1174,12 +1174,13 @@ program_check_common: .globl fp_unavailable_common fp_unavailable_common: EXCEPTION_PROLOG_COMMON(0x800, PACA_EXGEN) - bne .load_up_fpu /* if from user, just load it up */ + bne 1f /* if from user, just load it up */ bl .save_nvgprs addi r3,r1,STACK_FRAME_OVERHEAD ENABLE_INTS bl .kernel_fp_unavailable_exception BUG_OPCODE +1: b .load_up_fpu .align 7 .globl altivec_unavailable_common @@ -1279,10 +1280,10 @@ _GLOBAL(do_hash_page) std r4,_DSISR(r1) andis. r0,r4,0xa450 /* weird error? */ - bne- .handle_page_fault /* if not, try to insert a HPTE */ + bne- handle_page_fault /* if not, try to insert a HPTE */ BEGIN_FTR_SECTION andis. r0,r4,0x0020 /* Is it a segment table fault? */ - bne- .do_ste_alloc /* If so handle it */ + bne- do_ste_alloc /* If so handle it */ END_FTR_SECTION_IFCLR(CPU_FTR_SLB) /* @@ -1324,7 +1325,7 @@ BEGIN_FW_FTR_SECTION * because ret_from_except_lite will check for and handle pending * interrupts if necessary. */ - beq .ret_from_except_lite + beq 13f /* For a hash failure, we don't bother re-enabling interrupts */ ble- 12f @@ -1346,14 +1347,14 @@ BEGIN_FW_FTR_SECTION END_FW_FTR_SECTION_IFCLR(FW_FEATURE_ISERIES) /* Here we have a page fault that hash_page can't handle. */ -_GLOBAL(handle_page_fault) +handle_page_fault: ENABLE_INTS 11: ld r4,_DAR(r1) ld r5,_DSISR(r1) addi r3,r1,STACK_FRAME_OVERHEAD bl .do_page_fault cmpdi r3,0 - beq+ .ret_from_except_lite + beq+ 13f bl .save_nvgprs mr r5,r3 addi r3,r1,STACK_FRAME_OVERHEAD @@ -1370,12 +1371,14 @@ _GLOBAL(handle_page_fault) bl .low_hash_fault b .ret_from_except +13: b .ret_from_except_lite + /* here we have a segment miss */ -_GLOBAL(do_ste_alloc) +do_ste_alloc: bl .ste_allocate /* try to insert stab entry */ cmpdi r3,0 - beq+ fast_exception_return - b .handle_page_fault + bne- handle_page_fault + b fast_exception_return /* * r13 points to the PACA, r9 contains the saved CR, -- cgit v0.10.2 From d55b4c631e89a008e80b003e5aa4291d9ec800ac Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 31 Oct 2006 16:59:35 -0800 Subject: [TIPC] net/tipc/port.c: fix NULL dereference The correct order is: NULL check before dereference Spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller diff --git a/net/tipc/port.c b/net/tipc/port.c index c1a1a76..b7f3199 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -1136,11 +1136,12 @@ int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) int res = -EINVAL; p_ptr = tipc_port_lock(ref); + if (!p_ptr) + return -EINVAL; + dbg("tipc_publ %u, p_ptr = %x, conn = %x, scope = %x, " "lower = %u, upper = %u\n", ref, p_ptr, p_ptr->publ.connected, scope, seq->lower, seq->upper); - if (!p_ptr) - return -EINVAL; if (p_ptr->publ.connected) goto exit; if (seq->lower > seq->upper) -- cgit v0.10.2 From b1736a71404b3961f061c795a81210aa7f945fc0 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 31 Oct 2006 17:31:33 -0800 Subject: [TCP]: Set default congestion control when no sysctl. The setting of the default congestion control was buried in the sysctl code so it would not be done properly if SYSCTL was not enabled. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e82a5be..15061b3 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -129,13 +129,6 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, return ret; } -static int __init tcp_congestion_default(void) -{ - return tcp_set_default_congestion_control(CONFIG_DEFAULT_TCP_CONG); -} - -late_initcall(tcp_congestion_default); - ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index af0aca1..1e2982f 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -131,6 +131,14 @@ int tcp_set_default_congestion_control(const char *name) return ret; } +/* Set default value from kernel configuration at bootup */ +static int __init tcp_congestion_default(void) +{ + return tcp_set_default_congestion_control(CONFIG_DEFAULT_TCP_CONG); +} +late_initcall(tcp_congestion_default); + + /* Get current default congestion control */ void tcp_get_default_congestion_control(char *name) { -- cgit v0.10.2 From 5b1225454f7891970cb5ba87c8ef24edb1fa6c3a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 1 Nov 2006 15:28:58 -0800 Subject: [IPV6]: File the fingerprints off ah6->spi/esp6->spi In theory these are opaque 32bit values. However, we end up allocating them sequentially in host-endian and stick unchanged on the wire. Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index 7af227b..7931e4f 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -135,7 +135,7 @@ u32 xfrm6_tunnel_spi_lookup(xfrm_address_t *saddr) x6spi = __xfrm6_tunnel_spi_lookup(saddr); spi = x6spi ? x6spi->spi : 0; read_unlock_bh(&xfrm6_tunnel_spi_lock); - return spi; + return htonl(spi); } EXPORT_SYMBOL(xfrm6_tunnel_spi_lookup); @@ -210,7 +210,7 @@ u32 xfrm6_tunnel_alloc_spi(xfrm_address_t *saddr) spi = __xfrm6_tunnel_alloc_spi(saddr); write_unlock_bh(&xfrm6_tunnel_spi_lock); - return spi; + return htonl(spi); } EXPORT_SYMBOL(xfrm6_tunnel_alloc_spi); -- cgit v0.10.2 From c7fed9d75074f7c243ec8ff2c55d04de2839a6f6 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 1 Nov 2006 16:30:39 -0800 Subject: [SPARC64]: Fix futex_atomic_cmpxchg_inatomic implementation. I copied the logic from ll/sc arch implementations, but that was wrong and makes no sense at all. Just do a straight compare-exchange instruction, just like x86. Based upon bug reports from Dennis Gilmore and Fabio Massimo. Signed-off-by: David S. Miller diff --git a/include/asm-sparc64/futex.h b/include/asm-sparc64/futex.h index dee4020..7392fc4 100644 --- a/include/asm-sparc64/futex.h +++ b/include/asm-sparc64/futex.h @@ -87,24 +87,22 @@ static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval) { __asm__ __volatile__( - "\n1: lduwa [%2] %%asi, %0\n" - "2: casa [%2] %%asi, %0, %1\n" - "3:\n" + "\n1: casa [%3] %%asi, %2, %0\n" + "2:\n" " .section .fixup,#alloc,#execinstr\n" " .align 4\n" - "4: ba 3b\n" - " mov %3, %0\n" + "3: ba 2b\n" + " mov %4, %0\n" " .previous\n" " .section __ex_table,\"a\"\n" " .align 4\n" - " .word 1b, 4b\n" - " .word 2b, 4b\n" + " .word 1b, 3b\n" " .previous\n" - : "=&r" (oldval) - : "r" (newval), "r" (uaddr), "i" (-EFAULT) + : "=r" (newval) + : "0" (newval), "r" (oldval), "r" (uaddr), "i" (-EFAULT) : "memory"); - return oldval; + return newval; } #endif /* !(_SPARC64_FUTEX_H) */ -- cgit v0.10.2 From 6f5b7ef6b5816dc497094048d7d8a270004602d6 Mon Sep 17 00:00:00 2001 From: Meelis Roos Date: Wed, 1 Nov 2006 18:07:27 -0800 Subject: [NETFILTER]: silence a warning in ebtables net/bridge/netfilter/ebtables.c: In function 'ebt_dev_check': net/bridge/netfilter/ebtables.c:89: warning: initialization discards qualifiers from pointer target type So make the char* a const char * and the warning is gone. Signed-off-by: Meelis Roos Signed-off-by: David S. Miller diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 3df55b2..9f85666 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -86,7 +86,7 @@ static inline int ebt_do_match (struct ebt_entry_match *m, static inline int ebt_dev_check(char *entry, const struct net_device *device) { int i = 0; - char *devname = device->name; + const char *devname = device->name; if (*entry == '\0') return 0; -- cgit v0.10.2 From 732f74a46711c0724885703fb689c79139c84a3c Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 1 Nov 2006 22:09:21 -0500 Subject: Revert "[PATCH] Add 0x7110 piix to ata_piix.c" This reverts commit f833229c96c0bf53c05995e4bd58709d9e9edd67: According to reviewers and the lspci data provided in commit message itself, PCI ID 0x7110 should not have been added to ata_piix. Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 8385387..720174d 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -168,7 +168,6 @@ static const struct pci_device_id piix_pci_tbl[] = { #ifdef ATA_ENABLE_PATA /* Intel PIIX4 for the 430TX/440BX/MX chipset: UDMA 33 */ /* Also PIIX4E (fn3 rev 2) and PIIX4M (fn3 rev 3) */ - { 0x8086, 0x7110, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 }, { 0x8086, 0x7111, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix_pata_33 }, { 0x8086, 0x24db, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 }, { 0x8086, 0x25a2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich_pata_100 }, -- cgit v0.10.2 From 8e87d4dc159148f04f515bc072df22a2c089e7f2 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 2 Nov 2006 03:45:24 +0000 Subject: [CIFS] report rename failure when target file is locked by Windows Fixes Samba bugzilla bug # 4182 Rename by handle failures (retry after rename by path) were not being returned back. Signed-off-by: Steve French diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 50afab8..0b3c37e 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -3,7 +3,9 @@ Version 1.46 Support deep tree mounts. Better support OS/2, Win9x (DOS) time stamps. Allow null user to be specified on mount ("username="). Do not return EINVAL on readdir when filldir fails due to overwritten blocksize -(fixes FC problem) +(fixes FC problem). Return error in rename 2nd attempt retry (ie report +if rename by handle also fails, after rename by path fails, we were +not reporting whether the retry worked or not). Version 1.45 ------------ diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 35d54bb..dffe295 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -885,10 +885,14 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, kmalloc(2 * sizeof(FILE_UNIX_BASIC_INFO), GFP_KERNEL); if (info_buf_source != NULL) { info_buf_target = info_buf_source + 1; - rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, - info_buf_source, cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + if (pTcon->ses->capabilities & CAP_UNIX) + rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, + info_buf_source, + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + /* else rc is still EEXIST so will fall through to + unlink the target and retry rename */ if (rc == 0) { rc = CIFSSMBUnixQPathInfo(xid, pTcon, toName, info_buf_target, @@ -937,7 +941,7 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { - CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, + rc = CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, cifs_sb_source->local_nls, cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); -- cgit v0.10.2 From d572b87946f8c598b3cad86a7913862dd48daadb Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Thu, 2 Nov 2006 10:50:40 -0600 Subject: JFS: Remove redundant xattr permission checking The vfs handles most permissions for setting and retrieving xattrs. This patch removes a redundant and wrong check so that it won't override the correct behavior which is being fixed in the vfs. Signed-off-by: Dave Kleikamp diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c index 4c7985e..b753ba2 100644 --- a/fs/jfs/xattr.c +++ b/fs/jfs/xattr.c @@ -756,6 +756,11 @@ static int can_set_system_xattr(struct inode *inode, const char *name, return -EOPNOTSUPP; } +/* + * Most of the permission checking is done by xattr_permission in the vfs. + * The local file system is responsible for handling the system.* namespace. + * We also need to verify that this is a namespace that we recognize. + */ static int can_set_xattr(struct inode *inode, const char *name, const void *value, size_t value_len) { @@ -771,10 +776,6 @@ static int can_set_xattr(struct inode *inode, const char *name, strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN)) return -EOPNOTSUPP; - if (!S_ISREG(inode->i_mode) && - (!S_ISDIR(inode->i_mode) || inode->i_mode &S_ISVTX)) - return -EPERM; - return 0; } -- cgit v0.10.2 From 8427829711b35e0e62668618cec577f65c102935 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 2 Nov 2006 02:00:02 +0000 Subject: Revert "[MIPS] Make SPARSEMEM selectable on QEMU." This reverts commit 31473747bd441719f9f6a07385684dce547533e0. Another amazing example of patch(1) messing up - lmo changeset 66e8560d11d02bcadc261498471831a6375ad046 was merged twice to kernel.org and ended up doing this rubbish job. Signed-off-by: Ralf Baechle diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 74ba763..76f1cea 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -427,7 +427,6 @@ config MOMENCO_OCELOT_G select SYS_SUPPORTS_32BIT_KERNEL select SYS_SUPPORTS_64BIT_KERNEL select SYS_SUPPORTS_BIG_ENDIAN - select ARCH_SPARSEMEM_ENABLE help The Ocelot is a MIPS-based Single Board Computer (SBC) made by Momentum Computer . @@ -1633,9 +1632,6 @@ config ARCH_DISCONTIGMEM_ENABLE config ARCH_SPARSEMEM_ENABLE bool - -config ARCH_SPARSEMEM_ENABLE - bool select SPARSEMEM_STATIC config NUMA -- cgit v0.10.2 From 8a88ca8f7fd15d06e53a848c6b3558ed9973327c Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 2 Nov 2006 17:23:33 +0000 Subject: [MIPS] Fix merge screwup by patch(1) Patch happily applied an Ocelot G patch to Ocelot C when merging linux-mips.org changeset 91ee9a801e65d2981dfe327d2519c7fc6ab02e6b into kernel.org as 6ceb6d3ab2d402cea326320a4143db90a66fd216. Signed-off-by: Ralf Baechle diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 76f1cea..848089ec 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -408,7 +408,7 @@ config MOMENCO_OCELOT_C select SWAP_IO_SPACE select SYS_HAS_CPU_RM7000 select SYS_SUPPORTS_32BIT_KERNEL - select SYS_SUPPORTS_64BIT_KERNEL if BROKEN + select SYS_SUPPORTS_64BIT_KERNEL select SYS_SUPPORTS_BIG_ENDIAN help The Ocelot is a MIPS-based Single Board Computer (SBC) made by @@ -425,7 +425,7 @@ config MOMENCO_OCELOT_G select SWAP_IO_SPACE select SYS_HAS_CPU_RM7000 select SYS_SUPPORTS_32BIT_KERNEL - select SYS_SUPPORTS_64BIT_KERNEL + select SYS_SUPPORTS_64BIT_KERNEL if BROKEN select SYS_SUPPORTS_BIG_ENDIAN help The Ocelot is a MIPS-based Single Board Computer (SBC) made by -- cgit v0.10.2 From 1a5c5de1b64ec510a6ab6994702c295db00b9acc Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 2 Nov 2006 17:23:33 +0000 Subject: [MIPS] IP27: Allow SMP ;-) Another changeset messed up by patch. When lmo commit 4ef893e0515e8bf336dfbd200884f244869fbb43 was merged to kernel.org as e73ea273ef87a04ff59fc368fa33333dca275dde patch happily applied the IP27 segment to IP22. f63f36c18b11e166d0f362ac04dbcd7e6ea23f9e did fix the effects partially - and with a wrong log message. Now fixed for real (tm). Signed-off-by: Ralf Baechle diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 848089ec..1443024 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -559,6 +559,7 @@ config SGI_IP27 select SYS_SUPPORTS_64BIT_KERNEL select SYS_SUPPORTS_BIG_ENDIAN select SYS_SUPPORTS_NUMA + select SYS_SUPPORTS_SMP help This are the SGI Origin 200, Origin 2000 and Onyx 2 Graphics workstations. To compile a Linux kernel that runs on these, say Y -- cgit v0.10.2 From 9ba126cfbf505f4d5b39ed294cedd241321c7a91 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 13 Oct 2006 11:22:52 +0100 Subject: [MIPS] Fix warning about init_initrd() call if !CONFIG_BLK_DEV_INITRD. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c index fdbb508..8f6e896 100644 --- a/arch/mips/kernel/setup.c +++ b/arch/mips/kernel/setup.c @@ -223,7 +223,11 @@ disable: #else /* !CONFIG_BLK_DEV_INITRD */ -#define init_initrd() 0 +static unsigned long __init init_initrd(void) +{ + return 0; +} + #define finalize_initrd() do {} while (0) #endif -- cgit v0.10.2 From 8b922a851731037b2f1e1669e9b1a0baff3ab5dc Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 30 Oct 2006 12:48:04 +0000 Subject: [MIPS] Ocelot G: Fix : "CURRENTLY_UNUSED" is not defined warning. CC arch/mips/momentum/ocelot_g/gt-irq.o arch/mips/momentum/ocelot_g/gt-irq.c:30:5: warning: "CURRENTLY_UNUSED" is not defined arch/mips/momentum/ocelot_g/gt-irq.c:199:5: warning: "CURRENTLY_UNUSED" is not defined Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_g/gt-irq.c b/arch/mips/momentum/ocelot_g/gt-irq.c index 7b5cc66..e5576bd 100644 --- a/arch/mips/momentum/ocelot_g/gt-irq.c +++ b/arch/mips/momentum/ocelot_g/gt-irq.c @@ -27,7 +27,7 @@ unsigned long bus_clock; * be handled and ack'ed differently than other MIPS interrupts. */ -#if CURRENTLY_UNUSED +#if 0 struct tq_struct irq_handlers[MAX_CAUSE_REGS][MAX_CAUSE_REG_WIDTH]; void hook_irq_handler(int int_cause, int bit_num, void *isr_ptr); @@ -95,7 +95,7 @@ int disable_galileo_irq(int int_cause, int bit_num) return 0; return 1; } -#endif /* UNUSED */ +#endif /* 0 */ /* * Interrupt handler for interrupts coming from the Galileo chip via P0_INT#. @@ -196,7 +196,7 @@ void gt64240_time_init(void) void gt64240_irq_init(void) { -#if CURRENTLY_UNUSED +#if 0 int i, j; /* Reset irq handlers pointers to NULL */ @@ -208,5 +208,5 @@ void gt64240_irq_init(void) irq_handlers[i][j].data = NULL; } } -#endif +#endif /* 0 */ } -- cgit v0.10.2 From 904880e717c5466041485ca6d6e8c6c1ef06d0fd Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 13 Oct 2006 11:32:50 +0100 Subject: [MIPS] Don't use R10000 llsc workaround version for all llsc-full processors. Found and original patch by bile@landofbile.com. Signed-off-by: Ralf Baechle diff --git a/include/asm-mips/system.h b/include/asm-mips/system.h index dcb4701..3056fee 100644 --- a/include/asm-mips/system.h +++ b/include/asm-mips/system.h @@ -392,7 +392,7 @@ static inline unsigned long __cmpxchg_u64(volatile int * m, unsigned long old, { __u64 retval; - if (cpu_has_llsc) { + if (cpu_has_llsc && R10000_LLSC_WAR) { __asm__ __volatile__( " .set push \n" " .set noat \n" -- cgit v0.10.2 From 236d333c3c05c179e31f461285c09271256a1381 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Tue, 10 Oct 2006 22:13:55 +0900 Subject: [MIPS] Do not use -msym32 option for modules. On 64-bit kernel, modules are loaded into XKSEG for now. While XKSEG address is not a sign-extended 32-bit address, we can not use -msym32 option. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 641aa30..d580d46 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -63,7 +63,9 @@ cflags-y += -mabi=64 ifdef CONFIG_BUILD_ELF64 cflags-y += $(call cc-option,-mno-explicit-relocs) else -cflags-y += $(call cc-option,-msym32) +# -msym32 can not be used for modules since they are loaded into XKSEG +CFLAGS_MODULE += $(call cc-option,-mno-explicit-relocs) +CFLAGS_KERNEL += $(call cc-option,-msym32) endif endif -- cgit v0.10.2 From 7a118df3ea23820b9922a1b51cd2f24e464f4c17 Mon Sep 17 00:00:00 2001 From: Sean Hefty Date: Tue, 31 Oct 2006 11:12:59 -0800 Subject: RDMA/addr: Use client registration to fix module unload race Require registration with ib_addr module to prevent caller from unloading while a callback is in progress. Signed-off-by: Sean Hefty Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index 60d3fbd..e11187e 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -47,6 +47,7 @@ struct addr_req { struct sockaddr src_addr; struct sockaddr dst_addr; struct rdma_dev_addr *addr; + struct rdma_addr_client *client; void *context; void (*callback)(int status, struct sockaddr *src_addr, struct rdma_dev_addr *addr, void *context); @@ -61,6 +62,26 @@ static LIST_HEAD(req_list); static DECLARE_WORK(work, process_req, NULL); static struct workqueue_struct *addr_wq; +void rdma_addr_register_client(struct rdma_addr_client *client) +{ + atomic_set(&client->refcount, 1); + init_completion(&client->comp); +} +EXPORT_SYMBOL(rdma_addr_register_client); + +static inline void put_client(struct rdma_addr_client *client) +{ + if (atomic_dec_and_test(&client->refcount)) + complete(&client->comp); +} + +void rdma_addr_unregister_client(struct rdma_addr_client *client) +{ + put_client(client); + wait_for_completion(&client->comp); +} +EXPORT_SYMBOL(rdma_addr_unregister_client); + int rdma_copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev, const unsigned char *dst_dev_addr) { @@ -229,6 +250,7 @@ static void process_req(void *data) list_del(&req->list); req->callback(req->status, &req->src_addr, req->addr, req->context); + put_client(req->client); kfree(req); } } @@ -264,7 +286,8 @@ static int addr_resolve_local(struct sockaddr_in *src_in, return ret; } -int rdma_resolve_ip(struct sockaddr *src_addr, struct sockaddr *dst_addr, +int rdma_resolve_ip(struct rdma_addr_client *client, + struct sockaddr *src_addr, struct sockaddr *dst_addr, struct rdma_dev_addr *addr, int timeout_ms, void (*callback)(int status, struct sockaddr *src_addr, struct rdma_dev_addr *addr, void *context), @@ -285,6 +308,8 @@ int rdma_resolve_ip(struct sockaddr *src_addr, struct sockaddr *dst_addr, req->addr = addr; req->callback = callback; req->context = context; + req->client = client; + atomic_inc(&client->refcount); src_in = (struct sockaddr_in *) &req->src_addr; dst_in = (struct sockaddr_in *) &req->dst_addr; @@ -305,6 +330,7 @@ int rdma_resolve_ip(struct sockaddr *src_addr, struct sockaddr *dst_addr, break; default: ret = req->status; + atomic_dec(&client->refcount); kfree(req); break; } diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index d8ca3c1..845090b 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -63,6 +63,7 @@ static struct ib_client cma_client = { }; static struct ib_sa_client sa_client; +static struct rdma_addr_client addr_client; static LIST_HEAD(dev_list); static LIST_HEAD(listen_any_list); static DEFINE_MUTEX(lock); @@ -1625,8 +1626,8 @@ int rdma_resolve_addr(struct rdma_cm_id *id, struct sockaddr *src_addr, if (cma_any_addr(dst_addr)) ret = cma_resolve_loopback(id_priv); else - ret = rdma_resolve_ip(&id->route.addr.src_addr, dst_addr, - &id->route.addr.dev_addr, + ret = rdma_resolve_ip(&addr_client, &id->route.addr.src_addr, + dst_addr, &id->route.addr.dev_addr, timeout_ms, addr_handler, id_priv); if (ret) goto err; @@ -2217,6 +2218,7 @@ static int cma_init(void) return -ENOMEM; ib_sa_register_client(&sa_client); + rdma_addr_register_client(&addr_client); ret = ib_register_client(&cma_client); if (ret) @@ -2224,6 +2226,7 @@ static int cma_init(void) return 0; err: + rdma_addr_unregister_client(&addr_client); ib_sa_unregister_client(&sa_client); destroy_workqueue(cma_wq); return ret; @@ -2232,6 +2235,7 @@ err: static void cma_cleanup(void) { ib_unregister_client(&cma_client); + rdma_addr_unregister_client(&addr_client); ib_sa_unregister_client(&sa_client); destroy_workqueue(cma_wq); idr_destroy(&sdp_ps); diff --git a/include/rdma/ib_addr.h b/include/rdma/ib_addr.h index 81b62307..c094e50 100644 --- a/include/rdma/ib_addr.h +++ b/include/rdma/ib_addr.h @@ -36,6 +36,22 @@ #include #include +struct rdma_addr_client { + atomic_t refcount; + struct completion comp; +}; + +/** + * rdma_addr_register_client - Register an address client. + */ +void rdma_addr_register_client(struct rdma_addr_client *client); + +/** + * rdma_addr_unregister_client - Deregister an address client. + * @client: Client object to deregister. + */ +void rdma_addr_unregister_client(struct rdma_addr_client *client); + struct rdma_dev_addr { unsigned char src_dev_addr[MAX_ADDR_LEN]; unsigned char dst_dev_addr[MAX_ADDR_LEN]; @@ -52,6 +68,7 @@ int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr); /** * rdma_resolve_ip - Resolve source and destination IP addresses to * RDMA hardware addresses. + * @client: Address client associated with request. * @src_addr: An optional source address to use in the resolution. If a * source address is not provided, a usable address will be returned via * the callback. @@ -64,7 +81,8 @@ int rdma_translate_ip(struct sockaddr *addr, struct rdma_dev_addr *dev_addr); * or been canceled. A status of 0 indicates success. * @context: User-specified context associated with the call. */ -int rdma_resolve_ip(struct sockaddr *src_addr, struct sockaddr *dst_addr, +int rdma_resolve_ip(struct rdma_addr_client *client, + struct sockaddr *src_addr, struct sockaddr *dst_addr, struct rdma_dev_addr *addr, int timeout_ms, void (*callback)(int status, struct sockaddr *src_addr, struct rdma_dev_addr *addr, void *context), -- cgit v0.10.2 From 73218187e8650c08d4e4735dbeeeb860f438da7c Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Thu, 2 Nov 2006 23:44:24 +0100 Subject: [ARM] 3917/1: Fix dmabounce symbol exports dma_sync_single is no more (and to be removed in 2.7) so this export should be dma_sync_single_for_cpu. Also export dma_sync_single_for_device. Signed-off-by: Kevin Hilman Signed-off-by: Russell King diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c index 028bdc9..2e635b8 100644 --- a/arch/arm/common/dmabounce.c +++ b/arch/arm/common/dmabounce.c @@ -662,7 +662,8 @@ EXPORT_SYMBOL(dma_map_single); EXPORT_SYMBOL(dma_unmap_single); EXPORT_SYMBOL(dma_map_sg); EXPORT_SYMBOL(dma_unmap_sg); -EXPORT_SYMBOL(dma_sync_single); +EXPORT_SYMBOL(dma_sync_single_for_cpu); +EXPORT_SYMBOL(dma_sync_single_for_device); EXPORT_SYMBOL(dma_sync_sg); EXPORT_SYMBOL(dmabounce_register_dev); EXPORT_SYMBOL(dmabounce_unregister_dev); -- cgit v0.10.2 From 05e2867a7bcc76de37e103a97ed48ba6872db797 Mon Sep 17 00:00:00 2001 From: Peer Chen Date: Thu, 2 Nov 2006 17:58:21 -0500 Subject: [libata] Add support for PATA controllers of MCP67 to pata_amd.c. Signed-off-by: Peer Chen Signed-off-by: Jeff Garzik diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c index 29234c8..5c47a9e 100644 --- a/drivers/ata/pata_amd.c +++ b/drivers/ata/pata_amd.c @@ -677,6 +677,8 @@ static const struct pci_device_id amd[] = { { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE), 8 }, { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE), 8 }, { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE), 8 }, + { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE), 8 }, + { PCI_VDEVICE(NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE), 8 }, { PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_CS5536_IDE), 9 }, { }, -- cgit v0.10.2 From 94c52fde553260e86f263448034930abe364faec Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Mon, 30 Oct 2006 02:27:45 +0100 Subject: [ARM] 3915/1: S3C2412: Add s3c2410_gpio_getirq() to general gpio.c s3c2410_gpio_getirq() holds for the S3C2412 build, so ensure that it gets built for all the current S3C24XX architectures Signed-off-by: Ben Dooks Signed-off-by: Russell King diff --git a/arch/arm/mach-s3c2410/gpio.c b/arch/arm/mach-s3c2410/gpio.c index db6393c..ba34654 100644 --- a/arch/arm/mach-s3c2410/gpio.c +++ b/arch/arm/mach-s3c2410/gpio.c @@ -3,7 +3,7 @@ * Copyright (c) 2004-2005 Simtec Electronics * Ben Dooks * - * S3C2410 GPIO support + * S3C24XX GPIO support * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -163,3 +163,22 @@ unsigned int s3c2410_modify_misccr(unsigned int clear, unsigned int change) } EXPORT_SYMBOL(s3c2410_modify_misccr); + +int s3c2410_gpio_getirq(unsigned int pin) +{ + if (pin < S3C2410_GPF0 || pin > S3C2410_GPG15) + return -1; /* not valid interrupts */ + + if (pin < S3C2410_GPG0 && pin > S3C2410_GPF7) + return -1; /* not valid pin */ + + if (pin < S3C2410_GPF4) + return (pin - S3C2410_GPF0) + IRQ_EINT0; + + if (pin < S3C2410_GPG0) + return (pin - S3C2410_GPF4) + IRQ_EINT4; + + return (pin - S3C2410_GPG0) + IRQ_EINT8; +} + +EXPORT_SYMBOL(s3c2410_gpio_getirq); diff --git a/arch/arm/mach-s3c2410/s3c2410-gpio.c b/arch/arm/mach-s3c2410/s3c2410-gpio.c index a2098f6..ec3a276 100644 --- a/arch/arm/mach-s3c2410/s3c2410-gpio.c +++ b/arch/arm/mach-s3c2410/s3c2410-gpio.c @@ -69,22 +69,3 @@ int s3c2410_gpio_irqfilter(unsigned int pin, unsigned int on, } EXPORT_SYMBOL(s3c2410_gpio_irqfilter); - -int s3c2410_gpio_getirq(unsigned int pin) -{ - if (pin < S3C2410_GPF0 || pin > S3C2410_GPG15) - return -1; /* not valid interrupts */ - - if (pin < S3C2410_GPG0 && pin > S3C2410_GPF7) - return -1; /* not valid pin */ - - if (pin < S3C2410_GPF4) - return (pin - S3C2410_GPF0) + IRQ_EINT0; - - if (pin < S3C2410_GPG0) - return (pin - S3C2410_GPF4) + IRQ_EINT4; - - return (pin - S3C2410_GPG0) + IRQ_EINT8; -} - -EXPORT_SYMBOL(s3c2410_gpio_getirq); -- cgit v0.10.2 From 8f7f9435e6df0985c877d10259393bdfaac3655f Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Fri, 27 Oct 2006 05:13:19 +0100 Subject: [ARM] 3912/1: Make PXA270 advertise HWCAP_IWMMXT capability ARM patch 3756/1 added HWCAP_IWMMXT. This patch adds support for broadcasting that info via /proc/cpuinfo and sets it for the CPU features of the PXA270. I've booted 19rc3 on a pxa270 and confirmed that the /proc/cpuinfo shows "iwmmxt" in the Features. Signed-off-by: Paul Gortmaker Signed-off-by: Nicolas Pitre Signed-off-by: Russell King diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 6bbd93d..29efc9f 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -357,6 +357,9 @@ static void __init setup_processor(void) #ifndef CONFIG_VFP elf_hwcap &= ~HWCAP_VFP; #endif +#ifndef CONFIG_IWMMXT + elf_hwcap &= ~HWCAP_IWMMXT; +#endif cpu_proc_init(); } @@ -854,6 +857,7 @@ static const char *hwcap_str[] = { "vfp", "edsp", "java", + "iwmmxt", NULL }; diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S index e8b377d..2749c1f 100644 --- a/arch/arm/mm/proc-xscale.S +++ b/arch/arm/mm/proc-xscale.S @@ -909,7 +909,7 @@ __pxa270_proc_info: b __xscale_setup .long cpu_arch_name .long cpu_elf_name - .long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP + .long HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP|HWCAP_IWMMXT .long cpu_pxa270_name .long xscale_processor_functions .long v4wbi_tlb_fns -- cgit v0.10.2 From 895663cd92574367054e0eb604a7428852f359b8 Mon Sep 17 00:00:00 2001 From: Peer Chen Date: Thu, 2 Nov 2006 17:59:46 -0500 Subject: [libata] Add support for AHCI controllers of MCP67. Signed-off-by: Peer Chen Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 988f8bb..234197e 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -334,6 +334,14 @@ static const struct pci_device_id ahci_pci_tbl[] = { { PCI_VDEVICE(NVIDIA, 0x044d), board_ahci }, /* MCP65 */ { PCI_VDEVICE(NVIDIA, 0x044e), board_ahci }, /* MCP65 */ { PCI_VDEVICE(NVIDIA, 0x044f), board_ahci }, /* MCP65 */ + { PCI_VDEVICE(NVIDIA, 0x0554), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0555), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0556), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0557), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0558), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x0559), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x055a), board_ahci }, /* MCP67 */ + { PCI_VDEVICE(NVIDIA, 0x055b), board_ahci }, /* MCP67 */ /* SiS */ { PCI_VDEVICE(SI, 0x1184), board_ahci }, /* SiS 966 */ -- cgit v0.10.2 From 4fa2eeeac5e13a8579ee45bc172eed690d28fbb7 Mon Sep 17 00:00:00 2001 From: Peer Chen Date: Thu, 2 Nov 2006 18:55:48 -0500 Subject: pci_ids.h: Add NVIDIA PCI ID Signed-off-by: Jeff Garzik diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f3a168f..fa4e1d7 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1213,6 +1213,7 @@ #define PCI_DEVICE_ID_NVIDIA_NVENET_21 0x0451 #define PCI_DEVICE_ID_NVIDIA_NVENET_22 0x0452 #define PCI_DEVICE_ID_NVIDIA_NVENET_23 0x0453 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE 0x0560 #define PCI_VENDOR_ID_IMS 0x10e0 #define PCI_DEVICE_ID_IMS_TT128 0x9128 -- cgit v0.10.2 From 6c8c21b9119cfe68a99825085014bba4f9c0c768 Mon Sep 17 00:00:00 2001 From: Trent Piepho Date: Sat, 14 Oct 2006 16:21:02 -0300 Subject: V4L/DVB (4752): DVB: Add DVB_FE_CUSTOMISE support for MT2060 Let the MT2060 be customized like most of the other DVB PLLs/front-ends. Also, add a missing dependency on I2C. Signed-off-by: Trent Piepho Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/dvb-usb/Kconfig b/drivers/media/dvb/dvb-usb/Kconfig index 2cc5caa..a263b3f 100644 --- a/drivers/media/dvb/dvb-usb/Kconfig +++ b/drivers/media/dvb/dvb-usb/Kconfig @@ -26,7 +26,7 @@ config DVB_USB_A800 tristate "AVerMedia AverTV DVB-T USB 2.0 (A800)" depends on DVB_USB select DVB_DIB3000MC - select DVB_TUNER_MT2060 + select DVB_TUNER_MT2060 if !DVB_FE_CUSTOMISE help Say Y here to support the AVerMedia AverTV DVB-T USB 2.0 (A800) receiver. @@ -34,7 +34,7 @@ config DVB_USB_DIBUSB_MB tristate "DiBcom USB DVB-T devices (based on the DiB3000M-B) (see help for device list)" depends on DVB_USB select DVB_DIB3000MB - select DVB_TUNER_MT2060 + select DVB_TUNER_MT2060 if !DVB_FE_CUSTOMISE help Support for USB 1.1 and 2.0 DVB-T receivers based on reference designs made by DiBcom () equipped with a DiB3000M-B demodulator. @@ -55,7 +55,7 @@ config DVB_USB_DIBUSB_MC tristate "DiBcom USB DVB-T devices (based on the DiB3000M-C/P) (see help for device list)" depends on DVB_USB select DVB_DIB3000MC - select DVB_TUNER_MT2060 + select DVB_TUNER_MT2060 if !DVB_FE_CUSTOMISE help Support for USB2.0 DVB-T receivers based on reference designs made by DiBcom () equipped with a DiB3000M-C/P demodulator. @@ -70,7 +70,7 @@ config DVB_USB_DIB0700 tristate "DiBcom DiB0700 USB DVB devices (see help for supported devices)" depends on DVB_USB select DVB_DIB3000MC - select DVB_TUNER_MT2060 + select DVB_TUNER_MT2060 if !DVB_FE_CUSTOMISE help Support for USB2.0/1.1 DVB receivers based on the DiB0700 USB bridge. The USB bridge is also present in devices having the DiB7700 DVB-T-USB @@ -87,7 +87,7 @@ config DVB_USB_UMT_010 tristate "HanfTek UMT-010 DVB-T USB2.0 support" depends on DVB_USB select DVB_DIB3000MC - select DVB_TUNER_MT2060 + select DVB_TUNER_MT2060 if !DVB_FE_CUSTOMISE help Say Y here to support the HanfTek UMT-010 USB2.0 stick-sized DVB-T receiver. @@ -153,7 +153,7 @@ config DVB_USB_NOVA_T_USB2 tristate "Hauppauge WinTV-NOVA-T usb2 DVB-T USB2.0 support" depends on DVB_USB select DVB_DIB3000MC - select DVB_TUNER_MT2060 + select DVB_TUNER_MT2060 if !DVB_FE_CUSTOMISE help Say Y here to support the Hauppauge WinTV-NOVA-T usb2 DVB-T USB2.0 receiver. diff --git a/drivers/media/dvb/frontends/Kconfig b/drivers/media/dvb/frontends/Kconfig index 080fa25..aebb8d6f 100644 --- a/drivers/media/dvb/frontends/Kconfig +++ b/drivers/media/dvb/frontends/Kconfig @@ -276,6 +276,8 @@ config DVB_TDA826X config DVB_TUNER_MT2060 tristate "Microtune MT2060 silicon IF tuner" + depends on I2C + default m if DVB_FE_CUSTOMISE help A driver for the silicon IF tuner MT2060 from Microtune. diff --git a/drivers/media/dvb/frontends/mt2060.h b/drivers/media/dvb/frontends/mt2060.h index 34a37c2..0a86eab 100644 --- a/drivers/media/dvb/frontends/mt2060.h +++ b/drivers/media/dvb/frontends/mt2060.h @@ -30,6 +30,14 @@ struct mt2060_config { u8 clock_out; /* 0 = off, 1 = CLK/4, 2 = CLK/2, 3 = CLK/1 */ }; +#if defined(CONFIG_DVB_TUNER_MT2060) || (defined(CONFIG_DVB_TUNER_MT2060_MODULE) && defined(MODULE)) extern struct dvb_frontend * mt2060_attach(struct dvb_frontend *fe, struct i2c_adapter *i2c, struct mt2060_config *cfg, u16 if1); +#else +static inline struct dvb_frontend * mt2060_attach(struct dvb_frontend *fe, struct i2c_adapter *i2c, struct mt2060_config *cfg, u16 if1) +{ + printk(KERN_WARNING "%s: driver disabled by Kconfig\n", __FUNCTION__); + return NULL; +} +#endif // CONFIG_DVB_TUNER_MT2060 #endif -- cgit v0.10.2 From ecba77f246011344f0b8f46eb25ae01ab4ae282d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?David=20H=E4rdeman?= Date: Fri, 27 Oct 2006 20:56:51 -0300 Subject: V4L/DVB (4785): Budget-ci: Change DEBIADDR_IR to a safer default MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The IR chip has no address decoding, so the IR data is always present in the high byte when doing a read from the saa7146 chip. This means that the DEBI address used is irrelevant to the IR decoding logic. DEBI addresses 0x1XXX are mapped to the registers on the CI module itself, but only the lowest two bits are actually used (see EN50221, section A.2.2.1), meaning that 0x1234 is equivalent to 0x1000 which maps to register 0 (the data register). A read from the data register is supposed to be preceded by a read from the size register, so some CI modules will be confused (the AlphaCrypt CAM will hang completely). The attached patch changes the address used when reading the IR data to use 0x4000 instead. This is the CI version address, which is a safer default, works with the AlphaCrypt CAM and matches the behaviour of the Windows driver (AFAIK). Signed-off-by: David Härdeman Signed-off-by: Oliver Endriss Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/ttpci/budget-ci.c b/drivers/media/dvb/ttpci/budget-ci.c index 2a2e9b4..ac0cecb 100644 --- a/drivers/media/dvb/ttpci/budget-ci.c +++ b/drivers/media/dvb/ttpci/budget-ci.c @@ -46,7 +46,14 @@ #include "bsbe1.h" #include "bsru6.h" -#define DEBIADDR_IR 0x1234 +/* + * Regarding DEBIADDR_IR: + * Some CI modules hang if random addresses are read. + * Using address 0x4000 for the IR read means that we + * use the same address as for CI version, which should + * be a safe default. + */ +#define DEBIADDR_IR 0x4000 #define DEBIADDR_CICONTROL 0x0000 #define DEBIADDR_CIVERSION 0x4000 #define DEBIADDR_IO 0x1000 -- cgit v0.10.2 From c2625bff997f195e067ae11c9b0aa7217fb32991 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 29 Oct 2006 11:12:27 -0300 Subject: V4L/DVB (4786): Pvrusb2: use NULL instead of 0 Fix sparse NULL usage warnings: drivers/media/video/pvrusb2/pvrusb2-v4l2.c:714:14: warning: Using plain integer as NULL pointer drivers/media/video/pvrusb2/pvrusb2-v4l2.c:715:16: warning: Using plain integer as NULL pointer drivers/media/video/pvrusb2/pvrusb2-v4l2.c:1079:10: warning: Using plain integer as NULL pointer drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c:224:58: warning: Using plain integer as NULL pointer Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c index df8feac..c80c26b 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c +++ b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c @@ -221,7 +221,7 @@ static unsigned int decoder_describe(struct pvr2_v4l_cx2584x *ctxt, static void decoder_reset(struct pvr2_v4l_cx2584x *ctxt) { int ret; - ret = pvr2_i2c_client_cmd(ctxt->client,VIDIOC_INT_RESET,0); + ret = pvr2_i2c_client_cmd(ctxt->client,VIDIOC_INT_RESET,NULL); pvr2_trace(PVR2_TRACE_CHIPS,"i2c cx25840 decoder_reset (ret=%d)",ret); } diff --git a/drivers/media/video/pvrusb2/pvrusb2-v4l2.c b/drivers/media/video/pvrusb2/pvrusb2-v4l2.c index 97e974d..bb40e90 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-v4l2.c +++ b/drivers/media/video/pvrusb2/pvrusb2-v4l2.c @@ -711,8 +711,8 @@ static void pvr2_v4l2_dev_destroy(struct pvr2_v4l2_dev *dip) dip->devbase.minor,pvr2_config_get_name(dip->config)); /* Paranoia */ - dip->v4lp = 0; - dip->stream = 0; + dip->v4lp = NULL; + dip->stream = NULL; /* Actual deallocation happens later when all internal references are gone. */ @@ -1076,7 +1076,7 @@ struct pvr2_v4l2 *pvr2_v4l2_create(struct pvr2_context *mnp) vp->vdev = kmalloc(sizeof(*vp->vdev),GFP_KERNEL); if (!vp->vdev) { kfree(vp); - return 0; + return NULL; } memset(vp->vdev,0,sizeof(*vp->vdev)); pvr2_channel_init(&vp->channel,mnp); -- cgit v0.10.2 From 9e741b74afc975da51ec60c5a8147b2ebcf7e33a Mon Sep 17 00:00:00 2001 From: Raymond Mantchala Date: Mon, 30 Oct 2006 23:20:50 -0300 Subject: V4L/DVB (4787): Budget-ci: Inversion setting fixed for Technotrend 1500 T Technotrend 1500 T card have "inverted inversion". This patch fixes that. Many thanks to Martin Zwickel from Technotrend for his confirmation and correction proposal. Signed-off-by: Raymond Mantchala Signed-off-by: Perceval Anichini Signed-off-by: Oliver Endriss Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/ttpci/budget-ci.c b/drivers/media/dvb/ttpci/budget-ci.c index ac0cecb..cd5ec48 100644 --- a/drivers/media/dvb/ttpci/budget-ci.c +++ b/drivers/media/dvb/ttpci/budget-ci.c @@ -1035,6 +1035,7 @@ static void frontend_init(struct budget_ci *budget_ci) case 0x1012: // TT DVB-T CI budget (tda10046/Philips tdm1316l(tda6651tt)) budget_ci->tuner_pll_address = 0x60; + philips_tdm1316l_config.invert = 1; budget_ci->budget.dvb_frontend = dvb_attach(tda10046_attach, &philips_tdm1316l_config, &budget_ci->budget.i2c_adap); if (budget_ci->budget.dvb_frontend) { -- cgit v0.10.2 From 588f98312c7fd1d86290583189d2eb24da70f752 Mon Sep 17 00:00:00 2001 From: Hartmut Hackmann Date: Wed, 18 Oct 2006 17:30:42 -0300 Subject: V4L/DVB (4770): Fix mode switch of Compro Videomate T300 The board did not return to analog mode since the board specific "demod sleep" function was not called. Signed-off-by: Hartmut Hackmann Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/saa7134/saa7134-dvb.c b/drivers/media/video/saa7134/saa7134-dvb.c index 1ba53b5..6b61d9b 100644 --- a/drivers/media/video/saa7134/saa7134-dvb.c +++ b/drivers/media/video/saa7134/saa7134-dvb.c @@ -1147,6 +1147,8 @@ static int dvb_init(struct saa7134_dev *dev) &philips_europa_config, &dev->i2c_adap); if (dev->dvb.frontend) { + dev->original_demod_sleep = dev->dvb.frontend->ops.sleep; + dev->dvb.frontend->ops.sleep = philips_europa_demod_sleep; dev->dvb.frontend->ops.tuner_ops.init = philips_europa_tuner_init; dev->dvb.frontend->ops.tuner_ops.sleep = philips_europa_tuner_sleep; dev->dvb.frontend->ops.tuner_ops.set_params = philips_td1316_tuner_set_params; -- cgit v0.10.2 From 9bb6e2593ad4cb94944f547154baee64b4734598 Mon Sep 17 00:00:00 2001 From: Oliver Endriss Date: Fri, 27 Oct 2006 18:02:01 -0300 Subject: V4L/DVB (4784): [saa7146_i2c] short_delay mode fixed for fast machines TT DVB-C 2300 runs at 137 kHz I2C speed. short_delay mode did not work reliably on fast machines with that speed. Increased max loop count from 20 to 50. Moved dummy access out of the loop. Signed-off-by: Oliver Endriss Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/common/saa7146_i2c.c b/drivers/media/common/saa7146_i2c.c index d9953f7..5297a36 100644 --- a/drivers/media/common/saa7146_i2c.c +++ b/drivers/media/common/saa7146_i2c.c @@ -217,11 +217,9 @@ static int saa7146_i2c_writeout(struct saa7146_dev *dev, u32* dword, int short_d } /* wait until we get a transfer done or error */ timeout = jiffies + HZ/100 + 1; /* 10ms */ + /* first read usually delivers bogus results... */ + saa7146_i2c_status(dev); while(1) { - /** - * first read usually delivers bogus results... - */ - saa7146_i2c_status(dev); status = saa7146_i2c_status(dev); if ((status & 0x3) != 1) break; @@ -232,10 +230,10 @@ static int saa7146_i2c_writeout(struct saa7146_dev *dev, u32* dword, int short_d DEB_I2C(("saa7146_i2c_writeout: timed out waiting for end of xfer\n")); return -EIO; } - if ((++trial < 20) && short_delay) + if (++trial < 50 && short_delay) udelay(10); else - msleep(1); + msleep(1); } } -- cgit v0.10.2 From c5dec9fb248e3318f30a26f9984b3b064053a77f Mon Sep 17 00:00:00 2001 From: Trent Piepho Date: Sat, 14 Oct 2006 15:44:44 -0300 Subject: V4L/DVB (4751): Fix DBV_FE_CUSTOMISE for card drivers compiled into kernel When a front-end is disabled, card drivers that use it are compiled with a stub version of the front-end's attach function. This way they have no references to the front-end's code and don't need it to be loaded. If a card driver is compiled into the kernel, and a front-end is a module, then that front-end is effectively disabled wrt the card driver. In this case, the card driver should get the stub version. This was not happening. The stub vs real attach function selection is changed so that when the front-end is a module the real attach function is only used if the card driver is a module as well. This means a module front-end will be supported by card drivers that are modules and not supported by card drivers compiled into the kernel. Signed-off-by: Trent Piepho Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/frontends/bcm3510.h b/drivers/media/dvb/frontends/bcm3510.h index 6dfa839..7e4f95e 100644 --- a/drivers/media/dvb/frontends/bcm3510.h +++ b/drivers/media/dvb/frontends/bcm3510.h @@ -34,7 +34,7 @@ struct bcm3510_config int (*request_firmware)(struct dvb_frontend* fe, const struct firmware **fw, char* name); }; -#if defined(CONFIG_DVB_BCM3510) || defined(CONFIG_DVB_BCM3510_MODULE) +#if defined(CONFIG_DVB_BCM3510) || (defined(CONFIG_DVB_BCM3510_MODULE) && defined(MODULE)) extern struct dvb_frontend* bcm3510_attach(const struct bcm3510_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/cx22700.h b/drivers/media/dvb/frontends/cx22700.h index 10286cc..7ac3369 100644 --- a/drivers/media/dvb/frontends/cx22700.h +++ b/drivers/media/dvb/frontends/cx22700.h @@ -31,7 +31,7 @@ struct cx22700_config u8 demod_address; }; -#if defined(CONFIG_DVB_CX22700) || defined(CONFIG_DVB_CX22700_MODULE) +#if defined(CONFIG_DVB_CX22700) || (defined(CONFIG_DVB_CX22700_MODULE) && defined(MODULE)) extern struct dvb_frontend* cx22700_attach(const struct cx22700_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/cx22702.h b/drivers/media/dvb/frontends/cx22702.h index bc217dd..9cd64da 100644 --- a/drivers/media/dvb/frontends/cx22702.h +++ b/drivers/media/dvb/frontends/cx22702.h @@ -41,7 +41,7 @@ struct cx22702_config u8 output_mode; }; -#if defined(CONFIG_DVB_CX22702) || defined(CONFIG_DVB_CX22702_MODULE) +#if defined(CONFIG_DVB_CX22702) || (defined(CONFIG_DVB_CX22702_MODULE) && defined(MODULE)) extern struct dvb_frontend* cx22702_attach(const struct cx22702_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/cx24110.h b/drivers/media/dvb/frontends/cx24110.h index c9d5ae2..0ca3af4 100644 --- a/drivers/media/dvb/frontends/cx24110.h +++ b/drivers/media/dvb/frontends/cx24110.h @@ -41,7 +41,7 @@ static inline int cx24110_pll_write(struct dvb_frontend *fe, u32 val) { return r; } -#if defined(CONFIG_DVB_CX24110) || defined(CONFIG_DVB_CX24110_MODULE) +#if defined(CONFIG_DVB_CX24110) || (defined(CONFIG_DVB_CX24110_MODULE) && defined(MODULE)) extern struct dvb_frontend* cx24110_attach(const struct cx24110_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/cx24123.h b/drivers/media/dvb/frontends/cx24123.h index 57a1dae..84f9e4f 100644 --- a/drivers/media/dvb/frontends/cx24123.h +++ b/drivers/media/dvb/frontends/cx24123.h @@ -35,7 +35,7 @@ struct cx24123_config int lnb_polarity; }; -#if defined(CONFIG_DVB_CX24123) || defined(CONFIG_DVB_CX24123_MODULE) +#if defined(CONFIG_DVB_CX24123) || (defined(CONFIG_DVB_CX24123_MODULE) && defined(MODULE)) extern struct dvb_frontend* cx24123_attach(const struct cx24123_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/dib3000.h b/drivers/media/dvb/frontends/dib3000.h index 0caac3f..a6d3854 100644 --- a/drivers/media/dvb/frontends/dib3000.h +++ b/drivers/media/dvb/frontends/dib3000.h @@ -41,7 +41,7 @@ struct dib_fe_xfer_ops int (*tuner_pass_ctrl)(struct dvb_frontend *fe, int onoff, u8 pll_ctrl); }; -#if defined(CONFIG_DVB_DIB3000MB) || defined(CONFIG_DVB_DIB3000MB_MODULE) +#if defined(CONFIG_DVB_DIB3000MB) || (defined(CONFIG_DVB_DIB3000MB_MODULE) && defined(MODULE)) extern struct dvb_frontend* dib3000mb_attach(const struct dib3000_config* config, struct i2c_adapter* i2c, struct dib_fe_xfer_ops *xfer_ops); #else diff --git a/drivers/media/dvb/frontends/dib3000mc.h b/drivers/media/dvb/frontends/dib3000mc.h index 0d6fdef..72d4757 100644 --- a/drivers/media/dvb/frontends/dib3000mc.h +++ b/drivers/media/dvb/frontends/dib3000mc.h @@ -39,7 +39,7 @@ struct dib3000mc_config { #define DEFAULT_DIB3000MC_I2C_ADDRESS 16 #define DEFAULT_DIB3000P_I2C_ADDRESS 24 -#if defined(CONFIG_DVB_DIB3000MC) || defined(CONFIG_DVB_DIB3000MC_MODULE) +#if defined(CONFIG_DVB_DIB3000MC) || (defined(CONFIG_DVB_DIB3000MC_MODULE) && defined(MODULE)) extern struct dvb_frontend * dib3000mc_attach(struct i2c_adapter *i2c_adap, u8 i2c_addr, struct dib3000mc_config *cfg); #else static inline struct dvb_frontend * dib3000mc_attach(struct i2c_adapter *i2c_adap, u8 i2c_addr, struct dib3000mc_config *cfg) diff --git a/drivers/media/dvb/frontends/isl6421.h b/drivers/media/dvb/frontends/isl6421.h index 1916e3e..ea7f78a 100644 --- a/drivers/media/dvb/frontends/isl6421.h +++ b/drivers/media/dvb/frontends/isl6421.h @@ -39,7 +39,7 @@ #define ISL6421_ISEL1 0x20 #define ISL6421_DCL 0x40 -#if defined(CONFIG_DVB_ISL6421) || defined(CONFIG_DVB_ISL6421_MODULE) +#if defined(CONFIG_DVB_ISL6421) || (defined(CONFIG_DVB_ISL6421_MODULE) && defined(MODULE)) /* override_set and override_clear control which system register bits (above) to always set & clear */ extern struct dvb_frontend *isl6421_attach(struct dvb_frontend *fe, struct i2c_adapter *i2c, u8 i2c_addr, u8 override_set, u8 override_clear); diff --git a/drivers/media/dvb/frontends/l64781.h b/drivers/media/dvb/frontends/l64781.h index 21ba4a2..cd15f76 100644 --- a/drivers/media/dvb/frontends/l64781.h +++ b/drivers/media/dvb/frontends/l64781.h @@ -31,7 +31,7 @@ struct l64781_config u8 demod_address; }; -#if defined(CONFIG_DVB_L64781) || defined(CONFIG_DVB_L64781_MODULE) +#if defined(CONFIG_DVB_L64781) || (defined(CONFIG_DVB_L64781_MODULE) && defined(MODULE)) extern struct dvb_frontend* l64781_attach(const struct l64781_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/lgdt330x.h b/drivers/media/dvb/frontends/lgdt330x.h index 3f96b48..9950590 100644 --- a/drivers/media/dvb/frontends/lgdt330x.h +++ b/drivers/media/dvb/frontends/lgdt330x.h @@ -52,7 +52,7 @@ struct lgdt330x_config int clock_polarity_flip; }; -#if defined(CONFIG_DVB_LGDT330X) || defined(CONFIG_DVB_LGDT330X_MODULE) +#if defined(CONFIG_DVB_LGDT330X) || (defined(CONFIG_DVB_LGDT330X_MODULE) && defined(MODULE)) extern struct dvb_frontend* lgdt330x_attach(const struct lgdt330x_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/lnbp21.h b/drivers/media/dvb/frontends/lnbp21.h index 1fe1dd1..68906ac 100644 --- a/drivers/media/dvb/frontends/lnbp21.h +++ b/drivers/media/dvb/frontends/lnbp21.h @@ -39,7 +39,7 @@ #include -#if defined(CONFIG_DVB_LNBP21) || defined(CONFIG_DVB_LNBP21_MODULE) +#if defined(CONFIG_DVB_LNBP21) || (defined(CONFIG_DVB_LNBP21_MODULE) && defined(MODULE)) /* override_set and override_clear control which system register bits (above) to always set & clear */ extern struct dvb_frontend *lnbp21_attach(struct dvb_frontend *fe, struct i2c_adapter *i2c, u8 override_set, u8 override_clear); #else diff --git a/drivers/media/dvb/frontends/mt312.h b/drivers/media/dvb/frontends/mt312.h index 7112fb4..cf9a150 100644 --- a/drivers/media/dvb/frontends/mt312.h +++ b/drivers/media/dvb/frontends/mt312.h @@ -34,7 +34,7 @@ struct mt312_config u8 demod_address; }; -#if defined(CONFIG_DVB_MT312) || defined(CONFIG_DVB_MT312_MODULE) +#if defined(CONFIG_DVB_MT312) || (defined(CONFIG_DVB_MT312_MODULE) && defined(MODULE)) struct dvb_frontend* vp310_mt312_attach(const struct mt312_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/mt352.h b/drivers/media/dvb/frontends/mt352.h index 0035c2e..e996408 100644 --- a/drivers/media/dvb/frontends/mt352.h +++ b/drivers/media/dvb/frontends/mt352.h @@ -51,7 +51,7 @@ struct mt352_config int (*demod_init)(struct dvb_frontend* fe); }; -#if defined(CONFIG_DVB_MT352) || defined(CONFIG_DVB_MT352_MODULE) +#if defined(CONFIG_DVB_MT352) || (defined(CONFIG_DVB_MT352_MODULE) && defined(MODULE)) extern struct dvb_frontend* mt352_attach(const struct mt352_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/nxt200x.h b/drivers/media/dvb/frontends/nxt200x.h index 2eb220e..28bc559 100644 --- a/drivers/media/dvb/frontends/nxt200x.h +++ b/drivers/media/dvb/frontends/nxt200x.h @@ -45,7 +45,7 @@ struct nxt200x_config int (*set_ts_params)(struct dvb_frontend* fe, int is_punctured); }; -#if defined(CONFIG_DVB_NXT200X) || defined(CONFIG_DVB_NXT200X_MODULE) +#if defined(CONFIG_DVB_NXT200X) || (defined(CONFIG_DVB_NXT200X_MODULE) && defined(MODULE)) extern struct dvb_frontend* nxt200x_attach(const struct nxt200x_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/nxt6000.h b/drivers/media/dvb/frontends/nxt6000.h index 9397393..13d2251 100644 --- a/drivers/media/dvb/frontends/nxt6000.h +++ b/drivers/media/dvb/frontends/nxt6000.h @@ -33,7 +33,7 @@ struct nxt6000_config u8 clock_inversion:1; }; -#if defined(CONFIG_DVB_NXT6000) || defined(CONFIG_DVB_NXT6000_MODULE) +#if defined(CONFIG_DVB_NXT6000) || (defined(CONFIG_DVB_NXT6000_MODULE) && defined(MODULE)) extern struct dvb_frontend* nxt6000_attach(const struct nxt6000_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/or51132.h b/drivers/media/dvb/frontends/or51132.h index 9718be4..add24f0 100644 --- a/drivers/media/dvb/frontends/or51132.h +++ b/drivers/media/dvb/frontends/or51132.h @@ -34,7 +34,7 @@ struct or51132_config int (*set_ts_params)(struct dvb_frontend* fe, int is_punctured); }; -#if defined(CONFIG_DVB_OR51132) || defined(CONFIG_DVB_OR51132_MODULE) +#if defined(CONFIG_DVB_OR51132) || (defined(CONFIG_DVB_OR51132_MODULE) && defined(MODULE)) extern struct dvb_frontend* or51132_attach(const struct or51132_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/or51211.h b/drivers/media/dvb/frontends/or51211.h index 10a5419..8aad840 100644 --- a/drivers/media/dvb/frontends/or51211.h +++ b/drivers/media/dvb/frontends/or51211.h @@ -37,7 +37,7 @@ struct or51211_config void (*sleep)(struct dvb_frontend * fe); }; -#if defined(CONFIG_DVB_OR51211) || defined(CONFIG_DVB_OR51211_MODULE) +#if defined(CONFIG_DVB_OR51211) || (defined(CONFIG_DVB_OR51211_MODULE) && defined(MODULE)) extern struct dvb_frontend* or51211_attach(const struct or51211_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/s5h1420.h b/drivers/media/dvb/frontends/s5h1420.h index efc54d7..1555870 100644 --- a/drivers/media/dvb/frontends/s5h1420.h +++ b/drivers/media/dvb/frontends/s5h1420.h @@ -34,7 +34,7 @@ struct s5h1420_config u8 invert:1; }; -#if defined(CONFIG_DVB_S5H1420) || defined(CONFIG_DVB_S5H1420_MODULE) +#if defined(CONFIG_DVB_S5H1420) || (defined(CONFIG_DVB_S5H1420_MODULE) && defined(MODULE)) extern struct dvb_frontend* s5h1420_attach(const struct s5h1420_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/sp8870.h b/drivers/media/dvb/frontends/sp8870.h index 4cf27d3..909cefe 100644 --- a/drivers/media/dvb/frontends/sp8870.h +++ b/drivers/media/dvb/frontends/sp8870.h @@ -35,7 +35,7 @@ struct sp8870_config int (*request_firmware)(struct dvb_frontend* fe, const struct firmware **fw, char* name); }; -#if defined(CONFIG_DVB_SP8870) || defined(CONFIG_DVB_SP8870_MODULE) +#if defined(CONFIG_DVB_SP8870) || (defined(CONFIG_DVB_SP8870_MODULE) && defined(MODULE)) extern struct dvb_frontend* sp8870_attach(const struct sp8870_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/sp887x.h b/drivers/media/dvb/frontends/sp887x.h index cab7ea6..7ee78d7 100644 --- a/drivers/media/dvb/frontends/sp887x.h +++ b/drivers/media/dvb/frontends/sp887x.h @@ -17,7 +17,7 @@ struct sp887x_config int (*request_firmware)(struct dvb_frontend* fe, const struct firmware **fw, char* name); }; -#if defined(CONFIG_DVB_SP887X) || defined(CONFIG_DVB_SP887X_MODULE) +#if defined(CONFIG_DVB_SP887X) || (defined(CONFIG_DVB_SP887X_MODULE) && defined(MODULE)) extern struct dvb_frontend* sp887x_attach(const struct sp887x_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/stv0297.h b/drivers/media/dvb/frontends/stv0297.h index 760b80d..69f4515 100644 --- a/drivers/media/dvb/frontends/stv0297.h +++ b/drivers/media/dvb/frontends/stv0297.h @@ -42,7 +42,7 @@ struct stv0297_config u8 stop_during_read:1; }; -#if defined(CONFIG_DVB_STV0297) || defined(CONFIG_DVB_STV0297_MODULE) +#if defined(CONFIG_DVB_STV0297) || (defined(CONFIG_DVB_STV0297_MODULE) && defined(MODULE)) extern struct dvb_frontend* stv0297_attach(const struct stv0297_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/stv0299.h b/drivers/media/dvb/frontends/stv0299.h index 7ef2520..33df949 100644 --- a/drivers/media/dvb/frontends/stv0299.h +++ b/drivers/media/dvb/frontends/stv0299.h @@ -89,7 +89,7 @@ struct stv0299_config int (*set_symbol_rate)(struct dvb_frontend* fe, u32 srate, u32 ratio); }; -#if defined(CONFIG_DVB_STV0299) || defined(CONFIG_DVB_STV0299_MODULE) +#if defined(CONFIG_DVB_STV0299) || (defined(CONFIG_DVB_STV0299_MODULE) && defined(MODULE)) extern struct dvb_frontend* stv0299_attach(const struct stv0299_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/tda10021.h b/drivers/media/dvb/frontends/tda10021.h index d68ae20..e3da780 100644 --- a/drivers/media/dvb/frontends/tda10021.h +++ b/drivers/media/dvb/frontends/tda10021.h @@ -32,7 +32,7 @@ struct tda10021_config u8 demod_address; }; -#if defined(CONFIG_DVB_TDA10021) || defined(CONFIG_DVB_TDA10021_MODULE) +#if defined(CONFIG_DVB_TDA10021) || (defined(CONFIG_DVB_TDA10021_MODULE) && defined(MODULE)) extern struct dvb_frontend* tda10021_attach(const struct tda10021_config* config, struct i2c_adapter* i2c, u8 pwm); #else diff --git a/drivers/media/dvb/frontends/tda1004x.h b/drivers/media/dvb/frontends/tda1004x.h index e28fca0..605ad2d 100644 --- a/drivers/media/dvb/frontends/tda1004x.h +++ b/drivers/media/dvb/frontends/tda1004x.h @@ -71,7 +71,7 @@ struct tda1004x_config int (*request_firmware)(struct dvb_frontend* fe, const struct firmware **fw, char* name); }; -#if defined(CONFIG_DVB_TDA1004X) || defined(CONFIG_DVB_TDA1004X_MODULE) +#if defined(CONFIG_DVB_TDA1004X) || (defined(CONFIG_DVB_TDA1004X_MODULE) && defined(MODULE)) extern struct dvb_frontend* tda10045_attach(const struct tda1004x_config* config, struct i2c_adapter* i2c); diff --git a/drivers/media/dvb/frontends/tda10086.h b/drivers/media/dvb/frontends/tda10086.h index 18457ad..ed584a8 100644 --- a/drivers/media/dvb/frontends/tda10086.h +++ b/drivers/media/dvb/frontends/tda10086.h @@ -35,7 +35,7 @@ struct tda10086_config u8 invert; }; -#if defined(CONFIG_DVB_TDA10086) || defined(CONFIG_DVB_TDA10086_MODULE) +#if defined(CONFIG_DVB_TDA10086) || (defined(CONFIG_DVB_TDA10086_MODULE) && defined(MODULE)) extern struct dvb_frontend* tda10086_attach(const struct tda10086_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/tda8083.h b/drivers/media/dvb/frontends/tda8083.h index aae15bd..2d33079 100644 --- a/drivers/media/dvb/frontends/tda8083.h +++ b/drivers/media/dvb/frontends/tda8083.h @@ -35,7 +35,7 @@ struct tda8083_config u8 demod_address; }; -#if defined(CONFIG_DVB_TDA8083) || defined(CONFIG_DVB_TDA8083_MODULE) +#if defined(CONFIG_DVB_TDA8083) || (defined(CONFIG_DVB_TDA8083_MODULE) && defined(MODULE)) extern struct dvb_frontend* tda8083_attach(const struct tda8083_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/tda826x.h b/drivers/media/dvb/frontends/tda826x.h index 83998c0..ad99811 100644 --- a/drivers/media/dvb/frontends/tda826x.h +++ b/drivers/media/dvb/frontends/tda826x.h @@ -35,7 +35,7 @@ * @param has_loopthrough Set to 1 if the card has a loopthrough RF connector. * @return FE pointer on success, NULL on failure. */ -#if defined(CONFIG_DVB_TDA826X) || defined(CONFIG_DVB_TDA826X_MODULE) +#if defined(CONFIG_DVB_TDA826X) || (defined(CONFIG_DVB_TDA826X_MODULE) && defined(MODULE)) extern struct dvb_frontend* tda826x_attach(struct dvb_frontend *fe, int addr, struct i2c_adapter *i2c, int has_loopthrough); diff --git a/drivers/media/dvb/frontends/tua6100.h b/drivers/media/dvb/frontends/tua6100.h index 8f98033..03a665e 100644 --- a/drivers/media/dvb/frontends/tua6100.h +++ b/drivers/media/dvb/frontends/tua6100.h @@ -34,7 +34,7 @@ #include #include "dvb_frontend.h" -#if defined(CONFIG_DVB_TUA6100) || defined(CONFIG_DVB_TUA6100_MODULE) +#if defined(CONFIG_DVB_TUA6100) || (defined(CONFIG_DVB_TUA6100_MODULE) && defined(MODULE)) extern struct dvb_frontend *tua6100_attach(struct dvb_frontend *fe, int addr, struct i2c_adapter *i2c); #else static inline struct dvb_frontend* tua6100_attach(struct dvb_frontend *fe, int addr, struct i2c_adapter *i2c) diff --git a/drivers/media/dvb/frontends/ves1820.h b/drivers/media/dvb/frontends/ves1820.h index f0c9dde..e4a2a32 100644 --- a/drivers/media/dvb/frontends/ves1820.h +++ b/drivers/media/dvb/frontends/ves1820.h @@ -41,7 +41,7 @@ struct ves1820_config u8 selagc:1; }; -#if defined(CONFIG_DVB_VES1820) || defined(CONFIG_DVB_VES1820_MODULE) +#if defined(CONFIG_DVB_VES1820) || (defined(CONFIG_DVB_VES1820_MODULE) && defined(MODULE)) extern struct dvb_frontend* ves1820_attach(const struct ves1820_config* config, struct i2c_adapter* i2c, u8 pwm); #else diff --git a/drivers/media/dvb/frontends/ves1x93.h b/drivers/media/dvb/frontends/ves1x93.h index 395fed3..d507f89 100644 --- a/drivers/media/dvb/frontends/ves1x93.h +++ b/drivers/media/dvb/frontends/ves1x93.h @@ -40,7 +40,7 @@ struct ves1x93_config u8 invert_pwm:1; }; -#if defined(CONFIG_DVB_VES1X93) || defined(CONFIG_DVB_VES1X93_MODULE) +#if defined(CONFIG_DVB_VES1X93) || (defined(CONFIG_DVB_VES1X93_MODULE) && defined(MODULE)) extern struct dvb_frontend* ves1x93_attach(const struct ves1x93_config* config, struct i2c_adapter* i2c); #else diff --git a/drivers/media/dvb/frontends/zl10353.h b/drivers/media/dvb/frontends/zl10353.h index 79a9472..0bc0109 100644 --- a/drivers/media/dvb/frontends/zl10353.h +++ b/drivers/media/dvb/frontends/zl10353.h @@ -36,7 +36,7 @@ struct zl10353_config int parallel_ts; }; -#if defined(CONFIG_DVB_ZL10353) || defined(CONFIG_DVB_ZL10353_MODULE) +#if defined(CONFIG_DVB_ZL10353) || (defined(CONFIG_DVB_ZL10353_MODULE) && defined(MODULE)) extern struct dvb_frontend* zl10353_attach(const struct zl10353_config *config, struct i2c_adapter *i2c); #else -- cgit v0.10.2 From 6851ecc6e2fa4a01449a0fec9f4abd9aec43afde Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 2 Nov 2006 23:02:24 -0800 Subject: PCI: Revert "PCI: i386/x86_84: disable PCI resource decode on device disable" This reverts commit 53e4d30dd666d7f83598957ee4a415eefb47c9a6. It was found that it caused unneeded problems (see http://bugzilla.kernel.org/show_bug.cgi?id=7082 for details of one such issue. Signed-off-by: Greg Kroah-Hartman diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index 6d5ace8..cdfcf97 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -343,7 +343,6 @@ int pcibios_enable_device(struct pci_dev *dev, int mask) void pcibios_disable_device (struct pci_dev *dev) { - pcibios_disable_resources(dev); if (pcibios_disable_irq) pcibios_disable_irq(dev); } diff --git a/arch/i386/pci/i386.c b/arch/i386/pci/i386.c index 10154a2..9858029 100644 --- a/arch/i386/pci/i386.c +++ b/arch/i386/pci/i386.c @@ -242,15 +242,6 @@ int pcibios_enable_resources(struct pci_dev *dev, int mask) return 0; } -void pcibios_disable_resources(struct pci_dev *dev) -{ - u16 cmd; - - pci_read_config_word(dev, PCI_COMMAND, &cmd); - cmd &= ~(PCI_COMMAND_IO | PCI_COMMAND_MEMORY); - pci_write_config_word(dev, PCI_COMMAND, cmd); -} - /* * If we set up a device for bus mastering, we need to check the latency * timer as certain crappy BIOSes forget to set it properly. diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h index ad065ce..a0a2518 100644 --- a/arch/i386/pci/pci.h +++ b/arch/i386/pci/pci.h @@ -43,7 +43,6 @@ extern unsigned int pcibios_max_latency; void pcibios_resource_survey(void); int pcibios_enable_resources(struct pci_dev *, int); -void pcibios_disable_resources(struct pci_dev *); /* pci-pc.c */ -- cgit v0.10.2 From bb44c308ee37c14ab63251e27d6d8b4dc73a10a4 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 27 Oct 2006 16:12:30 -0700 Subject: PCI: Let PCI_MULTITHREAD_PROBE depend on BROKEN PCI_MULTITHREAD_PROBE is an interesting feature, but in its current state it seems to be more of a trap for users who accidentally enable it. This patch lets PCI_MULTITHREAD_PROBE depend on BROKEN for 2.6.19. The intention is to get this patch reversed in -mm as soon as it's in Linus' tree, and reverse it for 2.6.20 or 2.6.21 after the fallout of in-kernel problems PCI_MULTITHREAD_PROBE causes got fixed. (akpm: I get enough bug reports already) Signed-off-by: Adrian Bunk Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index ecc50db..5f1b9f5 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -19,7 +19,7 @@ config PCI_MSI config PCI_MULTITHREAD_PROBE bool "PCI Multi-threaded probe (EXPERIMENTAL)" - depends on PCI && EXPERIMENTAL + depends on PCI && EXPERIMENTAL && BROKEN help Say Y here if you want the PCI core to spawn a new thread for every PCI device that is probed. This can cause a huge -- cgit v0.10.2 From 984d115bbf2d731ed2264031fe49c1378d730db0 Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Fri, 3 Nov 2006 01:47:20 +0100 Subject: [ARM] 3918/1: ixp4xx irq-chip rework This is a rework of the ixp4xx irq_chip implementation. The use of two irq_chip structures and potentially switching between them is a violation of the intended use of the IRQ framework. The current implementation does not work with current in-kernel spinlock debugging or lockdep due to lock recursion problems caused by calling set_irq_chip/handler from within the chip's set_irq_type(). This patch goes back to using one irq_chip structure and handling the differences between edge/level, normal/GPIO interrupts inside the ack/mask/unmask routines themselves. Signed-off-by: Kevin Hilman Signed-off-by: Deepak Saxena Signed-off-by: Russell King diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c index c7513f6..fbe288a 100644 --- a/arch/arm/mach-ixp4xx/common.c +++ b/arch/arm/mach-ixp4xx/common.c @@ -86,7 +86,8 @@ enum ixp4xx_irq_type { IXP4XX_IRQ_LEVEL, IXP4XX_IRQ_EDGE }; -static void ixp4xx_config_irq(unsigned irq, enum ixp4xx_irq_type type); +/* Each bit represents an IRQ: 1: edge-triggered, 0: level triggered */ +static unsigned long long ixp4xx_irq_edge = 0; /* * IRQ -> GPIO mapping table @@ -135,7 +136,11 @@ static int ixp4xx_set_irq_type(unsigned int irq, unsigned int type) default: return -EINVAL; } - ixp4xx_config_irq(irq, irq_type); + + if (irq_type == IXP4XX_IRQ_EDGE) + ixp4xx_irq_edge |= (1 << irq); + else + ixp4xx_irq_edge &= ~(1 << irq); if (line >= 8) { /* pins 8-15 */ line -= 8; @@ -167,14 +172,6 @@ static void ixp4xx_irq_mask(unsigned int irq) *IXP4XX_ICMR &= ~(1 << irq); } -static void ixp4xx_irq_unmask(unsigned int irq) -{ - if (cpu_is_ixp46x() && irq >= 32) - *IXP4XX_ICMR2 |= (1 << (irq - 32)); - else - *IXP4XX_ICMR |= (1 << irq); -} - static void ixp4xx_irq_ack(unsigned int irq) { int line = (irq < 32) ? irq2gpio[irq] : -1; @@ -187,41 +184,25 @@ static void ixp4xx_irq_ack(unsigned int irq) * Level triggered interrupts on GPIO lines can only be cleared when the * interrupt condition disappears. */ -static void ixp4xx_irq_level_unmask(unsigned int irq) +static void ixp4xx_irq_unmask(unsigned int irq) { - ixp4xx_irq_ack(irq); - ixp4xx_irq_unmask(irq); -} + if (!(ixp4xx_irq_edge & (1 << irq))) + ixp4xx_irq_ack(irq); -static struct irqchip ixp4xx_irq_level_chip = { - .ack = ixp4xx_irq_mask, - .mask = ixp4xx_irq_mask, - .unmask = ixp4xx_irq_level_unmask, - .set_type = ixp4xx_set_irq_type, -}; + if (cpu_is_ixp46x() && irq >= 32) + *IXP4XX_ICMR2 |= (1 << (irq - 32)); + else + *IXP4XX_ICMR |= (1 << irq); +} -static struct irqchip ixp4xx_irq_edge_chip = { +static struct irqchip ixp4xx_irq_chip = { + .name = "IXP4xx", .ack = ixp4xx_irq_ack, .mask = ixp4xx_irq_mask, .unmask = ixp4xx_irq_unmask, .set_type = ixp4xx_set_irq_type, }; -static void ixp4xx_config_irq(unsigned irq, enum ixp4xx_irq_type type) -{ - switch (type) { - case IXP4XX_IRQ_LEVEL: - set_irq_chip(irq, &ixp4xx_irq_level_chip); - set_irq_handler(irq, do_level_IRQ); - break; - case IXP4XX_IRQ_EDGE: - set_irq_chip(irq, &ixp4xx_irq_edge_chip); - set_irq_handler(irq, do_edge_IRQ); - break; - } - set_irq_flags(irq, IRQF_VALID); -} - void __init ixp4xx_init_irq(void) { int i = 0; @@ -241,8 +222,11 @@ void __init ixp4xx_init_irq(void) } /* Default to all level triggered */ - for(i = 0; i < NR_IRQS; i++) - ixp4xx_config_irq(i, IXP4XX_IRQ_LEVEL); + for(i = 0; i < NR_IRQS; i++) { + set_irq_chip(i, &ixp4xx_irq_chip); + set_irq_handler(i, do_level_IRQ); + set_irq_flags(i, IRQF_VALID); + } } -- cgit v0.10.2 From 1f4a39319e9226c3b1d5b91a1e4d3559ef8740e4 Mon Sep 17 00:00:00 2001 From: Enrico Scholz Date: Fri, 3 Nov 2006 13:47:39 +0100 Subject: [ARM] 3919/1: Fixed definition of some PXA270 CIF related registers Fixed definition of some CIF registers; see PXA27x Developer\'s Manual. Signed-off-by: Enrico Scholz Signed-off-by: Russell King diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h index 68731e0..cff752f 100644 --- a/include/asm-arm/arch-pxa/pxa-regs.h +++ b/include/asm-arm/arch-pxa/pxa-regs.h @@ -2242,7 +2242,7 @@ #define CICR1_TBIT (1 << 31) /* Transparency bit */ #define CICR1_RGBT_CONV (0x3 << 30) /* RGBT conversion mask */ -#define CICR1_PPL (0x3f << 15) /* Pixels per line mask */ +#define CICR1_PPL (0x7ff << 15) /* Pixels per line mask */ #define CICR1_RGB_CONV (0x7 << 12) /* RGB conversion mask */ #define CICR1_RGB_F (1 << 11) /* RGB format */ #define CICR1_YCBCR_F (1 << 10) /* YCbCr format */ @@ -2268,7 +2268,7 @@ #define CICR3_VSW (0x3f << 10) /* Vertical sync pulse width mask */ #define CICR3_BFPW (0x3f << 3) /* Beginning-of-frame pixel clock wait count mask */ -#define CICR3_LPF (0x3ff << 0) /* Lines per frame mask */ +#define CICR3_LPF (0x7ff << 0) /* Lines per frame mask */ #define CICR4_MCLK_DLY (0x3 << 24) /* MCLK Data Capture Delay mask */ #define CICR4_PCLK_EN (1 << 23) /* Pixel clock enable */ @@ -2289,8 +2289,8 @@ #define CISR_EOL (1 << 8) /* End of line */ #define CISR_PAR_ERR (1 << 7) /* Parity error */ #define CISR_CQD (1 << 6) /* Camera interface quick disable */ -#define CISR_SOF (1 << 5) /* Start of frame */ -#define CISR_CDD (1 << 4) /* Camera interface disable done */ +#define CISR_CDD (1 << 5) /* Camera interface disable done */ +#define CISR_SOF (1 << 4) /* Start of frame */ #define CISR_EOF (1 << 3) /* End of frame */ #define CISR_IFO_2 (1 << 2) /* FIFO overrun for Channel 2 */ #define CISR_IFO_1 (1 << 1) /* FIFO overrun for Channel 1 */ -- cgit v0.10.2 From d91f75fb761e1e691796287889774cc8690034c1 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 3 Nov 2006 18:30:12 +0100 Subject: [ARM] 3920/1: S3C24XX: Remove smdk2410_defconfig Remove the smdk2410_defconifg as it is out of data and has not been touched since 2.6.11. Use the s3c2410_defconfig instead. Signed-off-by: Ben Dooks Signed-off-by: Russell King diff --git a/arch/arm/configs/smdk2410_defconfig b/arch/arm/configs/smdk2410_defconfig deleted file mode 100644 index 4d123d3..0000000 --- a/arch/arm/configs/smdk2410_defconfig +++ /dev/null @@ -1,735 +0,0 @@ -# -# Automatically generated make config: don't edit -# Linux kernel version: 2.6.12-rc1-bk2 -# Sun Mar 27 22:42:40 2005 -# -CONFIG_ARM=y -CONFIG_MMU=y -CONFIG_UID16=y -CONFIG_RWSEM_GENERIC_SPINLOCK=y -CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_GENERIC_IOMAP=y - -# -# Code maturity level options -# -CONFIG_EXPERIMENTAL=y -CONFIG_CLEAN_COMPILE=y -CONFIG_BROKEN_ON_SMP=y - -# -# General setup -# -CONFIG_LOCALVERSION="" -CONFIG_SWAP=y -CONFIG_SYSVIPC=y -# CONFIG_POSIX_MQUEUE is not set -# CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y -# CONFIG_AUDIT is not set -# CONFIG_HOTPLUG is not set -CONFIG_KOBJECT_UEVENT=y -# CONFIG_IKCONFIG is not set -# CONFIG_EMBEDDED is not set -CONFIG_KALLSYMS=y -# CONFIG_KALLSYMS_ALL is not set -# CONFIG_KALLSYMS_EXTRA_PASS is not set -CONFIG_BASE_FULL=y -CONFIG_FUTEX=y -CONFIG_EPOLL=y -CONFIG_CC_OPTIMIZE_FOR_SIZE=y -CONFIG_SHMEM=y -CONFIG_CC_ALIGN_FUNCTIONS=0 -CONFIG_CC_ALIGN_LABELS=0 -CONFIG_CC_ALIGN_LOOPS=0 -CONFIG_CC_ALIGN_JUMPS=0 -# CONFIG_TINY_SHMEM is not set -CONFIG_BASE_SMALL=0 - -# -# Loadable module support -# -# CONFIG_MODULES is not set - -# -# System Type -# -# CONFIG_ARCH_CLPS7500 is not set -# CONFIG_ARCH_CLPS711X is not set -# CONFIG_ARCH_CO285 is not set -# CONFIG_ARCH_EBSA110 is not set -# CONFIG_ARCH_FOOTBRIDGE is not set -# CONFIG_ARCH_INTEGRATOR is not set -# CONFIG_ARCH_IOP3XX is not set -# CONFIG_ARCH_IXP4XX is not set -# CONFIG_ARCH_IXP2000 is not set -# CONFIG_ARCH_L7200 is not set -# CONFIG_ARCH_PXA is not set -# CONFIG_ARCH_RPC is not set -# CONFIG_ARCH_SA1100 is not set -CONFIG_ARCH_S3C2410=y -# CONFIG_ARCH_SHARK is not set -# CONFIG_ARCH_LH7A40X is not set -# CONFIG_ARCH_OMAP is not set -# CONFIG_ARCH_VERSATILE is not set -# CONFIG_ARCH_IMX is not set -# CONFIG_ARCH_H720X is not set - -# -# S3C24XX Implementations -# -# CONFIG_ARCH_BAST is not set -# CONFIG_ARCH_H1940 is not set -# CONFIG_MACH_N30 is not set -CONFIG_ARCH_SMDK2410=y -# CONFIG_ARCH_S3C2440 is not set -# CONFIG_MACH_VR1000 is not set -# CONFIG_MACH_RX3715 is not set -# CONFIG_MACH_OTOM is not set -# CONFIG_MACH_NEXCODER_2440 is not set -CONFIG_CPU_S3C2410=y - -# -# S3C2410 Boot -# - -# -# S3C2410 Setup -# -# CONFIG_S3C2410_DMA is not set -CONFIG_S3C2410_LOWLEVEL_UART_PORT=0 - -# -# Processor Type -# -CONFIG_CPU_32=y -CONFIG_CPU_ARM920T=y -CONFIG_CPU_32v4=y -CONFIG_CPU_ABRT_EV4T=y -CONFIG_CPU_CACHE_V4WT=y -CONFIG_CPU_CACHE_VIVT=y -CONFIG_CPU_COPY_V4WB=y -CONFIG_CPU_TLB_V4WBI=y - -# -# Processor Features -# -CONFIG_ARM_THUMB=y -# CONFIG_CPU_ICACHE_DISABLE is not set -# CONFIG_CPU_DCACHE_DISABLE is not set -# CONFIG_CPU_DCACHE_WRITETHROUGH is not set - -# -# Bus support -# - -# -# PCCARD (PCMCIA/CardBus) support -# -# CONFIG_PCCARD is not set - -# -# Kernel Features -# -# CONFIG_PREEMPT is not set -CONFIG_ALIGNMENT_TRAP=y - -# -# Boot options -# -CONFIG_ZBOOT_ROM_TEXT=0x0 -CONFIG_ZBOOT_ROM_BSS=0x0 -CONFIG_CMDLINE="root=1f04 mem=32M" -# CONFIG_XIP_KERNEL is not set - -# -# Floating point emulation -# - -# -# At least one emulation must be selected -# -# CONFIG_FPE_NWFPE is not set -# CONFIG_FPE_FASTFPE is not set - -# -# Userspace binary formats -# -CONFIG_BINFMT_ELF=y -CONFIG_BINFMT_AOUT=y -# CONFIG_BINFMT_MISC is not set -# CONFIG_ARTHUR is not set - -# -# Power management options -# -# CONFIG_PM is not set - -# -# Device Drivers -# - -# -# Generic Driver Options -# -CONFIG_STANDALONE=y -CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set -# CONFIG_DEBUG_DRIVER is not set - -# -# Memory Technology Devices (MTD) -# -CONFIG_MTD=y -# CONFIG_MTD_DEBUG is not set -# CONFIG_MTD_CONCAT is not set -# CONFIG_MTD_PARTITIONS is not set - -# -# User Modules And Translation Layers -# -CONFIG_MTD_CHAR=y -CONFIG_MTD_BLOCK=y -# CONFIG_FTL is not set -# CONFIG_NFTL is not set -# CONFIG_INFTL is not set - -# -# RAM/ROM/Flash chip drivers -# -CONFIG_MTD_CFI=y -# CONFIG_MTD_JEDECPROBE is not set -CONFIG_MTD_GEN_PROBE=y -# CONFIG_MTD_CFI_ADV_OPTIONS is not set -CONFIG_MTD_MAP_BANK_WIDTH_1=y -CONFIG_MTD_MAP_BANK_WIDTH_2=y -CONFIG_MTD_MAP_BANK_WIDTH_4=y -# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set -CONFIG_MTD_CFI_I1=y -CONFIG_MTD_CFI_I2=y -# CONFIG_MTD_CFI_I4 is not set -# CONFIG_MTD_CFI_I8 is not set -CONFIG_MTD_CFI_INTELEXT=y -# CONFIG_MTD_CFI_AMDSTD is not set -# CONFIG_MTD_CFI_STAA is not set -CONFIG_MTD_CFI_UTIL=y -# CONFIG_MTD_RAM is not set -# CONFIG_MTD_ROM is not set -# CONFIG_MTD_ABSENT is not set -# CONFIG_MTD_XIP is not set - -# -# Mapping drivers for chip access -# -# CONFIG_MTD_COMPLEX_MAPPINGS is not set -# CONFIG_MTD_PHYSMAP is not set -# CONFIG_MTD_ARM_INTEGRATOR is not set -# CONFIG_MTD_EDB7312 is not set - -# -# Self-contained MTD device drivers -# -# CONFIG_MTD_SLRAM is not set -# CONFIG_MTD_PHRAM is not set -# CONFIG_MTD_MTDRAM is not set -# CONFIG_MTD_BLKMTD is not set -# CONFIG_MTD_BLOCK2MTD is not set - -# -# Disk-On-Chip Device Drivers -# -# CONFIG_MTD_DOC2000 is not set -# CONFIG_MTD_DOC2001 is not set -# CONFIG_MTD_DOC2001PLUS is not set - -# -# NAND Flash Device Drivers -# -# CONFIG_MTD_NAND is not set - -# -# Parallel port support -# -# CONFIG_PARPORT is not set - -# -# Plug and Play support -# - -# -# Block devices -# -# CONFIG_BLK_DEV_FD is not set -# CONFIG_BLK_DEV_COW_COMMON is not set -# CONFIG_BLK_DEV_LOOP is not set -# CONFIG_BLK_DEV_NBD is not set -CONFIG_BLK_DEV_RAM=y -CONFIG_BLK_DEV_RAM_COUNT=16 -CONFIG_BLK_DEV_RAM_SIZE=4096 -# CONFIG_BLK_DEV_INITRD is not set -CONFIG_INITRAMFS_SOURCE="" -# CONFIG_CDROM_PKTCDVD is not set - -# -# IO Schedulers -# -CONFIG_IOSCHED_NOOP=y -CONFIG_IOSCHED_AS=y -CONFIG_IOSCHED_DEADLINE=y -CONFIG_IOSCHED_CFQ=y -# CONFIG_ATA_OVER_ETH is not set - -# -# ATA/ATAPI/MFM/RLL support -# -# CONFIG_IDE is not set - -# -# SCSI device support -# -# CONFIG_SCSI is not set - -# -# Multi-device support (RAID and LVM) -# -# CONFIG_MD is not set - -# -# Fusion MPT device support -# - -# -# IEEE 1394 (FireWire) support -# - -# -# I2O device support -# - -# -# Networking support -# -CONFIG_NET=y - -# -# Networking options -# -# CONFIG_PACKET is not set -# CONFIG_NETLINK_DEV is not set -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_PNP=y -# CONFIG_IP_PNP_DHCP is not set -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -# CONFIG_SYN_COOKIES is not set -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -# CONFIG_IP_TCPDIAG is not set -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set -CONFIG_NETDEVICES=y -# CONFIG_DUMMY is not set -# CONFIG_BONDING is not set -# CONFIG_EQUALIZER is not set -# CONFIG_TUN is not set - -# -# Ethernet (10 or 100Mbit) -# -CONFIG_NET_ETHERNET=y -# CONFIG_MII is not set -# CONFIG_SMC91X is not set - -# -# Ethernet (1000 Mbit) -# - -# -# Ethernet (10000 Mbit) -# - -# -# Token Ring devices -# - -# -# Wireless LAN (non-hamradio) -# -# CONFIG_NET_RADIO is not set - -# -# Wan interfaces -# -# CONFIG_WAN is not set -# CONFIG_PPP is not set -# CONFIG_SLIP is not set -# CONFIG_SHAPER is not set -# CONFIG_NETCONSOLE is not set - -# -# ISDN subsystem -# -# CONFIG_ISDN is not set - -# -# Input device support -# -CONFIG_INPUT=y - -# -# Userland interfaces -# -CONFIG_INPUT_MOUSEDEV=y -CONFIG_INPUT_MOUSEDEV_PSAUX=y -CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024 -CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768 -# CONFIG_INPUT_JOYDEV is not set -# CONFIG_INPUT_TSDEV is not set -# CONFIG_INPUT_EVDEV is not set -# CONFIG_INPUT_EVBUG is not set - -# -# Input Device Drivers -# -CONFIG_INPUT_KEYBOARD=y -CONFIG_KEYBOARD_ATKBD=y -# CONFIG_KEYBOARD_SUNKBD is not set -# CONFIG_KEYBOARD_LKKBD is not set -# CONFIG_KEYBOARD_XTKBD is not set -# CONFIG_KEYBOARD_NEWTON is not set -CONFIG_INPUT_MOUSE=y -CONFIG_MOUSE_PS2=y -# CONFIG_MOUSE_SERIAL is not set -# CONFIG_MOUSE_VSXXXAA is not set -# CONFIG_INPUT_JOYSTICK is not set -# CONFIG_INPUT_TOUCHSCREEN is not set -# CONFIG_INPUT_MISC is not set - -# -# Hardware I/O ports -# -CONFIG_SERIO=y -CONFIG_SERIO_SERPORT=y -CONFIG_SERIO_LIBPS2=y -# CONFIG_SERIO_RAW is not set -# CONFIG_GAMEPORT is not set -CONFIG_SOUND_GAMEPORT=y - -# -# Character devices -# -CONFIG_VT=y -CONFIG_VT_CONSOLE=y -CONFIG_HW_CONSOLE=y -# CONFIG_SERIAL_NONSTANDARD is not set - -# -# Serial drivers -# -# CONFIG_SERIAL_8250 is not set - -# -# Non-8250 serial port support -# -CONFIG_SERIAL_S3C2410=y -CONFIG_SERIAL_S3C2410_CONSOLE=y -CONFIG_SERIAL_CORE=y -CONFIG_SERIAL_CORE_CONSOLE=y -CONFIG_UNIX98_PTYS=y -CONFIG_LEGACY_PTYS=y -CONFIG_LEGACY_PTY_COUNT=256 - -# -# IPMI -# -# CONFIG_IPMI_HANDLER is not set - -# -# Watchdog Cards -# -# CONFIG_WATCHDOG is not set -# CONFIG_NVRAM is not set -# CONFIG_RTC is not set -# CONFIG_S3C2410_RTC is not set -# CONFIG_DTLK is not set -# CONFIG_R3964 is not set - -# -# Ftape, the floppy tape device driver -# -# CONFIG_DRM is not set -# CONFIG_RAW_DRIVER is not set - -# -# TPM devices -# -# CONFIG_TCG_TPM is not set - -# -# I2C support -# -# CONFIG_I2C is not set - -# -# Misc devices -# - -# -# Multimedia devices -# -# CONFIG_VIDEO_DEV is not set - -# -# Digital Video Broadcasting Devices -# -# CONFIG_DVB is not set - -# -# Graphics support -# -CONFIG_FB=y -CONFIG_FB_CFB_FILLRECT=y -CONFIG_FB_CFB_COPYAREA=y -CONFIG_FB_CFB_IMAGEBLIT=y -CONFIG_FB_SOFT_CURSOR=y -# CONFIG_FB_MODE_HELPERS is not set -# CONFIG_FB_TILEBLITTING is not set -CONFIG_FB_VIRTUAL=y - -# -# Console display driver support -# -# CONFIG_VGA_CONSOLE is not set -CONFIG_DUMMY_CONSOLE=y -CONFIG_FRAMEBUFFER_CONSOLE=y -# CONFIG_FONTS is not set -CONFIG_FONT_8x8=y -CONFIG_FONT_8x16=y - -# -# Logo configuration -# -# CONFIG_LOGO is not set -# CONFIG_BACKLIGHT_LCD_SUPPORT is not set - -# -# Sound -# -# CONFIG_SOUND is not set - -# -# USB support -# -CONFIG_USB_ARCH_HAS_HCD=y -# CONFIG_USB_ARCH_HAS_OHCI is not set -# CONFIG_USB is not set - -# -# USB Gadget Support -# -# CONFIG_USB_GADGET is not set - -# -# MMC/SD Card support -# -# CONFIG_MMC is not set - -# -# File systems -# -CONFIG_EXT2_FS=y -# CONFIG_EXT2_FS_XATTR is not set -# CONFIG_EXT3_FS is not set -# CONFIG_JBD is not set -# CONFIG_REISERFS_FS is not set -# CONFIG_JFS_FS is not set - -# -# XFS support -# -# CONFIG_XFS_FS is not set -# CONFIG_MINIX_FS is not set -CONFIG_ROMFS_FS=y -# CONFIG_QUOTA is not set -CONFIG_DNOTIFY=y -# CONFIG_AUTOFS_FS is not set -# CONFIG_AUTOFS4_FS is not set - -# -# CD-ROM/DVD Filesystems -# -# CONFIG_ISO9660_FS is not set -# CONFIG_UDF_FS is not set - -# -# DOS/FAT/NT Filesystems -# -# CONFIG_MSDOS_FS is not set -# CONFIG_VFAT_FS is not set -# CONFIG_NTFS_FS is not set - -# -# Pseudo filesystems -# -CONFIG_PROC_FS=y -CONFIG_SYSFS=y -# CONFIG_DEVFS_FS is not set -# CONFIG_DEVPTS_FS_XATTR is not set -# CONFIG_TMPFS is not set -# CONFIG_HUGETLB_PAGE is not set -CONFIG_RAMFS=y - -# -# Miscellaneous filesystems -# -# CONFIG_ADFS_FS is not set -# CONFIG_AFFS_FS is not set -# CONFIG_HFS_FS is not set -# CONFIG_HFSPLUS_FS is not set -# CONFIG_BEFS_FS is not set -# CONFIG_BFS_FS is not set -# CONFIG_EFS_FS is not set -# CONFIG_JFFS_FS is not set -# CONFIG_JFFS2_FS is not set -# CONFIG_CRAMFS is not set -# CONFIG_VXFS_FS is not set -# CONFIG_HPFS_FS is not set -# CONFIG_QNX4FS_FS is not set -# CONFIG_SYSV_FS is not set -# CONFIG_UFS_FS is not set - -# -# Network File Systems -# -CONFIG_NFS_FS=y -# CONFIG_NFS_V3 is not set -# CONFIG_NFS_V4 is not set -# CONFIG_NFS_DIRECTIO is not set -# CONFIG_NFSD is not set -CONFIG_ROOT_NFS=y -CONFIG_LOCKD=y -CONFIG_SUNRPC=y -# CONFIG_RPCSEC_GSS_KRB5 is not set -# CONFIG_RPCSEC_GSS_SPKM3 is not set -# CONFIG_SMB_FS is not set -# CONFIG_CIFS is not set -# CONFIG_NCP_FS is not set -# CONFIG_CODA_FS is not set -# CONFIG_AFS_FS is not set - -# -# Partition Types -# -CONFIG_PARTITION_ADVANCED=y -# CONFIG_ACORN_PARTITION is not set -# CONFIG_OSF_PARTITION is not set -# CONFIG_AMIGA_PARTITION is not set -# CONFIG_ATARI_PARTITION is not set -# CONFIG_MAC_PARTITION is not set -# CONFIG_MSDOS_PARTITION is not set -# CONFIG_LDM_PARTITION is not set -# CONFIG_SGI_PARTITION is not set -# CONFIG_ULTRIX_PARTITION is not set -# CONFIG_SUN_PARTITION is not set -# CONFIG_EFI_PARTITION is not set - -# -# Native Language Support -# -# CONFIG_NLS is not set - -# -# Profiling support -# -# CONFIG_PROFILING is not set - -# -# Kernel hacking -# -# CONFIG_PRINTK_TIME is not set -CONFIG_DEBUG_KERNEL=y -# CONFIG_MAGIC_SYSRQ is not set -CONFIG_LOG_BUF_SHIFT=14 -# CONFIG_SCHEDSTATS is not set -# CONFIG_DEBUG_SLAB is not set -# CONFIG_DEBUG_SPINLOCK is not set -# CONFIG_DEBUG_SPINLOCK_SLEEP is not set -# CONFIG_DEBUG_KOBJECT is not set -CONFIG_DEBUG_BUGVERBOSE=y -# CONFIG_DEBUG_INFO is not set -# CONFIG_DEBUG_FS is not set -CONFIG_FRAME_POINTER=y -CONFIG_DEBUG_USER=y -# CONFIG_DEBUG_WAITQ is not set -# CONFIG_DEBUG_ERRORS is not set -CONFIG_DEBUG_LL=y -# CONFIG_DEBUG_ICEDCC is not set -CONFIG_DEBUG_S3C2410_PORT=y -CONFIG_DEBUG_S3C2410_UART=0 - -# -# Security options -# -# CONFIG_KEYS is not set -# CONFIG_SECURITY is not set - -# -# Cryptographic options -# -# CONFIG_CRYPTO is not set - -# -# Hardware crypto devices -# - -# -# Library routines -# -# CONFIG_CRC_CCITT is not set -CONFIG_CRC32=y -CONFIG_LIBCRC32C=y -- cgit v0.10.2 From 73d15da44f34626b4ad18eb4d56e1c2c4aaed578 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 3 Nov 2006 18:33:43 +0100 Subject: [ARM] 3921/1: S3C24XX: remove bast_defconfig Remove the bast_defconfig, as it has not been updated since 2.6.13. The s3c2410_defconfig should be a good replacement. Signed-off-by: Ben Dooks Signed-off-by: Russell King diff --git a/arch/arm/configs/bast_defconfig b/arch/arm/configs/bast_defconfig deleted file mode 100644 index 4a8564f..0000000 --- a/arch/arm/configs/bast_defconfig +++ /dev/null @@ -1,947 +0,0 @@ -# -# Automatically generated make config: don't edit -# Linux kernel version: 2.6.12-rc1-bk2 -# Sun Mar 27 02:24:16 2005 -# -CONFIG_ARM=y -CONFIG_MMU=y -CONFIG_UID16=y -CONFIG_RWSEM_GENERIC_SPINLOCK=y -CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_GENERIC_IOMAP=y - -# -# Code maturity level options -# -CONFIG_EXPERIMENTAL=y -CONFIG_CLEAN_COMPILE=y -CONFIG_BROKEN_ON_SMP=y - -# -# General setup -# -CONFIG_LOCALVERSION="" -CONFIG_SWAP=y -CONFIG_SYSVIPC=y -# CONFIG_POSIX_MQUEUE is not set -# CONFIG_BSD_PROCESS_ACCT is not set -CONFIG_SYSCTL=y -# CONFIG_AUDIT is not set -# CONFIG_HOTPLUG is not set -CONFIG_KOBJECT_UEVENT=y -# CONFIG_IKCONFIG is not set -# CONFIG_EMBEDDED is not set -CONFIG_KALLSYMS=y -# CONFIG_KALLSYMS_ALL is not set -# CONFIG_KALLSYMS_EXTRA_PASS is not set -CONFIG_BASE_FULL=y -CONFIG_FUTEX=y -CONFIG_EPOLL=y -CONFIG_CC_OPTIMIZE_FOR_SIZE=y -CONFIG_SHMEM=y -CONFIG_CC_ALIGN_FUNCTIONS=0 -CONFIG_CC_ALIGN_LABELS=0 -CONFIG_CC_ALIGN_LOOPS=0 -CONFIG_CC_ALIGN_JUMPS=0 -# CONFIG_TINY_SHMEM is not set -CONFIG_BASE_SMALL=0 - -# -# Loadable module support -# -CONFIG_MODULES=y -# CONFIG_MODULE_UNLOAD is not set -CONFIG_OBSOLETE_MODPARM=y -# CONFIG_MODVERSIONS is not set -# CONFIG_MODULE_SRCVERSION_ALL is not set -CONFIG_KMOD=y - -# -# System Type -# -# CONFIG_ARCH_CLPS7500 is not set -# CONFIG_ARCH_CLPS711X is not set -# CONFIG_ARCH_CO285 is not set -# CONFIG_ARCH_EBSA110 is not set -# CONFIG_ARCH_FOOTBRIDGE is not set -# CONFIG_ARCH_INTEGRATOR is not set -# CONFIG_ARCH_IOP3XX is not set -# CONFIG_ARCH_IXP4XX is not set -# CONFIG_ARCH_IXP2000 is not set -# CONFIG_ARCH_L7200 is not set -# CONFIG_ARCH_PXA is not set -# CONFIG_ARCH_RPC is not set -# CONFIG_ARCH_SA1100 is not set -CONFIG_ARCH_S3C2410=y -# CONFIG_ARCH_SHARK is not set -# CONFIG_ARCH_LH7A40X is not set -# CONFIG_ARCH_OMAP is not set -# CONFIG_ARCH_VERSATILE is not set -# CONFIG_ARCH_IMX is not set -# CONFIG_ARCH_H720X is not set - -# -# S3C24XX Implementations -# -CONFIG_ARCH_BAST=y -# CONFIG_ARCH_H1940 is not set -# CONFIG_MACH_N30 is not set -# CONFIG_ARCH_SMDK2410 is not set -# CONFIG_ARCH_S3C2440 is not set -CONFIG_MACH_VR1000=y -# CONFIG_MACH_RX3715 is not set -# CONFIG_MACH_OTOM is not set -# CONFIG_MACH_NEXCODER_2440 is not set -CONFIG_CPU_S3C2410=y - -# -# S3C2410 Boot -# -# CONFIG_S3C2410_BOOT_WATCHDOG is not set - -# -# S3C2410 Setup -# -CONFIG_S3C2410_DMA=y -# CONFIG_S3C2410_DMA_DEBUG is not set -# CONFIG_S3C2410_PM_DEBUG is not set -# CONFIG_S3C2410_PM_CHECK is not set -CONFIG_S3C2410_LOWLEVEL_UART_PORT=0 - -# -# Processor Type -# -CONFIG_CPU_32=y -CONFIG_CPU_ARM920T=y -CONFIG_CPU_32v4=y -CONFIG_CPU_ABRT_EV4T=y -CONFIG_CPU_CACHE_V4WT=y -CONFIG_CPU_CACHE_VIVT=y -CONFIG_CPU_COPY_V4WB=y -CONFIG_CPU_TLB_V4WBI=y - -# -# Processor Features -# -# CONFIG_ARM_THUMB is not set -# CONFIG_CPU_ICACHE_DISABLE is not set -# CONFIG_CPU_DCACHE_DISABLE is not set -# CONFIG_CPU_DCACHE_WRITETHROUGH is not set - -# -# Bus support -# - -# -# PCCARD (PCMCIA/CardBus) support -# -# CONFIG_PCCARD is not set - -# -# Kernel Features -# -# CONFIG_PREEMPT is not set -CONFIG_ALIGNMENT_TRAP=y - -# -# Boot options -# -CONFIG_ZBOOT_ROM_TEXT=0x0 -CONFIG_ZBOOT_ROM_BSS=0x0 -CONFIG_CMDLINE="root=/dev/hda1 ro init=/bin/bash console=ttySAC0" -# CONFIG_XIP_KERNEL is not set - -# -# Floating point emulation -# - -# -# At least one emulation must be selected -# -CONFIG_FPE_NWFPE=y -# CONFIG_FPE_NWFPE_XP is not set -# CONFIG_FPE_FASTFPE is not set - -# -# Userspace binary formats -# -CONFIG_BINFMT_ELF=y -CONFIG_BINFMT_AOUT=y -# CONFIG_BINFMT_MISC is not set -# CONFIG_ARTHUR is not set - -# -# Power management options -# -CONFIG_PM=y -CONFIG_APM=y - -# -# Device Drivers -# - -# -# Generic Driver Options -# -CONFIG_STANDALONE=y -CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set -# CONFIG_DEBUG_DRIVER is not set - -# -# Memory Technology Devices (MTD) -# -CONFIG_MTD=y -# CONFIG_MTD_DEBUG is not set -# CONFIG_MTD_CONCAT is not set -CONFIG_MTD_PARTITIONS=y -CONFIG_MTD_REDBOOT_PARTS=y -CONFIG_MTD_REDBOOT_DIRECTORY_BLOCK=-1 -CONFIG_MTD_REDBOOT_PARTS_UNALLOCATED=y -# CONFIG_MTD_REDBOOT_PARTS_READONLY is not set -CONFIG_MTD_CMDLINE_PARTS=y -# CONFIG_MTD_AFS_PARTS is not set - -# -# User Modules And Translation Layers -# -CONFIG_MTD_CHAR=y -CONFIG_MTD_BLOCK=y -# CONFIG_FTL is not set -# CONFIG_NFTL is not set -# CONFIG_INFTL is not set - -# -# RAM/ROM/Flash chip drivers -# -CONFIG_MTD_CFI=y -CONFIG_MTD_JEDECPROBE=y -CONFIG_MTD_GEN_PROBE=y -# CONFIG_MTD_CFI_ADV_OPTIONS is not set -CONFIG_MTD_MAP_BANK_WIDTH_1=y -CONFIG_MTD_MAP_BANK_WIDTH_2=y -CONFIG_MTD_MAP_BANK_WIDTH_4=y -# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set -CONFIG_MTD_MAP_BANK_WIDTH_16=y -# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set -CONFIG_MTD_CFI_I1=y -CONFIG_MTD_CFI_I2=y -# CONFIG_MTD_CFI_I4 is not set -# CONFIG_MTD_CFI_I8 is not set -CONFIG_MTD_CFI_INTELEXT=y -# CONFIG_MTD_CFI_AMDSTD is not set -# CONFIG_MTD_CFI_STAA is not set -CONFIG_MTD_CFI_UTIL=y -# CONFIG_MTD_RAM is not set -# CONFIG_MTD_ROM is not set -# CONFIG_MTD_ABSENT is not set -# CONFIG_MTD_OBSOLETE_CHIPS is not set -# CONFIG_MTD_XIP is not set - -# -# Mapping drivers for chip access -# -# CONFIG_MTD_COMPLEX_MAPPINGS is not set -# CONFIG_MTD_PHYSMAP is not set -# CONFIG_MTD_ARM_INTEGRATOR is not set -# CONFIG_MTD_EDB7312 is not set -# CONFIG_MTD_IMPA7 is not set -CONFIG_MTD_BAST=y -CONFIG_MTD_BAST_MAXSIZE=4 - -# -# Self-contained MTD device drivers -# -# CONFIG_MTD_SLRAM is not set -# CONFIG_MTD_PHRAM is not set -# CONFIG_MTD_MTDRAM is not set -# CONFIG_MTD_BLKMTD is not set -# CONFIG_MTD_BLOCK2MTD is not set - -# -# Disk-On-Chip Device Drivers -# -# CONFIG_MTD_DOC2000 is not set -# CONFIG_MTD_DOC2001 is not set -# CONFIG_MTD_DOC2001PLUS is not set - -# -# NAND Flash Device Drivers -# -CONFIG_MTD_NAND=y -# CONFIG_MTD_NAND_VERIFY_WRITE is not set -CONFIG_MTD_NAND_IDS=y -CONFIG_MTD_NAND_S3C2410=y -# CONFIG_MTD_NAND_S3C2410_DEBUG is not set -# CONFIG_MTD_NAND_S3C2410_HWECC is not set -# CONFIG_MTD_NAND_DISKONCHIP is not set -# CONFIG_MTD_NAND_NANDSIM is not set - -# -# Parallel port support -# -CONFIG_PARPORT=y -# CONFIG_PARPORT_PC is not set -# CONFIG_PARPORT_ARC is not set -# CONFIG_PARPORT_GSC is not set -CONFIG_PARPORT_1284=y - -# -# Plug and Play support -# - -# -# Block devices -# -# CONFIG_BLK_DEV_FD is not set -# CONFIG_PARIDE is not set -# CONFIG_BLK_DEV_COW_COMMON is not set -CONFIG_BLK_DEV_LOOP=y -# CONFIG_BLK_DEV_CRYPTOLOOP is not set -CONFIG_BLK_DEV_NBD=m -CONFIG_BLK_DEV_RAM=y -CONFIG_BLK_DEV_RAM_COUNT=16 -CONFIG_BLK_DEV_RAM_SIZE=4096 -CONFIG_BLK_DEV_INITRD=y -CONFIG_INITRAMFS_SOURCE="" -# CONFIG_CDROM_PKTCDVD is not set - -# -# IO Schedulers -# -CONFIG_IOSCHED_NOOP=y -CONFIG_IOSCHED_AS=y -CONFIG_IOSCHED_DEADLINE=y -CONFIG_IOSCHED_CFQ=y -# CONFIG_ATA_OVER_ETH is not set - -# -# ATA/ATAPI/MFM/RLL support -# -CONFIG_IDE=y -CONFIG_BLK_DEV_IDE=y - -# -# Please see Documentation/ide.txt for help/info on IDE drives -# -# CONFIG_BLK_DEV_IDE_SATA is not set -CONFIG_BLK_DEV_IDEDISK=y -# CONFIG_IDEDISK_MULTI_MODE is not set -CONFIG_BLK_DEV_IDECD=y -CONFIG_BLK_DEV_IDETAPE=m -CONFIG_BLK_DEV_IDEFLOPPY=m -# CONFIG_IDE_TASK_IOCTL is not set - -# -# IDE chipset support/bugfixes -# -CONFIG_IDE_GENERIC=y -# CONFIG_IDE_ARM is not set -CONFIG_BLK_DEV_IDE_BAST=y -# CONFIG_BLK_DEV_IDEDMA is not set -# CONFIG_IDEDMA_AUTO is not set -# CONFIG_BLK_DEV_HD is not set - -# -# SCSI device support -# -# CONFIG_SCSI is not set - -# -# Multi-device support (RAID and LVM) -# -# CONFIG_MD is not set - -# -# Fusion MPT device support -# - -# -# IEEE 1394 (FireWire) support -# - -# -# I2O device support -# - -# -# Networking support -# -CONFIG_NET=y - -# -# Networking options -# -# CONFIG_PACKET is not set -# CONFIG_NETLINK_DEV is not set -CONFIG_UNIX=y -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_PNP=y -# CONFIG_IP_PNP_DHCP is not set -CONFIG_IP_PNP_BOOTP=y -# CONFIG_IP_PNP_RARP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_ARPD is not set -# CONFIG_SYN_COOKIES is not set -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -CONFIG_IP_TCPDIAG=y -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set - -# -# SCTP Configuration (EXPERIMENTAL) -# -# CONFIG_IP_SCTP is not set -# CONFIG_ATM is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_X25 is not set -# CONFIG_LAPB is not set -# CONFIG_NET_DIVERT is not set -# CONFIG_ECONET is not set -# CONFIG_WAN_ROUTER is not set - -# -# QoS and/or fair queueing -# -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set -CONFIG_NETDEVICES=y -# CONFIG_DUMMY is not set -# CONFIG_BONDING is not set -# CONFIG_EQUALIZER is not set -# CONFIG_TUN is not set - -# -# Ethernet (10 or 100Mbit) -# -CONFIG_NET_ETHERNET=y -# CONFIG_MII is not set -# CONFIG_SMC91X is not set - -# -# Ethernet (1000 Mbit) -# - -# -# Ethernet (10000 Mbit) -# - -# -# Token Ring devices -# - -# -# Wireless LAN (non-hamradio) -# -# CONFIG_NET_RADIO is not set - -# -# Wan interfaces -# -# CONFIG_WAN is not set -# CONFIG_PLIP is not set -# CONFIG_PPP is not set -# CONFIG_SLIP is not set -# CONFIG_SHAPER is not set -# CONFIG_NETCONSOLE is not set - -# -# ISDN subsystem -# -# CONFIG_ISDN is not set - -# -# Input device support -# -CONFIG_INPUT=y - -# -# Userland interfaces -# -CONFIG_INPUT_MOUSEDEV=y -CONFIG_INPUT_MOUSEDEV_PSAUX=y -CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024 -CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768 -# CONFIG_INPUT_JOYDEV is not set -# CONFIG_INPUT_TSDEV is not set -# CONFIG_INPUT_EVDEV is not set -# CONFIG_INPUT_EVBUG is not set - -# -# Input Device Drivers -# -CONFIG_INPUT_KEYBOARD=y -CONFIG_KEYBOARD_ATKBD=y -# CONFIG_KEYBOARD_SUNKBD is not set -# CONFIG_KEYBOARD_LKKBD is not set -# CONFIG_KEYBOARD_XTKBD is not set -# CONFIG_KEYBOARD_NEWTON is not set -CONFIG_INPUT_MOUSE=y -CONFIG_MOUSE_PS2=y -# CONFIG_MOUSE_SERIAL is not set -# CONFIG_MOUSE_VSXXXAA is not set -# CONFIG_INPUT_JOYSTICK is not set -# CONFIG_INPUT_TOUCHSCREEN is not set -# CONFIG_INPUT_MISC is not set - -# -# Hardware I/O ports -# -CONFIG_SERIO=y -CONFIG_SERIO_SERPORT=y -# CONFIG_SERIO_PARKBD is not set -CONFIG_SERIO_LIBPS2=y -# CONFIG_SERIO_RAW is not set -# CONFIG_GAMEPORT is not set -CONFIG_SOUND_GAMEPORT=y - -# -# Character devices -# -CONFIG_VT=y -CONFIG_VT_CONSOLE=y -CONFIG_HW_CONSOLE=y -CONFIG_SERIAL_NONSTANDARD=y -# CONFIG_COMPUTONE is not set -# CONFIG_ROCKETPORT is not set -# CONFIG_CYCLADES is not set -# CONFIG_DIGIEPCA is not set -# CONFIG_MOXA_INTELLIO is not set -# CONFIG_MOXA_SMARTIO is not set -# CONFIG_ISI is not set -# CONFIG_SYNCLINKMP is not set -# CONFIG_N_HDLC is not set -# CONFIG_RISCOM8 is not set -# CONFIG_SPECIALIX is not set -# CONFIG_SX is not set -# CONFIG_RIO is not set -# CONFIG_STALDRV is not set - -# -# Serial drivers -# -CONFIG_SERIAL_8250=y -CONFIG_SERIAL_8250_CONSOLE=y -CONFIG_SERIAL_8250_NR_UARTS=8 -CONFIG_SERIAL_8250_EXTENDED=y -CONFIG_SERIAL_8250_MANY_PORTS=y -CONFIG_SERIAL_8250_SHARE_IRQ=y -# CONFIG_SERIAL_8250_DETECT_IRQ is not set -# CONFIG_SERIAL_8250_MULTIPORT is not set -# CONFIG_SERIAL_8250_RSA is not set - -# -# Non-8250 serial port support -# -CONFIG_SERIAL_S3C2410=y -CONFIG_SERIAL_S3C2410_CONSOLE=y -CONFIG_SERIAL_CORE=y -CONFIG_SERIAL_CORE_CONSOLE=y -CONFIG_UNIX98_PTYS=y -CONFIG_LEGACY_PTYS=y -CONFIG_LEGACY_PTY_COUNT=256 -CONFIG_PRINTER=y -# CONFIG_LP_CONSOLE is not set -CONFIG_PPDEV=y -# CONFIG_TIPAR is not set - -# -# IPMI -# -# CONFIG_IPMI_HANDLER is not set - -# -# Watchdog Cards -# -CONFIG_WATCHDOG=y -# CONFIG_WATCHDOG_NOWAYOUT is not set - -# -# Watchdog Device Drivers -# -# CONFIG_SOFT_WATCHDOG is not set -CONFIG_S3C2410_WATCHDOG=y -# CONFIG_NVRAM is not set -# CONFIG_RTC is not set -CONFIG_S3C2410_RTC=y -# CONFIG_DTLK is not set -# CONFIG_R3964 is not set - -# -# Ftape, the floppy tape device driver -# -# CONFIG_DRM is not set -# CONFIG_RAW_DRIVER is not set - -# -# TPM devices -# -# CONFIG_TCG_TPM is not set - -# -# I2C support -# -CONFIG_I2C=y -CONFIG_I2C_CHARDEV=m - -# -# I2C Algorithms -# -CONFIG_I2C_ALGOBIT=m -# CONFIG_I2C_ALGOPCF is not set -# CONFIG_I2C_ALGOPCA is not set - -# -# I2C Hardware Bus support -# -# CONFIG_I2C_ISA is not set -# CONFIG_I2C_PARPORT is not set -# CONFIG_I2C_PARPORT_LIGHT is not set -CONFIG_I2C_S3C2410=y -# CONFIG_I2C_STUB is not set -# CONFIG_I2C_PCA_ISA is not set - -# -# Hardware Sensors Chip support -# -CONFIG_I2C_SENSOR=m -# CONFIG_SENSORS_ADM1021 is not set -# CONFIG_SENSORS_ADM1025 is not set -# CONFIG_SENSORS_ADM1026 is not set -# CONFIG_SENSORS_ADM1031 is not set -# CONFIG_SENSORS_ASB100 is not set -# CONFIG_SENSORS_DS1621 is not set -# CONFIG_SENSORS_FSCHER is not set -# CONFIG_SENSORS_FSCPOS is not set -# CONFIG_SENSORS_GL518SM is not set -# CONFIG_SENSORS_GL520SM is not set -# CONFIG_SENSORS_IT87 is not set -# CONFIG_SENSORS_LM63 is not set -CONFIG_SENSORS_LM75=m -# CONFIG_SENSORS_LM77 is not set -CONFIG_SENSORS_LM78=m -# CONFIG_SENSORS_LM80 is not set -# CONFIG_SENSORS_LM83 is not set -CONFIG_SENSORS_LM85=m -# CONFIG_SENSORS_LM87 is not set -# CONFIG_SENSORS_LM90 is not set -# CONFIG_SENSORS_MAX1619 is not set -# CONFIG_SENSORS_PC87360 is not set -# CONFIG_SENSORS_SMSC47B397 is not set -# CONFIG_SENSORS_SMSC47M1 is not set -# CONFIG_SENSORS_W83781D is not set -# CONFIG_SENSORS_W83L785TS is not set -# CONFIG_SENSORS_W83627HF is not set - -# -# Other I2C Chip support -# -CONFIG_SENSORS_EEPROM=m -# CONFIG_SENSORS_PCF8574 is not set -# CONFIG_SENSORS_PCF8591 is not set -# CONFIG_SENSORS_RTC8564 is not set -# CONFIG_I2C_DEBUG_CORE is not set -# CONFIG_I2C_DEBUG_ALGO is not set -# CONFIG_I2C_DEBUG_BUS is not set -# CONFIG_I2C_DEBUG_CHIP is not set - -# -# Misc devices -# - -# -# Multimedia devices -# -# CONFIG_VIDEO_DEV is not set - -# -# Digital Video Broadcasting Devices -# -# CONFIG_DVB is not set - -# -# Graphics support -# -CONFIG_FB=y -# CONFIG_FB_CFB_FILLRECT is not set -# CONFIG_FB_CFB_COPYAREA is not set -# CONFIG_FB_CFB_IMAGEBLIT is not set -# CONFIG_FB_SOFT_CURSOR is not set -CONFIG_FB_MODE_HELPERS=y -# CONFIG_FB_TILEBLITTING is not set -# CONFIG_FB_VIRTUAL is not set - -# -# Console display driver support -# -# CONFIG_VGA_CONSOLE is not set -CONFIG_DUMMY_CONSOLE=y -# CONFIG_FRAMEBUFFER_CONSOLE is not set - -# -# Logo configuration -# -# CONFIG_LOGO is not set -# CONFIG_BACKLIGHT_LCD_SUPPORT is not set - -# -# Sound -# -# CONFIG_SOUND is not set - -# -# USB support -# -CONFIG_USB_ARCH_HAS_HCD=y -# CONFIG_USB_ARCH_HAS_OHCI is not set -# CONFIG_USB is not set - -# -# USB Gadget Support -# -# CONFIG_USB_GADGET is not set - -# -# MMC/SD Card support -# -# CONFIG_MMC is not set - -# -# File systems -# -CONFIG_EXT2_FS=y -# CONFIG_EXT2_FS_XATTR is not set -CONFIG_EXT3_FS=y -CONFIG_EXT3_FS_XATTR=y -# CONFIG_EXT3_FS_POSIX_ACL is not set -# CONFIG_EXT3_FS_SECURITY is not set -CONFIG_JBD=y -# CONFIG_JBD_DEBUG is not set -CONFIG_FS_MBCACHE=y -# CONFIG_REISERFS_FS is not set -# CONFIG_JFS_FS is not set - -# -# XFS support -# -# CONFIG_XFS_FS is not set -# CONFIG_MINIX_FS is not set -CONFIG_ROMFS_FS=y -# CONFIG_QUOTA is not set -CONFIG_DNOTIFY=y -# CONFIG_AUTOFS_FS is not set -# CONFIG_AUTOFS4_FS is not set - -# -# CD-ROM/DVD Filesystems -# -# CONFIG_ISO9660_FS is not set -# CONFIG_UDF_FS is not set - -# -# DOS/FAT/NT Filesystems -# -CONFIG_FAT_FS=y -CONFIG_MSDOS_FS=y -CONFIG_VFAT_FS=y -CONFIG_FAT_DEFAULT_CODEPAGE=437 -CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" -# CONFIG_NTFS_FS is not set - -# -# Pseudo filesystems -# -CONFIG_PROC_FS=y -CONFIG_SYSFS=y -# CONFIG_DEVFS_FS is not set -# CONFIG_DEVPTS_FS_XATTR is not set -# CONFIG_TMPFS is not set -# CONFIG_HUGETLB_PAGE is not set -CONFIG_RAMFS=y - -# -# Miscellaneous filesystems -# -# CONFIG_ADFS_FS is not set -# CONFIG_AFFS_FS is not set -# CONFIG_HFS_FS is not set -# CONFIG_HFSPLUS_FS is not set -# CONFIG_BEFS_FS is not set -# CONFIG_BFS_FS is not set -# CONFIG_EFS_FS is not set -CONFIG_JFFS_FS=y -CONFIG_JFFS_FS_VERBOSE=0 -# CONFIG_JFFS_PROC_FS is not set -CONFIG_JFFS2_FS=y -CONFIG_JFFS2_FS_DEBUG=0 -# CONFIG_JFFS2_FS_NAND is not set -# CONFIG_JFFS2_FS_NOR_ECC is not set -# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set -CONFIG_JFFS2_ZLIB=y -CONFIG_JFFS2_RTIME=y -# CONFIG_JFFS2_RUBIN is not set -# CONFIG_CRAMFS is not set -# CONFIG_VXFS_FS is not set -# CONFIG_HPFS_FS is not set -# CONFIG_QNX4FS_FS is not set -# CONFIG_SYSV_FS is not set -# CONFIG_UFS_FS is not set - -# -# Network File Systems -# -CONFIG_NFS_FS=y -# CONFIG_NFS_V3 is not set -# CONFIG_NFS_V4 is not set -# CONFIG_NFS_DIRECTIO is not set -# CONFIG_NFSD is not set -CONFIG_ROOT_NFS=y -CONFIG_LOCKD=y -CONFIG_SUNRPC=y -# CONFIG_RPCSEC_GSS_KRB5 is not set -# CONFIG_RPCSEC_GSS_SPKM3 is not set -# CONFIG_SMB_FS is not set -# CONFIG_CIFS is not set -# CONFIG_NCP_FS is not set -# CONFIG_CODA_FS is not set -# CONFIG_AFS_FS is not set - -# -# Partition Types -# -CONFIG_PARTITION_ADVANCED=y -# CONFIG_ACORN_PARTITION is not set -# CONFIG_OSF_PARTITION is not set -# CONFIG_AMIGA_PARTITION is not set -# CONFIG_ATARI_PARTITION is not set -# CONFIG_MAC_PARTITION is not set -CONFIG_MSDOS_PARTITION=y -CONFIG_BSD_DISKLABEL=y -# CONFIG_MINIX_SUBPARTITION is not set -CONFIG_SOLARIS_X86_PARTITION=y -# CONFIG_UNIXWARE_DISKLABEL is not set -# CONFIG_LDM_PARTITION is not set -# CONFIG_SGI_PARTITION is not set -# CONFIG_ULTRIX_PARTITION is not set -# CONFIG_SUN_PARTITION is not set -# CONFIG_EFI_PARTITION is not set - -# -# Native Language Support -# -CONFIG_NLS=y -CONFIG_NLS_DEFAULT="iso8859-1" -# CONFIG_NLS_CODEPAGE_437 is not set -# CONFIG_NLS_CODEPAGE_737 is not set -# CONFIG_NLS_CODEPAGE_775 is not set -# CONFIG_NLS_CODEPAGE_850 is not set -# CONFIG_NLS_CODEPAGE_852 is not set -# CONFIG_NLS_CODEPAGE_855 is not set -# CONFIG_NLS_CODEPAGE_857 is not set -# CONFIG_NLS_CODEPAGE_860 is not set -# CONFIG_NLS_CODEPAGE_861 is not set -# CONFIG_NLS_CODEPAGE_862 is not set -# CONFIG_NLS_CODEPAGE_863 is not set -# CONFIG_NLS_CODEPAGE_864 is not set -# CONFIG_NLS_CODEPAGE_865 is not set -# CONFIG_NLS_CODEPAGE_866 is not set -# CONFIG_NLS_CODEPAGE_869 is not set -# CONFIG_NLS_CODEPAGE_936 is not set -# CONFIG_NLS_CODEPAGE_950 is not set -# CONFIG_NLS_CODEPAGE_932 is not set -# CONFIG_NLS_CODEPAGE_949 is not set -# CONFIG_NLS_CODEPAGE_874 is not set -# CONFIG_NLS_ISO8859_8 is not set -# CONFIG_NLS_CODEPAGE_1250 is not set -# CONFIG_NLS_CODEPAGE_1251 is not set -# CONFIG_NLS_ASCII is not set -# CONFIG_NLS_ISO8859_1 is not set -# CONFIG_NLS_ISO8859_2 is not set -# CONFIG_NLS_ISO8859_3 is not set -# CONFIG_NLS_ISO8859_4 is not set -# CONFIG_NLS_ISO8859_5 is not set -# CONFIG_NLS_ISO8859_6 is not set -# CONFIG_NLS_ISO8859_7 is not set -# CONFIG_NLS_ISO8859_9 is not set -# CONFIG_NLS_ISO8859_13 is not set -# CONFIG_NLS_ISO8859_14 is not set -# CONFIG_NLS_ISO8859_15 is not set -# CONFIG_NLS_KOI8_R is not set -# CONFIG_NLS_KOI8_U is not set -# CONFIG_NLS_UTF8 is not set - -# -# Profiling support -# -# CONFIG_PROFILING is not set - -# -# Kernel hacking -# -# CONFIG_PRINTK_TIME is not set -CONFIG_DEBUG_KERNEL=y -# CONFIG_MAGIC_SYSRQ is not set -CONFIG_LOG_BUF_SHIFT=16 -# CONFIG_SCHEDSTATS is not set -# CONFIG_DEBUG_SLAB is not set -# CONFIG_DEBUG_SPINLOCK is not set -# CONFIG_DEBUG_SPINLOCK_SLEEP is not set -# CONFIG_DEBUG_KOBJECT is not set -CONFIG_DEBUG_BUGVERBOSE=y -CONFIG_DEBUG_INFO=y -# CONFIG_DEBUG_FS is not set -CONFIG_FRAME_POINTER=y -CONFIG_DEBUG_USER=y -# CONFIG_DEBUG_WAITQ is not set -# CONFIG_DEBUG_ERRORS is not set -CONFIG_DEBUG_LL=y -# CONFIG_DEBUG_ICEDCC is not set -CONFIG_DEBUG_S3C2410_PORT=y -CONFIG_DEBUG_S3C2410_UART=0 - -# -# Security options -# -# CONFIG_KEYS is not set -# CONFIG_SECURITY is not set - -# -# Cryptographic options -# -# CONFIG_CRYPTO is not set - -# -# Hardware crypto devices -# - -# -# Library routines -# -# CONFIG_CRC_CCITT is not set -CONFIG_CRC32=y -# CONFIG_LIBCRC32C is not set -CONFIG_ZLIB_INFLATE=y -CONFIG_ZLIB_DEFLATE=y -- cgit v0.10.2 From da86341856befac4e2f4b16965d6f78379dc960a Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 3 Nov 2006 18:38:02 +0100 Subject: [ARM] 3922/1: S3C24XX: update s3c2410_defconfig to 2.6.19-rc4 Update the s3c2410_defconfig to 2.6.19-rc4 Signed-off-by: Ben Dooks Signed-off-by: Russell King diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig index a832226..c773380 100644 --- a/arch/arm/configs/s3c2410_defconfig +++ b/arch/arm/configs/s3c2410_defconfig @@ -1,9 +1,10 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.18 -# Wed Sep 20 20:27:31 2006 +# Linux kernel version: 2.6.19-rc4 +# Fri Nov 3 17:34:07 2006 # CONFIG_ARM=y +# CONFIG_GENERIC_TIME is not set CONFIG_MMU=y CONFIG_GENERIC_HARDIRQS=y CONFIG_TRACE_IRQFLAGS_SUPPORT=y @@ -29,17 +30,20 @@ CONFIG_LOCALVERSION="" CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y +# CONFIG_IPC_NS is not set # CONFIG_POSIX_MQUEUE is not set # CONFIG_BSD_PROCESS_ACCT is not set # CONFIG_TASKSTATS is not set +# CONFIG_UTS_NS is not set # CONFIG_AUDIT is not set # CONFIG_IKCONFIG is not set # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" CONFIG_CC_OPTIMIZE_FOR_SIZE=y +CONFIG_SYSCTL=y # CONFIG_EMBEDDED is not set CONFIG_UID16=y -CONFIG_SYSCTL=y +# CONFIG_SYSCTL_SYSCALL is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -70,6 +74,7 @@ CONFIG_KMOD=y # # Block layer # +CONFIG_BLOCK=y # CONFIG_BLK_DEV_IO_TRACE is not set # @@ -120,6 +125,7 @@ CONFIG_ARCH_S3C2410=y # # S3C24XX Implementations # +# CONFIG_MACH_AML_M5900 is not set CONFIG_MACH_ANUBIS=y CONFIG_MACH_OSIRIS=y CONFIG_ARCH_BAST=y @@ -178,6 +184,8 @@ CONFIG_CPU_CACHE_V4WT=y CONFIG_CPU_CACHE_VIVT=y CONFIG_CPU_COPY_V4WB=y CONFIG_CPU_TLB_V4WBI=y +CONFIG_CPU_CP15=y +CONFIG_CPU_CP15_MMU=y # # Processor Features @@ -251,6 +259,7 @@ CONFIG_BINFMT_AOUT=y CONFIG_PM=y CONFIG_PM_LEGACY=y # CONFIG_PM_DEBUG is not set +# CONFIG_PM_SYSFS_DEPRECATED is not set CONFIG_APM=y # @@ -266,6 +275,7 @@ CONFIG_NET=y CONFIG_UNIX=y CONFIG_XFRM=y # CONFIG_XFRM_USER is not set +# CONFIG_XFRM_SUB_POLICY is not set # CONFIG_NET_KEY is not set CONFIG_INET=y # CONFIG_IP_MULTICAST is not set @@ -286,10 +296,12 @@ CONFIG_IP_PNP_BOOTP=y # CONFIG_INET_TUNNEL is not set CONFIG_INET_XFRM_MODE_TRANSPORT=y CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_INET_XFRM_MODE_BEET=y CONFIG_INET_DIAG=y CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set @@ -377,6 +389,7 @@ CONFIG_MTD_BLOCK=y # CONFIG_NFTL is not set # CONFIG_INFTL is not set # CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set # # RAM/ROM/Flash chip drivers @@ -512,6 +525,7 @@ CONFIG_BLK_DEV_IDE_BAST=y # # CONFIG_RAID_ATTRS is not set # CONFIG_SCSI is not set +# CONFIG_SCSI_NETLINK is not set # # Multi-device support (RAID and LVM) @@ -606,6 +620,7 @@ CONFIG_DM9000=y # Input device support # CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set # # Userland interfaces @@ -628,6 +643,7 @@ CONFIG_KEYBOARD_ATKBD=y # CONFIG_KEYBOARD_LKKBD is not set # CONFIG_KEYBOARD_XTKBD is not set # CONFIG_KEYBOARD_NEWTON is not set +# CONFIG_KEYBOARD_STOWAWAY is not set CONFIG_INPUT_MOUSE=y CONFIG_MOUSE_PS2=y # CONFIG_MOUSE_SERIAL is not set @@ -734,7 +750,6 @@ CONFIG_S3C2410_WATCHDOG=y # CONFIG_USBPCWATCHDOG is not set CONFIG_HW_RANDOM=y # CONFIG_NVRAM is not set -CONFIG_S3C2410_RTC=y # CONFIG_DTLK is not set # CONFIG_R3964 is not set @@ -747,7 +762,6 @@ CONFIG_S3C2410_RTC=y # TPM devices # # CONFIG_TCG_TPM is not set -# CONFIG_TELCLOCK is not set # # I2C support @@ -798,6 +812,7 @@ CONFIG_SENSORS_EEPROM=m # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -834,6 +849,7 @@ CONFIG_SENSORS_LM85=m # CONFIG_SENSORS_SMSC47M1 is not set # CONFIG_SENSORS_SMSC47M192 is not set # CONFIG_SENSORS_SMSC47B397 is not set +# CONFIG_SENSORS_VT1211 is not set # CONFIG_SENSORS_W83781D is not set # CONFIG_SENSORS_W83791D is not set # CONFIG_SENSORS_W83792D is not set @@ -845,6 +861,7 @@ CONFIG_SENSORS_LM85=m # # Misc devices # +# CONFIG_TIFM_CORE is not set # # LED devices @@ -863,7 +880,6 @@ CONFIG_SENSORS_LM85=m # Multimedia devices # # CONFIG_VIDEO_DEV is not set -CONFIG_VIDEO_V4L2=y # # Digital Video Broadcasting Devices @@ -876,6 +892,7 @@ CONFIG_VIDEO_V4L2=y # CONFIG_FIRMWARE_EDID=y CONFIG_FB=y +# CONFIG_FB_DDC is not set CONFIG_FB_CFB_FILLRECT=y CONFIG_FB_CFB_COPYAREA=y CONFIG_FB_CFB_IMAGEBLIT=y @@ -951,7 +968,6 @@ CONFIG_USB_OHCI_LITTLE_ENDIAN=y # # may also be needed; see USB_STORAGE Help for more information # -# CONFIG_USB_STORAGE is not set # CONFIG_USB_LIBUSUAL is not set # @@ -1007,6 +1023,7 @@ CONFIG_USB_MON=y # # CONFIG_USB_EMI62 is not set # CONFIG_USB_EMI26 is not set +# CONFIG_USB_ADUTUX is not set # CONFIG_USB_AUERSWALD is not set # CONFIG_USB_RIO500 is not set # CONFIG_USB_LEGOTOWER is not set @@ -1014,11 +1031,12 @@ CONFIG_USB_MON=y # CONFIG_USB_LED is not set # CONFIG_USB_CYPRESS_CY7C63 is not set # CONFIG_USB_CYTHERM is not set -# CONFIG_USB_PHIDGETKIT is not set -# CONFIG_USB_PHIDGETSERVO is not set +# CONFIG_USB_PHIDGET is not set # CONFIG_USB_IDMOUSE is not set +# CONFIG_USB_FTDI_ELAN is not set # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1051,6 +1069,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1058,6 +1077,7 @@ CONFIG_FS_MBCACHE=y # CONFIG_JFS_FS is not set # CONFIG_FS_POSIX_ACL is not set # CONFIG_XFS_FS is not set +# CONFIG_GFS2_FS is not set # CONFIG_OCFS2_FS is not set # CONFIG_MINIX_FS is not set CONFIG_ROMFS_FS=y @@ -1089,6 +1109,7 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1" # Pseudo filesystems # CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y CONFIG_SYSFS=y # CONFIG_TMPFS is not set # CONFIG_HUGETLB_PAGE is not set @@ -1219,6 +1240,7 @@ CONFIG_NLS_DEFAULT="iso8859-1" # Kernel hacking # # CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_MUST_CHECK=y CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_KERNEL=y @@ -1238,9 +1260,10 @@ CONFIG_DEBUG_BUGVERBOSE=y CONFIG_DEBUG_INFO=y # CONFIG_DEBUG_FS is not set # CONFIG_DEBUG_VM is not set +# CONFIG_DEBUG_LIST is not set CONFIG_FRAME_POINTER=y -# CONFIG_UNWIND_INFO is not set CONFIG_FORCED_INLINING=y +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set CONFIG_DEBUG_USER=y # CONFIG_DEBUG_WAITQ is not set @@ -1262,10 +1285,6 @@ CONFIG_DEBUG_S3C2410_UART=0 # CONFIG_CRYPTO is not set # -# Hardware crypto devices -# - -# # Library routines # # CONFIG_CRC_CCITT is not set -- cgit v0.10.2 From 3f84ada6c173d856b5e06b239931866ed7dd7a16 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 3 Nov 2006 18:47:10 +0100 Subject: [ARM] 3923/1: S3C24XX: update s3c2410_defconfig with new drivers Add the new drivers, such as SPI, LED and RTC core, to the s3c2410_defconfig. Signed-off-by: Ben Dooks Signed-off-by: Russell King diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig index c773380..0563c14 100644 --- a/arch/arm/configs/s3c2410_defconfig +++ b/arch/arm/configs/s3c2410_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit # Linux kernel version: 2.6.19-rc4 -# Fri Nov 3 17:34:07 2006 +# Fri Nov 3 17:41:31 2006 # CONFIG_ARM=y # CONFIG_GENERIC_TIME is not set @@ -66,7 +66,8 @@ CONFIG_BASE_SMALL=0 # Loadable module support # CONFIG_MODULES=y -# CONFIG_MODULE_UNLOAD is not set +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set # CONFIG_MODVERSIONS is not set # CONFIG_MODULE_SRCVERSION_ALL is not set CONFIG_KMOD=y @@ -431,6 +432,8 @@ CONFIG_MTD_BAST_MAXSIZE=4 # # Self-contained MTD device drivers # +# CONFIG_MTD_DATAFLASH is not set +# CONFIG_MTD_M25P80 is not set # CONFIG_MTD_SLRAM is not set # CONFIG_MTD_PHRAM is not set # CONFIG_MTD_MTDRAM is not set @@ -806,8 +809,21 @@ CONFIG_SENSORS_EEPROM=m # # SPI support # -# CONFIG_SPI is not set -# CONFIG_SPI_MASTER is not set +CONFIG_SPI=y +# CONFIG_SPI_DEBUG is not set +CONFIG_SPI_MASTER=y + +# +# SPI Master Controller Drivers +# +CONFIG_SPI_BITBANG=m +# CONFIG_SPI_BUTTERFLY is not set +CONFIG_SPI_S3C24XX_GPIO=m +CONFIG_SPI_S3C24XX=m + +# +# SPI Protocol Masters +# # # Dallas's 1-wire bus @@ -835,6 +851,7 @@ CONFIG_HWMON_VID=m # CONFIG_SENSORS_GL520SM is not set # CONFIG_SENSORS_IT87 is not set # CONFIG_SENSORS_LM63 is not set +# CONFIG_SENSORS_LM70 is not set CONFIG_SENSORS_LM75=m # CONFIG_SENSORS_LM77 is not set CONFIG_SENSORS_LM78=m @@ -866,15 +883,21 @@ CONFIG_SENSORS_LM85=m # # LED devices # -# CONFIG_NEW_LEDS is not set +CONFIG_NEW_LEDS=y +CONFIG_LEDS_CLASS=m # # LED drivers # +CONFIG_LEDS_S3C24XX=m # # LED Triggers # +CONFIG_LEDS_TRIGGERS=y +CONFIG_LEDS_TRIGGER_TIMER=m +# CONFIG_LEDS_TRIGGER_IDE_DISK is not set +CONFIG_LEDS_TRIGGER_HEARTBEAT=m # # Multimedia devices @@ -1057,7 +1080,37 @@ CONFIG_USB_MON=y # Real Time Clock # CONFIG_RTC_LIB=y -# CONFIG_RTC_CLASS is not set +CONFIG_RTC_CLASS=y +CONFIG_RTC_HCTOSYS=y +CONFIG_RTC_HCTOSYS_DEVICE="rtc0" +# CONFIG_RTC_DEBUG is not set + +# +# RTC interfaces +# +CONFIG_RTC_INTF_SYSFS=y +CONFIG_RTC_INTF_PROC=y +CONFIG_RTC_INTF_DEV=y +# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set + +# +# RTC drivers +# +# CONFIG_RTC_DRV_X1205 is not set +# CONFIG_RTC_DRV_DS1307 is not set +# CONFIG_RTC_DRV_DS1553 is not set +# CONFIG_RTC_DRV_ISL1208 is not set +# CONFIG_RTC_DRV_DS1672 is not set +# CONFIG_RTC_DRV_DS1742 is not set +# CONFIG_RTC_DRV_PCF8563 is not set +# CONFIG_RTC_DRV_PCF8583 is not set +# CONFIG_RTC_DRV_RS5C348 is not set +# CONFIG_RTC_DRV_RS5C372 is not set +CONFIG_RTC_DRV_S3C=y +# CONFIG_RTC_DRV_M48T86 is not set +# CONFIG_RTC_DRV_TEST is not set +# CONFIG_RTC_DRV_MAX6902 is not set +# CONFIG_RTC_DRV_V3020 is not set # # File systems -- cgit v0.10.2 From 90ac3c8124453fb355c10d3e1a27af5c0ab21099 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 9 Apr 2002 12:14:34 -0700 Subject: USB: add another sierra wireless device id As reported by Peter Kucmeroski and Jason Ganovsky. Cc: Peter Kucmeroski Cc: Jason Ganovsky Cc: Kevin Lloyd Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c index ea16572..69cc8fb 100644 --- a/drivers/usb/serial/sierra.c +++ b/drivers/usb/serial/sierra.c @@ -35,6 +35,7 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x1199, 0x0020) }, /* Sierra Wireless MC5725 */ { USB_DEVICE(0x1199, 0x0017) }, /* Sierra Wireless EM5625 */ { USB_DEVICE(0x1199, 0x0019) }, /* Sierra Wireless AirCard 595 */ + { USB_DEVICE(0x1199, 0x0218) }, /* Sierra Wireless MC5720 */ { USB_DEVICE(0x1199, 0x6802) }, /* Sierra Wireless MC8755 */ { USB_DEVICE(0x1199, 0x6803) }, /* Sierra Wireless MC8765 */ { USB_DEVICE(0x1199, 0x6804) }, /* Sierra Wireless MC8755 for Europe */ @@ -58,6 +59,7 @@ static struct usb_device_id id_table_3port [] = { { USB_DEVICE(0x1199, 0x0020) }, /* Sierra Wireless MC5725 */ { USB_DEVICE(0x1199, 0x0017) }, /* Sierra Wireless EM5625 */ { USB_DEVICE(0x1199, 0x0019) }, /* Sierra Wireless AirCard 595 */ + { USB_DEVICE(0x1199, 0x0218) }, /* Sierra Wireless MC5720 */ { USB_DEVICE(0x1199, 0x6802) }, /* Sierra Wireless MC8755 */ { USB_DEVICE(0x1199, 0x6803) }, /* Sierra Wireless MC8765 */ { USB_DEVICE(0x1199, 0x6812) }, /* Sierra Wireless MC8775 */ -- cgit v0.10.2 From bc724b98c5e782c2d6781428ed87768daa34921d Mon Sep 17 00:00:00 2001 From: Phil Dibowitz Date: Thu, 19 Oct 2006 00:11:17 -0700 Subject: USB: usb-storage: Unusual_dev update The protocol in this entry is needed for some versions of the device but not others. This adds the NEED_OVERRIDE flag to prevent it complaining to users who don't need it. Signed-off-by: Phil Dibowitz diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 37ed8e0..1e0d04f 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1236,7 +1236,7 @@ UNUSUAL_DEV( 0x0e21, 0x0520, 0x0100, 0x0100, "Cowon Systems", "iAUDIO M5", US_SC_DEVICE, US_PR_BULK, NULL, - 0 ), + US_FL_NEED_OVERRIDE ), /* Submitted by Antoine Mairesse */ UNUSUAL_DEV( 0x0ed1, 0x6660, 0x0100, 0x0300, -- cgit v0.10.2 From 68717950e11eab8ff754b2721d23e9cb3a47b56f Mon Sep 17 00:00:00 2001 From: Grant Grundler Date: Thu, 19 Oct 2006 15:09:51 -0700 Subject: hid-core: big-endian fix fix Adam Kropelin had posted 32-bit fix in June 2005 about two weeks after I originally had posted my fixes for big endian support. Adam has a UPS device which reports LINEV using 32-bits. Added comments to describe the limitations of the code. extract() is the same version I posted earlier and tested in user space. Made similar changes to implement() routine. I've written (and will shortly post) a test for implement(). Code tested on C3600 (parisc) with USB keyboard/mouse attached. I've dropped test_implement.c and a few other user space test programs on http://iou.parisc-linux.org/~grundler/tests/ -rw-r--r-- 1 grundler grundler 1750 Oct 18 09:13 test_extract.c -rw-r--r-- 1 grundler grundler 561 Jan 25 2006 test_ffs.c -rw-r--r-- 1 grundler users 7175 Apr 8 2005 test_fls.c -rw-r--r-- 1 grundler grundler 206 Sep 1 15:52 test_gettimeofday.c -rw-r--r-- 1 grundler grundler 1886 Oct 19 09:20 test_implement.c -rw-r--r-- 1 grundler users 2707 Jun 4 2005 test_unaligned.c I would appreciate if someone else would look at the output of test_implement.c to make it does The Right Thing. Signed-off-by: Grant Grundler Cc: Matthew Wilcox Cc: Dmitry Torokhov Acked-By: Adam Kropelin Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 45f44fe..6d42036 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -270,7 +270,7 @@ static int hid_add_field(struct hid_parser *parser, unsigned report_type, unsign * Read data value from item. */ -static __inline__ __u32 item_udata(struct hid_item *item) +static u32 item_udata(struct hid_item *item) { switch (item->size) { case 1: return item->data.u8; @@ -280,7 +280,7 @@ static __inline__ __u32 item_udata(struct hid_item *item) return 0; } -static __inline__ __s32 item_sdata(struct hid_item *item) +static s32 item_sdata(struct hid_item *item) { switch (item->size) { case 1: return item->data.s8; @@ -727,7 +727,7 @@ static struct hid_device *hid_parse_report(__u8 *start, unsigned size) * done by hand. */ -static __inline__ __s32 snto32(__u32 value, unsigned n) +static s32 snto32(__u32 value, unsigned n) { switch (n) { case 8: return ((__s8)value); @@ -741,9 +741,9 @@ static __inline__ __s32 snto32(__u32 value, unsigned n) * Convert a signed 32-bit integer to a signed n-bit integer. */ -static __inline__ __u32 s32ton(__s32 value, unsigned n) +static u32 s32ton(__s32 value, unsigned n) { - __s32 a = value >> (n - 1); + s32 a = value >> (n - 1); if (a && a != -1) return value < 0 ? 1 << (n - 1) : (1 << (n - 1)) - 1; return value & ((1 << n) - 1); @@ -751,30 +751,55 @@ static __inline__ __u32 s32ton(__s32 value, unsigned n) /* * Extract/implement a data field from/to a little endian report (bit array). + * + * Code sort-of follows HID spec: + * http://www.usb.org/developers/devclass_docs/HID1_11.pdf + * + * While the USB HID spec allows unlimited length bit fields in "report + * descriptors", most devices never use more than 16 bits. + * One model of UPS is claimed to report "LINEV" as a 32-bit field. + * Search linux-kernel and linux-usb-devel archives for "hid-core extract". */ static __inline__ __u32 extract(__u8 *report, unsigned offset, unsigned n) { - u32 x; + u64 x; + + WARN_ON(n > 32); report += offset >> 3; /* adjust byte index */ - offset &= 8 - 1; - x = get_unaligned((u32 *) report); - x = le32_to_cpu(x); - x = (x >> offset) & ((1 << n) - 1); - return x; + offset &= 7; /* now only need bit offset into one byte */ + x = get_unaligned((u64 *) report); + x = le64_to_cpu(x); + x = (x >> offset) & ((1ULL << n) - 1); /* extract bit field */ + return (u32) x; } +/* + * "implement" : set bits in a little endian bit stream. + * Same concepts as "extract" (see comments above). + * The data mangled in the bit stream remains in little endian + * order the whole time. It make more sense to talk about + * endianness of register values by considering a register + * a "cached" copy of the little endiad bit stream. + */ static __inline__ void implement(__u8 *report, unsigned offset, unsigned n, __u32 value) { - u32 x; + u64 x; + u64 m = (1ULL << n) - 1; + + WARN_ON(n > 32); + + WARN_ON(value > m); + value &= m; report += offset >> 3; - offset &= 8 - 1; - x = get_unaligned((u32 *)report); - x &= cpu_to_le32(~((((__u32) 1 << n) - 1) << offset)); - x |= cpu_to_le32(value << offset); - put_unaligned(x,(u32 *)report); + offset &= 7; + + x = get_unaligned((u64 *)report); + x &= cpu_to_le64(~(m << offset)); + x |= cpu_to_le64(((u64) value) << offset); + put_unaligned(x, (u64 *) report); } /* -- cgit v0.10.2 From 78001e3d75c5d3ae1e8dc9875892b9461e4c8d4b Mon Sep 17 00:00:00 2001 From: Bjorn Schneider Date: Sat, 28 Oct 2006 12:42:04 +0200 Subject: USB: new VID/PID-combos for cp2101 3 new VID/PID combinations (registered with Silicon Laboratories Inc.) added for devices made by Lipowsky Industrie Elektronik GmbH all using the CP2102 usb-to-serial converter (Baby-JTAG, Baby-LIN, HARP-1). Signed-off-by: Bjorn Schneider Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/cp2101.c b/drivers/usb/serial/cp2101.c index bbf6532..f95d42c 100644 --- a/drivers/usb/serial/cp2101.c +++ b/drivers/usb/serial/cp2101.c @@ -64,6 +64,9 @@ static struct usb_device_id id_table [] = { { USB_DEVICE(0x10C4, 0x80F6) }, /* Suunto sports instrument */ { USB_DEVICE(0x10C4, 0x813D) }, /* Burnside Telecom Deskmobile */ { USB_DEVICE(0x10C4, 0x815E) }, /* Helicomm IP-Link 1220-DVM */ + { USB_DEVICE(0x10C4, 0x81C8) }, /* Lipowsky Industrie Elektronik GmbH, Baby-JTAG */ + { USB_DEVICE(0x10C4, 0x81E2) }, /* Lipowsky Industrie Elektronik GmbH, Baby-LIN */ + { USB_DEVICE(0x10C4, 0x8218) }, /* Lipowsky Industrie Elektronik GmbH, HARP-1 */ { USB_DEVICE(0x10C4, 0xEA60) }, /* Silicon Labs factory default */ { USB_DEVICE(0x10C4, 0xEA61) }, /* Silicon Labs factory default */ { USB_DEVICE(0x16D6, 0x0001) }, /* Jablotron serial interface */ -- cgit v0.10.2 From baafe37c6a58d4ddb8c2c62cd0f20340b4c66b35 Mon Sep 17 00:00:00 2001 From: Jan Luebbe Date: Fri, 27 Oct 2006 18:59:24 +0200 Subject: USB: sierra: Fix id for Sierra Wireless MC8755 in new table The new version of sierra.c has introduced tables for the 1 port and 3 port variants. The device id i added in my last patch needs to be added to the 3 port table. Signed-off-by: Jan Luebbe Cc: Kevin Lloyd Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c index 69cc8fb..4b5097f 100644 --- a/drivers/usb/serial/sierra.c +++ b/drivers/usb/serial/sierra.c @@ -62,6 +62,7 @@ static struct usb_device_id id_table_3port [] = { { USB_DEVICE(0x1199, 0x0218) }, /* Sierra Wireless MC5720 */ { USB_DEVICE(0x1199, 0x6802) }, /* Sierra Wireless MC8755 */ { USB_DEVICE(0x1199, 0x6803) }, /* Sierra Wireless MC8765 */ + { USB_DEVICE(0x1199, 0x6804) }, /* Sierra Wireless MC8755 for Europe */ { USB_DEVICE(0x1199, 0x6812) }, /* Sierra Wireless MC8775 */ { USB_DEVICE(0x1199, 0x6820) }, /* Sierra Wireless AirCard 875 */ { } -- cgit v0.10.2 From d8fa59a8f6f7c9a1bc294154fd6805c6b247683d Mon Sep 17 00:00:00 2001 From: Daniel Ritz Date: Fri, 27 Oct 2006 22:46:03 +0200 Subject: usbtouchscreen: use endpoint address from endpoint descriptor use the endpoint address from the endpoint descriptor instead of the hardcoding it to 0x81. at least some ITM based screen use a different address and don't work without this. Signed-off-by: Daniel Ritz Cc: Ralf Lehmann Cc: J.P. Delport Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/usbtouchscreen.c b/drivers/usb/input/usbtouchscreen.c index 2902742..933cedd 100644 --- a/drivers/usb/input/usbtouchscreen.c +++ b/drivers/usb/input/usbtouchscreen.c @@ -640,7 +640,7 @@ static int usbtouch_probe(struct usb_interface *intf, type->max_press, 0, 0); usb_fill_int_urb(usbtouch->irq, usbtouch->udev, - usb_rcvintpipe(usbtouch->udev, 0x81), + usb_rcvintpipe(usbtouch->udev, endpoint->bEndpointAddress), usbtouch->data, type->rept_size, usbtouch_irq, usbtouch, endpoint->bInterval); -- cgit v0.10.2 From 6c8df79f8c0f8d861ea25e6e104a29398d8398f4 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Sat, 28 Oct 2006 11:36:59 +0200 Subject: USB: failure in usblp's error path if urb submission fails due to a transient error here eg. ENOMEM , the driver is dead. This fixes it. Regards Oliver Signed-off-by: Oliver Neukum Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c index 809d465..16353b6 100644 --- a/drivers/usb/class/usblp.c +++ b/drivers/usb/class/usblp.c @@ -722,6 +722,7 @@ static ssize_t usblp_write(struct file *file, const char __user *buffer, size_t usblp->wcomplete = 0; err = usb_submit_urb(usblp->writeurb, GFP_KERNEL); if (err) { + usblp->wcomplete = 1; if (err != -ENOMEM) count = -EIO; else -- cgit v0.10.2 From 5a69ebe1e90d9e8d43131f08d344751cf42254c5 Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Sat, 28 Oct 2006 18:07:25 +0200 Subject: USB: usblp: fix system suspend for some systems this has been confirmed to fix suspend problems with usblp. Signed-off-by: Oliver Neukum Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c index 16353b6..6303970 100644 --- a/drivers/usb/class/usblp.c +++ b/drivers/usb/class/usblp.c @@ -1203,8 +1203,6 @@ static int usblp_suspend (struct usb_interface *intf, pm_message_t message) down (&usblp->sem); /* we take no more IO */ usblp->sleeping = 1; - /* we wait for anything printing */ - wait_event (usblp->wait, usblp->wcomplete || !usblp->present); usblp_unlink_urbs(usblp); up (&usblp->sem); mutex_unlock (&usblp_mutex); -- cgit v0.10.2 From 23b0d968c2c82c2574ca97148ce092eff4ab84a6 Mon Sep 17 00:00:00 2001 From: Naranjo Manuel Francisco Date: Fri, 27 Oct 2006 16:08:54 -0300 Subject: USB: HID: add blacklist AIRcable USB, little beautification This patch add AIRcable USBto USB-HID blacklist, makes some little changes things in the Kconfig to make AIRcable USB look as all the rest of drivers. And it removes the readme part that was on Documentation/usb/usb-serial.txt because it is not needed anymore. Signed-off-by: Naranjo Manuel Francisco diff --git a/Documentation/usb/usb-serial.txt b/Documentation/usb/usb-serial.txt index 8dc2bac..50436e1 100644 --- a/Documentation/usb/usb-serial.txt +++ b/Documentation/usb/usb-serial.txt @@ -428,12 +428,6 @@ Options supported: See http://www.uuhaus.de/linux/palmconnect.html for up-to-date information on this driver. -AIRcable USB Dongle Bluetooth driver - If there is the cdc_acm driver loaded in the system, you will find that the - cdc_acm claims the device before AIRcable can. This is simply corrected - by unloading both modules and then loading the aircable module before - cdc_acm module - Generic Serial driver If your device is not one of the above listed devices, compatible with diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 6d42036..6daf85c 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1640,6 +1640,9 @@ void hid_init_reports(struct hid_device *hid) #define USB_VENDOR_ID_SUN 0x0430 #define USB_DEVICE_ID_RARITAN_KVM_DONGLE 0xcdab +#define USB_VENDOR_ID_AIRCABLE 0x16CA +#define USB_DEVICE_ID_AIRCABLE1 0x1502 + /* * Alphabetically sorted blacklist by quirk type. */ @@ -1657,6 +1660,7 @@ static const struct hid_blacklist { { USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_22, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_23, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_24, HID_QUIRK_IGNORE }, + { USB_VENDOR_ID_AIRCABLE, USB_DEVICE_ID_AIRCABLE1, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_ALCOR, USB_DEVICE_ID_ALCOR_USBRS232, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_BERKSHIRE, USB_DEVICE_ID_BERKSHIRE_PCWD, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW40, HID_QUIRK_IGNORE }, diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig index 9a6ec1b..2a8dd4c 100644 --- a/drivers/usb/serial/Kconfig +++ b/drivers/usb/serial/Kconfig @@ -54,10 +54,10 @@ config USB_SERIAL_GENERIC properly. config USB_SERIAL_AIRCABLE - tristate "AIRcable USB Bluetooth Dongle Driver (EXPERIMENTAL)" + tristate "USB AIRcable Bluetooth Dongle Driver (EXPERIMENTAL)" depends on USB_SERIAL && EXPERIMENTAL help - Say Y here if you want to use AIRcable USB Bluetoot Dongle. + Say Y here if you want to use USB AIRcable Bluetooth Dongle. To compile this driver as a module, choose M here: the module will be called aircable. -- cgit v0.10.2 From 11bd44abbd204f580ea91e75c84e012988971012 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 1 Nov 2006 14:26:26 -0800 Subject: USB: fix compiler issues with newer gcc versions Remove complaint from newer GCCs; they don't like forward function declarations except in top-level contexts. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 66bff18..ba165af 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -1188,6 +1188,7 @@ static inline void show_string(struct usb_device *udev, char *id, char *string) #ifdef CONFIG_USB_OTG #include "otg_whitelist.h" +static int __usb_port_suspend(struct usb_device *, int port1); #endif /** @@ -1289,8 +1290,6 @@ int usb_new_device(struct usb_device *udev) * (Includes HNP test device.) */ if (udev->bus->b_hnp_enable || udev->bus->is_b_host) { - static int __usb_port_suspend(struct usb_device *, - int port1); err = __usb_port_suspend(udev, udev->bus->otg_port); if (err < 0) dev_dbg(&udev->dev, "HNP fail, %d\n", err); -- cgit v0.10.2 From d518b2b48a9c11fc381b179709f5321bce1f3b39 Mon Sep 17 00:00:00 2001 From: Dominic Cerquetti Date: Fri, 20 Oct 2006 14:51:45 -0700 Subject: USB: xpad: additional USB id's added Adding additional USB vendor/product ID's for XBOX pads provided by the XBOX Linux team. Signed-off-by: Dominic Cerquetti Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/xpad.c b/drivers/usb/input/xpad.c index 6a12a94..df97e5c 100644 --- a/drivers/usb/input/xpad.c +++ b/drivers/usb/input/xpad.c @@ -2,6 +2,10 @@ * X-Box gamepad - v0.0.6 * * Copyright (c) 2002 Marko Friedemann + * 2004 Oliver Schwartz , + * Steven Toth , + * Franz Lehner , + * Ivan Hawkes * 2005 Dominic Cerquetti * 2006 Adam Buchbinder * @@ -29,6 +33,7 @@ * - ITO Takayuki for providing essential xpad information on his website * - Vojtech Pavlik - iforce driver / input subsystem * - Greg Kroah-Hartman - usb-skeleton driver + * - XBOX Linux project - extra USB id's * * TODO: * - fine tune axes (especially trigger axes) @@ -54,6 +59,13 @@ * - fixed d-pad to axes mapping * * 2002-07-17 - 0.0.5 : simplified d-pad handling + * + * 2004-10-02 - 0.0.6 : DDR pad support + * - borrowed from the XBOX linux kernel + * - USB id's for commonly used dance pads are present + * - dance pads will map D-PAD to buttons, not axes + * - pass the module paramater 'dpad_to_buttons' to force + * the D-PAD to map to buttons if your pad is not detected */ #include @@ -90,8 +102,35 @@ static const struct xpad_device { { 0x045e, 0x0202, "Microsoft X-Box pad v1 (US)", MAP_DPAD_TO_AXES }, { 0x045e, 0x0289, "Microsoft X-Box pad v2 (US)", MAP_DPAD_TO_AXES }, { 0x045e, 0x0285, "Microsoft X-Box pad (Japan)", MAP_DPAD_TO_AXES }, - { 0x05fd, 0x107a, "InterAct 'PowerPad Pro' X-Box pad (Germany)", MAP_DPAD_TO_AXES }, + { 0x045e, 0x0287, "Microsoft Xbox Controller S", MAP_DPAD_TO_AXES }, { 0x0c12, 0x8809, "RedOctane Xbox Dance Pad", MAP_DPAD_TO_BUTTONS }, + { 0x044f, 0x0f07, "Thrustmaster, Inc. Controller", MAP_DPAD_TO_AXES }, + { 0x046d, 0xca84, "Logitech Xbox Cordless Controller", MAP_DPAD_TO_AXES }, + { 0x046d, 0xca88, "Logitech Compact Controller for Xbox", MAP_DPAD_TO_AXES }, + { 0x05fd, 0x1007, "Mad Catz Controller (unverified)", MAP_DPAD_TO_AXES }, + { 0x05fd, 0x107a, "InterAct 'PowerPad Pro' X-Box pad (Germany)", MAP_DPAD_TO_AXES }, + { 0x0738, 0x4516, "Mad Catz Control Pad", MAP_DPAD_TO_AXES }, + { 0x0738, 0x4522, "Mad Catz LumiCON", MAP_DPAD_TO_AXES }, + { 0x0738, 0x4526, "Mad Catz Control Pad Pro", MAP_DPAD_TO_AXES }, + { 0x0738, 0x4536, "Mad Catz MicroCON", MAP_DPAD_TO_AXES }, + { 0x0738, 0x4540, "Mad Catz Beat Pad", MAP_DPAD_TO_BUTTONS }, + { 0x0738, 0x4556, "Mad Catz Lynx Wireless Controller", MAP_DPAD_TO_AXES }, + { 0x0738, 0x6040, "Mad Catz Beat Pad Pro", MAP_DPAD_TO_BUTTONS }, + { 0x0c12, 0x8802, "Zeroplus Xbox Controller", MAP_DPAD_TO_AXES }, + { 0x0c12, 0x8810, "Zeroplus Xbox Controller", MAP_DPAD_TO_AXES }, + { 0x0c12, 0x9902, "HAMA VibraX - *FAULTY HARDWARE*", MAP_DPAD_TO_AXES }, + { 0x0e4c, 0x1097, "Radica Gamester Controller", MAP_DPAD_TO_AXES }, + { 0x0e4c, 0x2390, "Radica Games Jtech Controller", MAP_DPAD_TO_AXES}, + { 0x0e6f, 0x0003, "Logic3 Freebird wireless Controller", MAP_DPAD_TO_AXES }, + { 0x0e6f, 0x0005, "Eclipse wireless Controller", MAP_DPAD_TO_AXES }, + { 0x0e6f, 0x0006, "Edge wireless Controller", MAP_DPAD_TO_AXES }, + { 0x0e8f, 0x0201, "SmartJoy Frag Xpad/PS2 adaptor", MAP_DPAD_TO_AXES }, + { 0x0f30, 0x0202, "Joytech Advanced Controller", MAP_DPAD_TO_AXES }, + { 0x0f30, 0x8888, "BigBen XBMiniPad Controller", MAP_DPAD_TO_AXES }, + { 0x102c, 0xff0c, "Joytech Wireless Advanced Controller", MAP_DPAD_TO_AXES }, + { 0x12ab, 0x8809, "Xbox DDR dancepad", MAP_DPAD_TO_BUTTONS }, + { 0x1430, 0x8888, "TX6500+ Dance Pad (first generation)", MAP_DPAD_TO_BUTTONS }, + { 0xffff, 0xffff, "Chinese-made Xbox Controller", MAP_DPAD_TO_AXES }, { 0x0000, 0x0000, "Generic X-Box pad", MAP_DPAD_UNKNOWN } }; -- cgit v0.10.2 From 9b823b43ff308c914530ec7fde5e2d79cb37b51a Mon Sep 17 00:00:00 2001 From: Jan Mate Date: Fri, 20 Oct 2006 14:51:44 -0700 Subject: USB Storage: unusual_devs.h entry for Sony Ericsson P990i USB Storage: this patch adds support for Sony Ericsson P990i Signed-off-by: Jan Mate Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 1e0d04f..bc1ac07 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1313,6 +1313,13 @@ UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), +/* Reported by Jan Mate */ +UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, + "Sony Ericsson", + "P990i", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + /* Reported by Kevin Cernekee * Tested on hardware version 1.10. * Entry is needed only for the initializer function override. -- cgit v0.10.2 From 18ee91fa9815fa3bb4e51cdcb8229bd0a0f11a70 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Thu, 2 Nov 2006 12:29:12 -0800 Subject: USB: use MII hooks only if CONFIG_MII is enabled Fix mcs7830 patch The recent mcs7830 update to make the MII support sharable goofed various pre-existing configurations in two ways: - it made the usbnet infrastructure reference MII symbols even when they're not needed in the kernel being built - it didn't enable MII along with the mcs7830 minidriver This patch fixes these two problems. However, there does seem to be a Kconfig reverse dependency bug in that MII gets wrongly enabled in some cases (like USBNET=y and USBNET_MII=n); I think I've noticed that same problem in other situations too. So the result can mean kernels being bloated by stuff that's needlessly enabled ... better than wrongly being disabled, but contributing to bloat. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/net/Kconfig b/drivers/usb/net/Kconfig index 454a186..e081836 100644 --- a/drivers/usb/net/Kconfig +++ b/drivers/usb/net/Kconfig @@ -92,8 +92,13 @@ config USB_RTL8150 To compile this driver as a module, choose M here: the module will be called rtl8150. +config USB_USBNET_MII + tristate + default n + config USB_USBNET tristate "Multi-purpose USB Networking Framework" + select MII if USBNET_MII != n ---help--- This driver supports several kinds of network links over USB, with "minidrivers" built around a common network driver core @@ -129,7 +134,7 @@ config USB_NET_AX8817X tristate "ASIX AX88xxx Based USB 2.0 Ethernet Adapters" depends on USB_USBNET && NET_ETHERNET select CRC32 - select MII + select USB_USBNET_MII default y help This option adds support for ASIX AX88xxx based USB 2.0 @@ -210,6 +215,7 @@ config USB_NET_PLUSB config USB_NET_MCS7830 tristate "MosChip MCS7830 based Ethernet adapters" depends on USB_USBNET + select USB_USBNET_MII help Choose this option if you're using a 10/100 Ethernet USB2 adapter based on the MosChip 7830 controller. This includes diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index 4087363..760b532 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -669,6 +669,9 @@ done: * they'll probably want to use this base set. */ +#if defined(CONFIG_MII) || defined(CONFIG_MII_MODULE) +#define HAVE_MII + int usbnet_get_settings (struct net_device *net, struct ethtool_cmd *cmd) { struct usbnet *dev = netdev_priv(net); @@ -699,20 +702,6 @@ int usbnet_set_settings (struct net_device *net, struct ethtool_cmd *cmd) } EXPORT_SYMBOL_GPL(usbnet_set_settings); - -void usbnet_get_drvinfo (struct net_device *net, struct ethtool_drvinfo *info) -{ - struct usbnet *dev = netdev_priv(net); - - /* REVISIT don't always return "usbnet" */ - strncpy (info->driver, driver_name, sizeof info->driver); - strncpy (info->version, DRIVER_VERSION, sizeof info->version); - strncpy (info->fw_version, dev->driver_info->description, - sizeof info->fw_version); - usb_make_path (dev->udev, info->bus_info, sizeof info->bus_info); -} -EXPORT_SYMBOL_GPL(usbnet_get_drvinfo); - u32 usbnet_get_link (struct net_device *net) { struct usbnet *dev = netdev_priv(net); @@ -730,40 +719,57 @@ u32 usbnet_get_link (struct net_device *net) } EXPORT_SYMBOL_GPL(usbnet_get_link); -u32 usbnet_get_msglevel (struct net_device *net) +int usbnet_nway_reset(struct net_device *net) { struct usbnet *dev = netdev_priv(net); - return dev->msg_enable; + if (!dev->mii.mdio_write) + return -EOPNOTSUPP; + + return mii_nway_restart(&dev->mii); } -EXPORT_SYMBOL_GPL(usbnet_get_msglevel); +EXPORT_SYMBOL_GPL(usbnet_nway_reset); -void usbnet_set_msglevel (struct net_device *net, u32 level) +#endif /* HAVE_MII */ + +void usbnet_get_drvinfo (struct net_device *net, struct ethtool_drvinfo *info) { struct usbnet *dev = netdev_priv(net); - dev->msg_enable = level; + /* REVISIT don't always return "usbnet" */ + strncpy (info->driver, driver_name, sizeof info->driver); + strncpy (info->version, DRIVER_VERSION, sizeof info->version); + strncpy (info->fw_version, dev->driver_info->description, + sizeof info->fw_version); + usb_make_path (dev->udev, info->bus_info, sizeof info->bus_info); } -EXPORT_SYMBOL_GPL(usbnet_set_msglevel); +EXPORT_SYMBOL_GPL(usbnet_get_drvinfo); -int usbnet_nway_reset(struct net_device *net) +u32 usbnet_get_msglevel (struct net_device *net) { struct usbnet *dev = netdev_priv(net); - if (!dev->mii.mdio_write) - return -EOPNOTSUPP; + return dev->msg_enable; +} +EXPORT_SYMBOL_GPL(usbnet_get_msglevel); - return mii_nway_restart(&dev->mii); +void usbnet_set_msglevel (struct net_device *net, u32 level) +{ + struct usbnet *dev = netdev_priv(net); + + dev->msg_enable = level; } -EXPORT_SYMBOL_GPL(usbnet_nway_reset); +EXPORT_SYMBOL_GPL(usbnet_set_msglevel); /* drivers may override default ethtool_ops in their bind() routine */ static struct ethtool_ops usbnet_ethtool_ops = { +#ifdef HAVE_MII .get_settings = usbnet_get_settings, .set_settings = usbnet_set_settings, - .get_drvinfo = usbnet_get_drvinfo, .get_link = usbnet_get_link, .nway_reset = usbnet_nway_reset, +#endif + .get_drvinfo = usbnet_get_drvinfo, .get_msglevel = usbnet_get_msglevel, .set_msglevel = usbnet_set_msglevel, }; -- cgit v0.10.2 From 7bd473fcc217adec000f213e8864bf9a161d57e1 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Thu, 2 Nov 2006 22:06:56 -0800 Subject: [PATCH] eCryptfs: Fix pointer deref I missed a pointer dereference in this kmalloc result check. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index f49f105..136175a 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -134,7 +134,7 @@ int ecryptfs_crypto_api_algify_cipher_name(char **algified_name, algified_name_len = (chaining_modifier_len + cipher_name_len + 3); (*algified_name) = kmalloc(algified_name_len, GFP_KERNEL); - if (!(algified_name)) { + if (!(*algified_name)) { rc = -ENOMEM; goto out; } -- cgit v0.10.2 From df66b8552be5fdab5c4b4d53ee08b99388b9bd02 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:06:56 -0800 Subject: [PATCH] tidy "md: check bio address after mapping through partitions" Neil's xterms are too wide. Cc: Neil Brown Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index c7b1dac..9eaee66 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -3075,11 +3075,12 @@ end_io: if (maxsector) { sector_t sector = bio->bi_sector; - if (maxsector < nr_sectors || maxsector - nr_sectors < sector) { + if (maxsector < nr_sectors || + maxsector - nr_sectors < sector) { /* - * This may well happen - partitions are not checked - * to make sure they are within the size of the - * whole device. + * This may well happen - partitions are not + * checked to make sure they are within the size + * of the whole device. */ handle_bad_sector(bio); goto end_io; -- cgit v0.10.2 From 7870db4c7fa1b03fec133c4f4e67fdaa04c5ac15 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 2 Nov 2006 22:06:57 -0800 Subject: [PATCH] md: send online/offline uevents when an md array starts/stops This allows udev to do something intelligent when an array becomes available. Acked-by: Greg KH Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index 50ab4a9..d111356 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3200,6 +3200,7 @@ static int do_md_run(mddev_t * mddev) mddev->changed = 1; md_new_event(mddev); + kobject_uevent(&mddev->gendisk->kobj, KOBJ_ONLINE); return 0; } @@ -3313,6 +3314,7 @@ static int do_md_stop(mddev_t * mddev, int mode) module_put(mddev->pers->owner); mddev->pers = NULL; + kobject_uevent(&mddev->gendisk->kobj, KOBJ_OFFLINE); if (mddev->ro) mddev->ro = 0; } -- cgit v0.10.2 From 87c2b7c045a44f6c1c7af23e64f2b286e6f7130a Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 2 Nov 2006 22:06:58 -0800 Subject: [PATCH] sys_pselect7 vs compat_sys_pselect7 uaccess error handling 758333458aa719bfc26ec16eafd4ad3a9e96014d fixes the not checked copy_to_user return value of compat_sys_pselect7. I ran into this too because of an old source tree, but my fix would look quite a bit different to Andi's fix. The reason is that the compat function IMHO should behave the very same as the non-compat function if possible. Since sys_pselect7 does not return -EFAULT in this specific case, change the compat code so it behaves like sys_pselect7. Cc: David Woodhouse Cc: Andi Kleen Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/compat.c b/fs/compat.c index 50624d4..8d0a001 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1835,9 +1835,12 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, } while (!ret && !timeout && tsp && (ts.tv_sec || ts.tv_nsec)); - if (ret == 0 && tsp && !(current->personality & STICKY_TIMEOUTS)) { + if (tsp) { struct compat_timespec rts; + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + rts.tv_sec = timeout / HZ; rts.tv_nsec = (timeout % HZ) * (NSEC_PER_SEC/HZ); if (rts.tv_nsec >= NSEC_PER_SEC) { @@ -1846,8 +1849,19 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, } if (compat_timespec_compare(&rts, &ts) >= 0) rts = ts; - if (copy_to_user(tsp, &rts, sizeof(rts))) - ret = -EFAULT; + if (copy_to_user(tsp, &rts, sizeof(rts))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND) + ret = -EINTR; + } } if (ret == -ERESTARTNOHAND) { -- cgit v0.10.2 From c6120938365df9976dc07c536e1c14190ead48e3 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 2 Nov 2006 22:07:01 -0800 Subject: [PATCH] update some docbook comments Correct a few comments in kernel-doc Doc and source files. (akpm: note: the patch removes a non-ascii character and might have to be applied by hand..) Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/kernel-doc-nano-HOWTO.txt b/Documentation/kernel-doc-nano-HOWTO.txt index c65233d..284e7e1 100644 --- a/Documentation/kernel-doc-nano-HOWTO.txt +++ b/Documentation/kernel-doc-nano-HOWTO.txt @@ -17,7 +17,7 @@ are: special place-holders for where the extracted documentation should go. -- scripts/docproc.c +- scripts/basic/docproc.c This is a program for converting SGML template files into SGML files. When a file is referenced it is searched for symbols diff --git a/scripts/basic/docproc.c b/scripts/basic/docproc.c index 4ab6cbf..d6071cb 100644 --- a/scripts/basic/docproc.c +++ b/scripts/basic/docproc.c @@ -250,7 +250,7 @@ void intfunc(char * filename) { docfunctions(filename, NOFUNCTION); } void extfunc(char * filename) { docfunctions(filename, FUNCTION); } /* - * Document spåecific function(s) in a file. + * Document specific function(s) in a file. * Call kernel-doc with the following parameters: * kernel-doc -docbook -function function1 [-function function2] */ -- cgit v0.10.2 From 733b72c31efb0d6b29577655939ccfe835381b52 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 2 Nov 2006 22:07:02 -0800 Subject: [PATCH] docbook: merge journal-api into filesystems.tmpl Move journal-api into filesystems.tmpl as a Chapter. Applies on top of the previous docbook: make a filesystems book patch. Remove trailing whitespace from journal-api chapter. Align some of the tags. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 3bf5086..db9499a 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -9,7 +9,7 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ kernel-hacking.xml kernel-locking.xml deviceiobook.xml \ procfs-guide.xml writing_usb_driver.xml \ - kernel-api.xml filesystems.xml journal-api.xml lsm.xml usb.xml \ + kernel-api.xml filesystems.xml lsm.xml usb.xml \ gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \ genericirq.xml diff --git a/Documentation/DocBook/filesystems.tmpl b/Documentation/DocBook/filesystems.tmpl index 4785032..39fa2ab 100644 --- a/Documentation/DocBook/filesystems.tmpl +++ b/Documentation/DocBook/filesystems.tmpl @@ -98,4 +98,304 @@ + + + The Linux Journalling API + + + + Roger + Gammans + +
+ rgammans@computer-surgery.co.uk +
+ + + + + + + Stephen + Tweedie + +
+ sct@redhat.com +
+ + + + + + 2002 + Roger Gammans + + + + The Linux Journalling API + + + Overview + + Details + +The journalling layer is easy to use. You need to +first of all create a journal_t data structure. There are +two calls to do this dependent on how you decide to allocate the physical +media on which the journal resides. The journal_init_inode() call +is for journals stored in filesystem inodes, or the journal_init_dev() +call can be use for journal stored on a raw device (in a continuous range +of blocks). A journal_t is a typedef for a struct pointer, so when +you are finally finished make sure you call journal_destroy() on it +to free up any used kernel memory. + + + +Once you have got your journal_t object you need to 'mount' or load the journal +file, unless of course you haven't initialised it yet - in which case you +need to call journal_create(). + + + +Most of the time however your journal file will already have been created, but +before you load it you must call journal_wipe() to empty the journal file. +Hang on, you say , what if the filesystem wasn't cleanly umount()'d . Well, it is the +job of the client file system to detect this and skip the call to journal_wipe(). + + + +In either case the next call should be to journal_load() which prepares the +journal file for use. Note that journal_wipe(..,0) calls journal_skip_recovery() +for you if it detects any outstanding transactions in the journal and similarly +journal_load() will call journal_recover() if necessary. +I would advise reading fs/ext3/super.c for examples on this stage. +[RGG: Why is the journal_wipe() call necessary - doesn't this needlessly +complicate the API. Or isn't a good idea for the journal layer to hide +dirty mounts from the client fs] + + + +Now you can go ahead and start modifying the underlying +filesystem. Almost. + + + + +You still need to actually journal your filesystem changes, this +is done by wrapping them into transactions. Additionally you +also need to wrap the modification of each of the buffers +with calls to the journal layer, so it knows what the modifications +you are actually making are. To do this use journal_start() which +returns a transaction handle. + + + +journal_start() +and its counterpart journal_stop(), which indicates the end of a transaction +are nestable calls, so you can reenter a transaction if necessary, +but remember you must call journal_stop() the same number of times as +journal_start() before the transaction is completed (or more accurately +leaves the update phase). Ext3/VFS makes use of this feature to simplify +quota support. + + + +Inside each transaction you need to wrap the modifications to the +individual buffers (blocks). Before you start to modify a buffer you +need to call journal_get_{create,write,undo}_access() as appropriate, +this allows the journalling layer to copy the unmodified data if it +needs to. After all the buffer may be part of a previously uncommitted +transaction. +At this point you are at last ready to modify a buffer, and once +you are have done so you need to call journal_dirty_{meta,}data(). +Or if you've asked for access to a buffer you now know is now longer +required to be pushed back on the device you can call journal_forget() +in much the same way as you might have used bforget() in the past. + + + +A journal_flush() may be called at any time to commit and checkpoint +all your transactions. + + + +Then at umount time , in your put_super() (2.4) or write_super() (2.5) +you can then call journal_destroy() to clean up your in-core journal object. + + + +Unfortunately there a couple of ways the journal layer can cause a deadlock. +The first thing to note is that each task can only have +a single outstanding transaction at any one time, remember nothing +commits until the outermost journal_stop(). This means +you must complete the transaction at the end of each file/inode/address +etc. operation you perform, so that the journalling system isn't re-entered +on another journal. Since transactions can't be nested/batched +across differing journals, and another filesystem other than +yours (say ext3) may be modified in a later syscall. + + + +The second case to bear in mind is that journal_start() can +block if there isn't enough space in the journal for your transaction +(based on the passed nblocks param) - when it blocks it merely(!) needs to +wait for transactions to complete and be committed from other tasks, +so essentially we are waiting for journal_stop(). So to avoid +deadlocks you must treat journal_start/stop() as if they +were semaphores and include them in your semaphore ordering rules to prevent +deadlocks. Note that journal_extend() has similar blocking behaviour to +journal_start() so you can deadlock here just as easily as on journal_start(). + + + +Try to reserve the right number of blocks the first time. ;-). This will +be the maximum number of blocks you are going to touch in this transaction. +I advise having a look at at least ext3_jbd.h to see the basis on which +ext3 uses to make these decisions. + + + +Another wriggle to watch out for is your on-disk block allocation strategy. +why? Because, if you undo a delete, you need to ensure you haven't reused any +of the freed blocks in a later transaction. One simple way of doing this +is make sure any blocks you allocate only have checkpointed transactions +listed against them. Ext3 does this in ext3_test_allocatable(). + + + +Lock is also providing through journal_{un,}lock_updates(), +ext3 uses this when it wants a window with a clean and stable fs for a moment. +eg. + + + + + journal_lock_updates() //stop new stuff happening.. + journal_flush() // checkpoint everything. + ..do stuff on stable fs + journal_unlock_updates() // carry on with filesystem use. + + + +The opportunities for abuse and DOS attacks with this should be obvious, +if you allow unprivileged userspace to trigger codepaths containing these +calls. + + + +A new feature of jbd since 2.5.25 is commit callbacks with the new +journal_callback_set() function you can now ask the journalling layer +to call you back when the transaction is finally committed to disk, so that +you can do some of your own management. The key to this is the journal_callback +struct, this maintains the internal callback information but you can +extend it like this:- + + + struct myfs_callback_s { + //Data structure element required by jbd.. + struct journal_callback for_jbd; + // Stuff for myfs allocated together. + myfs_inode* i_commited; + + } + + + +this would be useful if you needed to know when data was committed to a +particular inode. + + + + + + Summary + +Using the journal is a matter of wrapping the different context changes, +being each mount, each modification (transaction) and each changed buffer +to tell the journalling layer about them. + + + +Here is a some pseudo code to give you an idea of how it works, as +an example. + + + + journal_t* my_jnrl = journal_create(); + journal_init_{dev,inode}(jnrl,...) + if (clean) journal_wipe(); + journal_load(); + + foreach(transaction) { /*transactions must be + completed before + a syscall returns to + userspace*/ + + handle_t * xct=journal_start(my_jnrl); + foreach(bh) { + journal_get_{create,write,undo}_access(xact,bh); + if ( myfs_modify(bh) ) { /* returns true + if makes changes */ + journal_dirty_{meta,}data(xact,bh); + } else { + journal_forget(bh); + } + } + journal_stop(xct); + } + journal_destroy(my_jrnl); + + + + + + + Data Types + + The journalling layer uses typedefs to 'hide' the concrete definitions + of the structures used. As a client of the JBD layer you can + just rely on the using the pointer as a magic cookie of some sort. + + Obviously the hiding is not enforced as this is 'C'. + + Structures +!Iinclude/linux/jbd.h + + + + + Functions + + The functions here are split into two groups those that + affect a journal as a whole, and those which are used to + manage transactions + + Journal Level +!Efs/jbd/journal.c +!Ifs/jbd/recovery.c + + Transasction Level +!Efs/jbd/transaction.c + + + + See also + + + + Journaling the Linux ext2fs Filesystem, LinuxExpo 98, Stephen Tweedie + + + + + + + Ext3 Journalling FileSystem, OLS 2000, Dr. Stephen Tweedie + + + + + + + diff --git a/Documentation/DocBook/journal-api.tmpl b/Documentation/DocBook/journal-api.tmpl deleted file mode 100644 index 2077f9a..0000000 --- a/Documentation/DocBook/journal-api.tmpl +++ /dev/null @@ -1,333 +0,0 @@ - - - - - - The Linux Journalling API - - - Roger - Gammans - -
- rgammans@computer-surgery.co.uk -
- - - - - - - Stephen - Tweedie - -
- sct@redhat.com -
- - - - - - 2002 - Roger Gammans - - - - - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later - version. - - - - This program is distributed in the hope that it will be - useful, but WITHOUT ANY WARRANTY; without even the implied - warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - See the GNU General Public License for more details. - - - - You should have received a copy of the GNU General Public - License along with this program; if not, write to the Free - Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, - MA 02111-1307 USA - - - - For more details see the file COPYING in the source - distribution of Linux. - - - - - - - - Overview - - Details - -The journalling layer is easy to use. You need to -first of all create a journal_t data structure. There are -two calls to do this dependent on how you decide to allocate the physical -media on which the journal resides. The journal_init_inode() call -is for journals stored in filesystem inodes, or the journal_init_dev() -call can be use for journal stored on a raw device (in a continuous range -of blocks). A journal_t is a typedef for a struct pointer, so when -you are finally finished make sure you call journal_destroy() on it -to free up any used kernel memory. - - - -Once you have got your journal_t object you need to 'mount' or load the journal -file, unless of course you haven't initialised it yet - in which case you -need to call journal_create(). - - - -Most of the time however your journal file will already have been created, but -before you load it you must call journal_wipe() to empty the journal file. -Hang on, you say , what if the filesystem wasn't cleanly umount()'d . Well, it is the -job of the client file system to detect this and skip the call to journal_wipe(). - - - -In either case the next call should be to journal_load() which prepares the -journal file for use. Note that journal_wipe(..,0) calls journal_skip_recovery() -for you if it detects any outstanding transactions in the journal and similarly -journal_load() will call journal_recover() if necessary. -I would advise reading fs/ext3/super.c for examples on this stage. -[RGG: Why is the journal_wipe() call necessary - doesn't this needlessly -complicate the API. Or isn't a good idea for the journal layer to hide -dirty mounts from the client fs] - - - -Now you can go ahead and start modifying the underlying -filesystem. Almost. - - - - - -You still need to actually journal your filesystem changes, this -is done by wrapping them into transactions. Additionally you -also need to wrap the modification of each of the buffers -with calls to the journal layer, so it knows what the modifications -you are actually making are. To do this use journal_start() which -returns a transaction handle. - - - -journal_start() -and its counterpart journal_stop(), which indicates the end of a transaction -are nestable calls, so you can reenter a transaction if necessary, -but remember you must call journal_stop() the same number of times as -journal_start() before the transaction is completed (or more accurately -leaves the update phase). Ext3/VFS makes use of this feature to simplify -quota support. - - - -Inside each transaction you need to wrap the modifications to the -individual buffers (blocks). Before you start to modify a buffer you -need to call journal_get_{create,write,undo}_access() as appropriate, -this allows the journalling layer to copy the unmodified data if it -needs to. After all the buffer may be part of a previously uncommitted -transaction. -At this point you are at last ready to modify a buffer, and once -you are have done so you need to call journal_dirty_{meta,}data(). -Or if you've asked for access to a buffer you now know is now longer -required to be pushed back on the device you can call journal_forget() -in much the same way as you might have used bforget() in the past. - - - -A journal_flush() may be called at any time to commit and checkpoint -all your transactions. - - - -Then at umount time , in your put_super() (2.4) or write_super() (2.5) -you can then call journal_destroy() to clean up your in-core journal object. - - - - -Unfortunately there a couple of ways the journal layer can cause a deadlock. -The first thing to note is that each task can only have -a single outstanding transaction at any one time, remember nothing -commits until the outermost journal_stop(). This means -you must complete the transaction at the end of each file/inode/address -etc. operation you perform, so that the journalling system isn't re-entered -on another journal. Since transactions can't be nested/batched -across differing journals, and another filesystem other than -yours (say ext3) may be modified in a later syscall. - - - -The second case to bear in mind is that journal_start() can -block if there isn't enough space in the journal for your transaction -(based on the passed nblocks param) - when it blocks it merely(!) needs to -wait for transactions to complete and be committed from other tasks, -so essentially we are waiting for journal_stop(). So to avoid -deadlocks you must treat journal_start/stop() as if they -were semaphores and include them in your semaphore ordering rules to prevent -deadlocks. Note that journal_extend() has similar blocking behaviour to -journal_start() so you can deadlock here just as easily as on journal_start(). - - - -Try to reserve the right number of blocks the first time. ;-). This will -be the maximum number of blocks you are going to touch in this transaction. -I advise having a look at at least ext3_jbd.h to see the basis on which -ext3 uses to make these decisions. - - - -Another wriggle to watch out for is your on-disk block allocation strategy. -why? Because, if you undo a delete, you need to ensure you haven't reused any -of the freed blocks in a later transaction. One simple way of doing this -is make sure any blocks you allocate only have checkpointed transactions -listed against them. Ext3 does this in ext3_test_allocatable(). - - - -Lock is also providing through journal_{un,}lock_updates(), -ext3 uses this when it wants a window with a clean and stable fs for a moment. -eg. - - - - - journal_lock_updates() //stop new stuff happening.. - journal_flush() // checkpoint everything. - ..do stuff on stable fs - journal_unlock_updates() // carry on with filesystem use. - - - -The opportunities for abuse and DOS attacks with this should be obvious, -if you allow unprivileged userspace to trigger codepaths containing these -calls. - - - -A new feature of jbd since 2.5.25 is commit callbacks with the new -journal_callback_set() function you can now ask the journalling layer -to call you back when the transaction is finally committed to disk, so that -you can do some of your own management. The key to this is the journal_callback -struct, this maintains the internal callback information but you can -extend it like this:- - - - struct myfs_callback_s { - //Data structure element required by jbd.. - struct journal_callback for_jbd; - // Stuff for myfs allocated together. - myfs_inode* i_commited; - - } - - - -this would be useful if you needed to know when data was committed to a -particular inode. - - - - - -Summary - -Using the journal is a matter of wrapping the different context changes, -being each mount, each modification (transaction) and each changed buffer -to tell the journalling layer about them. - - - -Here is a some pseudo code to give you an idea of how it works, as -an example. - - - - journal_t* my_jnrl = journal_create(); - journal_init_{dev,inode}(jnrl,...) - if (clean) journal_wipe(); - journal_load(); - - foreach(transaction) { /*transactions must be - completed before - a syscall returns to - userspace*/ - - handle_t * xct=journal_start(my_jnrl); - foreach(bh) { - journal_get_{create,write,undo}_access(xact,bh); - if ( myfs_modify(bh) ) { /* returns true - if makes changes */ - journal_dirty_{meta,}data(xact,bh); - } else { - journal_forget(bh); - } - } - journal_stop(xct); - } - journal_destroy(my_jrnl); - - - - - - - Data Types - - The journalling layer uses typedefs to 'hide' the concrete definitions - of the structures used. As a client of the JBD layer you can - just rely on the using the pointer as a magic cookie of some sort. - - Obviously the hiding is not enforced as this is 'C'. - - Structures -!Iinclude/linux/jbd.h - - - - - Functions - - The functions here are split into two groups those that - affect a journal as a whole, and those which are used to - manage transactions - - Journal Level -!Efs/jbd/journal.c -!Ifs/jbd/recovery.c - - Transasction Level -!Efs/jbd/transaction.c - - - - See also - - - - Journaling the Linux ext2fs Filesystem,LinuxExpo 98, Stephen Tweedie - - - - - - - Ext3 Journalling FileSystem , OLS 2000, Dr. Stephen Tweedie - - - - - - -- cgit v0.10.2 From c7e12b838989b0e432c7a1cdf1e6c6fd936007f6 Mon Sep 17 00:00:00 2001 From: Pavel Emelianov Date: Thu, 2 Nov 2006 22:07:03 -0800 Subject: [PATCH] Fix ipc entries removal Fix two issuses related to ipc_ids->entries freeing. 1. When freeing ipc namespace we need to free entries allocated with ipc_init_ids(). 2. When removing old entries in grow_ary() ipc_rcu_putref() may be called on entries set to &ids->nullentry earlier in ipc_init_ids(). This is almost impossible without namespaces, but with them this situation becomes possible. Found during OpenVZ testing after obvious leaks in beancounters. Signed-off-by: Pavel Emelianov Cc: Kirill Korotaev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/ipc/msg.c b/ipc/msg.c index 5b213d9..600e06f 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -124,6 +124,7 @@ void msg_exit_ns(struct ipc_namespace *ns) } mutex_unlock(&msg_ids(ns).mutex); + ipc_fini_ids(ns->ids[IPC_MSG_IDS]); kfree(ns->ids[IPC_MSG_IDS]); ns->ids[IPC_MSG_IDS] = NULL; } diff --git a/ipc/sem.c b/ipc/sem.c index 0dafcc4..21b3289 100644 --- a/ipc/sem.c +++ b/ipc/sem.c @@ -161,6 +161,7 @@ void sem_exit_ns(struct ipc_namespace *ns) } mutex_unlock(&sem_ids(ns).mutex); + ipc_fini_ids(ns->ids[IPC_SEM_IDS]); kfree(ns->ids[IPC_SEM_IDS]); ns->ids[IPC_SEM_IDS] = NULL; } diff --git a/ipc/shm.c b/ipc/shm.c index bfbd317..d1198dd 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -116,6 +116,7 @@ void shm_exit_ns(struct ipc_namespace *ns) } mutex_unlock(&shm_ids(ns).mutex); + ipc_fini_ids(ns->ids[IPC_SHM_IDS]); kfree(ns->ids[IPC_SHM_IDS]); ns->ids[IPC_SHM_IDS] = NULL; } diff --git a/ipc/util.c b/ipc/util.c index 42479e4..cd8bb14 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -301,7 +301,7 @@ static int grow_ary(struct ipc_ids* ids, int newsize) */ rcu_assign_pointer(ids->entries, new); - ipc_rcu_putref(old); + __ipc_fini_ids(ids, old); return newsize; } diff --git a/ipc/util.h b/ipc/util.h index c8fd6b9..e3aa2c5 100644 --- a/ipc/util.h +++ b/ipc/util.h @@ -83,6 +83,18 @@ void* ipc_rcu_alloc(int size); void ipc_rcu_getref(void *ptr); void ipc_rcu_putref(void *ptr); +static inline void __ipc_fini_ids(struct ipc_ids *ids, + struct ipc_id_ary *entries) +{ + if (entries != &ids->nullentry) + ipc_rcu_putref(entries); +} + +static inline void ipc_fini_ids(struct ipc_ids *ids) +{ + __ipc_fini_ids(ids, ids->entries); +} + struct kern_ipc_perm* ipc_get(struct ipc_ids* ids, int id); struct kern_ipc_perm* ipc_lock(struct ipc_ids* ids, int id); void ipc_lock_by_ptr(struct kern_ipc_perm *ipcp); -- cgit v0.10.2 From 941c7105dc4f4961727acc518e18e00b9a03cbf3 Mon Sep 17 00:00:00 2001 From: nkalmala Date: Thu, 2 Nov 2006 22:07:04 -0800 Subject: [PATCH] mm: un-needed add-store operation wastes a few bytes Un-needed add-store operation wastes a few bytes. 8 bytes wasted with -O2, on a ppc. Signed-off-by: nkalmala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index b55bb35..bf2f6cf 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -853,7 +853,7 @@ again: pcp = &zone_pcp(zone, cpu)->pcp[cold]; local_irq_save(flags); if (!pcp->count) { - pcp->count += rmqueue_bulk(zone, 0, + pcp->count = rmqueue_bulk(zone, 0, pcp->batch, &pcp->list); if (unlikely(!pcp->count)) goto failed; -- cgit v0.10.2 From 86f4f0f9ba6e35fbbc409dfc3d8615c1a9822482 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 2 Nov 2006 22:07:05 -0800 Subject: [PATCH] fix UFS superblock alignment issues ufs2 fails to mount on x86_64, claiming bad magic. This is because ufs_super_block_third's fs_un1 member is padded out by 4 bytes for 8-byte alignment, pushing down the rest of the struct. Forcing this to be packed solves it. I took a quick look over other on-disk structures and didn't immediately find other problems. I was able to mount & ls a populated ufs2 filesystem w/ this change. Signed-off-by: Eric Sandeen Cc: Evgeniy Dushistov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/ufs_fs.h b/include/linux/ufs_fs.h index 61eef50..28967ed 100644 --- a/include/linux/ufs_fs.h +++ b/include/linux/ufs_fs.h @@ -908,7 +908,7 @@ struct ufs_super_block_third { __fs64 fs_csaddr; /* blk addr of cyl grp summary area */ __fs64 fs_pendingblocks;/* blocks in process of being freed */ __fs32 fs_pendinginodes;/*inodes in process of being freed */ - } fs_u2; + } __attribute__ ((packed)) fs_u2; } fs_un1; union { struct { -- cgit v0.10.2 From 5d861d920a86523bbeb56c19b9906c3fb1b58048 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 2 Nov 2006 22:07:06 -0800 Subject: [PATCH] lkdtm: cleanup headers and module_param/MODULE_PARM_DESC Fix module_param/sysfs file permission typo. Clean up MODULE_PARM_DESC strings to avoid fancy (and incorrect) formatting. Fix header includes for lkdtm; add some needed ones, remove unused ones; and fix this gcc warning: drivers/misc/lkdtm.c:150: warning: 'struct buffer_head' declared inside parameter list drivers/misc/lkdtm.c:150: warning: its scope is only this definition or declaration, which is probably not what you want Signed-off-by: Randy Dunlap Cc: Ankita Garg Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/misc/lkdtm.c b/drivers/misc/lkdtm.c index bbdba7b..46a9c35 100644 --- a/drivers/misc/lkdtm.c +++ b/drivers/misc/lkdtm.c @@ -44,12 +44,14 @@ */ #include +#include #include +#include #include -#include +#include #include -#include #include +#include #include #ifdef CONFIG_IDE @@ -116,16 +118,16 @@ static enum ctype cptype = NONE; static int count = DEFAULT_COUNT; module_param(recur_count, int, 0644); -MODULE_PARM_DESC(recur_count, "Recurcion level for the stack overflow test,\ - default is 10"); +MODULE_PARM_DESC(recur_count, " Recursion level for the stack overflow test, "\ + "default is 10"); module_param(cpoint_name, charp, 0644); -MODULE_PARM_DESC(cpoint_name, "Crash Point, where kernel is to be crashed"); -module_param(cpoint_type, charp, 06444); -MODULE_PARM_DESC(cpoint_type, "Crash Point Type, action to be taken on\ - hitting the crash point"); -module_param(cpoint_count, int, 06444); -MODULE_PARM_DESC(cpoint_count, "Crash Point Count, number of times the \ - crash point is to be hit to trigger action"); +MODULE_PARM_DESC(cpoint_name, " Crash Point, where kernel is to be crashed"); +module_param(cpoint_type, charp, 0644); +MODULE_PARM_DESC(cpoint_type, " Crash Point Type, action to be taken on "\ + "hitting the crash point"); +module_param(cpoint_count, int, 0644); +MODULE_PARM_DESC(cpoint_count, " Crash Point Count, number of times the "\ + "crash point is to be hit to trigger action"); unsigned int jp_do_irq(unsigned int irq) { -- cgit v0.10.2 From 029e332ea717810172e965ec50f942755ad0c58a Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 2 Nov 2006 22:07:06 -0800 Subject: [PATCH] Cleanup read_pages() Current read_pages() assume ->readpages() frees the passed pages. This patch free the pages in ->read_pages(), if those were remaining in the pages_list. So, readpages() just can ignore the remaining pages in pages_list. Signed-off-by: OGAWA Hirofumi Cc: Steven French Cc: Miklos Szeredi Cc: Steven Whitehouse Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/readahead.c b/mm/readahead.c index 1ba736a..23cb61a 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -173,6 +173,8 @@ static int read_pages(struct address_space *mapping, struct file *filp, if (mapping->a_ops->readpages) { ret = mapping->a_ops->readpages(filp, mapping, pages, nr_pages); + /* Clean up the remaining pages */ + put_pages_list(pages); goto out; } -- cgit v0.10.2 From 05ac9d4b3d7eac9e8542c83341a0e22d09aecf8f Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 2 Nov 2006 22:07:08 -0800 Subject: [PATCH] cifs: ->readpages() fixes This just ignore the remaining pages, and will fix a forgot put_pages_list(). Signed-off-by: OGAWA Hirofumi Cc: Steven French Cc: Miklos Szeredi Cc: Steven Whitehouse Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 976a691..7e056b9 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1806,13 +1806,6 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, } if ((rc < 0) || (smb_read_data == NULL)) { cFYI(1, ("Read error in readpages: %d", rc)); - /* clean up remaing pages off list */ - while (!list_empty(page_list) && (i < num_pages)) { - page = list_entry(page_list->prev, struct page, - lru); - list_del(&page->lru); - page_cache_release(page); - } break; } else if (bytes_read > 0) { pSMBr = (struct smb_com_read_rsp *)smb_read_data; @@ -1831,13 +1824,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, this case is ok - if we are at server EOF we will hit it on next read */ - /* while (!list_empty(page_list) && (i < num_pages)) { - page = list_entry(page_list->prev, - struct page, list); - list_del(&page->list); - page_cache_release(page); - } - break; */ + /* break; */ } } else { cFYI(1, ("No bytes read (%d) at offset %lld . " @@ -1845,14 +1832,6 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, bytes_read, offset)); /* BB turn off caching and do new lookup on file size at server? */ - while (!list_empty(page_list) && (i < num_pages)) { - page = list_entry(page_list->prev, struct page, - lru); - list_del(&page->lru); - - /* BB removeme - replace with zero of page? */ - page_cache_release(page); - } break; } if (smb_read_data) { -- cgit v0.10.2 From 2e990021bfc65b1a3778479a9e6b4811f9c1ff0e Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 2 Nov 2006 22:07:09 -0800 Subject: [PATCH] fuse: ->readpages() cleanup This just ignore the remaining pages. Signed-off-by: OGAWA Hirofumi Cc: Steven French Cc: Miklos Szeredi Cc: Steven Whitehouse Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 2bb5ace..763a50d 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -397,14 +397,14 @@ static int fuse_readpages(struct file *file, struct address_space *mapping, err = -EIO; if (is_bad_inode(inode)) - goto clean_pages_up; + goto out; data.file = file; data.inode = inode; data.req = fuse_get_req(fc); err = PTR_ERR(data.req); if (IS_ERR(data.req)) - goto clean_pages_up; + goto out; err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data); if (!err) { @@ -413,10 +413,7 @@ static int fuse_readpages(struct file *file, struct address_space *mapping, else fuse_put_request(fc, data.req); } - return err; - -clean_pages_up: - put_pages_list(pages); +out: return err; } -- cgit v0.10.2 From 7011774db8afca43be466f0f0428434a9edf053e Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 2 Nov 2006 22:07:10 -0800 Subject: [PATCH] gfs2: ->readpages() fixes This just ignore the remaining pages, and remove unneeded unlock_pages(). Signed-off-by: OGAWA Hirofumi Cc: Steven French Cc: Miklos Szeredi Acked-by: Steven Whitehouse Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 8d5963c..015640b 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -337,13 +337,6 @@ out: out_noerror: ret = 0; out_unlock: - /* unlock all pages, we can't do any I/O right now */ - for (page_idx = 0; page_idx < nr_pages; page_idx++) { - struct page *page = list_entry(pages->prev, struct page, lru); - list_del(&page->lru); - unlock_page(page); - page_cache_release(page); - } if (do_unlock) gfs2_holder_uninit(&gh); goto out; -- cgit v0.10.2 From 77d6e1397a004c9376fed855e4164ca2b1dba2ed Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 2 Nov 2006 22:07:10 -0800 Subject: [PATCH] edac_mc: fix error handling Call sysdev_class_unregister() on failure in edac_sysfs_memctrl_setup() and decrease identation level for clear logic. Acked-by: Doug Thompson Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index 4bde30b..75e9e38 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -230,34 +230,43 @@ static struct kobj_type ktype_memctrl = { */ static int edac_sysfs_memctrl_setup(void) { - int err=0; + int err = 0; debugf1("%s()\n", __func__); /* create the /sys/devices/system/edac directory */ err = sysdev_class_register(&edac_class); - if (!err) { - /* Init the MC's kobject */ - memset(&edac_memctrl_kobj, 0, sizeof (edac_memctrl_kobj)); - edac_memctrl_kobj.parent = &edac_class.kset.kobj; - edac_memctrl_kobj.ktype = &ktype_memctrl; + if (err) { + debugf1("%s() error=%d\n", __func__, err); + return err; + } - /* generate sysfs "..../edac/mc" */ - err = kobject_set_name(&edac_memctrl_kobj,"mc"); + /* Init the MC's kobject */ + memset(&edac_memctrl_kobj, 0, sizeof (edac_memctrl_kobj)); + edac_memctrl_kobj.parent = &edac_class.kset.kobj; + edac_memctrl_kobj.ktype = &ktype_memctrl; - if (!err) { - /* FIXME: maybe new sysdev_create_subdir() */ - err = kobject_register(&edac_memctrl_kobj); + /* generate sysfs "..../edac/mc" */ + err = kobject_set_name(&edac_memctrl_kobj,"mc"); - if (err) - debugf1("Failed to register '.../edac/mc'\n"); - else - debugf1("Registered '.../edac/mc' kobject\n"); - } - } else - debugf1("%s() error=%d\n", __func__, err); + if (err) + goto fail; + + /* FIXME: maybe new sysdev_create_subdir() */ + err = kobject_register(&edac_memctrl_kobj); + + if (err) { + debugf1("Failed to register '.../edac/mc'\n"); + goto fail; + } + debugf1("Registered '.../edac/mc' kobject\n"); + + return 0; + +fail: + sysdev_class_unregister(&edac_class); return err; } -- cgit v0.10.2 From 7ef55b8a05c02db7c07d81827c69fe8f124e8654 Mon Sep 17 00:00:00 2001 From: Srinivasa Ds Date: Thu, 2 Nov 2006 22:07:12 -0800 Subject: [PATCH] NFS4: fix for recursive locking problem When I was performing some operations on NFS, I got below error on server side. ============================================= [ INFO: possible recursive locking detected ] 2.6.19-prep #1 --------------------------------------------- nfsd4/3525 is trying to acquire lock: (&inode->i_mutex){--..}, at: [] mutex_lock+0x21/0x24 but task is already holding lock: (&inode->i_mutex){--..}, at: [] mutex_lock+0x21/0x24 other info that might help us debug this: 2 locks held by nfsd4/3525: #0: (client_mutex){--..}, at: [] mutex_lock+0x21/0x24 #1: (&inode->i_mutex){--..}, at: [] mutex_lock+0x21/0x24 stack backtrace: [] show_trace_log_lvl+0x58/0x16a [] show_trace+0xd/0x10 [] dump_stack+0x19/0x1b [] __lock_acquire+0x778/0x99c [] lock_acquire+0x4b/0x6d [] __mutex_lock_slowpath+0xbc/0x20a [] mutex_lock+0x21/0x24 [] vfs_rmdir+0x76/0xf8 [] nfsd4_clear_clid_dir+0x2c/0x41 [nfsd] [] nfsd4_remove_clid_dir+0xb1/0xe8 [nfsd] [] laundromat_main+0x9b/0x1c3 [nfsd] [] run_workqueue+0x7a/0xbb [] worker_thread+0xd2/0x107 [] kthread+0xc3/0xf2 [] kernel_thread_helper+0x5/0xb =================================================================== Cause for this problem was,2 successive mutex_lock calls on 2 diffrent inodes ,as shown below static int nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) { int status; /* For now this directory should already be empty, but we empty it of * any regular files anyway, just in case the directory was created by * a kernel from the future.... */ nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); mutex_lock(&dir->d_inode->i_mutex); status = vfs_rmdir(dir->d_inode, dentry); ... int vfs_rmdir(struct inode *dir, struct dentry *dentry) { int error = may_delete(dir, dentry, 1); if (error) return error; if (!dir->i_op || !dir->i_op->rmdir) return -EPERM; DQUOT_INIT(dir); mutex_lock(&dentry->d_inode->i_mutex); ... So I have developed the patch to overcome this problem. Signed-off-by: Srinivasa DS Cc: Neil Brown Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index e9d0770..81b8565 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -274,7 +274,7 @@ nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) * any regular files anyway, just in case the directory was created by * a kernel from the future.... */ nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); - mutex_lock(&dir->d_inode->i_mutex); + mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); status = vfs_rmdir(dir->d_inode, dentry); mutex_unlock(&dir->d_inode->i_mutex); return status; -- cgit v0.10.2 From d13adb604693374c5fce47cd1a2017bcf3178eae Mon Sep 17 00:00:00 2001 From: Yvan Seth Date: Thu, 2 Nov 2006 22:07:13 -0800 Subject: [PATCH] ipmi_si_intf.c sets bad class_mask with PCI_DEVICE_CLASS Taken from http://bugzilla.kernel.org/show_bug.cgi?id=7439 It looks like device registration in drivers/char/ipmi/ipmi_si_intf.c was cleaned up and a small error was made when setting the class_mask. The fix is simple as the correct mask value is defined in the code but is not used. Acked-by: Corey Minyard Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index e5cfb1f..157fa81 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -1867,7 +1867,7 @@ static int ipmi_pci_resume(struct pci_dev *pdev) static struct pci_device_id ipmi_pci_devices[] = { { PCI_DEVICE(PCI_HP_VENDOR_ID, PCI_MMC_DEVICE_ID) }, - { PCI_DEVICE_CLASS(PCI_ERMC_CLASSCODE, PCI_ERMC_CLASSCODE) } + { PCI_DEVICE_CLASS(PCI_ERMC_CLASSCODE, PCI_ERMC_CLASSCODE_MASK) } }; MODULE_DEVICE_TABLE(pci, ipmi_pci_devices); -- cgit v0.10.2 From 7f6b8876c7e66b0d15af134e2a5b87e55514eb6d Mon Sep 17 00:00:00 2001 From: Daniel Yeisley Date: Thu, 2 Nov 2006 22:07:14 -0800 Subject: [PATCH] init_reap_node() initialization fix It looks like there is a bug in init_reap_node() in slab.c that can cause multiple oops's on certain ES7000 configurations. The variable reap_node is defined per cpu, but only initialized on a single CPU. This causes an oops in next_reap_node() when __get_cpu_var(reap_node) returns the wrong value. Fix is below. Signed-off-by: Dan Yeisley Cc: Andi Kleen Acked-by: Christoph Lameter Cc: Pekka Enberg Cc: Manfred Spraul Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/slab.c b/mm/slab.c index 84c631f..3c4a7e3 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -883,7 +883,7 @@ static void init_reap_node(int cpu) if (node == MAX_NUMNODES) node = first_node(node_online_map); - __get_cpu_var(reap_node) = node; + per_cpu(reap_node, cpu) = node; } static void next_reap_node(void) -- cgit v0.10.2 From f46c483357c2d87606bbefb511321e3efd4baae0 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:07:16 -0800 Subject: [PATCH] Add printk_timed_ratelimit() printk_ratelimit() has global state which makes it not useful for callers which wish to perform ratelimiting at a particular frequency. Add a printk_timed_ratelimit() which utilises caller-provided state storage to permit more flexibility. This function can in fact be used for things other than printk ratelimiting and is perhaps poorly named. Cc: Ulrich Drepper Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 80f39ca..24b6111 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -171,6 +171,8 @@ __attribute_const__ roundup_pow_of_two(unsigned long x) extern int printk_ratelimit(void); extern int __printk_ratelimit(int ratelimit_jiffies, int ratelimit_burst); +extern bool printk_timed_ratelimit(unsigned long *caller_jiffies, + unsigned int interval_msec); static inline void console_silent(void) { diff --git a/kernel/printk.c b/kernel/printk.c index f7d427e..6642655 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -31,6 +31,7 @@ #include #include #include +#include #include @@ -1101,3 +1102,23 @@ int printk_ratelimit(void) printk_ratelimit_burst); } EXPORT_SYMBOL(printk_ratelimit); + +/** + * printk_timed_ratelimit - caller-controlled printk ratelimiting + * @caller_jiffies: pointer to caller's state + * @interval_msecs: minimum interval between prints + * + * printk_timed_ratelimit() returns true if more than @interval_msecs + * milliseconds have elapsed since the last time printk_timed_ratelimit() + * returned true. + */ +bool printk_timed_ratelimit(unsigned long *caller_jiffies, + unsigned int interval_msecs) +{ + if (*caller_jiffies == 0 || time_after(jiffies, *caller_jiffies)) { + *caller_jiffies = jiffies + msecs_to_jiffies(interval_msecs); + return true; + } + return false; +} +EXPORT_SYMBOL(printk_timed_ratelimit); -- cgit v0.10.2 From 19c6b6ed3f597a583f58e3fc99256cc01ae8c394 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:07:17 -0800 Subject: [PATCH] schedule removal of FUTEX_FD Apparently FUTEX_FD is unfixably racy and nothing uses it (or if it does, it shouldn't). Add a warning printk, give any remaining users six months to migrate off it. Cc: Ulrich Drepper Cc: Ingo Molnar Acked-by: Thomas Gleixner Cc: Rusty Russell Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/futex.c b/kernel/futex.c index b364e00..93ef30b 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1507,6 +1507,13 @@ static int futex_fd(u32 __user *uaddr, int signal) struct futex_q *q; struct file *filp; int ret, err; + static unsigned long printk_interval; + + if (printk_timed_ratelimit(&printk_interval, 60 * 60 * 1000)) { + printk(KERN_WARNING "Process `%s' used FUTEX_FD, which " + "will be removed from the kernel in June 2007\n", + current->comm); + } ret = -EINVAL; if (!valid_signal(signal)) -- cgit v0.10.2 From 90d53909443b3986569b38ef145f09ea2359af75 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:07:18 -0800 Subject: [PATCH] acpi_noirq section fix WARNING: vmlinux - Section mismatch: reference to .init.data:acpi_noirq from .text between 'pcibios_penalize_isa_irq' (at offset 0xc026ffa1) and 'pirq_serverworks_get' Acked-by: "Brown, Len" Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index ab974ff..22e4c466 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -70,7 +70,7 @@ static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return #define PREFIX "ACPI: " -int acpi_noirq __initdata; /* skip ACPI IRQ initialization */ +int acpi_noirq; /* skip ACPI IRQ initialization */ int acpi_pci_disabled __initdata; /* skip ACPI PCI scan and IRQ initialization */ int acpi_ht __initdata = 1; /* enable HT */ -- cgit v0.10.2 From b918f6e62cd46774f9fc0a3fbba6bd10ad85ee14 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 2 Nov 2006 22:07:19 -0800 Subject: [PATCH] swsusp: debugging Add a swsusp debugging mode. This does everything that's needed for a suspend except for actually suspending. So we can look in the log messages and work out a) what code is being slow and b) which drivers are misbehaving. (1) # echo testproc > /sys/power/disk # echo disk > /sys/power/state This should turn off the non-boot CPU, freeze all processes, wait for 5 seconds and then thaw the processes and the CPU. (2) # echo test > /sys/power/disk # echo disk > /sys/power/state This should turn off the non-boot CPU, freeze all processes, shrink memory, suspend all devices, wait for 5 seconds, resume the devices etc. Cc: Pavel Machek Cc: Stefan Seyfried Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power index d882f80..dcff4d0 100644 --- a/Documentation/ABI/testing/sysfs-power +++ b/Documentation/ABI/testing/sysfs-power @@ -21,7 +21,7 @@ Description: these states. What: /sys/power/disk -Date: August 2006 +Date: September 2006 Contact: Rafael J. Wysocki Description: The /sys/power/disk file controls the operating mode of the @@ -39,6 +39,19 @@ Description: 'reboot' - the memory image will be saved by the kernel and the system will be rebooted. + Additionally, /sys/power/disk can be used to turn on one of the + two testing modes of the suspend-to-disk mechanism: 'testproc' + or 'test'. If the suspend-to-disk mechanism is in the + 'testproc' mode, writing 'disk' to /sys/power/state will cause + the kernel to disable nonboot CPUs and freeze tasks, wait for 5 + seconds, unfreeze tasks and enable nonboot CPUs. If it is in + the 'test' mode, writing 'disk' to /sys/power/state will cause + the kernel to disable nonboot CPUs and freeze tasks, shrink + memory, suspend devices, wait for 5 seconds, resume devices, + unfreeze tasks and enable nonboot CPUs. Then, we are able to + look in the log messages and work out, for example, which code + is being slow and which device drivers are misbehaving. + The suspend-to-disk method may be chosen by writing to this file one of the accepted strings: @@ -46,6 +59,8 @@ Description: 'platform' 'shutdown' 'reboot' + 'testproc' + 'test' It will only change to 'firmware' or 'platform' if the system supports that. diff --git a/Documentation/power/interface.txt b/Documentation/power/interface.txt index a66bec2..74311d7 100644 --- a/Documentation/power/interface.txt +++ b/Documentation/power/interface.txt @@ -30,6 +30,17 @@ testing). The system will support either 'firmware' or 'platform', and that is known a priori. But, the user may choose 'shutdown' or 'reboot' as alternatives. +Additionally, /sys/power/disk can be used to turn on one of the two testing +modes of the suspend-to-disk mechanism: 'testproc' or 'test'. If the +suspend-to-disk mechanism is in the 'testproc' mode, writing 'disk' to +/sys/power/state will cause the kernel to disable nonboot CPUs and freeze +tasks, wait for 5 seconds, unfreeze tasks and enable nonboot CPUs. If it is +in the 'test' mode, writing 'disk' to /sys/power/state will cause the kernel +to disable nonboot CPUs and freeze tasks, shrink memory, suspend devices, wait +for 5 seconds, resume devices, unfreeze tasks and enable nonboot CPUs. Then, +we are able to look in the log messages and work out, for example, which code +is being slow and which device drivers are misbehaving. + Reading from this file will display what the mode is currently set to. Writing to this file will accept one of @@ -37,6 +48,8 @@ to. Writing to this file will accept one of 'platform' 'shutdown' 'reboot' + 'testproc' + 'test' It will only change to 'firmware' or 'platform' if the system supports it. diff --git a/include/linux/pm.h b/include/linux/pm.h index 6b27e07..070394e 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -116,7 +116,9 @@ typedef int __bitwise suspend_disk_method_t; #define PM_DISK_PLATFORM ((__force suspend_disk_method_t) 2) #define PM_DISK_SHUTDOWN ((__force suspend_disk_method_t) 3) #define PM_DISK_REBOOT ((__force suspend_disk_method_t) 4) -#define PM_DISK_MAX ((__force suspend_disk_method_t) 5) +#define PM_DISK_TEST ((__force suspend_disk_method_t) 5) +#define PM_DISK_TESTPROC ((__force suspend_disk_method_t) 6) +#define PM_DISK_MAX ((__force suspend_disk_method_t) 7) struct pm_ops { suspend_disk_method_t pm_disk_mode; diff --git a/kernel/power/disk.c b/kernel/power/disk.c index d3a158a..b1fb786 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -71,7 +71,7 @@ static inline void platform_finish(void) static int prepare_processes(void) { - int error; + int error = 0; pm_prepare_console(); @@ -84,6 +84,12 @@ static int prepare_processes(void) goto thaw; } + if (pm_disk_mode == PM_DISK_TESTPROC) { + printk("swsusp debug: Waiting for 5 seconds.\n"); + mdelay(5000); + goto thaw; + } + /* Free memory before shutting down devices. */ if (!(error = swsusp_shrink_memory())) return 0; @@ -120,13 +126,21 @@ int pm_suspend_disk(void) if (error) return error; + if (pm_disk_mode == PM_DISK_TESTPROC) + goto Thaw; + suspend_console(); error = device_suspend(PMSG_FREEZE); if (error) { resume_console(); printk("Some devices failed to suspend\n"); - unprepare_processes(); - return error; + goto Thaw; + } + + if (pm_disk_mode == PM_DISK_TEST) { + printk("swsusp debug: Waiting for 5 seconds.\n"); + mdelay(5000); + goto Done; } pr_debug("PM: snapshotting memory.\n"); @@ -143,16 +157,17 @@ int pm_suspend_disk(void) power_down(pm_disk_mode); else { swsusp_free(); - unprepare_processes(); - return error; + goto Thaw; } - } else + } else { pr_debug("PM: Image restored successfully.\n"); + } swsusp_free(); Done: device_resume(); resume_console(); + Thaw: unprepare_processes(); return error; } @@ -249,6 +264,8 @@ static const char * const pm_disk_modes[] = { [PM_DISK_PLATFORM] = "platform", [PM_DISK_SHUTDOWN] = "shutdown", [PM_DISK_REBOOT] = "reboot", + [PM_DISK_TEST] = "test", + [PM_DISK_TESTPROC] = "testproc", }; /** @@ -303,17 +320,19 @@ static ssize_t disk_store(struct subsystem * s, const char * buf, size_t n) } } if (mode) { - if (mode == PM_DISK_SHUTDOWN || mode == PM_DISK_REBOOT) + if (mode == PM_DISK_SHUTDOWN || mode == PM_DISK_REBOOT || + mode == PM_DISK_TEST || mode == PM_DISK_TESTPROC) { pm_disk_mode = mode; - else { + } else { if (pm_ops && pm_ops->enter && (mode == pm_ops->pm_disk_mode)) pm_disk_mode = mode; else error = -EINVAL; } - } else + } else { error = -EINVAL; + } pr_debug("PM: suspend-to-disk mode set to '%s'\n", pm_disk_modes[mode]); -- cgit v0.10.2 From d3e5a938e7ed718f6d191e8b6b176fcfeb88a294 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:07:20 -0800 Subject: [PATCH] spi section fix WARNING: vmlinux - Section mismatch: reference to .init.text:spi_register_board_info from __ksymtab_gpl between '__ksymtab_spi_register_board_info' (at offset 0xc032f7d0) and '__ksymtab_spi_alloc_master' Fix this by removing the export. Acked-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 146298a..c3c0626 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -281,7 +281,6 @@ spi_register_board_info(struct spi_board_info const *info, unsigned n) up(&board_lock); return 0; } -EXPORT_SYMBOL_GPL(spi_register_board_info); /* FIXME someone should add support for a __setup("spi", ...) that * creates board info from kernel command lines -- cgit v0.10.2 From d2c89a4284ea4ecfba77c6f2d7d6f96d52e801e5 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Thu, 2 Nov 2006 22:07:20 -0800 Subject: [PATCH] reiserfs: reset errval after initializing bitmap cache Callers after reiserfs_init_bitmap_cache() expect errval to contain -EINVAL until much later. If a condition fails before errval is reset later, reiserfs_fill_super() will mistakenly return 0, causing an Oops in do_add_mount(). This patch resets errval to -EINVAL after the call. I view this as a temporary fix and real error codes should be used throughout reiserfs_fill_super(). Signed-off-by: Jeff Mahoney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 9041802..1724999 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -1619,6 +1619,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) "jmacd-8: reiserfs_fill_super: unable to read bitmap"); goto error; } + errval = -EINVAL; #ifdef CONFIG_REISERFS_CHECK SWARN(silent, s, "CONFIG_REISERFS_CHECK is set ON"); SWARN(silent, s, "- it is slow mode for debugging."); -- cgit v0.10.2 From 53b173327d283b9bdbfb0c3b6de6f0eb197819d6 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Thu, 2 Nov 2006 22:07:22 -0800 Subject: [PATCH] uml: fix I/O hang Fix a UML hang in which everything would just stop until some I/O happened - a ping, someone whacking the keyboard - at which point everything would start up again as though nothing had happened. The cause was gcc reordering some code which absolutely needed to be executed in the order in the source. When unblock_signals switches signals from off to on, it needs to see if any interrupts had happened in the critical section. The interrupt handlers check signals_enabled - if it is zero, then the handler adds a bit to the "pending" bitmask and returns. unblock_signals checks this mask to see if any signals need to be delivered. The crucial part is this: signals_enabled = 1; save_pending = pending; if(save_pending == 0) return; pending = 0; In order to avoid an interrupt arriving between reading pending and setting it to zero, in which case, the record of the interrupt would be erased, signals are enabled. What happened was that gcc reordered this so that 'save_pending = pending' came before 'signals_enabled = 1', creating a one-instruction window within which an interrupt could arrive, set its bit in pending, and have it be immediately erased. When the I/O workload is purely disk-based, the loss of a block device interrupt stops the entire I/O system because the next block request will wait for the current one to finish. Thus the system hangs until something else causes some I/O to arrive, such as a network packet or console input. The fix to this particular problem is a memory barrier between enabling signals and reading the pending signal mask. An xchg would also probably work. Looking over this code for similar problems led me to do a few more things: - make signals_enabled and pending volatile so that they don't get cached in registers - add an mb() to the return paths of block_signals and unblock_signals so that the modification of signals_enabled doesn't get shuffled into the caller in the event that these are inlined in the future. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/include/sysdep-i386/barrier.h b/arch/um/include/sysdep-i386/barrier.h new file mode 100644 index 0000000..b58d52c --- /dev/null +++ b/arch/um/include/sysdep-i386/barrier.h @@ -0,0 +1,9 @@ +#ifndef __SYSDEP_I386_BARRIER_H +#define __SYSDEP_I386_BARRIER_H + +/* Copied from include/asm-i386 for use by userspace. i386 has the option + * of using mfence, but I'm just using this, which works everywhere, for now. + */ +#define mb() asm volatile("lock; addl $0,0(%esp)") + +#endif diff --git a/arch/um/include/sysdep-x86_64/barrier.h b/arch/um/include/sysdep-x86_64/barrier.h new file mode 100644 index 0000000..7b610be --- /dev/null +++ b/arch/um/include/sysdep-x86_64/barrier.h @@ -0,0 +1,7 @@ +#ifndef __SYSDEP_X86_64_BARRIER_H +#define __SYSDEP_X86_64_BARRIER_H + +/* Copied from include/asm-x86_64 for use by userspace. */ +#define mb() asm volatile("mfence":::"memory") + +#endif diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index 6b81739..b897e85 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -15,6 +15,7 @@ #include "user.h" #include "signal_kern.h" #include "sysdep/sigcontext.h" +#include "sysdep/barrier.h" #include "sigcontext.h" #include "mode.h" #include "os.h" @@ -34,8 +35,12 @@ #define SIGALRM_BIT 2 #define SIGALRM_MASK (1 << SIGALRM_BIT) -static int signals_enabled = 1; -static int pending = 0; +/* These are used by both the signal handlers and + * block/unblock_signals. I don't want modifications cached in a + * register - they must go straight to memory. + */ +static volatile int signals_enabled = 1; +static volatile int pending = 0; void sig_handler(int sig, struct sigcontext *sc) { @@ -152,6 +157,12 @@ int change_sig(int signal, int on) void block_signals(void) { signals_enabled = 0; + /* This must return with signals disabled, so this barrier + * ensures that writes are flushed out before the return. + * This might matter if gcc figures out how to inline this and + * decides to shuffle this code into the caller. + */ + mb(); } void unblock_signals(void) @@ -171,9 +182,23 @@ void unblock_signals(void) */ signals_enabled = 1; + /* Setting signals_enabled and reading pending must + * happen in this order. + */ + mb(); + save_pending = pending; - if(save_pending == 0) + if(save_pending == 0){ + /* This must return with signals enabled, so + * this barrier ensures that writes are + * flushed out before the return. This might + * matter if gcc figures out how to inline + * this (unlikely, given its size) and decides + * to shuffle this code into the caller. + */ + mb(); return; + } pending = 0; -- cgit v0.10.2 From 1f6f61649d8c64d7a3a4d143405df9a7bdd4af10 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Thu, 2 Nov 2006 22:07:23 -0800 Subject: [PATCH] uml: include tidying In order to get the __NR_* constants, we need sys/syscall.h. linux/unistd.h works as well since it includes syscall.h, however syscall.h is more parsimonious. We were inconsistent in this, and this patch adds syscall.h includes where necessary and removes linux/unistd.h includes where they are not needed. asm/unistd.h also includes the __NR_* constants, but these are not the glibc-sanctioned ones, so this also removes one such inclusion. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c index 51f0893..c692a19 100644 --- a/arch/um/os-Linux/process.c +++ b/arch/um/os-Linux/process.c @@ -7,7 +7,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index cb9ab54..9b34fe6 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include #include "user.h" #include "sysdep/ptrace.h" diff --git a/arch/um/os-Linux/tls.c b/arch/um/os-Linux/tls.c index 9f7999f..16215b9 100644 --- a/arch/um/os-Linux/tls.c +++ b/arch/um/os-Linux/tls.c @@ -1,7 +1,7 @@ #include +#include #include #include -#include #include #include "sysdep/tls.h" #include "uml-config.h" -- cgit v0.10.2 From 3fd593979802f81ff6452596ac61e3840f917589 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Thu, 2 Nov 2006 22:07:24 -0800 Subject: [PATCH] Create compat_sys_migrate_pages This is needed on bigendian 64bit architectures. Signed-off-by: Stephen Rothwell Acked-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/compat.h b/include/linux/compat.h index f155319..80b17f4 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -230,5 +230,9 @@ asmlinkage long compat_sys_adjtimex(struct compat_timex __user *utp); extern int compat_printk(const char *fmt, ...); extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat); +asmlinkage long compat_sys_migrate_pages(compat_pid_t pid, + compat_ulong_t maxnode, const compat_ulong_t __user *old_nodes, + const compat_ulong_t __user *new_nodes); + #endif /* CONFIG_COMPAT */ #endif /* _LINUX_COMPAT_H */ diff --git a/kernel/compat.c b/kernel/compat.c index d4898aa..6952dd0 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -982,4 +982,37 @@ asmlinkage long compat_sys_move_pages(pid_t pid, unsigned long nr_pages, } return sys_move_pages(pid, nr_pages, pages, nodes, status, flags); } + +asmlinkage long compat_sys_migrate_pages(compat_pid_t pid, + compat_ulong_t maxnode, + const compat_ulong_t __user *old_nodes, + const compat_ulong_t __user *new_nodes) +{ + unsigned long __user *old = NULL; + unsigned long __user *new = NULL; + nodemask_t tmp_mask; + unsigned long nr_bits; + unsigned long size; + + nr_bits = min_t(unsigned long, maxnode - 1, MAX_NUMNODES); + size = ALIGN(nr_bits, BITS_PER_LONG) / 8; + if (old_nodes) { + if (compat_get_bitmap(nodes_addr(tmp_mask), old_nodes, nr_bits)) + return -EFAULT; + old = compat_alloc_user_space(new_nodes ? size * 2 : size); + if (new_nodes) + new = old + size / sizeof(unsigned long); + if (copy_to_user(old, nodes_addr(tmp_mask), size)) + return -EFAULT; + } + if (new_nodes) { + if (compat_get_bitmap(nodes_addr(tmp_mask), new_nodes, nr_bits)) + return -EFAULT; + if (new == NULL) + new = compat_alloc_user_space(size); + if (copy_to_user(new, nodes_addr(tmp_mask), size)) + return -EFAULT; + } + return sys_migrate_pages(pid, nr_bits + 1, old, new); +} #endif diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 0e53314..d7306d0 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -135,6 +135,7 @@ cond_syscall(sys_madvise); cond_syscall(sys_mremap); cond_syscall(sys_remap_file_pages); cond_syscall(compat_sys_move_pages); +cond_syscall(compat_sys_migrate_pages); /* block-layer dependent */ cond_syscall(sys_bdflush); -- cgit v0.10.2 From 43530d2b04b63ac4bb4ac25deee5f1180ccedc2e Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Thu, 2 Nov 2006 22:07:24 -0800 Subject: [PATCH] powerpc: wire up sys_migrate_pages Signed-off-by: Stephen Rothwell Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-powerpc/systbl.h b/include/asm-powerpc/systbl.h index eac85ce..c6a0318 100644 --- a/include/asm-powerpc/systbl.h +++ b/include/asm-powerpc/systbl.h @@ -261,7 +261,7 @@ SYSX(sys_ni_syscall, ppc_fadvise64_64, ppc_fadvise64_64) PPC_SYS_SPU(rtas) OLDSYS(debug_setcontext) SYSCALL(ni_syscall) -SYSCALL(ni_syscall) +COMPAT_SYS(migrate_pages) COMPAT_SYS(mbind) COMPAT_SYS(get_mempolicy) COMPAT_SYS(set_mempolicy) diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h index 464a48c..b5fe932 100644 --- a/include/asm-powerpc/unistd.h +++ b/include/asm-powerpc/unistd.h @@ -276,7 +276,7 @@ #define __NR_rtas 255 #define __NR_sys_debug_setcontext 256 /* Number 257 is reserved for vserver */ -/* 258 currently unused */ +#define __NR_migrate_pages 258 #define __NR_mbind 259 #define __NR_get_mempolicy 260 #define __NR_set_mempolicy 261 -- cgit v0.10.2 From 1f604c4bc078213aa1c4576efa0e8dad98522fa7 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Thu, 2 Nov 2006 22:07:25 -0800 Subject: [PATCH] drivers/isdn/hysdn/hysdn_sched.c: sleep after taking spinlock fix spin_lock_irq{save,restore} is incorrectly called here (the function can sleep after acquring the lock). done the necessary corrections and removed unwanted cli/sti. Signed-off-by: Amol Lad Signed-off-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/hysdn/hysdn_sched.c b/drivers/isdn/hysdn/hysdn_sched.c index 1fadf01..1875877 100644 --- a/drivers/isdn/hysdn/hysdn_sched.c +++ b/drivers/isdn/hysdn/hysdn_sched.c @@ -155,21 +155,17 @@ hysdn_tx_cfgline(hysdn_card *card, unsigned char *line, unsigned short chan) if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg chan=%d len=%d", chan, strlen(line) + 1); - spin_lock_irqsave(&card->hysdn_lock, flags); while (card->async_busy) { - sti(); if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg delayed"); msleep_interruptible(20); /* Timeout 20ms */ - if (!--cnt) { - spin_unlock_irqrestore(&card->hysdn_lock, flags); + if (!--cnt) return (-ERR_ASYNC_TIME); /* timed out */ - } - cli(); } /* wait for buffer to become free */ + spin_lock_irqsave(&card->hysdn_lock, flags); strcpy(card->async_data, line); card->async_len = strlen(line) + 1; card->async_channel = chan; @@ -177,30 +173,23 @@ hysdn_tx_cfgline(hysdn_card *card, unsigned char *line, unsigned short chan) /* now queue the task */ schedule_work(&card->irq_queue); - sti(); + spin_unlock_irqrestore(&card->hysdn_lock, flags); if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg data queued"); cnt++; /* short delay */ - cli(); while (card->async_busy) { - sti(); if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg waiting for tx-ready"); msleep_interruptible(20); /* Timeout 20ms */ - if (!--cnt) { - spin_unlock_irqrestore(&card->hysdn_lock, flags); + if (!--cnt) return (-ERR_ASYNC_TIME); /* timed out */ - } - cli(); } /* wait for buffer to become free again */ - spin_unlock_irqrestore(&card->hysdn_lock, flags); - if (card->debug_flags & LOG_SCHED_ASYN) hysdn_addlog(card, "async tx-cfg data send"); -- cgit v0.10.2 From 8804023061b3447fbaddbd286d78170ad88d1a43 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 2 Nov 2006 22:07:26 -0800 Subject: [PATCH] fix Documentation/accounting/getdelays.c buf size getdelays reports a "fatal reply error, errno 258". We don't have enough room for multi-threaded exit (PID + TGID). Signed-off-by: Oleg Nesterov Cc: Balbir Singh Cc: Shailabh Nagar Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/accounting/getdelays.c b/Documentation/accounting/getdelays.c index b11792a..bf2b0e2 100644 --- a/Documentation/accounting/getdelays.c +++ b/Documentation/accounting/getdelays.c @@ -49,7 +49,7 @@ __u64 stime, utime; } /* Maximum size of response requested or message sent */ -#define MAX_MSG_SIZE 256 +#define MAX_MSG_SIZE 1024 /* Maximum number of cpus expected to be specified in a cpumask */ #define MAX_CPUS 32 /* Maximum length of pathname to log file */ -- cgit v0.10.2 From cda5e61a8e0b11826780b8e5a4155683f0557c8b Mon Sep 17 00:00:00 2001 From: Peer Chen Date: Thu, 2 Nov 2006 22:07:27 -0800 Subject: [PATCH] IDE: Add the support of nvidia PATA controllers of MCP67 to amd74xx.c Add support for PATA controllers of MCP67 to amd74xx.c. Signed-off-by: Peer Chen Cc: Jeff Garzik Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c index 2b0ea8b..753fe0e 100644 --- a/drivers/ide/pci/amd74xx.c +++ b/drivers/ide/pci/amd74xx.c @@ -75,6 +75,7 @@ static struct amd_ide_chip { { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE, 0x50, AMD_UDMA_133 }, + { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_AMD_CS5536_IDE, 0x40, AMD_UDMA_100 }, { 0 } }; @@ -491,7 +492,8 @@ static ide_pci_device_t amd74xx_chipsets[] __devinitdata = { /* 16 */ DECLARE_NV_DEV("NFORCE-MCP55"), /* 17 */ DECLARE_NV_DEV("NFORCE-MCP61"), /* 18 */ DECLARE_NV_DEV("NFORCE-MCP65"), - /* 19 */ DECLARE_AMD_DEV("AMD5536"), + /* 19 */ DECLARE_NV_DEV("NFORCE-MCP67"), + /* 20 */ DECLARE_AMD_DEV("AMD5536"), }; static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_id *id) @@ -530,7 +532,8 @@ static struct pci_device_id amd74xx_pci_tbl[] = { { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 16 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 17 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 18 }, - { PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 19 }, + { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 19 }, + { PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 20 }, { 0, }, }; MODULE_DEVICE_TABLE(pci, amd74xx_pci_tbl); -- cgit v0.10.2 From 8ce08464d2c749610a52c4d6c7c11080a7eaaef1 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Thu, 2 Nov 2006 22:07:28 -0800 Subject: [PATCH] Fix sys_move_pages when a NULL node list is passed sys_move_pages() uses vmalloc() to allocate an array of structures that is fills with information passed from user mode and then passes to do_stat_pages() (in the case the node list is NULL). do_stat_pages() depends on a marker in the node field of the structure to decide how large the array is and this marker is correctly inserted into the last element of the array. However, vmalloc() doesn't zero the memory it allocates and if the user passes NULL for the node list, then the node fields are not filled in (except for the end marker). If the memory the vmalloc() returned happend to have a word with the marker value in it in just the right place, do_pages_stat will fail to fill the status field of part of the array and we will return (random) kernel data to user mode. Signed-off-by: Stephen Rothwell Cc: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/migrate.c b/mm/migrate.c index ba2453f..b4979d4 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -952,7 +952,8 @@ asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages, goto out; pm[i].node = node; - } + } else + pm[i].node = 0; /* anything to not match MAX_NUMNODES */ } /* End marker */ pm[nr_pages].node = MAX_NUMNODES; -- cgit v0.10.2 From f1f2d8713d16a1e198880bbc716eb24fae09c858 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Thu, 2 Nov 2006 22:07:29 -0800 Subject: [PATCH] Fix user.* xattr permission check for sticky dirs The user.* extended attributes are only allowed on regular files and directories. Sticky directories further restrict write access to the owner and privileged users. (See the attr(5) man page for an explanation.) The original check in ext2/ext3 when user.* xattrs were merged was more restrictive than intended, and when the xattr permission checks were moved into the VFS, read access to user.* attributes on sticky directores ended up being denied in addition. Originally-from: Gerard Neil Signed-off-by: Andreas Gruenbacher Cc: Dave Kleikamp Cc: Jan Engelhardt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/xattr.c b/fs/xattr.c index 3956351..0901bdc 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -48,14 +48,21 @@ xattr_permission(struct inode *inode, const char *name, int mask) return 0; /* - * The trusted.* namespace can only accessed by a privilegued user. + * The trusted.* namespace can only be accessed by a privileged user. */ if (!strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN)) return (capable(CAP_SYS_ADMIN) ? 0 : -EPERM); + /* In user.* namespace, only regular files and directories can have + * extended attributes. For sticky directories, only the owner and + * privileged user can write attributes. + */ if (!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN)) { - if (!S_ISREG(inode->i_mode) && - (!S_ISDIR(inode->i_mode) || inode->i_mode & S_ISVTX)) + if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode)) + return -EPERM; + if (S_ISDIR(inode->i_mode) && (inode->i_mode & S_ISVTX) && + (mask & MAY_WRITE) && (current->fsuid != inode->i_uid) && + !capable(CAP_FOWNER)) return -EPERM; } -- cgit v0.10.2 From ddac0d39cf437d02fde9795ae57d9c4b4c146de9 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sat, 4 Nov 2006 12:49:32 +0100 Subject: [PATCH] splice: fix problem introduced with inode diet After the inode slimming patch that unionised i_pipe/i_bdev/i_cdev, it's no longer enough to check for existance of ->i_pipe to verify that this is a pipe. Original patch from Eric Dumazet Final solution suggested by Linus. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/fs/splice.c b/fs/splice.c index 8d70595..da74583 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1109,6 +1109,19 @@ out_release: EXPORT_SYMBOL(do_splice_direct); /* + * After the inode slimming patch, i_pipe/i_bdev/i_cdev share the same + * location, so checking ->i_pipe is not enough to verify that this is a + * pipe. + */ +static inline struct pipe_inode_info *pipe_info(struct inode *inode) +{ + if (S_ISFIFO(inode->i_mode)) + return inode->i_pipe; + + return NULL; +} + +/* * Determine where to splice to/from. */ static long do_splice(struct file *in, loff_t __user *off_in, @@ -1119,7 +1132,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, loff_t offset, *off; long ret; - pipe = in->f_dentry->d_inode->i_pipe; + pipe = pipe_info(in->f_dentry->d_inode); if (pipe) { if (off_in) return -ESPIPE; @@ -1140,7 +1153,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, return ret; } - pipe = out->f_dentry->d_inode->i_pipe; + pipe = pipe_info(out->f_dentry->d_inode); if (pipe) { if (off_out) return -ESPIPE; @@ -1298,7 +1311,7 @@ static int get_iovec_page_array(const struct iovec __user *iov, static long do_vmsplice(struct file *file, const struct iovec __user *iov, unsigned long nr_segs, unsigned int flags) { - struct pipe_inode_info *pipe = file->f_dentry->d_inode->i_pipe; + struct pipe_inode_info *pipe; struct page *pages[PIPE_BUFFERS]; struct partial_page partial[PIPE_BUFFERS]; struct splice_pipe_desc spd = { @@ -1308,7 +1321,8 @@ static long do_vmsplice(struct file *file, const struct iovec __user *iov, .ops = &user_page_pipe_buf_ops, }; - if (unlikely(!pipe)) + pipe = pipe_info(file->f_dentry->d_inode); + if (!pipe) return -EBADF; if (unlikely(nr_segs > UIO_MAXIOV)) return -EINVAL; @@ -1535,8 +1549,8 @@ static int link_pipe(struct pipe_inode_info *ipipe, static long do_tee(struct file *in, struct file *out, size_t len, unsigned int flags) { - struct pipe_inode_info *ipipe = in->f_dentry->d_inode->i_pipe; - struct pipe_inode_info *opipe = out->f_dentry->d_inode->i_pipe; + struct pipe_inode_info *ipipe = pipe_info(in->f_dentry->d_inode); + struct pipe_inode_info *opipe = pipe_info(out->f_dentry->d_inode); int ret = -EINVAL; /* -- cgit v0.10.2 From 80491eb90c750fcd7d13830062f27ae9b7cc5f75 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 4 Nov 2006 09:55:00 -0800 Subject: Revert unintentional "volatile" changes in ipc/msg.c Commit 5a06a363ef48444186f18095ae1b932dddbbfa89 ("[PATCH] ipc/msg.c: clean up coding style") breaks fakeroot on Alpha (variously hangs or oopses), according to a report by Falk Hueffner. The fact that the code seems to rely on compiler access ordering through the use of "volatile" is a pretty certain sign that the code has locking problems, and we should fix those properly and then remove the whole "volatile" entirely. But in the meantime, the movement of "volatile" was unintentional, and should be reverted. Cc: Falk Hueffner Cc: Andrew Morton Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds diff --git a/ipc/msg.c b/ipc/msg.c index 600e06f..1266b1d 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -52,7 +52,7 @@ struct msg_receiver { long r_msgtype; long r_maxsize; - volatile struct msg_msg *r_msg; + struct msg_msg *volatile r_msg; }; /* one msg_sender for each sleeping sender */ -- cgit v0.10.2 From 45c18b0bb579b5c1b89f8c99f1b6ffa4c586ba08 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 4 Nov 2006 10:06:02 -0800 Subject: Fix unlikely (but possible) race condition on task->user access There's a possible race condition when doing a "switch_uid()" from one user to another, which could race with another thread doing a signal allocation and looking at the old thread ->user pointer as it is freed. This explains an oops reported by Lukasz Trabinski: http://permalink.gmane.org/gmane.linux.kernel/462241 We fix this by delaying the (reference-counted) freeing of the user structure until the thread signal handler lock has been released, so that we know that the signal allocation has either seen the new value or has properly incremented the reference count of the old one. Race identified by Oleg Nesterov. Cc: Lukasz Trabinski Cc: Oleg Nesterov Cc: Andrew Morton Cc: Ingo Molnar Signed-off-by: Linus Torvalds diff --git a/kernel/user.c b/kernel/user.c index 6408c04..220e586 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -187,6 +187,17 @@ void switch_uid(struct user_struct *new_user) atomic_dec(&old_user->processes); switch_uid_keyring(new_user); current->user = new_user; + + /* + * We need to synchronize with __sigqueue_alloc() + * doing a get_uid(p->user).. If that saw the old + * user value, we need to wait until it has exited + * its critical region before we can free the old + * structure. + */ + smp_mb(); + spin_unlock_wait(¤t->sighand->siglock); + free_uid(old_user); suid_keys(current); } -- cgit v0.10.2 From 10b1fbdb0a0ca91847a534ad26d0bc250c25b74f Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 4 Nov 2006 13:03:00 -0800 Subject: Make sure "user->sigpending" count is in sync The previous commit (45c18b0bb579b5c1b89f8c99f1b6ffa4c586ba08, aka "Fix unlikely (but possible) race condition on task->user access") fixed a potential oops due to __sigqueue_alloc() getting its "user" pointer out of sync with switch_user(), and accessing a user pointer that had been de-allocated on another CPU. It still left another (much less serious) problem, where a concurrent __sigqueue_alloc and swich_user could cause sigqueue_alloc to do signal pending reference counting for a _different_ user than the one it then actually ended up using. No oops, but we'd end up with the wrong signal accounting. Another case of Oleg's eagle-eyes picking up the problem. This is trivially fixed by just making sure we load whichever "user" structure we decide to use (it doesn't matter _which_ one we pick, we just need to pick one) just once. Acked-by: Oleg Nesterov Cc: Andrew Morton Cc: Ingo Molnar Signed-off-by: Linus Torvalds diff --git a/kernel/signal.c b/kernel/signal.c index 7ed8d53..df18c16 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -267,18 +267,25 @@ static struct sigqueue *__sigqueue_alloc(struct task_struct *t, gfp_t flags, int override_rlimit) { struct sigqueue *q = NULL; + struct user_struct *user; - atomic_inc(&t->user->sigpending); + /* + * In order to avoid problems with "switch_user()", we want to make + * sure that the compiler doesn't re-load "t->user" + */ + user = t->user; + barrier(); + atomic_inc(&user->sigpending); if (override_rlimit || - atomic_read(&t->user->sigpending) <= + atomic_read(&user->sigpending) <= t->signal->rlim[RLIMIT_SIGPENDING].rlim_cur) q = kmem_cache_alloc(sigqueue_cachep, flags); if (unlikely(q == NULL)) { - atomic_dec(&t->user->sigpending); + atomic_dec(&user->sigpending); } else { INIT_LIST_HEAD(&q->list); q->flags = 0; - q->user = get_uid(t->user); + q->user = get_uid(user); } return(q); } -- cgit v0.10.2 From 4833ed094097323f5f219820f6ebdc8dd66f501f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:27:06 -0800 Subject: [IPX]: Trivial parts of endianness annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/include/linux/ipx.h b/include/linux/ipx.h index 4f29c60..eb19b4e 100644 --- a/include/linux/ipx.h +++ b/include/linux/ipx.h @@ -7,8 +7,8 @@ struct sockaddr_ipx { sa_family_t sipx_family; - __u16 sipx_port; - __u32 sipx_network; + __be16 sipx_port; + __be32 sipx_network; unsigned char sipx_node[IPX_NODE_LEN]; __u8 sipx_type; unsigned char sipx_zero; /* 16 byte fill */ @@ -23,13 +23,13 @@ struct sockaddr_ipx { #define IPX_CRTITF 1 struct ipx_route_definition { - __u32 ipx_network; - __u32 ipx_router_network; + __be32 ipx_network; + __be32 ipx_router_network; unsigned char ipx_router_node[IPX_NODE_LEN]; }; struct ipx_interface_definition { - __u32 ipx_network; + __be32 ipx_network; unsigned char ipx_device[16]; unsigned char ipx_dlink_type; #define IPX_FRAME_NONE 0 @@ -55,8 +55,8 @@ struct ipx_config_data { */ struct ipx_route_def { - __u32 ipx_network; - __u32 ipx_router_network; + __be32 ipx_network; + __be32 ipx_router_network; #define IPX_ROUTE_NO_ROUTER 0 unsigned char ipx_router_node[IPX_NODE_LEN]; unsigned char ipx_device[16]; diff --git a/include/net/ipx.h b/include/net/ipx.h index 5c0cf33..4a423d2 100644 --- a/include/net/ipx.h +++ b/include/net/ipx.h @@ -15,9 +15,9 @@ #include struct ipx_address { - __u32 net; + __be32 net; __u8 node[IPX_NODE_LEN]; - __u16 sock; + __be16 sock; }; #define ipx_broadcast_node "\377\377\377\377\377\377" @@ -28,7 +28,7 @@ struct ipx_address { struct ipxhdr { __u16 ipx_checksum __attribute__ ((packed)); #define IPX_NO_CHECKSUM 0xFFFF - __u16 ipx_pktsize __attribute__ ((packed)); + __be16 ipx_pktsize __attribute__ ((packed)); __u8 ipx_tctrl; __u8 ipx_type; #define IPX_TYPE_UNKNOWN 0x00 @@ -48,14 +48,14 @@ static __inline__ struct ipxhdr *ipx_hdr(struct sk_buff *skb) struct ipx_interface { /* IPX address */ - __u32 if_netnum; + __be32 if_netnum; unsigned char if_node[IPX_NODE_LEN]; atomic_t refcnt; /* physical device info */ struct net_device *if_dev; struct datalink_proto *if_dlink; - unsigned short if_dlink_type; + __be16 if_dlink_type; /* socket support */ unsigned short if_sknum; @@ -71,7 +71,7 @@ struct ipx_interface { }; struct ipx_route { - __u32 ir_net; + __be32 ir_net; struct ipx_interface *ir_intrfc; unsigned char ir_routed; unsigned char ir_router_node[IPX_NODE_LEN]; @@ -82,10 +82,10 @@ struct ipx_route { #ifdef __KERNEL__ struct ipx_cb { u8 ipx_tctrl; - u32 ipx_dest_net; - u32 ipx_source_net; + __be32 ipx_dest_net; + __be32 ipx_source_net; struct { - u32 netnum; + __be32 netnum; int index; } last_hop; }; @@ -97,7 +97,7 @@ struct ipx_sock { struct sock sk; struct ipx_address dest_addr; struct ipx_interface *intrfc; - unsigned short port; + __be16 port; #ifdef CONFIG_IPX_INTERN unsigned char node[IPX_NODE_LEN]; #endif @@ -132,7 +132,7 @@ extern struct ipx_interface *ipx_primary_net; extern int ipx_proc_init(void); extern void ipx_proc_exit(void); -extern const char *ipx_frame_name(unsigned short); +extern const char *ipx_frame_name(__be16); extern const char *ipx_device_name(struct ipx_interface *intrfc); static __inline__ void ipxitf_hold(struct ipx_interface *intrfc) diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index bef3f61..c272a38a 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -83,13 +83,13 @@ DEFINE_SPINLOCK(ipx_interfaces_lock); struct ipx_interface *ipx_primary_net; struct ipx_interface *ipx_internal_net; -extern int ipxrtr_add_route(__u32 network, struct ipx_interface *intrfc, +extern int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, unsigned char *node); extern void ipxrtr_del_routes(struct ipx_interface *intrfc); extern int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, struct iovec *iov, int len, int noblock); extern int ipxrtr_route_skb(struct sk_buff *skb); -extern struct ipx_route *ipxrtr_lookup(__u32 net); +extern struct ipx_route *ipxrtr_lookup(__be32 net); extern int ipxrtr_ioctl(unsigned int cmd, void __user *arg); #undef IPX_REFCNT_DEBUG @@ -177,7 +177,7 @@ static void ipxitf_clear_primary_net(void) } static struct ipx_interface *__ipxitf_find_using_phys(struct net_device *dev, - unsigned short datalink) + __be16 datalink) { struct ipx_interface *i; @@ -190,7 +190,7 @@ out: } static struct ipx_interface *ipxitf_find_using_phys(struct net_device *dev, - unsigned short datalink) + __be16 datalink) { struct ipx_interface *i; @@ -202,7 +202,7 @@ static struct ipx_interface *ipxitf_find_using_phys(struct net_device *dev, return i; } -struct ipx_interface *ipxitf_find_using_net(__u32 net) +struct ipx_interface *ipxitf_find_using_net(__be32 net) { struct ipx_interface *i; @@ -237,7 +237,7 @@ static void ipxitf_insert_socket(struct ipx_interface *intrfc, struct sock *sk) /* caller must hold intrfc->if_sklist_lock */ static struct sock *__ipxitf_find_socket(struct ipx_interface *intrfc, - unsigned short port) + __be16 port) { struct sock *s; struct hlist_node *node; @@ -252,7 +252,7 @@ found: /* caller must hold a reference to intrfc */ static struct sock *ipxitf_find_socket(struct ipx_interface *intrfc, - unsigned short port) + __be16 port) { struct sock *s; @@ -268,7 +268,7 @@ static struct sock *ipxitf_find_socket(struct ipx_interface *intrfc, #ifdef CONFIG_IPX_INTERN static struct sock *ipxitf_find_internal_socket(struct ipx_interface *intrfc, unsigned char *ipx_node, - unsigned short port) + __be16 port) { struct sock *s; struct hlist_node *node; @@ -600,10 +600,10 @@ int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node) /* see if we need to include the netnum in the route list */ if (IPX_SKB_CB(skb)->last_hop.index >= 0) { - u32 *last_hop = (u32 *)(((u8 *) skb->data) + + __be32 *last_hop = (__be32 *)(((u8 *) skb->data) + sizeof(struct ipxhdr) + IPX_SKB_CB(skb)->last_hop.index * - sizeof(u32)); + sizeof(__be32)); *last_hop = IPX_SKB_CB(skb)->last_hop.netnum; IPX_SKB_CB(skb)->last_hop.index = -1; } @@ -772,7 +772,7 @@ static void ipxitf_discover_netnum(struct ipx_interface *intrfc, } else { printk(KERN_WARNING "IPX: Network number collision " "%lx\n %s %s and %s %s\n", - (unsigned long) htonl(cb->ipx_source_net), + (unsigned long) ntohl(cb->ipx_source_net), ipx_device_name(i), ipx_frame_name(i->if_dlink_type), ipx_device_name(intrfc), @@ -812,7 +812,7 @@ static int ipxitf_pprop(struct ipx_interface *intrfc, struct sk_buff *skb) int i, rc = -EINVAL; struct ipx_interface *ifcs; char *c; - u32 *l; + __be32 *l; /* Illegal packet - too many hops or too short */ /* We decide to throw it away: no broadcasting, no local processing. @@ -833,7 +833,7 @@ static int ipxitf_pprop(struct ipx_interface *intrfc, struct sk_buff *skb) goto out; c = ((u8 *) ipx) + sizeof(struct ipxhdr); - l = (u32 *) c; + l = (__be32 *) c; /* Don't broadcast packet if already seen this net */ for (i = 0; i < IPX_SKB_CB(skb)->ipx_tctrl; i++) @@ -855,7 +855,7 @@ static int ipxitf_pprop(struct ipx_interface *intrfc, struct sk_buff *skb) /* That aren't in the list */ if (ifcs == intrfc) continue; - l = (__u32 *) c; + l = (__be32 *) c; /* don't consider the last entry in the packet list, * it is our netnum, and it is not there yet */ for (i = 0; i < IPX_SKB_CB(skb)->ipx_tctrl; i++) @@ -885,8 +885,8 @@ static void ipxitf_insert(struct ipx_interface *intrfc) ipx_primary_net = intrfc; } -static struct ipx_interface *ipxitf_alloc(struct net_device *dev, __u32 netnum, - unsigned short dlink_type, +static struct ipx_interface *ipxitf_alloc(struct net_device *dev, __be32 netnum, + __be16 dlink_type, struct datalink_proto *dlink, unsigned char internal, int ipx_offset) @@ -960,7 +960,7 @@ static __be16 ipx_map_frame_type(unsigned char type) static int ipxitf_create(struct ipx_interface_definition *idef) { struct net_device *dev; - unsigned short dlink_type = 0; + __be16 dlink_type = 0; struct datalink_proto *datalink = NULL; struct ipx_interface *intrfc; int rc; @@ -1073,7 +1073,7 @@ out: static int ipxitf_delete(struct ipx_interface_definition *idef) { struct net_device *dev = NULL; - unsigned short dlink_type = 0; + __be16 dlink_type = 0; struct ipx_interface *intrfc; int rc = 0; @@ -1110,7 +1110,7 @@ out: } static struct ipx_interface *ipxitf_auto_create(struct net_device *dev, - unsigned short dlink_type) + __be16 dlink_type) { struct ipx_interface *intrfc = NULL; struct datalink_proto *datalink; @@ -1122,7 +1122,7 @@ static struct ipx_interface *ipxitf_auto_create(struct net_device *dev, if (dev->addr_len > IPX_NODE_LEN) goto out; - switch (htons(dlink_type)) { + switch (ntohs(dlink_type)) { case ETH_P_IPX: datalink = pEII_datalink; break; case ETH_P_802_2: datalink = p8022_datalink; break; case ETH_P_SNAP: datalink = pSNAP_datalink; break; @@ -1266,7 +1266,7 @@ __u16 ipx_cksum(struct ipxhdr *packet, int length) return ~sum; } -const char *ipx_frame_name(unsigned short frame) +const char *ipx_frame_name(__be16 frame) { char* rc = "None"; @@ -1401,7 +1401,7 @@ out: /* caller must hold a reference to intrfc */ -static unsigned short ipx_first_free_socketnum(struct ipx_interface *intrfc) +static __be16 ipx_first_free_socketnum(struct ipx_interface *intrfc) { unsigned short socketNum = intrfc->if_sknum; @@ -1410,7 +1410,7 @@ static unsigned short ipx_first_free_socketnum(struct ipx_interface *intrfc) if (socketNum < IPX_MIN_EPHEMERAL_SOCKET) socketNum = IPX_MIN_EPHEMERAL_SOCKET; - while (__ipxitf_find_socket(intrfc, ntohs(socketNum))) + while (__ipxitf_find_socket(intrfc, htons(socketNum))) if (socketNum > IPX_MAX_EPHEMERAL_SOCKET) socketNum = IPX_MIN_EPHEMERAL_SOCKET; else @@ -1419,7 +1419,7 @@ static unsigned short ipx_first_free_socketnum(struct ipx_interface *intrfc) spin_unlock_bh(&intrfc->if_sklist_lock); intrfc->if_sknum = socketNum; - return ntohs(socketNum); + return htons(socketNum); } static int ipx_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) @@ -1473,7 +1473,7 @@ static int ipx_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) ipxs->port)) { SOCK_DEBUG(sk, "IPX: bind failed because port %X in use.\n", - ntohs((int)addr->sipx_port)); + ntohs(addr->sipx_port)); goto out_put; } } else { @@ -1488,7 +1488,7 @@ static int ipx_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (ipxitf_find_socket(intrfc, addr->sipx_port)) { SOCK_DEBUG(sk, "IPX: bind failed because port %X in use.\n", - ntohs((int)addr->sipx_port)); + ntohs(addr->sipx_port)); goto out_put; } } @@ -1665,7 +1665,7 @@ static int ipx_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty intrfc = ipxitf_find_using_phys(dev, pt->type); if (!intrfc) { if (ipxcfg_auto_create_interfaces && - ntohl(IPX_SKB_CB(skb)->ipx_dest_net)) { + IPX_SKB_CB(skb)->ipx_dest_net) { intrfc = ipxitf_auto_create(dev, pt->type); if (intrfc) ipxitf_hold(intrfc); diff --git a/net/ipx/ipx_proc.c b/net/ipx/ipx_proc.c index 4c0c712..b7463df 100644 --- a/net/ipx/ipx_proc.c +++ b/net/ipx/ipx_proc.c @@ -260,22 +260,22 @@ static int ipx_seq_socket_show(struct seq_file *seq, void *v) ipxs = ipx_sk(s); #ifdef CONFIG_IPX_INTERN seq_printf(seq, "%08lX:%02X%02X%02X%02X%02X%02X:%04X ", - (unsigned long)htonl(ipxs->intrfc->if_netnum), + (unsigned long)ntohl(ipxs->intrfc->if_netnum), ipxs->node[0], ipxs->node[1], ipxs->node[2], ipxs->node[3], - ipxs->node[4], ipxs->node[5], htons(ipxs->port)); + ipxs->node[4], ipxs->node[5], ntohs(ipxs->port)); #else - seq_printf(seq, "%08lX:%04X ", (unsigned long) htonl(ipxs->intrfc->if_netnum), - htons(ipxs->port)); + seq_printf(seq, "%08lX:%04X ", (unsigned long) ntohl(ipxs->intrfc->if_netnum), + ntohs(ipxs->port)); #endif /* CONFIG_IPX_INTERN */ if (s->sk_state != TCP_ESTABLISHED) seq_printf(seq, "%-28s", "Not_Connected"); else { seq_printf(seq, "%08lX:%02X%02X%02X%02X%02X%02X:%04X ", - (unsigned long)htonl(ipxs->dest_addr.net), + (unsigned long)ntohl(ipxs->dest_addr.net), ipxs->dest_addr.node[0], ipxs->dest_addr.node[1], ipxs->dest_addr.node[2], ipxs->dest_addr.node[3], ipxs->dest_addr.node[4], ipxs->dest_addr.node[5], - htons(ipxs->dest_addr.sock)); + ntohs(ipxs->dest_addr.sock)); } seq_printf(seq, "%08X %08X %02X %03d\n", diff --git a/net/ipx/ipx_route.c b/net/ipx/ipx_route.c index a30dbb1..8bfaefa 100644 --- a/net/ipx/ipx_route.c +++ b/net/ipx/ipx_route.c @@ -20,16 +20,16 @@ DEFINE_RWLOCK(ipx_routes_lock); extern struct ipx_interface *ipx_internal_net; extern __u16 ipx_cksum(struct ipxhdr *packet, int length); -extern struct ipx_interface *ipxitf_find_using_net(__u32 net); +extern struct ipx_interface *ipxitf_find_using_net(__be32 net); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); extern int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node); -extern struct ipx_interface *ipxitf_find_using_net(__u32 net); +extern struct ipx_interface *ipxitf_find_using_net(__be32 net); -struct ipx_route *ipxrtr_lookup(__u32 net) +struct ipx_route *ipxrtr_lookup(__be32 net) { struct ipx_route *r; @@ -48,7 +48,7 @@ unlock: /* * Caller must hold a reference to intrfc */ -int ipxrtr_add_route(__u32 network, struct ipx_interface *intrfc, +int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, unsigned char *node) { struct ipx_route *rt; @@ -118,7 +118,7 @@ out: return rc; } -static int ipxrtr_delete(__u32 net) +static int ipxrtr_delete(__be32 net) { struct ipx_route *r, *tmp; int rc; -- cgit v0.10.2 From 02e60370d4dac83f22d5ae75d5512bcb9a3f24b7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:28:23 -0800 Subject: [IPX]: Annotate and fix IPX checksum Calculation of IPX checksum got buggered about 2.4.0. The old variant mangled the packet; that got fixed, but calculation itself got buggered. Restored the correct logics, fixed a subtle breakage we used to have even back then: if the sum is 0 mod 0xffff, we want to return 0, not 0xffff. The latter has special meaning for IPX (cheksum disabled). Observation (and obvious fix) nicked from history of FreeBSD ipx_cksum.c... Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/include/net/ipx.h b/include/net/ipx.h index 4a423d2..c6b2ee6 100644 --- a/include/net/ipx.h +++ b/include/net/ipx.h @@ -26,8 +26,8 @@ struct ipx_address { #define IPX_MAX_PPROP_HOPS 8 struct ipxhdr { - __u16 ipx_checksum __attribute__ ((packed)); -#define IPX_NO_CHECKSUM 0xFFFF + __be16 ipx_checksum __attribute__ ((packed)); +#define IPX_NO_CHECKSUM __constant_htons(0xFFFF) __be16 ipx_pktsize __attribute__ ((packed)); __u8 ipx_tctrl; __u8 ipx_type; diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index c272a38a..76c6615 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -1234,27 +1234,27 @@ static int ipxitf_ioctl(unsigned int cmd, void __user *arg) /* Note: We assume ipx_tctrl==0 and htons(length)==ipx_pktsize */ /* This functions should *not* mess with packet contents */ -__u16 ipx_cksum(struct ipxhdr *packet, int length) +__be16 ipx_cksum(struct ipxhdr *packet, int length) { /* * NOTE: sum is a net byte order quantity, which optimizes the * loop. This only works on big and little endian machines. (I * don't know of a machine that isn't.) */ - /* start at ipx_dest - We skip the checksum field and start with - * ipx_type before the loop, not considering ipx_tctrl in the calc */ - __u16 *p = (__u16 *)&packet->ipx_dest; - __u32 i = (length >> 1) - 1; /* Number of complete words */ - __u32 sum = packet->ipx_type << sizeof(packet->ipx_tctrl); - - /* Loop through all complete words except the checksum field, - * ipx_type (accounted above) and ipx_tctrl (not used in the cksum) */ - while (--i) + /* handle the first 3 words separately; checksum should be skipped + * and ipx_tctrl masked out */ + __u16 *p = (__u16 *)packet; + __u32 sum = p[1] + (p[2] & (__force u16)htons(0x00ff)); + __u32 i = (length >> 1) - 3; /* Number of remaining complete words */ + + /* Loop through them */ + p += 3; + while (i--) sum += *p++; /* Add on the last part word if it exists */ if (packet->ipx_pktsize & htons(1)) - sum += ntohs(0xff00) & *p; + sum += (__force u16)htons(0xff00) & *p; /* Do final fixup */ sum = (sum & 0xffff) + (sum >> 16); @@ -1263,7 +1263,14 @@ __u16 ipx_cksum(struct ipxhdr *packet, int length) if (sum >= 0x10000) sum++; - return ~sum; + /* + * Leave 0 alone; we don't want 0xffff here. Note that we can't get + * here with 0x10000, so this check is the same as ((__u16)sum) + */ + if (sum) + sum = ~sum; + + return (__force __be16)sum; } const char *ipx_frame_name(__be16 frame) diff --git a/net/ipx/ipx_route.c b/net/ipx/ipx_route.c index 8bfaefa..68560ee 100644 --- a/net/ipx/ipx_route.c +++ b/net/ipx/ipx_route.c @@ -19,7 +19,7 @@ DEFINE_RWLOCK(ipx_routes_lock); extern struct ipx_interface *ipx_internal_net; -extern __u16 ipx_cksum(struct ipxhdr *packet, int length); +extern __be16 ipx_cksum(struct ipxhdr *packet, int length); extern struct ipx_interface *ipxitf_find_using_net(__be32 net); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); @@ -238,7 +238,7 @@ int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, /* Apply checksum. Not allowed on 802.3 links. */ if (sk->sk_no_check || intrfc->if_dlink_type == htons(IPX_FRAME_8023)) - ipx->ipx_checksum = 0xFFFF; + ipx->ipx_checksum = htons(0xFFFF); else ipx->ipx_checksum = ipx_cksum(ipx, len + sizeof(struct ipxhdr)); -- cgit v0.10.2 From 95026cd242bd4188a036f2eba20994113ed5a5d7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:55:35 -0800 Subject: [IPV6]: Fix ECN bug on big-endian __constant_htons(2<<4) is not a replacement for htonl(2<<20). Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h index d599c6b..7849844 100644 --- a/include/net/inet_ecn.h +++ b/include/net/inet_ecn.h @@ -48,7 +48,7 @@ static inline __u8 INET_ECN_encapsulate(__u8 outer, __u8 inner) #define IP6_ECN_flow_xmit(sk, label) do { \ if (INET_ECN_is_capable(inet_sk(sk)->tos)) \ - (label) |= __constant_htons(INET_ECN_ECT_0 << 4); \ + (label) |= htonl(INET_ECN_ECT_0 << 20); \ } while (0) static inline int IP_ECN_set_ce(struct iphdr *iph) -- cgit v0.10.2 From 7ac00a24f379f8ab9d3c968fea3dc030a45956fd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:58:17 -0800 Subject: [NETFILTER] bug: NFULA_CFG_QTHRESH uses 32bit Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index b59d3b2..fca946c 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -896,8 +896,8 @@ nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, } if (nfula[NFULA_CFG_QTHRESH-1]) { - u_int32_t qthresh = - *(u_int16_t *)NFA_DATA(nfula[NFULA_CFG_QTHRESH-1]); + __be32 qthresh = + *(__be32 *)NFA_DATA(nfula[NFULA_CFG_QTHRESH-1]); nfulnl_set_qthresh(inst, ntohl(qthresh)); } -- cgit v0.10.2 From d1208b999dd367b72168cc3c7f8d8d2c95143c67 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:58:41 -0800 Subject: [NETFILTER] bug: nfulnl_msg_config_mode ->copy_range is 32bit Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index fca946c..00580a8 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -878,7 +878,7 @@ nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, params = NFA_DATA(nfula[NFULA_CFG_MODE-1]); nfulnl_set_mode(inst, params->copy_mode, - ntohs(params->copy_range)); + ntohl(params->copy_range)); } if (nfula[NFULA_CFG_TIMEOUT-1]) { -- cgit v0.10.2 From febf0a431e42f5a1fdb2b763273700610552ddcc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:59:17 -0800 Subject: [NETFILTER] bug: skb->protocol is already net-endian htons() is not needed (and no, it's not misspelled ntohs() - userland expects net-endian here). Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 00580a8..b2bf8f2 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -427,7 +427,7 @@ __build_packet_message(struct nfulnl_instance *inst, nfmsg->version = NFNETLINK_V0; nfmsg->res_id = htons(inst->group_num); - pmsg.hw_protocol = htons(skb->protocol); + pmsg.hw_protocol = skb->protocol; pmsg.hook = hooknum; NFA_PUT(inst->skb, NFULA_PACKET_HDR, sizeof(pmsg), &pmsg); diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 8eb2473..6e4ada3 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -414,7 +414,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, nfmsg->res_id = htons(queue->queue_num); pmsg.packet_id = htonl(entry->id); - pmsg.hw_protocol = htons(entskb->protocol); + pmsg.hw_protocol = entskb->protocol; pmsg.hook = entinf->hook; NFA_PUT(skb, NFQA_PACKET_HDR, sizeof(pmsg), &pmsg); -- cgit v0.10.2 From 36da4d869f23bc7d1a70a3185218cb626537845c Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 3 Nov 2006 01:01:03 -0800 Subject: [TG3]: Fix 2nd ifup failure on 5752M. This fixes a bug reported in: http://bugzilla.kernel.org/show_bug.cgi?id=7438 tg3_close() turns off the PHY if WoL and ASF are both disabled. On the next tg3_open(), some devices such as the 5752M will not be brought up correctly without a PHY reset early in the reset sequence. The PHY clock is needed for some internal MAC blocks to function correctly. This problem is fixed by always resetting the PHY early in tg3_reset_hw() when it is called from tg3_open() or tg3_resume(). tg3_setup_phy() can then be called later in the sequence without the reset_phy parameter set to 1, since the PHY reset is already done. Update version to 3.68. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 8e39849..8f059b7 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -68,8 +68,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.67" -#define DRV_MODULE_RELDATE "October 18, 2006" +#define DRV_MODULE_VERSION "3.68" +#define DRV_MODULE_RELDATE "November 02, 2006" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -6014,7 +6014,7 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy) tg3_abort_hw(tp, 1); } - if ((tp->tg3_flags2 & TG3_FLG2_MII_SERDES) && reset_phy) + if (reset_phy) tg3_phy_reset(tp); err = tg3_chip_reset(tp); @@ -6574,7 +6574,7 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy) tw32(GRC_LOCAL_CTRL, tp->grc_local_ctrl); } - err = tg3_setup_phy(tp, reset_phy); + err = tg3_setup_phy(tp, 0); if (err) return err; -- cgit v0.10.2 From 0f37c6057414fb68024793966b1dcb6a135cb844 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 03:49:56 -0800 Subject: [PKTGEN]: TCI endianness fixes open-coded variant there works only for little-endian Signed-off-by: Al Viro Signed-off-by: David S. Miller diff --git a/net/core/pktgen.c b/net/core/pktgen.c index dd023fd..733d86d 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -2304,6 +2304,12 @@ static void mpls_push(__be32 *mpls, struct pktgen_dev *pkt_dev) *mpls |= MPLS_STACK_BOTTOM; } +static inline __be16 build_tci(unsigned int id, unsigned int cfi, + unsigned int prio) +{ + return htons(id | (cfi << 12) | (prio << 13)); +} + static struct sk_buff *fill_packet_ipv4(struct net_device *odev, struct pktgen_dev *pkt_dev) { @@ -2353,16 +2359,16 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev, if (pkt_dev->vlan_id != 0xffff) { if(pkt_dev->svlan_id != 0xffff) { svlan_tci = (__be16 *)skb_put(skb, sizeof(__be16)); - *svlan_tci = htons(pkt_dev->svlan_id); - *svlan_tci |= pkt_dev->svlan_p << 5; - *svlan_tci |= pkt_dev->svlan_cfi << 4; + *svlan_tci = build_tci(pkt_dev->svlan_id, + pkt_dev->svlan_cfi, + pkt_dev->svlan_p); svlan_encapsulated_proto = (__be16 *)skb_put(skb, sizeof(__be16)); *svlan_encapsulated_proto = __constant_htons(ETH_P_8021Q); } vlan_tci = (__be16 *)skb_put(skb, sizeof(__be16)); - *vlan_tci = htons(pkt_dev->vlan_id); - *vlan_tci |= pkt_dev->vlan_p << 5; - *vlan_tci |= pkt_dev->vlan_cfi << 4; + *vlan_tci = build_tci(pkt_dev->vlan_id, + pkt_dev->vlan_cfi, + pkt_dev->vlan_p); vlan_encapsulated_proto = (__be16 *)skb_put(skb, sizeof(__be16)); *vlan_encapsulated_proto = __constant_htons(ETH_P_IP); } @@ -2689,16 +2695,16 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev, if (pkt_dev->vlan_id != 0xffff) { if(pkt_dev->svlan_id != 0xffff) { svlan_tci = (__be16 *)skb_put(skb, sizeof(__be16)); - *svlan_tci = htons(pkt_dev->svlan_id); - *svlan_tci |= pkt_dev->svlan_p << 5; - *svlan_tci |= pkt_dev->svlan_cfi << 4; + *svlan_tci = build_tci(pkt_dev->svlan_id, + pkt_dev->svlan_cfi, + pkt_dev->svlan_p); svlan_encapsulated_proto = (__be16 *)skb_put(skb, sizeof(__be16)); *svlan_encapsulated_proto = __constant_htons(ETH_P_8021Q); } vlan_tci = (__be16 *)skb_put(skb, sizeof(__be16)); - *vlan_tci = htons(pkt_dev->vlan_id); - *vlan_tci |= pkt_dev->vlan_p << 5; - *vlan_tci |= pkt_dev->vlan_cfi << 4; + *vlan_tci = build_tci(pkt_dev->vlan_id, + pkt_dev->vlan_cfi, + pkt_dev->vlan_p); vlan_encapsulated_proto = (__be16 *)skb_put(skb, sizeof(__be16)); *vlan_encapsulated_proto = __constant_htons(ETH_P_IPV6); } -- cgit v0.10.2 From db38c179a759a9c4722525e8c9f09ac80e372377 Mon Sep 17 00:00:00 2001 From: Larry Woodman Date: Fri, 3 Nov 2006 16:05:45 -0800 Subject: [NET]: __alloc_pages() failures reported due to fragmentation We have seen a couple of __alloc_pages() failures due to fragmentation, there is plenty of free memory but no large order pages available. I think the problem is in sock_alloc_send_pskb(), the gfp_mask includes __GFP_REPEAT but its never used/passed to the page allocator. Shouldnt the gfp_mask be passed to alloc_skb() ? Signed-off-by: Larry Woodman Signed-off-by: David S. Miller diff --git a/net/core/sock.c b/net/core/sock.c index d472db4..ee6cd25 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1160,7 +1160,7 @@ static struct sk_buff *sock_alloc_send_pskb(struct sock *sk, goto failure; if (atomic_read(&sk->sk_wmem_alloc) < sk->sk_sndbuf) { - skb = alloc_skb(header_len, sk->sk_allocation); + skb = alloc_skb(header_len, gfp_mask); if (skb) { int npages; int i; -- cgit v0.10.2 From 36f73d0c3b7efa72cd8b89f2d429ff39bc12f15c Mon Sep 17 00:00:00 2001 From: Dmitry Mishin Date: Fri, 3 Nov 2006 16:08:19 -0800 Subject: [IPV6]: Add ndisc_netdev_notifier unregister. If inet6_init() fails later than ndisc_init() call, or IPv6 module is unloaded, ndisc_netdev_notifier call remains in the list and will follows in oops later. Signed-off-by: Dmitry Mishin Signed-off-by: David S. Miller diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 41a8a5f..73eb8c3 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1742,6 +1742,7 @@ int __init ndisc_init(struct net_proto_family *ops) void ndisc_cleanup(void) { + unregister_netdevice_notifier(&ndisc_netdev_notifier); #ifdef CONFIG_SYSCTL neigh_sysctl_unregister(&nd_tbl.parms); #endif -- cgit v0.10.2 From daccff024ffeb21caa2cc479ccc33b2ec50705b1 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sun, 5 Nov 2006 15:47:04 -0800 Subject: [IPV6]: Give sit driver an appropriate module alias. It would be nice to keep things working even with this built as a module, it took me some time to realize my IPv6 tunnel was broken because of the missing sit module. This module alias fixes things until distributions have added an appropriate alias to modprobe.conf. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index b481a4d..be699f8 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -854,3 +854,4 @@ int __init sit_init(void) module_init(sit_init); module_exit(sit_cleanup); MODULE_LICENSE("GPL"); +MODULE_ALIAS("sit0"); -- cgit v0.10.2 From 38c94377a36f70e86665231c9f477e445c806618 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Sun, 5 Nov 2006 16:44:06 -0800 Subject: [NETLABEL]: Fix build failure. > the build with the attached .config failed, make ends with: > ... > : undefined reference to `cipso_v4_sock_getattr' > net/built-in.o: In function `netlbl_socket_getattr': ... It looks like I was stupid and made NetLabel depend on CONFIG_NET and not CONFIG_INET, the patch below should fix this by making NetLabel depend on CONFIG_INET and CONFIG_SECURITY. Please review and apply for 2.6.19. Signed-off-by: Paul Moore Signed-off-by: David S. Miller diff --git a/net/Kconfig b/net/Kconfig index a81aca4..67e39ad 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -63,6 +63,7 @@ config INET if INET source "net/ipv4/Kconfig" source "net/ipv6/Kconfig" +source "net/netlabel/Kconfig" endif # if INET @@ -249,8 +250,6 @@ source "net/ieee80211/Kconfig" config WIRELESS_EXT bool -source "net/netlabel/Kconfig" - config FIB_RULES bool diff --git a/net/netlabel/Kconfig b/net/netlabel/Kconfig index 9f7121a..56958c8 100644 --- a/net/netlabel/Kconfig +++ b/net/netlabel/Kconfig @@ -4,7 +4,7 @@ config NETLABEL bool "NetLabel subsystem support" - depends on NET && SECURITY + depends on SECURITY default n ---help--- NetLabel provides support for explicit network packet labeling -- cgit v0.10.2 From 59359ff87700f5e742c96a55da9cf0819984c128 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 5 Nov 2006 16:51:03 -0800 Subject: [SPARC]: Fix robust futex syscalls and wire up migrate_pages. When I added the entries for the robust futex syscall entries, I forgot to bump NR_SYSCALLS. The current situation is error-prone because NR_SYSCALLS lives in entry.S where the system call limit checks are enforced. Move the definition to asm/unistd.h in order to make this mistake much more difficult to make. And wire up sys_migrate_pages since the powerpc folks implemented the compat wrapper for us. Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S index a4edff4..831f540 100644 --- a/arch/sparc/kernel/entry.S +++ b/arch/sparc/kernel/entry.S @@ -32,13 +32,12 @@ #include #include #include +#include #include #define curptr g6 -#define NR_SYSCALLS 300 /* Each OS is different... */ - /* These are just handy. */ #define _SV save %sp, -STACKFRAME_SZ, %sp #define _RS restore diff --git a/arch/sparc/kernel/systbls.S b/arch/sparc/kernel/systbls.S index 10df38e..ea75ca5 100644 --- a/arch/sparc/kernel/systbls.S +++ b/arch/sparc/kernel/systbls.S @@ -78,7 +78,7 @@ sys_call_table: /*285*/ .long sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_fstatat64 /*290*/ .long sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat /*295*/ .long sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll, sys_unshare -/*300*/ .long sys_set_robust_list, sys_get_robust_list +/*300*/ .long sys_set_robust_list, sys_get_robust_list, sys_migrate_pages #ifdef CONFIG_SUNOS_EMUL /* Now the SunOS syscall table. */ @@ -190,6 +190,7 @@ sunos_sys_table: /*290*/ .long sunos_nosys, sunos_nosys, sunos_nosys .long sunos_nosys, sunos_nosys, sunos_nosys .long sunos_nosys, sunos_nosys, sunos_nosys - .long sunos_nosys, sunos_nosys, sunos_nosys + .long sunos_nosys +/*300*/ .long sunos_nosys, sunos_nosys, sunos_nosys #endif diff --git a/arch/sparc64/kernel/entry.S b/arch/sparc64/kernel/entry.S index 0aaa35f..6f28bec 100644 --- a/arch/sparc64/kernel/entry.S +++ b/arch/sparc64/kernel/entry.S @@ -22,11 +22,10 @@ #include #include #include +#include #define curptr g6 -#define NR_SYSCALLS 300 /* Each OS is different... */ - .text .align 32 diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S index 419a63f..9a80267 100644 --- a/arch/sparc64/kernel/systbls.S +++ b/arch/sparc64/kernel/systbls.S @@ -79,7 +79,7 @@ sys_call_table32: .word sys_mkdirat, sys_mknodat, sys_fchownat, compat_sys_futimesat, compat_sys_fstatat64 /*290*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat .word sys_fchmodat, sys_faccessat, compat_sys_pselect6, compat_sys_ppoll, sys_unshare -/*300*/ .word compat_sys_set_robust_list, compat_sys_get_robust_list +/*300*/ .word compat_sys_set_robust_list, compat_sys_get_robust_list, compat_sys_migrate_pages #endif /* CONFIG_COMPAT */ @@ -149,7 +149,7 @@ sys_call_table: .word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_fstatat64 /*290*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat .word sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll, sys_unshare -/*300*/ .word sys_set_robust_list, sys_get_robust_list +/*300*/ .word sys_set_robust_list, sys_get_robust_list, sys_migrate_pages #if defined(CONFIG_SUNOS_EMUL) || defined(CONFIG_SOLARIS_EMUL) || \ defined(CONFIG_SOLARIS_EMUL_MODULE) @@ -262,5 +262,7 @@ sunos_sys_table: /*290*/ .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys .word sunos_nosys, sunos_nosys, sunos_nosys - .word sunos_nosys, sunos_nosys, sunos_nosys + .word sunos_nosys +/*300*/ .word sunos_nosys, sunos_nosys, sunos_nosys + #endif diff --git a/include/asm-sparc/unistd.h b/include/asm-sparc/unistd.h index c7a495a..f7827fa 100644 --- a/include/asm-sparc/unistd.h +++ b/include/asm-sparc/unistd.h @@ -318,12 +318,15 @@ #define __NR_unshare 299 #define __NR_set_robust_list 300 #define __NR_get_robust_list 301 +#define __NR_migrate_pages 302 + +#define NR_SYSCALLS 303 #ifdef __KERNEL__ -/* WARNING: You MAY NOT add syscall numbers larger than 301, since +/* WARNING: You MAY NOT add syscall numbers larger than 302, since * all of the syscall tables in the Sparc kernel are - * sized to have 301 entries (starting at zero). Therefore - * find a free slot in the 0-301 range. + * sized to have 302 entries (starting at zero). Therefore + * find a free slot in the 0-302 range. */ #define _syscall0(type,name) \ diff --git a/include/asm-sparc64/unistd.h b/include/asm-sparc64/unistd.h index 124cf07..63669da 100644 --- a/include/asm-sparc64/unistd.h +++ b/include/asm-sparc64/unistd.h @@ -320,12 +320,16 @@ #define __NR_unshare 299 #define __NR_set_robust_list 300 #define __NR_get_robust_list 301 +#define __NR_migrate_pages 302 + +#define NR_SYSCALLS 303 #ifdef __KERNEL__ -/* WARNING: You MAY NOT add syscall numbers larger than 301, since + +/* WARNING: You MAY NOT add syscall numbers larger than 302, since * all of the syscall tables in the Sparc kernel are - * sized to have 301 entries (starting at zero). Therefore - * find a free slot in the 0-301 range. + * sized to have 302 entries (starting at zero). Therefore + * find a free slot in the 0-302 range. */ #define _syscall0(type,name) \ -- cgit v0.10.2 From 1b5135d9b922fdcf46e1e7383167d93d42635fb4 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Fri, 3 Nov 2006 17:47:20 +0100 Subject: [PATCH] ehea: Nullpointer dereferencation fix Fix: Must check for nullpointer before dereferencing it - not afterwards. Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik diff --git a/drivers/net/ehea/ehea_qmr.c b/drivers/net/ehea/ehea_qmr.c index 3e18623..1615593 100644 --- a/drivers/net/ehea/ehea_qmr.c +++ b/drivers/net/ehea/ehea_qmr.c @@ -209,11 +209,11 @@ int ehea_destroy_cq(struct ehea_cq *cq) { u64 adapter_handle, hret; - adapter_handle = cq->adapter->handle; - if (!cq) return 0; + adapter_handle = cq->adapter->handle; + /* deregister all previous registered pages */ hret = ehea_h_free_resource(adapter_handle, cq->fw_handle); if (hret != H_SUCCESS) { -- cgit v0.10.2 From 07fd06b3bc1589e44aefd02eb28700a51b3c9d12 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Fri, 3 Nov 2006 17:47:52 +0100 Subject: [PATCH] ehea: Removed redundant define Removed define H_CB_ALIGNMENT which is already defined in include/asm-powerpc/hvcall.h Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h index b40724f..39ad9f7 100644 --- a/drivers/net/ehea/ehea.h +++ b/drivers/net/ehea/ehea.h @@ -39,7 +39,7 @@ #include #define DRV_NAME "ehea" -#define DRV_VERSION "EHEA_0034" +#define DRV_VERSION "EHEA_0043" #define EHEA_MSG_DEFAULT (NETIF_MSG_LINK | NETIF_MSG_TIMER \ | NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) @@ -105,9 +105,6 @@ #define EHEA_BCMC_VLANID_ALL 0x01 #define EHEA_BCMC_VLANID_SINGLE 0x00 -/* Use this define to kmallocate pHYP control blocks */ -#define H_CB_ALIGNMENT 4096 - #define EHEA_CACHE_LINE 128 /* Memory Regions */ -- cgit v0.10.2 From a1d261c561522151cb96c75f1dd1a51cf17665cf Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Fri, 3 Nov 2006 17:48:23 +0100 Subject: [PATCH] ehea: 64K page support fix This patch fixes 64k page support by using PAGE_MASK and appropriate pagesize defines in several places. Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik diff --git a/drivers/net/ehea/ehea_ethtool.c b/drivers/net/ehea/ehea_ethtool.c index 82eb2fb..9f57c2e 100644 --- a/drivers/net/ehea/ehea_ethtool.c +++ b/drivers/net/ehea/ehea_ethtool.c @@ -238,7 +238,7 @@ static void ehea_get_ethtool_stats(struct net_device *dev, data[i++] = port->port_res[0].swqe_refill_th; data[i++] = port->resets; - cb6 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb6 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb6) { ehea_error("no mem for cb6"); return; diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 4538c99..6ad6961 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -92,7 +92,7 @@ static struct net_device_stats *ehea_get_stats(struct net_device *dev) memset(stats, 0, sizeof(*stats)); - cb2 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb2 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb2) { ehea_error("no mem for cb2"); goto out; @@ -586,8 +586,8 @@ int ehea_sense_port_attr(struct ehea_port *port) u64 hret; struct hcp_ehea_port_cb0 *cb0; - cb0 = kzalloc(H_CB_ALIGNMENT, GFP_ATOMIC); /* May be called via */ - if (!cb0) { /* ehea_neq_tasklet() */ + cb0 = kzalloc(PAGE_SIZE, GFP_ATOMIC); /* May be called via */ + if (!cb0) { /* ehea_neq_tasklet() */ ehea_error("no mem for cb0"); ret = -ENOMEM; goto out; @@ -670,7 +670,7 @@ int ehea_set_portspeed(struct ehea_port *port, u32 port_speed) u64 hret; int ret = 0; - cb4 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb4 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb4) { ehea_error("no mem for cb4"); ret = -ENOMEM; @@ -985,7 +985,7 @@ static int ehea_configure_port(struct ehea_port *port) struct hcp_ehea_port_cb0 *cb0; ret = -ENOMEM; - cb0 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb0 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb0) goto out; @@ -1443,7 +1443,7 @@ static int ehea_set_mac_addr(struct net_device *dev, void *sa) goto out; } - cb0 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb0 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb0) { ehea_error("no mem for cb0"); ret = -ENOMEM; @@ -1501,7 +1501,7 @@ static void ehea_promiscuous(struct net_device *dev, int enable) if ((enable && port->promisc) || (!enable && !port->promisc)) return; - cb7 = kzalloc(H_CB_ALIGNMENT, GFP_ATOMIC); + cb7 = kzalloc(PAGE_SIZE, GFP_ATOMIC); if (!cb7) { ehea_error("no mem for cb7"); goto out; @@ -1870,7 +1870,7 @@ static void ehea_vlan_rx_register(struct net_device *dev, port->vgrp = grp; - cb1 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb1 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb1) { ehea_error("no mem for cb1"); goto out; @@ -1899,7 +1899,7 @@ static void ehea_vlan_rx_add_vid(struct net_device *dev, unsigned short vid) int index; u64 hret; - cb1 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb1 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb1) { ehea_error("no mem for cb1"); goto out; @@ -1935,7 +1935,7 @@ static void ehea_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid) if (port->vgrp) port->vgrp->vlan_devices[vid] = NULL; - cb1 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb1 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb1) { ehea_error("no mem for cb1"); goto out; @@ -1968,7 +1968,7 @@ int ehea_activate_qp(struct ehea_adapter *adapter, struct ehea_qp *qp) u64 dummy64 = 0; struct hcp_modify_qp_cb0* cb0; - cb0 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb0 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb0) { ret = -ENOMEM; goto out; @@ -2269,7 +2269,7 @@ int ehea_sense_adapter_attr(struct ehea_adapter *adapter) u64 hret; int ret; - cb = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb) { ret = -ENOMEM; goto out; @@ -2340,7 +2340,7 @@ static int ehea_setup_single_port(struct ehea_port *port, goto out; /* Enable Jumbo frames */ - cb4 = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + cb4 = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!cb4) { ehea_error("no mem for cb4"); } else { diff --git a/drivers/net/ehea/ehea_phyp.c b/drivers/net/ehea/ehea_phyp.c index 0b51a8c..0cfc2bc 100644 --- a/drivers/net/ehea/ehea_phyp.c +++ b/drivers/net/ehea/ehea_phyp.c @@ -506,7 +506,7 @@ u64 ehea_h_register_rpage_mr(const u64 adapter_handle, const u64 mr_handle, const u8 pagesize, const u8 queue_type, const u64 log_pageaddr, const u64 count) { - if ((count > 1) && (log_pageaddr & 0xfff)) { + if ((count > 1) && (log_pageaddr & ~PAGE_MASK)) { ehea_error("not on pageboundary"); return H_PARAMETER; } diff --git a/drivers/net/ehea/ehea_phyp.h b/drivers/net/ehea/ehea_phyp.h index fa51e3b..919f94b 100644 --- a/drivers/net/ehea/ehea_phyp.h +++ b/drivers/net/ehea/ehea_phyp.h @@ -81,14 +81,16 @@ static inline u32 get_longbusy_msecs(int long_busy_ret_code) static inline void hcp_epas_ctor(struct h_epas *epas, u64 paddr_kernel, u64 paddr_user) { - epas->kernel.addr = ioremap(paddr_kernel, PAGE_SIZE); + /* To support 64k pages we must round to 64k page boundary */ + epas->kernel.addr = ioremap((paddr_kernel & PAGE_MASK), PAGE_SIZE) + + (paddr_kernel & ~PAGE_MASK); epas->user.addr = paddr_user; } static inline void hcp_epas_dtor(struct h_epas *epas) { if (epas->kernel.addr) - iounmap(epas->kernel.addr); + iounmap((void __iomem*)((u64)epas->kernel.addr & PAGE_MASK)); epas->user.addr = 0; epas->kernel.addr = 0; diff --git a/drivers/net/ehea/ehea_qmr.c b/drivers/net/ehea/ehea_qmr.c index 1615593..72ef7bd 100644 --- a/drivers/net/ehea/ehea_qmr.c +++ b/drivers/net/ehea/ehea_qmr.c @@ -512,7 +512,7 @@ int ehea_reg_mr_adapter(struct ehea_adapter *adapter) start = KERNELBASE; end = (u64)high_memory; - nr_pages = (end - start) / PAGE_SIZE; + nr_pages = (end - start) / EHEA_PAGESIZE; pt = kzalloc(PAGE_SIZE, GFP_KERNEL); if (!pt) { @@ -538,9 +538,9 @@ int ehea_reg_mr_adapter(struct ehea_adapter *adapter) if (nr_pages > 1) { u64 num_pages = min(nr_pages, (u64)512); for (i = 0; i < num_pages; i++) - pt[i] = virt_to_abs((void*)(((u64)start) - + ((k++) * - PAGE_SIZE))); + pt[i] = virt_to_abs((void*)(((u64)start) + + ((k++) * + EHEA_PAGESIZE))); hret = ehea_h_register_rpage_mr(adapter->handle, adapter->mr.handle, 0, @@ -548,8 +548,9 @@ int ehea_reg_mr_adapter(struct ehea_adapter *adapter) num_pages); nr_pages -= num_pages; } else { - u64 abs_adr = virt_to_abs((void*)(((u64)start) - + (k * PAGE_SIZE))); + u64 abs_adr = virt_to_abs((void*)(((u64)start) + + (k * EHEA_PAGESIZE))); + hret = ehea_h_register_rpage_mr(adapter->handle, adapter->mr.handle, 0, 0, abs_adr,1); -- cgit v0.10.2 From a81c52a81d6dbe6a36bce18112da04f20b175192 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 1 Nov 2006 21:18:58 -0800 Subject: [PATCH] Kconfig: remove redundant NETDEVICES depends drivers/net/Kconfig says: # All the following symbols are dependent on NETDEVICES - do not repeat # that for each of the symbols. so remove duplicate 'depends' uses of NETDEVICES. Signed-off-by: Randy Dunlap Signed-off-by: Jeff Garzik diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 28c17d1..9cb3ca5 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -486,7 +486,7 @@ config SGI_IOC3_ETH_HW_TX_CSUM config MIPS_SIM_NET tristate "MIPS simulator Network device (EXPERIMENTAL)" - depends on NETDEVICES && MIPS_SIM && EXPERIMENTAL + depends on MIPS_SIM && EXPERIMENTAL help The MIPSNET device is a simple Ethernet network device which is emulated by the MIPS Simulator. @@ -2467,7 +2467,7 @@ config ISERIES_VETH config RIONET tristate "RapidIO Ethernet over messaging driver support" - depends on NETDEVICES && RAPIDIO + depends on RAPIDIO config RIONET_TX_SIZE int "Number of outbound queue entries" -- cgit v0.10.2 From e9a43850e6a99e335c8120579749389b376bc204 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Mon, 30 Oct 2006 09:31:27 +0100 Subject: AVR32: Get rid of board_early_init board_early_init() is left over from some early prototyping work where we had to initialize the SDRAM controller ourselves. This depends on the kernel being loaded into static RAM, which just isn't possible on any commercially available products today. In order to run without a boot loader, we need to create a zImage stub or have the debugger initialize the SDRAM for us (for really low-level debugging) Signed-off-by: Haavard Skinnemoen diff --git a/arch/avr32/boards/atstk1000/setup.c b/arch/avr32/boards/atstk1000/setup.c index 191ab85..272c011 100644 --- a/arch/avr32/boards/atstk1000/setup.c +++ b/arch/avr32/boards/atstk1000/setup.c @@ -21,15 +21,6 @@ struct tag *bootloader_tags __initdata; struct lcdc_platform_data __initdata atstk1000_fb0_data; -asmlinkage void __init board_early_init(void) -{ - extern void sdram_init(void); - -#ifdef CONFIG_LOADER_STANDALONE - sdram_init(); -#endif -} - void __init board_setup_fbmem(unsigned long fbmem_start, unsigned long fbmem_size) { diff --git a/arch/avr32/kernel/head.S b/arch/avr32/kernel/head.S index 773b7ad..6163bd0 100644 --- a/arch/avr32/kernel/head.S +++ b/arch/avr32/kernel/head.S @@ -30,9 +30,6 @@ kernel_entry: mov r7, 0 #endif - /* Set up the PIO, SDRAM controller, early printk, etc. */ - rcall board_early_init - /* Start the show */ lddpc pc, kernel_start_addr -- cgit v0.10.2 From 168c8fa32ff40d6caf7226a5abf2a0096f3393bf Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Tue, 31 Oct 2006 20:01:11 +0100 Subject: AVR32: Fix thinko in generic_find_next_zero_le_bit() The existing implementation of this function seems to be looking for a one although it should be looking for a zero. This causes trouble for the ext2 filesystem, which tends to report -ENOSPC without this patch. Fix this by complementing each word before scanning. Signed-off-by: Haavard Skinnemoen diff --git a/arch/avr32/lib/findbit.S b/arch/avr32/lib/findbit.S index 2b4856f..c6b91de 100644 --- a/arch/avr32/lib/findbit.S +++ b/arch/avr32/lib/findbit.S @@ -136,6 +136,7 @@ ENTRY(generic_find_next_zero_le_bit) /* offset is not word-aligned. Handle the first (32 - r10) bits */ ldswp.w r8, r12[0] sub r12, -4 + com r8 lsr r8, r8, r10 brne .L_found @@ -146,7 +147,7 @@ ENTRY(generic_find_next_zero_le_bit) /* Main loop. offset must be word-aligned */ 1: ldswp.w r8, r12[0] - cp.w r8, 0 + com r8 brne .L_found sub r12, -4 sub r9, 32 -- cgit v0.10.2 From 4b96b1a10cb00c867103b21f0f2a6c91b705db11 Mon Sep 17 00:00:00 2001 From: Gautham R Shenoy Date: Sun, 5 Nov 2006 23:52:04 -0800 Subject: [PATCH] Fix the spurious unlock_cpu_hotplug false warnings Cpu-hotplug locking has a minor race case caused because of setting the variable "recursive" to NULL *after* releasing the cpu_bitmask_lock in the function unlock_cpu_hotplug,instead of doing so before releasing the cpu_bitmask_lock. This was the cause of most of the recent false spurious lock_cpu_unlock warnings. This should fix the problem reported by Martin Lorenz reported in http://lkml.org/lkml/2006/10/29/127. Thanks to Srinivasa DS for pointing it out. Signed-off-by: Gautham R Shenoy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/cpu.c b/kernel/cpu.c index 663c920..272254f 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -58,8 +58,8 @@ void unlock_cpu_hotplug(void) recursive_depth--; return; } - mutex_unlock(&cpu_bitmask_lock); recursive = NULL; + mutex_unlock(&cpu_bitmask_lock); } EXPORT_SYMBOL_GPL(unlock_cpu_hotplug); -- cgit v0.10.2 From 18a61e4adbc4dbe209e0d154df5cd37ce17dc314 Mon Sep 17 00:00:00 2001 From: Ankita Garg Date: Sun, 5 Nov 2006 23:52:07 -0800 Subject: [PATCH] Fix for LKDTM MEM_SWAPOUT crashpoint The MEM_SWAPOUT crashpoint in LKDTM could be broken as some compilers inline the call to shrink_page_list() and symbol lookup for this function name fails. Replacing it with the function shrink_inactive_list(), which is the only function calling shrink_page_list(). Signed-off-by: Ankita Garg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/misc/lkdtm.c b/drivers/misc/lkdtm.c index 46a9c35..db9d7df 100644 --- a/drivers/misc/lkdtm.c +++ b/drivers/misc/lkdtm.c @@ -157,8 +157,8 @@ void jp_ll_rw_block(int rw, int nr, struct buffer_head *bhs[]) struct scan_control; -unsigned long jp_shrink_page_list(struct list_head *page_list, - struct scan_control *sc) +unsigned long jp_shrink_inactive_list(unsigned long max_scan, + struct zone *zone, struct scan_control *sc) { lkdtm_handler(); jprobe_return(); @@ -297,8 +297,8 @@ int lkdtm_module_init(void) lkdtm.entry = (kprobe_opcode_t*) jp_ll_rw_block; break; case MEM_SWAPOUT: - lkdtm.kp.symbol_name = "shrink_page_list"; - lkdtm.entry = (kprobe_opcode_t*) jp_shrink_page_list; + lkdtm.kp.symbol_name = "shrink_inactive_list"; + lkdtm.entry = (kprobe_opcode_t*) jp_shrink_inactive_list; break; case TIMERADD: lkdtm.kp.symbol_name = "hrtimer_start"; -- cgit v0.10.2 From e5b9a335fd2180c6db1bcc4b24e83aff7481ebe3 Mon Sep 17 00:00:00 2001 From: Tilman Schmidt Date: Sun, 5 Nov 2006 23:52:08 -0800 Subject: [PATCH] isdn/gigaset: convert warning message Make the failed-to-allocate-skb warning a non-debug message. Signed-off-by: Tilman Schmidt Cc: Hansjoerg Lipp Cc: Karsten Keil Cc: Kai Germaschewski Cc: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index d8d256d..5800bee 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -616,7 +616,7 @@ static struct bc_state *gigaset_initbcs(struct bc_state *bcs, } else if ((bcs->skb = dev_alloc_skb(SBUFSIZE + HW_HDR_LEN)) != NULL) skb_reserve(bcs->skb, HW_HDR_LEN); else { - gig_dbg(DEBUG_INIT, "could not allocate skb\n"); + warn("could not allocate skb\n"); bcs->inputstate |= INS_skip_frame; } -- cgit v0.10.2 From 64efade11cddc4237c1b95ea4ca18af122a7e19e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sun, 5 Nov 2006 23:52:10 -0800 Subject: [PATCH] lockdep: fix delayacct locking bug Make the delayacct lock irqsave; this avoids the possible deadlock where an interrupt is taken while holding the delayacct lock which needs to take the delayacct lock. Signed-off-by: Peter Zijlstra Acked-by: Oleg Nesterov Cc: Balbir Singh Cc: Shailabh Nagar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 36752f1..66a0ea4 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -66,6 +66,7 @@ static void delayacct_end(struct timespec *start, struct timespec *end, { struct timespec ts; s64 ns; + unsigned long flags; do_posix_clock_monotonic_gettime(end); ts = timespec_sub(*end, *start); @@ -73,10 +74,10 @@ static void delayacct_end(struct timespec *start, struct timespec *end, if (ns < 0) return; - spin_lock(¤t->delays->lock); + spin_lock_irqsave(¤t->delays->lock, flags); *total += ns; (*count)++; - spin_unlock(¤t->delays->lock); + spin_unlock_irqrestore(¤t->delays->lock, flags); } void __delayacct_blkio_start(void) @@ -104,6 +105,7 @@ int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) s64 tmp; struct timespec ts; unsigned long t1,t2,t3; + unsigned long flags; /* Though tsk->delays accessed later, early exit avoids * unnecessary returning of other data @@ -136,14 +138,14 @@ int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) /* zero XXX_total, non-zero XXX_count implies XXX stat overflowed */ - spin_lock(&tsk->delays->lock); + spin_lock_irqsave(&tsk->delays->lock, flags); tmp = d->blkio_delay_total + tsk->delays->blkio_delay; d->blkio_delay_total = (tmp < d->blkio_delay_total) ? 0 : tmp; tmp = d->swapin_delay_total + tsk->delays->swapin_delay; d->swapin_delay_total = (tmp < d->swapin_delay_total) ? 0 : tmp; d->blkio_count += tsk->delays->blkio_count; d->swapin_count += tsk->delays->swapin_count; - spin_unlock(&tsk->delays->lock); + spin_unlock_irqrestore(&tsk->delays->lock, flags); done: return 0; @@ -152,11 +154,12 @@ done: __u64 __delayacct_blkio_ticks(struct task_struct *tsk) { __u64 ret; + unsigned long flags; - spin_lock(&tsk->delays->lock); + spin_lock_irqsave(&tsk->delays->lock, flags); ret = nsec_to_clock_t(tsk->delays->blkio_delay + tsk->delays->swapin_delay); - spin_unlock(&tsk->delays->lock); + spin_unlock_irqrestore(&tsk->delays->lock, flags); return ret; } -- cgit v0.10.2 From 0e009be8a0c2309f3696df70f72ef0075aa34c9c Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 5 Nov 2006 23:52:11 -0800 Subject: [PATCH] Improve the removed sysctl warnings Don't warn about libpthread's access to kernel.version. When it receives -ENOSYS it will read /proc/sys/kernel/version. If anything else shows up print the sysctl number string. Signed-off-by: Eric W. Biederman Cc: Cal Peake Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 8bff2c1..0c8e805 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2680,13 +2680,33 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen, asmlinkage long sys_sysctl(struct __sysctl_args __user *args) { static int msg_count; + struct __sysctl_args tmp; + int name[CTL_MAXNAME]; + int i; + + /* Read in the sysctl name for better debug message logging */ + if (copy_from_user(&tmp, args, sizeof(tmp))) + return -EFAULT; + if (tmp.nlen <= 0 || tmp.nlen >= CTL_MAXNAME) + return -ENOTDIR; + for (i = 0; i < tmp.nlen; i++) + if (get_user(name[i], tmp.name + i)) + return -EFAULT; + + /* Ignore accesses to kernel.version */ + if ((tmp.nlen == 2) && (name[0] == CTL_KERN) && (name[1] == KERN_VERSION)) + goto out; if (msg_count < 5) { msg_count++; printk(KERN_INFO "warning: process `%s' used the removed sysctl " - "system call\n", current->comm); + "system call with ", current->comm); + for (i = 0; i < tmp.nlen; i++) + printk("%d.", name[i]); + printk("\n"); } +out: return -ENOSYS; } -- cgit v0.10.2 From d99f160ac53e51090f015a8f0617cea25f81a191 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 5 Nov 2006 23:52:12 -0800 Subject: [PATCH] sysctl: allow a zero ctl_name in the middle of a sysctl table Since it is becoming clear that there are just enough users of the binary sysctl interface that completely removing the binary interface from the kernel will not be an option for foreseeable future, we need to find a way to address the sysctl maintenance issues. The basic problem is that sysctl requires one central authority to allocate sysctl numbers, or else conflicts and ABI breakage occur. The proc interface to sysctl does not have that problem, as names are not densely allocated. By not terminating a sysctl table until I have neither a ctl_name nor a procname, it becomes simple to add sysctl entries that don't show up in the binary sysctl interface. Which allows people to avoid allocating a binary sysctl value when not needed. I have audited the kernel code and in my reading I have not found a single sysctl table that wasn't terminated by a completely zero filled entry. So this change in behavior should not affect anything. I think this mechanism eases the pain enough that combined with a little disciple we can solve the reoccurring sysctl ABI breakage. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 1b24bd4..c184732 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -961,8 +961,8 @@ extern ctl_handler sysctl_ms_jiffies; /* * Register a set of sysctl names by calling register_sysctl_table * with an initialised array of ctl_table's. An entry with zero - * ctl_name terminates the table. table->de will be set up by the - * registration and need not be initialised in advance. + * ctl_name and NULL procname terminates the table. table->de will be + * set up by the registration and need not be initialised in advance. * * sysctl names can be mirrored automatically under /proc/sys. The * procname supplied controls /proc naming. @@ -973,7 +973,10 @@ extern ctl_handler sysctl_ms_jiffies; * Leaf nodes in the sysctl tree will be represented by a single file * under /proc; non-leaf nodes will be represented by directories. A * null procname disables /proc mirroring at this node. - * + * + * sysctl entries with a zero ctl_name will not be available through + * the binary sysctl interface. + * * sysctl(2) can automatically manage read and write requests through * the sysctl table. The data and maxlen fields of the ctl_table * struct enable minimal validation of the values being written to be diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 0c8e805..09e569f4 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1315,7 +1315,9 @@ repeat: return -ENOTDIR; if (get_user(n, name)) return -EFAULT; - for ( ; table->ctl_name; table++) { + for ( ; table->ctl_name || table->procname; table++) { + if (!table->ctl_name) + continue; if (n == table->ctl_name || table->ctl_name == CTL_ANY) { int error; if (table->child) { @@ -1532,7 +1534,7 @@ static void register_proc_table(ctl_table * table, struct proc_dir_entry *root, int len; mode_t mode; - for (; table->ctl_name; table++) { + for (; table->ctl_name || table->procname; table++) { /* Can't do anything without a proc name. */ if (!table->procname) continue; @@ -1579,7 +1581,7 @@ static void register_proc_table(ctl_table * table, struct proc_dir_entry *root, static void unregister_proc_table(ctl_table * table, struct proc_dir_entry *root) { struct proc_dir_entry *de; - for (; table->ctl_name; table++) { + for (; table->ctl_name || table->procname; table++) { if (!(de = table->de)) continue; if (de->mode & S_IFDIR) { -- cgit v0.10.2 From 7cc13edc139108bb527b692f0548dce6bc648572 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 5 Nov 2006 23:52:13 -0800 Subject: [PATCH] sysctl: implement CTL_UNNUMBERED This patch takes the CTL_UNNUMBERD concept from NFS and makes it available to all new sysctl users. At the same time the sysctl binary interface maintenance documentation is updated to mention and to describe what is needed to successfully maintain the sysctl binary interface. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 6341392..8ca1808 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -353,9 +353,6 @@ EXPORT_SYMBOL(lockd_down); * Sysctl parameters (same as module parameters, different interface). */ -/* Something that isn't CTL_ANY, CTL_NONE or a value that may clash. */ -#define CTL_UNNUMBERED -2 - static ctl_table nlm_sysctls[] = { { .ctl_name = CTL_UNNUMBERED, diff --git a/fs/nfs/sysctl.c b/fs/nfs/sysctl.c index 2fe3403..3ea50ac 100644 --- a/fs/nfs/sysctl.c +++ b/fs/nfs/sysctl.c @@ -18,11 +18,6 @@ static const int nfs_set_port_min = 0; static const int nfs_set_port_max = 65535; static struct ctl_table_header *nfs_callback_sysctl_table; -/* - * Something that isn't CTL_ANY, CTL_NONE or a value that may clash. - * Use the same values as fs/lockd/svc.c - */ -#define CTL_UNNUMBERED -2 static ctl_table nfs_cb_sysctls[] = { #ifdef CONFIG_NFS_V4 diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index c184732..d98562f 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -6,10 +6,17 @@ **************************************************************** **************************************************************** ** + ** WARNING: ** The values in this file are exported to user space via - ** the sysctl() binary interface. However this interface - ** is unstable and deprecated and will be removed in the future. - ** For a stable interface use /proc/sys. + ** the sysctl() binary interface. Do *NOT* change the + ** numbering of any existing values here, and do not change + ** any numbers within any one set of values. If you have to + ** have to redefine an existing interface, use a new number for it. + ** The kernel will then return -ENOTDIR to any application using + ** the old binary interface. + ** + ** For new interfaces unless you really need a binary number + ** please use CTL_UNNUMBERED. ** **************************************************************** **************************************************************** @@ -48,6 +55,7 @@ struct __sysctl_args { #ifdef __KERNEL__ #define CTL_ANY -1 /* Matches any name */ #define CTL_NONE 0 +#define CTL_UNNUMBERED CTL_NONE /* sysctl without a binary number */ #endif enum -- cgit v0.10.2 From 0c7bb31db0e35d4b772fac452b722460ca368acf Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 5 Nov 2006 23:52:13 -0800 Subject: [PATCH] sunrpc: add missing spin_unlock auth_domain_put() forgot to unlock acquired spinlock. Cc: Olaf Kirch Cc: Andy Adamson Cc: J. Bruce Fields Acked-by: Trond Myklebust Signed-off-by: Akinobu Mita Cc: Neil Brown Acked-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index 0004c1f..ee9bb15 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -126,6 +126,7 @@ void auth_domain_put(struct auth_domain *dom) if (atomic_dec_and_lock(&dom->ref.refcount, &auth_domain_lock)) { hlist_del(&dom->hash); dom->flavour->domain_release(dom); + spin_unlock(&auth_domain_lock); } } -- cgit v0.10.2 From bcc8bcb1f0cc51c0042497d5de2d79743050e3bb Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 6 Nov 2006 10:49:00 +0100 Subject: [S390] revert add_active_range() usage patch. Commit 7676bef9c183fd573822cac9992927ef596d584c breaks DCSS support on s390. DCSS needs initialized struct pages to work. With the usage of add_active_range() only the struct pages for physically present pages are initialized. This could be fixed if the DCSS driver would initiliaze the struct pages itself, but this doesn't work too. This is because the mem_map array does not include holes after the last present memory area and therefore there is nothing that could be initialized. To fix this and to avoid some dirty hacks revert this patch for now. Will be added later when we move to a virtual mem_map. Cc: Carsten Otte Cc: Adrian Bunk Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 608193c..245b81b 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -236,9 +236,6 @@ config WARN_STACK_SIZE This allows you to specify the maximum frame size a function may have without the compiler complaining about it. -config ARCH_POPULATES_NODE_MAP - def_bool y - source "mm/Kconfig" comment "I/O subsystem configuration" diff --git a/arch/s390/defconfig b/arch/s390/defconfig index c313e9a..7cd51e7 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -119,7 +119,6 @@ CONFIG_PACK_STACK=y CONFIG_CHECK_STACK=y CONFIG_STACK_GUARD=256 # CONFIG_WARN_STACK is not set -CONFIG_ARCH_POPULATES_NODE_MAP=y CONFIG_SELECT_MEMORY_MODEL=y CONFIG_FLATMEM_MANUAL=y # CONFIG_DISCONTIGMEM_MANUAL is not set diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 49f2b68..a31abdd 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -70,6 +70,7 @@ struct { #define CHUNK_READ_WRITE 0 #define CHUNK_READ_ONLY 1 volatile int __cpu_logical_map[NR_CPUS]; /* logical cpu to cpu address */ +unsigned long __initdata zholes_size[MAX_NR_ZONES]; static unsigned long __initdata memory_end; /* @@ -357,6 +358,21 @@ void machine_power_off(void) */ void (*pm_power_off)(void) = machine_power_off; +static void __init +add_memory_hole(unsigned long start, unsigned long end) +{ + unsigned long dma_pfn = MAX_DMA_ADDRESS >> PAGE_SHIFT; + + if (end <= dma_pfn) + zholes_size[ZONE_DMA] += end - start + 1; + else if (start > dma_pfn) + zholes_size[ZONE_NORMAL] += end - start + 1; + else { + zholes_size[ZONE_DMA] += dma_pfn - start + 1; + zholes_size[ZONE_NORMAL] += end - dma_pfn; + } +} + static int __init early_parse_mem(char *p) { memory_end = memparse(p, &p); @@ -478,6 +494,7 @@ setup_memory(void) { unsigned long bootmap_size; unsigned long start_pfn, end_pfn, init_pfn; + unsigned long last_rw_end; int i; /* @@ -533,27 +550,39 @@ setup_memory(void) /* * Register RAM areas with the bootmem allocator. */ + last_rw_end = start_pfn; for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { - unsigned long start_chunk, end_chunk, pfn; + unsigned long start_chunk, end_chunk; if (memory_chunk[i].type != CHUNK_READ_WRITE) continue; - start_chunk = PFN_DOWN(memory_chunk[i].addr); - end_chunk = start_chunk + PFN_DOWN(memory_chunk[i].size) - 1; - end_chunk = min(end_chunk, end_pfn); - if (start_chunk >= end_chunk) - continue; - add_active_range(0, start_chunk, end_chunk); - pfn = max(start_chunk, start_pfn); - for (; pfn <= end_chunk; pfn++) - page_set_storage_key(PFN_PHYS(pfn), PAGE_DEFAULT_KEY); + start_chunk = (memory_chunk[i].addr + PAGE_SIZE - 1); + start_chunk >>= PAGE_SHIFT; + end_chunk = (memory_chunk[i].addr + memory_chunk[i].size); + end_chunk >>= PAGE_SHIFT; + if (start_chunk < start_pfn) + start_chunk = start_pfn; + if (end_chunk > end_pfn) + end_chunk = end_pfn; + if (start_chunk < end_chunk) { + /* Initialize storage key for RAM pages */ + for (init_pfn = start_chunk ; init_pfn < end_chunk; + init_pfn++) + page_set_storage_key(init_pfn << PAGE_SHIFT, + PAGE_DEFAULT_KEY); + free_bootmem(start_chunk << PAGE_SHIFT, + (end_chunk - start_chunk) << PAGE_SHIFT); + if (last_rw_end < start_chunk) + add_memory_hole(last_rw_end, start_chunk - 1); + last_rw_end = end_chunk; + } } psw_set_key(PAGE_DEFAULT_KEY); - free_bootmem_with_active_regions(0, max_pfn); - reserve_bootmem(0, PFN_PHYS(start_pfn)); + if (last_rw_end < end_pfn - 1) + add_memory_hole(last_rw_end, end_pfn - 1); /* * Reserve the bootmem bitmap itself as well. We do this in two diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index d998917..e1881c3 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -84,6 +84,7 @@ void show_mem(void) printk("%d pages swap cached\n",cached); } +extern unsigned long __initdata zholes_size[]; /* * paging_init() sets up the page tables */ @@ -100,15 +101,16 @@ void __init paging_init(void) unsigned long pgdir_k = (__pa(swapper_pg_dir) & PAGE_MASK) | _KERNSEG_TABLE; static const int ssm_mask = 0x04000000L; unsigned long ro_start_pfn, ro_end_pfn; - unsigned long max_zone_pfns[MAX_NR_ZONES]; + unsigned long zones_size[MAX_NR_ZONES]; ro_start_pfn = PFN_DOWN((unsigned long)&__start_rodata); ro_end_pfn = PFN_UP((unsigned long)&__end_rodata); - memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); - max_zone_pfns[ZONE_DMA] = max_low_pfn; - max_zone_pfns[ZONE_NORMAL] = max_low_pfn; - free_area_init_nodes(max_zone_pfns); + memset(zones_size, 0, sizeof(zones_size)); + zones_size[ZONE_DMA] = max_low_pfn; + free_area_init_node(0, &contig_page_data, zones_size, + __pa(PAGE_OFFSET) >> PAGE_SHIFT, + zholes_size); /* unmap whole virtual address space */ @@ -168,16 +170,26 @@ void __init paging_init(void) unsigned long pgdir_k = (__pa(swapper_pg_dir) & PAGE_MASK) | _KERN_REGION_TABLE; static const int ssm_mask = 0x04000000L; + unsigned long zones_size[MAX_NR_ZONES]; + unsigned long dma_pfn, high_pfn; unsigned long ro_start_pfn, ro_end_pfn; - unsigned long max_zone_pfns[MAX_NR_ZONES]; + memset(zones_size, 0, sizeof(zones_size)); + dma_pfn = MAX_DMA_ADDRESS >> PAGE_SHIFT; + high_pfn = max_low_pfn; ro_start_pfn = PFN_DOWN((unsigned long)&__start_rodata); ro_end_pfn = PFN_UP((unsigned long)&__end_rodata); - memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); - max_zone_pfns[ZONE_DMA] = PFN_DOWN(MAX_DMA_ADDRESS); - max_zone_pfns[ZONE_NORMAL] = max_low_pfn; - free_area_init_nodes(max_zone_pfns); + if (dma_pfn > high_pfn) + zones_size[ZONE_DMA] = high_pfn; + else { + zones_size[ZONE_DMA] = dma_pfn; + zones_size[ZONE_NORMAL] = high_pfn - dma_pfn; + } + + /* Initialize mem_map[]. */ + free_area_init_node(0, &contig_page_data, zones_size, + __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size); /* * map whole physical memory to virtual memory (identity mapping) -- cgit v0.10.2 From c4972f33169e3e1ce06ad00917ddaf2c3b49b36a Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 6 Nov 2006 10:49:02 +0100 Subject: [S390] IRQs too early enabled. setup_lowcore() calls ctl_set_bit() which returns withs interrupts enabled. The setup arch code is not supposed to enable interrupts that early. Therefore use the __ctl_set_bit() variant. This fixes the not working lock dependency validator on non 64 bit systems. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index a31abdd..2aa13e8 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -450,7 +450,7 @@ setup_lowcore(void) lc->extended_save_area_addr = (__u32) __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0); /* enable extended save area */ - ctl_set_bit(14, 29); + __ctl_set_bit(14, 29); } #endif set_prefix((u32)(unsigned long) lc); -- cgit v0.10.2 From bbfd2bf9028a55cb85ff51eb34dc663b7e49ed55 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Mon, 6 Nov 2006 14:02:44 +0100 Subject: AVR32: Wire up sys_epoll_pwait Signed-off-by: Haavard Skinnemoen diff --git a/arch/avr32/kernel/syscall-stubs.S b/arch/avr32/kernel/syscall-stubs.S index 7589a9b..890286a 100644 --- a/arch/avr32/kernel/syscall-stubs.S +++ b/arch/avr32/kernel/syscall-stubs.S @@ -100,3 +100,12 @@ __sys_splice: rcall sys_splice sub sp, -4 popm pc + + .global __sys_epoll_pwait + .type __sys_epoll_pwait,@function +__sys_epoll_pwait: + pushm lr + st.w --sp, ARG6 + rcall sys_epoll_pwait + sub sp, -4 + popm pc diff --git a/arch/avr32/kernel/syscall_table.S b/arch/avr32/kernel/syscall_table.S index 63b2069..db8f8b5 100644 --- a/arch/avr32/kernel/syscall_table.S +++ b/arch/avr32/kernel/syscall_table.S @@ -286,4 +286,5 @@ sys_call_table: .long sys_sync_file_range .long sys_tee .long sys_vmsplice + .long __sys_epoll_pwait /* 265 */ .long sys_ni_syscall /* r8 is saturated at nr_syscalls */ diff --git a/include/asm-avr32/unistd.h b/include/asm-avr32/unistd.h index a50e500..56ed1f9 100644 --- a/include/asm-avr32/unistd.h +++ b/include/asm-avr32/unistd.h @@ -280,9 +280,10 @@ #define __NR_sync_file_range 262 #define __NR_tee 263 #define __NR_vmsplice 264 +#define __NR_epoll_pwait 265 #ifdef __KERNEL__ -#define NR_syscalls 265 +#define NR_syscalls 266 #define __ARCH_WANT_IPC_PARSE_VERSION -- cgit v0.10.2 From 6af0f5f83169ef215b4579834ddfb2691e8a7bff Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Mon, 6 Nov 2006 14:06:45 +0100 Subject: AVR32: Add missing return instruction in __raw_writesb __raw_writesb ends with a conditional branch, which is obviously wrong. It should return after the last loop terminates. Signed-off-by: Haavard Skinnemoen diff --git a/arch/avr32/lib/io-readsb.S b/arch/avr32/lib/io-readsb.S index b319d5e..2be5da7 100644 --- a/arch/avr32/lib/io-readsb.S +++ b/arch/avr32/lib/io-readsb.S @@ -45,3 +45,5 @@ __raw_readsb: sub r10, 1 st.b r11++, r8 brne 3b + + retal r12 -- cgit v0.10.2 From eb1dc33aa235b0e44ada6716cda385883c6e6bff Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 28 Oct 2006 03:03:48 +0400 Subject: [GFS2] don't panic needlessly First, SLAB_PANIC is unjustified. Second, all error propagating and backing out is in place. Signed-off-by: Alexey Dobriyan Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 21508a1..9889c1e 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -84,8 +84,8 @@ static int __init init_gfs2_fs(void) gfs2_inode_cachep = kmem_cache_create("gfs2_inode", sizeof(struct gfs2_inode), - 0, (SLAB_RECLAIM_ACCOUNT| - SLAB_PANIC|SLAB_MEM_SPREAD), + 0, SLAB_RECLAIM_ACCOUNT| + SLAB_MEM_SPREAD, gfs2_init_inode_once, NULL); if (!gfs2_inode_cachep) goto fail; -- cgit v0.10.2 From 4a221953ed121692aa25998451a57c7f4be8b4f6 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 09:57:57 -0500 Subject: [GFS2] Fix incorrect fs sync behaviour. This adds a sync_fs superblock operation for GFS2 and removes the journal flush from write_super in favour of sync_fs where it ought to be. This is more or less identical to the way in which ext3 does this. This bug was pointed out by Russell Cattelan Cc: Russell Cattelan Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index 06f06f7..b47d959 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -138,16 +138,27 @@ static void gfs2_put_super(struct super_block *sb) } /** - * gfs2_write_super - disk commit all incore transactions - * @sb: the filesystem + * gfs2_write_super + * @sb: the superblock * - * This function is called every time sync(2) is called. - * After this exits, all dirty buffers are synced. */ static void gfs2_write_super(struct super_block *sb) { + sb->s_dirt = 0; +} + +/** + * gfs2_sync_fs - sync the filesystem + * @sb: the superblock + * + * Flushes the log to disk. + */ +static int gfs2_sync_fs(struct super_block *sb, int wait) +{ + sb->s_dirt = 0; gfs2_log_flush(sb->s_fs_info, NULL); + return 0; } /** @@ -452,17 +463,18 @@ static void gfs2_destroy_inode(struct inode *inode) } struct super_operations gfs2_super_ops = { - .alloc_inode = gfs2_alloc_inode, - .destroy_inode = gfs2_destroy_inode, - .write_inode = gfs2_write_inode, - .delete_inode = gfs2_delete_inode, - .put_super = gfs2_put_super, - .write_super = gfs2_write_super, - .write_super_lockfs = gfs2_write_super_lockfs, - .unlockfs = gfs2_unlockfs, - .statfs = gfs2_statfs, - .remount_fs = gfs2_remount_fs, - .clear_inode = gfs2_clear_inode, - .show_options = gfs2_show_options, + .alloc_inode = gfs2_alloc_inode, + .destroy_inode = gfs2_destroy_inode, + .write_inode = gfs2_write_inode, + .delete_inode = gfs2_delete_inode, + .put_super = gfs2_put_super, + .write_super = gfs2_write_super, + .sync_fs = gfs2_sync_fs, + .write_super_lockfs = gfs2_write_super_lockfs, + .unlockfs = gfs2_unlockfs, + .statfs = gfs2_statfs, + .remount_fs = gfs2_remount_fs, + .clear_inode = gfs2_clear_inode, + .show_options = gfs2_show_options, }; -- cgit v0.10.2 From 26d83dedf61d26d85f10bc34b92f4de7660fd746 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Mon, 30 Oct 2006 16:59:08 -0500 Subject: [GFS2] Fix OOM error handling Fix the OOM error handling in inode.c where it was possible for a NULL pointer to be dereferenced. Signed-off-by: Steven Whitehouse diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 57c43ac..d470e52 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -157,6 +157,9 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum *inum, struct gfs2_glock *io_gl; int error; + if (!inode) + return ERR_PTR(-ENOBUFS); + if (inode->i_state & I_NEW) { struct gfs2_sbd *sdp = GFS2_SB(inode); umode_t mode = DT2IF(type); -- cgit v0.10.2 From ba542e3b92f9ea7c482ae56b68b9122eebc53a39 Mon Sep 17 00:00:00 2001 From: Patrick Caulfield Date: Thu, 2 Nov 2006 14:41:23 +0000 Subject: [DLM] Fix kref_put oops This patch fixes the recounting on the lockspace kobject. Previously the lockspace was freed while userspace could have had a reference to one of its sysfs files, causing an oops in kref_put. Now the lockspace kfree is moved into the kobject release() function Signed-Off-By: Patrick Caulfield Signed-off-by: Steven Whitehouse diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 109333c..499ee11 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -143,6 +143,12 @@ static ssize_t dlm_attr_store(struct kobject *kobj, struct attribute *attr, return a->store ? a->store(ls, buf, len) : len; } +static void lockspace_kobj_release(struct kobject *k) +{ + struct dlm_ls *ls = container_of(k, struct dlm_ls, ls_kobj); + kfree(ls); +} + static struct sysfs_ops dlm_attr_ops = { .show = dlm_attr_show, .store = dlm_attr_store, @@ -151,6 +157,7 @@ static struct sysfs_ops dlm_attr_ops = { static struct kobj_type dlm_ktype = { .default_attrs = dlm_attrs, .sysfs_ops = &dlm_attr_ops, + .release = lockspace_kobj_release, }; static struct kset dlm_kset = { @@ -678,7 +685,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) dlm_clear_members_gone(ls); kfree(ls->ls_node_array); kobject_unregister(&ls->ls_kobj); - kfree(ls); + /* The ls structure will be freed when the kobject is done with */ mutex_lock(&ls_lock); ls_count--; -- cgit v0.10.2 From e2de7f565521a76fbbb927f701c5a1d381c71a93 Mon Sep 17 00:00:00 2001 From: Patrick Caulfield Date: Mon, 6 Nov 2006 08:53:28 +0000 Subject: [DLM] fix oops in kref_put when removing a lockspace Now that the lockspace struct is freed when the last sysfs object is released this patch prevents use of that lockspace by sysfs. We attempt to re-get the lockspace from the lockspace list and fail the request if it has been removed. Signed-Off-By: Patrick Caulfield Signed-off-by: Steven Whitehouse diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 499ee11..f8842ca 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -43,6 +43,10 @@ static ssize_t dlm_control_store(struct dlm_ls *ls, const char *buf, size_t len) ssize_t ret = len; int n = simple_strtol(buf, NULL, 0); + ls = dlm_find_lockspace_local(ls->ls_local_handle); + if (!ls) + return -EINVAL; + switch (n) { case 0: dlm_ls_stop(ls); @@ -53,6 +57,7 @@ static ssize_t dlm_control_store(struct dlm_ls *ls, const char *buf, size_t len) default: ret = -EINVAL; } + dlm_put_lockspace(ls); return ret; } -- cgit v0.10.2 From cca72333e71e348995859b88628c1abcb58b759e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 3 Nov 2006 17:42:47 +0000 Subject: [MIPS] Ocelot C: Fix large number of warnings. Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_c/ocelot_c_fpga.h b/arch/mips/momentum/ocelot_c/ocelot_c_fpga.h index 7228cd1..f0f5581 100644 --- a/arch/mips/momentum/ocelot_c/ocelot_c_fpga.h +++ b/arch/mips/momentum/ocelot_c/ocelot_c_fpga.h @@ -53,7 +53,9 @@ #define OCELOT_C_REG_INTSET 0xe #define OCELOT_C_REG_INTCLR 0xf -#define OCELOT_FPGA_WRITE(x, y) writeb(x, OCELOT_C_CS0_ADDR + OCELOT_C_REG_##y) -#define OCELOT_FPGA_READ(x) readb(OCELOT_C_CS0_ADDR + OCELOT_C_REG_##x) +#define __FPGA_REG_TO_ADDR(reg) \ + ((void *) OCELOT_C_CS0_ADDR + OCELOT_C_REG_##reg) +#define OCELOT_FPGA_WRITE(x, reg) writeb(x, __FPGA_REG_TO_ADDR(reg)) +#define OCELOT_FPGA_READ(reg) readb(__FPGA_REG_TO_ADDR(reg)) #endif -- cgit v0.10.2 From 9c422e2ad6a1d8bd03e2e9d49e5c63b82165d596 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 3 Nov 2006 17:45:25 +0000 Subject: [MIPS] Ocelot C: fix eth registration after conversion to platform_device Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_c/setup.c b/arch/mips/momentum/ocelot_c/setup.c index 9c0c462..0a3ffa3 100644 --- a/arch/mips/momentum/ocelot_c/setup.c +++ b/arch/mips/momentum/ocelot_c/setup.c @@ -50,6 +50,7 @@ #include #include #include +#include #include #include #include @@ -69,7 +70,6 @@ #include "ocelot_c_fpga.h" unsigned long marvell_base; -extern unsigned long mv64340_sram_base; unsigned long cpu_clock; /* These functions are used for rebooting or halting the machine*/ @@ -119,7 +119,6 @@ void PMON_v2_setup(void) add_wired_entry(ENTRYLO(0xfe000000), ENTRYLO(0xff000000), 0xfffffffffe000000, PM_16M); marvell_base = 0xfffffffff4000000; - mv64340_sram_base = 0xfffffffffe000000; #else /* marvell and extra space */ add_wired_entry(ENTRYLO(0xf4000000), ENTRYLO(0xf4010000), 0xf4000000, PM_64K); @@ -129,7 +128,6 @@ void PMON_v2_setup(void) add_wired_entry(ENTRYLO(0xfe000000), ENTRYLO(0xff000000), 0xfe000000, PM_16M); marvell_base = 0xf4000000; - mv64340_sram_base = 0xfe000000; #endif } @@ -365,3 +363,123 @@ static int io_base_ioremap(void) module_init(io_base_ioremap); #endif + +#if defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) + +static struct resource mv643xx_eth_shared_resources[] = { + [0] = { + .name = "ethernet shared base", + .start = 0xf1000000 + MV643XX_ETH_SHARED_REGS, + .end = 0xf1000000 + MV643XX_ETH_SHARED_REGS + + MV643XX_ETH_SHARED_REGS_SIZE - 1, + .flags = IORESOURCE_MEM, + }, +}; + +static struct platform_device mv643xx_eth_shared_device = { + .name = MV643XX_ETH_SHARED_NAME, + .id = 0, + .num_resources = ARRAY_SIZE(mv643xx_eth_shared_resources), + .resource = mv643xx_eth_shared_resources, +}; + +#define MV_SRAM_BASE 0xfe000000UL +#define MV_SRAM_SIZE (256 * 1024) + +#define MV_SRAM_RXRING_SIZE (MV_SRAM_SIZE / 4) +#define MV_SRAM_TXRING_SIZE (MV_SRAM_SIZE / 4) + +#define MV_SRAM_BASE_ETH0 MV_SRAM_BASE +#define MV_SRAM_BASE_ETH1 (MV_SRAM_BASE + (MV_SRAM_SIZE / 2)) + +#define MV64x60_IRQ_ETH_0 48 +#define MV64x60_IRQ_ETH_1 49 + +#ifdef CONFIG_MV643XX_ETH_0 + +static struct resource mv64x60_eth0_resources[] = { + [0] = { + .name = "eth0 irq", + .start = MV64x60_IRQ_ETH_0, + .end = MV64x60_IRQ_ETH_0, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct mv643xx_eth_platform_data eth0_pd = { + .tx_sram_addr = MV_SRAM_BASE_ETH0, + .tx_sram_size = MV_SRAM_TXRING_SIZE, + .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, + + .rx_sram_addr = MV_SRAM_BASE_ETH0 + MV_SRAM_TXRING_SIZE, + .rx_sram_size = MV_SRAM_RXRING_SIZE, + .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, +}; + +static struct platform_device eth0_device = { + .name = MV643XX_ETH_NAME, + .id = 0, + .num_resources = ARRAY_SIZE(mv64x60_eth0_resources), + .resource = mv64x60_eth0_resources, + .dev = { + .platform_data = ð0_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_0 */ + +#ifdef CONFIG_MV643XX_ETH_1 + +static struct resource mv64x60_eth1_resources[] = { + [0] = { + .name = "eth1 irq", + .start = MV64x60_IRQ_ETH_1, + .end = MV64x60_IRQ_ETH_1, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct mv643xx_eth_platform_data eth1_pd = { + .tx_sram_addr = MV_SRAM_BASE_ETH1, + .tx_sram_size = MV_SRAM_TXRING_SIZE, + .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, + + .rx_sram_addr = MV_SRAM_BASE_ETH1 + MV_SRAM_TXRING_SIZE, + .rx_sram_size = MV_SRAM_RXRING_SIZE, + .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, +}; + +static struct platform_device eth1_device = { + .name = MV643XX_ETH_NAME, + .id = 1, + .num_resources = ARRAY_SIZE(mv64x60_eth1_resources), + .resource = mv64x60_eth1_resources, + .dev = { + .platform_data = ð1_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_1 */ + +static struct platform_device *mv643xx_eth_pd_devs[] __initdata = { + &mv643xx_eth_shared_device, +#ifdef CONFIG_MV643XX_ETH_0 + ð0_device, +#endif +#ifdef CONFIG_MV643XX_ETH_1 + ð1_device, +#endif + /* The third port is not wired up on the Ocelot C */ +}; + +int mv643xx_eth_add_pds(void) +{ + int ret; + + ret = platform_add_devices(mv643xx_eth_pd_devs, + ARRAY_SIZE(mv643xx_eth_pd_devs)); + + return ret; +} + +device_initcall(mv643xx_eth_add_pds); + +#endif /* defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) */ -- cgit v0.10.2 From 2002d2bde1a9ad10e2521b8b117c11abfbc2ee93 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 3 Nov 2006 17:48:17 +0000 Subject: [MIPS] Ocelot C: Fix warning about missmatching format string. CC arch/mips/momentum/ocelot_c/setup.o arch/mips/momentum/ocelot_c/setup.c: In function 'momenco_time_init': arch/mips/momentum/ocelot_c/setup.c:223: warning: format '%d' expects type 'int', but argument 2 has type 'long unsigned int' Change data type to match format string; a 32-bit type better suits our needs. Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_c/prom.c b/arch/mips/momentum/ocelot_c/prom.c index 4c50a14..1f67728 100644 --- a/arch/mips/momentum/ocelot_c/prom.c +++ b/arch/mips/momentum/ocelot_c/prom.c @@ -29,7 +29,7 @@ struct callvectors* debug_vectors; extern unsigned long marvell_base; -extern unsigned long cpu_clock; +extern unsigned int cpu_clock; #ifdef CONFIG_MV643XX_ETH extern unsigned char prom_mac_addr_base[6]; diff --git a/arch/mips/momentum/ocelot_c/setup.c b/arch/mips/momentum/ocelot_c/setup.c index 0a3ffa3..7832d4e 100644 --- a/arch/mips/momentum/ocelot_c/setup.c +++ b/arch/mips/momentum/ocelot_c/setup.c @@ -70,7 +70,7 @@ #include "ocelot_c_fpga.h" unsigned long marvell_base; -unsigned long cpu_clock; +unsigned int cpu_clock; /* These functions are used for rebooting or halting the machine*/ extern void momenco_ocelot_restart(char *command); -- cgit v0.10.2 From ad0b365573718a4a83266f98c9a49305c8eaf0b8 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 3 Nov 2006 18:06:33 +0000 Subject: [MIPS] Ocelot C: Fix mapping of ioport address range. o Fix warnings o 768MB worth of I/O ports were insane o 64-bit kernels don't need special handling because ioremap does the magic Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_c/setup.c b/arch/mips/momentum/ocelot_c/setup.c index 7832d4e..b0e82a0 100644 --- a/arch/mips/momentum/ocelot_c/setup.c +++ b/arch/mips/momentum/ocelot_c/setup.c @@ -344,25 +344,23 @@ void __init plat_mem_setup(void) } } -#ifndef CONFIG_64BIT -/* This needs to be one of the first initcalls, because no I/O port access - can work before this */ +/* + * This needs to be one of the first initcalls, because no I/O port access + * can work before this + */ static int io_base_ioremap(void) { - /* we're mapping PCI accesses from 0xc0000000 to 0xf0000000 */ - void *io_remap_range = ioremap(0xc0000000, 0x30000000); + void __iomem * io_remap_range = ioremap(0xc0000000UL, 0x10000); - if (!io_remap_range) { + if (!io_remap_range) panic("Could not ioremap I/O port range"); - } - printk("io_remap_range set at 0x%08x\n", (uint32_t)io_remap_range); - set_io_port_base(io_remap_range - 0xc0000000); + + set_io_port_base((unsigned long) io_remap_range); return 0; } module_init(io_base_ioremap); -#endif #if defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) -- cgit v0.10.2 From d19f7befe929d400df89699eb51b8d7f4ef1b2d8 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 4 Nov 2006 13:02:46 +0000 Subject: [MIPS] Ocelot 3: Fix large number of warnings. Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_3/ocelot_3_fpga.h b/arch/mips/momentum/ocelot_3/ocelot_3_fpga.h index 227e429..5710a90 100644 --- a/arch/mips/momentum/ocelot_3/ocelot_3_fpga.h +++ b/arch/mips/momentum/ocelot_3/ocelot_3_fpga.h @@ -51,7 +51,9 @@ extern unsigned long ocelot_fpga_base; -#define OCELOT_FPGA_WRITE(x, y) writeb(x, ocelot_fpga_base + OCELOT_3_REG_##y) -#define OCELOT_FPGA_READ(x) readb(ocelot_fpga_base + OCELOT_3_REG_##x) +#define __FPGA_REG_TO_ADDR(reg) \ + ((void *) ocelot_fpga_base + OCELOT_3_REG_##reg) +#define OCELOT_FPGA_WRITE(x, reg) writeb(x, __FPGA_REG_TO_ADDR(reg)) +#define OCELOT_FPGA_READ(reg) readb(__FPGA_REG_TO_ADDR(reg)) #endif -- cgit v0.10.2 From d6b861c6402307e30c7df24dcda911df64a5f9d6 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 4 Nov 2006 23:26:27 +0000 Subject: [MIPS] SB1: On bootup only flush cache on local CPU. This fixes a warning on bootup warning in smp_call_function. Signed-off-by: Ralf Baechle diff --git a/arch/mips/mm/c-sb1.c b/arch/mips/mm/c-sb1.c index ea49a77..d0ddb4a 100644 --- a/arch/mips/mm/c-sb1.c +++ b/arch/mips/mm/c-sb1.c @@ -505,5 +505,5 @@ void sb1_cache_init(void) : : "memory"); - flush_cache_all(); + local_sb1___flush_cache_all(); } -- cgit v0.10.2 From 907c51b2d1705b022c3fb65b66cb4e5e09346433 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 5 Nov 2006 01:18:11 +0000 Subject: [MIPS] Ocelot C: Fix MAC address detection after platform_device conversion. Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_c/Makefile b/arch/mips/momentum/ocelot_c/Makefile index 94802b4..d69161a 100644 --- a/arch/mips/momentum/ocelot_c/Makefile +++ b/arch/mips/momentum/ocelot_c/Makefile @@ -2,7 +2,7 @@ # Makefile for Momentum Computer's Ocelot-C and -CS boards. # -obj-y += cpci-irq.o irq.o prom.o reset.o \ +obj-y += cpci-irq.o irq.o platform.o prom.o reset.o \ setup.o uart-irq.o obj-$(CONFIG_KGDB) += dbg_io.o diff --git a/arch/mips/momentum/ocelot_c/platform.c b/arch/mips/momentum/ocelot_c/platform.c new file mode 100644 index 0000000..6c495b2 --- /dev/null +++ b/arch/mips/momentum/ocelot_c/platform.c @@ -0,0 +1,201 @@ +#include +#include +#include +#include +#include + +#include "ocelot_c_fpga.h" + +#if defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) + +static struct resource mv643xx_eth_shared_resources[] = { + [0] = { + .name = "ethernet shared base", + .start = 0xf1000000 + MV643XX_ETH_SHARED_REGS, + .end = 0xf1000000 + MV643XX_ETH_SHARED_REGS + + MV643XX_ETH_SHARED_REGS_SIZE - 1, + .flags = IORESOURCE_MEM, + }, +}; + +static struct platform_device mv643xx_eth_shared_device = { + .name = MV643XX_ETH_SHARED_NAME, + .id = 0, + .num_resources = ARRAY_SIZE(mv643xx_eth_shared_resources), + .resource = mv643xx_eth_shared_resources, +}; + +#define MV_SRAM_BASE 0xfe000000UL +#define MV_SRAM_SIZE (256 * 1024) + +#define MV_SRAM_RXRING_SIZE (MV_SRAM_SIZE / 4) +#define MV_SRAM_TXRING_SIZE (MV_SRAM_SIZE / 4) + +#define MV_SRAM_BASE_ETH0 MV_SRAM_BASE +#define MV_SRAM_BASE_ETH1 (MV_SRAM_BASE + (MV_SRAM_SIZE / 2)) + +#define MV64x60_IRQ_ETH_0 48 +#define MV64x60_IRQ_ETH_1 49 + +#ifdef CONFIG_MV643XX_ETH_0 + +static struct resource mv64x60_eth0_resources[] = { + [0] = { + .name = "eth0 irq", + .start = MV64x60_IRQ_ETH_0, + .end = MV64x60_IRQ_ETH_0, + .flags = IORESOURCE_IRQ, + }, +}; + +static char eth0_mac_addr[ETH_ALEN]; + +static struct mv643xx_eth_platform_data eth0_pd = { + .mac_addr = eth0_mac_addr, + + .tx_sram_addr = MV_SRAM_BASE_ETH0, + .tx_sram_size = MV_SRAM_TXRING_SIZE, + .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, + + .rx_sram_addr = MV_SRAM_BASE_ETH0 + MV_SRAM_TXRING_SIZE, + .rx_sram_size = MV_SRAM_RXRING_SIZE, + .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, +}; + +static struct platform_device eth0_device = { + .name = MV643XX_ETH_NAME, + .id = 0, + .num_resources = ARRAY_SIZE(mv64x60_eth0_resources), + .resource = mv64x60_eth0_resources, + .dev = { + .platform_data = ð0_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_0 */ + +#ifdef CONFIG_MV643XX_ETH_1 + +static struct resource mv64x60_eth1_resources[] = { + [0] = { + .name = "eth1 irq", + .start = MV64x60_IRQ_ETH_1, + .end = MV64x60_IRQ_ETH_1, + .flags = IORESOURCE_IRQ, + }, +}; + +static char eth1_mac_addr[ETH_ALEN]; + +static struct mv643xx_eth_platform_data eth1_pd = { + .mac_addr = eth1_mac_addr, + + .tx_sram_addr = MV_SRAM_BASE_ETH1, + .tx_sram_size = MV_SRAM_TXRING_SIZE, + .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, + + .rx_sram_addr = MV_SRAM_BASE_ETH1 + MV_SRAM_TXRING_SIZE, + .rx_sram_size = MV_SRAM_RXRING_SIZE, + .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, +}; + +static struct platform_device eth1_device = { + .name = MV643XX_ETH_NAME, + .id = 1, + .num_resources = ARRAY_SIZE(mv64x60_eth1_resources), + .resource = mv64x60_eth1_resources, + .dev = { + .platform_data = ð1_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_1 */ + +static struct platform_device *mv643xx_eth_pd_devs[] __initdata = { + &mv643xx_eth_shared_device, +#ifdef CONFIG_MV643XX_ETH_0 + ð0_device, +#endif +#ifdef CONFIG_MV643XX_ETH_1 + ð1_device, +#endif + /* The third port is not wired up on the Ocelot C */ +}; + +static u8 __init exchange_bit(u8 val, u8 cs) +{ + /* place the data */ + OCELOT_FPGA_WRITE((val << 2) | cs, EEPROM_MODE); + udelay(1); + + /* turn the clock on */ + OCELOT_FPGA_WRITE((val << 2) | cs | 0x2, EEPROM_MODE); + udelay(1); + + /* turn the clock off and read-strobe */ + OCELOT_FPGA_WRITE((val << 2) | cs | 0x10, EEPROM_MODE); + + /* return the data */ + return (OCELOT_FPGA_READ(EEPROM_MODE) >> 3) & 0x1; +} + +static void __init get_mac(char dest[6]) +{ + u8 read_opcode[12] = {1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; + int i,j; + + for (i = 0; i < 12; i++) + exchange_bit(read_opcode[i], 1); + + for (j = 0; j < 6; j++) { + dest[j] = 0; + for (i = 0; i < 8; i++) { + dest[j] <<= 1; + dest[j] |= exchange_bit(0, 1); + } + } + + /* turn off CS */ + exchange_bit(0,0); +} + +/* + * Copy and increment ethernet MAC address by a small value. + * + * This is useful for systems where the only one MAC address is stored in + * non-volatile memory for multiple ports. + */ +static inline void eth_mac_add(unsigned char *dst, unsigned char *src, + unsigned int add) +{ + int i; + + BUG_ON(add >= 256); + + for (i = ETH_ALEN; i >= 0; i--) { + dst[i] = src[i] + add; + add = dst[i] < src[i]; /* compute carry */ + } + + WARN_ON(add); +} + +static int __init mv643xx_eth_add_pds(void) +{ + unsigned char mac[ETH_ALEN]; + int ret; + + get_mac(mac); +#ifdef CONFIG_MV643XX_ETH_0 + eth_mac_add(eth1_mac_addr, mac, 0); +#endif +#ifdef CONFIG_MV643XX_ETH_1 + eth_mac_add(eth1_mac_addr, mac, 1); +#endif + ret = platform_add_devices(mv643xx_eth_pd_devs, + ARRAY_SIZE(mv643xx_eth_pd_devs)); + + return ret; +} + +device_initcall(mv643xx_eth_add_pds); + +#endif /* defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) */ diff --git a/arch/mips/momentum/ocelot_c/prom.c b/arch/mips/momentum/ocelot_c/prom.c index 1f67728..d0b77e1 100644 --- a/arch/mips/momentum/ocelot_c/prom.c +++ b/arch/mips/momentum/ocelot_c/prom.c @@ -31,10 +31,6 @@ struct callvectors* debug_vectors; extern unsigned long marvell_base; extern unsigned int cpu_clock; -#ifdef CONFIG_MV643XX_ETH -extern unsigned char prom_mac_addr_base[6]; -#endif - const char *get_system_type(void) { #ifdef CONFIG_CPU_SR71000 @@ -44,55 +40,6 @@ const char *get_system_type(void) #endif } -#ifdef CONFIG_MV643XX_ETH -static void burn_clocks(void) -{ - int i; - - /* this loop should burn at least 1us -- this should be plenty */ - for (i = 0; i < 0x10000; i++) - ; -} - -static u8 exchange_bit(u8 val, u8 cs) -{ - /* place the data */ - OCELOT_FPGA_WRITE((val << 2) | cs, EEPROM_MODE); - burn_clocks(); - - /* turn the clock on */ - OCELOT_FPGA_WRITE((val << 2) | cs | 0x2, EEPROM_MODE); - burn_clocks(); - - /* turn the clock off and read-strobe */ - OCELOT_FPGA_WRITE((val << 2) | cs | 0x10, EEPROM_MODE); - - /* return the data */ - return ((OCELOT_FPGA_READ(EEPROM_MODE) >> 3) & 0x1); -} - -void get_mac(char dest[6]) -{ - u8 read_opcode[12] = {1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; - int i,j; - - for (i = 0; i < 12; i++) - exchange_bit(read_opcode[i], 1); - - for (j = 0; j < 6; j++) { - dest[j] = 0; - for (i = 0; i < 8; i++) { - dest[j] <<= 1; - dest[j] |= exchange_bit(0, 1); - } - } - - /* turn off CS */ - exchange_bit(0,0); -} -#endif - - #ifdef CONFIG_64BIT unsigned long signext(unsigned long addr) @@ -226,11 +173,6 @@ void __init prom_init(void) mips_machgroup = MACH_GROUP_MOMENCO; mips_machtype = MACH_MOMENCO_OCELOT_C; -#ifdef CONFIG_MV643XX_ETH - /* get the base MAC address for on-board ethernet ports */ - get_mac(prom_mac_addr_base); -#endif - #ifndef CONFIG_64BIT debug_vectors->printf("Booting Linux kernel...\n"); #endif diff --git a/arch/mips/momentum/ocelot_c/setup.c b/arch/mips/momentum/ocelot_c/setup.c index b0e82a0..0b6b233 100644 --- a/arch/mips/momentum/ocelot_c/setup.c +++ b/arch/mips/momentum/ocelot_c/setup.c @@ -50,7 +50,6 @@ #include #include #include -#include #include #include #include @@ -361,123 +360,3 @@ static int io_base_ioremap(void) } module_init(io_base_ioremap); - -#if defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) - -static struct resource mv643xx_eth_shared_resources[] = { - [0] = { - .name = "ethernet shared base", - .start = 0xf1000000 + MV643XX_ETH_SHARED_REGS, - .end = 0xf1000000 + MV643XX_ETH_SHARED_REGS + - MV643XX_ETH_SHARED_REGS_SIZE - 1, - .flags = IORESOURCE_MEM, - }, -}; - -static struct platform_device mv643xx_eth_shared_device = { - .name = MV643XX_ETH_SHARED_NAME, - .id = 0, - .num_resources = ARRAY_SIZE(mv643xx_eth_shared_resources), - .resource = mv643xx_eth_shared_resources, -}; - -#define MV_SRAM_BASE 0xfe000000UL -#define MV_SRAM_SIZE (256 * 1024) - -#define MV_SRAM_RXRING_SIZE (MV_SRAM_SIZE / 4) -#define MV_SRAM_TXRING_SIZE (MV_SRAM_SIZE / 4) - -#define MV_SRAM_BASE_ETH0 MV_SRAM_BASE -#define MV_SRAM_BASE_ETH1 (MV_SRAM_BASE + (MV_SRAM_SIZE / 2)) - -#define MV64x60_IRQ_ETH_0 48 -#define MV64x60_IRQ_ETH_1 49 - -#ifdef CONFIG_MV643XX_ETH_0 - -static struct resource mv64x60_eth0_resources[] = { - [0] = { - .name = "eth0 irq", - .start = MV64x60_IRQ_ETH_0, - .end = MV64x60_IRQ_ETH_0, - .flags = IORESOURCE_IRQ, - }, -}; - -static struct mv643xx_eth_platform_data eth0_pd = { - .tx_sram_addr = MV_SRAM_BASE_ETH0, - .tx_sram_size = MV_SRAM_TXRING_SIZE, - .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, - - .rx_sram_addr = MV_SRAM_BASE_ETH0 + MV_SRAM_TXRING_SIZE, - .rx_sram_size = MV_SRAM_RXRING_SIZE, - .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, -}; - -static struct platform_device eth0_device = { - .name = MV643XX_ETH_NAME, - .id = 0, - .num_resources = ARRAY_SIZE(mv64x60_eth0_resources), - .resource = mv64x60_eth0_resources, - .dev = { - .platform_data = ð0_pd, - }, -}; -#endif /* CONFIG_MV643XX_ETH_0 */ - -#ifdef CONFIG_MV643XX_ETH_1 - -static struct resource mv64x60_eth1_resources[] = { - [0] = { - .name = "eth1 irq", - .start = MV64x60_IRQ_ETH_1, - .end = MV64x60_IRQ_ETH_1, - .flags = IORESOURCE_IRQ, - }, -}; - -static struct mv643xx_eth_platform_data eth1_pd = { - .tx_sram_addr = MV_SRAM_BASE_ETH1, - .tx_sram_size = MV_SRAM_TXRING_SIZE, - .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, - - .rx_sram_addr = MV_SRAM_BASE_ETH1 + MV_SRAM_TXRING_SIZE, - .rx_sram_size = MV_SRAM_RXRING_SIZE, - .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, -}; - -static struct platform_device eth1_device = { - .name = MV643XX_ETH_NAME, - .id = 1, - .num_resources = ARRAY_SIZE(mv64x60_eth1_resources), - .resource = mv64x60_eth1_resources, - .dev = { - .platform_data = ð1_pd, - }, -}; -#endif /* CONFIG_MV643XX_ETH_1 */ - -static struct platform_device *mv643xx_eth_pd_devs[] __initdata = { - &mv643xx_eth_shared_device, -#ifdef CONFIG_MV643XX_ETH_0 - ð0_device, -#endif -#ifdef CONFIG_MV643XX_ETH_1 - ð1_device, -#endif - /* The third port is not wired up on the Ocelot C */ -}; - -int mv643xx_eth_add_pds(void) -{ - int ret; - - ret = platform_add_devices(mv643xx_eth_pd_devs, - ARRAY_SIZE(mv643xx_eth_pd_devs)); - - return ret; -} - -device_initcall(mv643xx_eth_add_pds); - -#endif /* defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) */ -- cgit v0.10.2 From ff28cbd2804105b144a7054e0302615e1da6749f Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sun, 5 Nov 2006 01:18:43 +0000 Subject: [MIPS] Ocelot 3: Fix MAC address detection after platform_device conversion. Signed-off-by: Ralf Baechle diff --git a/arch/mips/momentum/ocelot_3/Makefile b/arch/mips/momentum/ocelot_3/Makefile index 8bcea64d..d5a090a 100644 --- a/arch/mips/momentum/ocelot_3/Makefile +++ b/arch/mips/momentum/ocelot_3/Makefile @@ -5,4 +5,4 @@ # removes any old dependencies. DON'T put your own dependencies here # unless it's something special (ie not a .c file). # -obj-y += irq.o prom.o reset.o setup.o +obj-y += irq.o platform.o prom.o reset.o setup.o diff --git a/arch/mips/momentum/ocelot_3/platform.c b/arch/mips/momentum/ocelot_3/platform.c new file mode 100644 index 0000000..eefe584 --- /dev/null +++ b/arch/mips/momentum/ocelot_3/platform.c @@ -0,0 +1,235 @@ +#include +#include +#include +#include +#include + +#include "ocelot_3_fpga.h" + +#if defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) + +static struct resource mv643xx_eth_shared_resources[] = { + [0] = { + .name = "ethernet shared base", + .start = 0xf1000000 + MV643XX_ETH_SHARED_REGS, + .end = 0xf1000000 + MV643XX_ETH_SHARED_REGS + + MV643XX_ETH_SHARED_REGS_SIZE - 1, + .flags = IORESOURCE_MEM, + }, +}; + +static struct platform_device mv643xx_eth_shared_device = { + .name = MV643XX_ETH_SHARED_NAME, + .id = 0, + .num_resources = ARRAY_SIZE(mv643xx_eth_shared_resources), + .resource = mv643xx_eth_shared_resources, +}; + +#define MV_SRAM_BASE 0xfe000000UL +#define MV_SRAM_SIZE (256 * 1024) + +#define MV_SRAM_RXRING_SIZE (MV_SRAM_SIZE / 4) +#define MV_SRAM_TXRING_SIZE (MV_SRAM_SIZE / 4) + +#define MV_SRAM_BASE_ETH0 MV_SRAM_BASE +#define MV_SRAM_BASE_ETH1 (MV_SRAM_BASE + (MV_SRAM_SIZE / 2)) + +#define MV64x60_IRQ_ETH_0 48 +#define MV64x60_IRQ_ETH_1 49 +#define MV64x60_IRQ_ETH_2 50 + +#ifdef CONFIG_MV643XX_ETH_0 + +static struct resource mv64x60_eth0_resources[] = { + [0] = { + .name = "eth0 irq", + .start = MV64x60_IRQ_ETH_0, + .end = MV64x60_IRQ_ETH_0, + .flags = IORESOURCE_IRQ, + }, +}; + +static char eth0_mac_addr[ETH_ALEN]; + +static struct mv643xx_eth_platform_data eth0_pd = { + .mac_addr = eth0_mac_addr, + + .tx_sram_addr = MV_SRAM_BASE_ETH0, + .tx_sram_size = MV_SRAM_TXRING_SIZE, + .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, + + .rx_sram_addr = MV_SRAM_BASE_ETH0 + MV_SRAM_TXRING_SIZE, + .rx_sram_size = MV_SRAM_RXRING_SIZE, + .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, +}; + +static struct platform_device eth0_device = { + .name = MV643XX_ETH_NAME, + .id = 0, + .num_resources = ARRAY_SIZE(mv64x60_eth0_resources), + .resource = mv64x60_eth0_resources, + .dev = { + .platform_data = ð0_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_0 */ + +#ifdef CONFIG_MV643XX_ETH_1 + +static struct resource mv64x60_eth1_resources[] = { + [0] = { + .name = "eth1 irq", + .start = MV64x60_IRQ_ETH_1, + .end = MV64x60_IRQ_ETH_1, + .flags = IORESOURCE_IRQ, + }, +}; + +static char eth1_mac_addr[ETH_ALEN]; + +static struct mv643xx_eth_platform_data eth1_pd = { + .mac_addr = eth1_mac_addr, + + .tx_sram_addr = MV_SRAM_BASE_ETH1, + .tx_sram_size = MV_SRAM_TXRING_SIZE, + .tx_queue_size = MV_SRAM_TXRING_SIZE / 16, + + .rx_sram_addr = MV_SRAM_BASE_ETH1 + MV_SRAM_TXRING_SIZE, + .rx_sram_size = MV_SRAM_RXRING_SIZE, + .rx_queue_size = MV_SRAM_RXRING_SIZE / 16, +}; + +static struct platform_device eth1_device = { + .name = MV643XX_ETH_NAME, + .id = 1, + .num_resources = ARRAY_SIZE(mv64x60_eth1_resources), + .resource = mv64x60_eth1_resources, + .dev = { + .platform_data = ð1_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_1 */ + +#ifdef CONFIG_MV643XX_ETH_2 + +static struct resource mv64x60_eth2_resources[] = { + [0] = { + .name = "eth2 irq", + .start = MV64x60_IRQ_ETH_2, + .end = MV64x60_IRQ_ETH_2, + .flags = IORESOURCE_IRQ, + }, +}; + +static char eth2_mac_addr[ETH_ALEN]; + +static struct mv643xx_eth_platform_data eth2_pd = { + .mac_addr = eth2_mac_addr, +}; + +static struct platform_device eth2_device = { + .name = MV643XX_ETH_NAME, + .id = 1, + .num_resources = ARRAY_SIZE(mv64x60_eth2_resources), + .resource = mv64x60_eth2_resources, + .dev = { + .platform_data = ð2_pd, + }, +}; +#endif /* CONFIG_MV643XX_ETH_2 */ + +static struct platform_device *mv643xx_eth_pd_devs[] __initdata = { + &mv643xx_eth_shared_device, +#ifdef CONFIG_MV643XX_ETH_0 + ð0_device, +#endif +#ifdef CONFIG_MV643XX_ETH_1 + ð1_device, +#endif +#ifdef CONFIG_MV643XX_ETH_2 + ð2_device, +#endif +}; + +static u8 __init exchange_bit(u8 val, u8 cs) +{ + /* place the data */ + OCELOT_FPGA_WRITE((val << 2) | cs, EEPROM_MODE); + udelay(1); + + /* turn the clock on */ + OCELOT_FPGA_WRITE((val << 2) | cs | 0x2, EEPROM_MODE); + udelay(1); + + /* turn the clock off and read-strobe */ + OCELOT_FPGA_WRITE((val << 2) | cs | 0x10, EEPROM_MODE); + + /* return the data */ + return (OCELOT_FPGA_READ(EEPROM_MODE) >> 3) & 0x1; +} + +static void __init get_mac(char dest[6]) +{ + u8 read_opcode[12] = {1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; + int i,j; + + for (i = 0; i < 12; i++) + exchange_bit(read_opcode[i], 1); + + for (j = 0; j < 6; j++) { + dest[j] = 0; + for (i = 0; i < 8; i++) { + dest[j] <<= 1; + dest[j] |= exchange_bit(0, 1); + } + } + + /* turn off CS */ + exchange_bit(0,0); +} + +/* + * Copy and increment ethernet MAC address by a small value. + * + * This is useful for systems where the only one MAC address is stored in + * non-volatile memory for multiple ports. + */ +static inline void eth_mac_add(unsigned char *dst, unsigned char *src, + unsigned int add) +{ + int i; + + BUG_ON(add >= 256); + + for (i = ETH_ALEN; i >= 0; i--) { + dst[i] = src[i] + add; + add = dst[i] < src[i]; /* compute carry */ + } + + WARN_ON(add); +} + +static int __init mv643xx_eth_add_pds(void) +{ + unsigned char mac[ETH_ALEN]; + int ret; + + get_mac(mac); +#ifdef CONFIG_MV643XX_ETH_0 + eth_mac_add(eth1_mac_addr, mac, 0); +#endif +#ifdef CONFIG_MV643XX_ETH_1 + eth_mac_add(eth1_mac_addr, mac, 1); +#endif +#ifdef CONFIG_MV643XX_ETH_2 + eth_mac_add(eth2_mac_addr, mac, 2); +#endif + ret = platform_add_devices(mv643xx_eth_pd_devs, + ARRAY_SIZE(mv643xx_eth_pd_devs)); + + return ret; +} + +device_initcall(mv643xx_eth_add_pds); + +#endif /* defined(CONFIG_MV643XX_ETH) || defined(CONFIG_MV643XX_ETH_MODULE) */ diff --git a/arch/mips/momentum/ocelot_3/prom.c b/arch/mips/momentum/ocelot_3/prom.c index 296d945..6ce9b7f 100644 --- a/arch/mips/momentum/ocelot_3/prom.c +++ b/arch/mips/momentum/ocelot_3/prom.c @@ -34,64 +34,11 @@ struct callvectors* debug_vectors; extern unsigned long marvell_base; extern unsigned long cpu_clock; -#ifdef CONFIG_MV643XX_ETH -extern unsigned char prom_mac_addr_base[6]; -#endif - const char *get_system_type(void) { return "Momentum Ocelot-3"; } -#ifdef CONFIG_MV643XX_ETH -void burn_clocks(void) -{ - int i; - - /* this loop should burn at least 1us -- this should be plenty */ - for (i = 0; i < 0x10000; i++) - ; -} - -u8 exchange_bit(u8 val, u8 cs) -{ - /* place the data */ - OCELOT_FPGA_WRITE((val << 2) | cs, EEPROM_MODE); - burn_clocks(); - - /* turn the clock on */ - OCELOT_FPGA_WRITE((val << 2) | cs | 0x2, EEPROM_MODE); - burn_clocks(); - - /* turn the clock off and read-strobe */ - OCELOT_FPGA_WRITE((val << 2) | cs | 0x10, EEPROM_MODE); - - /* return the data */ - return ((OCELOT_FPGA_READ(EEPROM_MODE) >> 3) & 0x1); -} - -void get_mac(char dest[6]) -{ - u8 read_opcode[12] = {1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}; - int i,j; - - for (i = 0; i < 12; i++) - exchange_bit(read_opcode[i], 1); - - for (j = 0; j < 6; j++) { - dest[j] = 0; - for (i = 0; i < 8; i++) { - dest[j] <<= 1; - dest[j] |= exchange_bit(0, 1); - } - } - - /* turn off CS */ - exchange_bit(0,0); -} -#endif - - #ifdef CONFIG_64BIT unsigned long signext(unsigned long addr) @@ -228,11 +175,6 @@ void __init prom_init(void) mips_machgroup = MACH_GROUP_MOMENCO; mips_machtype = MACH_MOMENCO_OCELOT_3; -#ifdef CONFIG_MV643XX_ETH - /* get the base MAC address for on-board ethernet ports */ - get_mac(prom_mac_addr_base); -#endif - #ifndef CONFIG_64BIT debug_vectors->printf("Booting Linux kernel...\n"); #endif diff --git a/arch/mips/momentum/ocelot_3/setup.c b/arch/mips/momentum/ocelot_3/setup.c index 7d74f8c..ff0829f 100644 --- a/arch/mips/momentum/ocelot_3/setup.c +++ b/arch/mips/momentum/ocelot_3/setup.c @@ -4,7 +4,7 @@ * BRIEF MODULE DESCRIPTION * Momentum Computer Ocelot-3 board dependent boot routines * - * Copyright (C) 1996, 1997, 01, 05 Ralf Baechle + * Copyright (C) 1996, 1997, 01, 05 - 06 Ralf Baechle * Copyright (C) 2000 RidgeRun, Inc. * Copyright (C) 2001 Red Hat, Inc. * Copyright (C) 2002 Momentum Computer -- cgit v0.10.2 From 325d08d1a44b601fbf70c259fb61c38d2af7d309 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 6 Nov 2006 13:32:39 +0000 Subject: [MIPS] EV64120: Fix timer initialization for HZ != 100. Signed-off-by: Ralf Baechle diff --git a/arch/mips/gt64120/common/time.c b/arch/mips/gt64120/common/time.c index c83ae6a..e6e4832 100644 --- a/arch/mips/gt64120/common/time.c +++ b/arch/mips/gt64120/common/time.c @@ -71,7 +71,7 @@ void gt64120_time_init(void) /* Disable timer first */ GT_WRITE(GT_TC_CONTROL_OFS, 0); /* Load timer value for 100 Hz */ - GT_WRITE(GT_TC3_OFS, Sys_clock / 100); + GT_WRITE(GT_TC3_OFS, Sys_clock / HZ); /* * Create the IRQ structure entry for the timer. Since we're too early -- cgit v0.10.2 From 4a4cf77923eeb3cec40a302656d6ab5ced04ba48 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 6 Nov 2006 17:41:06 +0000 Subject: [MIPS] Make irq number allocator generally available for fixing EV64120. Signed-off-by: Ralf Baechle diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c index dd24434..9b0e49d 100644 --- a/arch/mips/kernel/irq.c +++ b/arch/mips/kernel/irq.c @@ -26,6 +26,48 @@ #include #include +static unsigned long irq_map[NR_IRQS / BITS_PER_LONG]; + +int __devinit allocate_irqno(void) +{ + int irq; + +again: + irq = find_first_zero_bit(irq_map, NR_IRQS); + + if (irq >= NR_IRQS) + return -ENOSPC; + + if (test_and_set_bit(irq, irq_map)) + goto again; + + return irq; +} + +EXPORT_SYMBOL_GPL(allocate_irqno); + +/* + * Allocate the 16 legacy interrupts for i8259 devices. This happens early + * in the kernel initialization so treating allocation failure as BUG() is + * ok. + */ +void __init alloc_legacy_irqno(void) +{ + int i; + + for (i = 0; i <= 16; i++) + BUG_ON(test_and_set_bit(i, irq_map)); +} + +void __devinit free_irqno(unsigned int irq) +{ + smp_mb__before_clear_bit(); + clear_bit(irq, irq_map); + smp_mb__after_clear_bit(); +} + +EXPORT_SYMBOL_GPL(free_irqno); + /* * 'what should we do if we get a hw irq event on an illegal vector'. * each architecture has to answer this themselves. diff --git a/arch/mips/sgi-ip27/ip27-irq.c b/arch/mips/sgi-ip27/ip27-irq.c index f01ba1f..270ecd3 100644 --- a/arch/mips/sgi-ip27/ip27-irq.c +++ b/arch/mips/sgi-ip27/ip27-irq.c @@ -354,29 +354,6 @@ static struct irq_chip bridge_irq_type = { .end = end_bridge_irq, }; -static unsigned long irq_map[NR_IRQS / BITS_PER_LONG]; - -int allocate_irqno(void) -{ - int irq; - -again: - irq = find_first_zero_bit(irq_map, NR_IRQS); - - if (irq >= NR_IRQS) - return -ENOSPC; - - if (test_and_set_bit(irq, irq_map)) - goto again; - - return irq; -} - -void free_irqno(unsigned int irq) -{ - clear_bit(irq, irq_map); -} - void __devinit register_bridge_irq(unsigned int irq) { irq_desc[irq].status = IRQ_DISABLED; diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c index c965705..5e82a26 100644 --- a/arch/mips/sgi-ip27/ip27-timer.c +++ b/arch/mips/sgi-ip27/ip27-timer.c @@ -214,8 +214,6 @@ static struct irqaction rt_irqaction = { .name = "timer" }; -extern int allocate_irqno(void); - void __init plat_timer_setup(struct irqaction *irq) { int irqno = allocate_irqno(); diff --git a/include/asm-mips/irq.h b/include/asm-mips/irq.h index 0ce2a80..35a05ca 100644 --- a/include/asm-mips/irq.h +++ b/include/asm-mips/irq.h @@ -74,4 +74,8 @@ extern int setup_irq_smtc(unsigned int irq, struct irqaction * new, unsigned long hwmask); #endif /* CONFIG_MIPS_MT_SMTC */ +extern int allocate_irqno(void); +extern void alloc_legacy_irqno(void); +extern void free_irqno(unsigned int irq); + #endif /* _ASM_IRQ_H */ -- cgit v0.10.2 From 4e5852f31a22094a19bbc305e42651b6c92f3008 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 6 Nov 2006 18:05:08 +0000 Subject: [MIPS] EV64120: Fix PCI interrupt allocation. Signed-off-by: Ralf Baechle diff --git a/arch/mips/pci/Makefile b/arch/mips/pci/Makefile index 3cf0dd4..70cb55b 100644 --- a/arch/mips/pci/Makefile +++ b/arch/mips/pci/Makefile @@ -26,7 +26,7 @@ obj-$(CONFIG_DDB5477) += fixup-ddb5477.o pci-ddb5477.o ops-ddb5477.o obj-$(CONFIG_LASAT) += pci-lasat.o obj-$(CONFIG_MIPS_ATLAS) += fixup-atlas.o obj-$(CONFIG_MIPS_COBALT) += fixup-cobalt.o -obj-$(CONFIG_MIPS_EV64120) += fixup-ev64120.o +obj-$(CONFIG_MIPS_EV64120) += pci-ev64120.o obj-$(CONFIG_SOC_AU1500) += fixup-au1000.o ops-au1000.o obj-$(CONFIG_SOC_AU1550) += fixup-au1000.o ops-au1000.o obj-$(CONFIG_SOC_PNX8550) += fixup-pnx8550.o ops-pnx8550.o diff --git a/arch/mips/pci/fixup-ev64120.c b/arch/mips/pci/fixup-ev64120.c deleted file mode 100644 index 8dbb90d..0000000 --- a/arch/mips/pci/fixup-ev64120.c +++ /dev/null @@ -1,34 +0,0 @@ -#include -#include - -int pci_range_ck(unsigned char bus, unsigned char dev) -{ - if (((bus == 0) || (bus == 1)) && (dev >= 6) && (dev <= 8)) - return 0; - - return -1; -} - -/* - * After detecting all agents over the PCI , this function is called - * in order to give an interrupt number for each PCI device starting - * from IRQ 20. It does also enables master for each device. - */ -void __devinit pcibios_fixup_bus(struct pci_bus *bus) -{ - unsigned int irq = 20; - struct pci_bus *current_bus = bus; - struct pci_dev *dev; - struct list_head *devices_link; - - list_for_each(devices_link, &(current_bus->devices)) { - dev = pci_dev_b(devices_link); - if (dev != NULL) { - dev->irq = irq++; - - /* Assign an interrupt number for the device */ - pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq); - pcibios_set_master(dev); - } - } -} diff --git a/arch/mips/pci/pci-ev64120.c b/arch/mips/pci/pci-ev64120.c new file mode 100644 index 0000000..9cd859e --- /dev/null +++ b/arch/mips/pci/pci-ev64120.c @@ -0,0 +1,21 @@ +#include + +int __init pcibios_map_irq(struct pci_dev *dev, u8 slot, u8 pin) +{ + int irq; + + if (!pin) + return 0; + + irq = allocate_irqno(); + if (irq < 0) + return 0; + + return irq; +} + +/* Do platform specific device initialization at pci_enable_device() time */ +int pcibios_plat_dev_init(struct pci_dev *dev) +{ + return 0; +} -- cgit v0.10.2 From 73f4388aedade209650ed629ab767d3e2b636f7b Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 6 Nov 2006 18:17:35 +0000 Subject: [MIPS] Fix EV64120 and Ocelot builds by providing a plat_timer_setup(). Signed-off-by: Ralf Baechle diff --git a/arch/mips/gt64120/common/time.c b/arch/mips/gt64120/common/time.c index e6e4832..c47eeb7 100644 --- a/arch/mips/gt64120/common/time.c +++ b/arch/mips/gt64120/common/time.c @@ -64,7 +64,7 @@ static irqreturn_t gt64120_irq(int irq, void *dev_id) * as *irq (=irq0 in ../kernel/time.c). We will do our own timer interrupt * handling. */ -void gt64120_time_init(void) +void __init plat_timer_setup(struct irqaction *irq) { static struct irqaction timer; diff --git a/arch/mips/gt64120/ev64120/setup.c b/arch/mips/gt64120/ev64120/setup.c index 91c2d3f..99c8d42 100644 --- a/arch/mips/gt64120/ev64120/setup.c +++ b/arch/mips/gt64120/ev64120/setup.c @@ -68,7 +68,6 @@ unsigned long __init prom_free_prom_memory(void) * Initializes basic routines and structures pointers, memory size (as * given by the bios and saves the command line. */ -extern void gt64120_time_init(void); void __init plat_mem_setup(void) { @@ -76,7 +75,6 @@ void __init plat_mem_setup(void) _machine_halt = galileo_machine_halt; pm_power_off = galileo_machine_power_off; - board_time_init = gt64120_time_init; set_io_port_base(KSEG1); } diff --git a/arch/mips/gt64120/momenco_ocelot/setup.c b/arch/mips/gt64120/momenco_ocelot/setup.c index 0e5bbee..94f94eb 100644 --- a/arch/mips/gt64120/momenco_ocelot/setup.c +++ b/arch/mips/gt64120/momenco_ocelot/setup.c @@ -70,7 +70,6 @@ extern void momenco_ocelot_restart(char *command); extern void momenco_ocelot_halt(void); extern void momenco_ocelot_power_off(void); -extern void gt64120_time_init(void); extern void momenco_ocelot_irq_setup(void); static char reset_reason; @@ -156,8 +155,6 @@ void __init plat_mem_setup(void) void (*l3func)(unsigned long)=KSEG1ADDR(&setup_l3cache); unsigned int tmpword; - board_time_init = gt64120_time_init; - _machine_restart = momenco_ocelot_restart; _machine_halt = momenco_ocelot_halt; pm_power_off = momenco_ocelot_power_off; -- cgit v0.10.2 From e78181feb0b94fb6afeaef3b28d4f5df1b847c98 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 6 Nov 2006 23:17:20 +0100 Subject: [PATCH] b44: change comment about irq mask register Through some experimentation with the similarly built bcm43xx I came to the conclusion that if the hw/firmware sets a bit in the interrupt register, an interrupt will only be raised if that bit is included in the interrupt mask. Hence, the interrupt mask is more like an interrupt control mask. This patch changes the comment to reflect that. Signed-off-by: Johannes Berg Signed-off-by: Jeff Garzik diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 1ec2174..474a4e34 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -908,8 +908,9 @@ static irqreturn_t b44_interrupt(int irq, void *dev_id) istat = br32(bp, B44_ISTAT); imask = br32(bp, B44_IMASK); - /* ??? What the fuck is the purpose of the interrupt mask - * ??? register if we have to mask it out by hand anyways? + /* The interrupt mask register controls which interrupt bits + * will actually raise an interrupt to the CPU when set by hw/firmware, + * but doesn't mask off the bits. */ istat &= imask; if (istat) { -- cgit v0.10.2 From edd106fc8ac1826dbe231b70ce0762db24133e5c Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Mon, 6 Nov 2006 08:57:12 -0800 Subject: [PATCH] e1000: Fix regression: garbled stats and irq allocation during swsusp e1000: Fix suspend/resume powerup and irq allocation From: Auke Kok After 7.0.33/2.6.16, e1000 suspend/resume left the user with an enabled device showing garbled statistics and undetermined irq allocation state, where `ifconfig eth0 down` would display `trying to free already freed irq`. Explicitly free and allocate irq as well as powerup the PHY during resume fixes when needed. Signed-off-by: Auke Kok Signed-off-by: Jeff Garzik diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 8d04752..726ec5e 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -4800,6 +4800,9 @@ e1000_suspend(struct pci_dev *pdev, pm_message_t state) if (adapter->hw.phy_type == e1000_phy_igp_3) e1000_phy_powerdown_workaround(&adapter->hw); + if (netif_running(netdev)) + e1000_free_irq(adapter); + /* Release control of h/w to f/w. If f/w is AMT enabled, this * would have already happened in close and is redundant. */ e1000_release_hw_control(adapter); @@ -4830,6 +4833,10 @@ e1000_resume(struct pci_dev *pdev) pci_enable_wake(pdev, PCI_D3hot, 0); pci_enable_wake(pdev, PCI_D3cold, 0); + if (netif_running(netdev) && (err = e1000_request_irq(adapter))) + return err; + + e1000_power_up_phy(adapter); e1000_reset(adapter); E1000_WRITE_REG(&adapter->hw, WUS, ~0); -- cgit v0.10.2 From 4891d53956abd4ad9d94127c50d931124319a324 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 7 Nov 2006 16:31:16 +0000 Subject: [CIFS] NFS stress test generates flood of "close with pending write" messages Informational/debug message was being logged too often. The error case of logging having to send a close with (presumably stuck on buggy server) pending writes is still logged. Signed-off-by: Steve French diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 7e056b9..4b07a8c 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -492,10 +492,13 @@ int cifs_close(struct inode *inode, struct file *file) the struct would be in each open file, but this should give enough time to clear the socket */ - cERROR(1,("close with pending writes")); +#ifdef CONFIG_CIFS_DEBUG2 + cFYI(1,("close delay, write pending")); +#endif /* DEBUG2 */ msleep(timeout); timeout *= 4; - } + } + cERROR(1,("close with pending writes")); rc = CIFSSMBClose(xid, pTcon, pSMBFile->netfid); } -- cgit v0.10.2 From 5fe14c851efedf95b0e7652a3a7b93ec899d1599 Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 7 Nov 2006 19:26:33 +0000 Subject: [CIFS] Explicitly set stat->blksize CIFS may perform I/O over the network in larger chunks than the page size, so it should explicitly set stat->blksize to ensure optimal I/O bandwidth Signed-off-by: Dave Kleikamp Signed-off-by: Steve French diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index dffe295..1ad8c9f 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1089,8 +1089,10 @@ int cifs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { int err = cifs_revalidate(dentry); - if (!err) + if (!err) { generic_fillattr(dentry->d_inode, stat); + stat->blksize = CIFS_MAX_MSGSIZE; + } return err; } -- cgit v0.10.2 From 6d15cb42fe4f8c07c80c9d49db721fcfe2da0e90 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Mon, 6 Nov 2006 19:29:16 +0100 Subject: [ARM] 3926/1: make timer led handle HZ != 100 The timer LED is unusable at HZ=large, since it's got a hard-wired value of 100 ticks per cycle; when HZ=1024 (for example) it's essentially always-on. This patch just makes that be HZ ticks per cycle. Signed-off-by: David Brownell Signed-off-by: Russell King diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c index 0c5a609..6ff5e3f 100644 --- a/arch/arm/kernel/time.c +++ b/arch/arm/kernel/time.c @@ -220,10 +220,10 @@ EXPORT_SYMBOL(leds_event); #ifdef CONFIG_LEDS_TIMER static inline void do_leds(void) { - static unsigned int count = 50; + static unsigned int count = HZ/2; if (--count == 0) { - count = 50; + count = HZ/2; leds_event(led_timer); } } -- cgit v0.10.2 From 5e7098275094ec405f2b19285ec0c38aead42d53 Mon Sep 17 00:00:00 2001 From: Ray Lehtiniemi Date: Tue, 7 Nov 2006 03:19:15 +0100 Subject: [ARM] 3927/1: Allow show_mem() to work with holes in memory map. show_mem() was not correctly handling holes in the memory map. It was treating the freed sections of the map as though they contained valid struct page entries. This could cause incorrect debugging output or even a kernel panic. This patch keeps the struct meminfo around after system initialization so that show_mem() can use it when scanning memory. show_mem() now walks over each bank of each online node, rather than assuming that each node contains a single contiguous bank. Signed-off-by: Ray Lehtiniemi Signed-off-by: Russell King diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index 22217fe..b5814b4 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -32,40 +32,51 @@ extern unsigned long phys_initrd_start; extern unsigned long phys_initrd_size; /* - * The sole use of this is to pass memory configuration - * data from paging_init to mem_init. + * This is used to pass memory configuration data from paging_init + * to mem_init, and by show_mem() to skip holes in the memory map. */ -static struct meminfo meminfo __initdata = { 0, }; +static struct meminfo meminfo = { 0, }; + +#define for_each_nodebank(iter,mi,no) \ + for (iter = 0; iter < mi->nr_banks; iter++) \ + if (mi->bank[iter].node == no) void show_mem(void) { int free = 0, total = 0, reserved = 0; - int shared = 0, cached = 0, slab = 0, node; + int shared = 0, cached = 0, slab = 0, node, i; + struct meminfo * mi = &meminfo; printk("Mem-info:\n"); show_free_areas(); printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); for_each_online_node(node) { - struct page *page, *end; - - page = NODE_MEM_MAP(node); - end = page + NODE_DATA(node)->node_spanned_pages; - - do { - total++; - if (PageReserved(page)) - reserved++; - else if (PageSwapCache(page)) - cached++; - else if (PageSlab(page)) - slab++; - else if (!page_count(page)) - free++; - else - shared += page_count(page) - 1; - page++; - } while (page < end); + for_each_nodebank (i,mi,node) { + unsigned int pfn1, pfn2; + struct page *page, *end; + + pfn1 = mi->bank[i].start >> PAGE_SHIFT; + pfn2 = (mi->bank[i].size + mi->bank[i].start) >> PAGE_SHIFT; + + page = NODE_MEM_MAP(node) + pfn1; + end = NODE_MEM_MAP(node) + pfn2; + + do { + total++; + if (PageReserved(page)) + reserved++; + else if (PageSwapCache(page)) + cached++; + else if (PageSlab(page)) + slab++; + else if (!page_count(page)) + free++; + else + shared += page_count(page) - 1; + page++; + } while (page < end); + } } printk("%d pages of RAM\n", total); @@ -76,10 +87,6 @@ void show_mem(void) printk("%d pages swap cached\n", cached); } -#define for_each_nodebank(iter,mi,no) \ - for (iter = 0; iter < mi->nr_banks; iter++) \ - if (mi->bank[iter].node == no) - /* * FIXME: We really want to avoid allocating the bootmap bitmap * over the top of the initrd. Hopefully, this is located towards -- cgit v0.10.2 From 16b7f4dcd340875625714438a812ea06400f9666 Mon Sep 17 00:00:00 2001 From: Jan-Benedict Glaw Date: Tue, 7 Nov 2006 23:50:37 +0100 Subject: Update for the srm_env driver. This patch contains a fix for a bug introduced more than a year ago (not setting *eof) and updates whitespace a bit. Signed-off-by: Jan-Benedict Glaw diff --git a/arch/alpha/kernel/srm_env.c b/arch/alpha/kernel/srm_env.c index 990ac61..f7dd081 100644 --- a/arch/alpha/kernel/srm_env.c +++ b/arch/alpha/kernel/srm_env.c @@ -2,7 +2,7 @@ * srm_env.c - Access to SRM environment * variables through linux' procfs * - * Copyright (C) 2001-2002 Jan-Benedict Glaw + * (C) 2001,2002,2006 by Jan-Benedict Glaw * * This driver is at all a modified version of Erik Mouw's * Documentation/DocBook/procfs_example.c, so: thank @@ -21,7 +21,7 @@ * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR * PURPOSE. See the GNU General Public License for more * details. - * + * * You should have received a copy of the GNU General Public * License along with this program; if not, write to the * Free Software Foundation, Inc., 59 Temple Place, @@ -29,33 +29,6 @@ * */ -/* - * Changelog - * ~~~~~~~~~ - * - * Thu, 22 Aug 2002 15:10:43 +0200 - * - Update Config.help entry. I got a number of emails asking - * me to tell their senders if they could make use of this - * piece of code... So: "SRM is something like BIOS for your - * Alpha" - * - Update code formatting a bit to better conform CodingStyle - * rules. - * - So this is v0.0.5, with no changes (except formatting) - * - * Wed, 22 May 2002 00:11:21 +0200 - * - Fix typo on comment (SRC -> SRM) - * - Call this "Version 0.0.4" - * - * Tue, 9 Apr 2002 18:44:40 +0200 - * - Implement access by variable name and additionally - * by number. This is done by creating two subdirectories - * where one holds all names (like the old directory - * did) and the other holding 256 files named like "0", - * "1" and so on. - * - Call this "Version 0.0.3" - * - */ - #include #include #include @@ -67,7 +40,7 @@ #define BASE_DIR "srm_environment" /* Subdir in /proc/ */ #define NAMED_DIR "named_variables" /* Subdir for known variables */ #define NUMBERED_DIR "numbered_variables" /* Subdir for all variables */ -#define VERSION "0.0.5" /* Module version */ +#define VERSION "0.0.6" /* Module version */ #define NAME "srm_env" /* Module name */ MODULE_AUTHOR("Jan-Benedict Glaw "); @@ -106,7 +79,6 @@ static srm_env_t srm_named_entries[] = { static srm_env_t srm_numbered_entries[256]; - static int srm_env_read(char *page, char **start, off_t off, int count, int *eof, void *data) @@ -115,21 +87,23 @@ srm_env_read(char *page, char **start, off_t off, int count, int *eof, unsigned long ret; srm_env_t *entry; - if(off != 0) - return -EFAULT; + if (off != 0) { + *eof = 1; + return 0; + } entry = (srm_env_t *) data; ret = callback_getenv(entry->id, page, count); - if((ret >> 61) == 0) + if ((ret >> 61) == 0) { nbytes = (int) ret; - else + *eof = 1; + } else nbytes = -EFAULT; return nbytes; } - static int srm_env_write(struct file *file, const char __user *buffer, unsigned long count, void *data) @@ -155,7 +129,7 @@ srm_env_write(struct file *file, const char __user *buffer, unsigned long count, ret1 = callback_setenv(entry->id, buf, count); if ((ret1 >> 61) == 0) { - do + do ret2 = callback_save_env(); while((ret2 >> 61) == 1); res = (int) ret1; @@ -172,14 +146,14 @@ srm_env_cleanup(void) srm_env_t *entry; unsigned long var_num; - if(base_dir) { + if (base_dir) { /* * Remove named entries */ - if(named_dir) { + if (named_dir) { entry = srm_named_entries; - while(entry->name != NULL && entry->id != 0) { - if(entry->proc_entry) { + while (entry->name != NULL && entry->id != 0) { + if (entry->proc_entry) { remove_proc_entry(entry->name, named_dir); entry->proc_entry = NULL; @@ -192,11 +166,11 @@ srm_env_cleanup(void) /* * Remove numbered entries */ - if(numbered_dir) { - for(var_num = 0; var_num <= 255; var_num++) { + if (numbered_dir) { + for (var_num = 0; var_num <= 255; var_num++) { entry = &srm_numbered_entries[var_num]; - if(entry->proc_entry) { + if (entry->proc_entry) { remove_proc_entry(entry->name, numbered_dir); entry->proc_entry = NULL; @@ -212,7 +186,6 @@ srm_env_cleanup(void) return; } - static int __init srm_env_init(void) { @@ -222,7 +195,7 @@ srm_env_init(void) /* * Check system */ - if(!alpha_using_srm) { + if (!alpha_using_srm) { printk(KERN_INFO "%s: This Alpha system doesn't " "know about SRM (or you've booted " "SRM->MILO->Linux, which gets " @@ -233,14 +206,14 @@ srm_env_init(void) /* * Init numbers */ - for(var_num = 0; var_num <= 255; var_num++) + for (var_num = 0; var_num <= 255; var_num++) sprintf(number[var_num], "%ld", var_num); /* * Create base directory */ base_dir = proc_mkdir(BASE_DIR, NULL); - if(base_dir == NULL) { + if (!base_dir) { printk(KERN_ERR "Couldn't create base dir /proc/%s\n", BASE_DIR); goto cleanup; @@ -251,7 +224,7 @@ srm_env_init(void) * Create per-name subdirectory */ named_dir = proc_mkdir(NAMED_DIR, base_dir); - if(named_dir == NULL) { + if (!named_dir) { printk(KERN_ERR "Couldn't create dir /proc/%s/%s\n", BASE_DIR, NAMED_DIR); goto cleanup; @@ -262,7 +235,7 @@ srm_env_init(void) * Create per-number subdirectory */ numbered_dir = proc_mkdir(NUMBERED_DIR, base_dir); - if(numbered_dir == NULL) { + if (!numbered_dir) { printk(KERN_ERR "Couldn't create dir /proc/%s/%s\n", BASE_DIR, NUMBERED_DIR); goto cleanup; @@ -274,10 +247,10 @@ srm_env_init(void) * Create all named nodes */ entry = srm_named_entries; - while(entry->name != NULL && entry->id != 0) { + while (entry->name && entry->id) { entry->proc_entry = create_proc_entry(entry->name, 0644, named_dir); - if(entry->proc_entry == NULL) + if (!entry->proc_entry) goto cleanup; entry->proc_entry->data = (void *) entry; @@ -291,13 +264,13 @@ srm_env_init(void) /* * Create all numbered nodes */ - for(var_num = 0; var_num <= 255; var_num++) { + for (var_num = 0; var_num <= 255; var_num++) { entry = &srm_numbered_entries[var_num]; entry->name = number[var_num]; entry->proc_entry = create_proc_entry(entry->name, 0644, numbered_dir); - if(entry->proc_entry == NULL) + if (!entry->proc_entry) goto cleanup; entry->id = var_num; @@ -318,7 +291,6 @@ cleanup: return -ENOMEM; } - static void __exit srm_env_exit(void) { @@ -328,7 +300,5 @@ srm_env_exit(void) return; } - module_init(srm_env_init); module_exit(srm_env_exit); - -- cgit v0.10.2 From 3b6a792f6ace33584897d1af08630c9acc0ce221 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Mon, 6 Nov 2006 14:34:48 -0800 Subject: [NET]: kconfig, correct traffic shaper As Patrick McHardy suggested, Traffic Shaper is now obsolete and alternative to it is no longer CBQ, since its problems with virtual devices, alter Kconfig text to reflect this -- put a link to the traffic schedulers as a whole. Signed-off-by: Jiri Slaby Acked-by: Patrick McHardy Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 9cb3ca5..6e863aa 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -2833,7 +2833,7 @@ config NET_FC "SCSI generic support". config SHAPER - tristate "Traffic Shaper (EXPERIMENTAL)" + tristate "Traffic Shaper (OBSOLETE)" depends on EXPERIMENTAL ---help--- The traffic shaper is a virtual network device that allows you to @@ -2842,9 +2842,9 @@ config SHAPER these virtual devices. See for more information. - An alternative to this traffic shaper is the experimental - Class-Based Queuing (CBQ) scheduling support which you get if you - say Y to "QoS and/or fair queuing" above. + An alternative to this traffic shaper are traffic schedulers which + you'll get if you say Y to "QoS and/or fair queuing" in + "Networking options". To compile this driver as a module, choose M here: the module will be called shaper. If unsure, say N. -- cgit v0.10.2 From 9e950efa20dc8037c27509666cba6999da9368e8 Mon Sep 17 00:00:00 2001 From: John Heffner Date: Mon, 6 Nov 2006 23:10:51 -0800 Subject: [TCP]: Don't use highmem in tcp hash size calculation. This patch removes consideration of high memory when determining TCP hash table sizes. Taking into account high memory results in tcp_mem values that are too large. Signed-off-by: John Heffner Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 66e9a72..4322318 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2270,7 +2270,7 @@ void __init tcp_init(void) thash_entries, (num_physpages >= 128 * 1024) ? 13 : 15, - HASH_HIGHMEM, + 0, &tcp_hashinfo.ehash_size, NULL, 0); @@ -2286,7 +2286,7 @@ void __init tcp_init(void) tcp_hashinfo.ehash_size, (num_physpages >= 128 * 1024) ? 13 : 15, - HASH_HIGHMEM, + 0, &tcp_hashinfo.bhash_size, NULL, 64 * 1024); -- cgit v0.10.2 From da33e3eb4876c43b78fdc7b7068653239f28714a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 7 Nov 2006 14:54:46 -0800 Subject: [PKT_SCHED] sch_htb: Use hlist_del_init(). Otherwise we can hit paths that (legally) do multiple deletes on the same node and OOPS with the HLIST poison values there instead of NULL. Signed-off-by: David S. Miller diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 9b9c555..4b52fa7 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1284,8 +1284,7 @@ static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl) struct htb_class, sibling)); /* note: this delete may happen twice (see htb_delete) */ - if (!hlist_unhashed(&cl->hlist)) - hlist_del(&cl->hlist); + hlist_del_init(&cl->hlist); list_del(&cl->sibling); if (cl->prio_activity) @@ -1333,8 +1332,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) sch_tree_lock(sch); /* delete from hash and active; remainder in destroy_class */ - if (!hlist_unhashed(&cl->hlist)) - hlist_del(&cl->hlist); + hlist_del_init(&cl->hlist); if (cl->prio_activity) htb_deactivate(q, cl); -- cgit v0.10.2 From 8e365eec04c86899fab5db0f49a9c98554069bd0 Mon Sep 17 00:00:00 2001 From: Chris Lalancette Date: Tue, 7 Nov 2006 14:56:19 -0800 Subject: [NETPOLL]: Compute checksum properly in netpoll_send_udp(). Signed-off-by: Chris Lalancette Signed-off-by: David S. Miller diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 9308af0..6589adb 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -340,6 +340,12 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) udph->dest = htons(np->remote_port); udph->len = htons(udp_len); udph->check = 0; + udph->check = csum_tcpudp_magic(htonl(np->local_ip), + htonl(np->remote_ip), + udp_len, IPPROTO_UDP, + csum_partial((unsigned char *)udph, udp_len, 0)); + if (udph->check == 0) + udph->check = -1; skb->nh.iph = iph = (struct iphdr *)skb_push(skb, sizeof(*iph)); -- cgit v0.10.2 From 25f484a62e41be8020b9a31bf50a792baa58d2d4 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 7 Nov 2006 14:57:15 -0800 Subject: [NET]: Set truesize in pskb_copy Since pskb_copy tacks on the non-linear bits from the original skb, it needs to count them in the truesize field of the new skb. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f735455..b8b1063 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -639,6 +639,7 @@ struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask) n->csum = skb->csum; n->ip_summed = skb->ip_summed; + n->truesize += skb->data_len; n->data_len = skb->data_len; n->len = skb->len; -- cgit v0.10.2 From af2c6a4aaa2253f1e29df8fb59a3d92174d30a33 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Tue, 7 Nov 2006 14:57:51 -0800 Subject: [TG3]: Fix array overrun in tg3_read_partno(). Use proper upper limits for the loops and check for all error conditions. The problem was noticed by Adrian Bunk. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 8f059b7..06e4f77 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -10212,7 +10212,7 @@ skip_phy_reset: static void __devinit tg3_read_partno(struct tg3 *tp) { unsigned char vpd_data[256]; - int i; + unsigned int i; u32 magic; if (tg3_nvram_read_swab(tp, 0x0, &magic)) @@ -10258,9 +10258,9 @@ static void __devinit tg3_read_partno(struct tg3 *tp) } /* Now parse and find the part number. */ - for (i = 0; i < 256; ) { + for (i = 0; i < 254; ) { unsigned char val = vpd_data[i]; - int block_end; + unsigned int block_end; if (val == 0x82 || val == 0x91) { i = (i + 3 + @@ -10276,21 +10276,26 @@ static void __devinit tg3_read_partno(struct tg3 *tp) (vpd_data[i + 1] + (vpd_data[i + 2] << 8))); i += 3; - while (i < block_end) { + + if (block_end > 256) + goto out_not_found; + + while (i < (block_end - 2)) { if (vpd_data[i + 0] == 'P' && vpd_data[i + 1] == 'N') { int partno_len = vpd_data[i + 2]; - if (partno_len > 24) + i += 3; + if (partno_len > 24 || (partno_len + i) > 256) goto out_not_found; memcpy(tp->board_part_number, - &vpd_data[i + 3], - partno_len); + &vpd_data[i], partno_len); /* Success. */ return; } + i += 3 + vpd_data[i + 2]; } /* Part number not found. */ -- cgit v0.10.2 From 375d9d71838970030c8e0bf0ac2abcc1a3487df8 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 7 Nov 2006 15:09:17 -0800 Subject: [DECNET]: Endianess fixes (try #2) Here are some fixes to endianess problems spotted by Al Viro. Signed-off-by: Steven Whitehouse Signed-off-by: David S. Miller diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 3456cd3..21f20f2 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -166,7 +166,7 @@ static struct hlist_head *dn_find_list(struct sock *sk) if (scp->addr.sdn_flags & SDF_WILD) return hlist_empty(&dn_wild_sk) ? &dn_wild_sk : NULL; - return &dn_sk_hash[scp->addrloc & DN_SK_HASH_MASK]; + return &dn_sk_hash[dn_ntohs(scp->addrloc) & DN_SK_HASH_MASK]; } /* @@ -180,7 +180,7 @@ static int check_port(__le16 port) if (port == 0) return -1; - sk_for_each(sk, node, &dn_sk_hash[port & DN_SK_HASH_MASK]) { + sk_for_each(sk, node, &dn_sk_hash[dn_ntohs(port) & DN_SK_HASH_MASK]) { struct dn_scp *scp = DN_SK(sk); if (scp->addrloc == port) return -1; @@ -194,12 +194,12 @@ static unsigned short port_alloc(struct sock *sk) static unsigned short port = 0x2000; unsigned short i_port = port; - while(check_port(++port) != 0) { + while(check_port(dn_htons(++port)) != 0) { if (port == i_port) return 0; } - scp->addrloc = port; + scp->addrloc = dn_htons(port); return 1; } @@ -418,7 +418,7 @@ struct sock *dn_find_by_skb(struct sk_buff *skb) struct dn_scp *scp; read_lock(&dn_hash_lock); - sk_for_each(sk, node, &dn_sk_hash[cb->dst_port & DN_SK_HASH_MASK]) { + sk_for_each(sk, node, &dn_sk_hash[dn_ntohs(cb->dst_port) & DN_SK_HASH_MASK]) { scp = DN_SK(sk); if (cb->src != dn_saddr2dn(&scp->peer)) continue; @@ -1016,13 +1016,14 @@ static void dn_access_copy(struct sk_buff *skb, struct accessdata_dn *acc) static void dn_user_copy(struct sk_buff *skb, struct optdata_dn *opt) { - unsigned char *ptr = skb->data; - - opt->opt_optl = *ptr++; - opt->opt_status = 0; - memcpy(opt->opt_data, ptr, opt->opt_optl); - skb_pull(skb, dn_ntohs(opt->opt_optl) + 1); - + unsigned char *ptr = skb->data; + u16 len = *ptr++; /* yes, it's 8bit on the wire */ + + BUG_ON(len > 16); /* we've checked the contents earlier */ + opt->opt_optl = dn_htons(len); + opt->opt_status = 0; + memcpy(opt->opt_data, ptr, len); + skb_pull(skb, len + 1); } static struct sk_buff *dn_wait_for_connect(struct sock *sk, long *timeo) diff --git a/net/decnet/dn_nsp_in.c b/net/decnet/dn_nsp_in.c index 72ecc6e..7683d4f 100644 --- a/net/decnet/dn_nsp_in.c +++ b/net/decnet/dn_nsp_in.c @@ -360,9 +360,9 @@ static void dn_nsp_conn_conf(struct sock *sk, struct sk_buff *skb) scp->max_window = decnet_no_fc_max_cwnd; if (skb->len > 0) { - unsigned char dlen = *skb->data; + u16 dlen = *skb->data; if ((dlen <= 16) && (dlen <= skb->len)) { - scp->conndata_in.opt_optl = dn_htons((__u16)dlen); + scp->conndata_in.opt_optl = dn_htons(dlen); memcpy(scp->conndata_in.opt_data, skb->data + 1, dlen); } } @@ -404,9 +404,9 @@ static void dn_nsp_disc_init(struct sock *sk, struct sk_buff *skb) memset(scp->discdata_in.opt_data, 0, 16); if (skb->len > 0) { - unsigned char dlen = *skb->data; + u16 dlen = *skb->data; if ((dlen <= 16) && (dlen <= skb->len)) { - scp->discdata_in.opt_optl = dn_htons((__u16)dlen); + scp->discdata_in.opt_optl = dn_htons(dlen); memcpy(scp->discdata_in.opt_data, skb->data + 1, dlen); } } diff --git a/net/decnet/dn_nsp_out.c b/net/decnet/dn_nsp_out.c index c2e21cd..b342e4e 100644 --- a/net/decnet/dn_nsp_out.c +++ b/net/decnet/dn_nsp_out.c @@ -526,7 +526,7 @@ void dn_send_conn_conf(struct sock *sk, gfp_t gfp) struct nsp_conn_init_msg *msg; __u8 len = (__u8)dn_ntohs(scp->conndata_out.opt_optl); - if ((skb = dn_alloc_skb(sk, 50 + dn_ntohs(scp->conndata_out.opt_optl), gfp)) == NULL) + if ((skb = dn_alloc_skb(sk, 50 + len, gfp)) == NULL) return; msg = (struct nsp_conn_init_msg *)skb_put(skb, sizeof(*msg)); diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index 3e0c882..590e0a7 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -124,8 +124,8 @@ static struct nla_policy dn_fib_rule_policy[FRA_MAX+1] __read_mostly = { static int dn_fib_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) { struct dn_fib_rule *r = (struct dn_fib_rule *)rule; - u16 daddr = fl->fld_dst; - u16 saddr = fl->fld_src; + __le16 daddr = fl->fld_dst; + __le16 saddr = fl->fld_src; if (((saddr ^ r->src) & r->srcmask) || ((daddr ^ r->dst) & r->dstmask)) -- cgit v0.10.2 From ecac598bcd1f151ee4760489bded625c147fb366 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Mon, 6 Nov 2006 09:45:31 -0600 Subject: [PATCH] bcm43xx: Drain TX status before starting IRQs Drain the Microcode TX-status-FIFO before we enable IRQs. This is required, because the FIFO may still have entries left from a previous run. Those would immediately fire after enabling IRQs and would lead to an oops in the DMA TXstatus handling code. Signed-off-by: Michael Buesch Signed-off-by: Larry Finger Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index 65edb56..62c2ff8 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -1463,6 +1463,23 @@ static void handle_irq_transmit_status(struct bcm43xx_private *bcm) } } +static void drain_txstatus_queue(struct bcm43xx_private *bcm) +{ + u32 dummy; + + if (bcm->current_core->rev < 5) + return; + /* Read all entries from the microcode TXstatus FIFO + * and throw them away. + */ + while (1) { + dummy = bcm43xx_read32(bcm, BCM43xx_MMIO_XMITSTAT_0); + if (!dummy) + break; + dummy = bcm43xx_read32(bcm, BCM43xx_MMIO_XMITSTAT_1); + } +} + static void bcm43xx_generate_noise_sample(struct bcm43xx_private *bcm) { bcm43xx_shm_write16(bcm, BCM43xx_SHM_SHARED, 0x408, 0x7F7F); @@ -3532,6 +3549,7 @@ int bcm43xx_select_wireless_core(struct bcm43xx_private *bcm, bcm43xx_macfilter_clear(bcm, BCM43xx_MACFILTER_ASSOC); bcm43xx_macfilter_set(bcm, BCM43xx_MACFILTER_SELF, (u8 *)(bcm->net_dev->dev_addr)); bcm43xx_security_init(bcm); + drain_txstatus_queue(bcm); ieee80211softmac_start(bcm->net_dev); /* Let's go! Be careful after enabling the IRQs. -- cgit v0.10.2 From 3406118cd34762a7bf6b1a4f1095f9ea7576a354 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 6 Nov 2006 09:48:48 -0600 Subject: [PATCH] bcm43xx: Add error checking in bcm43xx_sprom_write() The Coverity checker noted that these "if (err)"'s couldn't ever be true. It seems the intention was to check the return values of the bcm43xx_pci_write_config32()'s? Signed-off-by: Adrian Bunk Signed-off-by: Larry Finger Signed-off-by: Michael Buesch Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index 62c2ff8..a1b7838 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -746,7 +746,7 @@ int bcm43xx_sprom_write(struct bcm43xx_private *bcm, const u16 *sprom) if (err) goto err_ctlreg; spromctl |= 0x10; /* SPROM WRITE enable. */ - bcm43xx_pci_write_config32(bcm, BCM43xx_PCICFG_SPROMCTL, spromctl); + err = bcm43xx_pci_write_config32(bcm, BCM43xx_PCICFG_SPROMCTL, spromctl); if (err) goto err_ctlreg; /* We must burn lots of CPU cycles here, but that does not @@ -768,7 +768,7 @@ int bcm43xx_sprom_write(struct bcm43xx_private *bcm, const u16 *sprom) mdelay(20); } spromctl &= ~0x10; /* SPROM WRITE enable. */ - bcm43xx_pci_write_config32(bcm, BCM43xx_PCICFG_SPROMCTL, spromctl); + err = bcm43xx_pci_write_config32(bcm, BCM43xx_PCICFG_SPROMCTL, spromctl); if (err) goto err_ctlreg; mdelay(500); -- cgit v0.10.2 From 80c218812786f619c9a1ce50d0e7c32c7afde4de Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 7 Nov 2006 18:24:20 -0800 Subject: Linux 2.6.19-rc5 Ok, things are clearly starting to calm down.. Finally. diff --git a/Makefile b/Makefile index 9557619..8484be1 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 19 -EXTRAVERSION =-rc4 +EXTRAVERSION =-rc5 NAME=Avast! A bilge rat! # *DOCUMENTATION* -- cgit v0.10.2 From 68ff6e8e0e203580ecb118319b5a3b53962edf5a Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Wed, 8 Nov 2006 07:46:02 -0500 Subject: [libata] sata_via: fix obvious typo Spotted by Martin Devera. Signed-off-by: Jeff Garzik diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c index f4455a1..1c7f19a 100644 --- a/drivers/ata/sata_via.c +++ b/drivers/ata/sata_via.c @@ -230,7 +230,7 @@ static int vt6420_prereset(struct ata_port *ap) int online; /* don't do any SCR stuff if we're not loading */ - if (!ATA_PFLAG_LOADING) + if (!(ap->pflags & ATA_PFLAG_LOADING)) goto skip_scr; /* Resume phy. This is the old resume sequence from -- cgit v0.10.2 From de8e7c12430a73654ae3cedbc45428d56c6b777b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 8 Nov 2006 10:09:28 -0800 Subject: Revert "[PATCH] i386: Add MMCFG resources to i386 too" This reverts commit de09bddb9d6f96785be470c832b881e6d72d589f. It tried to reserve the MMCONFIG mmio memory ranges, but since the MMCONFIG information is broken and often bogus (which is why we don't dare use it most of the time _anyway_), it does more harm than good. Cc: Jeff Chua Cc: Adrian Bunk Cc: Andi Kleen Signed-off-by: Linus Torvalds diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c index d0c3da3..c6b6d9b 100644 --- a/arch/i386/pci/mmconfig.c +++ b/arch/i386/pci/mmconfig.c @@ -154,38 +154,6 @@ static struct pci_raw_ops pci_mmcfg = { .write = pci_mmcfg_write, }; - -static __init void pci_mmcfg_insert_resources(void) -{ -#define PCI_MMCFG_RESOURCE_NAME_LEN 19 - int i; - struct resource *res; - char *names; - unsigned num_buses; - - res = kcalloc(PCI_MMCFG_RESOURCE_NAME_LEN + sizeof(*res), - pci_mmcfg_config_num, GFP_KERNEL); - - if (!res) { - printk(KERN_ERR "PCI: Unable to allocate MMCONFIG resources\n"); - return; - } - - names = (void *)&res[pci_mmcfg_config_num]; - for (i = 0; i < pci_mmcfg_config_num; i++, res++) { - num_buses = pci_mmcfg_config[i].end_bus_number - - pci_mmcfg_config[i].start_bus_number + 1; - res->name = names; - snprintf(names, PCI_MMCFG_RESOURCE_NAME_LEN, "PCI MMCONFIG %u", - pci_mmcfg_config[i].pci_segment_group_number); - res->start = pci_mmcfg_config[i].base_address; - res->end = res->start + (num_buses << 20) - 1; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - insert_resource(&iomem_resource, res); - names += PCI_MMCFG_RESOURCE_NAME_LEN; - } -} - /* K8 systems have some devices (typically in the builtin northbridge) that are only accessible using type1 Normally this can be expressed in the MCFG by not listing them @@ -222,8 +190,6 @@ static __init void unreachable_devices(void) } } - - void __init pci_mmcfg_init(int type) { if ((pci_probe & PCI_PROBE_MMCONF) == 0) @@ -251,5 +217,4 @@ void __init pci_mmcfg_init(int type) pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF; unreachable_devices(); - pci_mmcfg_insert_resources(); } -- cgit v0.10.2 From 6c0ffb9d2fd987c79c6cbb81c3f3011c63749b1a Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 8 Nov 2006 10:23:03 -0800 Subject: x86-64: clean up io-apic accesses This is just commit 130fe05dbc0114609cfef9815c0c5580b42decfa ported to x86-64, for all the same reasons. It cleans up the IO-APIC accesses in order to then fix the ordering issues. We move the accessor functions (that were only used by io_apic.c) out of a header file, and use proper memory-mapped accesses rather than making up our own "volatile" pointers. Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index fe429e5..96e02d8 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -88,6 +88,52 @@ static struct irq_pin_list { short apic, pin, next; } irq_2_pin[PIN_MAP_SIZE]; +struct io_apic { + unsigned int index; + unsigned int unused[3]; + unsigned int data; +}; + +static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) +{ + return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) + + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); +} + +static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + return readl(&io_apic->data); +} + +static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + +/* + * Re-write a value: to be used for read-modify-write + * cycles where the read already set up the index register. + */ +static inline void io_apic_modify(unsigned int apic, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(value, &io_apic->data); +} + +/* + * Synchronize the IO-APIC and the CPU by doing + * a dummy read from the IO-APIC + */ +static inline void io_apic_sync(unsigned int apic) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + readl(&io_apic->data); +} + #define __DO_ACTION(R, ACTION, FINAL) \ \ { \ diff --git a/include/asm-x86_64/io_apic.h b/include/asm-x86_64/io_apic.h index 171ec2d..561ecbf 100644 --- a/include/asm-x86_64/io_apic.h +++ b/include/asm-x86_64/io_apic.h @@ -12,10 +12,6 @@ #define APIC_MISMATCH_DEBUG -#define IO_APIC_BASE(idx) \ - ((volatile int *)(__fix_to_virt(FIX_IO_APIC_BASE_0 + idx) \ - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK))) - /* * The structure of the IO-APIC: */ @@ -119,36 +115,6 @@ extern struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* non-0 if default (table-less) MP configuration */ extern int mpc_default_type; -static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) -{ - *IO_APIC_BASE(apic) = reg; - return *(IO_APIC_BASE(apic)+4); -} - -static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) -{ - *IO_APIC_BASE(apic) = reg; - *(IO_APIC_BASE(apic)+4) = value; -} - -/* - * Re-write a value: to be used for read-modify-write - * cycles where the read already set up the index register. - */ -static inline void io_apic_modify(unsigned int apic, unsigned int value) -{ - *(IO_APIC_BASE(apic)+4) = value; -} - -/* - * Synchronize the IO-APIC and the CPU by doing - * a dummy read from the IO-APIC - */ -static inline void io_apic_sync(unsigned int apic) -{ - (void) *(IO_APIC_BASE(apic)+4); -} - /* 1 if "noapic" boot option passed */ extern int skip_ioapic_setup; -- cgit v0.10.2 From 48797ebd9e8b16fddcd4ef062f792314a6b9219a Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 8 Nov 2006 10:27:54 -0800 Subject: x86-64: write IO APIC irq routing entries in correct order This is the x86-64 version of f9dadfa71bc594df09044da61d1c72701121d802 that did the same thing on i386. Since the "mask" bit is in the low word, when we write a new entry, we need to write the high word first, before we potentially unmask it. The exception is when we actually want to mask the interrupt, in which case we want to write the low word first to make sure that the high word doesn't change while the interrupt routing is still active. Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 96e02d8..3b8f9c6 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -172,12 +172,34 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) return eu.entry; } +/* + * When we write a new IO APIC routing entry, we need to write the high + * word first! If the mask bit in the low word is clear, we will enable + * the interrupt, and we need to make sure the entry is fully populated + * before that happens. + */ static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; union entry_union eu; eu.entry = e; spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0x11 + 2*pin, eu.w2); + io_apic_write(apic, 0x10 + 2*pin, eu.w1); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +/* + * When we mask an IO APIC routing entry, we need to write the low + * word first, in order to set the mask bit before we change the + * high bits! + */ +static void ioapic_mask_entry(int apic, int pin) +{ + unsigned long flags; + union entry_union eu = { .entry.mask = 1 }; + + spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x10 + 2*pin, eu.w1); io_apic_write(apic, 0x11 + 2*pin, eu.w2); spin_unlock_irqrestore(&ioapic_lock, flags); @@ -302,9 +324,7 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) /* * Disable it in the IO-APIC irq-routing table: */ - memset(&entry, 0, sizeof(entry)); - entry.mask = 1; - ioapic_write_entry(apic, pin, entry); + ioapic_mask_entry(apic, pin); } static void clear_IO_APIC (void) -- cgit v0.10.2 From 6e659c63998881e8f4a842edbe86ac8c5cdaee41 Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 8 Nov 2006 23:10:46 +0000 Subject: [CIFS] Fix mount failure when domain not specified Fixes Samba bugzilla #4176 When users do not specify their domain on mount, 2.6.18 started sending default domain instead of a null domain (which was the only way on some servers to use a default domain). Users of 2.6.18 who did not specify their domain name on mounts to certain common Windows servers that were members of a domain, but not the domain controller, would get mount failures which they did not get in 2.6.18 This fixes that issue and should remove complaints about mount behavior changing. Signed-off-by: Steve French diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index a8a0835..bbdda99 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -90,7 +90,9 @@ static void unicode_ssetup_strings(char ** pbcc_area, struct cifsSesInfo *ses, } */ /* copy user */ if(ses->userName == NULL) { - /* BB what about null user mounts - check that we do this BB */ + /* null user mount */ + *bcc_ptr = 0; + *(bcc_ptr+1) = 0; } else { /* 300 should be long enough for any conceivable user name */ bytes_ret = cifs_strtoUCS((__le16 *) bcc_ptr, ses->userName, 300, nls_cp); @@ -98,10 +100,13 @@ static void unicode_ssetup_strings(char ** pbcc_area, struct cifsSesInfo *ses, bcc_ptr += 2 * bytes_ret; bcc_ptr += 2; /* account for null termination */ /* copy domain */ - if(ses->domainName == NULL) - bytes_ret = cifs_strtoUCS((__le16 *) bcc_ptr, - "CIFS_LINUX_DOM", 32, nls_cp); - else + if(ses->domainName == NULL) { + /* Sending null domain better than using a bogus domain name (as + we did briefly in 2.6.18) since server will use its default */ + *bcc_ptr = 0; + *(bcc_ptr+1) = 0; + bytes_ret = 0; + } else bytes_ret = cifs_strtoUCS((__le16 *) bcc_ptr, ses->domainName, 256, nls_cp); bcc_ptr += 2 * bytes_ret; @@ -144,13 +149,11 @@ static void ascii_ssetup_strings(char ** pbcc_area, struct cifsSesInfo *ses, /* copy domain */ - if(ses->domainName == NULL) { - strcpy(bcc_ptr, "CIFS_LINUX_DOM"); - bcc_ptr += 14; /* strlen(CIFS_LINUX_DOM) */ - } else { + if(ses->domainName != NULL) { strncpy(bcc_ptr, ses->domainName, 256); bcc_ptr += strnlen(ses->domainName, 256); - } + } /* else we will send a null domain name + so the server will default to its own domain */ *bcc_ptr = 0; bcc_ptr++; -- cgit v0.10.2 From d654c673d6394bc26e159b1057b357371b4ce1dc Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Wed, 8 Nov 2006 17:44:37 -0800 Subject: [PATCH] Regression in 2.6.19-rc microcode driver If the microcode driver is built in (rather than module) there are some, ehm, interesting effects happening due to the new "call out to userspace" behavior that is introduced.. and which runs too early. The result is a boot hang; which is really nasty. The patch below is a minimally safe patch to fix this regression for 2.6.19 by just not requesting actual microcode updates during early boot. (That is a good idea in general anyway) The "real" fix is a lot more complex given the entire cpu hotplug scenario (during cpu hotplug you normally need to load the microcode as well); but the interactions for that are just really messy at this point; this fix at least makes it work and avoids a full detangle of hotplug. Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index c4d0291..23f5984 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -577,7 +577,7 @@ static void microcode_init_cpu(int cpu) set_cpus_allowed(current, cpumask_of_cpu(cpu)); mutex_lock(µcode_mutex); collect_cpu_info(cpu); - if (uci->valid) + if (uci->valid && system_state == SYSTEM_RUNNING) cpu_request_microcode(cpu); mutex_unlock(µcode_mutex); set_cpus_allowed(current, old); -- cgit v0.10.2 From 75b2bd55bd7724c727856fbdf3ab71d2e4287ac8 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 8 Nov 2006 17:44:38 -0800 Subject: [PATCH] A minor fix for set_mb() in Documentation/memory-barriers.txt set_mb() is used by set_current_state() which needs mb(), not wmb(). I think it would be right to assume that set_mb() implies mb(), all arches seem to do just this. Signed-off-by: Oleg Nesterov Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt index 7f790f6..7751704 100644 --- a/Documentation/memory-barriers.txt +++ b/Documentation/memory-barriers.txt @@ -1016,7 +1016,7 @@ There are some more advanced barrier functions: (*) set_mb(var, value) - This assigns the value to the variable and then inserts at least a write + This assigns the value to the variable and then inserts a full memory barrier after it, depending on the function. It isn't guaranteed to insert anything more than a compiler barrier in a UP compilation. -- cgit v0.10.2 From af85852de0b32d92b14295aa6f5ba3a9ad044cf6 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 8 Nov 2006 17:44:39 -0800 Subject: [PATCH] nfsd4: reindent do_open_lookup() Minor rearrangement, cleanup of do_open_lookup(). No change in behavior. Signed-off-by: J. Bruce Fields Acked-by: Neil Brown Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 0a7bbdc..4a73f5b 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -106,27 +106,25 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o open->op_fname.len, &open->op_iattr, &resfh, open->op_createmode, (u32 *)open->op_verf.data, &open->op_truncate); - } - else { + } else { status = nfsd_lookup(rqstp, current_fh, open->op_fname.data, open->op_fname.len, &resfh); fh_unlock(current_fh); } + if (status) + goto out; - if (!status) { - set_change_info(&open->op_cinfo, current_fh); + set_change_info(&open->op_cinfo, current_fh); - /* set reply cache */ - fh_dup2(current_fh, &resfh); - open->op_stateowner->so_replay.rp_openfh_len = - resfh.fh_handle.fh_size; - memcpy(open->op_stateowner->so_replay.rp_openfh, - &resfh.fh_handle.fh_base, - resfh.fh_handle.fh_size); + /* set reply cache */ + fh_dup2(current_fh, &resfh); + open->op_stateowner->so_replay.rp_openfh_len = resfh.fh_handle.fh_size; + memcpy(open->op_stateowner->so_replay.rp_openfh, + &resfh.fh_handle.fh_base, resfh.fh_handle.fh_size); - status = do_open_permission(rqstp, current_fh, open, MAY_NOP); - } + status = do_open_permission(rqstp, current_fh, open, MAY_NOP); +out: fh_put(&resfh); return status; } -- cgit v0.10.2 From 81ac95c5569d7a60ab5db6c1ccec56c12b3ebcb5 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 8 Nov 2006 17:44:40 -0800 Subject: [PATCH] nfsd4: fix open-create permissions In the case where an open creates the file, we shouldn't be rechecking permissions to open the file; the open succeeds regardless of what the new file's mode bits say. This patch fixes the problem, but only by introducing yet another parameter to nfsd_create_v3. This is ugly. This will be fixed by later patches. Signed-off-by: J. Bruce Fields Acked-by: Neil Brown Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 64db601..7f5bad0 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -258,7 +258,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, /* Now create the file and set attributes */ nfserr = nfsd_create_v3(rqstp, dirfhp, argp->name, argp->len, attr, newfhp, - argp->createmode, argp->verf, NULL); + argp->createmode, argp->verf, NULL, NULL); RETURN_STATUS(nfserr); } diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 4a73f5b..50bc942 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -93,6 +93,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o { struct svc_fh resfh; __be32 status; + int created = 0; fh_init(&resfh, NFS4_FHSIZE); open->op_truncate = 0; @@ -105,7 +106,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o status = nfsd_create_v3(rqstp, current_fh, open->op_fname.data, open->op_fname.len, &open->op_iattr, &resfh, open->op_createmode, - (u32 *)open->op_verf.data, &open->op_truncate); + (u32 *)open->op_verf.data, &open->op_truncate, &created); } else { status = nfsd_lookup(rqstp, current_fh, open->op_fname.data, open->op_fname.len, &resfh); @@ -122,7 +123,8 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o memcpy(open->op_stateowner->so_replay.rp_openfh, &resfh.fh_handle.fh_base, resfh.fh_handle.fh_size); - status = do_open_permission(rqstp, current_fh, open, MAY_NOP); + if (!created) + status = do_open_permission(rqstp, current_fh, open, MAY_NOP); out: fh_put(&resfh); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index f21e917..1a7ad8c 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1237,7 +1237,7 @@ __be32 nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, struct svc_fh *resfhp, int createmode, u32 *verifier, - int *truncp) + int *truncp, int *created) { struct dentry *dentry, *dchild = NULL; struct inode *dirp; @@ -1331,6 +1331,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); if (host_err < 0) goto out_nfserr; + if (created) + *created = 1; if (EX_ISSYNC(fhp->fh_export)) { err = nfserrno(nfsd_sync_dir(dentry)); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index eb23114..edb54c3 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -89,7 +89,7 @@ __be32 nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); __be32 nfsd_create_v3(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, struct svc_fh *res, int createmode, - u32 *verifier, int *truncp); + u32 *verifier, int *truncp, int *created); __be32 nfsd_commit(struct svc_rqst *, struct svc_fh *, loff_t, unsigned long); #endif /* CONFIG_NFSD_V3 */ -- cgit v0.10.2 From c06cb8b1c4d25e5b4d7a2d7c2462619de1e0dbc4 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Wed, 8 Nov 2006 17:44:41 -0800 Subject: [PATCH] i386: Force data segment to be 4K aligned o Currently there is no specific alignment restriction in linker script and in some cases it can be placed non 4K aligned addresses. This fails kexec which checks that segment to be loaded is page aligned. o I guess, it does not harm data segment to be 4K aligned. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S index adc1f23..c6f84a0 100644 --- a/arch/i386/kernel/vmlinux.lds.S +++ b/arch/i386/kernel/vmlinux.lds.S @@ -51,6 +51,7 @@ SECTIONS __tracedata_end = .; /* writeable */ + . = ALIGN(4096); .data : AT(ADDR(.data) - LOAD_OFFSET) { /* Data */ *(.data) CONSTRUCTORS -- cgit v0.10.2 From bfc5ecdf48b529f6a2bd98ba26bfac39ca8cd8a5 Mon Sep 17 00:00:00 2001 From: Alasdair G Kergon Date: Wed, 8 Nov 2006 17:44:42 -0800 Subject: [PATCH] dm: fix find_device race There is a race between dev_create() and find_device(). If the mdptr has not yet been stored against a device, find_device() needs to behave as though no device was found. It already returns NULL, but there is a dm_put() missing: it must drop the reference dm_get_md() took. The bug was introduced by dm-fix-mapped-device-ref-counting.patch. It manifests itself if another dm ioctl attempts to reference a newly-created device while the device creation ioctl is still running. The consequence is that the device cannot be removed until the machine is rebooted. Certain udev configurations can lead to this happening. Signed-off-by: Alasdair G Kergon Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index d13bb15..4510ad8 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -606,9 +606,14 @@ static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param) return __get_name_cell(param->name); md = dm_get_md(huge_decode_dev(param->dev)); - if (md) - mdptr = dm_get_mdptr(md); + if (!md) + goto out; + mdptr = dm_get_mdptr(md); + if (!mdptr) + dm_put(md); + +out: return mdptr; } -- cgit v0.10.2 From d287483d6d7a2d5b313aee155285f89b57d9cd4a Mon Sep 17 00:00:00 2001 From: Alasdair G Kergon Date: Wed, 8 Nov 2006 17:44:43 -0800 Subject: [PATCH] dm: suspend: fix error path If the device is already suspended, just return the error and skip the code that would incorrectly wipe md->suspended_bdev. (This isn't currently a problem because existing code avoids calling this function if the device is already suspended.) Signed-off-by: Alasdair G Kergon Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/dm.c b/drivers/md/dm.c index b5764a8..fc4f743 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1285,7 +1285,7 @@ int dm_suspend(struct mapped_device *md, int do_lockfs) down(&md->suspend_lock); if (dm_suspended(md)) - goto out; + goto out_unlock; map = dm_get_table(md); @@ -1361,6 +1361,8 @@ out: } dm_table_put(map); + +out_unlock: up(&md->suspend_lock); return r; } -- cgit v0.10.2 From 5d55fdf94998db1df9ee7f1def8806bfd0e5ff73 Mon Sep 17 00:00:00 2001 From: Jonathan E Brassow Date: Wed, 8 Nov 2006 17:44:43 -0800 Subject: [PATCH] dm: multipath: fix rr_add_path order When adding paths to the round-robin path selector, their order gets inverted, which is not desirable. Fix by replacing list_add() with list_add_tail(). Signed-off-by: Jonathan E Brassow Signed-off-by: Alasdair G Kergon Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/dm-round-robin.c b/drivers/md/dm-round-robin.c index c5a16c5..6f9fcd4 100644 --- a/drivers/md/dm-round-robin.c +++ b/drivers/md/dm-round-robin.c @@ -136,7 +136,7 @@ static int rr_add_path(struct path_selector *ps, struct path *path, path->pscontext = pi; - list_add(&pi->list, &s->valid_paths); + list_add_tail(&pi->list, &s->valid_paths); return 0; } -- cgit v0.10.2 From 33184048dc4f9d5550d3b6a88c8e0ff92033eb6e Mon Sep 17 00:00:00 2001 From: Jonathan E Brassow Date: Wed, 8 Nov 2006 17:44:44 -0800 Subject: [PATCH] dm: raid1: fix waiting for io on suspend All device-mapper targets must complete outstanding I/O before suspending. The mirror target generates I/O in its recovery phase and fails to wait for it. It needs to be tracked so we can ensure that it has completed before we suspend. [akpm@osdl.org: cleanup] Signed-off-by: Jonathan E Brassow Signed-off-by: Alasdair G Kergon Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 659224c..48a653b 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -24,6 +24,7 @@ static struct workqueue_struct *_kmirrord_wq; static struct work_struct _kmirrord_work; +static DECLARE_WAIT_QUEUE_HEAD(_kmirrord_recovery_stopped); static inline void wake(void) { @@ -83,6 +84,7 @@ struct region_hash { struct list_head *buckets; spinlock_t region_lock; + atomic_t recovery_in_flight; struct semaphore recovery_count; struct list_head clean_regions; struct list_head quiesced_regions; @@ -191,6 +193,7 @@ static int rh_init(struct region_hash *rh, struct mirror_set *ms, spin_lock_init(&rh->region_lock); sema_init(&rh->recovery_count, 0); + atomic_set(&rh->recovery_in_flight, 0); INIT_LIST_HEAD(&rh->clean_regions); INIT_LIST_HEAD(&rh->quiesced_regions); INIT_LIST_HEAD(&rh->recovered_regions); @@ -382,6 +385,8 @@ static void rh_update_states(struct region_hash *rh) rh->log->type->clear_region(rh->log, reg->key); rh->log->type->complete_resync_work(rh->log, reg->key, 1); dispatch_bios(rh->ms, ®->delayed_bios); + if (atomic_dec_and_test(&rh->recovery_in_flight)) + wake_up_all(&_kmirrord_recovery_stopped); up(&rh->recovery_count); mempool_free(reg, rh->region_pool); } @@ -502,11 +507,21 @@ static int __rh_recovery_prepare(struct region_hash *rh) static void rh_recovery_prepare(struct region_hash *rh) { - while (!down_trylock(&rh->recovery_count)) + /* Extra reference to avoid race with rh_stop_recovery */ + atomic_inc(&rh->recovery_in_flight); + + while (!down_trylock(&rh->recovery_count)) { + atomic_inc(&rh->recovery_in_flight); if (__rh_recovery_prepare(rh) <= 0) { + atomic_dec(&rh->recovery_in_flight); up(&rh->recovery_count); break; } + } + + /* Drop the extra reference */ + if (atomic_dec_and_test(&rh->recovery_in_flight)) + wake_up_all(&_kmirrord_recovery_stopped); } /* @@ -1177,6 +1192,11 @@ static void mirror_postsuspend(struct dm_target *ti) struct dirty_log *log = ms->rh.log; rh_stop_recovery(&ms->rh); + + /* Wait for all I/O we generated to complete */ + wait_event(_kmirrord_recovery_stopped, + !atomic_read(&ms->rh.recovery_in_flight)); + if (log->type->suspend && log->type->suspend(log)) /* FIXME: need better error handling */ DMWARN("log suspend failed"); -- cgit v0.10.2 From b196872cd65a06ad65853c4513e0d0f24452d32e Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 8 Nov 2006 17:44:45 -0800 Subject: [PATCH] drivers/telephony/ixj: fix an array overrun The Coverity checker noted that in drivers/telephony/ixj.c:ixj_build_filter_cadence(), filter_en[4] or filter_en[5] could be written to. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/telephony/ixj.h b/drivers/telephony/ixj.h index fbea454..8d69bcd 100644 --- a/drivers/telephony/ixj.h +++ b/drivers/telephony/ixj.h @@ -1295,7 +1295,7 @@ typedef struct { Proc_Info_Type Info_write; unsigned short frame_count; unsigned int filter_hist[4]; - unsigned char filter_en[4]; + unsigned char filter_en[6]; unsigned short proc_load; unsigned long framesread; unsigned long frameswritten; -- cgit v0.10.2 From b5b9df697b166948f2770fb65d1b8809637a199b Mon Sep 17 00:00:00 2001 From: Tigran Aivazian Date: Wed, 8 Nov 2006 17:44:46 -0800 Subject: [PATCH] Tigran has moved Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/CREDITS b/CREDITS index 5329ead..606d407 100644 --- a/CREDITS +++ b/CREDITS @@ -45,7 +45,7 @@ S: Longford, Ireland S: Sydney, Australia N: Tigran A. Aivazian -E: tigran@veritas.com +E: tigran@aivazian.fsnet.co.uk W: http://www.moses.uklinux.net/patches D: BFS filesystem D: Intel IA32 CPU microcode update support diff --git a/MAINTAINERS b/MAINTAINERS index d708702..8732dae 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -493,7 +493,7 @@ S: Maintained BFS FILE SYSTEM P: Tigran A. Aivazian -M: tigran@veritas.com +M: tigran@aivazian.fsnet.co.uk L: linux-kernel@vger.kernel.org S: Maintained @@ -1513,7 +1513,7 @@ S: Maintained INTEL IA32 MICROCODE UPDATE SUPPORT P: Tigran Aivazian -M: tigran@veritas.com +M: tigran@aivazian.fsnet.co.uk S: Maintained INTEL IXP4XX RANDOM NUMBER GENERATOR SUPPORT -- cgit v0.10.2 From 2f4713036114dd13d1f4fe433b7f236250b65f5a Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 8 Nov 2006 17:44:47 -0800 Subject: [PATCH] md: change ONLINE/OFFLINE events to a single CHANGE event It turns out that CHANGE is preferred to ONLINE/OFFLINE for various reasons (not least of which being that udev understands it already). So remove the recently added KOBJ_OFFLINE (no-one is likely to care anyway) and change the ONLINE to a CHANGE event Cc: Kay Sievers Signed-off-by: Neil Brown Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index d111356..ae50a24 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3200,7 +3200,7 @@ static int do_md_run(mddev_t * mddev) mddev->changed = 1; md_new_event(mddev); - kobject_uevent(&mddev->gendisk->kobj, KOBJ_ONLINE); + kobject_uevent(&mddev->gendisk->kobj, KOBJ_CHANGE); return 0; } @@ -3314,7 +3314,6 @@ static int do_md_stop(mddev_t * mddev, int mode) module_put(mddev->pers->owner); mddev->pers = NULL; - kobject_uevent(&mddev->gendisk->kobj, KOBJ_OFFLINE); if (mddev->ro) mddev->ro = 0; } -- cgit v0.10.2 From 0692c6b1cf5537b190f90fb5903f1af89a41b0a8 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 8 Nov 2006 17:44:48 -0800 Subject: [PATCH] md: fix sizing problem with raid5-reshape and CONFIG_LBD=n I forgot to has the size-in-blocks to (loff_t) before shifting up to a size-in-bytes. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index e14f457..69c3e20 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3659,7 +3659,7 @@ static void end_reshape(raid5_conf_t *conf) bdev = bdget_disk(conf->mddev->gendisk, 0); if (bdev) { mutex_lock(&bdev->bd_inode->i_mutex); - i_size_write(bdev->bd_inode, conf->mddev->array_size << 10); + i_size_write(bdev->bd_inode, (loff_t)conf->mddev->array_size << 10); mutex_unlock(&bdev->bd_inode->i_mutex); bdput(bdev); } -- cgit v0.10.2 From 4b438a23fb05b6566393f9f0a3987ea3dcc1c0c4 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 8 Nov 2006 17:44:48 -0800 Subject: [PATCH] md: do not freeze md threads for suspend If there's a swap file on a software RAID, it should be possible to use this file for saving the swsusp's suspend image. Also, this file should be available to the memory management subsystem when memory is being freed before the suspend image is created. For the above reasons it seems that md_threads should not be frozen during the suspend and the appended patch makes this happen, but then there is the question if they don't cause any data to be written to disks after the suspend image has been created, provided that all filesystems are frozen at that time. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/md.c b/drivers/md/md.c index ae50a24..8cbf9c9 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -4486,6 +4486,7 @@ static int md_thread(void * arg) * many dirty RAID5 blocks. */ + current->flags |= PF_NOFREEZE; allow_signal(SIGKILL); while (!kthread_should_stop()) { @@ -4502,7 +4503,6 @@ static int md_thread(void * arg) test_bit(THREAD_WAKEUP, &thread->flags) || kthread_should_stop(), thread->timeout); - try_to_freeze(); clear_bit(THREAD_WAKEUP, &thread->flags); -- cgit v0.10.2 From 8bdc052eccdc7893d075d3f1f7103594a458c8c4 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 8 Nov 2006 17:44:49 -0800 Subject: [PATCH] kretprobe: fix kretprobe-booster to save regs and set status There are two bugs in the kretprobe-booster. 1) It doesn't make room for gs registers. 2) It doesn't change status of the current kprobe. This status will effect the fault handling. This patch fixes these bugs and, additionally, saves skipped registers for compatibility with the original kretprobe. Signed-off-by: Masami Hiramatsu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index d98e44b..fc79e1e 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -361,8 +361,11 @@ no_kprobe: asm volatile ( ".global kretprobe_trampoline\n" "kretprobe_trampoline: \n" " pushf\n" - /* skip cs, eip, orig_eax, es, ds */ - " subl $20, %esp\n" + /* skip cs, eip, orig_eax */ + " subl $12, %esp\n" + " pushl %gs\n" + " pushl %ds\n" + " pushl %es\n" " pushl %eax\n" " pushl %ebp\n" " pushl %edi\n" @@ -373,10 +376,10 @@ no_kprobe: " movl %esp, %eax\n" " call trampoline_handler\n" /* move eflags to cs */ - " movl 48(%esp), %edx\n" - " movl %edx, 44(%esp)\n" + " movl 52(%esp), %edx\n" + " movl %edx, 48(%esp)\n" /* save true return address on eflags */ - " movl %eax, 48(%esp)\n" + " movl %eax, 52(%esp)\n" " popl %ebx\n" " popl %ecx\n" " popl %edx\n" @@ -384,8 +387,8 @@ no_kprobe: " popl %edi\n" " popl %ebp\n" " popl %eax\n" - /* skip eip, orig_eax, es, ds */ - " addl $16, %esp\n" + /* skip eip, orig_eax, es, ds, gs */ + " addl $20, %esp\n" " popf\n" " ret\n"); } @@ -404,6 +407,10 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs) INIT_HLIST_HEAD(&empty_rp); spin_lock_irqsave(&kretprobe_lock, flags); head = kretprobe_inst_table_head(current); + /* fixup registers */ + regs->xcs = __KERNEL_CS; + regs->eip = trampoline_address; + regs->orig_eax = 0xffffffff; /* * It is possible to have multiple instances associated with a given @@ -425,6 +432,7 @@ fastcall void *__kprobes trampoline_handler(struct pt_regs *regs) if (ri->rp && ri->rp->handler){ __get_cpu_var(current_kprobe) = &ri->rp->kp; + get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE; ri->rp->handler(ri, regs); __get_cpu_var(current_kprobe) = NULL; } -- cgit v0.10.2 From 6c33eb39976b67628452ebc791834c7d590e545e Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 8 Nov 2006 17:44:50 -0800 Subject: [PATCH] ia64: select ACPI_NUMA if ACPI When ACPI && NUMA, pxm_to_node is used and it exists in drivers/acpi/numa.c Tony said: The patch makes sense ... if you pick both of "ACPI" and "NUMA", then you need (and should automatically be given) ACPI_NUMA too. The only open question is whether there is a better way of getting there. Perhaps with less configuration options in the first place? We are heading towards a future where so many systems will be NUMA that there would seem to be little benefit in keeping ACPI_NUMA separate from ACPI ... but perhaps we aren't quite there yet. Signed-off-by: KAMEZAWA Hiroyuki Cc: Len Brown Acked-by: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 70f7eb9..1468239 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -341,6 +341,7 @@ config NUMA bool "NUMA support" depends on !IA64_HP_SIM && !FLATMEM default y if IA64_SGI_SN2 + select ACPI_NUMA if ACPI help Say Y to compile the kernel to support NUMA (Non-Uniform Memory Access). This option is for configuring high-end multiprocessor -- cgit v0.10.2 From 13bb7e37e5081d03643e2bd64f3f5d21f32e7221 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 8 Nov 2006 17:44:51 -0800 Subject: [PATCH] sysctl: Undeprecate sys_sysctl The basic issue is that despite have been deprecated and warned about as a very bad thing in the man pages since its inception there are a few real users of sys_sysctl. It was my assumption that because sysctl had been deprecated for all of 2.6 there would be no user space users by this point, so I initially gave sys_sysctl a very short deprecation period. Now that I know there are a few real users the only sane way to proceed with deprecation is to push the time limit out to a year or two work and work with distributions that have big testing pools like fedora core to find these last remaining users. Which means that the sys_sysctl interface needs to be maintained in the meantime. Since I have provided a technical measure that allows us to add new sysctl entries without reserving more binary numbers I believe that is enough to fix the sys_sysctl binary interface maintenance problems, because there is no longer a need to change the binary interface at all. Since the sys_sysctl implementation needs to stay around for a while and the worst of the maintenance issues that caused us to occasionally break the ABI have been addressed I don't see any advantage in continuing with the removal of sys_sysctl. So instead of merely increasing the deprecation period this patch removes the deprecation of sys_sysctl and modifies the kernel to compile the code in by default. With committing to maintain sys_sysctl we get all of the advantages of a fast interface for anything that needs it. Currently sys_sysctl is about 5x faster than /proc/sys, for the same string data. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 1ac3c74..d52c4aa 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -53,18 +53,6 @@ Who: Mauro Carvalho Chehab --------------------------- -What: sys_sysctl -When: January 2007 -Why: The same information is available through /proc/sys and that is the - interface user space prefers to use. And there do not appear to be - any existing user in user space of sys_sysctl. The additional - maintenance overhead of keeping a set of binary names gets - in the way of doing a good job of maintaining this interface. - -Who: Eric Biederman - ---------------------------- - What: PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl]) When: November 2005 Files: drivers/pcmcia/: pcmcia_ioctl.c diff --git a/init/Kconfig b/init/Kconfig index c8b2624..176f7e5 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -304,20 +304,19 @@ config UID16 config SYSCTL_SYSCALL bool "Sysctl syscall support" if EMBEDDED - default n + default y select SYSCTL ---help--- - Enable the deprecated sysctl system call. sys_sysctl uses - binary paths that have been found to be a major pain to maintain - and use. The interface in /proc/sys is now the primary and what - everyone uses. + sys_sysctl uses binary paths that have been found challenging + to properly maintain and use. The interface in /proc/sys + using paths with ascii names is now the primary path to this + information. - Nothing has been using the binary sysctl interface for some - time now so nothing should break if you disable sysctl syscall - support, and your kernel will get marginally smaller. + Almost nothing using the binary sysctl interface so if you are + trying to save some space it is probably safe to disable this, + making your kernel marginally smaller. - Unless you have an application that uses the sys_sysctl interface - you should probably say N here. + If unsure say Y here. config KALLSYMS bool "Load all symbols for debugging/kksymoops" if EMBEDDED -- cgit v0.10.2 From f3ce6a0ead8c557e9acdc733addd23cbc206c7e3 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 8 Nov 2006 17:44:52 -0800 Subject: [PATCH] IPMI: Clean up the waiting message queue properly on unload A wrong function was being used to free a list; this fixes the problem. Otherwise, an oops at unload time was possible. But not likely, since you can't have any users when you unload the modules and it is very hard to get messages into this queue without users. Signed-off-by: Corey Minyard Cc: Patrick Schoeller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 34a4fd1..e55a0d2 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -376,13 +376,23 @@ static void free_recv_msg_list(struct list_head *q) } } +static void free_smi_msg_list(struct list_head *q) +{ + struct ipmi_smi_msg *msg, *msg2; + + list_for_each_entry_safe(msg, msg2, q, link) { + list_del(&msg->link); + ipmi_free_smi_msg(msg); + } +} + static void clean_up_interface_data(ipmi_smi_t intf) { int i; struct cmd_rcvr *rcvr, *rcvr2; struct list_head list; - free_recv_msg_list(&intf->waiting_msgs); + free_smi_msg_list(&intf->waiting_msgs); free_recv_msg_list(&intf->waiting_events); /* Wholesale remove all the entries from the list in the -- cgit v0.10.2 From 46d52b09fa6a2d1e313cb75ca352d6f466e67bd1 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 8 Nov 2006 17:44:55 -0800 Subject: [PATCH] IPMI: retry messages on certain error returns Some more errors from the IPMI send message command are retryable, but are not being retried by the IPMI code. Make sure they get retried. Signed-off-by: Corey Minyard Cc: Frederic Lelievre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index e55a0d2..0b07ca1 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3242,7 +3242,9 @@ void ipmi_smi_msg_received(ipmi_smi_t intf, report the error immediately. */ if ((msg->rsp_size >= 3) && (msg->rsp[2] != 0) && (msg->rsp[2] != IPMI_NODE_BUSY_ERR) - && (msg->rsp[2] != IPMI_LOST_ARBITRATION_ERR)) + && (msg->rsp[2] != IPMI_LOST_ARBITRATION_ERR) + && (msg->rsp[2] != IPMI_BUS_ERR) + && (msg->rsp[2] != IPMI_NAK_ON_WRITE_ERR)) { int chan = msg->rsp[3] & 0xf; diff --git a/include/linux/ipmi_msgdefs.h b/include/linux/ipmi_msgdefs.h index 22f5e2a..4d04d8b 100644 --- a/include/linux/ipmi_msgdefs.h +++ b/include/linux/ipmi_msgdefs.h @@ -75,6 +75,8 @@ #define IPMI_INVALID_COMMAND_ERR 0xc1 #define IPMI_ERR_MSG_TRUNCATED 0xc6 #define IPMI_LOST_ARBITRATION_ERR 0x81 +#define IPMI_BUS_ERR 0x82 +#define IPMI_NAK_ON_WRITE_ERR 0x83 #define IPMI_ERR_UNSPECIFIED 0xff #define IPMI_CHANNEL_PROTOCOL_IPMB 1 -- cgit v0.10.2 From 64d9fe6973a9348e5211f3cc9f04b899329caeb4 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 8 Nov 2006 17:44:56 -0800 Subject: [PATCH] ipmi_si_intf.c: fix "&& 0xff" typos Signed-off-by: Alexey Dobriyan Acked-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 157fa81..abc5149 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -1211,7 +1211,7 @@ static void intf_mem_outb(struct si_sm_io *io, unsigned int offset, static unsigned char intf_mem_inw(struct si_sm_io *io, unsigned int offset) { return (readw((io->addr)+(offset * io->regspacing)) >> io->regshift) - && 0xff; + & 0xff; } static void intf_mem_outw(struct si_sm_io *io, unsigned int offset, @@ -1223,7 +1223,7 @@ static void intf_mem_outw(struct si_sm_io *io, unsigned int offset, static unsigned char intf_mem_inl(struct si_sm_io *io, unsigned int offset) { return (readl((io->addr)+(offset * io->regspacing)) >> io->regshift) - && 0xff; + & 0xff; } static void intf_mem_outl(struct si_sm_io *io, unsigned int offset, @@ -1236,7 +1236,7 @@ static void intf_mem_outl(struct si_sm_io *io, unsigned int offset, static unsigned char mem_inq(struct si_sm_io *io, unsigned int offset) { return (readq((io->addr)+(offset * io->regspacing)) >> io->regshift) - && 0xff; + & 0xff; } static void mem_outq(struct si_sm_io *io, unsigned int offset, -- cgit v0.10.2 From ec68307cc5a8dc499e48693843bb42f6b6028458 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 8 Nov 2006 17:44:57 -0800 Subject: [PATCH] htirq: refactor so we only have one function that writes to the chip This refactoring actually optimizes the code a little by caching the value that we think the device is programmed with instead of reading it back from the hardware. Which simplifies the code a little and should speed things up a bit. This patch introduces the concept of a ht_irq_msg and modifies the architecture read/write routines to update this code. There is a minor consistency fix here as well as x86_64 forgot to initialize the htirq as masked. Signed-off-by: Eric W. Biederman Cc: Andi Kleen Acked-by: Bryan O'Sullivan Cc: Cc: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 507983c..ad84bc2 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2624,18 +2624,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -2673,7 +2671,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; cpumask_t tmp; @@ -2681,9 +2679,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) cpu_set(vector >> 8, tmp); dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2695,8 +2694,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_MT_ARBITRATED) | HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 3b8f9c6..41bfc49 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1955,18 +1955,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -1987,7 +1985,7 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) dest = cpu_mask_to_apicid(tmp); - target_ht_irq(irq, dest, vector & 0xff); + target_ht_irq(irq, dest, vector); set_native_irq_info(irq, mask); } #endif @@ -2010,14 +2008,15 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq, TARGET_CPUS, &tmp); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2026,10 +2025,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_RQEOI_EDGE | ((INT_DELIVERY_MODE != dest_LowestPrio) ? HT_IRQ_LOW_MT_FIXED : - HT_IRQ_LOW_MT_ARBITRATED); + HT_IRQ_LOW_MT_ARBITRATED) | + HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index 0e27f24..e346fe3 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -27,82 +27,55 @@ struct ht_irq_cfg { struct pci_dev *dev; unsigned pos; unsigned idx; + struct ht_irq_msg msg; }; -void write_ht_irq_low(unsigned int irq, u32 data) -{ - struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); -} - -void write_ht_irq_high(unsigned int irq, u32 data) -{ - struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); -} -u32 read_ht_irq_low(unsigned int irq) +void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) { struct ht_irq_cfg *cfg = get_irq_data(irq); unsigned long flags; - u32 data; spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); + if (cfg->msg.address_lo != msg->address_lo) { + pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); + pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_lo); + } + if (cfg->msg.address_hi != msg->address_hi) { + pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); + pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_hi); + } spin_unlock_irqrestore(&ht_irq_lock, flags); - return data; + cfg->msg = *msg; } -u32 read_ht_irq_high(unsigned int irq) +void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) { struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - u32 data; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - spin_unlock_irqrestore(&ht_irq_lock, flags); - return data; + *msg = cfg->msg; } void mask_ht_irq(unsigned int irq) { struct ht_irq_cfg *cfg; - unsigned long flags; - u32 data; + struct ht_irq_msg msg; cfg = get_irq_data(irq); - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - data |= 1; - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); + msg = cfg->msg; + msg.address_lo |= 1; + write_ht_irq_msg(irq, &msg); } void unmask_ht_irq(unsigned int irq) { struct ht_irq_cfg *cfg; - unsigned long flags; - u32 data; + struct ht_irq_msg msg; cfg = get_irq_data(irq); - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - data &= ~1; - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); + msg = cfg->msg; + msg.address_lo &= ~1; + write_ht_irq_msg(irq, &msg); } /** @@ -152,6 +125,9 @@ int ht_create_irq(struct pci_dev *dev, int idx) cfg->dev = dev; cfg->pos = pos; cfg->idx = 0x10 + (idx * 2); + /* Initialize msg to a value that will never match the first write. */ + cfg->msg.address_lo = 0xffffffff; + cfg->msg.address_hi = 0xffffffff; irq = create_irq(); if (irq < 0) { diff --git a/include/linux/htirq.h b/include/linux/htirq.h index 1f15ce2..108f0d9 100644 --- a/include/linux/htirq.h +++ b/include/linux/htirq.h @@ -1,11 +1,14 @@ #ifndef LINUX_HTIRQ_H #define LINUX_HTIRQ_H +struct ht_irq_msg { + u32 address_lo; /* low 32 bits of the ht irq message */ + u32 address_hi; /* high 32 bits of the it irq message */ +}; + /* Helper functions.. */ -void write_ht_irq_low(unsigned int irq, u32 data); -void write_ht_irq_high(unsigned int irq, u32 data); -u32 read_ht_irq_low(unsigned int irq); -u32 read_ht_irq_high(unsigned int irq); +void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); +void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); void mask_ht_irq(unsigned int irq); void unmask_ht_irq(unsigned int irq); -- cgit v0.10.2 From 43539c38cd8edb915d1f0e1f55dcb70638b4cc8e Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 8 Nov 2006 17:44:57 -0800 Subject: [PATCH] htirq: allow buggy drivers of buggy hardware to write the registers This patch adds a variant of ht_create_irq __ht_create_irq that takes an aditional parameter update that is a function that is called whenever we want to write to a drivers htirq configuration registers. This is needed to support the ipath_iba6110 because it's registers in the proper location are not actually conected to the hardware that controlls interrupt delivery. [bos@serpentine.com: fixes] Signed-off-by: Eric W. Biederman Cc: Andi Kleen Cc: Cc: Roland Dreier Signed-off-by: Bryan O'Sullivan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index e346fe3..0a8d1cc 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -25,6 +25,8 @@ static DEFINE_SPINLOCK(ht_irq_lock); struct ht_irq_cfg { struct pci_dev *dev; + /* Update callback used to cope with buggy hardware */ + ht_irq_update_t *update; unsigned pos; unsigned idx; struct ht_irq_msg msg; @@ -44,6 +46,8 @@ void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_hi); } + if (cfg->update) + cfg->update(cfg->dev, irq, msg); spin_unlock_irqrestore(&ht_irq_lock, flags); cfg->msg = *msg; } @@ -79,16 +83,14 @@ void unmask_ht_irq(unsigned int irq) } /** - * ht_create_irq - create an irq and attach it to a device. + * __ht_create_irq - create an irq and attach it to a device. * @dev: The hypertransport device to find the irq capability on. * @idx: Which of the possible irqs to attach to. - * - * ht_create_irq is needs to be called for all hypertransport devices - * that generate irqs. + * @update: Function to be called when changing the htirq message * * The irq number of the new irq or a negative error value is returned. */ -int ht_create_irq(struct pci_dev *dev, int idx) +int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update) { struct ht_irq_cfg *cfg; unsigned long flags; @@ -123,6 +125,7 @@ int ht_create_irq(struct pci_dev *dev, int idx) return -ENOMEM; cfg->dev = dev; + cfg->update = update; cfg->pos = pos; cfg->idx = 0x10 + (idx * 2); /* Initialize msg to a value that will never match the first write. */ @@ -145,6 +148,21 @@ int ht_create_irq(struct pci_dev *dev, int idx) } /** + * ht_create_irq - create an irq and attach it to a device. + * @dev: The hypertransport device to find the irq capability on. + * @idx: Which of the possible irqs to attach to. + * + * ht_create_irq needs to be called for all hypertransport devices + * that generate irqs. + * + * The irq number of the new irq or a negative error value is returned. + */ +int ht_create_irq(struct pci_dev *dev, int idx) +{ + return __ht_create_irq(dev, idx, NULL); +} + +/** * ht_destroy_irq - destroy an irq created with ht_create_irq * * This reverses ht_create_irq removing the specified irq from @@ -162,5 +180,6 @@ void ht_destroy_irq(unsigned int irq) kfree(cfg); } +EXPORT_SYMBOL(__ht_create_irq); EXPORT_SYMBOL(ht_create_irq); EXPORT_SYMBOL(ht_destroy_irq); diff --git a/include/linux/htirq.h b/include/linux/htirq.h index 108f0d9..c96ea46 100644 --- a/include/linux/htirq.h +++ b/include/linux/htirq.h @@ -15,4 +15,9 @@ void unmask_ht_irq(unsigned int irq); /* The arch hook for getting things started */ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev); +/* For drivers of buggy hardware */ +typedef void (ht_irq_update_t)(struct pci_dev *dev, int irq, + struct ht_irq_msg *msg); +int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update); + #endif /* LINUX_HTIRQ_H */ -- cgit v0.10.2 From 51f65ebccf55121832c265838f93949f898b12ff Mon Sep 17 00:00:00 2001 From: Bryan O'Sullivan Date: Wed, 8 Nov 2006 17:44:58 -0800 Subject: [PATCH] IB/ipath - program intconfig register using new HT irq hook Eric's changes to the htirq infrastructure require corresponding modifications to the ipath HT driver code so that interrupts are still delivered properly. Signed-off-by: Bryan O'Sullivan Cc: Eric W. Biederman Cc: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/infiniband/hw/ipath/ipath_driver.c b/drivers/infiniband/hw/ipath/ipath_driver.c index b4ffaa7..09a13c1 100644 --- a/drivers/infiniband/hw/ipath/ipath_driver.c +++ b/drivers/infiniband/hw/ipath/ipath_driver.c @@ -304,7 +304,7 @@ static int __devinit ipath_init_one(struct pci_dev *pdev, } addr = pci_resource_start(pdev, 0); len = pci_resource_len(pdev, 0); - ipath_cdbg(VERBOSE, "regbase (0) %llx len %d irq %x, vend %x/%x " + ipath_cdbg(VERBOSE, "regbase (0) %llx len %d pdev->irq %d, vend %x/%x " "driver_data %lx\n", addr, len, pdev->irq, ent->vendor, ent->device, ent->driver_data); @@ -467,15 +467,15 @@ static int __devinit ipath_init_one(struct pci_dev *pdev, * check 0 irq after we return from chip-specific bus setup, since * that can affect this due to setup */ - if (!pdev->irq) + if (!dd->ipath_irq) ipath_dev_err(dd, "irq is 0, BIOS error? Interrupts won't " "work\n"); else { - ret = request_irq(pdev->irq, ipath_intr, IRQF_SHARED, + ret = request_irq(dd->ipath_irq, ipath_intr, IRQF_SHARED, IPATH_DRV_NAME, dd); if (ret) { ipath_dev_err(dd, "Couldn't setup irq handler, " - "irq=%u: %d\n", pdev->irq, ret); + "irq=%d: %d\n", dd->ipath_irq, ret); goto bail_iounmap; } } @@ -637,11 +637,10 @@ static void __devexit ipath_remove_one(struct pci_dev *pdev) * free up port 0 (kernel) rcvhdr, egr bufs, and eventually tid bufs * for all versions of the driver, if they were allocated */ - if (pdev->irq) { - ipath_cdbg(VERBOSE, - "unit %u free_irq of irq %x\n", - dd->ipath_unit, pdev->irq); - free_irq(pdev->irq, dd); + if (dd->ipath_irq) { + ipath_cdbg(VERBOSE, "unit %u free irq %d\n", + dd->ipath_unit, dd->ipath_irq); + dd->ipath_f_free_irq(dd); } else ipath_dbg("irq is 0, not doing free_irq " "for unit %u\n", dd->ipath_unit); diff --git a/drivers/infiniband/hw/ipath/ipath_iba6110.c b/drivers/infiniband/hw/ipath/ipath_iba6110.c index 9e4e8d4..e57c7a3 100644 --- a/drivers/infiniband/hw/ipath/ipath_iba6110.c +++ b/drivers/infiniband/hw/ipath/ipath_iba6110.c @@ -38,6 +38,7 @@ #include #include +#include #include "ipath_kernel.h" #include "ipath_registers.h" @@ -913,49 +914,40 @@ static void slave_or_pri_blk(struct ipath_devdata *dd, struct pci_dev *pdev, } } -static int set_int_handler(struct ipath_devdata *dd, struct pci_dev *pdev, - int pos) +static int ipath_ht_intconfig(struct ipath_devdata *dd) { - u32 int_handler_addr_lower; - u32 int_handler_addr_upper; - u64 ihandler; - u32 intvec; + int ret; - /* use indirection register to get the intr handler */ - pci_write_config_byte(pdev, pos + HT_INTR_REG_INDEX, 0x10); - pci_read_config_dword(pdev, pos + 4, &int_handler_addr_lower); - pci_write_config_byte(pdev, pos + HT_INTR_REG_INDEX, 0x11); - pci_read_config_dword(pdev, pos + 4, &int_handler_addr_upper); + if (dd->ipath_intconfig) { + ipath_write_kreg(dd, dd->ipath_kregs->kr_interruptconfig, + dd->ipath_intconfig); /* interrupt address */ + ret = 0; + } else { + ipath_dev_err(dd, "No interrupts enabled, couldn't setup " + "interrupt address\n"); + ret = -EINVAL; + } - ihandler = (u64) int_handler_addr_lower | - ((u64) int_handler_addr_upper << 32); + return ret; +} + +static void ipath_ht_irq_update(struct pci_dev *dev, int irq, + struct ht_irq_msg *msg) +{ + struct ipath_devdata *dd = pci_get_drvdata(dev); + u64 prev_intconfig = dd->ipath_intconfig; + + dd->ipath_intconfig = msg->address_lo; + dd->ipath_intconfig |= ((u64) msg->address_hi) << 32; /* - * kernels with CONFIG_PCI_MSI set the vector in the irq field of - * struct pci_device, so we use that to program the internal - * interrupt register (not config space) with that value. The BIOS - * must still have done the basic MSI setup. - */ - intvec = pdev->irq; - /* - * clear any vector bits there; normally not set but we'll overload - * this for some debug purposes (setting the HTC debug register - * value from software, rather than GPIOs), so it might be set on a - * driver reload. + * If the previous value of dd->ipath_intconfig is zero, we're + * getting configured for the first time, and must not program the + * intconfig register here (it will be programmed later, when the + * hardware is ready). Otherwise, we should. */ - ihandler &= ~0xff0000; - /* x86 vector goes in intrinfo[23:16] */ - ihandler |= intvec << 16; - ipath_cdbg(VERBOSE, "ihandler lower %x, upper %x, intvec %x, " - "interruptconfig %llx\n", int_handler_addr_lower, - int_handler_addr_upper, intvec, - (unsigned long long) ihandler); - - /* can't program yet, so save for interrupt setup */ - dd->ipath_intconfig = ihandler; - /* keep going, so we find link control stuff also */ - - return ihandler != 0; + if (prev_intconfig) + ipath_ht_intconfig(dd); } /** @@ -971,12 +963,19 @@ static int set_int_handler(struct ipath_devdata *dd, struct pci_dev *pdev, static int ipath_setup_ht_config(struct ipath_devdata *dd, struct pci_dev *pdev) { - int pos, ret = 0; - int ihandler = 0; + int pos, ret; + + ret = __ht_create_irq(pdev, 0, ipath_ht_irq_update); + if (ret < 0) { + ipath_dev_err(dd, "Couldn't create interrupt handler: " + "err %d\n", ret); + goto bail; + } + dd->ipath_irq = ret; + ret = 0; /* - * Read the capability info to find the interrupt info, and also - * handle clearing CRC errors in linkctrl register if necessary. We + * Handle clearing CRC errors in linkctrl register if necessary. We * do this early, before we ever enable errors or hardware errors, * mostly to avoid causing the chip to enter freeze mode. */ @@ -1000,17 +999,9 @@ static int ipath_setup_ht_config(struct ipath_devdata *dd, } if (!(cap_type & 0xE0)) slave_or_pri_blk(dd, pdev, pos, cap_type); - else if (cap_type == HT_INTR_DISC_CONFIG) - ihandler = set_int_handler(dd, pdev, pos); } while ((pos = pci_find_next_capability(pdev, pos, PCI_CAP_ID_HT))); - if (!ihandler) { - ipath_dev_err(dd, "Couldn't find interrupt handler in " - "config space\n"); - ret = -ENODEV; - } - bail: return ret; } @@ -1360,25 +1351,6 @@ static void ipath_ht_quiet_serdes(struct ipath_devdata *dd) ipath_write_kreg(dd, dd->ipath_kregs->kr_serdesconfig0, val); } -static int ipath_ht_intconfig(struct ipath_devdata *dd) -{ - int ret; - - if (!dd->ipath_intconfig) { - ipath_dev_err(dd, "No interrupts enabled, couldn't setup " - "interrupt address\n"); - ret = 1; - goto bail; - } - - ipath_write_kreg(dd, dd->ipath_kregs->kr_interruptconfig, - dd->ipath_intconfig); /* interrupt address */ - ret = 0; - -bail: - return ret; -} - /** * ipath_pe_put_tid - write a TID in chip * @dd: the infinipath device @@ -1575,6 +1547,14 @@ static int ipath_ht_get_base_info(struct ipath_portdata *pd, void *kbase) return 0; } +static void ipath_ht_free_irq(struct ipath_devdata *dd) +{ + free_irq(dd->ipath_irq, dd); + ht_destroy_irq(dd->ipath_irq); + dd->ipath_irq = 0; + dd->ipath_intconfig = 0; +} + /** * ipath_init_iba6110_funcs - set up the chip-specific function pointers * @dd: the infinipath device @@ -1598,6 +1578,7 @@ void ipath_init_iba6110_funcs(struct ipath_devdata *dd) dd->ipath_f_cleanup = ipath_setup_ht_cleanup; dd->ipath_f_setextled = ipath_setup_ht_setextled; dd->ipath_f_get_base_info = ipath_ht_get_base_info; + dd->ipath_f_free_irq = ipath_ht_free_irq; /* * initialize chip-specific variables diff --git a/drivers/infiniband/hw/ipath/ipath_iba6120.c b/drivers/infiniband/hw/ipath/ipath_iba6120.c index a72ab9d..6af8968 100644 --- a/drivers/infiniband/hw/ipath/ipath_iba6120.c +++ b/drivers/infiniband/hw/ipath/ipath_iba6120.c @@ -851,6 +851,7 @@ static int ipath_setup_pe_config(struct ipath_devdata *dd, int pos, ret; dd->ipath_msi_lo = 0; /* used as a flag during reset processing */ + dd->ipath_irq = pdev->irq; ret = pci_enable_msi(dd->pcidev); if (ret) ipath_dev_err(dd, "pci_enable_msi failed: %d, " @@ -1323,6 +1324,12 @@ done: return 0; } +static void ipath_pe_free_irq(struct ipath_devdata *dd) +{ + free_irq(dd->ipath_irq, dd); + dd->ipath_irq = 0; +} + /** * ipath_init_iba6120_funcs - set up the chip-specific function pointers * @dd: the infinipath device @@ -1349,6 +1356,7 @@ void ipath_init_iba6120_funcs(struct ipath_devdata *dd) dd->ipath_f_cleanup = ipath_setup_pe_cleanup; dd->ipath_f_setextled = ipath_setup_pe_setextled; dd->ipath_f_get_base_info = ipath_pe_get_base_info; + dd->ipath_f_free_irq = ipath_pe_free_irq; /* initialize chip-specific variables */ dd->ipath_f_tidtemplate = ipath_pe_tidtemplate; diff --git a/drivers/infiniband/hw/ipath/ipath_intr.c b/drivers/infiniband/hw/ipath/ipath_intr.c index d9079ee..5652a55 100644 --- a/drivers/infiniband/hw/ipath/ipath_intr.c +++ b/drivers/infiniband/hw/ipath/ipath_intr.c @@ -710,14 +710,14 @@ static void ipath_bad_intr(struct ipath_devdata *dd, u32 * unexpectp) * linuxbios development work, and it may happen in * the future again. */ - if (dd->pcidev && dd->pcidev->irq) { + if (dd->pcidev && dd->ipath_irq) { ipath_dev_err(dd, "Now %u unexpected " "interrupts, unregistering " "interrupt handler\n", *unexpectp); - ipath_dbg("free_irq of irq %x\n", - dd->pcidev->irq); - free_irq(dd->pcidev->irq, dd); + ipath_dbg("free_irq of irq %d\n", + dd->ipath_irq); + dd->ipath_f_free_irq(dd); } } if (ipath_read_kreg32(dd, dd->ipath_kregs->kr_intmask)) { @@ -753,7 +753,7 @@ static void ipath_bad_regread(struct ipath_devdata *dd) if (allbits == 2) { ipath_dev_err(dd, "Still bad interrupt status, " "unregistering interrupt\n"); - free_irq(dd->pcidev->irq, dd); + dd->ipath_f_free_irq(dd); } else if (allbits > 2) { if ((allbits % 10000) == 0) printk("."); diff --git a/drivers/infiniband/hw/ipath/ipath_kernel.h b/drivers/infiniband/hw/ipath/ipath_kernel.h index 06d5020..986b212 100644 --- a/drivers/infiniband/hw/ipath/ipath_kernel.h +++ b/drivers/infiniband/hw/ipath/ipath_kernel.h @@ -213,6 +213,8 @@ struct ipath_devdata { void (*ipath_f_setextled)(struct ipath_devdata *, u64, u64); /* fill out chip-specific fields */ int (*ipath_f_get_base_info)(struct ipath_portdata *, void *); + /* free irq */ + void (*ipath_f_free_irq)(struct ipath_devdata *); struct ipath_ibdev *verbs_dev; struct timer_list verbs_timer; /* total dwords sent (summed from counter) */ @@ -328,6 +330,8 @@ struct ipath_devdata { /* so we can rewrite it after a chip reset */ u32 ipath_pcibar1; + /* interrupt number */ + int ipath_irq; /* HT/PCI Vendor ID (here for NodeInfo) */ u16 ipath_vendorid; /* HT/PCI Device ID (here for NodeInfo) */ -- cgit v0.10.2 From 088406bcf66d6c7fd8a5c04c00aa410ae9077403 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 8 Nov 2006 17:44:59 -0800 Subject: [PATCH] nfsd: fix spurious error return from nfsd_create in async case Commit 6264d69d7df654ca64f625e9409189a0e50734e9 modified the nfsd_create() error handling in such a way that nfsd_create will usually return nfserr_perm even when succesful, if the export has the async export option. This introduced a regression that could cause mkdir() to always return a permissions error, even though the directory in question was actually succesfully created. Signed-off-by: J. Bruce Fields Acked-by: NeilBrown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 1a7ad8c..bb4d926 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1177,7 +1177,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, /* * Get the dir op function pointer. */ - err = nfserr_perm; + err = 0; switch (type) { case S_IFREG: host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); -- cgit v0.10.2 From 621da0f8af228525e4b40390e36fbdc44a587cf1 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 9 Nov 2006 16:00:06 +1100 Subject: [POWERPC] Make sure initrd and dtb sections get into zImage correctly The "wrapper" script was using the wrong names for the initrd and dtb (device-tree blob) sections. This fixes it, and also ensures the symbols for the start and end of the dtb get defined correctly. Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/boot/wrapper b/arch/powerpc/boot/wrapper index eab7318..b5fb1fe 100755 --- a/arch/powerpc/boot/wrapper +++ b/arch/powerpc/boot/wrapper @@ -179,11 +179,11 @@ if [ -z "$cacheit" ]; then fi if [ -n "$initrd" ]; then - addsec $tmp "$initrd" initrd + addsec $tmp "$initrd" $isection fi if [ -n "$dtb" ]; then - addsec $tmp "$dtb" dtb + addsec $tmp "$dtb" .kernel:dtb fi if [ "$platform" != "miboot" ]; then diff --git a/arch/powerpc/boot/zImage.lds.S b/arch/powerpc/boot/zImage.lds.S index 4b6bb3f..4be3c64 100644 --- a/arch/powerpc/boot/zImage.lds.S +++ b/arch/powerpc/boot/zImage.lds.S @@ -21,6 +21,11 @@ SECTIONS __got2_end = .; } + . = ALIGN(8); + _dtb_start = .; + .kernel:dtb : { *(.kernel:dtb) } + _dtb_end = .; + . = ALIGN(4096); _vmlinux_start = .; .kernel:vmlinux.strip : { *(.kernel:vmlinux.strip) } -- cgit v0.10.2 From 3f048109d9c4f8bb028ccb0d256ab65eb44f5988 Mon Sep 17 00:00:00 2001 From: "malahal@us.ibm.com" Date: Wed, 4 Oct 2006 17:28:37 -0700 Subject: [SCSI] aic94xx SCSI timeout fix The patch updates DDB0 in the aic94xx driver itself. It doesn't supply or use lldd_port_formed field. DDB0 is updated prior to posting notification to libsas layer. Signed-off-by: Malahal Naineni Signed-off-by: James Bottomley diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c index 3c2d7a3..af7e011 100644 --- a/drivers/scsi/aic94xx/aic94xx_hwi.c +++ b/drivers/scsi/aic94xx/aic94xx_hwi.c @@ -112,6 +112,21 @@ static int asd_init_phy(struct asd_phy *phy) return 0; } +static void asd_init_ports(struct asd_ha_struct *asd_ha) +{ + int i; + + spin_lock_init(&asd_ha->asd_ports_lock); + for (i = 0; i < ASD_MAX_PHYS; i++) { + struct asd_port *asd_port = &asd_ha->asd_ports[i]; + + memset(asd_port->sas_addr, 0, SAS_ADDR_SIZE); + memset(asd_port->attached_sas_addr, 0, SAS_ADDR_SIZE); + asd_port->phy_mask = 0; + asd_port->num_phys = 0; + } +} + static int asd_init_phys(struct asd_ha_struct *asd_ha) { u8 i; @@ -121,6 +136,7 @@ static int asd_init_phys(struct asd_ha_struct *asd_ha) struct asd_phy *phy = &asd_ha->phys[i]; phy->phy_desc = &asd_ha->hw_prof.phy_desc[i]; + phy->asd_port = NULL; phy->sas_phy.enabled = 0; phy->sas_phy.id = i; @@ -658,6 +674,8 @@ int asd_init_hw(struct asd_ha_struct *asd_ha) goto Out; } + asd_init_ports(asd_ha); + err = asd_init_scbs(asd_ha); if (err) { asd_printk("couldn't initialize scbs for %s\n", diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.h b/drivers/scsi/aic94xx/aic94xx_hwi.h index 7b6aca0..c6c3d18 100644 --- a/drivers/scsi/aic94xx/aic94xx_hwi.h +++ b/drivers/scsi/aic94xx/aic94xx_hwi.h @@ -193,6 +193,16 @@ struct asd_seq_data { struct asd_ascb **escb_arr; /* array of pointers to escbs */ }; +/* This is an internal port structure. These are used to get accurate + * phy_mask for updating DDB 0. + */ +struct asd_port { + u8 sas_addr[SAS_ADDR_SIZE]; + u8 attached_sas_addr[SAS_ADDR_SIZE]; + u32 phy_mask; + int num_phys; +}; + /* This is the Host Adapter structure. It describes the hardware * SAS adapter. */ @@ -211,6 +221,8 @@ struct asd_ha_struct { struct hw_profile hw_prof; struct asd_phy phys[ASD_MAX_PHYS]; + spinlock_t asd_ports_lock; + struct asd_port asd_ports[ASD_MAX_PHYS]; struct asd_sas_port ports[ASD_MAX_PHYS]; struct dma_pool *scb_pool; diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c index a4cc432..57c5ba4 100644 --- a/drivers/scsi/aic94xx/aic94xx_init.c +++ b/drivers/scsi/aic94xx/aic94xx_init.c @@ -786,8 +786,6 @@ static void asd_remove_driver_attrs(struct device_driver *driver) } static struct sas_domain_function_template aic94xx_transport_functions = { - .lldd_port_formed = asd_update_port_links, - .lldd_dev_found = asd_dev_found, .lldd_dev_gone = asd_dev_gone, diff --git a/drivers/scsi/aic94xx/aic94xx_sas.h b/drivers/scsi/aic94xx/aic94xx_sas.h index 64d2317..9050e93 100644 --- a/drivers/scsi/aic94xx/aic94xx_sas.h +++ b/drivers/scsi/aic94xx/aic94xx_sas.h @@ -733,6 +733,7 @@ struct asd_phy { struct sas_identify_frame *identify_frame; struct asd_dma_tok *id_frm_tok; + struct asd_port *asd_port; u8 frame_rcvd[ASD_EDB_SIZE]; }; diff --git a/drivers/scsi/aic94xx/aic94xx_scb.c b/drivers/scsi/aic94xx/aic94xx_scb.c index 7ee49b5..b15caf1 100644 --- a/drivers/scsi/aic94xx/aic94xx_scb.c +++ b/drivers/scsi/aic94xx/aic94xx_scb.c @@ -168,6 +168,70 @@ static inline void asd_get_attached_sas_addr(struct asd_phy *phy, u8 *sas_addr) } } +static void asd_form_port(struct asd_ha_struct *asd_ha, struct asd_phy *phy) +{ + int i; + struct asd_port *free_port = NULL; + struct asd_port *port; + struct asd_sas_phy *sas_phy = &phy->sas_phy; + unsigned long flags; + + spin_lock_irqsave(&asd_ha->asd_ports_lock, flags); + if (!phy->asd_port) { + for (i = 0; i < ASD_MAX_PHYS; i++) { + port = &asd_ha->asd_ports[i]; + + /* Check for wide port */ + if (port->num_phys > 0 && + memcmp(port->sas_addr, sas_phy->sas_addr, + SAS_ADDR_SIZE) == 0 && + memcmp(port->attached_sas_addr, + sas_phy->attached_sas_addr, + SAS_ADDR_SIZE) == 0) { + break; + } + + /* Find a free port */ + if (port->num_phys == 0 && free_port == NULL) { + free_port = port; + } + } + + /* Use a free port if this doesn't form a wide port */ + if (i >= ASD_MAX_PHYS) { + port = free_port; + BUG_ON(!port); + memcpy(port->sas_addr, sas_phy->sas_addr, + SAS_ADDR_SIZE); + memcpy(port->attached_sas_addr, + sas_phy->attached_sas_addr, + SAS_ADDR_SIZE); + } + port->num_phys++; + port->phy_mask |= (1U << sas_phy->id); + phy->asd_port = port; + } + ASD_DPRINTK("%s: updating phy_mask 0x%x for phy%d\n", + __FUNCTION__, phy->asd_port->phy_mask, sas_phy->id); + asd_update_port_links(asd_ha, phy); + spin_unlock_irqrestore(&asd_ha->asd_ports_lock, flags); +} + +static void asd_deform_port(struct asd_ha_struct *asd_ha, struct asd_phy *phy) +{ + struct asd_port *port = phy->asd_port; + struct asd_sas_phy *sas_phy = &phy->sas_phy; + unsigned long flags; + + spin_lock_irqsave(&asd_ha->asd_ports_lock, flags); + if (port) { + port->num_phys--; + port->phy_mask &= ~(1U << sas_phy->id); + phy->asd_port = NULL; + } + spin_unlock_irqrestore(&asd_ha->asd_ports_lock, flags); +} + static inline void asd_bytes_dmaed_tasklet(struct asd_ascb *ascb, struct done_list_struct *dl, int edb_id, int phy_id) @@ -187,6 +251,7 @@ static inline void asd_bytes_dmaed_tasklet(struct asd_ascb *ascb, asd_get_attached_sas_addr(phy, phy->sas_phy.attached_sas_addr); spin_unlock_irqrestore(&phy->sas_phy.frame_rcvd_lock, flags); asd_dump_frame_rcvd(phy, dl); + asd_form_port(ascb->ha, phy); sas_ha->notify_port_event(&phy->sas_phy, PORTE_BYTES_DMAED); } @@ -197,6 +262,7 @@ static inline void asd_link_reset_err_tasklet(struct asd_ascb *ascb, struct asd_ha_struct *asd_ha = ascb->ha; struct sas_ha_struct *sas_ha = &asd_ha->sas_ha; struct asd_sas_phy *sas_phy = sas_ha->sas_phy[phy_id]; + struct asd_phy *phy = &asd_ha->phys[phy_id]; u8 lr_error = dl->status_block[1]; u8 retries_left = dl->status_block[2]; @@ -221,6 +287,7 @@ static inline void asd_link_reset_err_tasklet(struct asd_ascb *ascb, asd_turn_led(asd_ha, phy_id, 0); sas_phy_disconnected(sas_phy); + asd_deform_port(asd_ha, phy); sas_ha->notify_port_event(sas_phy, PORTE_LINK_RESET_ERR); if (retries_left == 0) { @@ -248,6 +315,8 @@ static inline void asd_primitive_rcvd_tasklet(struct asd_ascb *ascb, unsigned long flags; struct sas_ha_struct *sas_ha = &ascb->ha->sas_ha; struct asd_sas_phy *sas_phy = sas_ha->sas_phy[phy_id]; + struct asd_ha_struct *asd_ha = ascb->ha; + struct asd_phy *phy = &asd_ha->phys[phy_id]; u8 reg = dl->status_block[1]; u32 cont = dl->status_block[2] << ((reg & 3)*8); @@ -284,6 +353,7 @@ static inline void asd_primitive_rcvd_tasklet(struct asd_ascb *ascb, phy_id); /* The sequencer disables all phys on that port. * We have to re-enable the phys ourselves. */ + asd_deform_port(asd_ha, phy); sas_ha->notify_port_event(sas_phy, PORTE_HARD_RESET); break; @@ -351,6 +421,7 @@ static void escb_tasklet_complete(struct asd_ascb *ascb, u8 sb_opcode = dl->status_block[0]; int phy_id = sb_opcode & DL_PHY_MASK; struct asd_sas_phy *sas_phy = sas_ha->sas_phy[phy_id]; + struct asd_phy *phy = &asd_ha->phys[phy_id]; if (edb > 6 || edb < 0) { ASD_DPRINTK("edb is 0x%x! dl->opcode is 0x%x\n", @@ -395,6 +466,7 @@ static void escb_tasklet_complete(struct asd_ascb *ascb, asd_turn_led(asd_ha, phy_id, 0); /* the device is gone */ sas_phy_disconnected(sas_phy); + asd_deform_port(asd_ha, phy); sas_ha->notify_port_event(sas_phy, PORTE_TIMER_EVENT); break; case REQ_TASK_ABORT: diff --git a/drivers/scsi/aic94xx/aic94xx_seq.c b/drivers/scsi/aic94xx/aic94xx_seq.c index 56e4b3b..8451125 100644 --- a/drivers/scsi/aic94xx/aic94xx_seq.c +++ b/drivers/scsi/aic94xx/aic94xx_seq.c @@ -1369,10 +1369,9 @@ int asd_start_seqs(struct asd_ha_struct *asd_ha) * port_map_by_links is also used as the conn_mask byte in the * initiator/target port DDB. */ -void asd_update_port_links(struct asd_sas_phy *sas_phy) +void asd_update_port_links(struct asd_ha_struct *asd_ha, struct asd_phy *phy) { - struct asd_ha_struct *asd_ha = sas_phy->ha->lldd_ha; - const u8 phy_mask = (u8) sas_phy->port->phy_mask; + const u8 phy_mask = (u8) phy->asd_port->phy_mask; u8 phy_is_up; u8 mask; int i, err; diff --git a/drivers/scsi/aic94xx/aic94xx_seq.h b/drivers/scsi/aic94xx/aic94xx_seq.h index 42281c3..9e715e5 100644 --- a/drivers/scsi/aic94xx/aic94xx_seq.h +++ b/drivers/scsi/aic94xx/aic94xx_seq.h @@ -64,7 +64,7 @@ int asd_unpause_lseq(struct asd_ha_struct *asd_ha, u8 lseq_mask); int asd_init_seqs(struct asd_ha_struct *asd_ha); int asd_start_seqs(struct asd_ha_struct *asd_ha); -void asd_update_port_links(struct asd_sas_phy *phy); +void asd_update_port_links(struct asd_ha_struct *asd_ha, struct asd_phy *phy); #endif #endif -- cgit v0.10.2 From 42961ee8fc4b05f5ca4d96ab34abd5149afe3541 Mon Sep 17 00:00:00 2001 From: "malahal@us.ibm.com" Date: Wed, 4 Oct 2006 17:34:03 -0700 Subject: [SCSI] aic94xx SCSI timeout fix: SMP retry fix. Updating DDB0 inside aic94xx driver itself caused SMP command timeout. I hit this SMP timeout problem twice but I am not able to reproduce it since then. Here is a fix that retries an SMP command. Signed-off-by: Malahal Naineni Signed-off-by: James Bottomley diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index 30b8014..e34a934 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -71,55 +71,65 @@ static void smp_task_done(struct sas_task *task) static int smp_execute_task(struct domain_device *dev, void *req, int req_size, void *resp, int resp_size) { - int res; - struct sas_task *task = sas_alloc_task(GFP_KERNEL); + int res, retry; + struct sas_task *task = NULL; struct sas_internal *i = to_sas_internal(dev->port->ha->core.shost->transportt); - if (!task) - return -ENOMEM; - - task->dev = dev; - task->task_proto = dev->tproto; - sg_init_one(&task->smp_task.smp_req, req, req_size); - sg_init_one(&task->smp_task.smp_resp, resp, resp_size); + for (retry = 0; retry < 3; retry++) { + task = sas_alloc_task(GFP_KERNEL); + if (!task) + return -ENOMEM; - task->task_done = smp_task_done; + task->dev = dev; + task->task_proto = dev->tproto; + sg_init_one(&task->smp_task.smp_req, req, req_size); + sg_init_one(&task->smp_task.smp_resp, resp, resp_size); - task->timer.data = (unsigned long) task; - task->timer.function = smp_task_timedout; - task->timer.expires = jiffies + SMP_TIMEOUT*HZ; - add_timer(&task->timer); + task->task_done = smp_task_done; - res = i->dft->lldd_execute_task(task, 1, GFP_KERNEL); + task->timer.data = (unsigned long) task; + task->timer.function = smp_task_timedout; + task->timer.expires = jiffies + SMP_TIMEOUT*HZ; + add_timer(&task->timer); - if (res) { - del_timer(&task->timer); - SAS_DPRINTK("executing SMP task failed:%d\n", res); - goto ex_err; - } + res = i->dft->lldd_execute_task(task, 1, GFP_KERNEL); - wait_for_completion(&task->completion); - res = -ETASK; - if ((task->task_state_flags & SAS_TASK_STATE_ABORTED)) { - SAS_DPRINTK("smp task timed out or aborted\n"); - i->dft->lldd_abort_task(task); - if (!(task->task_state_flags & SAS_TASK_STATE_DONE)) { - SAS_DPRINTK("SMP task aborted and not done\n"); + if (res) { + del_timer(&task->timer); + SAS_DPRINTK("executing SMP task failed:%d\n", res); goto ex_err; } + + wait_for_completion(&task->completion); + res = -ETASK; + if ((task->task_state_flags & SAS_TASK_STATE_ABORTED)) { + SAS_DPRINTK("smp task timed out or aborted\n"); + i->dft->lldd_abort_task(task); + if (!(task->task_state_flags & SAS_TASK_STATE_DONE)) { + SAS_DPRINTK("SMP task aborted and not done\n"); + goto ex_err; + } + } + if (task->task_status.resp == SAS_TASK_COMPLETE && + task->task_status.stat == SAM_GOOD) { + res = 0; + break; + } else { + SAS_DPRINTK("%s: task to dev %016llx response: 0x%x " + "status 0x%x\n", __FUNCTION__, + SAS_ADDR(dev->sas_addr), + task->task_status.resp, + task->task_status.stat); + sas_free_task(task); + task = NULL; + } } - if (task->task_status.resp == SAS_TASK_COMPLETE && - task->task_status.stat == SAM_GOOD) - res = 0; - else - SAS_DPRINTK("%s: task to dev %016llx response: 0x%x " - "status 0x%x\n", __FUNCTION__, - SAS_ADDR(dev->sas_addr), - task->task_status.resp, - task->task_status.stat); ex_err: - sas_free_task(task); + BUG_ON(retry == 3 && task != NULL); + if (task != NULL) { + sas_free_task(task); + } return res; } -- cgit v0.10.2 From 4039c30ef5d9189ff8dc72aaf610d1c933877e20 Mon Sep 17 00:00:00 2001 From: adam radford Date: Thu, 26 Oct 2006 18:01:06 -0700 Subject: [SCSI] 3ware 9000 add support for 9650SE Updates the 3ware 9000 driver: - Free irq handler in __twa_shutdown(). - Serialize reset code. - Add support for 9650SE controllers. Signed-off-by: Adam Radford Signed-off-by: James Bottomley diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c index 5f8c26c..b091a0f 100644 --- a/drivers/scsi/3w-9xxx.c +++ b/drivers/scsi/3w-9xxx.c @@ -66,6 +66,9 @@ 2.26.02.006 - Fix 9550SX pchip reset timeout. Add big endian support. 2.26.02.007 - Disable local interrupts during kmap/unmap_atomic(). + 2.26.02.008 - Free irq handler in __twa_shutdown(). + Serialize reset code. + Add support for 9650SE controllers. */ #include @@ -89,7 +92,7 @@ #include "3w-9xxx.h" /* Globals */ -#define TW_DRIVER_VERSION "2.26.02.007" +#define TW_DRIVER_VERSION "2.26.02.008" static TW_Device_Extension *twa_device_extension_list[TW_MAX_SLOT]; static unsigned int twa_device_extension_count; static int twa_major = -1; @@ -566,9 +569,9 @@ static int twa_check_srl(TW_Device_Extension *tw_dev, int *flashed) goto out; } - tw_dev->working_srl = fw_on_ctlr_srl; - tw_dev->working_branch = fw_on_ctlr_branch; - tw_dev->working_build = fw_on_ctlr_build; + tw_dev->tw_compat_info.working_srl = fw_on_ctlr_srl; + tw_dev->tw_compat_info.working_branch = fw_on_ctlr_branch; + tw_dev->tw_compat_info.working_build = fw_on_ctlr_build; /* Try base mode compatibility */ if (!(init_connect_result & TW_CTLR_FW_COMPATIBLE)) { @@ -590,10 +593,23 @@ static int twa_check_srl(TW_Device_Extension *tw_dev, int *flashed) } goto out; } - tw_dev->working_srl = TW_BASE_FW_SRL; - tw_dev->working_branch = TW_BASE_FW_BRANCH; - tw_dev->working_build = TW_BASE_FW_BUILD; - } + tw_dev->tw_compat_info.working_srl = TW_BASE_FW_SRL; + tw_dev->tw_compat_info.working_branch = TW_BASE_FW_BRANCH; + tw_dev->tw_compat_info.working_build = TW_BASE_FW_BUILD; + } + + /* Load rest of compatibility struct */ + strncpy(tw_dev->tw_compat_info.driver_version, TW_DRIVER_VERSION, strlen(TW_DRIVER_VERSION)); + tw_dev->tw_compat_info.driver_srl_high = TW_CURRENT_DRIVER_SRL; + tw_dev->tw_compat_info.driver_branch_high = TW_CURRENT_DRIVER_BRANCH; + tw_dev->tw_compat_info.driver_build_high = TW_CURRENT_DRIVER_BUILD; + tw_dev->tw_compat_info.driver_srl_low = TW_BASE_FW_SRL; + tw_dev->tw_compat_info.driver_branch_low = TW_BASE_FW_BRANCH; + tw_dev->tw_compat_info.driver_build_low = TW_BASE_FW_BUILD; + tw_dev->tw_compat_info.fw_on_ctlr_srl = fw_on_ctlr_srl; + tw_dev->tw_compat_info.fw_on_ctlr_branch = fw_on_ctlr_branch; + tw_dev->tw_compat_info.fw_on_ctlr_build = fw_on_ctlr_build; + retval = 0; out: return retval; @@ -631,7 +647,7 @@ static int twa_chrdev_ioctl(struct inode *inode, struct file *file, unsigned int goto out2; /* Check data buffer size */ - if (driver_command.buffer_length > TW_MAX_SECTORS * 512) { + if (driver_command.buffer_length > TW_MAX_SECTORS * 2048) { retval = TW_IOCTL_ERROR_OS_EINVAL; goto out2; } @@ -680,13 +696,6 @@ static int twa_chrdev_ioctl(struct inode *inode, struct file *file, unsigned int /* Now wait for command to complete */ timeout = wait_event_timeout(tw_dev->ioctl_wqueue, tw_dev->chrdev_request_id == TW_IOCTL_CHRDEV_FREE, timeout); - /* See if we reset while waiting for the ioctl to complete */ - if (test_bit(TW_IN_RESET, &tw_dev->flags)) { - clear_bit(TW_IN_RESET, &tw_dev->flags); - retval = TW_IOCTL_ERROR_OS_ERESTARTSYS; - goto out3; - } - /* We timed out, and didn't get an interrupt */ if (tw_dev->chrdev_request_id != TW_IOCTL_CHRDEV_FREE) { /* Now we need to reset the board */ @@ -694,11 +703,6 @@ static int twa_chrdev_ioctl(struct inode *inode, struct file *file, unsigned int tw_dev->host->host_no, TW_DRIVER, 0xc, cmd); retval = TW_IOCTL_ERROR_OS_EIO; - spin_lock_irqsave(tw_dev->host->host_lock, flags); - tw_dev->state[request_id] = TW_S_COMPLETED; - twa_free_request_id(tw_dev, request_id); - tw_dev->posted_request_count--; - spin_unlock_irqrestore(tw_dev->host->host_lock, flags); twa_reset_device_extension(tw_dev, 1); goto out3; } @@ -717,16 +721,7 @@ static int twa_chrdev_ioctl(struct inode *inode, struct file *file, unsigned int tw_ioctl->driver_command.status = 0; /* Copy compatiblity struct into ioctl data buffer */ tw_compat_info = (TW_Compatibility_Info *)tw_ioctl->data_buffer; - strncpy(tw_compat_info->driver_version, TW_DRIVER_VERSION, strlen(TW_DRIVER_VERSION)); - tw_compat_info->working_srl = tw_dev->working_srl; - tw_compat_info->working_branch = tw_dev->working_branch; - tw_compat_info->working_build = tw_dev->working_build; - tw_compat_info->driver_srl_high = TW_CURRENT_DRIVER_SRL; - tw_compat_info->driver_branch_high = TW_CURRENT_DRIVER_BRANCH; - tw_compat_info->driver_build_high = TW_CURRENT_DRIVER_BUILD; - tw_compat_info->driver_srl_low = TW_BASE_FW_SRL; - tw_compat_info->driver_branch_low = TW_BASE_FW_BRANCH; - tw_compat_info->driver_build_low = TW_BASE_FW_BUILD; + memcpy(tw_compat_info, &tw_dev->tw_compat_info, sizeof(TW_Compatibility_Info)); break; case TW_IOCTL_GET_LAST_EVENT: if (tw_dev->event_queue_wrapped) { @@ -895,7 +890,8 @@ static int twa_decode_bits(TW_Device_Extension *tw_dev, u32 status_reg_value) } if (status_reg_value & TW_STATUS_QUEUE_ERROR) { - TW_PRINTK(tw_dev->host, TW_DRIVER, 0xe, "Controller Queue Error: clearing"); + if ((tw_dev->tw_pci_dev->device != PCI_DEVICE_ID_3WARE_9650SE) || (!test_bit(TW_IN_RESET, &tw_dev->flags))) + TW_PRINTK(tw_dev->host, TW_DRIVER, 0xe, "Controller Queue Error: clearing"); writel(TW_CONTROL_CLEAR_QUEUE_ERROR, TW_CONTROL_REG_ADDR(tw_dev)); } @@ -939,10 +935,12 @@ static int twa_empty_response_queue_large(TW_Device_Extension *tw_dev) unsigned long before; int retval = 1; - if (tw_dev->tw_pci_dev->device == PCI_DEVICE_ID_3WARE_9550SX) { + if ((tw_dev->tw_pci_dev->device == PCI_DEVICE_ID_3WARE_9550SX) || + (tw_dev->tw_pci_dev->device == PCI_DEVICE_ID_3WARE_9650SE)) { before = jiffies; while ((response_que_value & TW_9550SX_DRAIN_COMPLETED) != TW_9550SX_DRAIN_COMPLETED) { response_que_value = readl(TW_RESPONSE_QUEUE_REG_ADDR_LARGE(tw_dev)); + msleep(1); if (time_after(jiffies, before + HZ * 30)) goto out; } @@ -1214,6 +1212,10 @@ static irqreturn_t twa_interrupt(int irq, void *dev_instance) handled = 1; + /* If we are resetting, bail */ + if (test_bit(TW_IN_RESET, &tw_dev->flags)) + goto twa_interrupt_bail; + /* Check controller for errors */ if (twa_check_bits(status_reg_value)) { if (twa_decode_bits(tw_dev, status_reg_value)) { @@ -1355,8 +1357,8 @@ static void twa_load_sgl(TW_Command_Full *full_command_packet, int request_id, d if (TW_OP_OUT(full_command_packet->command.newcommand.opcode__reserved) == TW_OP_EXECUTE_SCSI) { newcommand = &full_command_packet->command.newcommand; - newcommand->request_id__lunl = - TW_REQ_LUN_IN(TW_LUN_OUT(newcommand->request_id__lunl), request_id); + newcommand->request_id__lunl = + cpu_to_le16(TW_REQ_LUN_IN(TW_LUN_OUT(newcommand->request_id__lunl), request_id)); newcommand->sg_list[0].address = TW_CPU_TO_SGL(dma_handle + sizeof(TW_Ioctl_Buf_Apache) - 1); newcommand->sg_list[0].length = cpu_to_le32(length); newcommand->sgl_entries__lunh = @@ -1531,6 +1533,13 @@ static int twa_post_command_packet(TW_Device_Extension *tw_dev, int request_id, int retval = 1; command_que_value = tw_dev->command_packet_phys[request_id]; + + /* For 9650SE write low 4 bytes first */ + if (tw_dev->tw_pci_dev->device == PCI_DEVICE_ID_3WARE_9650SE) { + command_que_value += TW_COMMAND_OFFSET; + writel((u32)command_que_value, TW_COMMAND_QUEUE_REG_ADDR_LARGE(tw_dev)); + } + status_reg_value = readl(TW_STATUS_REG_ADDR(tw_dev)); if (twa_check_bits(status_reg_value)) @@ -1557,13 +1566,17 @@ static int twa_post_command_packet(TW_Device_Extension *tw_dev, int request_id, TW_UNMASK_COMMAND_INTERRUPT(tw_dev); goto out; } else { - /* We successfully posted the command packet */ - if (sizeof(dma_addr_t) > 4) { - command_que_value += TW_COMMAND_OFFSET; - writel((u32)command_que_value, TW_COMMAND_QUEUE_REG_ADDR(tw_dev)); - writel((u32)((u64)command_que_value >> 32), TW_COMMAND_QUEUE_REG_ADDR(tw_dev) + 0x4); + if (tw_dev->tw_pci_dev->device == PCI_DEVICE_ID_3WARE_9650SE) { + /* Now write upper 4 bytes */ + writel((u32)((u64)command_que_value >> 32), TW_COMMAND_QUEUE_REG_ADDR_LARGE(tw_dev) + 0x4); } else { - writel(TW_COMMAND_OFFSET + command_que_value, TW_COMMAND_QUEUE_REG_ADDR(tw_dev)); + if (sizeof(dma_addr_t) > 4) { + command_que_value += TW_COMMAND_OFFSET; + writel((u32)command_que_value, TW_COMMAND_QUEUE_REG_ADDR(tw_dev)); + writel((u32)((u64)command_que_value >> 32), TW_COMMAND_QUEUE_REG_ADDR(tw_dev) + 0x4); + } else { + writel(TW_COMMAND_OFFSET + command_que_value, TW_COMMAND_QUEUE_REG_ADDR(tw_dev)); + } } tw_dev->state[request_id] = TW_S_POSTED; tw_dev->posted_request_count++; @@ -1620,14 +1633,9 @@ static int twa_reset_device_extension(TW_Device_Extension *tw_dev, int ioctl_res goto out; TW_ENABLE_AND_CLEAR_INTERRUPTS(tw_dev); + clear_bit(TW_IN_RESET, &tw_dev->flags); + tw_dev->chrdev_request_id = TW_IOCTL_CHRDEV_FREE; - /* Wake up any ioctl that was pending before the reset */ - if ((tw_dev->chrdev_request_id == TW_IOCTL_CHRDEV_FREE) || (ioctl_reset)) { - clear_bit(TW_IN_RESET, &tw_dev->flags); - } else { - tw_dev->chrdev_request_id = TW_IOCTL_CHRDEV_FREE; - wake_up(&tw_dev->ioctl_wqueue); - } retval = 0; out: return retval; @@ -1736,6 +1744,9 @@ static int twa_scsi_eh_reset(struct scsi_cmnd *SCpnt) "WARNING: (0x%02X:0x%04X): Command (0x%x) timed out, resetting card.\n", TW_DRIVER, 0x2c, SCpnt->cmnd[0]); + /* Make sure we are not issuing an ioctl or resetting from ioctl */ + mutex_lock(&tw_dev->ioctl_lock); + /* Now reset the card and some of the device extension data */ if (twa_reset_device_extension(tw_dev, 0)) { TW_PRINTK(tw_dev->host, TW_DRIVER, 0x2b, "Controller reset failed during scsi host reset"); @@ -1744,6 +1755,7 @@ static int twa_scsi_eh_reset(struct scsi_cmnd *SCpnt) retval = SUCCESS; out: + mutex_unlock(&tw_dev->ioctl_lock); return retval; } /* End twa_scsi_eh_reset() */ @@ -1753,8 +1765,14 @@ static int twa_scsi_queue(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_cmnd int request_id, retval; TW_Device_Extension *tw_dev = (TW_Device_Extension *)SCpnt->device->host->hostdata; + /* If we are resetting due to timed out ioctl, report as busy */ + if (test_bit(TW_IN_RESET, &tw_dev->flags)) { + retval = SCSI_MLQUEUE_HOST_BUSY; + goto out; + } + /* Check if this FW supports luns */ - if ((SCpnt->device->lun != 0) && (tw_dev->working_srl < TW_FW_SRL_LUNS_SUPPORTED)) { + if ((SCpnt->device->lun != 0) && (tw_dev->tw_compat_info.working_srl < TW_FW_SRL_LUNS_SUPPORTED)) { SCpnt->result = (DID_BAD_TARGET << 16); done(SCpnt); retval = 0; @@ -1960,6 +1978,9 @@ static void __twa_shutdown(TW_Device_Extension *tw_dev) /* Disable interrupts */ TW_DISABLE_INTERRUPTS(tw_dev); + /* Free up the IRQ */ + free_irq(tw_dev->tw_pci_dev->irq, tw_dev); + printk(KERN_WARNING "3w-9xxx: Shutting down host %d.\n", tw_dev->host->host_no); /* Tell the card we are shutting down */ @@ -2091,21 +2112,25 @@ static int __devinit twa_probe(struct pci_dev *pdev, const struct pci_device_id /* Initialize the card */ if (twa_reset_sequence(tw_dev, 0)) - goto out_release_mem_region; + goto out_iounmap; /* Set host specific parameters */ - host->max_id = TW_MAX_UNITS; + if (pdev->device == PCI_DEVICE_ID_3WARE_9650SE) + host->max_id = TW_MAX_UNITS_9650SE; + else + host->max_id = TW_MAX_UNITS; + host->max_cmd_len = TW_MAX_CDB_LEN; /* Channels aren't supported by adapter */ - host->max_lun = TW_MAX_LUNS(tw_dev->working_srl); + host->max_lun = TW_MAX_LUNS(tw_dev->tw_compat_info.working_srl); host->max_channel = 0; /* Register the card with the kernel SCSI layer */ retval = scsi_add_host(host, &pdev->dev); if (retval) { TW_PRINTK(tw_dev->host, TW_DRIVER, 0x27, "scsi add host failed"); - goto out_release_mem_region; + goto out_iounmap; } pci_set_drvdata(pdev, host); @@ -2145,6 +2170,8 @@ static int __devinit twa_probe(struct pci_dev *pdev, const struct pci_device_id out_remove_host: scsi_remove_host(host); +out_iounmap: + iounmap(tw_dev->base_addr); out_release_mem_region: pci_release_regions(pdev); out_free_device_extension: @@ -2170,12 +2197,12 @@ static void twa_remove(struct pci_dev *pdev) twa_major = -1; } - /* Free up the IRQ */ - free_irq(tw_dev->tw_pci_dev->irq, tw_dev); - /* Shutdown the card */ __twa_shutdown(tw_dev); + /* Free IO remapping */ + iounmap(tw_dev->base_addr); + /* Free up the mem region */ pci_release_regions(pdev); @@ -2193,6 +2220,8 @@ static struct pci_device_id twa_pci_tbl[] __devinitdata = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { PCI_VENDOR_ID_3WARE, PCI_DEVICE_ID_3WARE_9550SX, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_3WARE, PCI_DEVICE_ID_3WARE_9650SE, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { } }; MODULE_DEVICE_TABLE(pci, twa_pci_tbl); diff --git a/drivers/scsi/3w-9xxx.h b/drivers/scsi/3w-9xxx.h index e5685be..7901517 100644 --- a/drivers/scsi/3w-9xxx.h +++ b/drivers/scsi/3w-9xxx.h @@ -289,7 +289,6 @@ static twa_message_type twa_error_table[] = { #define TW_STATUS_VALID_INTERRUPT 0x00DF0000 /* PCI related defines */ -#define TW_NUMDEVICES 1 #define TW_PCI_CLEAR_PARITY_ERRORS 0xc100 #define TW_PCI_CLEAR_PCI_ABORT 0x2000 @@ -335,6 +334,7 @@ static twa_message_type twa_error_table[] = { #define TW_ALIGNMENT_9000 4 /* 4 bytes */ #define TW_ALIGNMENT_9000_SGL 0x3 #define TW_MAX_UNITS 16 +#define TW_MAX_UNITS_9650SE 32 #define TW_INIT_MESSAGE_CREDITS 0x100 #define TW_INIT_COMMAND_PACKET_SIZE 0x3 #define TW_INIT_COMMAND_PACKET_SIZE_EXTENDED 0x6 @@ -354,7 +354,6 @@ static twa_message_type twa_error_table[] = { #define TW_MAX_RESPONSE_DRAIN 256 #define TW_MAX_AEN_DRAIN 40 #define TW_IN_RESET 2 -#define TW_IN_CHRDEV_IOCTL 3 #define TW_IN_ATTENTION_LOOP 4 #define TW_MAX_SECTORS 256 #define TW_AEN_WAIT_TIME 1000 @@ -417,6 +416,9 @@ static twa_message_type twa_error_table[] = { #ifndef PCI_DEVICE_ID_3WARE_9550SX #define PCI_DEVICE_ID_3WARE_9550SX 0x1003 #endif +#ifndef PCI_DEVICE_ID_3WARE_9650SE +#define PCI_DEVICE_ID_3WARE_9650SE 0x1004 +#endif /* Bitmask macros to eliminate bitfields */ @@ -442,6 +444,7 @@ static twa_message_type twa_error_table[] = { #define TW_CONTROL_REG_ADDR(x) (x->base_addr) #define TW_STATUS_REG_ADDR(x) ((unsigned char __iomem *)x->base_addr + 0x4) #define TW_COMMAND_QUEUE_REG_ADDR(x) (sizeof(dma_addr_t) > 4 ? ((unsigned char __iomem *)x->base_addr + 0x20) : ((unsigned char __iomem *)x->base_addr + 0x8)) +#define TW_COMMAND_QUEUE_REG_ADDR_LARGE(x) ((unsigned char __iomem *)x->base_addr + 0x20) #define TW_RESPONSE_QUEUE_REG_ADDR(x) ((unsigned char __iomem *)x->base_addr + 0xC) #define TW_RESPONSE_QUEUE_REG_ADDR_LARGE(x) ((unsigned char __iomem *)x->base_addr + 0x30) #define TW_CLEAR_ALL_INTERRUPTS(x) (writel(TW_STATUS_VALID_INTERRUPT, TW_CONTROL_REG_ADDR(x))) @@ -626,6 +629,9 @@ typedef struct TAG_TW_Compatibility_Info unsigned short driver_srl_low; unsigned short driver_branch_low; unsigned short driver_build_low; + unsigned short fw_on_ctlr_srl; + unsigned short fw_on_ctlr_branch; + unsigned short fw_on_ctlr_build; } TW_Compatibility_Info; #pragma pack() @@ -668,9 +674,7 @@ typedef struct TAG_TW_Device_Extension { wait_queue_head_t ioctl_wqueue; struct mutex ioctl_lock; char aen_clobber; - unsigned short working_srl; - unsigned short working_branch; - unsigned short working_build; + TW_Compatibility_Info tw_compat_info; } TW_Device_Extension; #endif /* _3W_9XXX_H */ -- cgit v0.10.2 From 7ca63cb470f23a197f187afe936d4bf806197d6e Mon Sep 17 00:00:00 2001 From: Douglas Gilbert Date: Fri, 27 Oct 2006 17:47:49 -0400 Subject: [SCSI] sg: fix incorrect last scatg length For certain LLDs the sg driver can cause on oops when the transfer length is large and not a multiple of PAGE_SIZE. ChangeLog: - correct the length of the last scatter gather list element. - fix some printk()s that have the wrong function name. Signed-off-by: Douglas Gilbert Signed-off-by: James Bottomley diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 3f8b931..81e3bc7 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -60,7 +60,7 @@ static int sg_version_num = 30534; /* 2 digits for each component */ #ifdef CONFIG_SCSI_PROC_FS #include -static char *sg_version_date = "20060920"; +static char *sg_version_date = "20061027"; static int sg_proc_init(void); static void sg_proc_cleanup(void); @@ -710,12 +710,12 @@ sg_common_write(Sg_fd * sfp, Sg_request * srp, (int) cmnd[0], (int) hp->cmd_len)); if ((k = sg_start_req(srp))) { - SCSI_LOG_TIMEOUT(1, printk("sg_write: start_req err=%d\n", k)); + SCSI_LOG_TIMEOUT(1, printk("sg_common_write: start_req err=%d\n", k)); sg_finish_rem_req(srp); return k; /* probably out of space --> ENOMEM */ } if ((k = sg_write_xfer(srp))) { - SCSI_LOG_TIMEOUT(1, printk("sg_write: write_xfer, bad address\n")); + SCSI_LOG_TIMEOUT(1, printk("sg_common_write: write_xfer, bad address\n")); sg_finish_rem_req(srp); return k; } @@ -746,7 +746,7 @@ sg_common_write(Sg_fd * sfp, Sg_request * srp, hp->dxfer_len, srp->data.k_use_sg, timeout, SG_DEFAULT_RETRIES, srp, sg_cmd_done, GFP_ATOMIC)) { - SCSI_LOG_TIMEOUT(1, printk("sg_write: scsi_execute_async failed\n")); + SCSI_LOG_TIMEOUT(1, printk("sg_common_write: scsi_execute_async failed\n")); /* * most likely out of mem, but could also be a bad map */ @@ -1283,7 +1283,7 @@ sg_cmd_done(void *data, char *sense, int result, int resid) sg_finish_rem_req(srp); srp = NULL; if (NULL == sfp->headrp) { - SCSI_LOG_TIMEOUT(1, printk("sg...bh: already closed, final cleanup\n")); + SCSI_LOG_TIMEOUT(1, printk("sg_cmd_done: already closed, final cleanup\n")); if (0 == sg_remove_sfp(sdp, sfp)) { /* device still present */ scsi_device_put(sdp->device); } @@ -1512,12 +1512,12 @@ sg_remove(struct class_device *cl_dev, struct class_interface *cl_intf) POLL_HUP); } } - SCSI_LOG_TIMEOUT(3, printk("sg_detach: dev=%d, dirty\n", k)); + SCSI_LOG_TIMEOUT(3, printk("sg_remove: dev=%d, dirty\n", k)); if (NULL == sdp->headfp) { sg_dev_arr[k] = NULL; } } else { /* nothing active, simple case */ - SCSI_LOG_TIMEOUT(3, printk("sg_detach: dev=%d\n", k)); + SCSI_LOG_TIMEOUT(3, printk("sg_remove: dev=%d\n", k)); sg_dev_arr[k] = NULL; } sg_nr_dev--; @@ -1876,14 +1876,15 @@ sg_build_indirect(Sg_scatter_hold * schp, Sg_fd * sfp, int buff_size) } } sg->page = p; - sg->length = ret_sz; + sg->length = (ret_sz > num) ? num : ret_sz; - SCSI_LOG_TIMEOUT(5, printk("sg_build_build: k=%d, a=0x%p, len=%d\n", - k, p, ret_sz)); + SCSI_LOG_TIMEOUT(5, printk("sg_build_indirect: k=%d, num=%d, " + "ret_sz=%d\n", k, num, ret_sz)); } /* end of for loop */ schp->k_use_sg = k; - SCSI_LOG_TIMEOUT(5, printk("sg_build_indirect: k_use_sg=%d, rem_sz=%d\n", k, rem_sz)); + SCSI_LOG_TIMEOUT(5, printk("sg_build_indirect: k_use_sg=%d, " + "rem_sz=%d\n", k, rem_sz)); schp->bufflen = blk_size; if (rem_sz > 0) /* must have failed */ @@ -2014,7 +2015,7 @@ sg_remove_scat(Sg_scatter_hold * schp) for (k = 0; (k < schp->k_use_sg) && sg->page; ++k, ++sg) { SCSI_LOG_TIMEOUT(5, printk( - "sg_remove_scat: k=%d, a=0x%p, len=%d\n", + "sg_remove_scat: k=%d, pg=0x%p, len=%d\n", k, sg->page, sg->length)); sg_page_free(sg->page, sg->length); } -- cgit v0.10.2 From 25a122fd0d28b48782b9524a85895573e7ccf304 Mon Sep 17 00:00:00 2001 From: Timo Teras Date: Wed, 25 Oct 2006 09:37:41 +0300 Subject: MMC: Poll card status after rescanning cards Some broken cards seem to process CMD1 even in stand-by state. The result is that the card replies with ILLEGAL_COMMAND error for the next command sent after rescanning. Currently the next command is select card, which would return the error. But CMD7 does actually succeed and retries of the command will timeout. The workaround is to poll card status after CMD1 to clear the pending error. Signed-off-by: Timo Teras Signed-off-by: Pierre Ossman diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index ee8863c..ec8168a 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -1178,14 +1178,29 @@ static void mmc_rescan(void *data) { struct mmc_host *host = data; struct list_head *l, *n; + unsigned char power_mode; mmc_claim_host(host); - if (host->ios.power_mode == MMC_POWER_ON) + /* + * Check for removed cards and newly inserted ones. We check for + * removed cards first so we can intelligently re-select the VDD. + */ + power_mode = host->ios.power_mode; + if (power_mode == MMC_POWER_ON) mmc_check_cards(host); mmc_setup(host); + /* + * Some broken cards process CMD1 even in stand-by state. There is + * no reply, but an ILLEGAL_COMMAND error is cached and returned + * after next command. We poll for card status here to clear any + * possibly pending error. + */ + if (power_mode == MMC_POWER_ON) + mmc_check_cards(host); + if (!list_empty(&host->cards)) { /* * (Re-)calculate the fastest clock rate which the -- cgit v0.10.2 From 63ef731aa6a81e286de78dcc92241d123424ed39 Mon Sep 17 00:00:00 2001 From: Timo Teras Date: Thu, 2 Nov 2006 19:43:27 +0100 Subject: MMC: Do not set unsupported bits in OCR response The card might go to inactive state (according to specification), if there are unsupported bits set in the OCR. Signed-off-by: Timo Teras Signed-off-by: Pierre Ossman diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index ec8168a..766bc544 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -475,7 +475,7 @@ static u32 mmc_select_voltage(struct mmc_host *host, u32 ocr) if (bit) { bit -= 1; - ocr = 3 << bit; + ocr &= 3 << bit; host->ios.vdd = bit; mmc_set_ios(host); -- cgit v0.10.2 From 8cc05f79d2a4e33d5cb621f36a448d5157f46893 Mon Sep 17 00:00:00 2001 From: Vitaly Wool Date: Tue, 17 Oct 2006 11:16:22 +0100 Subject: [ARM] 3857/2: pnx4008: add devices' registration This patch adds platform devices' registration for the devices which drivers either have been added to the mainline or on the way to. arch/arm/mach-pnx4008/core.c | 69 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 69 insertions(+) Signed-off-by: Vitaly Wool Signed-off-by: Russell King diff --git a/arch/arm/mach-pnx4008/core.c b/arch/arm/mach-pnx4008/core.c index 3d73c1e..429c796 100644 --- a/arch/arm/mach-pnx4008/core.c +++ b/arch/arm/mach-pnx4008/core.c @@ -133,10 +133,79 @@ static struct platform_device serial_device = { }, }; +static struct platform_device nand_flash_device = { + .name = "pnx4008-flash", + .id = -1, + .dev = { + .coherent_dma_mask = 0xFFFFFFFF, + }, +}; + +/* The dmamask must be set for OHCI to work */ +static u64 ohci_dmamask = ~(u32) 0; + +static struct resource ohci_resources[] = { + { + .start = IO_ADDRESS(PNX4008_USB_CONFIG_BASE), + .end = IO_ADDRESS(PNX4008_USB_CONFIG_BASE + 0x100), + .flags = IORESOURCE_MEM, + }, { + .start = USB_HOST_INT, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device ohci_device = { + .name = "pnx4008-usb-ohci", + .id = -1, + .dev = { + .dma_mask = &ohci_dmamask, + .coherent_dma_mask = 0xffffffff, + }, + .num_resources = ARRAY_SIZE(ohci_resources), + .resource = ohci_resources, +}; + +static struct platform_device sdum_device = { + .name = "pnx4008-sdum", + .id = 0, + .dev = { + .coherent_dma_mask = 0xffffffff, + }, +}; + +static struct platform_device rgbfb_device = { + .name = "pnx4008-rgbfb", + .id = 0, + .dev = { + .coherent_dma_mask = 0xffffffff, + } +}; + +struct resource watchdog_resources[] = { + { + .start = PNX4008_WDOG_BASE, + .end = PNX4008_WDOG_BASE + SZ_4K - 1, + .flags = IORESOURCE_MEM, + }, +}; + +static struct platform_device watchdog_device = { + .name = "pnx4008-watchdog", + .id = -1, + .num_resources = ARRAY_SIZE(watchdog_resources), + .resource = watchdog_resources, +}; + static struct platform_device *devices[] __initdata = { &spipnx_1, &spipnx_2, &serial_device, + &ohci_device, + &nand_flash_device, + &sdum_device, + &rgbfb_device, + &watchdog_device, }; -- cgit v0.10.2 From 7e28db5d8ff63b1cabc221c5cb84a5f45752f1c2 Mon Sep 17 00:00:00 2001 From: Hoang-Nam Nguyen Date: Tue, 7 Nov 2006 00:56:39 +0100 Subject: IB/ehca: Assure 4K alignment for firmware control blocks Assure 4K alignment for firmware control blocks in 64K page mode, because kzalloc()'s result address might not be 4K aligned if 64K pages are enabled. Thus, we introduce wrappers called ehca_{alloc,free}_fw_ctrlblock(), which use a slab cache for objects with 4K length and 4K alignment in order to alloc/free firmware control blocks in 64K page mode. In 4K page mode those wrappers just are defines of get_zeroed_page() and free_page(). Signed-off-by: Hoang-Nam Nguyen Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/ehca/ehca_hca.c b/drivers/infiniband/hw/ehca/ehca_hca.c index 5eae6ac..e1b618c 100644 --- a/drivers/infiniband/hw/ehca/ehca_hca.c +++ b/drivers/infiniband/hw/ehca/ehca_hca.c @@ -40,6 +40,7 @@ */ #include "ehca_tools.h" +#include "ehca_iverbs.h" #include "hcp_if.h" int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props) @@ -49,7 +50,7 @@ int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props) ib_device); struct hipz_query_hca *rblock; - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_err(&shca->ib_device, "Can't allocate rblock memory."); return -ENOMEM; @@ -96,7 +97,7 @@ int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props) = min_t(int, rblock->max_total_mcast_qp_attach, INT_MAX); query_device1: - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); return ret; } @@ -109,7 +110,7 @@ int ehca_query_port(struct ib_device *ibdev, ib_device); struct hipz_query_port *rblock; - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_err(&shca->ib_device, "Can't allocate rblock memory."); return -ENOMEM; @@ -162,7 +163,7 @@ int ehca_query_port(struct ib_device *ibdev, props->active_speed = 0x1; query_port1: - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); return ret; } @@ -178,7 +179,7 @@ int ehca_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 *pkey) return -EINVAL; } - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_err(&shca->ib_device, "Can't allocate rblock memory."); return -ENOMEM; @@ -193,7 +194,7 @@ int ehca_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 *pkey) memcpy(pkey, &rblock->pkey_entries + index, sizeof(u16)); query_pkey1: - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); return ret; } @@ -211,7 +212,7 @@ int ehca_query_gid(struct ib_device *ibdev, u8 port, return -EINVAL; } - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_err(&shca->ib_device, "Can't allocate rblock memory."); return -ENOMEM; @@ -227,7 +228,7 @@ int ehca_query_gid(struct ib_device *ibdev, u8 port, memcpy(&gid->raw[8], &rblock->guid_entries[index], sizeof(u64)); query_gid1: - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); return ret; } diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c index 048cc44..c3ea746 100644 --- a/drivers/infiniband/hw/ehca/ehca_irq.c +++ b/drivers/infiniband/hw/ehca/ehca_irq.c @@ -45,6 +45,7 @@ #include "ehca_tools.h" #include "hcp_if.h" #include "hipz_fns.h" +#include "ipz_pt_fn.h" #define EQE_COMPLETION_EVENT EHCA_BMASK_IBM(1,1) #define EQE_CQ_QP_NUMBER EHCA_BMASK_IBM(8,31) @@ -137,38 +138,36 @@ int ehca_error_data(struct ehca_shca *shca, void *data, u64 *rblock; unsigned long block_count; - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_err(&shca->ib_device, "Cannot allocate rblock memory."); ret = -ENOMEM; goto error_data1; } + /* rblock must be 4K aligned and should be 4K large */ ret = hipz_h_error_data(shca->ipz_hca_handle, resource, rblock, &block_count); - if (ret == H_R_STATE) { + if (ret == H_R_STATE) ehca_err(&shca->ib_device, "No error data is available: %lx.", resource); - } else if (ret == H_SUCCESS) { int length; length = EHCA_BMASK_GET(ERROR_DATA_LENGTH, rblock[0]); - if (length > PAGE_SIZE) - length = PAGE_SIZE; + if (length > EHCA_PAGESIZE) + length = EHCA_PAGESIZE; print_error_data(shca, data, rblock, length); - } - else { + } else ehca_err(&shca->ib_device, "Error data could not be fetched: %lx", resource); - } - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); error_data1: return ret; diff --git a/drivers/infiniband/hw/ehca/ehca_iverbs.h b/drivers/infiniband/hw/ehca/ehca_iverbs.h index 319c39d..3720e30 100644 --- a/drivers/infiniband/hw/ehca/ehca_iverbs.h +++ b/drivers/infiniband/hw/ehca/ehca_iverbs.h @@ -179,4 +179,12 @@ int ehca_mmap_register(u64 physical,void **mapped, int ehca_munmap(unsigned long addr, size_t len); +#ifdef CONFIG_PPC_64K_PAGES +void *ehca_alloc_fw_ctrlblock(void); +void ehca_free_fw_ctrlblock(void *ptr); +#else +#define ehca_alloc_fw_ctrlblock() ((void *) get_zeroed_page(GFP_KERNEL)) +#define ehca_free_fw_ctrlblock(ptr) free_page((unsigned long)(ptr)) +#endif + #endif diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c index 024d511..01f5aa9 100644 --- a/drivers/infiniband/hw/ehca/ehca_main.c +++ b/drivers/infiniband/hw/ehca/ehca_main.c @@ -40,6 +40,9 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#ifdef CONFIG_PPC_64K_PAGES +#include +#endif #include "ehca_classes.h" #include "ehca_iverbs.h" #include "ehca_mrmw.h" @@ -49,7 +52,7 @@ MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Christoph Raisch "); MODULE_DESCRIPTION("IBM eServer HCA InfiniBand Device Driver"); -MODULE_VERSION("SVNEHCA_0017"); +MODULE_VERSION("SVNEHCA_0018"); int ehca_open_aqp1 = 0; int ehca_debug_level = 0; @@ -94,11 +97,31 @@ spinlock_t ehca_cq_idr_lock; DEFINE_IDR(ehca_qp_idr); DEFINE_IDR(ehca_cq_idr); + static struct list_head shca_list; /* list of all registered ehcas */ static spinlock_t shca_list_lock; static struct timer_list poll_eqs_timer; +#ifdef CONFIG_PPC_64K_PAGES +static struct kmem_cache *ctblk_cache = NULL; + +void *ehca_alloc_fw_ctrlblock(void) +{ + void *ret = kmem_cache_zalloc(ctblk_cache, SLAB_KERNEL); + if (!ret) + ehca_gen_err("Out of memory for ctblk"); + return ret; +} + +void ehca_free_fw_ctrlblock(void *ptr) +{ + if (ptr) + kmem_cache_free(ctblk_cache, ptr); + +} +#endif + static int ehca_create_slab_caches(void) { int ret; @@ -133,6 +156,17 @@ static int ehca_create_slab_caches(void) goto create_slab_caches5; } +#ifdef CONFIG_PPC_64K_PAGES + ctblk_cache = kmem_cache_create("ehca_cache_ctblk", + EHCA_PAGESIZE, H_CB_ALIGNMENT, + SLAB_HWCACHE_ALIGN, + NULL, NULL); + if (!ctblk_cache) { + ehca_gen_err("Cannot create ctblk SLAB cache."); + ehca_cleanup_mrmw_cache(); + goto create_slab_caches5; + } +#endif return 0; create_slab_caches5: @@ -157,6 +191,10 @@ static void ehca_destroy_slab_caches(void) ehca_cleanup_qp_cache(); ehca_cleanup_cq_cache(); ehca_cleanup_pd_cache(); +#ifdef CONFIG_PPC_64K_PAGES + if (ctblk_cache) + kmem_cache_destroy(ctblk_cache); +#endif } #define EHCA_HCAAVER EHCA_BMASK_IBM(32,39) @@ -168,7 +206,7 @@ int ehca_sense_attributes(struct ehca_shca *shca) u64 h_ret; struct hipz_query_hca *rblock; - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_gen_err("Cannot allocate rblock memory."); return -ENOMEM; @@ -211,7 +249,7 @@ int ehca_sense_attributes(struct ehca_shca *shca) shca->sport[1].rate = IB_RATE_30_GBPS; num_ports1: - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); return ret; } @@ -220,7 +258,7 @@ static int init_node_guid(struct ehca_shca *shca) int ret = 0; struct hipz_query_hca *rblock; - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + rblock = ehca_alloc_fw_ctrlblock(); if (!rblock) { ehca_err(&shca->ib_device, "Can't allocate rblock memory."); return -ENOMEM; @@ -235,7 +273,7 @@ static int init_node_guid(struct ehca_shca *shca) memcpy(&shca->ib_device.node_guid, &rblock->node_guid, sizeof(u64)); init_node_guid1: - kfree(rblock); + ehca_free_fw_ctrlblock(rblock); return ret; } @@ -431,7 +469,7 @@ static ssize_t ehca_show_##name(struct device *dev, \ \ shca = dev->driver_data; \ \ - rblock = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); \ + rblock = ehca_alloc_fw_ctrlblock(); \ if (!rblock) { \ dev_err(dev, "Can't allocate rblock memory."); \ return 0; \ @@ -439,12 +477,12 @@ static ssize_t ehca_show_##name(struct device *dev, \ \ if (hipz_h_query_hca(shca->ipz_hca_handle, rblock) != H_SUCCESS) { \ dev_err(dev, "Can't query device properties"); \ - kfree(rblock); \ + ehca_free_fw_ctrlblock(rblock); \ return 0; \ } \ \ data = rblock->name; \ - kfree(rblock); \ + ehca_free_fw_ctrlblock(rblock); \ \ if ((strcmp(#name, "num_ports") == 0) && (ehca_nr_ports == 1)) \ return snprintf(buf, 256, "1\n"); \ @@ -752,7 +790,7 @@ int __init ehca_module_init(void) int ret; printk(KERN_INFO "eHCA Infiniband Device Driver " - "(Rel.: SVNEHCA_0017)\n"); + "(Rel.: SVNEHCA_0018)\n"); idr_init(&ehca_qp_idr); idr_init(&ehca_cq_idr); spin_lock_init(&ehca_qp_idr_lock); diff --git a/drivers/infiniband/hw/ehca/ehca_mrmw.c b/drivers/infiniband/hw/ehca/ehca_mrmw.c index 5ca6544..abce676 100644 --- a/drivers/infiniband/hw/ehca/ehca_mrmw.c +++ b/drivers/infiniband/hw/ehca/ehca_mrmw.c @@ -1013,7 +1013,7 @@ int ehca_reg_mr_rpages(struct ehca_shca *shca, u32 i; u64 *kpage; - kpage = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + kpage = ehca_alloc_fw_ctrlblock(); if (!kpage) { ehca_err(&shca->ib_device, "kpage alloc failed"); ret = -ENOMEM; @@ -1092,7 +1092,7 @@ int ehca_reg_mr_rpages(struct ehca_shca *shca, ehca_reg_mr_rpages_exit1: - kfree(kpage); + ehca_free_fw_ctrlblock(kpage); ehca_reg_mr_rpages_exit0: if (ret) ehca_err(&shca->ib_device, "ret=%x shca=%p e_mr=%p pginfo=%p " @@ -1124,7 +1124,7 @@ inline int ehca_rereg_mr_rereg1(struct ehca_shca *shca, ehca_mrmw_map_acl(acl, &hipz_acl); ehca_mrmw_set_pgsize_hipz_acl(&hipz_acl); - kpage = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); + kpage = ehca_alloc_fw_ctrlblock(); if (!kpage) { ehca_err(&shca->ib_device, "kpage alloc failed"); ret = -ENOMEM; @@ -1181,7 +1181,7 @@ inline int ehca_rereg_mr_rereg1(struct ehca_shca *shca, } ehca_rereg_mr_rereg1_exit1: - kfree(kpage); + ehca_free_fw_ctrlblock(kpage); ehca_rereg_mr_rereg1_exit0: if ( ret && (ret != -EAGAIN) ) ehca_err(&shca->ib_device, "ret=%x lkey=%x rkey=%x " diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c index 4394123..cf3e50e 100644 --- a/drivers/infiniband/hw/ehca/ehca_qp.c +++ b/drivers/infiniband/hw/ehca/ehca_qp.c @@ -811,8 +811,8 @@ static int internal_modify_qp(struct ib_qp *ibqp, unsigned long spl_flags = 0; /* do query_qp to obtain current attr values */ - mqpcb = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL); - if (mqpcb == NULL) { + mqpcb = ehca_alloc_fw_ctrlblock(); + if (!mqpcb) { ehca_err(ibqp->device, "Could not get zeroed page for mqpcb " "ehca_qp=%p qp_num=%x ", my_qp, ibqp->qp_num); return -ENOMEM; @@ -1225,7 +1225,7 @@ modify_qp_exit2: } modify_qp_exit1: - kfree(mqpcb); + ehca_free_fw_ctrlblock(mqpcb); return ret; } @@ -1277,7 +1277,7 @@ int ehca_query_qp(struct ib_qp *qp, return -EINVAL; } - qpcb = kzalloc(H_CB_ALIGNMENT, GFP_KERNEL ); + qpcb = ehca_alloc_fw_ctrlblock(); if (!qpcb) { ehca_err(qp->device,"Out of memory for qpcb " "ehca_qp=%p qp_num=%x", my_qp, qp->qp_num); @@ -1401,7 +1401,7 @@ int ehca_query_qp(struct ib_qp *qp, ehca_dmp(qpcb, 4*70, "qp_num=%x", qp->qp_num); query_qp_exit1: - kfree(qpcb); + ehca_free_fw_ctrlblock(qpcb); return ret; } -- cgit v0.10.2 From f7b2e8c76b3423a1d2501b9399261e9c9a33e100 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 9 Nov 2006 21:25:37 +0000 Subject: [CIFS] Fix minor problem with previous patch The patch NFS stress test generates flood of "close with pending write was missing an if Signed-off-by: Steve French diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 4b07a8c..2436ed8 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -498,7 +498,8 @@ int cifs_close(struct inode *inode, struct file *file) msleep(timeout); timeout *= 4; } - cERROR(1,("close with pending writes")); + if(atomic_read(&pSMBFile->wrtPending)) + cERROR(1,("close with pending writes")); rc = CIFSSMBClose(xid, pTcon, pSMBFile->netfid); } -- cgit v0.10.2 From 534284a09b3f58cd92acd0652b7267ee142932ba Mon Sep 17 00:00:00 2001 From: Pete Wyckoff Date: Wed, 8 Nov 2006 15:58:31 -0600 Subject: [SCSI] iscsi: always release crypto Unconditionally free crypto state, as it is always allocated during TCP connection creation. Without this, crypto structures leak and crc32c module refcounts grow as connections are created and destroyed. Signed-off-by: Pete Wyckoff Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c index 0a9dbc5..c0b8b33 100644 --- a/drivers/scsi/iscsi_tcp.c +++ b/drivers/scsi/iscsi_tcp.c @@ -1816,21 +1816,14 @@ iscsi_tcp_conn_destroy(struct iscsi_cls_conn *cls_conn) { struct iscsi_conn *conn = cls_conn->dd_data; struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - int digest = 0; - - if (conn->hdrdgst_en || conn->datadgst_en) - digest = 1; iscsi_tcp_release_conn(conn); iscsi_conn_teardown(cls_conn); - /* now free tcp_conn */ - if (digest) { - if (tcp_conn->tx_hash.tfm) - crypto_free_hash(tcp_conn->tx_hash.tfm); - if (tcp_conn->rx_hash.tfm) - crypto_free_hash(tcp_conn->rx_hash.tfm); - } + if (tcp_conn->tx_hash.tfm) + crypto_free_hash(tcp_conn->tx_hash.tfm); + if (tcp_conn->rx_hash.tfm) + crypto_free_hash(tcp_conn->rx_hash.tfm); kfree(tcp_conn); } -- cgit v0.10.2 From d6e24d1c8a197cc9c2a1568224474f4b7af50803 Mon Sep 17 00:00:00 2001 From: Pete Wyckoff Date: Wed, 8 Nov 2006 15:58:32 -0600 Subject: [SCSI] iscsi: add newlines to debug messages Some messages from debug_scsi do not have trailing newlines, making console messages difficult to read. Fix that. Signed-off-by: Pete Wyckoff Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 2865ebd..5d88621 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -975,13 +975,13 @@ int iscsi_eh_host_reset(struct scsi_cmnd *sc) if (session->state == ISCSI_STATE_TERMINATE) { failed: debug_scsi("failing host reset: session terminated " - "[CID %d age %d]", conn->id, session->age); + "[CID %d age %d]\n", conn->id, session->age); spin_unlock_bh(&session->lock); return FAILED; } if (sc->SCp.phase == session->age) { - debug_scsi("failing connection CID %d due to SCSI host reset", + debug_scsi("failing connection CID %d due to SCSI host reset\n", conn->id); fail_session = 1; } @@ -1054,7 +1054,8 @@ static int iscsi_exec_abort_task(struct scsi_cmnd *sc, NULL, 0); if (rc) { iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - debug_scsi("abort sent failure [itt 0x%x] %d", ctask->itt, rc); + debug_scsi("abort sent failure [itt 0x%x] %d\n", ctask->itt, + rc); return rc; } @@ -1071,7 +1072,7 @@ static int iscsi_exec_abort_task(struct scsi_cmnd *sc, conn->tmabort_timer.function = iscsi_tmabort_timedout; conn->tmabort_timer.data = (unsigned long)ctask; add_timer(&conn->tmabort_timer); - debug_scsi("abort set timeout [itt 0x%x]", ctask->itt); + debug_scsi("abort set timeout [itt 0x%x]\n", ctask->itt); } spin_unlock_bh(&session->lock); mutex_unlock(&conn->xmitmutex); -- cgit v0.10.2 From db37c505e5dfc1a26d6c82f1ce0c3ae06641c3e0 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Wed, 8 Nov 2006 15:58:33 -0600 Subject: [SCSI] iscsi_tcp: fix xmittask oops XMSTATE_SOL_HDR could be set when the xmit thread tests it, but there may not be anything on the r2tqueue yet. Move the XMSTATE_SOL_HDR set before the addition to the queue to make sure that when we pull something off it it is valid. This does not add locks around the xmstate test or make that a atmoic_t because this is a fast path and if it is set when we test it we can handle it there without the overhead. Later on we check the xmitqueue for all requests with the session lock so we will not miss it. Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c index c0b8b33..d0b139c 100644 --- a/drivers/scsi/iscsi_tcp.c +++ b/drivers/scsi/iscsi_tcp.c @@ -415,8 +415,8 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) iscsi_solicit_data_init(conn, ctask, r2t); tcp_ctask->exp_r2tsn = r2tsn + 1; - tcp_ctask->xmstate |= XMSTATE_SOL_HDR; __kfifo_put(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*)); + tcp_ctask->xmstate |= XMSTATE_SOL_HDR; list_move_tail(&ctask->running, &conn->xmitqueue); scsi_queue_work(session->host, &conn->xmitwork); @@ -1627,9 +1627,12 @@ static int iscsi_send_sol_pdu(struct iscsi_conn *conn, if (tcp_ctask->xmstate & XMSTATE_SOL_HDR) { tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR; tcp_ctask->xmstate |= XMSTATE_SOL_DATA; - if (!tcp_ctask->r2t) + if (!tcp_ctask->r2t) { + spin_lock_bh(&session->lock); __kfifo_get(tcp_ctask->r2tqueue, (void*)&tcp_ctask->r2t, sizeof(void*)); + spin_unlock_bh(&session->lock); + } send_hdr: r2t = tcp_ctask->r2t; dtask = &r2t->dtask; -- cgit v0.10.2 From 82a0d7b5829ebd033b7f808c026ab43509913692 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Wed, 8 Nov 2006 15:58:34 -0600 Subject: [SCSI] iscsi class: update version Update version number Signed-off-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 2d3baa9..9b25124 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -33,7 +33,7 @@ #define ISCSI_SESSION_ATTRS 11 #define ISCSI_CONN_ATTRS 11 #define ISCSI_HOST_ATTRS 0 -#define ISCSI_TRANSPORT_VERSION "2.0-685" +#define ISCSI_TRANSPORT_VERSION "2.0-724" struct iscsi_internal { int daemon_pid; -- cgit v0.10.2 From 107e716b3487df5e2940ebe3338d935306efc78b Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 9 Nov 2006 21:45:09 +0100 Subject: [SCSI] gdth: Fix && typos Fix uses of "&&" where "&" was obviously intended instead. Signed-off-by: Jean Delvare Signed-off-by: James Bottomley diff --git a/drivers/scsi/gdth.c b/drivers/scsi/gdth.c index 4bc14ad..4c698a7 100644 --- a/drivers/scsi/gdth.c +++ b/drivers/scsi/gdth.c @@ -3531,7 +3531,7 @@ static irqreturn_t gdth_interrupt(int irq,void *dev_id) IStatus &= ~0x80; #ifdef INT_COAL if (coalesced) - ha->status = pcs->ext_status && 0xffff; + ha->status = pcs->ext_status & 0xffff; else #endif ha->status = gdth_readw(&dp6m_ptr->i960r.status); @@ -3543,7 +3543,7 @@ static irqreturn_t gdth_interrupt(int irq,void *dev_id) if (coalesced) { ha->info = pcs->info0; ha->info2 = pcs->info1; - ha->service = (pcs->ext_status >> 16) && 0xffff; + ha->service = (pcs->ext_status >> 16) & 0xffff; } else #endif { -- cgit v0.10.2 From 05052f7f130b1232faeee1674a5bc41f67746cff Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 8 Nov 2006 19:56:37 -0800 Subject: [SCSI] psi240i.c: fix an array overrun Fix an array overrun spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: James Bottomley diff --git a/drivers/scsi/psi240i.c b/drivers/scsi/psi240i.c index ac0419e..899e89d 100644 --- a/drivers/scsi/psi240i.c +++ b/drivers/scsi/psi240i.c @@ -328,7 +328,7 @@ static void Irq_Handler (int irq, void *dev_id) pinquiryData->AdditionalLength = 35 - 4; // Fill in vendor identification fields. - for ( z = 0; z < 20; z += 2 ) + for ( z = 0; z < 8; z += 2 ) { pinquiryData->VendorId[z] = ((UCHAR *)identifyData.ModelNumber)[z + 1]; pinquiryData->VendorId[z + 1] = ((UCHAR *)identifyData.ModelNumber)[z]; -- cgit v0.10.2 From da413908d5e9ebdd0889a599e80d21d7237021c6 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 9 Nov 2006 20:00:55 -0800 Subject: [IPVS]: Compile fix for annotations in userland. This change makes __beXX available to user-space applications, such as ipvsadm, which include ip_vs.h Signed-Off-By: Simon Horman Signed-off-by: David S. Miller diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index 49c717e..903108e 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -7,6 +7,7 @@ #define _IP_VS_H #include /* For __uXX types */ +#include /* For __beXX types in userland */ #define IP_VS_VERSION_CODE 0x010201 #define NVERSION(version) \ -- cgit v0.10.2 From 599540a85595bd5950354bd95f5ebf9c6e07c971 Mon Sep 17 00:00:00 2001 From: Kalle Pokki Date: Wed, 1 Nov 2006 09:52:41 +0200 Subject: [POWERPC] CPM_UART: Fix non-console transmit The SMC and SCC hardware transmitter is enabled at the wrong place. Simply writing twice to the non-console port, like $ echo asdf > /dev/ttyCPM1 $ echo asdf > /dev/ttyCPM1 puts the shell into endless uninterruptible sleep, since the transmitter is stopped after the first write, and is not enabled before the shutdown function of the second write. Thus the transmit buffers are never emptied. Signed-off-by: Kalle Pokki Signed-off-by: Vitaly Bordug Signed-off-by: Paul Mackerras diff --git a/drivers/serial/cpm_uart/cpm_uart_core.c b/drivers/serial/cpm_uart/cpm_uart_core.c index 0abb544..32fd8c8 100644 --- a/drivers/serial/cpm_uart/cpm_uart_core.c +++ b/drivers/serial/cpm_uart/cpm_uart_core.c @@ -195,10 +195,8 @@ static void cpm_uart_start_tx(struct uart_port *port) if (cpm_uart_tx_pump(port) != 0) { if (IS_SMC(pinfo)) { smcp->smc_smcm |= SMCM_TX; - smcp->smc_smcmr |= SMCMR_TEN; } else { sccp->scc_sccm |= UART_SCCM_TX; - pinfo->sccp->scc_gsmrl |= SCC_GSMRL_ENT; } } } @@ -421,9 +419,10 @@ static int cpm_uart_startup(struct uart_port *port) /* Startup rx-int */ if (IS_SMC(pinfo)) { pinfo->smcp->smc_smcm |= SMCM_RX; - pinfo->smcp->smc_smcmr |= SMCMR_REN; + pinfo->smcp->smc_smcmr |= (SMCMR_REN | SMCMR_TEN); } else { pinfo->sccp->scc_sccm |= UART_SCCM_RX; + pinfo->sccp->scc_gsmrl |= (SCC_GSMRL_ENR | SCC_GSMRL_ENT); } if (!(pinfo->flags & FLAG_CONSOLE)) -- cgit v0.10.2 From 0091cf5a6ae6e52fc95ceb53200975ef2c81c206 Mon Sep 17 00:00:00 2001 From: Kalle Pokki Date: Wed, 1 Nov 2006 15:08:13 +0200 Subject: [POWERPC] CPM_UART: Fix non-console initialisation The cpm_uart driver is initialised incorrectly, if there is a frame buffer console, and CONFIG_SERIAL_CPM_CONSOLE is defined. The driver fails to call cpm_uart_init_portdesc() and set_lineif() in this case. Signed-off-by: Kalle Pokki Signed-off-by: Vitaly Bordug Signed-off-by: Paul Mackerras diff --git a/drivers/serial/cpm_uart/cpm_uart.h b/drivers/serial/cpm_uart/cpm_uart.h index a8f894c..69715e5 100644 --- a/drivers/serial/cpm_uart/cpm_uart.h +++ b/drivers/serial/cpm_uart/cpm_uart.h @@ -88,7 +88,7 @@ extern struct uart_cpm_port cpm_uart_ports[UART_NR]; /* these are located in their respective files */ void cpm_line_cr_cmd(int line, int cmd); -int cpm_uart_init_portdesc(void); +int __init cpm_uart_init_portdesc(void); int cpm_uart_allocbuf(struct uart_cpm_port *pinfo, unsigned int is_con); void cpm_uart_freebuf(struct uart_cpm_port *pinfo); diff --git a/drivers/serial/cpm_uart/cpm_uart_core.c b/drivers/serial/cpm_uart/cpm_uart_core.c index 32fd8c8..7a3b97f 100644 --- a/drivers/serial/cpm_uart/cpm_uart_core.c +++ b/drivers/serial/cpm_uart/cpm_uart_core.c @@ -1349,11 +1349,10 @@ static int cpm_uart_init(void) { pr_info("cpm_uart: WARNING: no UART devices found on platform bus!\n"); pr_info( "cpm_uart: the driver will guess configuration, but this mode is no longer supported.\n"); -#ifndef CONFIG_SERIAL_CPM_CONSOLE - ret = cpm_uart_init_portdesc(); - if (ret) - return ret; -#endif + + /* Don't run this again, if the console driver did it already */ + if (cpm_uart_nr == 0) + cpm_uart_init_portdesc(); cpm_reg.nr = cpm_uart_nr; ret = uart_register_driver(&cpm_reg); @@ -1365,6 +1364,8 @@ static int cpm_uart_init(void) { int con = cpm_uart_port_map[i]; cpm_uart_ports[con].port.line = i; cpm_uart_ports[con].port.flags = UPF_BOOT_AUTOCONF; + if (cpm_uart_ports[con].set_lineif) + cpm_uart_ports[con].set_lineif(&cpm_uart_ports[con]); uart_add_one_port(&cpm_reg, &cpm_uart_ports[con].port); } diff --git a/drivers/serial/cpm_uart/cpm_uart_cpm1.c b/drivers/serial/cpm_uart/cpm_uart_cpm1.c index 95afc37..08e55fd 100644 --- a/drivers/serial/cpm_uart/cpm_uart_cpm1.c +++ b/drivers/serial/cpm_uart/cpm_uart_cpm1.c @@ -184,7 +184,7 @@ void cpm_uart_freebuf(struct uart_cpm_port *pinfo) } /* Setup any dynamic params in the uart desc */ -int cpm_uart_init_portdesc(void) +int __init cpm_uart_init_portdesc(void) { pr_debug("CPM uart[-]:init portdesc\n"); -- cgit v0.10.2 From ae883cab9457aad0fb3342249e1207873d3b64de Mon Sep 17 00:00:00 2001 From: John Rose Date: Wed, 8 Nov 2006 10:07:30 -0600 Subject: [POWERPC] pseries: Force 4k update_flash block and list sizes The enablement of 64k pages on pseries platforms exposed a bug in the RTAS mechanism for updating firmware. RTAS assumes 4k for flash block and list sizes, and use of any other sizes results in a failure, even though PAPR does not specify any such requirement. This patch changes the rtas_flash module to force the use of 4k memory block and list sizes when preparing and sending a firmware image to RTAS. The rtas_flash function now uses a slab cache of 4k blocks with 4k alignment, rather than get_zeroed_page(), to allocate the memory for the flash blocks and lists. The 4k alignment requirement is specified in PAPR. Signed-off-by: John Rose Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c index 1442b63..6f6fc97 100644 --- a/arch/powerpc/kernel/rtas_flash.c +++ b/arch/powerpc/kernel/rtas_flash.c @@ -72,6 +72,10 @@ #define VALIDATE_BUF_SIZE 4096 #define RTAS_MSG_MAXLEN 64 +/* Quirk - RTAS requires 4k list length and block size */ +#define RTAS_BLKLIST_LENGTH 4096 +#define RTAS_BLK_SIZE 4096 + struct flash_block { char *data; unsigned long length; @@ -83,7 +87,7 @@ struct flash_block { * into a version/length and translate the pointers * to absolute. */ -#define FLASH_BLOCKS_PER_NODE ((PAGE_SIZE - 16) / sizeof(struct flash_block)) +#define FLASH_BLOCKS_PER_NODE ((RTAS_BLKLIST_LENGTH - 16) / sizeof(struct flash_block)) struct flash_block_list { unsigned long num_blocks; struct flash_block_list *next; @@ -96,6 +100,9 @@ struct flash_block_list_header { /* just the header of flash_block_list */ static struct flash_block_list_header rtas_firmware_flash_list = {0, NULL}; +/* Use slab cache to guarantee 4k alignment */ +static kmem_cache_t *flash_block_cache = NULL; + #define FLASH_BLOCK_LIST_VERSION (1UL) /* Local copy of the flash block list. @@ -153,7 +160,7 @@ static int flash_list_valid(struct flash_block_list *flist) return FLASH_IMG_NULL_DATA; } block_size = f->blocks[i].length; - if (block_size <= 0 || block_size > PAGE_SIZE) { + if (block_size <= 0 || block_size > RTAS_BLK_SIZE) { return FLASH_IMG_BAD_LEN; } image_size += block_size; @@ -177,9 +184,9 @@ static void free_flash_list(struct flash_block_list *f) while (f) { for (i = 0; i < f->num_blocks; i++) - free_page((unsigned long)(f->blocks[i].data)); + kmem_cache_free(flash_block_cache, f->blocks[i].data); next = f->next; - free_page((unsigned long)f); + kmem_cache_free(flash_block_cache, f); f = next; } } @@ -278,6 +285,12 @@ static ssize_t rtas_flash_read(struct file *file, char __user *buf, return msglen; } +/* constructor for flash_block_cache */ +void rtas_block_ctor(void *ptr, kmem_cache_t *cache, unsigned long flags) +{ + memset(ptr, 0, RTAS_BLK_SIZE); +} + /* We could be much more efficient here. But to keep this function * simple we allocate a page to the block list no matter how small the * count is. If the system is low on memory it will be just as well @@ -302,7 +315,7 @@ static ssize_t rtas_flash_write(struct file *file, const char __user *buffer, * proc file */ if (uf->flist == NULL) { - uf->flist = (struct flash_block_list *) get_zeroed_page(GFP_KERNEL); + uf->flist = kmem_cache_alloc(flash_block_cache, GFP_KERNEL); if (!uf->flist) return -ENOMEM; } @@ -313,21 +326,21 @@ static ssize_t rtas_flash_write(struct file *file, const char __user *buffer, next_free = fl->num_blocks; if (next_free == FLASH_BLOCKS_PER_NODE) { /* Need to allocate another block_list */ - fl->next = (struct flash_block_list *)get_zeroed_page(GFP_KERNEL); + fl->next = kmem_cache_alloc(flash_block_cache, GFP_KERNEL); if (!fl->next) return -ENOMEM; fl = fl->next; next_free = 0; } - if (count > PAGE_SIZE) - count = PAGE_SIZE; - p = (char *)get_zeroed_page(GFP_KERNEL); + if (count > RTAS_BLK_SIZE) + count = RTAS_BLK_SIZE; + p = kmem_cache_alloc(flash_block_cache, GFP_KERNEL); if (!p) return -ENOMEM; if(copy_from_user(p, buffer, count)) { - free_page((unsigned long)p); + kmem_cache_free(flash_block_cache, p); return -EFAULT; } fl->blocks[next_free].data = p; @@ -791,6 +804,16 @@ int __init rtas_flash_init(void) goto cleanup; rtas_flash_term_hook = rtas_flash_firmware; + + flash_block_cache = kmem_cache_create("rtas_flash_cache", + RTAS_BLK_SIZE, RTAS_BLK_SIZE, 0, + rtas_block_ctor, NULL); + if (!flash_block_cache) { + printk(KERN_ERR "%s: failed to create block cache\n", + __FUNCTION__); + rc = -ENOMEM; + goto cleanup; + } return 0; cleanup: @@ -805,6 +828,10 @@ cleanup: void __exit rtas_flash_cleanup(void) { rtas_flash_term_hook = NULL; + + if (flash_block_cache) + kmem_cache_destroy(flash_block_cache); + remove_flash_pde(firmware_flash_pde); remove_flash_pde(firmware_update_pde); remove_flash_pde(validate_pde); -- cgit v0.10.2 From ab56dbddc8a23ff3f4602855aaf0fcb3c814118b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 10 Nov 2006 15:11:20 +1100 Subject: [POWERPC] Fix cell "new style" mapping and add debug This fixes a typo in the "new style" code for mapping SPE resources, which causes it to try to map the same resource 4 times. It also adds some pr_debug's that are useful to track down issues with the firmware when bringinh up new machines. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c index d0fb959..7aa809d 100644 --- a/arch/powerpc/platforms/cell/spu_base.c +++ b/arch/powerpc/platforms/cell/spu_base.c @@ -655,14 +655,19 @@ static int __init spu_map_interrupts(struct spu *spu, struct device_node *np) for (i=0; i < 3; i++) { ret = of_irq_map_one(np, i, &oirq); - if (ret) + if (ret) { + pr_debug("spu_new: failed to get irq %d\n", i); goto err; - + } ret = -EINVAL; + pr_debug(" irq %d no 0x%x on %s\n", i, oirq.specifier[0], + oirq.controller->full_name); spu->irqs[i] = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); - if (spu->irqs[i] == NO_IRQ) + if (spu->irqs[i] == NO_IRQ) { + pr_debug("spu_new: failed to map it !\n"); goto err; + } } return 0; @@ -681,7 +686,7 @@ static int spu_map_resource(struct device_node *node, int nr, struct resource resource = { }; int ret; - ret = of_address_to_resource(node, 0, &resource); + ret = of_address_to_resource(node, nr, &resource); if (ret) goto out; @@ -704,22 +709,42 @@ static int __init spu_map_device(struct spu *spu, struct device_node *node) ret = spu_map_resource(node, 0, (void __iomem**)&spu->local_store, &spu->local_store_phys); - if (ret) + if (ret) { + pr_debug("spu_new: failed to map %s resource 0\n", + node->full_name); goto out; + } ret = spu_map_resource(node, 1, (void __iomem**)&spu->problem, &spu->problem_phys); - if (ret) + if (ret) { + pr_debug("spu_new: failed to map %s resource 1\n", + node->full_name); goto out_unmap; + } ret = spu_map_resource(node, 2, (void __iomem**)&spu->priv2, NULL); - if (ret) + if (ret) { + pr_debug("spu_new: failed to map %s resource 2\n", + node->full_name); goto out_unmap; + } if (!firmware_has_feature(FW_FEATURE_LPAR)) ret = spu_map_resource(node, 3, (void __iomem**)&spu->priv1, NULL); - if (ret) + if (ret) { + pr_debug("spu_new: failed to map %s resource 3\n", + node->full_name); goto out_unmap; + } + pr_debug("spu_new: %s maps:\n", node->full_name); + pr_debug(" local store : 0x%016lx -> 0x%p\n", + spu->local_store_phys, spu->local_store); + pr_debug(" problem state : 0x%016lx -> 0x%p\n", + spu->problem_phys, spu->problem); + pr_debug(" priv2 : 0x%p\n", spu->priv2); + pr_debug(" priv1 : 0x%p\n", spu->priv1); + return 0; out_unmap: -- cgit v0.10.2 From 36b600f2649e3be49039efe31edeeb64277dbd99 Mon Sep 17 00:00:00 2001 From: Geoff Levand Date: Thu, 2 Nov 2006 21:08:45 -0800 Subject: [POWERPC] cell: set ARCH_SPARSEMEM_DEFAULT in Kconfig The current cell processor support needs sparsemem, so set it as the default memory model. Signed-off-by: Geoff Levand Acked-by: Arnd Bergmann Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 2bd9b7f..0673dbe 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -740,7 +740,7 @@ config ARCH_SPARSEMEM_ENABLE config ARCH_SPARSEMEM_DEFAULT def_bool y - depends on SMP && PPC_PSERIES + depends on (SMP && PPC_PSERIES) || PPC_CELL config ARCH_POPULATES_NODE_MAP def_bool y -- cgit v0.10.2 From 0daa2303028a63dbd1b2e38f10854f0f7bf1ef9a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Nov 2006 19:51:01 -0800 Subject: [PATCH] bonding: lockdep annotation ============================================= [ INFO: possible recursive locking detected ] 2.6.17-1.2600.fc6 #1 Signed-off-by: Jeff Garzik diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index c0bbdda..17a4611 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4692,6 +4692,8 @@ static int bond_check_params(struct bond_params *params) return 0; } +static struct lock_class_key bonding_netdev_xmit_lock_key; + /* Create a new bond based on the specified name and bonding parameters. * Caller must NOT hold rtnl_lock; we need to release it here before we * set up our sysfs entries. @@ -4727,6 +4729,9 @@ int bond_create(char *name, struct bond_params *params, struct bonding **newbond if (res < 0) { goto out_bond; } + + lockdep_set_class(&bond_dev->_xmit_lock, &bonding_netdev_xmit_lock_key); + if (newbond) *newbond = bond_dev->priv; -- cgit v0.10.2 From ace48ffb5d6c927c5a98048d93543e1cae0eebd0 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 8 Nov 2006 19:51:03 -0800 Subject: [PATCH] com20020 build fix com20020.c needs to export functions if either of the ISA or PCI modules are built as loadable modules. Or they could always be exported. WARNING: "com20020_found" [drivers/net/arcnet/com20020-pci.ko] undefined! WARNING: "com20020_check" [drivers/net/arcnet/com20020-pci.ko] undefined! Signed-off-by: Randy Dunlap Cc: Toralf Forster Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/arcnet/com20020.c b/drivers/net/arcnet/com20020.c index 0dc70c7..aa9dd8f 100644 --- a/drivers/net/arcnet/com20020.c +++ b/drivers/net/arcnet/com20020.c @@ -337,13 +337,16 @@ static void com20020_set_mc_list(struct net_device *dev) } } -#ifdef MODULE - +#if defined(CONFIG_ARCNET_COM20020_PCI_MODULE) || \ + defined(CONFIG_ARCNET_COM20020_ISA_MODULE) EXPORT_SYMBOL(com20020_check); EXPORT_SYMBOL(com20020_found); +#endif MODULE_LICENSE("GPL"); +#ifdef MODULE + int init_module(void) { BUGLVL(D_NORMAL) printk(VERSION); -- cgit v0.10.2 From 92b1f905637bbd79fcd430a09737fd97061eb405 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Wed, 8 Nov 2006 19:49:15 -0800 Subject: [PATCH] drivers cris: return on NULL dev_alloc_skb() If the next descriptor array entry cannot be allocated by dev_alloc_skb(), return immediately so it is not dereferenced later. We cannot register the device with a partial descriptor list. Cc: Mikael Starvik Signed-off-by: David Rientjes Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik diff --git a/drivers/net/cris/eth_v10.c b/drivers/net/cris/eth_v10.c index 966b563..a03d781 100644 --- a/drivers/net/cris/eth_v10.c +++ b/drivers/net/cris/eth_v10.c @@ -509,6 +509,8 @@ etrax_ethernet_init(void) * does not share cacheline with any other data (to avoid cache bug) */ RxDescList[i].skb = dev_alloc_skb(MAX_MEDIA_DATA_SIZE + 2 * L1_CACHE_BYTES); + if (!RxDescList[i].skb) + return -ENOMEM; RxDescList[i].descr.ctrl = 0; RxDescList[i].descr.sw_len = MAX_MEDIA_DATA_SIZE; RxDescList[i].descr.next = virt_to_phys(&RxDescList[i + 1]); -- cgit v0.10.2 From bb831eb2027c12a740ad4e4ad78f2a0e841ae808 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Fri, 10 Nov 2006 14:57:37 -0800 Subject: [IPVS]: More endianness fixed. - make sure port in FTP data is in network order (in fact it was looking buggy for big endian boxes before Viro's changes) - htonl -> htons for port Signed-off-by: Julian Anastasov Signed-off-by: David S. Miller diff --git a/net/ipv4/ipvs/ip_vs_ftp.c b/net/ipv4/ipvs/ip_vs_ftp.c index 6d398f1..687c1de 100644 --- a/net/ipv4/ipvs/ip_vs_ftp.c +++ b/net/ipv4/ipvs/ip_vs_ftp.c @@ -200,7 +200,7 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, from = n_cp->vaddr; port = n_cp->vport; sprintf(buf,"%d,%d,%d,%d,%d,%d", NIPQUAD(from), - ntohs(port)&255, (ntohs(port)>>8)&255); + (ntohs(port)>>8)&255, ntohs(port)&255); buf_len = strlen(buf); /* diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c index bfe779e..6ff05c3 100644 --- a/net/ipv4/ipvs/ip_vs_proto_tcp.c +++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c @@ -117,7 +117,7 @@ tcp_fast_csum_update(struct tcphdr *tcph, __be32 oldip, __be32 newip, { tcph->check = ip_vs_check_diff(~oldip, newip, - ip_vs_check_diff(oldport ^ htonl(0xFFFF), + ip_vs_check_diff(oldport ^ htons(0xFFFF), newport, tcph->check)); } diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c index 54aa760..691c8b6 100644 --- a/net/ipv4/ipvs/ip_vs_proto_udp.c +++ b/net/ipv4/ipvs/ip_vs_proto_udp.c @@ -122,10 +122,10 @@ udp_fast_csum_update(struct udphdr *uhdr, __be32 oldip, __be32 newip, { uhdr->check = ip_vs_check_diff(~oldip, newip, - ip_vs_check_diff(oldport ^ htonl(0xFFFF), + ip_vs_check_diff(oldport ^ htons(0xFFFF), newport, uhdr->check)); if (!uhdr->check) - uhdr->check = htonl(0xFFFF); + uhdr->check = -1; } static int @@ -173,7 +173,7 @@ udp_snat_handler(struct sk_buff **pskb, cp->protocol, (*pskb)->csum); if (udph->check == 0) - udph->check = htonl(0xFFFF); + udph->check = -1; IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", pp->name, udph->check, (char*)&(udph->check) - (char*)udph); @@ -228,7 +228,7 @@ udp_dnat_handler(struct sk_buff **pskb, cp->protocol, (*pskb)->csum); if (udph->check == 0) - udph->check = 0xFFFF; + udph->check = -1; (*pskb)->ip_summed = CHECKSUM_UNNECESSARY; } return 1; -- cgit v0.10.2 From 93c189c1148a5e39bcc8f62568f42a77f93477c5 Mon Sep 17 00:00:00 2001 From: Vlad Apostolov Date: Sat, 11 Nov 2006 18:03:49 +1100 Subject: [XFS] 956618: Linux crashes on boot with XFS-DMAPI filesystem when CONFIG_XFS_TRACE is on SGI-PV: 956618 SGI-Modid: xfs-linux-melb:xfs-kern:27196a Signed-off-by: Vlad Apostolov Signed-off-by: Tim Shimmin diff --git a/fs/xfs/Makefile-linux-2.6 b/fs/xfs/Makefile-linux-2.6 index 291948d..b49989b 100644 --- a/fs/xfs/Makefile-linux-2.6 +++ b/fs/xfs/Makefile-linux-2.6 @@ -21,22 +21,7 @@ EXTRA_CFLAGS += -Ifs/xfs -Ifs/xfs/linux-2.6 -funsigned-char XFS_LINUX := linux-2.6 ifeq ($(CONFIG_XFS_DEBUG),y) - EXTRA_CFLAGS += -g -DSTATIC="" -DDEBUG - EXTRA_CFLAGS += -DXFS_BUF_LOCK_TRACKING -endif -ifeq ($(CONFIG_XFS_TRACE),y) - EXTRA_CFLAGS += -DXFS_ALLOC_TRACE - EXTRA_CFLAGS += -DXFS_ATTR_TRACE - EXTRA_CFLAGS += -DXFS_BLI_TRACE - EXTRA_CFLAGS += -DXFS_BMAP_TRACE - EXTRA_CFLAGS += -DXFS_BMBT_TRACE - EXTRA_CFLAGS += -DXFS_DIR2_TRACE - EXTRA_CFLAGS += -DXFS_DQUOT_TRACE - EXTRA_CFLAGS += -DXFS_ILOCK_TRACE - EXTRA_CFLAGS += -DXFS_LOG_TRACE - EXTRA_CFLAGS += -DXFS_RW_TRACE - EXTRA_CFLAGS += -DXFS_BUF_TRACE - EXTRA_CFLAGS += -DXFS_VNODE_TRACE + EXTRA_CFLAGS += -g endif obj-$(CONFIG_XFS_FS) += xfs.o diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index db5f5a3..d338284 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -15,6 +15,7 @@ * along with this program; if not, write the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ +#include "xfs.h" #include #include #include @@ -31,7 +32,6 @@ #include #include #include -#include "xfs_linux.h" STATIC kmem_zone_t *xfs_buf_zone; STATIC kmem_shaker_t xfs_buf_shake; @@ -1406,7 +1406,7 @@ xfs_alloc_bufhash( btp->bt_hashshift = external ? 3 : 8; /* 8 or 256 buckets */ btp->bt_hashmask = (1 << btp->bt_hashshift) - 1; btp->bt_hash = kmem_zalloc((1 << btp->bt_hashshift) * - sizeof(xfs_bufhash_t), KM_SLEEP); + sizeof(xfs_bufhash_t), KM_SLEEP | KM_LARGE); for (i = 0; i < (1 << btp->bt_hashshift); i++) { spin_lock_init(&btp->bt_hash[i].bh_lock); INIT_LIST_HEAD(&btp->bt_hash[i].bh_list); diff --git a/fs/xfs/support/debug.c b/fs/xfs/support/debug.c index c75f683..4363512 100644 --- a/fs/xfs/support/debug.c +++ b/fs/xfs/support/debug.c @@ -15,11 +15,9 @@ * along with this program; if not, write the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ +#include #include "debug.h" #include "spin.h" -#include -#include -#include static char message[256]; /* keep it off the stack */ static DEFINE_SPINLOCK(xfs_err_lock); diff --git a/fs/xfs/xfs.h b/fs/xfs/xfs.h index 1a48dbb..bf0a120 100644 --- a/fs/xfs/xfs.h +++ b/fs/xfs/xfs.h @@ -17,5 +17,28 @@ */ #ifndef __XFS_H__ #define __XFS_H__ + +#ifdef CONFIG_XFS_DEBUG +#define STATIC +#define DEBUG 1 +#define XFS_BUF_LOCK_TRACKING 1 +/* #define QUOTADEBUG 1 */ +#endif + +#ifdef CONFIG_XFS_TRACE +#define XFS_ALLOC_TRACE 1 +#define XFS_ATTR_TRACE 1 +#define XFS_BLI_TRACE 1 +#define XFS_BMAP_TRACE 1 +#define XFS_BMBT_TRACE 1 +#define XFS_DIR2_TRACE 1 +#define XFS_DQUOT_TRACE 1 +#define XFS_ILOCK_TRACE 1 +#define XFS_LOG_TRACE 1 +#define XFS_RW_TRACE 1 +#define XFS_BUF_TRACE 1 +#define XFS_VNODE_TRACE 1 +#endif + #include #endif /* __XFS_H__ */ -- cgit v0.10.2 From 439b8434792d0b62e32ab1416f214a18a640cc03 Mon Sep 17 00:00:00 2001 From: Tim Shimmin Date: Sat, 11 Nov 2006 18:04:34 +1100 Subject: [XFS] Keep lockdep happy. SGI-PV: 956964 SGI-Modid: xfs-linux-melb:xfs-kern:27200a Signed-off-by: Tim Shimmin Signed-off-by: David Chinner Signed-off-by: Eric Sandeen diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index b73d216..1562ac2 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -683,6 +683,7 @@ xfs_ireclaim(xfs_inode_t *ip) /* * Free all memory associated with the inode. */ + xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); xfs_idestroy(ip); } -- cgit v0.10.2 From 70a505285f9859f77e07f7c12371b0d29ecf3d82 Mon Sep 17 00:00:00 2001 From: Vlad Apostolov Date: Sat, 11 Nov 2006 18:04:41 +1100 Subject: [XFS] rename uio_read() to xfs_uio_read() SGI-PV: 957004 SGI-Modid: xfs-linux-melb:xfs-kern:27231a Signed-off-by: Vlad Apostolov Signed-off-by: Tim Shimmin diff --git a/fs/xfs/support/move.c b/fs/xfs/support/move.c index caefa17..ac8617c 100644 --- a/fs/xfs/support/move.c +++ b/fs/xfs/support/move.c @@ -22,7 +22,7 @@ * as we go. */ int -uio_read(caddr_t src, size_t len, struct uio *uio) +xfs_uio_read(caddr_t src, size_t len, struct uio *uio) { size_t count; diff --git a/fs/xfs/support/move.h b/fs/xfs/support/move.h index 97a2498..977879c 100644 --- a/fs/xfs/support/move.h +++ b/fs/xfs/support/move.h @@ -65,6 +65,6 @@ struct uio { typedef struct uio uio_t; typedef struct iovec iovec_t; -extern int uio_read (caddr_t, size_t, uio_t *); +extern int xfs_uio_read (caddr_t, size_t, uio_t *); #endif /* __XFS_SUPPORT_MOVE_H__ */ diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index 8edbe1a..8e8e527 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -678,7 +678,7 @@ xfs_dir2_put_dirent64_uio( idbp->d_off = pa->cook; idbp->d_name[namelen] = '\0'; memcpy(idbp->d_name, pa->name, namelen); - rval = uio_read((caddr_t)idbp, reclen, uio); + rval = xfs_uio_read((caddr_t)idbp, reclen, uio); pa->done = (rval == 0); return rval; } diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 061e2ff..2c79c36 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1013,7 +1013,7 @@ xfs_readlink( pathlen = (int)ip->i_d.di_size; if (ip->i_df.if_flags & XFS_IFINLINE) { - error = uio_read(ip->i_df.if_u1.if_data, pathlen, uiop); + error = xfs_uio_read(ip->i_df.if_u1.if_data, pathlen, uiop); } else { /* @@ -1044,7 +1044,7 @@ xfs_readlink( byte_cnt = pathlen; pathlen -= byte_cnt; - error = uio_read(XFS_BUF_PTR(bp), byte_cnt, uiop); + error = xfs_uio_read(XFS_BUF_PTR(bp), byte_cnt, uiop); xfs_buf_relse (bp); } -- cgit v0.10.2 From 2e2e7bb1fd857b9fc83b0cd77b6b647ebb423301 Mon Sep 17 00:00:00 2001 From: Vlad Apostolov Date: Sat, 11 Nov 2006 18:04:47 +1100 Subject: [XFS] 956664: dm_read_invis() changes i_atime SGI-PV: 956664 SGI-Modid: xfs-linux-melb:xfs-kern:27315a Signed-off-by: Vlad Apostolov Signed-off-by: Sam Vaughan Signed-off-by: Tim Shimmin diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index a74f854..74d09482 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -341,8 +341,11 @@ xfs_open_by_handle( put_unused_fd(new_fd); return -XFS_ERROR(-PTR_ERR(filp)); } - if (inode->i_mode & S_IFREG) + if (inode->i_mode & S_IFREG) { + /* invisible operation should not change atime */ + filp->f_flags |= O_NOATIME; filp->f_op = &xfs_invis_file_operations; + } fd_install(new_fd, filp); return new_fd; -- cgit v0.10.2 From 7a18c386078eaf17ae54595f66c0d64d9c1cb29c Mon Sep 17 00:00:00 2001 From: David Chinner Date: Sat, 11 Nov 2006 18:04:54 +1100 Subject: [XFS] Clean up i_flags and i_flags_lock handling. SGI-PV: 956832 SGI-Modid: xfs-linux-melb:xfs-kern:27358a Signed-off-by: David Chinner Signed-off-by: Nathan Scott Signed-off-by: Tim Shimmin diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 38c4d12..de05abb 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -227,9 +227,7 @@ xfs_initialize_vnode( xfs_revalidate_inode(XFS_BHVTOM(bdp), vp, ip); xfs_set_inodeops(inode); - spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~XFS_INEW; - spin_unlock(&ip->i_flags_lock); + xfs_iflags_clear(ip, XFS_INEW); barrier(); unlock_new_inode(inode); diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 1562ac2..4b0c188 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -215,7 +215,7 @@ again: * If INEW is set this inode is being set up * we need to pause and try again. */ - if (ip->i_flags & XFS_INEW) { + if (xfs_iflags_test(ip, XFS_INEW)) { read_unlock(&ih->ih_lock); delay(1); XFS_STATS_INC(xs_ig_frecycle); @@ -230,7 +230,7 @@ again: * on its way out of the system, * we need to pause and try again. */ - if (ip->i_flags & XFS_IRECLAIM) { + if (xfs_iflags_test(ip, XFS_IRECLAIM)) { read_unlock(&ih->ih_lock); delay(1); XFS_STATS_INC(xs_ig_frecycle); @@ -243,9 +243,7 @@ again: XFS_STATS_INC(xs_ig_found); - spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~XFS_IRECLAIMABLE; - spin_unlock(&ip->i_flags_lock); + xfs_iflags_clear(ip, XFS_IRECLAIMABLE); version = ih->ih_version; read_unlock(&ih->ih_lock); xfs_ihash_promote(ih, ip, version); @@ -299,10 +297,7 @@ finish_inode: if (lock_flags != 0) xfs_ilock(ip, lock_flags); - spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~XFS_ISTALE; - spin_unlock(&ip->i_flags_lock); - + xfs_iflags_clear(ip, XFS_ISTALE); vn_trace_exit(vp, "xfs_iget.found", (inst_t *)__return_address); goto return_ip; @@ -371,10 +366,7 @@ finish_inode: ih->ih_next = ip; ip->i_udquot = ip->i_gdquot = NULL; ih->ih_version++; - spin_lock(&ip->i_flags_lock); - ip->i_flags |= XFS_INEW; - spin_unlock(&ip->i_flags_lock); - + xfs_iflags_set(ip, XFS_INEW); write_unlock(&ih->ih_lock); /* @@ -625,7 +617,7 @@ xfs_iput_new(xfs_inode_t *ip, vn_trace_entry(vp, "xfs_iput_new", (inst_t *)__return_address); if ((ip->i_d.di_mode == 0)) { - ASSERT(!(ip->i_flags & XFS_IRECLAIMABLE)); + ASSERT(!xfs_iflags_test(ip, XFS_IRECLAIMABLE)); vn_mark_bad(vp); } if (inode->i_state & I_NEW) diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index c27d7d49..17d2a47 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2193,7 +2193,7 @@ xfs_ifree_cluster( /* Inode not in memory or we found it already, * nothing to do */ - if (!ip || (ip->i_flags & XFS_ISTALE)) { + if (!ip || xfs_iflags_test(ip, XFS_ISTALE)) { read_unlock(&ih->ih_lock); continue; } @@ -2215,10 +2215,7 @@ xfs_ifree_cluster( if (ip == free_ip) { if (xfs_iflock_nowait(ip)) { - spin_lock(&ip->i_flags_lock); - ip->i_flags |= XFS_ISTALE; - spin_unlock(&ip->i_flags_lock); - + xfs_iflags_set(ip, XFS_ISTALE); if (xfs_inode_clean(ip)) { xfs_ifunlock(ip); } else { @@ -2231,9 +2228,7 @@ xfs_ifree_cluster( if (xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) { if (xfs_iflock_nowait(ip)) { - spin_lock(&ip->i_flags_lock); - ip->i_flags |= XFS_ISTALE; - spin_unlock(&ip->i_flags_lock); + xfs_iflags_set(ip, XFS_ISTALE); if (xfs_inode_clean(ip)) { xfs_ifunlock(ip); @@ -2263,9 +2258,7 @@ xfs_ifree_cluster( AIL_LOCK(mp,s); iip->ili_flush_lsn = iip->ili_item.li_lsn; AIL_UNLOCK(mp, s); - spin_lock(&iip->ili_inode->i_flags_lock); - iip->ili_inode->i_flags |= XFS_ISTALE; - spin_unlock(&iip->ili_inode->i_flags_lock); + xfs_iflags_set(ip, XFS_ISTALE); pre_flushed++; } lip = lip->li_bio_list; @@ -2764,7 +2757,7 @@ xfs_iunpin( struct inode *inode = NULL; spin_lock(&ip->i_flags_lock); - if (!(ip->i_flags & (XFS_IRECLAIM|XFS_IRECLAIMABLE))) { + if (!__xfs_iflags_test(ip, XFS_IRECLAIM|XFS_IRECLAIMABLE)) { bhv_vnode_t *vp = XFS_ITOV_NULL(ip); /* make sync come back and flush this inode */ diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index e96eb08..bc82372 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -305,6 +305,47 @@ typedef struct xfs_inode { #endif } xfs_inode_t; + +/* + * i_flags helper functions + */ +static inline void +__xfs_iflags_set(xfs_inode_t *ip, unsigned short flags) +{ + ip->i_flags |= flags; +} + +static inline void +xfs_iflags_set(xfs_inode_t *ip, unsigned short flags) +{ + spin_lock(&ip->i_flags_lock); + __xfs_iflags_set(ip, flags); + spin_unlock(&ip->i_flags_lock); +} + +static inline void +xfs_iflags_clear(xfs_inode_t *ip, unsigned short flags) +{ + spin_lock(&ip->i_flags_lock); + ip->i_flags &= ~flags; + spin_unlock(&ip->i_flags_lock); +} + +static inline int +__xfs_iflags_test(xfs_inode_t *ip, unsigned short flags) +{ + return (ip->i_flags & flags); +} + +static inline int +xfs_iflags_test(xfs_inode_t *ip, unsigned short flags) +{ + int ret; + spin_lock(&ip->i_flags_lock); + ret = __xfs_iflags_test(ip, flags); + spin_unlock(&ip->i_flags_lock); + return ret; +} #endif /* __KERNEL__ */ diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 2c79c36..4c5d73c 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -3844,9 +3844,7 @@ xfs_reclaim( XFS_MOUNT_ILOCK(mp); vn_bhv_remove(VN_BHV_HEAD(vp), XFS_ITOBHV(ip)); list_add_tail(&ip->i_reclaim, &mp->m_del_inodes); - spin_lock(&ip->i_flags_lock); - ip->i_flags |= XFS_IRECLAIMABLE; - spin_unlock(&ip->i_flags_lock); + xfs_iflags_set(ip, XFS_IRECLAIMABLE); XFS_MOUNT_IUNLOCK(mp); } return 0; @@ -3872,8 +3870,8 @@ xfs_finish_reclaim( */ write_lock(&ih->ih_lock); spin_lock(&ip->i_flags_lock); - if ((ip->i_flags & XFS_IRECLAIM) || - (!(ip->i_flags & XFS_IRECLAIMABLE) && vp == NULL)) { + if (__xfs_iflags_test(ip, XFS_IRECLAIM) || + (!__xfs_iflags_test(ip, XFS_IRECLAIMABLE) && vp == NULL)) { spin_unlock(&ip->i_flags_lock); write_unlock(&ih->ih_lock); if (locked) { @@ -3882,7 +3880,7 @@ xfs_finish_reclaim( } return 1; } - ip->i_flags |= XFS_IRECLAIM; + __xfs_iflags_set(ip, XFS_IRECLAIM); spin_unlock(&ip->i_flags_lock); write_unlock(&ih->ih_lock); -- cgit v0.10.2 From 4c60658e0f4e253cf275f12b7c76bf128515a774 Mon Sep 17 00:00:00 2001 From: David Chinner Date: Sat, 11 Nov 2006 18:05:00 +1100 Subject: [XFS] Prevent a deadlock when xfslogd unpins inodes. The previous fixes for the use after free in xfs_iunpin left a nasty log deadlock when xfslogd unpinned the inode and dropped the last reference to the inode. the ->clear_inode() method can issue transactions, and if the log was full, the transaction could push on the log and get stuck trying to push the inode it was currently unpinning. To fix this, we provide xfs_iunpin a guarantee that it will always have a valid xfs_inode <-> linux inode link or a particular flag will be set on the inode. We then use log forces during lookup to ensure transactions are completed before we recycle the inode. This ensures that xfs_iunpin will never use the linux inode after it is being freed, and any lookup on an inode on the reclaim list will wait until it is safe to attach a new linux inode to the xfs inode. SGI-PV: 956832 SGI-Modid: xfs-linux-melb:xfs-kern:27359a Signed-off-by: David Chinner Signed-off-by: Shailendra Tripathi Signed-off-by: Takenori Nagano Signed-off-by: Tim Shimmin diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 4b0c188..c1c89da 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -237,6 +237,36 @@ again: goto again; } + ASSERT(xfs_iflags_test(ip, XFS_IRECLAIMABLE)); + + /* + * If lookup is racing with unlink, then we + * should return an error immediately so we + * don't remove it from the reclaim list and + * potentially leak the inode. + */ + if ((ip->i_d.di_mode == 0) && + !(flags & XFS_IGET_CREATE)) { + read_unlock(&ih->ih_lock); + return ENOENT; + } + + /* + * There may be transactions sitting in the + * incore log buffers or being flushed to disk + * at this time. We can't clear the + * XFS_IRECLAIMABLE flag until these + * transactions have hit the disk, otherwise we + * will void the guarantee the flag provides + * xfs_iunpin() + */ + if (xfs_ipincount(ip)) { + read_unlock(&ih->ih_lock); + xfs_log_force(mp, 0, + XFS_LOG_FORCE|XFS_LOG_SYNC); + XFS_STATS_INC(xs_ig_frecycle); + goto again; + } vn_trace_exit(vp, "xfs_iget.alloc", (inst_t *)__return_address); diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 17d2a47..d72c80d 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2741,42 +2741,39 @@ xfs_iunpin( { ASSERT(atomic_read(&ip->i_pincount) > 0); - if (atomic_dec_and_test(&ip->i_pincount)) { + if (atomic_dec_and_lock(&ip->i_pincount, &ip->i_flags_lock)) { + /* - * If the inode is currently being reclaimed, the - * linux inode _and_ the xfs vnode may have been - * freed so we cannot reference either of them safely. - * Hence we should not try to do anything to them - * if the xfs inode is currently in the reclaim - * path. + * If the inode is currently being reclaimed, the link between + * the bhv_vnode and the xfs_inode will be broken after the + * XFS_IRECLAIM* flag is set. Hence, if these flags are not + * set, then we can move forward and mark the linux inode dirty + * knowing that it is still valid as it won't freed until after + * the bhv_vnode<->xfs_inode link is broken in xfs_reclaim. The + * i_flags_lock is used to synchronise the setting of the + * XFS_IRECLAIM* flags and the breaking of the link, and so we + * can execute atomically w.r.t to reclaim by holding this lock + * here. * - * However, we still need to issue the unpin wakeup - * call as the inode reclaim may be blocked waiting for - * the inode to become unpinned. + * However, we still need to issue the unpin wakeup call as the + * inode reclaim may be blocked waiting for the inode to become + * unpinned. */ - struct inode *inode = NULL; - spin_lock(&ip->i_flags_lock); if (!__xfs_iflags_test(ip, XFS_IRECLAIM|XFS_IRECLAIMABLE)) { bhv_vnode_t *vp = XFS_ITOV_NULL(ip); + struct inode *inode = NULL; + + BUG_ON(vp == NULL); + inode = vn_to_inode(vp); + BUG_ON(inode->i_state & I_CLEAR); /* make sync come back and flush this inode */ - if (vp) { - inode = vn_to_inode(vp); - - if (!(inode->i_state & - (I_NEW|I_FREEING|I_CLEAR))) { - inode = igrab(inode); - if (inode) - mark_inode_dirty_sync(inode); - } else - inode = NULL; - } + if (!(inode->i_state & (I_NEW|I_FREEING))) + mark_inode_dirty_sync(inode); } spin_unlock(&ip->i_flags_lock); wake_up(&ip->i_ipin_wait); - if (inode) - iput(inode); } } diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 4c5d73c..bda774a 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -3827,11 +3827,16 @@ xfs_reclaim( */ xfs_synchronize_atime(ip); - /* If we have nothing to flush with this inode then complete the - * teardown now, otherwise break the link between the xfs inode - * and the linux inode and clean up the xfs inode later. This - * avoids flushing the inode to disk during the delete operation - * itself. + /* + * If we have nothing to flush with this inode then complete the + * teardown now, otherwise break the link between the xfs inode and the + * linux inode and clean up the xfs inode later. This avoids flushing + * the inode to disk during the delete operation itself. + * + * When breaking the link, we need to set the XFS_IRECLAIMABLE flag + * first to ensure that xfs_iunpin() will never see an xfs inode + * that has a linux inode being reclaimed. Synchronisation is provided + * by the i_flags_lock. */ if (!ip->i_update_core && (ip->i_itemp == NULL)) { xfs_ilock(ip, XFS_ILOCK_EXCL); @@ -3840,11 +3845,13 @@ xfs_reclaim( } else { xfs_mount_t *mp = ip->i_mount; - /* Protect sync from us */ + /* Protect sync and unpin from us */ XFS_MOUNT_ILOCK(mp); + spin_lock(&ip->i_flags_lock); + __xfs_iflags_set(ip, XFS_IRECLAIMABLE); vn_bhv_remove(VN_BHV_HEAD(vp), XFS_ITOBHV(ip)); + spin_unlock(&ip->i_flags_lock); list_add_tail(&ip->i_reclaim, &mp->m_del_inodes); - xfs_iflags_set(ip, XFS_IRECLAIMABLE); XFS_MOUNT_IUNLOCK(mp); } return 0; -- cgit v0.10.2 From 050e714eb2bc662e9df6bf048ce86b4fbdd9bcd3 Mon Sep 17 00:00:00 2001 From: David Chinner Date: Sat, 11 Nov 2006 18:05:06 +1100 Subject: [XFS] Remove KERNEL_VERSION macros from xfs_dmapi.h SGI-PV: 957005 SGI-Modid: xfs-linux-melb:xfs-kern:27398a Signed-off-by: David Chinner Signed-off-by: Michal Piotrowski Signed-off-by: Tim Shimmin diff --git a/fs/xfs/linux-2.6/xfs_dmapi_priv.h b/fs/xfs/linux-2.6/xfs_dmapi_priv.h new file mode 100644 index 0000000..a8b0b16 --- /dev/null +++ b/fs/xfs/linux-2.6/xfs_dmapi_priv.h @@ -0,0 +1,28 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_DMAPI_PRIV_H__ +#define __XFS_DMAPI_PRIV_H__ + +/* + * Based on IO_ISDIRECT, decide which i_ flag is set. + */ +#define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ + DM_FLAGS_IMUX : 0) +#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_IMUX) + +#endif /*__XFS_DMAPI_PRIV_H__*/ diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h index 4e7865a..adc3d25 100644 --- a/fs/xfs/xfs_dmapi.h +++ b/fs/xfs/xfs_dmapi.h @@ -157,27 +157,9 @@ typedef enum { #define DM_FLAGS_IALLOCSEM_WR 0x020 /* thread holds i_alloc_sem wr */ /* - * Based on IO_ISDIRECT, decide which i_ flag is set. + * Pull in platform specific event flags defines */ -#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,0) -#define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ - DM_FLAGS_IMUX : 0) -#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_IMUX) -#endif - -#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0)) && \ - (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,22)) -#define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ - DM_FLAGS_IALLOCSEM_RD : DM_FLAGS_IMUX) -#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_IMUX) -#endif - -#if LINUX_VERSION_CODE <= KERNEL_VERSION(2,4,21) -#define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ - 0 : DM_FLAGS_IMUX) -#define DM_SEM_FLAG_WR (DM_FLAGS_IMUX) -#endif - +#include "xfs_dmapi_priv.h" /* * Macros to turn caller specified delay/block flags into -- cgit v0.10.2 From 6751718932be7325c898e5908daec2ee917afd09 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 11 Nov 2006 22:54:07 +0000 Subject: [ARM] Remove PM_LEGACY=y from selected ARM defconfigs Most ARM defconfigs don't actually need to have PM_LEGACY enabled. Disable it for ATEB9200, Collie, IXP4xx, OMAP H2, S3C2410 and Versatile. Signed-off-by: Russell King diff --git a/arch/arm/configs/ateb9200_defconfig b/arch/arm/configs/ateb9200_defconfig index 15e6b0b..3de5c64 100644 --- a/arch/arm/configs/ateb9200_defconfig +++ b/arch/arm/configs/ateb9200_defconfig @@ -217,7 +217,7 @@ CONFIG_BINFMT_ELF=y # Power management options # CONFIG_PM=y -CONFIG_PM_LEGACY=y +# CONFIG_PM_LEGACY is not set # CONFIG_PM_DEBUG is not set # CONFIG_APM is not set diff --git a/arch/arm/configs/collie_defconfig b/arch/arm/configs/collie_defconfig index 074c47a..a375891 100644 --- a/arch/arm/configs/collie_defconfig +++ b/arch/arm/configs/collie_defconfig @@ -219,7 +219,7 @@ CONFIG_BINFMT_ELF=y # Power management options # CONFIG_PM=y -CONFIG_PM_LEGACY=y +# CONFIG_PM_LEGACY is not set # CONFIG_PM_DEBUG is not set CONFIG_APM=y diff --git a/arch/arm/configs/ixp4xx_defconfig b/arch/arm/configs/ixp4xx_defconfig index 4975b91..fac7c3b 100644 --- a/arch/arm/configs/ixp4xx_defconfig +++ b/arch/arm/configs/ixp4xx_defconfig @@ -206,10 +206,8 @@ CONFIG_BINFMT_ELF=y # # Power management options # -CONFIG_PM=y -CONFIG_PM_LEGACY=y -# CONFIG_PM_DEBUG is not set -CONFIG_APM=y +# CONFIG_PM is not set +# CONFIG_APM is not set # # Networking diff --git a/arch/arm/configs/omap_h2_1610_defconfig b/arch/arm/configs/omap_h2_1610_defconfig index 05adb0b..b0efd4c 100644 --- a/arch/arm/configs/omap_h2_1610_defconfig +++ b/arch/arm/configs/omap_h2_1610_defconfig @@ -257,7 +257,7 @@ CONFIG_BINFMT_AOUT=y # Power management options # CONFIG_PM=y -CONFIG_PM_LEGACY=y +# CONFIG_PM_LEGACY is not set # CONFIG_PM_DEBUG is not set # CONFIG_APM is not set diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig index 0563c14..c0152393 100644 --- a/arch/arm/configs/s3c2410_defconfig +++ b/arch/arm/configs/s3c2410_defconfig @@ -258,7 +258,7 @@ CONFIG_BINFMT_AOUT=y # Power management options # CONFIG_PM=y -CONFIG_PM_LEGACY=y +# CONFIG_PM_LEGACY is not set # CONFIG_PM_DEBUG is not set # CONFIG_PM_SYSFS_DEPRECATED is not set CONFIG_APM=y diff --git a/arch/arm/configs/versatile_defconfig b/arch/arm/configs/versatile_defconfig index 96b7a77..f7bf6ef 100644 --- a/arch/arm/configs/versatile_defconfig +++ b/arch/arm/configs/versatile_defconfig @@ -200,7 +200,7 @@ CONFIG_BINFMT_ELF=y # Power management options # CONFIG_PM=y -CONFIG_PM_LEGACY=y +# CONFIG_PM_LEGACY is not set # CONFIG_PM_DEBUG is not set # CONFIG_APM is not set -- cgit v0.10.2 From d027c4dc7d6e35a4e43dbcc178f0bf3359814306 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 3 Nov 2006 07:14:32 -0300 Subject: V4L/DVB (4795): Tda826x: use correct max frequency sparse "defined twice" warning Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/frontends/tda826x.c b/drivers/media/dvb/frontends/tda826x.c index eeab26b..34815b0 100644 --- a/drivers/media/dvb/frontends/tda826x.c +++ b/drivers/media/dvb/frontends/tda826x.c @@ -121,7 +121,7 @@ static struct dvb_tuner_ops tda826x_tuner_ops = { .info = { .name = "Philips TDA826X", .frequency_min = 950000, - .frequency_min = 2175000 + .frequency_max = 2175000 }, .release = tda826x_release, .sleep = tda826x_sleep, -- cgit v0.10.2 From ff97d93d6a311759db1b74b9b90dd6bcb8ce0aee Mon Sep 17 00:00:00 2001 From: Hermann Pitton Date: Fri, 3 Nov 2006 10:45:52 -0300 Subject: V4L/DVB (4802): Cx88: fix remote control on WinFast 2000XP Expert fix remote control on WinFast 2000XP Expert by setting timing back to 1 ms, like it was in the original patch by Robert Reid. Signed-off-by: Hermann Pitton Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c index ee48995..57e1c02 100644 --- a/drivers/media/video/cx88/cx88-input.c +++ b/drivers/media/video/cx88/cx88-input.c @@ -202,13 +202,19 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci) ir->sampling = 1; break; case CX88_BOARD_WINFAST_DTV2000H: - case CX88_BOARD_WINFAST2000XP_EXPERT: ir_codes = ir_codes_winfast; ir->gpio_addr = MO_GP0_IO; ir->mask_keycode = 0x8f8; ir->mask_keyup = 0x100; ir->polling = 50; /* ms */ break; + case CX88_BOARD_WINFAST2000XP_EXPERT: + ir_codes = ir_codes_winfast; + ir->gpio_addr = MO_GP0_IO; + ir->mask_keycode = 0x8f8; + ir->mask_keyup = 0x100; + ir->polling = 1; /* ms */ + break; case CX88_BOARD_IODATA_GVBCTV7E: ir_codes = ir_codes_iodata_bctv7e; ir->gpio_addr = MO_GP0_IO; @@ -216,7 +222,7 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci) ir->mask_keydown = 0x02; ir->polling = 5; /* ms */ break; - case CX88_BOARD_PROLINK_PLAYTVPVR: + case CX88_BOARD_PROLINK_PLAYTVPVR: case CX88_BOARD_PIXELVIEW_PLAYTV_ULTRA_PRO: ir_codes = ir_codes_pixelview; ir->gpio_addr = MO_GP1_IO; -- cgit v0.10.2 From ce48d5ecf3f52378064f317e0094b601508e9b3e Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Sun, 5 Nov 2006 09:02:13 -0300 Subject: V4L/DVB (4804): Fix missing i2c dependency for saa7110 drivers/media/video/saa7110.c:112: undefined reference to `i2c_master_send' drivers/built-in.o: In function `saa7110_read': drivers/media/video/saa7110.c:130: undefined reference to `i2c_smbus_read_byte' drivers/media/video/saa7110.c:130: undefined reference to `i2c_smbus_read_byte' Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index fbe5b61..bf26755 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -186,7 +186,7 @@ config VIDEO_KS0127 config VIDEO_SAA7110 tristate "Philips SAA7110 video decoder" - depends on VIDEO_V4L1 + depends on VIDEO_V4L1 && I2C ---help--- Support for the Philips SAA7110 video decoders. -- cgit v0.10.2 From 450efcfd2e1d941e302a8c89322fbfcef237be98 Mon Sep 17 00:00:00 2001 From: "pasky@ucw.cz" Date: Sun, 12 Nov 2006 14:22:32 -0300 Subject: V4L/DVB (4814): Remote support for Avermedia 777 I didn't test it personally since I don't have this card, but A16AR uses the same interface and that one certainly does work perfectly (see the next patch). This patch was originally sent in http://marc.theaimsgroup.com/?l=linux-video&m=114743413825375&w=2 https://www.redhat.com/mailman/private/video4linux-list/2006-May/msg00103.html but never got applied. This version has some trivial modifications and drops the weird gpio hack (it's not clear what practical purpose does it serve). Signed-off-by: Jose Alberto Reguero Signed-off-by: Petr Baudis Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index c9d8e3b..94324b3 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -3718,6 +3718,7 @@ int saa7134_board_init1(struct saa7134_dev *dev) case SAA7134_BOARD_AVERMEDIA_STUDIO_307: case SAA7134_BOARD_AVERMEDIA_307: case SAA7134_BOARD_AVERMEDIA_GO_007_FM: + case SAA7134_BOARD_AVERMEDIA_777: /* case SAA7134_BOARD_SABRENT_SBTTVFM: */ /* not finished yet */ case SAA7134_BOARD_VIDEOMATE_TV_PVR: case SAA7134_BOARD_VIDEOMATE_GOLD_PLUS: diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c index ff59911..e8dcb6f 100644 --- a/drivers/media/video/saa7134/saa7134-input.c +++ b/drivers/media/video/saa7134/saa7134-input.c @@ -194,6 +194,14 @@ int saa7134_input_init1(struct saa7134_dev *dev) saa_setb(SAA7134_GPIO_GPMODE0, 0x4); saa_setb(SAA7134_GPIO_GPSTATUS0, 0x4); break; + case SAA7134_BOARD_AVERMEDIA_777: + ir_codes = ir_codes_avermedia; + mask_keycode = 0x02F200; + mask_keydown = 0x000400; + polling = 50; // ms + /* Without this we won't receive key up events */ + saa_setb(SAA7134_GPIO_GPMODE1, 0x1); + saa_setb(SAA7134_GPIO_GPSTATUS1, 0x1); case SAA7134_BOARD_KWORLD_TERMINATOR: ir_codes = ir_codes_pixelview; mask_keycode = 0x00001f; -- cgit v0.10.2 From 29e0f1a136d39c5683d998741911b769d0172d52 Mon Sep 17 00:00:00 2001 From: "pasky@ucw.cz" Date: Sun, 12 Nov 2006 14:23:32 -0300 Subject: V4L/DVB (4815): Remote support for Avermedia A16AR The remote as well as the GPIO interface is the same as what comes with 777. For an example of mplayer lirc configuration, see http://pasky.or.cz/~pasky/dev/v4l/lircrc Signed-off-by: Petr Baudis Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 94324b3..1a402e4 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -3735,6 +3735,7 @@ int saa7134_board_init1(struct saa7134_dev *dev) case SAA7134_BOARD_FLYDVBT_LR301: case SAA7134_BOARD_FLYDVBTDUO: case SAA7134_BOARD_PROTEUS_2309: + case SAA7134_BOARD_AVERMEDIA_A16AR: dev->has_remote = SAA7134_REMOTE_GPIO; break; case SAA7134_BOARD_FLYDVBS_LR300: @@ -3773,7 +3774,6 @@ int saa7134_board_init1(struct saa7134_dev *dev) saa_writeb(SAA7134_GPIO_GPMODE3, 0x08); saa_writeb(SAA7134_GPIO_GPSTATUS3, 0x00); break; - case SAA7134_BOARD_AVERMEDIA_A16AR: case SAA7134_BOARD_AVERMEDIA_CARDBUS: /* power-up tuner chip */ saa_andorl(SAA7134_GPIO_GPMODE0 >> 2, 0xffffffff, 0xffffffff); diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c index e8dcb6f..7f62403 100644 --- a/drivers/media/video/saa7134/saa7134-input.c +++ b/drivers/media/video/saa7134/saa7134-input.c @@ -185,7 +185,6 @@ int saa7134_input_init1(struct saa7134_dev *dev) case SAA7134_BOARD_AVERMEDIA_STUDIO_305: case SAA7134_BOARD_AVERMEDIA_STUDIO_307: case SAA7134_BOARD_AVERMEDIA_GO_007_FM: - case SAA7134_BOARD_AVERMEDIA_A16AR: ir_codes = ir_codes_avermedia; mask_keycode = 0x0007C8; mask_keydown = 0x000010; @@ -195,6 +194,7 @@ int saa7134_input_init1(struct saa7134_dev *dev) saa_setb(SAA7134_GPIO_GPSTATUS0, 0x4); break; case SAA7134_BOARD_AVERMEDIA_777: + case SAA7134_BOARD_AVERMEDIA_A16AR: ir_codes = ir_codes_avermedia; mask_keycode = 0x02F200; mask_keydown = 0x000400; -- cgit v0.10.2 From 0871a8849b80646074cd28b2b078c8e002e51282 Mon Sep 17 00:00:00 2001 From: "pasky@ucw.cz" Date: Sun, 12 Nov 2006 14:24:57 -0300 Subject: V4L/DVB (4816): Change tuner type for Avermedia A16AR This changes it from TDA8290 which is allegedly very unlikely to TD1316 which is allegedly very likely. I didn't get it to work with either, but expected that this got applied when Mauro sent it to me, so here it goes again; feel free to drop it to the floor. :-) Signed-off-by: Petr Baudis Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 1a402e4..51f0cfd 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -2969,7 +2969,7 @@ struct saa7134_board saa7134_boards[] = { /* Petr Baudis */ .name = "AVerMedia TV Hybrid A16AR", .audio_clock = 0x187de7, - .tuner_type = TUNER_PHILIPS_TDA8290, /* untested */ + .tuner_type = TUNER_PHILIPS_TD1316, /* untested */ .radio_type = TUNER_TEA5767, /* untested */ .tuner_addr = ADDR_UNSET, .radio_addr = ADDR_UNSET, -- cgit v0.10.2 From fef4fa1475db6a53237e29451c88c15167d69cc4 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 9 Nov 2006 17:25:28 -0300 Subject: V4L/DVB (4817): Fix uses of "&&" where "&" was intended Fix uses of "&&" where "&" was intended in bttv-cards.c and tveeprom.c Signed-off-by: Jean Delvare Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/video/bt8xx/bttv-cards.c b/drivers/media/video/bt8xx/bttv-cards.c index a84903e..21ebe8f 100644 --- a/drivers/media/video/bt8xx/bttv-cards.c +++ b/drivers/media/video/bt8xx/bttv-cards.c @@ -4001,7 +4001,7 @@ static void __devinit init_PXC200(struct bttv *btv) * - sleep 1ms * - write 0x0E * read from GPIO_DATA into buf (uint_32) - * - if ( buf>>18 & 0x01 ) || ( buf>>19 && 0x01 != 0 ) + * - if ( buf>>18 & 0x01 ) || ( buf>>19 & 0x01 != 0 ) * error. ERROR_CPLD_Check_Failed. */ /* ----------------------------------------------------------------------- */ diff --git a/drivers/media/video/tveeprom.c b/drivers/media/video/tveeprom.c index e6baaee..6b9ef73 100644 --- a/drivers/media/video/tveeprom.c +++ b/drivers/media/video/tveeprom.c @@ -468,7 +468,7 @@ void tveeprom_hauppauge_analog(struct i2c_client *c, struct tveeprom *tvee, (eeprom_data[i+6] << 8) + (eeprom_data[i+7] << 16); - if ( (eeprom_data[i + 8] && 0xf0) && + if ( (eeprom_data[i + 8] & 0xf0) && (tvee->serial_number < 0xffffff) ) { tvee->MAC_address[0] = 0x00; tvee->MAC_address[1] = 0x0D; -- cgit v0.10.2 From 6f36fbb242442184d314e305199bb9a449be4f67 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 9 Nov 2006 17:36:44 -0300 Subject: V4L/DVB (4818): Flexcop-usb: fix debug printk .. fix debug printk. Why, oh why, one would want to do (u16 & 0xff) << 8 and print it with %02x format? Acked-by: Patrick Boettcher Signed-off-by: Alexey Dobriyan Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab diff --git a/drivers/media/dvb/b2c2/flexcop-usb.c b/drivers/media/dvb/b2c2/flexcop-usb.c index 2853ea1..87fb75f 100644 --- a/drivers/media/dvb/b2c2/flexcop-usb.c +++ b/drivers/media/dvb/b2c2/flexcop-usb.c @@ -246,7 +246,7 @@ static int flexcop_usb_i2c_req(struct flexcop_usb *fc_usb, wIndex = (chipaddr << 8 ) | addr; deb_i2c("i2c %2d: %02x %02x %02x %02x %02x %02x\n",func,request_type,req, - ((wValue && 0xff) << 8),wValue >> 8,((wIndex && 0xff) << 8),wIndex >> 8); + wValue & 0xff, wValue >> 8, wIndex & 0xff, wIndex >> 8); len = usb_control_msg(fc_usb->udev,pipe, req, -- cgit v0.10.2 From 2b4ac44e7c7e16cf9411b81693ff3e604f332bf1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Nov 2006 12:27:48 -0800 Subject: [PATCH] vmalloc: optimization, cleanup, bugfixes - reorder 'struct vm_struct' to speedup lookups on CPUS with small cache lines. The fields 'next,addr,size' should be now in the same cache line, to speedup lookups. - One minor cleanup in __get_vm_area_node() - Bugfixes in vmalloc_user() and vmalloc_32_user() NULL returns from __vmalloc() and __find_vm_area() were not tested. [akpm@osdl.org: remove redundant BUG_ONs] Signed-off-by: Eric Dumazet Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index dc9a29d..924e502 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -23,13 +23,14 @@ struct vm_area_struct; #endif struct vm_struct { + /* keep next,addr,size together to speedup lookups */ + struct vm_struct *next; void *addr; unsigned long size; unsigned long flags; struct page **pages; unsigned int nr_pages; unsigned long phys_addr; - struct vm_struct *next; }; /* diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 46606c1..7dc6aa7 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -186,10 +186,8 @@ static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long fl if (unlikely(!area)) return NULL; - if (unlikely(!size)) { - kfree (area); + if (unlikely(!size)) return NULL; - } /* * We always allocate a guard page. @@ -532,11 +530,12 @@ void *vmalloc_user(unsigned long size) void *ret; ret = __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, PAGE_KERNEL); - write_lock(&vmlist_lock); - area = __find_vm_area(ret); - area->flags |= VM_USERMAP; - write_unlock(&vmlist_lock); - + if (ret) { + write_lock(&vmlist_lock); + area = __find_vm_area(ret); + area->flags |= VM_USERMAP; + write_unlock(&vmlist_lock); + } return ret; } EXPORT_SYMBOL(vmalloc_user); @@ -605,11 +604,12 @@ void *vmalloc_32_user(unsigned long size) void *ret; ret = __vmalloc(size, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL); - write_lock(&vmlist_lock); - area = __find_vm_area(ret); - area->flags |= VM_USERMAP; - write_unlock(&vmlist_lock); - + if (ret) { + write_lock(&vmlist_lock); + area = __find_vm_area(ret); + area->flags |= VM_USERMAP; + write_unlock(&vmlist_lock); + } return ret; } EXPORT_SYMBOL(vmalloc_32_user); -- cgit v0.10.2 From d67afe5ed00070de0965bfc98de5f6ed3a80a73e Mon Sep 17 00:00:00 2001 From: David Miller Date: Fri, 10 Nov 2006 12:27:48 -0800 Subject: [PATCH] pci: don't try to remove sysfs files before they are setup. The PCI sysfs attributes are created after the initial PCI bus scan. With the addition of more return value checking and assertions in the device and sysfs layers we now can get dumps like this on sparc64: [ 20.135032] Call Trace: [ 20.135042] [0000000000537f88] pci_remove_bus_device+0x30/0xc0 [ 20.135076] [000000000078f890] pci_fill_in_pbm_cookies+0x98/0x440 [ 20.135109] [000000000042e828] sabre_scan_bus+0x230/0x400 [ 20.135139] [000000000078c710] pcibios_init+0x58/0xa0 [ 20.135159] [0000000000416f14] init+0x9c/0x2e0 [ 20.135190] [0000000000417a50] kernel_thread+0x38/0x60 [ 20.135211] [0000000000417170] rest_init+0x18/0x40 [ 20.135514] PCI0(PBMB): Bus running at 33MHz It's triggering because removal of the "config" PCI sysfs file for the device fails. On sparc64, after probing the device, we'll delete the PCI device via pci_remove_bus_device() if we cannot find the firmware device tree node corresponding to it. This is fine, but at this point the sysfs files for the PCI device won't be setup yet. So we should not try to do anything in pci_remove_sysfs_dev_files() if pci_sysfs_init() has not run yet. Signed-off-by: David S. Miller Acked-by: Greg Kroah-Hartman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index a1d2e97..f952bfe 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -642,6 +642,9 @@ err: */ void pci_remove_sysfs_dev_files(struct pci_dev *pdev) { + if (!sysfs_initialized) + return; + if (pdev->cfg_size < 4096) sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr); else -- cgit v0.10.2 From 1a4b0fc503ff4149f5915be4aeb179b9453cf485 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Fri, 10 Nov 2006 12:27:49 -0800 Subject: [PATCH] mspec driver build fix Fix MSPEC driver to build for non SN2 enabled configs as the driver should work in cached and uncached modes (no fetchop) on these systems. In addition make MSPEC select IA64_UNCACHED_ALLOCATOR, which is required for it and move it to arch/ia64/Kconfig to avoid warnings on non ia64 architectures running allmodconfig. Once the Kconfig code is fixed, we can move it back. Signed-off-by: Jes Sorensen Cc: Fernando Luis Vzquez Cao Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 1468239..683b12c 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -484,6 +484,15 @@ source "net/Kconfig" source "drivers/Kconfig" +config MSPEC + tristate "Memory special operations driver" + depends on IA64 + select IA64_UNCACHED_ALLOCATOR + help + If you have an ia64 and you want to enable memory special + operations support (formerly known as fetchop), say Y here, + otherwise say N. + source "fs/Kconfig" source "lib/Kconfig" diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 39a9f8c..2af12fc 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -409,14 +409,6 @@ config SGI_MBCS If you have an SGI Altix with an attached SABrick say Y or M here, otherwise say N. -config MSPEC - tristate "Memory special operations driver" - depends on IA64 - help - If you have an ia64 and you want to enable memory special - operations support (formerly known as fetchop), say Y here, - otherwise say N. - source "drivers/serial/Kconfig" config UNIX98_PTYS diff --git a/drivers/char/mspec.c b/drivers/char/mspec.c index 5c0dec3..235e892 100644 --- a/drivers/char/mspec.c +++ b/drivers/char/mspec.c @@ -72,7 +72,11 @@ enum { MSPEC_UNCACHED }; +#ifdef CONFIG_SGI_SN static int is_sn2; +#else +#define is_sn2 0 +#endif /* * One of these structures is allocated when an mspec region is mmaped. The @@ -211,7 +215,7 @@ mspec_nopfn(struct vm_area_struct *vma, unsigned long address) if (vdata->type == MSPEC_FETCHOP) paddr = TO_AMO(maddr); else - paddr = __pa(TO_CAC(maddr)); + paddr = maddr & ~__IA64_UNCACHED_OFFSET; pfn = paddr >> PAGE_SHIFT; @@ -335,6 +339,7 @@ mspec_init(void) * The fetchop device only works on SN2 hardware, uncached and cached * memory drivers should both be valid on all ia64 hardware */ +#ifdef CONFIG_SGI_SN if (ia64_platform_is("sn2")) { is_sn2 = 1; if (is_shub2()) { @@ -363,6 +368,7 @@ mspec_init(void) goto free_scratch_pages; } } +#endif ret = misc_register(&cached_miscdev); if (ret) { printk(KERN_ERR "%s: failed to register device %i\n", diff --git a/include/asm-ia64/sn/addrs.h b/include/asm-ia64/sn/addrs.h index 1d9efe5..e715c79 100644 --- a/include/asm-ia64/sn/addrs.h +++ b/include/asm-ia64/sn/addrs.h @@ -136,9 +136,13 @@ */ #define TO_PHYS(x) (TO_PHYS_MASK & (x)) #define TO_CAC(x) (CAC_BASE | TO_PHYS(x)) +#ifdef CONFIG_SGI_SN #define TO_AMO(x) (AMO_BASE | TO_PHYS(x)) #define TO_GET(x) (GET_BASE | TO_PHYS(x)) - +#else +#define TO_AMO(x) ({ BUG(); x; }) +#define TO_GET(x) ({ BUG(); x; }) +#endif /* * Covert from processor physical address to II/TIO physical address: -- cgit v0.10.2 From 7947d2cc2c2e01125a393de83862d02b621999fe Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Fri, 10 Nov 2006 12:27:50 -0800 Subject: [PATCH] IPMI: Fix more && typos Fix improper use of "&&" when "&" was intended. Signed-off-by: Jean Delvare Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 0b07ca1..a41b8df 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -1854,7 +1854,7 @@ static ssize_t provides_dev_sdrs_show(struct device *dev, struct bmc_device *bmc = dev_get_drvdata(dev); return snprintf(buf, 10, "%u\n", - bmc->id.device_revision && 0x80 >> 7); + (bmc->id.device_revision & 0x80) >> 7); } static ssize_t revision_show(struct device *dev, struct device_attribute *attr, @@ -1863,7 +1863,7 @@ static ssize_t revision_show(struct device *dev, struct device_attribute *attr, struct bmc_device *bmc = dev_get_drvdata(dev); return snprintf(buf, 20, "%u\n", - bmc->id.device_revision && 0x0F); + bmc->id.device_revision & 0x0F); } static ssize_t firmware_rev_show(struct device *dev, -- cgit v0.10.2 From e40c67597eac7a0b0e676867517b01a5a57f7b4b Mon Sep 17 00:00:00 2001 From: Wink Saville Date: Fri, 10 Nov 2006 12:27:52 -0800 Subject: [PATCH] Patch for nvidia divide by zero error for 7600 pci-express card The following patch resolves the divide by zero error I encountered on my system: http://marc.10east.com/?l=linux-fbdev-devel&m=116058257024413&w=2 I accomplished this by merging what I thought was appropriate from: http://webcvs.freedesktop.org/xorg/driver/xf86-video-nv/src/ Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/nvidia/nv_hw.c b/drivers/video/nvidia/nv_hw.c index 9ed640d..ea42611 100644 --- a/drivers/video/nvidia/nv_hw.c +++ b/drivers/video/nvidia/nv_hw.c @@ -145,12 +145,18 @@ static void nvGetClocks(struct nvidia_par *par, unsigned int *MClk, if (par->Architecture >= NV_ARCH_40) { pll = NV_RD32(par->PMC, 0x4020); - P = (pll >> 16) & 0x03; + P = (pll >> 16) & 0x07; pll = NV_RD32(par->PMC, 0x4024); M = pll & 0xFF; N = (pll >> 8) & 0xFF; - MB = (pll >> 16) & 0xFF; - NB = (pll >> 24) & 0xFF; + if (((par->Chipset & 0xfff0) == 0x0290) || + ((par->Chipset & 0xfff0) == 0x0390)) { + MB = 1; + NB = 1; + } else { + MB = (pll >> 16) & 0xFF; + NB = (pll >> 24) & 0xFF; + } *MClk = ((N * NB * par->CrystalFreqKHz) / (M * MB)) >> P; pll = NV_RD32(par->PMC, 0x4000); diff --git a/drivers/video/nvidia/nv_setup.c b/drivers/video/nvidia/nv_setup.c index a18a9ae..61dc46f 100644 --- a/drivers/video/nvidia/nv_setup.c +++ b/drivers/video/nvidia/nv_setup.c @@ -359,6 +359,7 @@ int NVCommonSetup(struct fb_info *info) case 0x0186: case 0x0187: case 0x018D: + case 0x0228: case 0x0286: case 0x028C: case 0x0316: @@ -382,6 +383,10 @@ int NVCommonSetup(struct fb_info *info) case 0x034C: case 0x0160: case 0x0166: + case 0x0169: + case 0x016B: + case 0x016C: + case 0x016D: case 0x00C8: case 0x00CC: case 0x0144: @@ -639,12 +644,23 @@ int NVCommonSetup(struct fb_info *info) par->fpHeight = NV_RD32(par->PRAMDAC, 0x0800) + 1; par->fpSyncs = NV_RD32(par->PRAMDAC, 0x0848) & 0x30000033; - printk("Panel size is %i x %i\n", par->fpWidth, par->fpHeight); + printk("nvidiafb: Panel size is %i x %i\n", par->fpWidth, par->fpHeight); } if (monA) info->monspecs = *monA; + if (!par->FlatPanel || !par->twoHeads) + par->FPDither = 0; + + par->LVDS = 0; + if (par->FlatPanel && par->twoHeads) { + NV_WR32(par->PRAMDAC0, 0x08B0, 0x00010004); + if (par->PRAMDAC0[0x08b4] & 1) + par->LVDS = 1; + printk("nvidiafb: Panel is %s\n", par->LVDS ? "LVDS" : "TMDS"); + } + kfree(edidA); kfree(edidB); done: diff --git a/drivers/video/nvidia/nv_type.h b/drivers/video/nvidia/nv_type.h index acdc266..86e65de 100644 --- a/drivers/video/nvidia/nv_type.h +++ b/drivers/video/nvidia/nv_type.h @@ -129,6 +129,7 @@ struct nvidia_par { int fpHeight; int PanelTweak; int paneltweak; + int LVDS; int pm_state; u32 crtcSync_read; u32 fpSyncs; diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c index eb24107..538e947 100644 --- a/drivers/video/nvidia/nvidia.c +++ b/drivers/video/nvidia/nvidia.c @@ -1160,20 +1160,20 @@ static u32 __devinit nvidia_get_arch(struct fb_info *info) case 0x0340: /* GeForceFX 5700 */ arch = NV_ARCH_30; break; - case 0x0040: - case 0x00C0: - case 0x0120: + case 0x0040: /* GeForce 6800 */ + case 0x00C0: /* GeForce 6800 */ + case 0x0120: /* GeForce 6800 */ case 0x0130: - case 0x0140: - case 0x0160: - case 0x01D0: - case 0x0090: - case 0x0210: - case 0x0220: + case 0x0140: /* GeForce 6600 */ + case 0x0160: /* GeForce 6200 */ + case 0x01D0: /* GeForce 7200, 7300, 7400 */ + case 0x0090: /* GeForce 7800 */ + case 0x0210: /* GeForce 6800 */ + case 0x0220: /* GeForce 6200 */ case 0x0230: - case 0x0240: - case 0x0290: - case 0x0390: + case 0x0240: /* GeForce 6100 */ + case 0x0290: /* GeForce 7900 */ + case 0x0390: /* GeForce 7600 */ arch = NV_ARCH_40; break; case 0x0020: /* TNT, TNT2 */ -- cgit v0.10.2 From d8b295f29091310d746509bb6d5828aaf4907a18 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 10 Nov 2006 12:27:53 -0800 Subject: [PATCH] Fix missing parens in set_personality() If you call set_personality() with an expression such as: set_personality(foo ? PERS_FOO1 : PERS_FOO2); then this evaluates to: ((current->personality == foo ? PERS_FOO1 : PERS_FOO2) ? ... which is obviously not the intended result. Add the missing parents to ensure this gets evaluated as expected: ((current->personality == (foo ? PERS_FOO1 : PERS_FOO2)) ? ... Signed-off-by: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/personality.h b/include/linux/personality.h index bf4cf20..012cd55 100644 --- a/include/linux/personality.h +++ b/include/linux/personality.h @@ -114,7 +114,7 @@ struct exec_domain { * Change personality of the currently running process. */ #define set_personality(pers) \ - ((current->personality == pers) ? 0 : __set_personality(pers)) + ((current->personality == (pers)) ? 0 : __set_personality(pers)) #endif /* __KERNEL__ */ -- cgit v0.10.2 From 5fd934a9a1b72c718ad5c388fd2e0fd90486cbf3 Mon Sep 17 00:00:00 2001 From: Franck Bui-Huu Date: Fri, 10 Nov 2006 12:27:53 -0800 Subject: [PATCH] .gitignore: add miscellaneous files Prevent git from reporting this useless status: On branch refs/heads/master Untracked files: (use "git add" to add to commit) TAGS scripts/kconfig/lkc_defs.h scripts/kconfig/qconf.moc nothing to commit Signed-off-by: Franck Bui-Huu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/.gitignore b/.gitignore index e1d5c17..9eb4b77 100644 --- a/.gitignore +++ b/.gitignore @@ -20,6 +20,7 @@ # Top-level generic files # tags +TAGS vmlinux* System.map Module.symvers diff --git a/scripts/kconfig/.gitignore b/scripts/kconfig/.gitignore index e8ad1f6..b49584c 100644 --- a/scripts/kconfig/.gitignore +++ b/scripts/kconfig/.gitignore @@ -6,6 +6,8 @@ lex.*.c *.tab.c *.tab.h zconf.hash.c +*.moc +lkc_defs.h # # configuration programs -- cgit v0.10.2 From 0130b0b32ee53dc7add773fcea984f6a26ef1da3 Mon Sep 17 00:00:00 2001 From: Sharyathi Nagesh Date: Fri, 10 Nov 2006 12:27:54 -0800 Subject: [PATCH] fix Data Acess error in dup_fd On running the Stress Test on machine for more than 72 hours following error message was observed. 0:mon> e cpu 0x0: Vector: 300 (Data Access) at [c00000007ce2f7f0] pc: c000000000060d90: .dup_fd+0x240/0x39c lr: c000000000060d6c: .dup_fd+0x21c/0x39c sp: c00000007ce2fa70 msr: 800000000000b032 dar: ffffffff00000028 dsisr: 40000000 current = 0xc000000074950980 paca = 0xc000000000454500 pid = 27330, comm = bash 0:mon> t [c00000007ce2fa70] c000000000060d28 .dup_fd+0x1d8/0x39c (unreliable) [c00000007ce2fb30] c000000000060f48 .copy_files+0x5c/0x88 [c00000007ce2fbd0] c000000000061f5c .copy_process+0x574/0x1520 [c00000007ce2fcd0] c000000000062f88 .do_fork+0x80/0x1c4 [c00000007ce2fdc0] c000000000011790 .sys_clone+0x5c/0x74 [c00000007ce2fe30] c000000000008950 .ppc_clone+0x8/0xc The problem is because of race window. When if(expand) block is executed in dup_fd unlocking of oldf->file_lock give a window for fdtable in oldf to be modified. So actual open_files in oldf may not match with open_files variable. Cc: Vadim Lobanov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/fork.c b/kernel/fork.c index 3da978e..4b4eab2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -687,6 +687,7 @@ static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) * the latest pointer. */ spin_lock(&oldf->file_lock); + open_files = count_open_files(old_fdt); old_fdt = files_fdtable(oldf); } -- cgit v0.10.2 From f72fa707604c015a6625e80f269506032d5430dc Mon Sep 17 00:00:00 2001 From: Pavel Emelianov Date: Fri, 10 Nov 2006 12:27:56 -0800 Subject: [PATCH] Fix misrouted interrupts deadlocks While testing kernel on machine with "irqpoll" option I've caught such a lockup: __do_IRQ() spin_lock(&desc->lock); desc->chip->ack(); /* IRQ is ACKed */ note_interrupt() misrouted_irq() handle_IRQ_event() if (...) local_irq_enable_in_hardirq(); /* interrupts are enabled from now */ ... __do_IRQ() /* same IRQ we've started from */ spin_lock(&desc->lock); /* LOCKUP */ Looking at misrouted_irq() code I've found that a potential deadlock like this can also take place: 1CPU: __do_IRQ() spin_lock(&desc->lock); /* irq = A */ misrouted_irq() for (i = 1; i < NR_IRQS; i++) { spin_lock(&desc->lock); /* irq = B */ if (desc->status & IRQ_INPROGRESS) { 2CPU: __do_IRQ() spin_lock(&desc->lock); /* irq = B */ misrouted_irq() for (i = 1; i < NR_IRQS; i++) { spin_lock(&desc->lock); /* irq = A */ if (desc->status & IRQ_INPROGRESS) { As the second lock on both CPUs is taken before checking that this irq is being handled in another processor this may cause a deadlock. This issue is only theoretical. I propose the attached patch to fix booth problems: when trying to handle misrouted IRQ active desc->lock may be unlocked. Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 543ea2e..9c7e2e4 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -147,7 +147,11 @@ void note_interrupt(unsigned int irq, struct irq_desc *desc, if (unlikely(irqfixup)) { /* Don't punish working computers */ if ((irqfixup == 2 && irq == 0) || action_ret == IRQ_NONE) { - int ok = misrouted_irq(irq); + int ok; + + spin_unlock(&desc->lock); + ok = misrouted_irq(irq); + spin_lock(&desc->lock); if (action_ret == IRQ_NONE) desc->irqs_unhandled -= ok; } -- cgit v0.10.2 From 09123d230a294cd3b860f4ea042235b988277f0a Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Fri, 10 Nov 2006 12:27:57 -0800 Subject: [PATCH] SCSI core: always store >= 36 bytes of INQUIRY data This patch (as810c) copies a minimum of 36 bytes of INQUIRY data, even if the device claims that not all of them are valid. Often badly behaved devices put plausible data in the Vendor, Product, and Revision strings but set the Additional Length byte to a small value. Using potentially valid data is certainly better than allocating a short buffer and then reading beyond the end of it, which is what we do now. Signed-off-by: Alan Stern Cc: James Bottomley Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index fd9e281..94a2746 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -631,12 +631,22 @@ static int scsi_add_lun(struct scsi_device *sdev, unsigned char *inq_result, * scanning run at their own risk, or supply a user level program * that can correctly scan. */ - sdev->inquiry = kmalloc(sdev->inquiry_len, GFP_ATOMIC); - if (sdev->inquiry == NULL) { + + /* + * Copy at least 36 bytes of INQUIRY data, so that we don't + * dereference unallocated memory when accessing the Vendor, + * Product, and Revision strings. Badly behaved devices may set + * the INQUIRY Additional Length byte to a small value, indicating + * these strings are invalid, but often they contain plausible data + * nonetheless. It doesn't matter if the device sent < 36 bytes + * total, since scsi_probe_lun() initializes inq_result with 0s. + */ + sdev->inquiry = kmemdup(inq_result, + max_t(size_t, sdev->inquiry_len, 36), + GFP_ATOMIC); + if (sdev->inquiry == NULL) return SCSI_SCAN_NO_RESPONSE; - } - memcpy(sdev->inquiry, inq_result, sdev->inquiry_len); sdev->vendor = (char *) (sdev->inquiry + 8); sdev->model = (char *) (sdev->inquiry + 16); sdev->rev = (char *) (sdev->inquiry + 32); -- cgit v0.10.2 From c58121143f87930621c1a6fa9683b6862f2b42c9 Mon Sep 17 00:00:00 2001 From: Hoang-Nam Nguyen Date: Sun, 5 Nov 2006 21:42:56 +0100 Subject: IB/ehca: Use named constant for max mtu Define and use a constant EHCA_MAX_MTU instead hardcoded value. Signed-off-by: Hoang-Nam Nguyen Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/ehca/ehca_av.c b/drivers/infiniband/hw/ehca/ehca_av.c index 3bac197..214e2fd 100644 --- a/drivers/infiniband/hw/ehca/ehca_av.c +++ b/drivers/infiniband/hw/ehca/ehca_av.c @@ -118,8 +118,7 @@ struct ib_ah *ehca_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) } memcpy(&av->av.grh.word_1, &gid, sizeof(gid)); } - /* for the time being we use a hard coded PMTU of 2048 Bytes */ - av->av.pmtu = 4; + av->av.pmtu = EHCA_MAX_MTU; /* dgid comes in grh.word_3 */ memcpy(&av->av.grh.word_3, &ah_attr->grh.dgid, @@ -193,7 +192,7 @@ int ehca_modify_ah(struct ib_ah *ah, struct ib_ah_attr *ah_attr) memcpy(&new_ehca_av.grh.word_1, &gid, sizeof(gid)); } - new_ehca_av.pmtu = 4; /* see also comment in create_ah() */ + new_ehca_av.pmtu = EHCA_MAX_MTU; memcpy(&new_ehca_av.grh.word_3, &ah_attr->grh.dgid, sizeof(ah_attr->grh.dgid)); diff --git a/drivers/infiniband/hw/ehca/hipz_hw.h b/drivers/infiniband/hw/ehca/hipz_hw.h index 3fc92b0..fad9136 100644 --- a/drivers/infiniband/hw/ehca/hipz_hw.h +++ b/drivers/infiniband/hw/ehca/hipz_hw.h @@ -45,6 +45,8 @@ #include "ehca_tools.h" +#define EHCA_MAX_MTU 4 + /* QP Table Entry Memory Map */ struct hipz_qptemm { u64 qpx_hcr; -- cgit v0.10.2 From f2c238a0c5e155acd49752c5fb93fb8d8534232b Mon Sep 17 00:00:00 2001 From: Hoang-Nam Nguyen Date: Sun, 5 Nov 2006 21:42:20 +0100 Subject: IB/ehca: Activate scaling code by default Change ehca's Kconfig to activates scaling code as default. After several measurements we saw that this feature prevents dropped packets (UD) in stress situation. Thus, enabling it helps to improve ehca's bandwidth through IPoIB. Signed-off-by: Hoang-Nam Nguyen Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/ehca/Kconfig b/drivers/infiniband/hw/ehca/Kconfig index 922389b..727b10d 100644 --- a/drivers/infiniband/hw/ehca/Kconfig +++ b/drivers/infiniband/hw/ehca/Kconfig @@ -10,6 +10,7 @@ config INFINIBAND_EHCA config INFINIBAND_EHCA_SCALING bool "Scaling support (EXPERIMENTAL)" depends on IBMEBUS && INFINIBAND_EHCA && HOTPLUG_CPU && EXPERIMENTAL + default y ---help--- eHCA scaling support schedules the CQ callbacks to different CPUs. -- cgit v0.10.2 From 2ffcab6ae44b02679229ca1852526d0a6e062dd2 Mon Sep 17 00:00:00 2001 From: Tom Tucker Date: Wed, 8 Nov 2006 14:23:22 -0600 Subject: RDMA/amso1100: Fix unitialized pseudo_netdev accessed in c2_register_device Rework some load-time error handling: c2_register_device() leaked when it failed, and the function that called it didn't check the return code. Signed-off-by: Tom Tucker Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/amso1100/c2.c b/drivers/infiniband/hw/amso1100/c2.c index 9e7bd94..27fe242 100644 --- a/drivers/infiniband/hw/amso1100/c2.c +++ b/drivers/infiniband/hw/amso1100/c2.c @@ -1155,7 +1155,8 @@ static int __devinit c2_probe(struct pci_dev *pcidev, goto bail10; } - c2_register_device(c2dev); + if (c2_register_device(c2dev)) + goto bail10; return 0; diff --git a/drivers/infiniband/hw/amso1100/c2_provider.c b/drivers/infiniband/hw/amso1100/c2_provider.c index da98d9f..fef9727 100644 --- a/drivers/infiniband/hw/amso1100/c2_provider.c +++ b/drivers/infiniband/hw/amso1100/c2_provider.c @@ -757,20 +757,17 @@ static struct net_device *c2_pseudo_netdev_init(struct c2_dev *c2dev) int c2_register_device(struct c2_dev *dev) { - int ret; + int ret = -ENOMEM; int i; /* Register pseudo network device */ dev->pseudo_netdev = c2_pseudo_netdev_init(dev); - if (dev->pseudo_netdev) { - ret = register_netdev(dev->pseudo_netdev); - if (ret) { - printk(KERN_ERR PFX - "Unable to register netdev, ret = %d\n", ret); - free_netdev(dev->pseudo_netdev); - return ret; - } - } + if (!dev->pseudo_netdev) + goto out3; + + ret = register_netdev(dev->pseudo_netdev); + if (ret) + goto out2; pr_debug("%s:%u\n", __FUNCTION__, __LINE__); strlcpy(dev->ibdev.name, "amso%d", IB_DEVICE_NAME_MAX); @@ -848,21 +845,25 @@ int c2_register_device(struct c2_dev *dev) ret = ib_register_device(&dev->ibdev); if (ret) - return ret; + goto out1; for (i = 0; i < ARRAY_SIZE(c2_class_attributes); ++i) { ret = class_device_create_file(&dev->ibdev.class_dev, c2_class_attributes[i]); - if (ret) { - unregister_netdev(dev->pseudo_netdev); - free_netdev(dev->pseudo_netdev); - ib_unregister_device(&dev->ibdev); - return ret; - } + if (ret) + goto out0; } + goto out3; - pr_debug("%s:%u\n", __FUNCTION__, __LINE__); - return 0; +out0: + ib_unregister_device(&dev->ibdev); +out1: + unregister_netdev(dev->pseudo_netdev); +out2: + free_netdev(dev->pseudo_netdev); +out3: + pr_debug("%s:%u ret=%d\n", __FUNCTION__, __LINE__, ret); + return ret; } void c2_unregister_device(struct c2_dev *dev) -- cgit v0.10.2 From b26c791e9ca3365616d40836000285931ca033d0 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 9 Nov 2006 21:02:26 +0100 Subject: RDMA/amso1100: Fix && typo Fix the AMSO1100 firmware version computation, which was broken due to "&&" being used where "&" should have. Signed-off-by: Jean Delvare Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c index 21d9612..623dc95 100644 --- a/drivers/infiniband/hw/amso1100/c2_rnic.c +++ b/drivers/infiniband/hw/amso1100/c2_rnic.c @@ -157,8 +157,8 @@ static int c2_rnic_query(struct c2_dev *c2dev, struct ib_device_attr *props) props->fw_ver = ((u64)be32_to_cpu(reply->fw_ver_major) << 32) | - ((be32_to_cpu(reply->fw_ver_minor) && 0xFFFF) << 16) | - (be32_to_cpu(reply->fw_ver_patch) && 0xFFFF); + ((be32_to_cpu(reply->fw_ver_minor) & 0xFFFF) << 16) | + (be32_to_cpu(reply->fw_ver_patch) & 0xFFFF); memcpy(&props->sys_image_guid, c2dev->netdev->dev_addr, 6); props->max_mr_size = 0xFFFFFFFF; props->page_size_cap = ~(C2_MIN_PAGESIZE-1); -- cgit v0.10.2 From 39798695b4bcc7b145f8910ca56195808d3a7637 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Mon, 13 Nov 2006 09:38:07 -0800 Subject: IB/mad: Fix race between cancel and receive completion When ib_cancel_mad() is called, it puts the canceled send on a list and schedules a "flushed" callback from process context. However, this leaves a window where a receive completion could be processed before the send is fully flushed. This is fine, except that ib_find_send_mad() will find the MAD and return it to the receive processing, which results in the sender getting both a successful receive and a "flushed" send completion for the same request. Understandably, this confuses the sender, which is expecting only one of these two callbacks, and leads to grief such as a use-after-free in IPoIB. Fix this by changing ib_find_send_mad() to return a send struct only if the status is still successful (and not "flushed"). The search of the send_list already had this check, so this patch just adds the same check to the search of the wait_list. Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 493f4c6..a72bcea 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -1750,7 +1750,7 @@ ib_find_send_mad(struct ib_mad_agent_private *mad_agent_priv, */ (is_direct(wc->recv_buf.mad->mad_hdr.mgmt_class) || rcv_has_same_gid(mad_agent_priv, wr, wc))) - return wr; + return (wr->status == IB_WC_SUCCESS) ? wr : NULL; } /* -- cgit v0.10.2 From 616e8a091a035c0bd9b871695f4af191df123caa Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 13 Nov 2006 18:04:59 +0100 Subject: [PATCH] Fix bad data direction in SG_IO Contrary to what the name misleads you to believe, SG_DXFER_TO_FROM_DEV is really just a normal read seen from the device side. This patch fixes http://lkml.org/lkml/2006/10/13/100 Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index 2dc3264..a19338e 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -246,10 +246,10 @@ static int sg_io(struct file *file, request_queue_t *q, switch (hdr->dxfer_direction) { default: return -EINVAL; - case SG_DXFER_TO_FROM_DEV: case SG_DXFER_TO_DEV: writing = 1; break; + case SG_DXFER_TO_FROM_DEV: case SG_DXFER_FROM_DEV: break; } -- cgit v0.10.2 From b71567312976305cc1ce7e9b71e7378c8bfcb40f Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 13 Nov 2006 18:05:02 +0100 Subject: [PATCH] ide-cd: only set rq->errors SCSI style for block pc requests We should only set ->errors to CHECK_CONDITION and so on for requests that use this field in the SCSI manner. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index bddfebd..8821494 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -724,7 +724,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret) * if we have an error, pass back CHECK_CONDITION as the * scsi status byte */ - if (!rq->errors) + if (blk_pc_request(rq) && !rq->errors) rq->errors = SAM_STAT_CHECK_CONDITION; /* Check for tray open. */ -- cgit v0.10.2 From 4dd7406e9c7e7a5422425ef699780463490b8745 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 13 Nov 2006 09:50:11 -0800 Subject: [dvb saa7134] Fix missing 'break' for avermedia card case MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 450efcfd2e1d941e302a8c89322fbfcef237be98 broke Avermedia 777 support. Added obvious missing "break" statement. Cc: José Suárez Cc: Michael Krufky Cc: Mauro Carvalho Chehab Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c index 7f62403..dee8355 100644 --- a/drivers/media/video/saa7134/saa7134-input.c +++ b/drivers/media/video/saa7134/saa7134-input.c @@ -202,6 +202,7 @@ int saa7134_input_init1(struct saa7134_dev *dev) /* Without this we won't receive key up events */ saa_setb(SAA7134_GPIO_GPMODE1, 0x1); saa_setb(SAA7134_GPIO_GPSTATUS1, 0x1); + break; case SAA7134_BOARD_KWORLD_TERMINATOR: ir_codes = ir_codes_pixelview; mask_keycode = 0x00001f; -- cgit v0.10.2 From b5bf24b94c65536d3cc2bf9039ab05b3967f7b7f Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 8 Nov 2006 16:18:26 +0000 Subject: [PATCH] hpt37x: Check the enablebits Helps for PATA but SATA bridged devices lie and always set all the bits so will need the error handling fixes from Tejun. Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c index 7350443..fce3fcd 100644 --- a/drivers/ata/pata_hpt37x.c +++ b/drivers/ata/pata_hpt37x.c @@ -25,7 +25,7 @@ #include #define DRV_NAME "pata_hpt37x" -#define DRV_VERSION "0.5" +#define DRV_VERSION "0.5.1" struct hpt_clock { u8 xfer_speed; @@ -453,7 +453,13 @@ static int hpt37x_pre_reset(struct ata_port *ap) { u8 scr2, ata66; struct pci_dev *pdev = to_pci_dev(ap->host->dev); - + static const struct pci_bits hpt37x_enable_bits[] = { + { 0x50, 1, 0x04, 0x04 }, + { 0x54, 1, 0x04, 0x04 } + }; + if (!pci_test_config_bits(pdev, &hpt37x_enable_bits[ap->port_no])) + return -ENOENT; + pci_read_config_byte(pdev, 0x5B, &scr2); pci_write_config_byte(pdev, 0x5B, scr2 & ~0x01); /* Cable register now active */ @@ -488,10 +494,17 @@ static void hpt37x_error_handler(struct ata_port *ap) static int hpt374_pre_reset(struct ata_port *ap) { + static const struct pci_bits hpt37x_enable_bits[] = { + { 0x50, 1, 0x04, 0x04 }, + { 0x54, 1, 0x04, 0x04 } + }; u16 mcr3, mcr6; u8 ata66; - struct pci_dev *pdev = to_pci_dev(ap->host->dev); + + if (!pci_test_config_bits(pdev, &hpt37x_enable_bits[ap->port_no])) + return -ENOENT; + /* Do the extra channel work */ pci_read_config_word(pdev, 0x52, &mcr3); pci_read_config_word(pdev, 0x56, &mcr6); -- cgit v0.10.2 From 3f9dd27a22ff79b6b6c4eccd19e4063bff0ddc7e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 10 Nov 2006 22:52:46 +0300 Subject: [PATCH] pata_artop: fix "& (1 >>" typo Signed-off-by: Alexey Dobriyan Signed-off-by: Jeff Garzik diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c index 690828e..96a0980 100644 --- a/drivers/ata/pata_artop.c +++ b/drivers/ata/pata_artop.c @@ -92,7 +92,7 @@ static int artop6260_pre_reset(struct ata_port *ap) return -ENOENT; pci_read_config_byte(pdev, 0x49, &tmp); - if (tmp & (1 >> ap->port_no)) + if (tmp & (1 << ap->port_no)) ap->cbl = ATA_CBL_PATA40; else ap->cbl = ATA_CBL_PATA80; -- cgit v0.10.2 From 253b92ecbd3d2e9f5a79fc7632c89ac74bff16c4 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 14 Nov 2006 09:55:41 -0500 Subject: libata: fix double-completion on error A curious thing happens, however, when ata_qc_new_init fails to get an ata_queued_cmd: First, ata_qc_new_init handles the failure like this: cmd->result = (DID_OK << 16) | (QUEUE_FULL << 1); done(cmd); Then, we return to ata_scsi_translate and do this: err_mem: cmd->result = (DID_ERROR << 16); done(cmd); It appears to me that first we set a status code indicating that we're ok but the device queue is full and finish the command, but then we blow away that status code and replace it with an error flag and finish the command a second time! That does not seem to be desirable behavior since we merely want the I/O to wait until a command slot frees up, not send errors up the block layer. In the err_mem case, we should simply exit out of ata_scsi_translate instead. Signed-off-by: Darrick J. Wong Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 7af2a4b..5c1fc46 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1612,9 +1612,9 @@ early_finish: err_did: ata_qc_free(qc); -err_mem: cmd->result = (DID_ERROR << 16); done(cmd); +err_mem: DPRINTK("EXIT - internal\n"); return 0; -- cgit v0.10.2 From aa8d18731508bb0a16f69b77e5f4541f0d6fb553 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 14 Nov 2006 15:15:40 +0000 Subject: [ARM] Remove OP_MAX_COUNTER OP_MAX_COUNTER never referenced, and is a reminant of an earlier oprofile implementation. Remove it. Signed-off-by: Russell King diff --git a/arch/arm/oprofile/op_counter.h b/arch/arm/oprofile/op_counter.h index 8c5351d..ca942a6 100644 --- a/arch/arm/oprofile/op_counter.h +++ b/arch/arm/oprofile/op_counter.h @@ -10,8 +10,6 @@ #ifndef OP_COUNTER_H #define OP_COUNTER_H -#define OP_MAX_COUNTER 5 - /* Per performance monitor configuration as set via * oprofilefs. */ -- cgit v0.10.2 From 64e72e41acae0dab733fb0d5d789b76d115210c0 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:56:33 +0100 Subject: Revert "[PATCH] MMCONFIG and new Intel motherboards" This reverts 4c6e052adfe285ede5884e4e8c4d33af33932c13 commit. Following Linus' i386 change: revert resource reservation for mmcfg config now. Will be revisited in .20 hopefully. diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index e61093b..f8b6b28 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -163,37 +163,6 @@ static __init void unreachable_devices(void) } } -static __init void pci_mmcfg_insert_resources(void) -{ -#define PCI_MMCFG_RESOURCE_NAME_LEN 19 - int i; - struct resource *res; - char *names; - unsigned num_buses; - - res = kcalloc(PCI_MMCFG_RESOURCE_NAME_LEN + sizeof(*res), - pci_mmcfg_config_num, GFP_KERNEL); - - if (!res) { - printk(KERN_ERR "PCI: Unable to allocate MMCONFIG resources\n"); - return; - } - - names = (void *)&res[pci_mmcfg_config_num]; - for (i = 0; i < pci_mmcfg_config_num; i++, res++) { - num_buses = pci_mmcfg_config[i].end_bus_number - - pci_mmcfg_config[i].start_bus_number + 1; - res->name = names; - snprintf(names, PCI_MMCFG_RESOURCE_NAME_LEN, "PCI MMCONFIG %u", - pci_mmcfg_config[i].pci_segment_group_number); - res->start = pci_mmcfg_config[i].base_address; - res->end = res->start + (num_buses << 20) - 1; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - insert_resource(&iomem_resource, res); - names += PCI_MMCFG_RESOURCE_NAME_LEN; - } -} - void __init pci_mmcfg_init(int type) { int i; @@ -237,7 +206,6 @@ void __init pci_mmcfg_init(int type) } unreachable_devices(); - pci_mmcfg_insert_resources(); raw_pci_ops = &pci_mmcfg; pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF; -- cgit v0.10.2 From 14f448e36192d6d2cd7dfd81cb044977b2f9dd9b Mon Sep 17 00:00:00 2001 From: Aaron Durbin Date: Tue, 14 Nov 2006 16:57:45 +0100 Subject: [PATCH] x86-64: Fix partial page check to ensure unusable memory is not being marked usable. Fix partial page check in e820_register_active_regions to ensure partial pages are not being marked as active in the memory pool. Signed-off-by: Aaron Durbin Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index a75c829..855b561 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -278,7 +278,7 @@ e820_register_active_regions(int nid, unsigned long start_pfn, >> PAGE_SHIFT; /* Skip map entries smaller than a page */ - if (ei_startpfn > ei_endpfn) + if (ei_startpfn >= ei_endpfn) continue; /* Check if end_pfn_map should be updated */ -- cgit v0.10.2 From 14679eb3c50897889ba62f9a37e3bcd8a205b5e7 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Fix PTRACE_[SG]ET_THREAD_AREA regression with ia32 emulation. ptrace(PTRACE_[SG]ET_THREAD_AREA) calls from ia32 code should be passed onto the x86_64 implementation. The default case in sys32_ptrace used to call to sys_ptrace(), but is now EINVAL. This patch fixes a regression caused by that changed. Signed-off-by: Mike McCormack Signed-off-by: Andi Kleen diff --git a/arch/x86_64/ia32/ptrace32.c b/arch/x86_64/ia32/ptrace32.c index 3a7561d..04566fe 100644 --- a/arch/x86_64/ia32/ptrace32.c +++ b/arch/x86_64/ia32/ptrace32.c @@ -244,6 +244,8 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data) case PTRACE_DETACH: case PTRACE_SYSCALL: case PTRACE_SETOPTIONS: + case PTRACE_SET_THREAD_AREA: + case PTRACE_GET_THREAD_AREA: return sys_ptrace(request, pid, addr, data); default: -- cgit v0.10.2 From 51d67a488b53a5cc8401460480c124eaec71e2d4 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: shorten the x86_64 boot setup GDT to what the comment says Stephen Tweedie, Herbert Xu, and myself have been struggling with a very nasty bug in Xen. But it also pointed out a small bug in the x86_64 kernel boot setup. The GDT limit being setup by the initial bzImage code when entering into protected mode is way too big. The comment by the code states that the size of the GDT is 2048, but the actual size being set up is much bigger (32768). This happens simply because of one extra '0'. Instead of setting up a 0x800 size, 0x8000 is set up. On bare metal this is fine because the CPU wont load any segments unless they are explicitly used. But unfortunately, this breaks Xen on vmx FV, since it (for now) blindly loads all the segments into the VMCS if they are less than the gdt limit. Since the real mode segments are around 0x3000, we are getting junk into the VMCS and that later causes an exception. Stephen Tweedie has written up a patch to fix the Xen side and will be submitting that to those folks. But that doesn't excuse the GDT limit being a magnitude too big. AK: changed to compute true gdt size in assembler, fixed comment Signed-off-by: Steven Rostedt Signed-off-by: Andi Kleen diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S index c3bfd22..770940c 100644 --- a/arch/x86_64/boot/setup.S +++ b/arch/x86_64/boot/setup.S @@ -836,13 +836,12 @@ gdt: .word 0x9200 # data read/write .word 0x00CF # granularity = 4096, 386 # (+5th nibble of limit) +gdt_end: idt_48: .word 0 # idt limit = 0 .word 0, 0 # idt base = 0L gdt_48: - .word 0x8000 # gdt limit=2048, - # 256 GDT entries - + .word gdt_end-gdt-1 # gdt limit .word 0, 0 # gdt base (filled in later) # Include video setup & detection code -- cgit v0.10.2 From 5e58a02a8f6a7a1c9ae41f39286bcd3aea0d6f24 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Handle reserve_bootmem_generic beyond end_pfn This can happen on kexec kernels with some configurations, in particularly on Unisys ES7000 systems. Analysis by Amul Shah Cc: Amul Shah Signed-off-by: Andi Kleen diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 971dc11..f1f977a 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -655,9 +655,22 @@ void free_initrd_mem(unsigned long start, unsigned long end) void __init reserve_bootmem_generic(unsigned long phys, unsigned len) { - /* Should check here against the e820 map to avoid double free */ #ifdef CONFIG_NUMA int nid = phys_to_nid(phys); +#endif + unsigned long pfn = phys >> PAGE_SHIFT; + if (pfn >= end_pfn) { + /* This can happen with kdump kernels when accessing firmware + tables. */ + if (pfn < end_pfn_map) + return; + printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %u\n", + phys, len); + return; + } + + /* Should check here against the e820 map to avoid double free */ +#ifdef CONFIG_NUMA reserve_bootmem_node(NODE_DATA(nid), phys, len); #else reserve_bootmem(phys, len); -- cgit v0.10.2 From 15803a43288da434d34d41c4ed650c3c1728d42c Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: setup saved_max_pfn correctly (kdump) x86_64: setup saved_max_pfn correctly 2.6.19-rc4 has broken CONFIG_CRASH_DUMP support on x86_64. It is impossible to read out the kernel contents from /proc/vmcore because saved_max_pfn is set to zero instead of the max_pfn value before the user map is setup. This happens because saved_max_pfn is initialized at parse_early_param() time, and at this time no active regions have been registered. save_max_pfn is setup from e820_end_of_ram(), more exact find_max_pfn_with_active_regions() which returns 0 because no regions exist. This patch fixes this by registering before and removing after the call to e820_end_of_ram(). Signed-off-by: Magnus Damm Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index 855b561..6fe191c 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -594,7 +594,9 @@ static int __init parse_memmap_opt(char *p) * size before original memory map is * reset. */ + e820_register_active_regions(0, 0, -1UL); saved_max_pfn = e820_end_of_ram(); + remove_all_active_ranges(); #endif end_pfn_map = 0; e820.nr_map = 0; -- cgit v0.10.2 From fa18f477d0987c011cce047a7c3cd1284f547a14 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86: Add acpi_user_timer_override option for Asus boards Timer overrides are normally disabled on Nvidia board because they are commonly wrong, except on new ones with HPET support. Unfortunately there are quite some Asus boards around that don't have HPET, but need a timer override. We don't know yet how to handle this transparently, but at least add a command line option to force the timer override and let them boot. Cc: len.brown@intel.com Signed-off-by: Andi Kleen diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index dd00fd5..6747384 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -164,6 +164,10 @@ and is between 256 and 4096 characters. It is defined in the file acpi_skip_timer_override [HW,ACPI] Recognize and ignore IRQ0/pin2 Interrupt Override. For broken nForce2 BIOS resulting in XT-PIC timer. + acpi_use_timer_override [HW,ACPI} + Use timer override. For some broken Nvidia NF5 boards + that require a timer override, but don't have + HPET acpi_dbg_layer= [HW,ACPI] Format: diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index 22e4c466..d12fb97 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -82,6 +82,7 @@ EXPORT_SYMBOL(acpi_strict); acpi_interrupt_flags acpi_sci_flags __initdata; int acpi_sci_override_gsi __initdata; int acpi_skip_timer_override __initdata; +int acpi_use_timer_override __initdata; #ifdef CONFIG_X86_LOCAL_APIC static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; @@ -1300,6 +1301,13 @@ static int __init parse_acpi_skip_timer_override(char *arg) return 0; } early_param("acpi_skip_timer_override", parse_acpi_skip_timer_override); + +static int __init parse_acpi_use_timer_override(char *arg) +{ + acpi_use_timer_override = 1; + return 0; +} +early_param("acpi_use_timer_override", parse_acpi_use_timer_override); #endif /* CONFIG_X86_IO_APIC */ static int __init setup_acpi_sci(char *s) diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c index fe799b1..c984169 100644 --- a/arch/i386/kernel/acpi/earlyquirk.c +++ b/arch/i386/kernel/acpi/earlyquirk.c @@ -27,11 +27,17 @@ static int __init check_bridge(int vendor, int device) #ifdef CONFIG_ACPI /* According to Nvidia all timer overrides are bogus unless HPET is enabled. */ - if (vendor == PCI_VENDOR_ID_NVIDIA) { + if (!acpi_use_timer_override && vendor == PCI_VENDOR_ID_NVIDIA) { nvidia_hpet_detected = 0; acpi_table_parse(ACPI_HPET, nvidia_hpet_check); if (nvidia_hpet_detected == 0) { acpi_skip_timer_override = 1; + printk(KERN_INFO "Nvidia board " + "detected. Ignoring ACPI " + "timer override.\n"); + printk(KERN_INFO "If you got timer trouble " + "try acpi_use_timer_override\n"); + } } #endif diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index 2b1245d..68273bf 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -45,7 +45,13 @@ static void nvidia_bugs(void) /* * All timer overrides on Nvidia are * wrong unless HPET is enabled. + * Unfortunately that's not true on many Asus boards. + * We don't know yet how to detect this automatically, but + * at least allow a command line override. */ + if (acpi_use_timer_override) + return; + nvidia_hpet_detected = 0; acpi_table_parse(ACPI_HPET, nvidia_hpet_check); if (nvidia_hpet_detected == 0) { @@ -53,6 +59,8 @@ static void nvidia_bugs(void) printk(KERN_INFO "Nvidia board " "detected. Ignoring ACPI " "timer override.\n"); + printk(KERN_INFO "If you got timer trouble " + "try acpi_use_timer_override\n"); } #endif /* RED-PEN skip them on mptables too? */ diff --git a/include/asm-i386/acpi.h b/include/asm-i386/acpi.h index 6016632..c80b3a9 100644 --- a/include/asm-i386/acpi.h +++ b/include/asm-i386/acpi.h @@ -132,6 +132,7 @@ extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq); #ifdef CONFIG_X86_IO_APIC extern int acpi_skip_timer_override; +extern int acpi_use_timer_override; #endif static inline void acpi_noirq_set(void) { acpi_noirq = 1; } diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h index ed59aa4..9d1916e 100644 --- a/include/asm-x86_64/acpi.h +++ b/include/asm-x86_64/acpi.h @@ -163,6 +163,7 @@ extern u8 x86_acpiid_to_apicid[]; #define ARCH_HAS_POWER_INIT 1 extern int acpi_skip_timer_override; +extern int acpi_use_timer_override; #endif /*__KERNEL__*/ -- cgit v0.10.2 From 8c131af1db510793f87dc43edbc8950a35370df3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Fix vgetcpu when CONFIG_HOTPLUG_CPU is disabled The vgetcpu per CPU initialization previously relied on CPU hotplug events for all CPUs to initialize the per CPU state. That only worked only on kernels with CONFIG_HOTPLUG_CPU enabled. On the others some CPUs didn't get their state initialized properly and vgetcpu wouldn't work. Change the initialization sequence to instead run in a normal initcall (which runs after the normal CPU bootup) and initialize all running CPUs there. Later hotplug CPUs are still handled with an hotplug notifier. This actually simplifies the code somewhat. Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 4f67697..9f74c88 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -376,9 +376,8 @@ int smp_call_function_single (int cpu, void (*func) (void *info), void *info, /* prevent preemption and reschedule on another processor */ int me = get_cpu(); if (cpu == me) { - WARN_ON(1); put_cpu(); - return -EBUSY; + return 0; } spin_lock_bh(&call_lock); __smp_call_function_single(cpu, func, info, nonatomic, wait); diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 88722f1..e3ef544 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -876,15 +876,6 @@ static struct irqaction irq0 = { timer_interrupt, IRQF_DISABLED, CPU_MASK_NONE, "timer", NULL, NULL }; -static int __cpuinit -time_cpu_notifier(struct notifier_block *nb, unsigned long action, void *hcpu) -{ - unsigned cpu = (unsigned long) hcpu; - if (action == CPU_ONLINE) - vsyscall_set_cpu(cpu); - return NOTIFY_DONE; -} - void __init time_init(void) { if (nohpet) @@ -925,8 +916,6 @@ void __init time_init(void) vxtime.last_tsc = get_cycles_sync(); set_cyc2ns_scale(cpu_khz); setup_irq(0, &irq0); - hotcpu_notifier(time_cpu_notifier, 0); - time_cpu_notifier(NULL, CPU_ONLINE, (void *)(long)smp_processor_id()); #ifndef CONFIG_SMP time_init_gtod(); diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index a98b460..a730bac 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -27,6 +27,9 @@ #include #include #include +#include +#include +#include #include #include @@ -243,32 +246,17 @@ static ctl_table kernel_root_table2[] = { #endif -static void __cpuinit write_rdtscp_cb(void *info) -{ - write_rdtscp_aux((unsigned long)info); -} - -void __cpuinit vsyscall_set_cpu(int cpu) +/* Assume __initcall executes before all user space. Hopefully kmod + doesn't violate that. We'll find out if it does. */ +static void __cpuinit vsyscall_set_cpu(int cpu) { unsigned long *d; unsigned long node = 0; #ifdef CONFIG_NUMA node = cpu_to_node[cpu]; #endif - if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) { - void *info = (void *)((node << 12) | cpu); - /* Can happen on preemptive kernel */ - if (get_cpu() == cpu) - write_rdtscp_cb(info); -#ifdef CONFIG_SMP - else { - /* the notifier is unfortunately not executed on the - target CPU */ - smp_call_function_single(cpu,write_rdtscp_cb,info,0,1); - } -#endif - put_cpu(); - } + if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) + write_rdtscp_aux((node << 12) | cpu); /* Store cpu number in limit so that it can be loaded quickly in user space in vgetcpu. @@ -280,6 +268,21 @@ void __cpuinit vsyscall_set_cpu(int cpu) *d |= (node >> 4) << 48; } +static void __cpuinit cpu_vsyscall_init(void *arg) +{ + /* preemption should be already off */ + vsyscall_set_cpu(raw_smp_processor_id()); +} + +static int __cpuinit +cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) +{ + long cpu = (long)arg; + if (action == CPU_ONLINE) + smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); + return NOTIFY_DONE; +} + static void __init map_vsyscall(void) { extern char __vsyscall_0; @@ -299,6 +302,8 @@ static int __init vsyscall_init(void) #ifdef CONFIG_SYSCTL register_sysctl_table(kernel_root_table2, 0); #endif + on_each_cpu(cpu_vsyscall_init, NULL, 0, 1); + hotcpu_notifier(cpu_vsyscall_notifier, 0); return 0; } diff --git a/include/asm-x86_64/vsyscall.h b/include/asm-x86_64/vsyscall.h index fd452fc..01d1c17 100644 --- a/include/asm-x86_64/vsyscall.h +++ b/include/asm-x86_64/vsyscall.h @@ -59,8 +59,6 @@ extern seqlock_t xtime_lock; extern int sysctl_vsyscall; -extern void vsyscall_set_cpu(int cpu); - #define ARCH_HAVE_XTIME_LOCK 1 #endif /* __KERNEL__ */ -- cgit v0.10.2 From 9446868b5383eb87f76b2d4389dea4bb968a6657 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Fix race in exit_idle When another interrupt happens in exit_idle the exit idle notifier could be called an incorrect number of times. Add a test_and_clear_bit_pda and use it handle the bit atomically against interrupts to avoid this. Pointed out by Stephane Eranian Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 49f7fac..f622605 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -88,9 +88,8 @@ void enter_idle(void) static void __exit_idle(void) { - if (read_pda(isidle) == 0) + if (test_and_clear_bit_pda(0, isidle) == 0) return; - write_pda(isidle, 0); atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL); } diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h index 14996d9..5642634 100644 --- a/include/asm-x86_64/pda.h +++ b/include/asm-x86_64/pda.h @@ -109,6 +109,15 @@ extern struct x8664_pda _proxy_pda; #define sub_pda(field,val) pda_to_op("sub",field,val) #define or_pda(field,val) pda_to_op("or",field,val) +/* This is not atomic against other CPUs -- CPU preemption needs to be off */ +#define test_and_clear_bit_pda(bit,field) ({ \ + int old__; \ + asm volatile("btr %2,%%gs:%c3\n\tsbbl %0,%0" \ + : "=r" (old__), "+m" (_proxy_pda.field) \ + : "dIr" (bit), "i" (pda_offset(field)) : "memory"); \ + old__; \ +}) + #endif #define PDA_STACKOFFSET (5*8) -- cgit v0.10.2 From 8b126b77536186eef69d408eb7959ce7f558f251 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 14 Nov 2006 02:03:23 -0800 Subject: [PATCH] setup_irq(): better mismatch debugging When we get a mismatch between handlers on the same IRQ, all we get is "IRQ handler type mismatch for IRQ n". Let's print the name of the presently-registered handler with which we got the mismatch. Cc: Ingo Molnar Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 6879202..b385878 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -216,6 +216,7 @@ int setup_irq(unsigned int irq, struct irqaction *new) { struct irq_desc *desc = irq_desc + irq; struct irqaction *old, **p; + const char *old_name = NULL; unsigned long flags; int shared = 0; @@ -255,8 +256,10 @@ int setup_irq(unsigned int irq, struct irqaction *new) * set the trigger type must match. */ if (!((old->flags & new->flags) & IRQF_SHARED) || - ((old->flags ^ new->flags) & IRQF_TRIGGER_MASK)) + ((old->flags ^ new->flags) & IRQF_TRIGGER_MASK)) { + old_name = old->name; goto mismatch; + } #if defined(CONFIG_IRQ_PER_CPU) /* All handlers must agree on per-cpuness */ @@ -322,11 +325,13 @@ int setup_irq(unsigned int irq, struct irqaction *new) return 0; mismatch: - spin_unlock_irqrestore(&desc->lock, flags); if (!(new->flags & IRQF_PROBE_SHARED)) { printk(KERN_ERR "IRQ handler type mismatch for IRQ %d\n", irq); + if (old_name) + printk(KERN_ERR "current handler: %s\n", old_name); dump_stack(); } + spin_unlock_irqrestore(&desc->lock, flags); return -EBUSY; } -- cgit v0.10.2 From f3ac84324fd949f671e6cf5620add5de02498386 Mon Sep 17 00:00:00 2001 From: Daniel Ritz Date: Tue, 14 Nov 2006 02:03:25 -0800 Subject: [PATCH] fix via586 irq routing for pirq 5 Fix interrupt routing for via 586 bridges. pirq can be 5 which needs to be mapped to INTD. But currently the access functions can handle only pirq 1-4. this is similar to the other via chipsets where pirq 4 and 5 are both mapped to INTD. Fixes bugzilla #7490 Cc: Daniel Paschka Cc: Adrian Bunk Signed-off-by: Daniel Ritz Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index dbc4aae..6916399 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -255,13 +255,13 @@ static int pirq_via_set(struct pci_dev *router, struct pci_dev *dev, int pirq, i */ static int pirq_via586_get(struct pci_dev *router, struct pci_dev *dev, int pirq) { - static const unsigned int pirqmap[4] = { 3, 2, 5, 1 }; + static const unsigned int pirqmap[5] = { 3, 2, 5, 1, 1 }; return read_config_nybble(router, 0x55, pirqmap[pirq-1]); } static int pirq_via586_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq) { - static const unsigned int pirqmap[4] = { 3, 2, 5, 1 }; + static const unsigned int pirqmap[5] = { 3, 2, 5, 1, 1 }; write_config_nybble(router, 0x55, pirqmap[pirq-1], irq); return 1; } -- cgit v0.10.2 From d8f7975159f35846754d3845c9701b612c5c0624 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 14 Nov 2006 02:03:26 -0800 Subject: [PATCH] revert "PCI: quirk for IBM Dock II cardbus controllers" Fix http://bugzilla.kernel.org/show_bug.cgi?id=7264 We need to target this quirk a little more tightly, using the T20 DMI string. Cc: Pavel Kysilka Acked-by: Kristen Carlson Accardi Cc: Greg Kroah-Hartman Cc: Dominik Brodowski Cc: Daniel Ritz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 204b1c8..5b44838 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1460,33 +1460,6 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, 0x2609, quirk_intel_pcie_pm); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, 0x260a, quirk_intel_pcie_pm); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, 0x260b, quirk_intel_pcie_pm); -/* - * Fixup the cardbus bridges on the IBM Dock II docking station - */ -static void __devinit quirk_ibm_dock2_cardbus(struct pci_dev *dev) -{ - u32 val; - - /* - * tie the 2 interrupt pins to INTA, and configure the - * multifunction routing register to handle this. - */ - if ((dev->subsystem_vendor == PCI_VENDOR_ID_IBM) && - (dev->subsystem_device == 0x0148)) { - printk(KERN_INFO "PCI: Found IBM Dock II Cardbus Bridge " - "applying quirk\n"); - pci_read_config_dword(dev, 0x8c, &val); - val = ((val & 0xffffff00) | 0x1002); - pci_write_config_dword(dev, 0x8c, val); - pci_read_config_dword(dev, 0x80, &val); - val = ((val & 0x00ffff00) | 0x2864c077); - pci_write_config_dword(dev, 0x80, val); - } -} - -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_TI, PCI_DEVICE_ID_TI_1420, - quirk_ibm_dock2_cardbus); - static void __devinit quirk_netmos(struct pci_dev *dev) { unsigned int num_parallel = (dev->subsystem_device & 0xf0) >> 4; -- cgit v0.10.2 From 6a34b57bec41c95f1e38f700cd9b81324baaffc7 Mon Sep 17 00:00:00 2001 From: Nicolas Kaiser Date: Tue, 14 Nov 2006 02:03:28 -0800 Subject: [PATCH] drivers/ide: stray bracket Stray bracket in debug code. Signed-off-by: Nicolas Kaiser Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/legacy/hd.c b/drivers/ide/legacy/hd.c index b1d5291..45ed035 100644 --- a/drivers/ide/legacy/hd.c +++ b/drivers/ide/legacy/hd.c @@ -459,7 +459,7 @@ ok_to_read: #ifdef DEBUG printk("%s: read: sector %ld, remaining = %ld, buffer=%p\n", req->rq_disk->disk_name, req->sector, req->nr_sectors, - req->buffer+512)); + req->buffer+512); #endif if (req->current_nr_sectors <= 0) end_request(req, 1); -- cgit v0.10.2 From ba8df43c0ee93ec05fc526278a80aaf4cb5ab1fa Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Tue, 14 Nov 2006 02:03:29 -0800 Subject: [PATCH] autofs4: panic after mount fail Resolve the panic on failed mount of an autofs filesystem originally reported by Mao Bibo. It addresses two issues that happen after the mount fail. The first a NULL pointer reference to a field (pipe) in the autofs superblock info structure and second the lack of super block cleanup by the autofs and autofs4 modules. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index 54c518c..38ede5c 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -25,6 +25,14 @@ void autofs_kill_sb(struct super_block *sb) struct autofs_sb_info *sbi = autofs_sbi(sb); unsigned int n; + /* + * In the event of a failure in get_sb_nodev the superblock + * info is not present so nothing else has been setup, so + * just exit when we are called from deactivate_super. + */ + if (!sbi) + return; + if ( !sbi->catatonic ) autofs_catatonic_mode(sbi); /* Free wait queues, close pipe */ @@ -136,7 +144,8 @@ int autofs_fill_super(struct super_block *s, void *data, int silent) s->s_fs_info = sbi; sbi->magic = AUTOFS_SBI_MAGIC; - sbi->catatonic = 0; + sbi->pipe = NULL; + sbi->catatonic = 1; sbi->exp_timeout = 0; sbi->oz_pgrp = process_group(current); autofs_initialize_hash(&sbi->dirhash); @@ -180,6 +189,7 @@ int autofs_fill_super(struct super_block *s, void *data, int silent) if ( !pipe->f_op || !pipe->f_op->write ) goto fail_fput; sbi->pipe = pipe; + sbi->catatonic = 0; /* * Success! Install the root dentry now to indicate completion. @@ -198,6 +208,8 @@ fail_iput: iput(root_inode); fail_free: kfree(sbi); + s->s_fs_info = NULL; + kill_anon_super(s); fail_unlock: return -EINVAL; } diff --git a/fs/autofs/waitq.c b/fs/autofs/waitq.c index 633f628..19a9caf 100644 --- a/fs/autofs/waitq.c +++ b/fs/autofs/waitq.c @@ -41,6 +41,7 @@ void autofs_catatonic_mode(struct autofs_sb_info *sbi) wq = nwq; } fput(sbi->pipe); /* Close the pipe */ + sbi->pipe = NULL; autofs_hash_dputall(&sbi->dirhash); /* Remove all dentry pointers */ } diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 51fd859..ce7c0f1 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -99,6 +99,9 @@ static void autofs4_force_release(struct autofs_sb_info *sbi) struct dentry *this_parent = sbi->sb->s_root; struct list_head *next; + if (!sbi->sb->s_root) + return; + spin_lock(&dcache_lock); repeat: next = this_parent->d_subdirs.next; @@ -146,6 +149,14 @@ void autofs4_kill_sb(struct super_block *sb) { struct autofs_sb_info *sbi = autofs4_sbi(sb); + /* + * In the event of a failure in get_sb_nodev the superblock + * info is not present so nothing else has been setup, so + * just exit when we are called from deactivate_super. + */ + if (!sbi) + return; + sb->s_fs_info = NULL; if ( !sbi->catatonic ) @@ -310,7 +321,8 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) s->s_fs_info = sbi; sbi->magic = AUTOFS_SBI_MAGIC; sbi->pipefd = -1; - sbi->catatonic = 0; + sbi->pipe = NULL; + sbi->catatonic = 1; sbi->exp_timeout = 0; sbi->oz_pgrp = process_group(current); sbi->sb = s; @@ -388,6 +400,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) goto fail_fput; sbi->pipe = pipe; sbi->pipefd = pipefd; + sbi->catatonic = 0; /* * Success! Install the root dentry now to indicate completion. @@ -412,6 +425,8 @@ fail_ino: kfree(ino); fail_free: kfree(sbi); + s->s_fs_info = NULL; + kill_anon_super(s); fail_unlock: return -EINVAL; } diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index c0a6c8d..1e4a539 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -41,10 +41,8 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) wake_up_interruptible(&wq->queue); wq = nwq; } - if (sbi->pipe) { - fput(sbi->pipe); /* Close the pipe */ - sbi->pipe = NULL; - } + fput(sbi->pipe); /* Close the pipe */ + sbi->pipe = NULL; } static int autofs4_write(struct file *file, const void *addr, int bytes) -- cgit v0.10.2 From d6e89cb6cd3a10eb203914093642f580c20476d4 Mon Sep 17 00:00:00 2001 From: Nathan Lynch Date: Tue, 14 Nov 2006 02:03:30 -0800 Subject: [PATCH] nvidiafb: fix unreachable code in nv10GetConfig Fix binary/logical operator typo which leads to unreachable code. Noticed while looking at other issues; I don't have the relevant hardware to test this. Signed-off-by: Nathan Lynch Cc: "Antonino A. Daplas" Acked-by: James Simmons Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/nvidia/nv_setup.c b/drivers/video/nvidia/nv_setup.c index 61dc46f..eab3e28 100644 --- a/drivers/video/nvidia/nv_setup.c +++ b/drivers/video/nvidia/nv_setup.c @@ -262,7 +262,7 @@ static void nv10GetConfig(struct nvidia_par *par) #endif dev = pci_find_slot(0, 1); - if ((par->Chipset && 0xffff) == 0x01a0) { + if ((par->Chipset & 0xffff) == 0x01a0) { int amt = 0; pci_read_config_dword(dev, 0x7c, &amt); -- cgit v0.10.2 From 69ae9e3ee4ce99140a7db424bebf55d8d180da2f Mon Sep 17 00:00:00 2001 From: David Brownell Date: Tue, 14 Nov 2006 02:03:31 -0800 Subject: [PATCH] usb: MAINTAINERS updates Looks like I still take care of the USB gadget/peripheral framework. Signed-off-by: David Brownell Acked-by: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 8732dae..a5508f9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3072,6 +3072,13 @@ L: video4linux-list@redhat.com W: http://www.linux-projects.org S: Maintained +USB GADGET/PERIPHERAL SUBSYSTEM +P: David Brownell +M: dbrownell@users.sourceforge.net +L: linux-usb-devel@lists.sourceforge.net +W: http://www.linux-usb.org/gadget +S: Maintained + USB HID/HIDBP DRIVERS P: Vojtech Pavlik M: vojtech@suse.cz @@ -3255,10 +3262,11 @@ L: linux-usb-users@lists.sourceforge.net L: linux-usb-devel@lists.sourceforge.net S: Maintained -USB "USBNET" DRIVER +USB "USBNET" DRIVER FRAMEWORK P: David Brownell M: dbrownell@users.sourceforge.net L: linux-usb-devel@lists.sourceforge.net +W: http://www.linux-usb.org/usbnet S: Maintained USB W996[87]CF DRIVER -- cgit v0.10.2 From 68589bc353037f233fe510ad9ff432338c95db66 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 14 Nov 2006 02:03:32 -0800 Subject: [PATCH] hugetlb: prepare_hugepage_range check offset too (David:) If hugetlbfs_file_mmap() returns a failure to do_mmap_pgoff() - for example, because the given file offset is not hugepage aligned - then do_mmap_pgoff will go to the unmap_and_free_vma backout path. But at this stage the vma hasn't been marked as hugepage, and the backout path will call unmap_region() on it. That will eventually call down to the non-hugepage version of unmap_page_range(). On ppc64, at least, that will cause serious problems if there are any existing hugepage pagetable entries in the vicinity - for example if there are any other hugepage mappings under the same PUD. unmap_page_range() will trigger a bad_pud() on the hugepage pud entries. I suspect this will also cause bad problems on ia64, though I don't have a machine to test it on. (Hugh:) prepare_hugepage_range() should check file offset alignment when it checks virtual address and length, to stop MAP_FIXED with a bad huge offset from unmapping before it fails further down. PowerPC should apply the same prepare_hugepage_range alignment checks as ia64 and all the others do. Then none of the alignment checks in hugetlbfs_file_mmap are required (nor is the check for too small a mapping); but even so, move up setting of VM_HUGETLB and add a comment to warn of what David Gibson discovered - if hugetlbfs_file_mmap fails before setting it, do_mmap_pgoff's unmap_region when unwinding from error will go the non-huge way, which may cause bad behaviour on architectures (powerpc and ia64) which segregate their huge mappings into a separate region of the address space. Signed-off-by: Hugh Dickins Cc: "Luck, Tony" Cc: "David S. Miller" Acked-by: Adam Litke Acked-by: David Gibson Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index eee5c1cf..f3a9585 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -70,8 +70,10 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) * Don't actually need to do any preparation, but need to make sure * the address is in the right region. */ -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff) { + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; if (len & ~HPAGE_MASK) return -EINVAL; if (addr & ~HPAGE_MASK) diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index fd68b74..506d897 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -491,11 +491,15 @@ static int open_high_hpage_areas(struct mm_struct *mm, u16 newareas) return 0; } -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff) { int err = 0; - if ( (addr+len) < addr ) + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; + if (len & ~HPAGE_MASK) + return -EINVAL; + if (addr & ~HPAGE_MASK) return -EINVAL; if (addr < 0x100000000UL) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 0bea6a6..7f47569 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -62,24 +62,19 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) loff_t len, vma_len; int ret; - if (vma->vm_pgoff & (HPAGE_SIZE / PAGE_SIZE - 1)) - return -EINVAL; - - if (vma->vm_start & ~HPAGE_MASK) - return -EINVAL; - - if (vma->vm_end & ~HPAGE_MASK) - return -EINVAL; - - if (vma->vm_end - vma->vm_start < HPAGE_SIZE) - return -EINVAL; + /* + * vma alignment has already been checked by prepare_hugepage_range. + * If you add any error returns here, do so after setting VM_HUGETLB, + * so is_vm_hugetlb_page tests below unmap_region go the right way + * when do_mmap_pgoff unwinds (may be important on powerpc and ia64). + */ + vma->vm_flags |= VM_HUGETLB | VM_RESERVED; + vma->vm_ops = &hugetlb_vm_ops; vma_len = (loff_t)(vma->vm_end - vma->vm_start); mutex_lock(&inode->i_mutex); file_accessed(file); - vma->vm_flags |= VM_HUGETLB | VM_RESERVED; - vma->vm_ops = &hugetlb_vm_ops; ret = -ENOMEM; len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 5081d27..ace64e5 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -60,8 +60,11 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(unsigned long addr, unsigned long len, + pgoff_t pgoff) { + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; if (len & ~HPAGE_MASK) return -EINVAL; if (addr & ~HPAGE_MASK) @@ -69,7 +72,8 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) return 0; } #else -int prepare_hugepage_range(unsigned long addr, unsigned long len); +int prepare_hugepage_range(unsigned long addr, unsigned long len, + pgoff_t pgoff); #endif #ifndef ARCH_HAS_SETCLEAR_HUGE_PTE @@ -107,7 +111,7 @@ static inline unsigned long hugetlb_total_pages(void) #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL -#define prepare_hugepage_range(addr, len) (-EINVAL) +#define prepare_hugepage_range(addr,len,pgoff) (-EINVAL) #define pmd_huge(x) 0 #define is_hugepage_only_range(mm, addr, len) 0 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; }) diff --git a/mm/mmap.c b/mm/mmap.c index 497e502..bdace87 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1379,7 +1379,7 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, * Check if the given range is hugepage aligned, and * can be made suitable for hugepages. */ - ret = prepare_hugepage_range(addr, len); + ret = prepare_hugepage_range(addr, len, pgoff); } else { /* * Ensure that a normal request is not falling in a -- cgit v0.10.2 From cb07c9a1864a8eac9f3123e428100d5b2a16e65a Mon Sep 17 00:00:00 2001 From: David Gibson Date: Tue, 14 Nov 2006 02:03:38 -0800 Subject: [PATCH] hugetlb: check for brk() entering a hugepage region Unlike mmap(), the codepath for brk() creates a vma without first checking that it doesn't touch a region exclusively reserved for hugepages. On powerpc, this can allow it to create a normal page vma in a hugepage region, causing oopses and other badness. Add a test to prevent this. With this patch, brk() will simply fail if it attempts to move the break into a hugepage reserved region. Signed-off-by: David Gibson Cc: Adam Litke Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/mmap.c b/mm/mmap.c index bdace87..2526463 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1880,6 +1880,10 @@ unsigned long do_brk(unsigned long addr, unsigned long len) if ((addr + len) > TASK_SIZE || (addr + len) < addr) return -EINVAL; + error = is_hugepage_only_range(current->mm, addr, len); + if (error) + return error; + flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; error = arch_mmap_check(addr, len, flags); -- cgit v0.10.2 From a4625085445b86951d8482c0cdd6d52719f7c323 Mon Sep 17 00:00:00 2001 From: Brian King Date: Mon, 13 Nov 2006 16:32:36 -0600 Subject: [PATCH] libata: Convert from module_init to subsys_initcall When building a monolithic kernel, the load order of drivers does not work for SAS libata users, resulting in a kernel oops. Convert libata to use subsys_initcall instead of module_init, which ensures that libata gets loaded before any LLDD. This is the same thing that scsi core does to solve the problem. The load order problem was observed on ipr SAS adapters and should exist for other SAS users as well. Signed-off-by: Brian King Acked-by: Jeff Garzik Signed-off-by: Linus Torvalds diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index a8fd0c3..915a55a 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -5957,7 +5957,7 @@ static void __exit ata_exit(void) destroy_workqueue(ata_aux_wq); } -module_init(ata_init); +subsys_initcall(ata_init); module_exit(ata_exit); static unsigned long ratelimit_time; -- cgit v0.10.2 From b369c2cfa47bc0ad495a95fe9a17c9888781d615 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 14 Nov 2006 12:36:03 +0100 Subject: [PATCH] cciss: fix iostat cciss needs to call disk_stat_add() for iostat to work. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 6ffe2b2..4105c3b 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -1300,6 +1300,12 @@ static void cciss_softirq_done(struct request *rq) complete_buffers(rq->bio, rq->errors); + if (blk_fs_request(rq)) { + const int rw = rq_data_dir(rq); + + disk_stat_add(rq->rq_disk, sectors[rw], rq->nr_sectors); + } + #ifdef CCISS_DEBUG printk("Done with %p\n", rq); #endif /* CCISS_DEBUG */ -- cgit v0.10.2 From 1f794b6082a5ff88f7c48d1634056026acf806f4 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 14 Nov 2006 12:36:45 +0100 Subject: [PATCH] cpqarray: fix iostat cpqarray needs to call disk_stat_add() for iostat to work. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index 570d2f0..d5f519e 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -998,6 +998,7 @@ static inline void complete_buffers(struct bio *bio, int ok) */ static inline void complete_command(cmdlist_t *cmd, int timeout) { + struct request *rq = cmd->rq; int ok=1; int i, ddir; @@ -1029,12 +1030,18 @@ static inline void complete_command(cmdlist_t *cmd, int timeout) pci_unmap_page(hba[cmd->ctlr]->pci_dev, cmd->req.sg[i].addr, cmd->req.sg[i].size, ddir); - complete_buffers(cmd->rq->bio, ok); + complete_buffers(rq->bio, ok); - add_disk_randomness(cmd->rq->rq_disk); + if (blk_fs_request(rq)) { + const int rw = rq_data_dir(rq); - DBGPX(printk("Done with %p\n", cmd->rq);); - end_that_request_last(cmd->rq, ok ? 1 : -EIO); + disk_stat_add(rq->rq_disk, sectors[rw], rq->nr_sectors); + } + + add_disk_randomness(rq->rq_disk); + + DBGPX(printk("Done with %p\n", rq);); + end_that_request_last(rq, ok ? 1 : -EIO); } /* -- cgit v0.10.2 From cd2579d7aa7bfc966cc271a88e77f8cfc3b0b7ba Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 14 Nov 2006 13:43:38 +0000 Subject: [PATCH] hugetlb: fix error return for brk() entering a hugepage region Commit cb07c9a1864a8eac9f3123e428100d5b2a16e65a causes the wrong return value. is_hugepage_only_range() is a boolean, so we should return -EINVAL rather than 1. Also - we can use "mm" instead of looking up "current->mm" again. Signed-off-by: Hugh Dickins Signed-off-by: Linus Torvalds diff --git a/mm/mmap.c b/mm/mmap.c index 2526463..7b40abd 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1880,9 +1880,8 @@ unsigned long do_brk(unsigned long addr, unsigned long len) if ((addr + len) > TASK_SIZE || (addr + len) < addr) return -EINVAL; - error = is_hugepage_only_range(current->mm, addr, len); - if (error) - return error; + if (is_hugepage_only_range(mm, addr, len)) + return -EINVAL; flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; -- cgit v0.10.2 From 9a3a04ac386f44175b6a4142eaeab3d4170a57f3 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 14 Nov 2006 15:20:51 -0800 Subject: Revert "[PATCH] fix Data Acess error in dup_fd" This reverts commit 0130b0b32ee53dc7add773fcea984f6a26ef1da3. Sergey Vlasov points out (and Vadim Lobanov concurs) that the bug it was supposed to fix must be some unrelated memory corruption, and the "fix" actually causes more problems: "However, the new code does not look safe in all cases. If some other task has opened more files while dup_fd() released oldf->file_lock, the new code will update open_files to the new larger value. But newf was allocated with the old smaller value of open_files, therefore subsequent accesses to newf may try to write into unallocated memory." so revert it. Cc: Sharyathi Nagesh Cc: Sergey Vlasov Cc: Vadim Lobanov Cc: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/fork.c b/kernel/fork.c index 4b4eab2..3da978e 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -687,7 +687,6 @@ static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) * the latest pointer. */ spin_lock(&oldf->file_lock); - open_files = count_open_files(old_fdt); old_fdt = files_fdtable(oldf); } -- cgit v0.10.2 From c387fd85f84b9d89a75596325d8d6a0f730baf64 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Wed, 15 Nov 2006 00:30:17 +0100 Subject: [PATCH] Char: isicom, fix close bug port is dereferenced even if it is NULL. Dereference it _after_ the check if (!port)... Thanks Eric for reporting this. This fixes http://bugzilla.kernel.org/show_bug.cgi?id=7527 Signed-off-by: Jiri Slaby Signed-off-by: Linus Torvalds diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c index e9e9bf3..58c955e 100644 --- a/drivers/char/isicom.c +++ b/drivers/char/isicom.c @@ -1062,11 +1062,12 @@ static void isicom_shutdown_port(struct isi_port *port) static void isicom_close(struct tty_struct *tty, struct file *filp) { struct isi_port *port = tty->driver_data; - struct isi_board *card = port->card; + struct isi_board *card; unsigned long flags; if (!port) return; + card = port->card; if (isicom_paranoia_check(port, tty->name, "isicom_close")) return; -- cgit v0.10.2 From 134a11f0c37c043d3ea557ea15b95b084e3cc2c8 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 10 Nov 2006 12:08:37 +0100 Subject: [PATCH] ALSA: hda-intel - Disable MSI support by default Disable MSI support on HD-audio driver as default since there are too many broken devices. The module option is changed from disable_msi to enable_msi, too. For turning MSI support on, pass enable_msi=1, instead. Signed-off-by: Takashi Iwai Signed-off-by: Linus Torvalds diff --git a/Documentation/sound/alsa/ALSA-Configuration.txt b/Documentation/sound/alsa/ALSA-Configuration.txt index 138673a..3472d9c 100644 --- a/Documentation/sound/alsa/ALSA-Configuration.txt +++ b/Documentation/sound/alsa/ALSA-Configuration.txt @@ -753,7 +753,7 @@ Prior to version 0.9.0rc4 options had a 'snd_' prefix. This was removed. position_fix - Fix DMA pointer (0 = auto, 1 = none, 2 = POSBUF, 3 = FIFO size) single_cmd - Use single immediate commands to communicate with codecs (for debugging only) - disable_msi - Disable Message Signaled Interrupt (MSI) + enable_msi - Enable Message Signaled Interrupt (MSI) (default = off) This module supports one card and autoprobe. diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 0e292dc..e35cfd3 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -55,7 +55,7 @@ static char *model; static int position_fix; static int probe_mask = -1; static int single_cmd; -static int disable_msi; +static int enable_msi; module_param(index, int, 0444); MODULE_PARM_DESC(index, "Index value for Intel HD audio interface."); @@ -69,8 +69,8 @@ module_param(probe_mask, int, 0444); MODULE_PARM_DESC(probe_mask, "Bitmask to probe codecs (default = -1)."); module_param(single_cmd, bool, 0444); MODULE_PARM_DESC(single_cmd, "Use single command to communicate with codecs (for debugging only)."); -module_param(disable_msi, int, 0); -MODULE_PARM_DESC(disable_msi, "Disable Message Signaled Interrupt (MSI)"); +module_param(enable_msi, int, 0); +MODULE_PARM_DESC(enable_msi, "Enable Message Signaled Interrupt (MSI)"); /* just for backward compatibility */ @@ -1531,7 +1531,7 @@ static int __devinit azx_create(struct snd_card *card, struct pci_dev *pci, chip->pci = pci; chip->irq = -1; chip->driver_type = driver_type; - chip->msi = !disable_msi; + chip->msi = enable_msi; chip->position_fix = position_fix; chip->single_cmd = single_cmd; -- cgit v0.10.2 From 45c99533252ef2297f37c5fdd672a3e0eb566870 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 14 Nov 2006 10:52:12 -0700 Subject: [PATCH] Use delayed disable mode of ioapic edge triggered interrupts Komuro reports that ISA interrupts do not work after a disable_irq(), causing some PCMCIA drivers to not work, with messages like eth0: Asix AX88190: io 0x300, irq 3, hw_addr xx:xx:xx:xx:xx:xx eth0: found link beat eth0: autonegotiation complete: 100baseT-FD selected eth0: interrupt(s) dropped! eth0: interrupt(s) dropped! eth0: interrupt(s) dropped! ... Linus Torvalds said: "Now, edge-triggered interrupts are a _lot_ harder to mask, because the Intel APIC is an unbelievable piece of sh*t, and has the edge-detect logic _before_ the mask logic, so if a edge happens _while_ the device is masked, you'll never ever see the edge ever again (unmasking will not cause a new edge, so you simply lost the interrupt). So when you "mask" an edge-triggered IRQ, you can't really mask it at all, because if you did that, you'd lose it forever if the IRQ comes in while you masked it. Instead, we're supposed to leave it active, and set a flag, and IF the IRQ comes in, we just remember it, and mask it at that point instead, and then on unmasking, we have to replay it by sending a self-IPI." This trivial patch solves the problem. Signed-off-by: Eric W. Biederman Cc: Ingo Molnar Acked-by: Komuro Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index ad84bc2..3b7a63e 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -1287,9 +1287,11 @@ static void ioapic_register_intr(int irq, int vector, unsigned long trigger) trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); - else + else { + irq_desc[irq].status |= IRQ_DELAYED_DISABLE; set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_edge_irq, "edge"); + } set_intr_gate(vector, interrupt[irq]); } diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 41bfc49..14654e6 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -790,9 +790,11 @@ static void ioapic_register_intr(int irq, int vector, unsigned long trigger) trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); - else + else { + irq_desc[irq].status |= IRQ_DELAYED_DISABLE; set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_edge_irq, "edge"); + } } static void __init setup_IO_APIC_irqs(void) -- cgit v0.10.2 From cbf093e8c7447a202e376199cc017161262bd7cd Mon Sep 17 00:00:00 2001 From: Robin Holt Date: Tue, 14 Nov 2006 20:50:59 -0600 Subject: [IA64] bte_unaligned_copy() transfers one extra cache line. When called to do a transfer that has a start offset within the cache line which is uneven between source and destination and a length which terminates the source of the copy exactly on a cache line, one extra line gets copied into a temporary buffer. This is normally not an issue since the buffer is a kernel buffer and only the requested information gets copied into the user buffer. The problem arises when the source ends at the very last physical page of memory. That last cache line does not exist and results in the SHUB chip raising an MCA. Signed-off-by: Robin Holt Signed-off-by: Dean Nelson Signed-off-by: Tony Luck diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c index 7f73ad4..ff1c556 100644 --- a/arch/ia64/sn/kernel/bte.c +++ b/arch/ia64/sn/kernel/bte.c @@ -381,14 +381,13 @@ bte_result_t bte_unaligned_copy(u64 src, u64 dest, u64 len, u64 mode) * bcopy to the destination. */ - /* Add the leader from source */ - headBteLen = len + (src & L1_CACHE_MASK); - /* Add the trailing bytes from footer. */ - headBteLen += L1_CACHE_BYTES - (headBteLen & L1_CACHE_MASK); - headBteSource = src & ~L1_CACHE_MASK; headBcopySrcOffset = src & L1_CACHE_MASK; headBcopyDest = dest; headBcopyLen = len; + + headBteSource = src - headBcopySrcOffset; + /* Add the leading and trailing bytes from source */ + headBteLen = L1_CACHE_ALIGN(len + headBcopySrcOffset); } if (headBcopyLen > 0) { -- cgit v0.10.2 From 056f4faa572f64fa926491a7d42c627c9dc507a7 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Mon, 13 Nov 2006 14:51:46 +1100 Subject: [POWERPC] Add the thread_siblings files to sysfs This adds the /sys/devices/system/cpu/*/topology/thread_siblings files on powerpc. These files are already available on other architectures. Signed-off-by: Stephen Rothwell Signed-off-by: Paul Mackerras diff --git a/include/asm-powerpc/topology.h b/include/asm-powerpc/topology.h index 8f7ee16..9fe7894 100644 --- a/include/asm-powerpc/topology.h +++ b/include/asm-powerpc/topology.h @@ -96,7 +96,13 @@ static inline void sysfs_remove_device_from_node(struct sys_device *dev, #ifdef CONFIG_SMP #include -#define smt_capable() (cpu_has_feature(CPU_FTR_SMT)) +#define smt_capable() (cpu_has_feature(CPU_FTR_SMT)) + +#ifdef CONFIG_PPC64 +#include + +#define topology_thread_siblings(cpu) (cpu_sibling_map[cpu]) +#endif #endif #endif /* __KERNEL__ */ -- cgit v0.10.2 From 9716a340310a383751a06589d0775fad04bd3f54 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Mon, 13 Nov 2006 14:52:42 +1100 Subject: [POWERPC] Wire up sys_move_pages All the infrastructure is already in place for this, so we only need to allocate a syscall number and hook it up. Signed-off-by: Stephen Rothwell Signed-off-by: Paul Mackerras diff --git a/include/asm-powerpc/systbl.h b/include/asm-powerpc/systbl.h index c6a0318..97b4354 100644 --- a/include/asm-powerpc/systbl.h +++ b/include/asm-powerpc/systbl.h @@ -304,3 +304,4 @@ SYSCALL_SPU(fchmodat) SYSCALL_SPU(faccessat) COMPAT_SYS_SPU(get_robust_list) COMPAT_SYS_SPU(set_robust_list) +COMPAT_SYS(move_pages) diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h index b5fe932..0e4ea37 100644 --- a/include/asm-powerpc/unistd.h +++ b/include/asm-powerpc/unistd.h @@ -323,10 +323,11 @@ #define __NR_faccessat 298 #define __NR_get_robust_list 299 #define __NR_set_robust_list 300 +#define __NR_move_pages 301 #ifdef __KERNEL__ -#define __NR_syscalls 301 +#define __NR_syscalls 302 #define __NR__exit __NR_exit #define NR_syscalls __NR_syscalls -- cgit v0.10.2 From d31e817183a4c1ee2e5fc0635ac075381f5c4419 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 16 Nov 2006 14:00:57 +1100 Subject: [PATCH] powerpc: windfarm shall request it's sub modules The windfarm code, in it's current incarnation, uses request_module() to load the various submodules it needs for a given platform so that only the main platform control module needs to be modprobed. However, it was missing various bits. This fixes it. In the future, we'll use some hotplug mecanisms to try to get all of this auto-loaded on the platforms where it matters but that isn't ready yet. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds diff --git a/drivers/macintosh/windfarm_pm112.c b/drivers/macintosh/windfarm_pm112.c index fa4b13f..b3fbb45 100644 --- a/drivers/macintosh/windfarm_pm112.c +++ b/drivers/macintosh/windfarm_pm112.c @@ -685,6 +685,17 @@ static int __init wf_pm112_init(void) ++nr_cores; printk(KERN_INFO "windfarm: initializing for dual-core desktop G5\n"); + +#ifdef MODULE + request_module("windfarm_smu_controls"); + request_module("windfarm_smu_sensors"); + request_module("windfarm_smu_sat"); + request_module("windfarm_lm75_sensor"); + request_module("windfarm_max6690_sensor"); + request_module("windfarm_cpufreq_clamp"); + +#endif /* MODULE */ + platform_driver_register(&wf_pm112_driver); return 0; } diff --git a/drivers/macintosh/windfarm_pm81.c b/drivers/macintosh/windfarm_pm81.c index 2a94485..f24fa73 100644 --- a/drivers/macintosh/windfarm_pm81.c +++ b/drivers/macintosh/windfarm_pm81.c @@ -788,6 +788,7 @@ static int __init wf_smu_init(void) request_module("windfarm_smu_controls"); request_module("windfarm_smu_sensors"); request_module("windfarm_lm75_sensor"); + request_module("windfarm_cpufreq_clamp"); #endif /* MODULE */ platform_driver_register(&wf_smu_driver); diff --git a/drivers/macintosh/windfarm_pm91.c b/drivers/macintosh/windfarm_pm91.c index 9961a67..26eee69 100644 --- a/drivers/macintosh/windfarm_pm91.c +++ b/drivers/macintosh/windfarm_pm91.c @@ -719,6 +719,7 @@ static int __init wf_smu_init(void) request_module("windfarm_smu_controls"); request_module("windfarm_smu_sensors"); request_module("windfarm_lm75_sensor"); + request_module("windfarm_cpufreq_clamp"); #endif /* MODULE */ platform_driver_register(&wf_smu_driver); -- cgit v0.10.2 From 44597f65f6af3c692560a639f61d25398d13d1b6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 15 Nov 2006 20:03:40 -0800 Subject: Linux 2.6.19-rc6 Getting there. Hopefully the MSI and other interrupt problems are all solved now. Signed-off-by: Linus Torvalds diff --git a/Makefile b/Makefile index 8484be1..958fad6 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 19 -EXTRAVERSION =-rc5 +EXTRAVERSION =-rc6 NAME=Avast! A bilge rat! # *DOCUMENTATION* -- cgit v0.10.2 From 0ccead1869444891ae6b41f2c5fc8498521c908e Mon Sep 17 00:00:00 2001 From: Gary Zambrano Date: Tue, 14 Nov 2006 16:34:00 -0800 Subject: [TG3]: Increase 5906 firmware poll time. Newer 5906 bootcode needs about 7ms to finish resetting so the poll firmware loop was changed to maximum 20ms. Signed-off-by: Gary Zambrano Signed-off-by: Michael Chan Acked-by: Jeff Garzik Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 06e4f77..6e86866 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -4728,10 +4728,11 @@ static int tg3_poll_fw(struct tg3 *tp) u32 val; if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) { - for (i = 0; i < 400; i++) { + /* Wait up to 20ms for init done. */ + for (i = 0; i < 200; i++) { if (tr32(VCPU_STATUS) & VCPU_STATUS_INIT_DONE) return 0; - udelay(10); + udelay(100); } return -ENODEV; } -- cgit v0.10.2 From 7fdeaf68a16f8102659cf1b30f57247d8ccbeef0 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 14 Nov 2006 19:47:09 -0800 Subject: [NETFILTER]: nfnetlink_log: fix byteorder of NFULA_SEQ_GLOBAL NFULA_SEQ_GLOBAL should be in network byteorder. Spotted by Al Viro. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index b2bf8f2..1e5207b 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -544,7 +544,7 @@ __build_packet_message(struct nfulnl_instance *inst, } /* global sequence number */ if (inst->flags & NFULNL_CFG_F_SEQ_GLOBAL) { - tmp_uint = atomic_inc_return(&global_seq); + tmp_uint = htonl(atomic_inc_return(&global_seq)); NFA_PUT(inst->skb, NFULA_SEQ_GLOBAL, sizeof(tmp_uint), &tmp_uint); } -- cgit v0.10.2 From d8a585d78efdf191a64ca655136ac1e49fd27cf4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 14 Nov 2006 19:48:09 -0800 Subject: [NETFILTER]: Use pskb_trim in {ip,ip6,nfnetlink}_queue Based on patch by James D. Nurmi: I've got some code very dependant on nfnetlink_queue, and turned up a large number of warns coming from skb_trim. While it's quite possibly my code, having not seen it on older kernels made me a bit suspect. Anyhow, based on some googling I turned up this thread: http://lkml.org/lkml/2006/8/13/56 And believe the issue to be related, so attached is a small patch to the kernel -- not sure if this is completely correct, but for anyone else hitting the WARN_ON(1) in skbuff.h, it might be helpful.. Signed-off-by: James D. Nurmi Ported to ip6_queue and nfnetlink_queue and added return value checks. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index 7edad79..97556cc 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -351,9 +351,10 @@ ipq_mangle_ipv4(ipq_verdict_msg_t *v, struct ipq_queue_entry *e) if (v->data_len < sizeof(*user_iph)) return 0; diff = v->data_len - e->skb->len; - if (diff < 0) - skb_trim(e->skb, v->data_len); - else if (diff > 0) { + if (diff < 0) { + if (pskb_trim(e->skb, v->data_len)) + return -ENOMEM; + } else if (diff > 0) { if (v->data_len > 0xFFFF) return -EINVAL; if (diff > skb_tailroom(e->skb)) { diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index 9510c24..9fec832 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -349,9 +349,10 @@ ipq_mangle_ipv6(ipq_verdict_msg_t *v, struct ipq_queue_entry *e) if (v->data_len < sizeof(*user_iph)) return 0; diff = v->data_len - e->skb->len; - if (diff < 0) - skb_trim(e->skb, v->data_len); - else if (diff > 0) { + if (diff < 0) { + if (pskb_trim(e->skb, v->data_len)) + return -ENOMEM; + } else if (diff > 0) { if (v->data_len > 0xFFFF) return -EINVAL; if (diff > skb_tailroom(e->skb)) { diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 6e4ada3..e815a9a 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -622,9 +622,10 @@ nfqnl_mangle(void *data, int data_len, struct nfqnl_queue_entry *e) int diff; diff = data_len - e->skb->len; - if (diff < 0) - skb_trim(e->skb, data_len); - else if (diff > 0) { + if (diff < 0) { + if (pskb_trim(e->skb, data_len)) + return -ENOMEM; + } else if (diff > 0) { if (data_len > 0xFFFF) return -EINVAL; if (diff > skb_tailroom(e->skb)) { -- cgit v0.10.2 From b96e7ecbd052a0916b6078e7600604d7e276a336 Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Tue, 14 Nov 2006 19:48:48 -0800 Subject: [NETFILTER]: ip6_tables: fixed conflicted optname for getsockopt 66 and 67 for getsockopt on IPv6 socket is doubly used for IPv6 Advanced API and ip6tables. This moves numbers for ip6tables to 68 and 69. This also kills XT_SO_* because {ip,ip6,arp}_tables doesn't have so much common numbers now. The old userland tools keep to behave as ever, because old kernel always calls functions of IPv6 Advanced API for their numbers. Signed-off-by: Yasuyuki Kozakai Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/include/linux/in6.h b/include/linux/in6.h index 9be6a47..f28621f 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -225,7 +225,7 @@ struct in6_flowlabel_req #endif /* - * Netfilter + * Netfilter (1) * * Following socket options are used in ip6_tables; * see include/linux/netfilter_ipv6/ip6_tables.h. @@ -240,4 +240,14 @@ struct in6_flowlabel_req #define IPV6_RECVTCLASS 66 #define IPV6_TCLASS 67 +/* + * Netfilter (2) + * + * Following socket options are used in ip6_tables; + * see include/linux/netfilter_ipv6/ip6_tables.h. + * + * IP6T_SO_GET_REVISION_MATCH 68 + * IP6T_SO_GET_REVISION_TARGET 69 + */ + #endif diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 04319a7..022edfa 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -96,22 +96,6 @@ struct _xt_align /* Error verdict. */ #define XT_ERROR_TARGET "ERROR" -/* - * New IP firewall options for [gs]etsockopt at the RAW IP level. - * Unlike BSD Linux inherits IP options so you don't have to use a raw - * socket for this. Instead we check rights in the calls. */ -#define XT_BASE_CTL 64 /* base for firewall socket options */ - -#define XT_SO_SET_REPLACE (XT_BASE_CTL) -#define XT_SO_SET_ADD_COUNTERS (XT_BASE_CTL + 1) -#define XT_SO_SET_MAX XT_SO_SET_ADD_COUNTERS - -#define XT_SO_GET_INFO (XT_BASE_CTL) -#define XT_SO_GET_ENTRIES (XT_BASE_CTL + 1) -#define XT_SO_GET_REVISION_MATCH (XT_BASE_CTL + 2) -#define XT_SO_GET_REVISION_TARGET (XT_BASE_CTL + 3) -#define XT_SO_GET_MAX XT_SO_GET_REVISION_TARGET - #define SET_COUNTER(c,b,p) do { (c).bcnt = (b); (c).pcnt = (p); } while(0) #define ADD_COUNTER(c,b,p) do { (c).bcnt += (b); (c).pcnt += (p); } while(0) diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index 44e39b6..0be2354 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -112,19 +112,20 @@ struct arpt_entry * New IP firewall options for [gs]etsockopt at the RAW IP level. * Unlike BSD Linux inherits IP options so you don't have to use a raw * socket for this. Instead we check rights in the calls. + * + * ATTENTION: check linux/in.h before adding new number here. */ -#define ARPT_CTL_OFFSET 32 -#define ARPT_BASE_CTL (XT_BASE_CTL+ARPT_CTL_OFFSET) - -#define ARPT_SO_SET_REPLACE (XT_SO_SET_REPLACE+ARPT_CTL_OFFSET) -#define ARPT_SO_SET_ADD_COUNTERS (XT_SO_SET_ADD_COUNTERS+ARPT_CTL_OFFSET) -#define ARPT_SO_SET_MAX (XT_SO_SET_MAX+ARPT_CTL_OFFSET) - -#define ARPT_SO_GET_INFO (XT_SO_GET_INFO+ARPT_CTL_OFFSET) -#define ARPT_SO_GET_ENTRIES (XT_SO_GET_ENTRIES+ARPT_CTL_OFFSET) -/* #define ARPT_SO_GET_REVISION_MATCH XT_SO_GET_REVISION_MATCH */ -#define ARPT_SO_GET_REVISION_TARGET (XT_SO_GET_REVISION_TARGET+ARPT_CTL_OFFSET) -#define ARPT_SO_GET_MAX (XT_SO_GET_REVISION_TARGET+ARPT_CTL_OFFSET) +#define ARPT_BASE_CTL 96 + +#define ARPT_SO_SET_REPLACE (ARPT_BASE_CTL) +#define ARPT_SO_SET_ADD_COUNTERS (ARPT_BASE_CTL + 1) +#define ARPT_SO_SET_MAX ARPT_SO_SET_ADD_COUNTERS + +#define ARPT_SO_GET_INFO (ARPT_BASE_CTL) +#define ARPT_SO_GET_ENTRIES (ARPT_BASE_CTL + 1) +/* #define ARPT_SO_GET_REVISION_MATCH (APRT_BASE_CTL + 2) */ +#define ARPT_SO_GET_REVISION_TARGET (ARPT_BASE_CTL + 3) +#define ARPT_SO_GET_MAX (ARPT_SO_GET_REVISION_TARGET) /* CONTINUE verdict for targets */ #define ARPT_CONTINUE XT_CONTINUE diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index a536bbd..4f06dad 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -101,18 +101,21 @@ struct ipt_entry /* * New IP firewall options for [gs]etsockopt at the RAW IP level. * Unlike BSD Linux inherits IP options so you don't have to use a raw - * socket for this. Instead we check rights in the calls. */ -#define IPT_BASE_CTL XT_BASE_CTL - -#define IPT_SO_SET_REPLACE XT_SO_SET_REPLACE -#define IPT_SO_SET_ADD_COUNTERS XT_SO_SET_ADD_COUNTERS -#define IPT_SO_SET_MAX XT_SO_SET_MAX - -#define IPT_SO_GET_INFO XT_SO_GET_INFO -#define IPT_SO_GET_ENTRIES XT_SO_GET_ENTRIES -#define IPT_SO_GET_REVISION_MATCH XT_SO_GET_REVISION_MATCH -#define IPT_SO_GET_REVISION_TARGET XT_SO_GET_REVISION_TARGET -#define IPT_SO_GET_MAX XT_SO_GET_REVISION_TARGET + * socket for this. Instead we check rights in the calls. + * + * ATTENTION: check linux/in.h before adding new number here. + */ +#define IPT_BASE_CTL 64 + +#define IPT_SO_SET_REPLACE (IPT_BASE_CTL) +#define IPT_SO_SET_ADD_COUNTERS (IPT_BASE_CTL + 1) +#define IPT_SO_SET_MAX IPT_SO_SET_ADD_COUNTERS + +#define IPT_SO_GET_INFO (IPT_BASE_CTL) +#define IPT_SO_GET_ENTRIES (IPT_BASE_CTL + 1) +#define IPT_SO_GET_REVISION_MATCH (IPT_BASE_CTL + 2) +#define IPT_SO_GET_REVISION_TARGET (IPT_BASE_CTL + 3) +#define IPT_SO_GET_MAX IPT_SO_GET_REVISION_TARGET #define IPT_CONTINUE XT_CONTINUE #define IPT_RETURN XT_RETURN diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index d7a8e9c0..4aed340 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -107,18 +107,21 @@ struct ip6t_entry /* * New IP firewall options for [gs]etsockopt at the RAW IP level. * Unlike BSD Linux inherits IP options so you don't have to use - * a raw socket for this. Instead we check rights in the calls. */ -#define IP6T_BASE_CTL XT_BASE_CTL - -#define IP6T_SO_SET_REPLACE XT_SO_SET_REPLACE -#define IP6T_SO_SET_ADD_COUNTERS XT_SO_SET_ADD_COUNTERS -#define IP6T_SO_SET_MAX XT_SO_SET_MAX - -#define IP6T_SO_GET_INFO XT_SO_GET_INFO -#define IP6T_SO_GET_ENTRIES XT_SO_GET_ENTRIES -#define IP6T_SO_GET_REVISION_MATCH XT_SO_GET_REVISION_MATCH -#define IP6T_SO_GET_REVISION_TARGET XT_SO_GET_REVISION_TARGET -#define IP6T_SO_GET_MAX XT_SO_GET_REVISION_TARGET + * a raw socket for this. Instead we check rights in the calls. + * + * ATTENTION: check linux/in6.h before adding new number here. + */ +#define IP6T_BASE_CTL 64 + +#define IP6T_SO_SET_REPLACE (IP6T_BASE_CTL) +#define IP6T_SO_SET_ADD_COUNTERS (IP6T_BASE_CTL + 1) +#define IP6T_SO_SET_MAX IP6T_SO_SET_ADD_COUNTERS + +#define IP6T_SO_GET_INFO (IP6T_BASE_CTL) +#define IP6T_SO_GET_ENTRIES (IP6T_BASE_CTL + 1) +#define IP6T_SO_GET_REVISION_MATCH (IP6T_BASE_CTL + 4) +#define IP6T_SO_GET_REVISION_TARGET (IP6T_BASE_CTL + 5) +#define IP6T_SO_GET_MAX IP6T_SO_GET_REVISION_TARGET /* CONTINUE verdict for targets */ #define IP6T_CONTINUE XT_CONTINUE -- cgit v0.10.2 From 337dde798dd1f3e4d4df4e684a0b4355dd65103d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 14 Nov 2006 19:49:13 -0800 Subject: [NETFILTER]: ip6_tables: use correct nexthdr value in ipv6_find_hdr() nexthdr is NEXTHDR_FRAGMENT, the nexthdr value from the fragment header is hp->nexthdr. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 167c2ea..204e021 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1494,7 +1494,7 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, if (_frag_off) { if (target < 0 && ((!ipv6_ext_hdr(hp->nexthdr)) || - nexthdr == NEXTHDR_NONE)) { + hp->nexthdr == NEXTHDR_NONE)) { if (fragoff) *fragoff = _frag_off; return hp->nexthdr; -- cgit v0.10.2 From 52bf376c63eebe72e862a1a6e713976b038c3f50 Mon Sep 17 00:00:00 2001 From: John Heffner Date: Tue, 14 Nov 2006 20:25:17 -0800 Subject: [TCP]: Fix up sysctl_tcp_mem initialization. Fix up tcp_mem initial settings to take into account the size of the hash entries (different on SMP and non-SMP systems). Signed-off-by: John Heffner Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4322318..c05e8ed 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2316,9 +2316,10 @@ void __init tcp_init(void) sysctl_max_syn_backlog = 128; } - sysctl_tcp_mem[0] = 768 << order; - sysctl_tcp_mem[1] = 1024 << order; - sysctl_tcp_mem[2] = 1536 << order; + /* Allow no more than 3/4 kernel memory (usually less) allocated to TCP */ + sysctl_tcp_mem[0] = (1536 / sizeof (struct inet_bind_hashbucket)) << order; + sysctl_tcp_mem[1] = sysctl_tcp_mem[0] * 4 / 3; + sysctl_tcp_mem[2] = sysctl_tcp_mem[0] * 2; limit = ((unsigned long)sysctl_tcp_mem[1]) << (PAGE_SHIFT - 7); max_share = min(4UL*1024*1024, limit); -- cgit v0.10.2 From c7835a77c86422d276b0d1a4c70924d933014c13 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 15 Nov 2006 21:14:42 -0800 Subject: [TG3]: Disable TSO on 5906 if CLKREQ is enabled. Due to hardware errata, TSO must be disabled if the PCI Express clock request is enabled on 5906. The chip may hang when transmitting TSO frames if CLKREQ is enabled. Update version to 3.69. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 6e86866..1dbdd6b 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -68,8 +68,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.68" -#define DRV_MODULE_RELDATE "November 02, 2006" +#define DRV_MODULE_VERSION "3.69" +#define DRV_MODULE_RELDATE "November 15, 2006" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -10366,7 +10366,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) u32 pci_state_reg, grc_misc_cfg; u32 val; u16 pci_cmd; - int err; + int err, pcie_cap; /* Force memory write invalidate off. If we leave it on, * then on 5700_BX chips we have to enable a workaround. @@ -10541,8 +10541,19 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906) tp->tg3_flags2 |= TG3_FLG2_JUMBO_CAPABLE; - if (pci_find_capability(tp->pdev, PCI_CAP_ID_EXP) != 0) + pcie_cap = pci_find_capability(tp->pdev, PCI_CAP_ID_EXP); + if (pcie_cap != 0) { tp->tg3_flags2 |= TG3_FLG2_PCI_EXPRESS; + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) { + u16 lnkctl; + + pci_read_config_word(tp->pdev, + pcie_cap + PCI_EXP_LNKCTL, + &lnkctl); + if (lnkctl & PCI_EXP_LNKCTL_CLKREQ_EN) + tp->tg3_flags2 &= ~TG3_FLG2_HW_TSO_2; + } + } /* If we have an AMD 762 or VIA K8T800 chipset, write * reordering to the mailbox registers done by the host @@ -11809,6 +11820,7 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5700 || GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5701 || tp->pci_chip_rev_id == CHIPREV_ID_5705_A0 || + GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906 || (tp->tg3_flags & TG3_FLAG_ENABLE_ASF) != 0) { tp->tg3_flags2 &= ~TG3_FLG2_TSO_CAPABLE; } else { diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index c312a12..c321316 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -371,6 +371,7 @@ #define PCI_EXP_DEVSTA_TRPND 0x20 /* Transactions Pending */ #define PCI_EXP_LNKCAP 12 /* Link Capabilities */ #define PCI_EXP_LNKCTL 16 /* Link Control */ +#define PCI_EXP_LNKCTL_CLKREQ_EN 0x100 /* Enable clkreq */ #define PCI_EXP_LNKSTA 18 /* Link Status */ #define PCI_EXP_SLTCAP 20 /* Slot Capabilities */ #define PCI_EXP_SLTCTL 24 /* Slot Control */ -- cgit v0.10.2 From 351a58390aad350bd5e22c7cc468f44a9330685c Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 16 Nov 2006 00:42:58 -0800 Subject: [IA64] irqs: use `name' not `typename' `typename' is going away and is usually uninitialised anwyay. Signed-off-by: Andrew Morton Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 9bf15fe..c195faf 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -664,7 +664,7 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, printk(KERN_WARNING "%s: changing vector %d from %s to %s\n", __FUNCTION__, vector, - idesc->chip->typename, irq_type->typename); + idesc->chip->name, irq_type->name); idesc->chip = irq_type; } return 0; diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index f07c086..6fe404d 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -76,7 +76,7 @@ int show_interrupts(struct seq_file *p, void *v) seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); } #endif - seq_printf(p, " %14s", irq_desc[i].chip->typename); + seq_printf(p, " %14s", irq_desc[i].chip->name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) -- cgit v0.10.2 From 06344db3162323247e9ac8d76d5acdb79d05f3c0 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 16 Nov 2006 00:43:02 -0800 Subject: [IA64] typename -> name conversion convert irq chip typename -> name. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Tony Luck diff --git a/arch/ia64/hp/sim/hpsim_irq.c b/arch/ia64/hp/sim/hpsim_irq.c index 8145547..c2f58ff 100644 --- a/arch/ia64/hp/sim/hpsim_irq.c +++ b/arch/ia64/hp/sim/hpsim_irq.c @@ -27,7 +27,7 @@ hpsim_set_affinity_noop (unsigned int a, cpumask_t b) } static struct hw_interrupt_type irq_type_hp_sim = { - .typename = "hpsim", + .name = "hpsim", .startup = hpsim_irq_startup, .shutdown = hpsim_irq_noop, .enable = hpsim_irq_noop, diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index c195faf..60d6495 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -426,7 +426,7 @@ iosapic_end_level_irq (unsigned int irq) #define iosapic_ack_level_irq nop struct hw_interrupt_type irq_type_iosapic_level = { - .typename = "IO-SAPIC-level", + .name = "IO-SAPIC-level", .startup = iosapic_startup_level_irq, .shutdown = iosapic_shutdown_level_irq, .enable = iosapic_enable_level_irq, @@ -473,7 +473,7 @@ iosapic_ack_edge_irq (unsigned int irq) #define iosapic_end_edge_irq nop struct hw_interrupt_type irq_type_iosapic_edge = { - .typename = "IO-SAPIC-edge", + .name = "IO-SAPIC-edge", .startup = iosapic_startup_edge_irq, .shutdown = iosapic_disable_edge_irq, .enable = iosapic_enable_edge_irq, diff --git a/arch/ia64/kernel/irq_lsapic.c b/arch/ia64/kernel/irq_lsapic.c index 1ab58b0..c2f07be 100644 --- a/arch/ia64/kernel/irq_lsapic.c +++ b/arch/ia64/kernel/irq_lsapic.c @@ -34,7 +34,7 @@ static int lsapic_retrigger(unsigned int irq) } struct hw_interrupt_type irq_type_ia64_lsapic = { - .typename = "LSAPIC", + .name = "LSAPIC", .startup = lsapic_noop_startup, .shutdown = lsapic_noop, .enable = lsapic_noop, diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c index 7bb6ad1..0b49459 100644 --- a/arch/ia64/sn/kernel/irq.c +++ b/arch/ia64/sn/kernel/irq.c @@ -201,7 +201,7 @@ static void sn_set_affinity_irq(unsigned int irq, cpumask_t mask) } struct hw_interrupt_type irq_type_sn = { - .typename = "SN hub", + .name = "SN hub", .startup = sn_startup_irq, .shutdown = sn_shutdown_irq, .enable = sn_enable_irq, -- cgit v0.10.2 From 5fbb004aba15bbca64c3fb611113295bda7ee9ea Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 16 Nov 2006 00:43:07 -0800 Subject: [IA64] use generic_handle_irq() Use generic_handle_irq() to handle mixed-type irq handling. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Tony Luck diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 6fe404d..54d55e4 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -197,7 +197,7 @@ void fixup_irqs(void) struct pt_regs *old_regs = set_irq_regs(NULL); vectors_in_migration[irq]=0; - __do_IRQ(irq); + generic_handle_irq(irq); set_irq_regs(old_regs); } } diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index 9c6dafa..ba3ba8b 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -186,7 +186,7 @@ ia64_handle_irq (ia64_vector vector, struct pt_regs *regs) ia64_setreg(_IA64_REG_CR_TPR, vector); ia64_srlz_d(); - __do_IRQ(local_vector_to_irq(vector)); + generic_handle_irq(local_vector_to_irq(vector)); /* * Disable interrupts and send EOI: @@ -242,7 +242,7 @@ void ia64_process_pending_intr(void) * Probably could shared code. */ vectors_in_migration[local_vector_to_irq(vector)]=0; - __do_IRQ(local_vector_to_irq(vector)); + generic_handle_irq(local_vector_to_irq(vector)); set_irq_regs(old_regs); /* -- cgit v0.10.2 From cea196bb2eb918c30f97a2b752bcec929d761f64 Mon Sep 17 00:00:00 2001 From: "Luck, Tony" Date: Thu, 16 Nov 2006 00:43:12 -0800 Subject: [IA64] a fix towards allmodconfig build The HP_SIMSCSI driver can't be built as a module (unhealthy dependencies on things that shouldn't really be exported). Signed-off-by: Andrew Morton Signed-off-by: Tony Luck diff --git a/arch/ia64/hp/sim/Kconfig b/arch/ia64/hp/sim/Kconfig index 18ccb12..f92306b 100644 --- a/arch/ia64/hp/sim/Kconfig +++ b/arch/ia64/hp/sim/Kconfig @@ -13,8 +13,8 @@ config HP_SIMSERIAL_CONSOLE depends on HP_SIMSERIAL config HP_SIMSCSI - tristate "Simulated SCSI disk" - depends on SCSI + bool "Simulated SCSI disk" + depends on SCSI=y endmenu -- cgit v0.10.2 From b48f5457b4e9d64d9c1117a4ece247d98b4db49f Mon Sep 17 00:00:00 2001 From: "Zhang, Yanmin" Date: Thu, 16 Nov 2006 01:19:08 -0800 Subject: [PATCH] ipmi: use platform_device_add() instead of platform_device_register() to register device allocated dynamically I got below warning when running 2.6.19-rc5-mm1 on my ia64 machine. WARNING at lib/kobject.c:172 kobject_init() Call Trace: [] show_stack+0x40/0xa0 sp=e0000002ff9f7bc0 bsp=e0000002ff9f0d10 [] dump_stack+0x30/0x60 sp=e0000002ff9f7d90 bsp=e0000002ff9f0cf8 [] kobject_init+0x90/0x160 sp=e0000002ff9f7d90 bsp=e0000002ff9f0cd0 [] device_initialize+0x40/0x1c0 sp=e0000002ff9f7da0 bsp=e0000002ff9f0cb0 [] platform_device_register+0x20/0x60 sp=e0000002ff9f7dd0 bsp=e0000002ff9f0c90 [] try_smi_init+0xbc0/0x11e0 sp=e0000002ff9f7dd0 bsp=e0000002ff9f0c50 [] init_ipmi_si+0xaa0/0x12e0 sp=e0000002ff9f7de0 bsp=e0000002ff9f0bd8 [] init+0x350/0x780 sp=e0000002ff9f7e00 bsp=e0000002ff9f0ba8 [] kernel_thread_helper+0x30/0x60 sp=e0000002ff9f7e30 bsp=e0000002ff9f0b80 [] start_kernel_thread+0x20/0x40 sp=e0000002ff9f7e30 bsp=e0000002ff9f0b80 WARNING at lib/kobject.c:172 kobject_init() Call Trace: [] show_stack+0x40/0xa0 sp=e0000002ff9f7b40 bsp=e0000002ff9f0db0 [] dump_stack+0x30/0x60 sp=e0000002ff9f7d10 bsp=e0000002ff9f0d98 [] kobject_init+0x90/0x160 sp=e0000002ff9f7d10 bsp=e0000002ff9f0d70 [] device_initialize+0x40/0x1c0 sp=e0000002ff9f7d20 bsp=e0000002ff9f0d50 [] platform_device_register+0x20/0x60 sp=e0000002ff9f7d50 bsp=e0000002ff9f0d30 [] ipmi_register_smi+0xcc0/0x18e0 sp=e0000002ff9f7d50 bsp=e0000002ff9f0c90 [] try_smi_init+0xc60/0x11e0 sp=e0000002ff9f7dd0 bsp=e0000002ff9f0c50 [] init_ipmi_si+0xaa0/0x12e0 sp=e0000002ff9f7de0 bsp=e0000002ff9f0bd8 [] init+0x350/0x780 sp=e0000002ff9f7e00 bsp=e0000002ff9f0ba8 [] kernel_thread_helper+0x30/0x60 sp=e0000002ff9f7e30 bsp=e0000002ff9f0b80 [] start_kernel_thread+0x20/0x40 sp=e0000002ff9f7e30 bsp=e0000002ff9f0b80 The root cause is the device struct is initialized twice. If the device is allocated dynamically by platform_device_alloc, platform_device_alloc will initialize struct device, then, platform_device_add should be used to register the device. The difference between platform_device_register and platform_device_add is platform_device_register will initiate the device while platform_device_add won't. Signed-off-by: Zhang Yanmin Cc: Corey Minyard Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index a41b8df..c47add8 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -2118,7 +2118,7 @@ static int ipmi_bmc_register(ipmi_smi_t intf) dev_set_drvdata(&bmc->dev->dev, bmc); kref_init(&bmc->refcount); - rv = platform_device_register(bmc->dev); + rv = platform_device_add(bmc->dev); mutex_unlock(&ipmidriver_mutex); if (rv) { printk(KERN_ERR diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index abc5149..bb1fac1 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2346,7 +2346,7 @@ static int try_smi_init(struct smi_info *new_smi) new_smi->dev = &new_smi->pdev->dev; new_smi->dev->driver = &ipmi_driver; - rv = platform_device_register(new_smi->pdev); + rv = platform_device_add(new_smi->pdev); if (rv) { printk(KERN_ERR "ipmi_si_intf:" -- cgit v0.10.2 From b86432b42eba5671969a9e6483ee219674b7ee25 Mon Sep 17 00:00:00 2001 From: "Zhang, Yanmin" Date: Thu, 16 Nov 2006 01:19:10 -0800 Subject: [PATCH] some irq_chip variables point to NULL I got an oops when booting 2.6.19-rc5-mm1 on my ia64 machine. Below is the log. Oops 11012296146944 [1] Modules linked in: binfmt_misc dm_mirror dm_multipath dm_mod thermal processor f an container button sg eepro100 e100 mii Pid: 0, CPU 0, comm: swapper psr : 0000121008022038 ifs : 800000000000040b ip : [] Not tainted ip is at __do_IRQ+0x371/0x3e0 unat: 0000000000000000 pfs : 000000000000040b rsc : 0000000000000003 rnat: 656960155aa56aa5 bsps: a00000010058b890 pr : 656960155aa55a65 ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c0270033f csd : 0000000000000000 ssd : 0000000000000000 b0 : a0000001000e1390 b6 : a0000001005beac0 b7 : e00000007f01aa00 f6 : 000000000000000000000 f7 : 0ffe69090000000000000 f8 : 1000a9090000000000000 f9 : 0ffff8000000000000000 f10 : 1000a908ffffff6f70000 f11 : 1003e0000000000000909 r1 : a000000100fbbff0 r2 : 0000000000010002 r3 : 0000000000010001 r8 : fffffffffffbffff r9 : a000000100bd8060 r10 : a000000100dd83b8 r11 : fffffffffffeffff r12 : a000000100bcbbb0 r13 : a000000100bc4000 r14 : 0000000000010000 r15 : 0000000000010000 r16 : a000000100c01aa8 r17 : a000000100d2c350 r18 : 0000000000000000 r19 : a000000100d2c300 r20 : a000000100c01a88 r21 : 0000000080010100 r22 : a000000100c01ac0 r23 : a0000001000108e0 r24 : e000000477980004 r25 : 0000000000000000 r26 : 0000000000000000 r27 : e00000000913400c r28 : e0000004799ee51c r29 : e0000004778b87f0 r30 : a000000100d2c300 r31 : a00000010005c7e0 Call Trace: [] show_stack+0x40/0xa0 sp=a000000100bcb760 bsp=a000000100bc4f40 [] show_regs+0x840/0x880 sp=a000000100bcb930 bsp=a000000100bc4ee8 [] die+0x250/0x320 sp=a000000100bcb930 bsp=a000000100bc4ea0 [] ia64_do_page_fault+0x8d0/0xa20 sp=a000000100bcb950 bsp=a000000100bc4e50 [] ia64_leave_kernel+0x0/0x290 sp=a000000100bcb9e0 bsp=a000000100bc4e50 [] __do_IRQ+0x370/0x3e0 sp=a000000100bcbbb0 bsp=a000000100bc4df0 [] ia64_handle_irq+0x170/0x220 sp=a000000100bcbbb0 bsp=a000000100bc4dc0 [] ia64_leave_kernel+0x0/0x290 sp=a000000100bcbbb0 bsp=a000000100bc4dc0 [] ia64_pal_call_static+0x90/0xc0 sp=a000000100bcbd80 bsp=a000000100bc4d78 [] default_idle+0x90/0x160 sp=a000000100bcbd80 bsp=a000000100bc4d58 [] cpu_idle+0x1f0/0x440 sp=a000000100bcbe20 bsp=a000000100bc4d18 [] rest_init+0xc0/0xe0 sp=a000000100bcbe20 bsp=a000000100bc4d00 [] start_kernel+0x6a0/0x6c0 sp=a000000100bcbe20 bsp=a000000100bc4ca0 [] __end_ivt_text+0x6d0/0x6f0 sp=a000000100bcbe30 bsp=a000000100bc4c00 <0>Kernel panic - not syncing: Aiee, killing interrupt handler! The root cause is that some irq_chip variables, especially ia64_msi_chip, initiate their memeber end to point to NULL. __do_IRQ doesn't check if irq_chip->end is null and just calls it after processing the interrupt. As irq_chip->end is called at many places, so I fix it by reinitiating irq_chip->end to dummy_irq_chip.end, e.g., a noop function. Signed-off-by: Zhang Yanmin Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 2d0dc3e..ebfd24a 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -233,6 +233,8 @@ void irq_chip_set_defaults(struct irq_chip *chip) chip->shutdown = chip->disable; if (!chip->name) chip->name = chip->typename; + if (!chip->end) + chip->end = dummy_irq_chip.end; } static inline void mask_ack_irq(struct irq_desc *desc, int irq) -- cgit v0.10.2 From 84a763e3d1a47fa9308b8817f265e936e5f1000a Mon Sep 17 00:00:00 2001 From: Vitaly Wool Date: Thu, 16 Nov 2006 01:19:11 -0800 Subject: [PATCH] pnx4008: rename driver Make the drivers' names less generic to avoid possible confusion in future, as was requested by Russell King. Signed-off-by: Vitaly Wool Acked-by: James Simmons Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/pnx4008/pnxrgbfb.c b/drivers/video/pnx4008/pnxrgbfb.c index 7d9453c..bf36b68 100644 --- a/drivers/video/pnx4008/pnxrgbfb.c +++ b/drivers/video/pnx4008/pnxrgbfb.c @@ -191,7 +191,7 @@ err: static struct platform_driver rgbfb_driver = { .driver = { - .name = "rgbfb", + .name = "pnx4008-rgbfb", }, .probe = rgbfb_probe, .remove = rgbfb_remove, diff --git a/drivers/video/pnx4008/sdum.c b/drivers/video/pnx4008/sdum.c index 51f0ecc..d23bf0d 100644 --- a/drivers/video/pnx4008/sdum.c +++ b/drivers/video/pnx4008/sdum.c @@ -848,7 +848,7 @@ static int sdum_remove(struct platform_device *pdev) static struct platform_driver sdum_driver = { .driver = { - .name = "sdum", + .name = "pnx4008-sdum", }, .probe = sdum_probe, .remove = sdum_remove, -- cgit v0.10.2 From 3b9c10dc59eaaef23e5a47110c20fb554f7dba28 Mon Sep 17 00:00:00 2001 From: Vitaly Wool Date: Thu, 16 Nov 2006 01:19:14 -0800 Subject: [PATCH] pnx4008:fix NULL dereference in rgbfb Fix possible NULL dereference in pnxrgbfb. Signed-off-by: Vitaly Wool Cc: James Simmons Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/pnx4008/pnxrgbfb.c b/drivers/video/pnx4008/pnxrgbfb.c index bf36b68..f29e66e 100644 --- a/drivers/video/pnx4008/pnxrgbfb.c +++ b/drivers/video/pnx4008/pnxrgbfb.c @@ -154,7 +154,8 @@ static int __devinit rgbfb_probe(struct platform_device *pdev) goto err1; } - if (!fb_get_options("pnxrgbfb", &option) && !strcmp(option, "nocursor")) + if (!fb_get_options("pnxrgbfb", &option) && option && + !strcmp(option, "nocursor")) rgbfb_ops.fb_cursor = no_cursor; info->node = -1; -- cgit v0.10.2 From a9083081b56800691e79d541638fd2fc6366db0f Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Thu, 16 Nov 2006 01:19:16 -0800 Subject: [PATCH] eCryptfs: dput() lower d_parent on rename On rename, for both the old and new lower dentry objects, eCryptfs is missing a dput on the lower parent directory dentry. This patch will prevent the BUG() at fs/dcache.c:613 from being hit after renaming a file inside eCryptfs and then doing a umount on the lower filesystem. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index ff4865d..ebec8cf 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -630,6 +630,8 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry, ecryptfs_copy_attr_all(old_dir, lower_old_dir_dentry->d_inode); out_lock: unlock_rename(lower_old_dir_dentry, lower_new_dir_dentry); + dput(lower_new_dentry->d_parent); + dput(lower_old_dentry->d_parent); dput(lower_new_dentry); dput(lower_old_dentry); return rc; -- cgit v0.10.2 From 3b46f0396c76a61526dec57a782a061c197ac337 Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Thu, 16 Nov 2006 01:19:17 -0800 Subject: [PATCH] set default video mode on PowerBook Wallstreet Finally add the third PowerBook Wallstreet 233MHz model to the list of known display resolutions. Without this change, a 640x480 video mode is used. A workaround so far was to boot with 'video=atyfb:vmode:14' Signed-off-by: Olaf Hering Cc: Benjamin Herrenschmidt Cc: "Antonino A. Daplas" Cc: Solomon Peachy Cc: James Simmons Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c index b77b309..e815b35 100644 --- a/drivers/video/aty/atyfb_base.c +++ b/drivers/video/aty/atyfb_base.c @@ -406,7 +406,7 @@ static struct { { PCI_CHIP_MACH64LB, "3D RAGE LT PRO (Mach64 LB, AGP)", 236, 75, 100, 135, ATI_CHIP_264LTPRO }, { PCI_CHIP_MACH64LD, "3D RAGE LT PRO (Mach64 LD, AGP)", 230, 100, 100, 135, ATI_CHIP_264LTPRO }, { PCI_CHIP_MACH64LI, "3D RAGE LT PRO (Mach64 LI, PCI)", 230, 100, 100, 135, ATI_CHIP_264LTPRO | M64F_G3_PB_1_1 | M64F_G3_PB_1024x768 }, - { PCI_CHIP_MACH64LP, "3D RAGE LT PRO (Mach64 LP, PCI)", 230, 100, 100, 135, ATI_CHIP_264LTPRO }, + { PCI_CHIP_MACH64LP, "3D RAGE LT PRO (Mach64 LP, PCI)", 230, 100, 100, 135, ATI_CHIP_264LTPRO | M64F_G3_PB_1024x768 }, { PCI_CHIP_MACH64LQ, "3D RAGE LT PRO (Mach64 LQ, PCI)", 230, 100, 100, 135, ATI_CHIP_264LTPRO }, { PCI_CHIP_MACH64GM, "3D RAGE XL (Mach64 GM, AGP 2x)", 230, 83, 63, 135, ATI_CHIP_264XL }, -- cgit v0.10.2 From e757bef270e21453bf507df200e2fb477c076da6 Mon Sep 17 00:00:00 2001 From: Bryan O'Sullivan Date: Thu, 16 Nov 2006 01:19:19 -0800 Subject: [PATCH] IB/ipath - fix driver build for platforms with PCI, but not HT The PCI Express and Hypertransport chip-specific source files should only be built when the kernel has the capability of actually compiling them. This fixes the driver build on, for example, ia64. Signed-off-by: Bryan O'Sullivan Cc: "Eric W. Biederman" Cc: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/infiniband/hw/ipath/Kconfig b/drivers/infiniband/hw/ipath/Kconfig index 574a678..5ca471a 100644 --- a/drivers/infiniband/hw/ipath/Kconfig +++ b/drivers/infiniband/hw/ipath/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_IPATH tristate "QLogic InfiniPath Driver" - depends on PCI_MSI && 64BIT && INFINIBAND + depends on (PCI_MSI || HT_IRQ) && 64BIT && INFINIBAND ---help--- This is a driver for QLogic InfiniPath host channel adapters, including InfiniBand verbs support. This driver allows these diff --git a/drivers/infiniband/hw/ipath/Makefile b/drivers/infiniband/hw/ipath/Makefile index 5e29cb0..7dc1055 100644 --- a/drivers/infiniband/hw/ipath/Makefile +++ b/drivers/infiniband/hw/ipath/Makefile @@ -10,8 +10,6 @@ ib_ipath-y := \ ipath_eeprom.o \ ipath_file_ops.o \ ipath_fs.o \ - ipath_iba6110.o \ - ipath_iba6120.o \ ipath_init_chip.o \ ipath_intr.o \ ipath_keys.o \ @@ -31,5 +29,8 @@ ib_ipath-y := \ ipath_verbs_mcast.o \ ipath_verbs.o +ib_ipath-$(CONFIG_HT_IRQ) += ipath_iba6110.o +ib_ipath-$(CONFIG_PCI_MSI) += ipath_iba6120.o + ib_ipath-$(CONFIG_X86_64) += ipath_wc_x86_64.o ib_ipath-$(CONFIG_PPC64) += ipath_wc_ppc64.o diff --git a/drivers/infiniband/hw/ipath/ipath_driver.c b/drivers/infiniband/hw/ipath/ipath_driver.c index 09a13c1..1aeddb4 100644 --- a/drivers/infiniband/hw/ipath/ipath_driver.c +++ b/drivers/infiniband/hw/ipath/ipath_driver.c @@ -390,12 +390,16 @@ static int __devinit ipath_init_one(struct pci_dev *pdev, /* setup the chip-specific functions, as early as possible. */ switch (ent->device) { +#ifdef CONFIG_HT_IRQ case PCI_DEVICE_ID_INFINIPATH_HT: ipath_init_iba6110_funcs(dd); break; +#endif +#ifdef CONFIG_PCI_MSI case PCI_DEVICE_ID_INFINIPATH_PE800: ipath_init_iba6120_funcs(dd); break; +#endif default: ipath_dev_err(dd, "Found unknown QLogic deviceid 0x%x, " "failing\n", ent->device); -- cgit v0.10.2 From 4c1b6d18bf2fdeb5ac725126c6928aaa98c8e22f Mon Sep 17 00:00:00 2001 From: Arnaud Giersch Date: Thu, 16 Nov 2006 01:19:21 -0800 Subject: [PATCH] parport: fix compilation failure Fix compilation failure. Signed-off-by: Arnaud Giersch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/parport/parport_ip32.c b/drivers/parport/parport_ip32.c index e3e1927..ec44efd 100644 --- a/drivers/parport/parport_ip32.c +++ b/drivers/parport/parport_ip32.c @@ -780,7 +780,7 @@ static irqreturn_t parport_ip32_interrupt(int irq, void *dev_id) enum parport_ip32_irq_mode irq_mode = priv->irq_mode; switch (irq_mode) { case PARPORT_IP32_IRQ_FWD: - parport_generic_irq(irq, p, regs); + parport_generic_irq(irq, p); break; case PARPORT_IP32_IRQ_HERE: parport_ip32_wakeup(p); -- cgit v0.10.2 From d6ddf55440833fd9404138026af246c51ebeef22 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 16 Nov 2006 01:19:22 -0800 Subject: [PATCH] hfs_fill_super returns success even if no root inode http://kernelfun.blogspot.com/2006/11/mokb-14-11-2006-linux-26x-selinux.html mount that image... fs: filesystem was not cleanly unmounted, running fsck.hfs is recommended. mounting read-only. hfs: get root inode failed. BUG: unable to handle kernel NULL pointer dereference at virtual address 00000018 printing eip ... EIP is at superblock_doinit+0x21/0x767 ... [] selinux_sb_kern_mount+0xc/0x4b [] vfs_kern_mount+0x99/0xf6 [] do_kern_mount+0x2d/0x3e [] do_mount+0x5fa/0x66d [] sys_mount+0x77/0xae [] syscall_call+0x7/0xb DWARF2 unwinder stuck at syscall_call+0x7/0xb hfs_fill_super() returns success even if root_inode = hfs_iget(sb, &fd.search_key->cat, &rec); or sb->s_root = d_alloc_root(root_inode); fails. This superblock finds its way to superblock_doinit() which does: struct dentry *root = sb->s_root; struct inode *inode = root->d_inode; and boom. Need to make sure the error cases return an error, I think. [akpm@osdl.org: return -ENOMEM on oom] Signed-off-by: Eric Sandeen Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hfs/super.c b/fs/hfs/super.c index d43b4fc..85b17b3 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -390,11 +390,13 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) hfs_find_exit(&fd); goto bail_no_root; } + res = -EINVAL; root_inode = hfs_iget(sb, &fd.search_key->cat, &rec); hfs_find_exit(&fd); if (!root_inode) goto bail_no_root; + res = -ENOMEM; sb->s_root = d_alloc_root(root_inode); if (!sb->s_root) goto bail_iput; -- cgit v0.10.2 From 55aa601e14cbec987fa577a895e9297df1d0a71d Mon Sep 17 00:00:00 2001 From: Phillip Susi Date: Thu, 16 Nov 2006 01:19:23 -0800 Subject: [PATCH] Update udf documentation to reflect current state of read/write support Change Documentation/filesystems/udf.txt from saying that read/write mounts on cd media are not supported to instead state the current level of support. Specifically that it works fine on dvd+rw media and can be made to work on cd-rw media via the pktcdvd device. Cc: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/filesystems/udf.txt b/Documentation/filesystems/udf.txt index 511b423..fde829a 100644 --- a/Documentation/filesystems/udf.txt +++ b/Documentation/filesystems/udf.txt @@ -7,8 +7,17 @@ If you encounter problems with reading UDF discs using this driver, please report them to linux_udf@hpesjro.fc.hp.com, which is the developer's list. -Write support requires a block driver which supports writing. The current -scsi and ide cdrom drivers do not support writing. +Write support requires a block driver which supports writing. Currently +dvd+rw drives and media support true random sector writes, and so a udf +filesystem on such devices can be directly mounted read/write. CD-RW +media however, does not support this. Instead the media can be formatted +for packet mode using the utility cdrwtool, then the pktcdvd driver can +be bound to the underlying cd device to provide the required buffering +and read-modify-write cycles to allow the filesystem random sector writes +while providing the hardware with only full packet writes. While not +required for dvd+rw media, use of the pktcdvd driver often enhances +performance due to very poor read-modify-write support supplied internally +by drive firmware. ------------------------------------------------------------------------------- The following mount options are supported: -- cgit v0.10.2 From 6897083abfb0156b533ab8ac42c47f68c550ca9e Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 16 Nov 2006 01:19:25 -0800 Subject: [PATCH] dell_rbu: fix error check platform_device_register_simple() returns error code as pointer when it fails. The return value should be checked by IS_ERR(). Cc: Abhay Salunke Signed-off-by: Akinobu Mita Cc: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/firmware/dell_rbu.c b/drivers/firmware/dell_rbu.c index 08b1617..fc702e4 100644 --- a/drivers/firmware/dell_rbu.c +++ b/drivers/firmware/dell_rbu.c @@ -705,17 +705,16 @@ static struct bin_attribute rbu_packet_size_attr = { static int __init dcdrbu_init(void) { - int rc = 0; + int rc; spin_lock_init(&rbu_data.lock); init_packet_head(); - rbu_device = - platform_device_register_simple("dell_rbu", -1, NULL, 0); - if (!rbu_device) { + rbu_device = platform_device_register_simple("dell_rbu", -1, NULL, 0); + if (IS_ERR(rbu_device)) { printk(KERN_ERR "dell_rbu:%s:platform_device_register_simple " "failed\n", __FUNCTION__); - return -EIO; + return PTR_ERR(rbu_device); } rc = sysfs_create_bin_file(&rbu_device->dev.kobj, &rbu_data_attr); -- cgit v0.10.2 From 64aaa4f8b7e676ccd3f6448df6867097b873b0a6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 16 Nov 2006 01:19:27 -0800 Subject: [PATCH] AFS: Amend the AFS configuration options Amend the text of AFS configuration options. Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/Kconfig b/fs/Kconfig index 133dcc8..7b1511d 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -2060,8 +2060,7 @@ config CODA_FS_OLD_API For most cases you probably want to say N. config AFS_FS -# for fs/nls/Config.in - tristate "Andrew File System support (AFS) (Experimental)" + tristate "Andrew File System support (AFS) (EXPERIMENTAL)" depends on INET && EXPERIMENTAL select RXRPC help -- cgit v0.10.2 From 565762f3fae23ec4db26607cf4726de7b5075b3b Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 16 Nov 2006 01:19:28 -0800 Subject: [PATCH] Don't give bad kprobes example aka ") < 0))" typo Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt index ba26201..d71faff 100644 --- a/Documentation/kprobes.txt +++ b/Documentation/kprobes.txt @@ -442,9 +442,10 @@ static int __init kprobe_init(void) kp.fault_handler = handler_fault; kp.symbol_name = "do_fork"; - if ((ret = register_kprobe(&kp) < 0)) { + ret = register_kprobe(&kp); + if (ret < 0) { printk("register_kprobe failed, returned %d\n", ret); - return -1; + return ret; } printk("kprobe registered\n"); return 0; -- cgit v0.10.2 From da63fc7ce63b43426dc3c69c05e28de2872c159a Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 16 Nov 2006 01:19:28 -0800 Subject: [PATCH] fat: add fat_getattr() This adds fat_getattr() for setting stat->blksize. (FAT uses the size of cluster for proper I/O) Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fat/file.c b/fs/fat/file.c index 8337451..0aa813d 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -303,7 +303,17 @@ void fat_truncate(struct inode *inode) fat_flush_inodes(inode->i_sb, inode, NULL); } +int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + generic_fillattr(inode, stat); + stat->blksize = MSDOS_SB(inode->i_sb)->cluster_size; + return 0; +} +EXPORT_SYMBOL_GPL(fat_getattr); + struct inode_operations fat_file_inode_operations = { .truncate = fat_truncate, .setattr = fat_notify_change, + .getattr = fat_getattr, }; diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c index b0f01b3..4524619 100644 --- a/fs/msdos/namei.c +++ b/fs/msdos/namei.c @@ -654,6 +654,7 @@ static struct inode_operations msdos_dir_inode_operations = { .rmdir = msdos_rmdir, .rename = msdos_rename, .setattr = fat_notify_change, + .getattr = fat_getattr, }; static int msdos_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index edb711f..0afd745 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -1004,6 +1004,7 @@ static struct inode_operations vfat_dir_inode_operations = { .rmdir = vfat_rmdir, .rename = vfat_rename, .setattr = fat_notify_change, + .getattr = fat_getattr, }; static int vfat_fill_super(struct super_block *sb, void *data, int silent) diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h index ce6c858..24a9ef1 100644 --- a/include/linux/msdos_fs.h +++ b/include/linux/msdos_fs.h @@ -402,6 +402,8 @@ extern const struct file_operations fat_file_operations; extern struct inode_operations fat_file_inode_operations; extern int fat_notify_change(struct dentry * dentry, struct iattr * attr); extern void fat_truncate(struct inode *inode); +extern int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat); /* fat/inode.c */ extern void fat_attach(struct inode *inode, loff_t i_pos); -- cgit v0.10.2 From 31be8309532a6743f301cb2e83bd12ca07988b09 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 16 Nov 2006 01:19:29 -0800 Subject: [PATCH] Fix strange size check in __get_vm_area_node() Recently, __get_vm_area_node() was changed like following if (unlikely(!area)) return NULL; - if (unlikely(!size)) { - kfree (area); + if (unlikely(!size)) return NULL; - } It is leaking `area', also original code seems strange already. Probably, we wanted to do this patch. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 7dc6aa7..86897ee 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -181,14 +181,13 @@ static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long fl } addr = ALIGN(start, align); size = PAGE_ALIGN(size); + if (unlikely(!size)) + return NULL; area = kmalloc_node(sizeof(*area), gfp_mask & GFP_LEVEL_MASK, node); if (unlikely(!area)) return NULL; - if (unlikely(!size)) - return NULL; - /* * We always allocate a guard page. */ -- cgit v0.10.2 From ae56fb16337c882c52806508f93ead4034004c7a Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Thu, 16 Nov 2006 01:19:30 -0800 Subject: [PATCH] eCryptfs: CIFS nlink fixes When CIFS is the lower filesystem, the old lower dentry needs to be explicitly dropped from inside eCryptfs to force a revalidate. In addition, when CIFS is the lower filesystem, the inode attributes need to be copied back up from the lower inode to the eCryptfs inode on an eCryptfs revalidate. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c index 0b9992a..52d1e36 100644 --- a/fs/ecryptfs/dentry.c +++ b/fs/ecryptfs/dentry.c @@ -57,6 +57,12 @@ static int ecryptfs_d_revalidate(struct dentry *dentry, struct nameidata *nd) rc = lower_dentry->d_op->d_revalidate(lower_dentry, nd); nd->dentry = dentry_save; nd->mnt = vfsmount_save; + if (dentry->d_inode) { + struct inode *lower_inode = + ecryptfs_inode_to_lower(dentry->d_inode); + + ecryptfs_copy_attr_all(dentry->d_inode, lower_inode); + } out: return rc; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index ebec8cf..dfcc684 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -470,6 +470,7 @@ out_lock: unlock_dir(lower_dir_dentry); dput(lower_new_dentry); dput(lower_old_dentry); + d_drop(lower_old_dentry); d_drop(new_dentry); d_drop(old_dentry); return rc; @@ -484,7 +485,7 @@ static int ecryptfs_unlink(struct inode *dir, struct dentry *dentry) lock_parent(lower_dentry); rc = vfs_unlink(lower_dir_inode, lower_dentry); if (rc) { - ecryptfs_printk(KERN_ERR, "Error in vfs_unlink\n"); + printk(KERN_ERR "Error in vfs_unlink; rc = [%d]\n", rc); goto out_unlock; } ecryptfs_copy_attr_times(dir, lower_dir_inode); -- cgit v0.10.2 From 097b8457dafe7efc22201b4062e2d1e82e494067 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 16 Nov 2006 01:19:31 -0800 Subject: [PATCH] scsi: clear garbage after CDBs on SG_IO ATAPI devices transfer fixed number of bytes for CDBs (12 or 16). Some ATAPI devices choke when shorter CDB is used and the left bytes contain garbage. Block SG_IO cleared left bytes but SCSI SG_IO didn't. This patch makes SCSI SG_IO clear it and simplify CDB clearing in block SG_IO. Signed-off-by: Tejun Heo Cc: Mathieu Fluhr Cc: James Bottomley Cc: Douglas Gilbert Acked-by: Jens Axboe Cc: Acked-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index a19338e..e55a756 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -286,9 +286,8 @@ static int sg_io(struct file *file, request_queue_t *q, * fill in request structure */ rq->cmd_len = hdr->cmd_len; + memset(rq->cmd, 0, BLK_MAX_CDB); /* ATAPI hates garbage after CDB */ memcpy(rq->cmd, cmd, hdr->cmd_len); - if (sizeof(rq->cmd) != hdr->cmd_len) - memset(rq->cmd + hdr->cmd_len, 0, sizeof(rq->cmd) - hdr->cmd_len); memset(sense, 0, sizeof(sense)); rq->sense = sense; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index d2c02df..3ac4890 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -410,6 +410,7 @@ int scsi_execute_async(struct scsi_device *sdev, const unsigned char *cmd, goto free_req; req->cmd_len = cmd_len; + memset(req->cmd, 0, BLK_MAX_CDB); /* ATAPI hates garbage after CDB */ memcpy(req->cmd, cmd, req->cmd_len); req->sense = sioc->sense; req->sense_len = 0; -- cgit v0.10.2 From 073ae841d6a5098f7c6e17fc1f329350d950d1ce Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 16 Nov 2006 10:59:12 +0200 Subject: IPoIB: Clear high octet in QP number IPoIB assumes that high (reserved) octet in the hardware address is 0, and copies it into the QPN. This violates RFC 4391 (which requires that the high 8 bits are ignored on receive), and will result in an invalid QPN being used when interoperating with IPoIB connected mode. Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 1eaf00e..85522da 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -49,6 +49,8 @@ #include +#define IPOIB_QPN(ha) (be32_to_cpup((__be32 *) ha) & 0xffffff) + MODULE_AUTHOR("Roland Dreier"); MODULE_DESCRIPTION("IP-over-InfiniBand net driver"); MODULE_LICENSE("Dual BSD/GPL"); @@ -520,8 +522,7 @@ static void neigh_add_path(struct sk_buff *skb, struct net_device *dev) memcpy(&neigh->dgid.raw, &path->pathrec.dgid.raw, sizeof(union ib_gid)); - ipoib_send(dev, skb, path->ah, - be32_to_cpup((__be32 *) skb->dst->neighbour->ha)); + ipoib_send(dev, skb, path->ah, IPOIB_QPN(skb->dst->neighbour->ha)); } else { neigh->ah = NULL; __skb_queue_tail(&neigh->queue, skb); @@ -599,8 +600,7 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, ipoib_dbg(priv, "Send unicast ARP to %04x\n", be16_to_cpu(path->pathrec.dlid)); - ipoib_send(dev, skb, path->ah, - be32_to_cpup((__be32 *) phdr->hwaddr)); + ipoib_send(dev, skb, path->ah, IPOIB_QPN(phdr->hwaddr)); } else if ((path->query || !path_rec_start(dev, path)) && skb_queue_len(&path->queue) < IPOIB_MAX_PATH_REC_QUEUE) { /* put pseudoheader back on for next time */ @@ -661,8 +661,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) goto out; } - ipoib_send(dev, skb, neigh->ah, - be32_to_cpup((__be32 *) skb->dst->neighbour->ha)); + ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(skb->dst->neighbour->ha)); goto out; } @@ -694,7 +693,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) IPOIB_GID_FMT "\n", skb->dst ? "neigh" : "dst", be16_to_cpup((__be16 *) skb->data), - be32_to_cpup((__be32 *) phdr->hwaddr), + IPOIB_QPN(phdr->hwaddr), IPOIB_GID_RAW_ARG(phdr->hwaddr + 4)); dev_kfree_skb_any(skb); ++priv->stats.tx_dropped; @@ -777,7 +776,7 @@ static void ipoib_neigh_destructor(struct neighbour *n) ipoib_dbg(priv, "neigh_destructor for %06x " IPOIB_GID_FMT "\n", - be32_to_cpup((__be32 *) n->ha), + IPOIB_QPN(n->ha), IPOIB_GID_RAW_ARG(n->ha + 4)); spin_lock_irqsave(&priv->lock, flags); -- cgit v0.10.2 From 6b3d1a95ba714bfb1cc81362f7f3e01b7654b4f3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 16 Nov 2006 10:22:03 +0100 Subject: [PATCH] x86-64: Fix vsyscall.c compilation on UP Broken by earlier patch by me. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index a730bac..92546c1 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -274,6 +274,7 @@ static void __cpuinit cpu_vsyscall_init(void *arg) vsyscall_set_cpu(raw_smp_processor_id()); } +#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) { @@ -282,6 +283,7 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); return NOTIFY_DONE; } +#endif static void __init map_vsyscall(void) { -- cgit v0.10.2 From ccf9ff524ccb195d648ecb0b168340560b42532c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 16 Nov 2006 11:49:16 +0100 Subject: [PATCH] x86_64: fix CONFIG_CC_STACKPROTECTOR build bug on x86_64, the CONFIG_CC_STACKPROTECTOR build fails if used in a distcc setup that has "CC" defined to "distcc gcc": gcc: gcc: linker input file unused because linking not done gcc: gcc: linker input file unused because linking not done gcc: gcc: linker input file unused because linking not done this is because the gcc-x86_64-has-stack-protector.sh script has a 2-parameters assumption. Fix this by passing $(CC) as a single parameter. Signed-off-by: Ingo Molnar Please-Use-Me-More: make randconfig Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 1397214..6e38d4d 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -66,8 +66,8 @@ AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) -cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector ) -cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector-all ) +cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh "$(CC)" -fstack-protector ) +cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh "$(CC)" -fstack-protector-all ) CFLAGS += $(cflags-y) CFLAGS_KERNEL += $(cflags-kernel-y) -- cgit v0.10.2 From 3da2495c0a92723d58cacaaff48dc60a29ddaae6 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 14 Nov 2006 16:28:01 -0500 Subject: OHCI: disallow autostop when wakeup is not available This patch (as822) prevents the OHCI autostop mechanism from kicking in if the root hub is not able or not allowed to issue wakeup requests. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ohci-hub.c b/drivers/usb/host/ohci-hub.c index 6f11359..da09e79 100644 --- a/drivers/usb/host/ohci-hub.c +++ b/drivers/usb/host/ohci-hub.c @@ -422,7 +422,8 @@ ohci_hub_status_data (struct usb_hcd *hcd, char *buf) ohci->autostop = 0; ohci->next_statechange = jiffies + STATECHANGE_DELAY; - } else if (time_after_eq (jiffies, + } else if (device_may_wakeup(&hcd->self.root_hub->dev) + && time_after_eq(jiffies, ohci->next_statechange) && !ohci->ed_rm_list && !(ohci->hc_control & -- cgit v0.10.2 From 40c36092f75ae2026e35feb4f85caa143b64423a Mon Sep 17 00:00:00 2001 From: Kjell Myksvoll Date: Sun, 22 Oct 2006 23:26:42 +0200 Subject: USB: ftdi_sio: adds vendor/product id for a RFID construction kit Adds the vendor and prodcut id for a RFID construction kit from the Elektor Electronics magazine, september 2006. From: Kjell Myksvoll Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index bd76b4c..c971d78 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -511,6 +511,7 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(FTDI_VID, FTDI_TACTRIX_OPENPORT_13M_PID) }, { USB_DEVICE(FTDI_VID, FTDI_TACTRIX_OPENPORT_13S_PID) }, { USB_DEVICE(FTDI_VID, FTDI_TACTRIX_OPENPORT_13U_PID) }, + { USB_DEVICE(ELEKTOR_VID, ELEKTOR_FT323R_PID) }, { }, /* Optional parameter entry */ { } /* Terminating entry */ }; diff --git a/drivers/usb/serial/ftdi_sio.h b/drivers/usb/serial/ftdi_sio.h index f0edb87..30921f5 100644 --- a/drivers/usb/serial/ftdi_sio.h +++ b/drivers/usb/serial/ftdi_sio.h @@ -175,9 +175,15 @@ #define FTDI_ASK_RDR400_PID 0xC991 /* ASK RDR 400 series card reader */ /* + * FTDI USB UART chips used in construction projects from the + * Elektor Electronics magazine (http://elektor-electronics.co.uk) + */ +#define ELEKTOR_VID 0x0C7D +#define ELEKTOR_FT323R_PID 0x0005 /* RFID-Reader, issue 09-2006 */ + +/* * DSS-20 Sync Station for Sony Ericsson P800 */ - #define FTDI_DSS20_PID 0xFC82 /* -- cgit v0.10.2 From fad14a0da885714c8610982045a6d04a4886865e Mon Sep 17 00:00:00 2001 From: Frank Sievertsen Date: Fri, 20 Oct 2006 09:43:53 +0200 Subject: USB: ftdi driver pid for dmx-interfaces Please add a usb pid to the ftdi_sio driver. The pid is used by dmx4all dmx-interfaces (for stage lighting). The interfaces are using the usb-id 0403:c850. I added the id to the driver and it works perfectly. I added a patch for linux 2.6.18.1, too. From: Frank Sievertsen Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index c971d78..c186b4e 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -311,6 +311,7 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(FTDI_VID, FTDI_ACTZWAVE_PID) }, { USB_DEVICE(FTDI_VID, FTDI_IRTRANS_PID) }, { USB_DEVICE(FTDI_VID, FTDI_IPLUS_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_DMX4ALL) }, { USB_DEVICE(FTDI_VID, FTDI_SIO_PID) }, { USB_DEVICE(FTDI_VID, FTDI_8U232AM_PID) }, { USB_DEVICE(FTDI_VID, FTDI_8U232AM_ALT_PID) }, diff --git a/drivers/usb/serial/ftdi_sio.h b/drivers/usb/serial/ftdi_sio.h index 30921f5..bae117d 100644 --- a/drivers/usb/serial/ftdi_sio.h +++ b/drivers/usb/serial/ftdi_sio.h @@ -55,6 +55,9 @@ /* iPlus device */ #define FTDI_IPLUS_PID 0xD070 /* Product Id */ +/* DMX4ALL DMX Interfaces */ +#define FTDI_DMX4ALL 0xC850 + /* www.crystalfontz.com devices - thanx for providing free devices for evaluation ! */ /* they use the ftdi chipset for the USB interface and the vendor id is the same */ #define FTDI_XF_632_PID 0xFC08 /* 632: 16x2 Character Display */ -- cgit v0.10.2 From 51b5bce8c253b82d4789161cc3b0c74bee313bb1 Mon Sep 17 00:00:00 2001 From: Phil Dibowitz Date: Thu, 2 Nov 2006 23:14:10 -0800 Subject: USB: Fix UCR-61S2B unusual_dev entry Recently this entry's bcd scope was narrowed so as not to falsly apply to bcd's other than 0x0110. But while it breaks those of a larger bcd, it is still needed for those of a smaller bcd - so this changes the lower bcd limit to 0x0000. Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index bc1ac07..e87fb538 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1323,8 +1323,10 @@ UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, /* Reported by Kevin Cernekee * Tested on hardware version 1.10. * Entry is needed only for the initializer function override. + * Devices with bcd > 110 seem to not need it while those + * with bcd < 110 appear to need it. */ -UNUSUAL_DEV( 0x1019, 0x0c55, 0x0110, 0x0110, +UNUSUAL_DEV( 0x1019, 0x0c55, 0x0000, 0x0110, "Desknote", "UCR-61S2B", US_SC_DEVICE, US_PR_DEVICE, usb_stor_ucr61s2b_init, -- cgit v0.10.2 From 583ceada075597a5b6acab1140d61ac81586a2a6 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 24 Oct 2006 12:04:22 -0400 Subject: USB: OHCI: fix root-hub resume bug When a suspended OHCI controller sees a port's status change, it sets both the Root-Hub-Status-Change and the Resume-Detect bits in the Interrupt Status register. Processing both these bits, the driver tries to resume the root hub twice! This patch (as807) fixes the bug by ignoring RD if RHSC is set. It also prints a slightly more informative log message when a remote-wakeup event occurs. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c index 9be6b30..ea4714e 100644 --- a/drivers/usb/host/ohci-hcd.c +++ b/drivers/usb/host/ohci-hcd.c @@ -715,13 +715,6 @@ static irqreturn_t ohci_irq (struct usb_hcd *hcd) return IRQ_NOTMINE; } - if (ints & OHCI_INTR_RHSC) { - ohci_vdbg (ohci, "rhsc\n"); - ohci->next_statechange = jiffies + STATECHANGE_DELAY; - ohci_writel (ohci, OHCI_INTR_RHSC, ®s->intrstatus); - usb_hcd_poll_rh_status(hcd); - } - if (ints & OHCI_INTR_UE) { disable (ohci); ohci_err (ohci, "OHCI Unrecoverable Error, disabled\n"); @@ -731,9 +724,21 @@ static irqreturn_t ohci_irq (struct usb_hcd *hcd) ohci_usb_reset (ohci); } - if (ints & OHCI_INTR_RD) { - ohci_vdbg (ohci, "resume detect\n"); - ohci_writel (ohci, OHCI_INTR_RD, ®s->intrstatus); + if (ints & OHCI_INTR_RHSC) { + ohci_vdbg(ohci, "rhsc\n"); + ohci->next_statechange = jiffies + STATECHANGE_DELAY; + ohci_writel(ohci, OHCI_INTR_RD | OHCI_INTR_RHSC, + ®s->intrstatus); + usb_hcd_poll_rh_status(hcd); + } + + /* For connect and disconnect events, we expect the controller + * to turn on RHSC along with RD. But for remote wakeup events + * this might not happen. + */ + else if (ints & OHCI_INTR_RD) { + ohci_vdbg(ohci, "resume detect\n"); + ohci_writel(ohci, OHCI_INTR_RD, ®s->intrstatus); hcd->poll_rh = 1; if (ohci->autostop) { spin_lock (&ohci->lock); diff --git a/drivers/usb/host/ohci-hub.c b/drivers/usb/host/ohci-hub.c index da09e79..6995ea3 100644 --- a/drivers/usb/host/ohci-hub.c +++ b/drivers/usb/host/ohci-hub.c @@ -169,7 +169,8 @@ __acquires(ohci->lock) break; case OHCI_USB_RESUME: /* HCFS changes sometime after INTR_RD */ - ohci_info (ohci, "wakeup\n"); + ohci_info(ohci, "%swakeup\n", + autostopped ? "auto-" : ""); break; case OHCI_USB_OPER: /* this can happen after resuming a swsusp snapshot */ -- cgit v0.10.2 From bb7eef6eea53633a8a49f014fd27c08f7d5fda1a Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Wed, 8 Nov 2006 19:58:07 -0800 Subject: USB: correct keymapping on Powerbook built-in USB ISO keyboards MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit similar to the version in adbhid_input_register(): The '<>' key and the '^°' key on a german keyboard is swapped. Provide correct keys to userland, external USB keyboards will not work correctly when the 'badmap'/'goodmap' workarounds from xkeyboard-config are used. It is expected that distributions drop the badmap/goodmap part from keycodes/macintosh in the xkeyboard-config package. This is probably 2.6.18.x material, if major distros settle on 2.6.18. Signed-off-by: Olaf Hering Cc: Dmitry Torokhov Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 6daf85c..8fde85c 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1797,10 +1797,10 @@ static const struct hid_blacklist { { USB_VENDOR_ID_APPLE, 0x020E, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x020F, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x0214, HID_QUIRK_POWERBOOK_HAS_FN }, - { USB_VENDOR_ID_APPLE, 0x0215, HID_QUIRK_POWERBOOK_HAS_FN }, + { USB_VENDOR_ID_APPLE, 0x0215, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_POWERBOOK_ISO_KEYBOARD}, { USB_VENDOR_ID_APPLE, 0x0216, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x0217, HID_QUIRK_POWERBOOK_HAS_FN }, - { USB_VENDOR_ID_APPLE, 0x0218, HID_QUIRK_POWERBOOK_HAS_FN }, + { USB_VENDOR_ID_APPLE, 0x0218, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_POWERBOOK_ISO_KEYBOARD}, { USB_VENDOR_ID_APPLE, 0x0219, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x030A, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x030B, HID_QUIRK_POWERBOOK_HAS_FN }, diff --git a/drivers/usb/input/hid-input.c b/drivers/usb/input/hid-input.c index 9a808a3..68e7ebb 100644 --- a/drivers/usb/input/hid-input.c +++ b/drivers/usb/input/hid-input.c @@ -121,6 +121,12 @@ static struct hidinput_key_translation powerbook_numlock_keys[] = { { } }; +static struct hidinput_key_translation powerbook_iso_keyboard[] = { + { KEY_GRAVE, KEY_102ND }, + { KEY_102ND, KEY_GRAVE }, + { } +}; + static int usbhid_pb_fnmode = 1; module_param_named(pb_fnmode, usbhid_pb_fnmode, int, 0644); MODULE_PARM_DESC(pb_fnmode, @@ -195,6 +201,14 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, } } + if (hid->quirks & HID_QUIRK_POWERBOOK_ISO_KEYBOARD) { + trans = find_translation(powerbook_iso_keyboard, usage->code); + if (trans) { + input_event(input, usage->type, trans->to, value); + return 1; + } + } + return 0; } @@ -210,6 +224,9 @@ static void hidinput_pb_setup(struct input_dev *input) for (trans = powerbook_numlock_keys; trans->from; trans++) set_bit(trans->to, input->keybit); + + for (trans = powerbook_iso_keyboard; trans->from; trans++) + set_bit(trans->to, input->keybit); } #else static inline int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, diff --git a/drivers/usb/input/hid.h b/drivers/usb/input/hid.h index 9b50eff..0e76e6d 100644 --- a/drivers/usb/input/hid.h +++ b/drivers/usb/input/hid.h @@ -260,6 +260,7 @@ struct hid_item { #define HID_QUIRK_POWERBOOK_HAS_FN 0x00001000 #define HID_QUIRK_POWERBOOK_FN_ON 0x00002000 #define HID_QUIRK_INVERT_HWHEEL 0x00004000 +#define HID_QUIRK_POWERBOOK_ISO_KEYBOARD 0x00008000 /* * This is the global environment of the parser. This information is -- cgit v0.10.2 From a3878f11ed29c50b7da1336adcac089e9c741fc2 Mon Sep 17 00:00:00 2001 From: Jan Mate Date: Wed, 8 Nov 2006 19:58:04 -0800 Subject: USB Storage: unusual_devs.h entry for Sony Ericsson P990i USB Storage: this patch adds support for Sony Ericsson P990i Signed-off-by: Jan Mate Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index e87fb538..cc701e8 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1320,6 +1320,13 @@ UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), +/* Reported by Jan Mate */ +UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, + "Sony Ericsson", + "P990i", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + /* Reported by Kevin Cernekee * Tested on hardware version 1.10. * Entry is needed only for the initializer function override. -- cgit v0.10.2 From a7dc4eeac8f18de5fc6bea1a0f46e67f42b83509 Mon Sep 17 00:00:00 2001 From: Julien BLACHE Date: Sun, 12 Nov 2006 11:22:42 +0100 Subject: USB: hid-core: Add quirk for new Apple keyboard/trackpad The new Core2 Duo MacBook Pro have a new keyboard+trackpad device. The following patch adds the needed HID quirk for the Fn key. Signed-off-by: Julien BLACHE Signed-off-by: Vojtech Pavlik Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 8fde85c..6d08a3b 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1802,6 +1802,7 @@ static const struct hid_blacklist { { USB_VENDOR_ID_APPLE, 0x0217, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x0218, HID_QUIRK_POWERBOOK_HAS_FN | HID_QUIRK_POWERBOOK_ISO_KEYBOARD}, { USB_VENDOR_ID_APPLE, 0x0219, HID_QUIRK_POWERBOOK_HAS_FN }, + { USB_VENDOR_ID_APPLE, 0x021B, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x030A, HID_QUIRK_POWERBOOK_HAS_FN }, { USB_VENDOR_ID_APPLE, 0x030B, HID_QUIRK_POWERBOOK_HAS_FN }, -- cgit v0.10.2 From 70708f2c2a3c164e9aa80345919a22c838b3b314 Mon Sep 17 00:00:00 2001 From: Sergey Vlasov Date: Mon, 6 Nov 2006 16:33:07 +0300 Subject: usb-storage: Remove duplicated unusual_devs.h entries for Sony Ericsson P990i For some reason the unusual_devs.h entry for Sony Ericsson P990i had three identical copies in a wrong place in the file in addition to the correct entry. Signed-off-by: Sergey Vlasov Signed-off-by: Phil Dibowitz Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index cc701e8..efb047f 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -1306,27 +1306,6 @@ UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), -/* Reported by Jan Mate */ -UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, - "Sony Ericsson", - "P990i", - US_SC_DEVICE, US_PR_DEVICE, NULL, - US_FL_FIX_CAPACITY ), - -/* Reported by Jan Mate */ -UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, - "Sony Ericsson", - "P990i", - US_SC_DEVICE, US_PR_DEVICE, NULL, - US_FL_FIX_CAPACITY ), - -/* Reported by Jan Mate */ -UNUSUAL_DEV( 0x0fce, 0xe030, 0x0000, 0x0000, - "Sony Ericsson", - "P990i", - US_SC_DEVICE, US_PR_DEVICE, NULL, - US_FL_FIX_CAPACITY ), - /* Reported by Kevin Cernekee * Tested on hardware version 1.10. * Entry is needed only for the initializer function override. -- cgit v0.10.2 From 6ab16a9029b0b26c23a4806d90ca76be6d6beae3 Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Tue, 7 Nov 2006 10:16:25 +0100 Subject: USB: Fixed outdated usb_get_device_descriptor() documentation usb_get_device_descriptor() used to convert several descriptor fields to host CPU's byte order. Now that it doesn't convert them anymore, update the documentation to reflect this. Signed-off-by: Laurent Pinchart Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index fccd195..7729c07 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -828,10 +828,7 @@ char *usb_cache_string(struct usb_device *udev, int index) * Context: !in_interrupt () * * Updates the copy of the device descriptor stored in the device structure, - * which dedicates space for this purpose. Note that several fields are - * converted to the host CPU's byte order: the USB version (bcdUSB), and - * vendors product and version fields (idVendor, idProduct, and bcdDevice). - * That lets device drivers compare against non-byteswapped constants. + * which dedicates space for this purpose. * * Not exported, only for use by the core. If drivers really want to read * the device descriptor directly, they can call usb_get_descriptor() with -- cgit v0.10.2 From 0029908ba9661ef26f7020309966aae23c2027b8 Mon Sep 17 00:00:00 2001 From: Alex Sanks Date: Sun, 29 Oct 2006 16:38:31 -0800 Subject: USB: ipaq: Add HTC Modem Support Adds support for HTC Smart Phones in modem mode (as opposed to sync mode). Loads and works with pppd on my T-Mobile SDA. Signed-off-by: Alex Sanks Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/ipaq.c b/drivers/usb/serial/ipaq.c index 6238aff..d72cf8b 100644 --- a/drivers/usb/serial/ipaq.c +++ b/drivers/usb/serial/ipaq.c @@ -320,6 +320,7 @@ static struct usb_device_id ipaq_id_table [] = { { USB_DEVICE(0x0B05, 0x9200) }, /* ASUS USB Sync */ { USB_DEVICE(0x0B05, 0x9202) }, /* ASUS USB Sync */ { USB_DEVICE(0x0BB4, 0x00CE) }, /* HTC USB Sync */ + { USB_DEVICE(0x0BB4, 0x00CF) }, /* HTC USB Modem */ { USB_DEVICE(0x0BB4, 0x0A01) }, /* PocketPC USB Sync */ { USB_DEVICE(0x0BB4, 0x0A02) }, /* PocketPC USB Sync */ { USB_DEVICE(0x0BB4, 0x0A03) }, /* PocketPC USB Sync */ -- cgit v0.10.2 From 5a3fcf5c7f035de8e2b28d144d67b7bebac8a723 Mon Sep 17 00:00:00 2001 From: Mariusz Kozlowski Date: Tue, 7 Nov 2006 00:31:51 +0100 Subject: USB: auerswald possible memleak fix fix possible memory leak in auerbuf_setup(). Regards, Mariusz Kozlowski Signed-off-by: Mariusz Kozlowski Signed-off-by: Wolfgang Muees Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/misc/auerswald.c b/drivers/usb/misc/auerswald.c index 0be9d62..e4971d6 100644 --- a/drivers/usb/misc/auerswald.c +++ b/drivers/usb/misc/auerswald.c @@ -780,7 +780,7 @@ static int auerbuf_setup (pauerbufctl_t bcp, unsigned int numElements, unsigned bl_fail:/* not enough memory. Free allocated elements */ dbg ("auerbuf_setup: no more memory"); - kfree(bep); + auerbuf_free(bep); auerbuf_free_buffers (bcp); return -ENOMEM; } -- cgit v0.10.2 From e45413eb708c1cf21082764457692c8eeac0ca97 Mon Sep 17 00:00:00 2001 From: Amol Lad Date: Thu, 5 Oct 2006 14:26:02 +0400 Subject: W1: ioremap balanced with iounmap ioremap must be balanced with iounmap in error path. Please consider for 2.6.19. Signed-off-by: Amol Lad Signed-off-by: Evgeniy Polyakov Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/w1/masters/matrox_w1.c b/drivers/w1/masters/matrox_w1.c index 2788b8c..6f9d880 100644 --- a/drivers/w1/masters/matrox_w1.c +++ b/drivers/w1/masters/matrox_w1.c @@ -215,6 +215,8 @@ static int __devinit matrox_w1_probe(struct pci_dev *pdev, const struct pci_devi return 0; err_out_free_device: + if (dev->virt_addr) + iounmap(dev->virt_addr); kfree(dev); return err; -- cgit v0.10.2 From 7bb0386f102ece8819182ccf7fffe8bbebc32b19 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Thu, 19 Oct 2006 12:24:42 +0900 Subject: debugfs: check return value correctly The return value is stored in "*dentry", not in "dentry". Signed-off-by: Akinobu Mita Signed-off-by: Greg Kroah-Hartman diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index e77676d..a736d44 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -147,13 +147,13 @@ static int debugfs_create_by_name(const char *name, mode_t mode, *dentry = NULL; mutex_lock(&parent->d_inode->i_mutex); *dentry = lookup_one_len(name, parent, strlen(name)); - if (!IS_ERR(dentry)) { + if (!IS_ERR(*dentry)) { if ((mode & S_IFMT) == S_IFDIR) error = debugfs_mkdir(parent->d_inode, *dentry, mode); else error = debugfs_create(parent->d_inode, *dentry, mode); } else - error = PTR_ERR(dentry); + error = PTR_ERR(*dentry); mutex_unlock(&parent->d_inode->i_mutex); return error; -- cgit v0.10.2 From d355c3c23ce56ab83e41f2bfb30d02fb90618530 Mon Sep 17 00:00:00 2001 From: Dennis Stosberg Date: Mon, 13 Nov 2006 09:15:20 +0100 Subject: aoe: Add forgotten NULL at end of attribute list in aoeblk.c This caused the system to stall when the aoe module was loaded. The error was introduced in commit 4ca5224f3ea4779054d96e885ca9b3980801ce13 Signed-off-by: Dennis Stosberg Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index d433f27..aa25f8b 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -68,6 +68,7 @@ static struct attribute *aoe_attrs[] = { &disk_attr_mac.attr, &disk_attr_netif.attr, &disk_attr_fwver.attr, + NULL }; static const struct attribute_group attr_group = { -- cgit v0.10.2 From 4f71c5de19c27f2198105d3b26b398494d5c353b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 17 Nov 2006 15:35:00 +1100 Subject: [PATCH] Fix radeon DDC regression When radeonfb was changed to use the new "generic" ddc, a bit of code initializing the GPIO lines was lost, causing it to not work if the firmware didn't configure them properly, which seems to happen on some cards. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds diff --git a/drivers/video/aty/radeon_i2c.c b/drivers/video/aty/radeon_i2c.c index 6767545..869725a 100644 --- a/drivers/video/aty/radeon_i2c.c +++ b/drivers/video/aty/radeon_i2c.c @@ -139,7 +139,13 @@ void radeon_delete_i2c_busses(struct radeonfb_info *rinfo) int radeon_probe_i2c_connector(struct radeonfb_info *rinfo, int conn, u8 **out_edid) { - u8 *edid = fb_ddc_read(&rinfo->i2c[conn-1].adapter); + u32 reg = rinfo->i2c[conn-1].ddc_reg; + u8 *edid; + + OUTREG(reg, INREG(reg) & + ~(VGA_DDC_DATA_OUTPUT | VGA_DDC_CLK_OUTPUT)); + + edid = fb_ddc_read(&rinfo->i2c[conn-1].adapter); if (out_edid) *out_edid = edid; -- cgit v0.10.2 From 4be703906cffd5902028d20626e636ba21fb0b61 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 16 Nov 2006 22:18:28 -0800 Subject: Fix generic fb_ddc i2c edid probe msg Benh points out that the msgs[0].flags entry never got initialized, and since it's an automatic stack allocation, it could have any random value, which is bad. Rewrite the initializer to explicitly initialize all fields of the small i2c_msg structure array we generate. Just to keep it all obvious, let's handle msgs[1].buf in the same initializer while we're at it, instead of initializing that one separately later. Signed-off-by: Linus Torvalds diff --git a/drivers/video/fb_ddc.c b/drivers/video/fb_ddc.c index 3aa6ebf..f836137 100644 --- a/drivers/video/fb_ddc.c +++ b/drivers/video/fb_ddc.c @@ -20,26 +20,26 @@ static unsigned char *fb_do_probe_ddc_edid(struct i2c_adapter *adapter) { unsigned char start = 0x0; + unsigned char *buf = kmalloc(EDID_LENGTH, GFP_KERNEL); struct i2c_msg msgs[] = { { .addr = DDC_ADDR, + .flags = 0, .len = 1, .buf = &start, }, { .addr = DDC_ADDR, .flags = I2C_M_RD, .len = EDID_LENGTH, + .buf = buf, } }; - unsigned char *buf; - buf = kmalloc(EDID_LENGTH, GFP_KERNEL); if (!buf) { dev_warn(&adapter->dev, "unable to allocate memory for EDID " "block.\n"); return NULL; } - msgs[1].buf = buf; if (i2c_transfer(adapter, msgs, 2) == 2) return buf; -- cgit v0.10.2 From 1d08811d0c05cd54a778f45588ec22eee027ff89 Mon Sep 17 00:00:00 2001 From: Jan-Benedict Glaw Date: Fri, 17 Nov 2006 10:32:04 +0100 Subject: lkkbd: Remove my old snail-mail address I moved to a different town and my old snail-mail address is invalid now. Also, there's no need at all to have any address like that in the sources, so remove it completely. Signed-off-by: Jan-Benedict Glaw diff --git a/drivers/input/keyboard/lkkbd.c b/drivers/input/keyboard/lkkbd.c index 708d5a1..979b93e 100644 --- a/drivers/input/keyboard/lkkbd.c +++ b/drivers/input/keyboard/lkkbd.c @@ -59,11 +59,6 @@ * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Should you need to contact me, the author, you can do so either by - * email or by paper mail: - * Jan-Benedict Glaw, Lilienstraße 16, 33790 Hörste (near Halle/Westf.), - * Germany. */ #include -- cgit v0.10.2 From 0796bdb7e9e4a48b401f4fba1ee5dc79a45528ef Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 17 Nov 2006 05:57:49 +0100 Subject: [PATCH] x86_64: stack unwinder crash fix the new dwarf2 unwinder crashes while trying to dump the stack: Leftover inexact backtrace: Unable to handle kernel paging request at ffffffff82800000 RIP: [] dump_trace+0x35b/0x3d2 PGD 203027 PUD 205027 PMD 0 Oops: 0000 [2] PREEMPT SMP CPU 0 Modules linked in: Pid: 30, comm: khelper Not tainted 2.6.19-rc6-rt1 #11 RIP: 0010:[] [] dump_trace+0x35b/0x3d2 RSP: 0000:ffff81003fb9d848 EFLAGS: 00010006 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffffff805b3520 RDI: 0000000000000000 RBP: ffffffff827ffff9 R08: ffffffff80aad000 R09: 0000000000000005 R10: ffffffff80aae000 R11: ffffffff8037961b R12: ffff81003fb9d858 R13: 0000000000000000 R14: ffffffff80598460 R15: ffffffff80ab1fc0 FS: 0000000000000000(0000) GS:ffffffff806c4200(0000) knlGS:0000000000000000 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: ffffffff82800000 CR3: 0000000000201000 CR4: 00000000000006e0 this crash happened because it did not sanitize the dwarf2 data it got, and got an unaligned stack pointer - which happily walked past the process stack (and eventually reached the end of kernel memory and pagefaulted there) due to this naive iteration condition: HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0); note that i386 is alot more conservative when it comes to trusting stack pointers: static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) { return p > (void *)tinfo && p < (void *)tinfo + THREAD_SIZE - 3; } but the x86_64 code did not take this bit of i386 code. The fix is to align the stack pointer. Signed-off-by: Ingo Molnar Cc: Andi Kleen Cc: Jan Beulich Cc: Thomas Gleixner Cc: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 7819022a..a153d0a 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -290,6 +290,12 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s if (tsk && tsk != current) stack = (unsigned long *)tsk->thread.rsp; } + /* + * Align the stack pointer on word boundary, later loops + * rely on that (and corruption / debug info bugs can cause + * unaligned values here): + */ + stack = (unsigned long *)((unsigned long)stack & ~(sizeof(long)-1)); /* * Print function call entries within a stack. 'cond' is the -- cgit v0.10.2 From dc1829a4c378d793fb3b95d56135d89a0d7ff72a Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 17 Nov 2006 14:26:18 +0100 Subject: [PATCH] i386/x86_64: ACPI cpu_idle_wait() fix The scheduler on Andreas Friedrich's hyperthreading system stopped working properly: the scheduler would never move tasks to another CPU! The lask known working kernel was 2.6.8. After a couple of attempts to corner the bug, the following smoking gun was found: BIOS reported wrong ACPI idfor the processor CPU#1: set_cpus_allowed(), swapper:1, 3 -> 2 [] show_trace_log_lvl+0x34/0x4a [] show_trace+0x2c/0x2e [] dump_stack+0x2b/0x2d [] set_cpus_allowed+0x52/0xec [] cpu_idle_wait+0x2e/0x100 [] acpi_processor_power_exit+0x45/0x58 [] acpi_processor_remove+0x46/0xea [] acpi_start_single_object+0x47/0x54 [] acpi_bus_register_driver+0xa4/0xd3 [] acpi_processor_init+0x57/0x77 [] init+0x146/0x2fd [] kernel_thread_helper+0x7/0x10 a quick look at cpu_idle_wait() shows how broken that code is on i386: it changes the init task's affinity map but never restores it ... and because all userspace tasks get forked by init, they all inherited that single-CPU affinity mask. x86_64 cloned this bug too. Signed-off-by: Ingo Molnar Cc: Andreas Friedrich Cc: Wolfgang Erig Cc: Andrew Morton Cc: Adrian Bunk Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 1e1fa3e..dd53c58 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -205,7 +205,7 @@ void cpu_idle(void) void cpu_idle_wait(void) { unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map; + cpumask_t map, tmp = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); put_cpu(); @@ -227,6 +227,8 @@ void cpu_idle_wait(void) } cpus_and(map, map, cpu_online_map); } while (!cpus_empty(map)); + + set_cpus_allowed(current, tmp); } EXPORT_SYMBOL_GPL(cpu_idle_wait); diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index f622605..7451a4c 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -144,7 +144,7 @@ static void poll_idle (void) void cpu_idle_wait(void) { unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map; + cpumask_t map, tmp = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); put_cpu(); @@ -167,6 +167,8 @@ void cpu_idle_wait(void) } cpus_and(map, map, cpu_online_map); } while (!cpus_empty(map)); + + set_cpus_allowed(current, tmp); } EXPORT_SYMBOL_GPL(cpu_idle_wait); -- cgit v0.10.2 From 1ff5683043196b9ad628a5de6bf8eeca52ee8bfd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 17 Nov 2006 19:57:22 +0100 Subject: [PATCH] lockdep: fix static keys in module-allocated percpu areas lockdep got confused by certain locks in modules: INFO: trying to register non-static key. the code is fine but needs lockdep annotation. turning off the locking correctness validator. Call Trace: [] dump_trace+0xaa/0x3f2 [] show_trace+0x3a/0x60 [] dump_stack+0x15/0x17 [] __lock_acquire+0x724/0x9bb [] lock_acquire+0x4d/0x67 [] rt_spin_lock+0x3d/0x41 [] :ip_conntrack:__ip_ct_refresh_acct+0x131/0x174 [] :ip_conntrack:udp_packet+0xbf/0xcf [] :ip_conntrack:ip_conntrack_in+0x394/0x4a7 [] nf_iterate+0x41/0x7f [] nf_hook_slow+0x64/0xd5 [] ip_rcv+0x24e/0x506 [...] Steven Rostedt found the bug: static_obj() check did not take PERCPU_ENOUGH_ROOM into account, so in-module DEFINE_PER_CPU-area locks were triggering this message. Signed-off-by: Ingo Molnar Signed-off-by: Steven Rostedt Signed-off-by: Linus Torvalds diff --git a/kernel/lockdep.c b/kernel/lockdep.c index b739be2..c9fefdb 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -1081,7 +1081,8 @@ static int static_obj(void *obj) */ for_each_possible_cpu(i) { start = (unsigned long) &__per_cpu_start + per_cpu_offset(i); - end = (unsigned long) &__per_cpu_end + per_cpu_offset(i); + end = (unsigned long) &__per_cpu_start + PERCPU_ENOUGH_ROOM + + per_cpu_offset(i); if ((addr >= start) && (addr < end)) return 1; -- cgit v0.10.2 From 808dbbb6bb61173bf52946a28f99089d2efa4c55 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 17 Nov 2006 11:14:56 -0800 Subject: x86: be more careful when walking back the frame pointer chain When showing the stack backtrace, make sure that we never accept not only an unchanging frame pointer, but also a frame pointer that moves back down the stack frame. It must always grow up (toward older stack frames). I doubt this has triggered, but a subtly corrupt stack with extremely unlucky contents could cause us to loop forever on a bogus endless frame pointer chain. This review was triggered by much worse problems happening in some of the other stack unwinding code. Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 00489b7..fe9c5e8 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -129,15 +129,19 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo, #ifdef CONFIG_FRAME_POINTER while (valid_stack_ptr(tinfo, (void *)ebp)) { + unsigned long new_ebp; addr = *(unsigned long *)(ebp + 4); ops->address(data, addr); /* * break out of recursive entries (such as - * end_of_stack_stop_unwind_function): + * end_of_stack_stop_unwind_function). Also, + * we can never allow a frame pointer to + * move downwards! */ - if (ebp == *(unsigned long *)ebp) + new_ebp = *(unsigned long *)ebp; + if (new_ebp <= ebp) break; - ebp = *(unsigned long *)ebp; + ebp = new_ebp; } #else while (valid_stack_ptr(tinfo, stack)) { -- cgit v0.10.2 From b976fe19acc565e5137e6f12af7b6633a23e6b7c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 17 Nov 2006 19:31:09 -0800 Subject: Revert "ACPI: created a dedicated workqueue for notify() execution" This reverts commit 37605a6900f6b4d886d995751fcfeef88c4e462c. Again. This same bug has now been introduced twice: it was done earlier by commit b8d35192c55fb055792ff0641408eaaec7c88988, only to be reverted last time in commit 72945b2b90a5554975b8f72673ab7139d232a121. We must NOT try to queue up notify handlers to another thread than the normal ACPI execution thread, because the notifications on some systems seem to just keep on accumulating until we run out of memory and/or threads. Keeping events within the one deferred execution thread automatically throttles the events properly. At least the Compaq N620c will lock up completely on the first thermal event without this patch reverted. Cc: David Brownell Cc: Len Brown Cc: Alexey Starikovskiy Signed-off-by: Linus Torvalds diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index c84286c..068fe4f 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -73,7 +73,6 @@ static unsigned int acpi_irq_irq; static acpi_osd_handler acpi_irq_handler; static void *acpi_irq_context; static struct workqueue_struct *kacpid_wq; -static struct workqueue_struct *kacpi_notify_wq; acpi_status acpi_os_initialize(void) { @@ -92,9 +91,8 @@ acpi_status acpi_os_initialize1(void) return AE_NULL_ENTRY; } kacpid_wq = create_singlethread_workqueue("kacpid"); - kacpi_notify_wq = create_singlethread_workqueue("kacpi_notify"); BUG_ON(!kacpid_wq); - BUG_ON(!kacpi_notify_wq); + return AE_OK; } @@ -106,7 +104,6 @@ acpi_status acpi_os_terminate(void) } destroy_workqueue(kacpid_wq); - destroy_workqueue(kacpi_notify_wq); return AE_OK; } @@ -569,7 +566,10 @@ void acpi_os_derive_pci_id(acpi_handle rhandle, /* upper bound */ static void acpi_os_execute_deferred(void *context) { - struct acpi_os_dpc *dpc = (struct acpi_os_dpc *)context; + struct acpi_os_dpc *dpc = NULL; + + + dpc = (struct acpi_os_dpc *)context; if (!dpc) { printk(KERN_ERR PREFIX "Invalid (NULL) context\n"); return; @@ -604,12 +604,14 @@ acpi_status acpi_os_execute(acpi_execute_type type, struct acpi_os_dpc *dpc; struct work_struct *task; + ACPI_FUNCTION_TRACE("os_queue_for_execution"); + ACPI_DEBUG_PRINT((ACPI_DB_EXEC, "Scheduling function [%p(%p)] for deferred execution.\n", function, context)); if (!function) - return AE_BAD_PARAMETER; + return_ACPI_STATUS(AE_BAD_PARAMETER); /* * Allocate/initialize DPC structure. Note that this memory will be @@ -622,20 +624,26 @@ acpi_status acpi_os_execute(acpi_execute_type type, * from the same memory. */ - dpc = kmalloc(sizeof(struct acpi_os_dpc) + - sizeof(struct work_struct), GFP_ATOMIC); + dpc = + kmalloc(sizeof(struct acpi_os_dpc) + sizeof(struct work_struct), + GFP_ATOMIC); if (!dpc) - return AE_NO_MEMORY; + return_ACPI_STATUS(AE_NO_MEMORY); + dpc->function = function; dpc->context = context; + task = (void *)(dpc + 1); INIT_WORK(task, acpi_os_execute_deferred, (void *)dpc); - if (!queue_work((type == OSL_NOTIFY_HANDLER)? - kacpi_notify_wq : kacpid_wq, task)) { - status = AE_ERROR; + + if (!queue_work(kacpid_wq, task)) { + ACPI_DEBUG_PRINT((ACPI_DB_ERROR, + "Call to queue_work() failed.\n")); kfree(dpc); + status = AE_ERROR; } - return status; + + return_ACPI_STATUS(status); } EXPORT_SYMBOL(acpi_os_execute); -- cgit v0.10.2 From ba9b1cd713b33720ca0f035dab2fdbc10bf7328d Mon Sep 17 00:00:00 2001 From: David Weinehall Date: Sat, 18 Nov 2006 03:58:58 +0100 Subject: [PATCH] Update my CREDITS entry I moved from Sweden to Finland 2.5 years ago, thought it might be time to update my CREDITS entry (simply removing the address completely seemed the sanest option). Signed-off-by: David Weinehall Signed-off-by: Linus Torvalds diff --git a/CREDITS b/CREDITS index 606d407..ccd4f9f 100644 --- a/CREDITS +++ b/CREDITS @@ -3511,14 +3511,12 @@ D: The Linux Support Team Erlangen N: David Weinehall E: tao@acc.umu.se +P: 1024D/DC47CA16 7ACE 0FB0 7A74 F994 9B36 E1D1 D14E 8526 DC47 CA16 W: http://www.acc.umu.se/~tao/ -W: http://www.acc.umu.se/~mcalinux/ +D: v2.0 kernel maintainer D: Fixes for the NE/2-driver D: Miscellaneous MCA-support D: Cleanup of the Config-files -S: Axtorpsvagen 40:20 -S: S-903 37 UMEA -S: Sweden N: Matt Welsh E: mdw@metalab.unc.edu -- cgit v0.10.2 From 610a5b742e9df4e59047f22d13d8bd83cafce388 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Fri, 17 Nov 2006 11:51:41 +1100 Subject: [CRYPTO] api: Remove one too many semicolon This patch has removed one too many semicolon in crypto.h. Signed-off-by: Yoichi Yuasa Signed-off-by: Herbert Xu diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 8f2ffa4..6485e97 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -245,7 +245,7 @@ int crypto_alg_available(const char *name, u32 flags) __deprecated_for_modules; int crypto_has_alg(const char *name, u32 type, u32 mask); #else -static int crypto_alg_available(const char *name, u32 flags); +static int crypto_alg_available(const char *name, u32 flags) __deprecated_for_modules; static inline int crypto_alg_available(const char *name, u32 flags) { -- cgit v0.10.2 From dfbc9e9d33adb1ac9910dd7f8ceb911947039a52 Mon Sep 17 00:00:00 2001 From: Daniel Ritz Date: Sat, 18 Nov 2006 22:19:34 -0800 Subject: [PATCH] pcmcia: fix 'rmmod pcmcia' with unbound devices Having unbound PCMCIA devices: doing a 'find /sys' after a 'rmmod pcmcia' gives an oops because the pcmcia_device is not unregisterd from the driver core. fixes bugzilla #7481 Signed-off-by: Daniel Ritz Dominik Brodowski Cc: Pavol Gono Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index 0f70192..a20d84d 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -1271,6 +1271,9 @@ static void pcmcia_bus_remove_socket(struct class_device *class_dev, socket->pcmcia_state.dead = 1; pccard_register_pcmcia(socket, NULL); + /* unregister any unbound devices */ + pcmcia_card_remove(socket, NULL); + pcmcia_put_socket(socket); return; -- cgit v0.10.2 From a6cd2d94e1072a5756b5e5ab647d3223cba7e555 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 18 Nov 2006 22:19:36 -0800 Subject: [PATCH] i2c-ixp4xx: fix ") != 0))" typo i2c_bit_add_bus() returns -E; -E != 0 => err = 1 probe fails with positive error code Signed-off-by: Alexey Dobriyan Cc: Deepak Saxena Acked-by: Jean Delvare Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/i2c/busses/i2c-ixp4xx.c b/drivers/i2c/busses/i2c-ixp4xx.c index 1ce01fb..05fffb9 100644 --- a/drivers/i2c/busses/i2c-ixp4xx.c +++ b/drivers/i2c/busses/i2c-ixp4xx.c @@ -137,7 +137,8 @@ static int ixp4xx_i2c_probe(struct platform_device *plat_dev) gpio_line_set(gpio->scl_pin, 0); gpio_line_set(gpio->sda_pin, 0); - if ((err = i2c_bit_add_bus(&drv_data->adapter) != 0)) { + err = i2c_bit_add_bus(&drv_data->adapter); + if (err != 0) printk(KERN_ERR "ERROR: Could not install %s\n", plat_dev->dev.bus_id); kfree(drv_data); -- cgit v0.10.2 From ffb3d1348605816de10d4e57281e02f606508b6c Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Sat, 18 Nov 2006 22:19:39 -0800 Subject: [PATCH] scx200_acb: handle PCI errors Signed-off-by: Jeff Garzik Signed-off-by: Jean Delvare Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/i2c/busses/scx200_acb.c b/drivers/i2c/busses/scx200_acb.c index 32aab0d..714bae7 100644 --- a/drivers/i2c/busses/scx200_acb.c +++ b/drivers/i2c/busses/scx200_acb.c @@ -494,11 +494,12 @@ static __init int scx200_create_pci(const char *text, struct pci_dev *pdev, iface->pdev = pdev; iface->bar = bar; - pci_enable_device_bars(iface->pdev, 1 << iface->bar); + rc = pci_enable_device_bars(iface->pdev, 1 << iface->bar); + if (rc) + goto errout_free; rc = pci_request_region(iface->pdev, iface->bar, iface->adapter.name); - - if (rc != 0) { + if (rc) { printk(KERN_ERR NAME ": can't allocate PCI BAR %d\n", iface->bar); goto errout_free; -- cgit v0.10.2 From 8243229f0940ab4e9f501879d3ffb7476b02ee6a Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Sat, 18 Nov 2006 22:19:40 -0800 Subject: [PATCH] x86_64: fix memory hotplug build with NUMA=n This is to fix compile error of x86-64 memory hotplug without any NUMA option. CC arch/x86_64/mm/init.o arch/x86_64/mm/init.c:501: error: redefinition of 'memory_add_physaddr_to_nid' include/linux/memory_hotplug.h:71: error: previous definition of 'memory_add_phys addr_to_nid' was here arch/x86_64/mm/init.c:509: error: redefinition of 'memory_add_physaddr_to_nid' arch/x86_64/mm/init.c:501: error: previous definition of 'memory_add_physaddr_to_ nid' was here I confirmed compile completion with !NUMA, (NUMA & !ACPI_NUMA), or (NUMA & ACPI_NUMA). Signed-off-by: Yasunori Goto Acked-by: Andi Kleen Cc: "Randy.Dunlap" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index f1f977a..4c0c00e 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -496,7 +496,7 @@ int remove_memory(u64 start, u64 size) } EXPORT_SYMBOL_GPL(remove_memory); -#ifndef CONFIG_ACPI_NUMA +#if !defined(CONFIG_ACPI_NUMA) && defined(CONFIG_NUMA) int memory_add_physaddr_to_nid(u64 start) { return 0; @@ -504,13 +504,6 @@ int memory_add_physaddr_to_nid(u64 start) EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); #endif -#ifndef CONFIG_ACPI_NUMA -int memory_add_physaddr_to_nid(u64 start) -{ - return 0; -} -#endif - #endif /* CONFIG_MEMORY_HOTPLUG */ #ifdef CONFIG_MEMORY_HOTPLUG_RESERVE -- cgit v0.10.2 From f0c69c4ee796a2d2277c3a000e24f29a25a00060 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 18 Nov 2006 22:19:41 -0800 Subject: [PATCH] ftape: fix printk format warnings Fix printk format warnings: drivers/char/ftape/zftape/zftape-buffers.c:87: warning: format '%d' expects type 'int', but argument 3 has type 'size_t' drivers/char/ftape/zftape/zftape-buffers.c:104: warning: format '%d' expects type 'int', but argument 3 has type 'size_t' Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ftape/zftape/zftape-buffers.c b/drivers/char/ftape/zftape/zftape-buffers.c index da06f13..7ebce2e 100644 --- a/drivers/char/ftape/zftape/zftape-buffers.c +++ b/drivers/char/ftape/zftape/zftape-buffers.c @@ -85,7 +85,7 @@ int zft_vmalloc_once(void *new, size_t size) peak_memory = used_memory; } TRACE_ABORT(0, ft_t_noise, - "allocated buffer @ %p, %d bytes", *(void **)new, size); + "allocated buffer @ %p, %zd bytes", *(void **)new, size); } int zft_vmalloc_always(void *new, size_t size) { @@ -101,7 +101,7 @@ void zft_vfree(void *old, size_t size) if (*(void **)old) { vfree(*(void **)old); used_memory -= size; - TRACE(ft_t_noise, "released buffer @ %p, %d bytes", + TRACE(ft_t_noise, "released buffer @ %p, %zd bytes", *(void **)old, size); *(void **)old = NULL; } -- cgit v0.10.2 From 49a1cd00b599d12c3f397e5a32f81f6e2aab0d74 Mon Sep 17 00:00:00 2001 From: Toralf Foerster Date: Sat, 18 Nov 2006 22:19:41 -0800 Subject: [PATCH] fix build error for HISAX_NETJET Fix a build error for the enter:now PCI card. Signed-off-by: Toralf Foerster Acked-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/isdn/hisax/Kconfig b/drivers/isdn/hisax/Kconfig index eb57a98..cfd2718 100644 --- a/drivers/isdn/hisax/Kconfig +++ b/drivers/isdn/hisax/Kconfig @@ -344,7 +344,7 @@ config HISAX_HFC_SX config HISAX_ENTERNOW_PCI bool "Formula-n enter:now PCI card" - depends on PCI && (BROKEN || !(SPARC || PPC || PARISC || M68K || FRV)) + depends on HISAX_NETJET && PCI && (BROKEN || !(SPARC || PPC || PARISC || M68K || FRV)) help This enables HiSax support for the Formula-n enter:now PCI ISDN card. -- cgit v0.10.2 From ace5f1d425beaa272b6e91cecc87b2c075d7feb2 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Mon, 20 Nov 2006 15:46:22 +1000 Subject: [PATCH] m68knommu: fix up for the irq_handler_t changes Switch to using irq_handler_t for interrupt function handler pointers. Change name of m68knommu's irq_hanlder_t data structure so it doesn't clash with the common type (include/linux/interrupt.h). Signed-off-by: Greg Ungerer Signed-off-by: Linus Torvalds diff --git a/arch/m68knommu/kernel/setup.c b/arch/m68knommu/kernel/setup.c index bde9811..7b21959 100644 --- a/arch/m68knommu/kernel/setup.c +++ b/arch/m68knommu/kernel/setup.c @@ -62,7 +62,7 @@ int (*mach_kbdrate) (struct kbd_repeat *); void (*mach_kbd_leds) (unsigned int); /* machine dependent irq functions */ void (*mach_init_IRQ) (void); -irqreturn_t (*(*mach_default_handler)[]) (int, void *, struct pt_regs *); +irq_handler_t mach_default_handler; int (*mach_get_irq_list) (struct seq_file *, void *); void (*mach_process_int) (int irq, struct pt_regs *fp); void (*mach_trap_init) (void); diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c index c5667bd..9226264 100644 --- a/arch/m68knommu/kernel/time.c +++ b/arch/m68knommu/kernel/time.c @@ -54,7 +54,7 @@ static irqreturn_t timer_interrupt(int irq, void *dummy, struct pt_regs * regs) update_process_times(user_mode(regs)); #endif if (current->pid) - profile_tick(CPU_PROFILING, regs); + profile_tick(CPU_PROFILING); /* * If we have an externally synchronized Linux clock, then update diff --git a/arch/m68knommu/platform/5307/ints.c b/arch/m68knommu/platform/5307/ints.c index b4b5509..a57239e 100644 --- a/arch/m68knommu/platform/5307/ints.c +++ b/arch/m68knommu/platform/5307/ints.c @@ -33,7 +33,7 @@ /* * This table stores the address info for each vector handler. */ -irq_handler_t irq_list[SYS_IRQS]; +struct irq_entry irq_list[SYS_IRQS]; #define NUM_IRQ_NODES 16 static irq_node_t nodes[NUM_IRQ_NODES]; @@ -44,7 +44,7 @@ volatile unsigned int num_spurious; unsigned int local_bh_count[NR_CPUS]; unsigned int local_irq_count[NR_CPUS]; -static irqreturn_t default_irq_handler(int irq, void *ptr, struct pt_regs *regs) +static irqreturn_t default_irq_handler(int irq, void *ptr) { #if 1 printk(KERN_INFO "%s(%d): default irq handler vec=%d [0x%x]\n", @@ -70,7 +70,7 @@ void __init init_IRQ(void) for (i = 0; i < SYS_IRQS; i++) { if (mach_default_handler) - irq_list[i].handler = (*mach_default_handler)[i]; + irq_list[i].handler = mach_default_handler; else irq_list[i].handler = default_irq_handler; irq_list[i].flags = IRQ_FLG_STD; @@ -100,7 +100,7 @@ irq_node_t *new_irq_node(void) int request_irq( unsigned int irq, - irqreturn_t (*handler)(int, void *, struct pt_regs *), + irq_handler_t handler, unsigned long flags, const char *devname, void *dev_id) @@ -157,7 +157,7 @@ void free_irq(unsigned int irq, void *dev_id) } if (mach_default_handler) - irq_list[irq].handler = (*mach_default_handler)[irq]; + irq_list[irq].handler = mach_default_handler; else irq_list[irq].handler = default_irq_handler; irq_list[irq].flags = IRQ_FLG_STD; @@ -168,8 +168,7 @@ void free_irq(unsigned int irq, void *dev_id) EXPORT_SYMBOL(free_irq); -int sys_request_irq(unsigned int irq, - irqreturn_t (*handler)(int, void *, struct pt_regs *), +int sys_request_irq(unsigned int irq, irq_handler_t handler, unsigned long flags, const char *devname, void *dev_id) { if (irq > IRQ7) { @@ -211,7 +210,7 @@ void sys_free_irq(unsigned int irq, void *dev_id) printk(KERN_WARNING "%s: Removing probably wrong IRQ %d from %s\n", __FUNCTION__, irq, irq_list[irq].devname); - irq_list[irq].handler = (*mach_default_handler)[irq]; + irq_list[irq].handler = mach_default_handler; irq_list[irq].flags = 0; irq_list[irq].dev_id = NULL; irq_list[irq].devname = NULL; @@ -241,7 +240,7 @@ asmlinkage void process_int(unsigned long vec, struct pt_regs *fp) if (vec >= VEC_INT1 && vec <= VEC_INT7) { vec -= VEC_SPUR; kstat_cpu(0).irqs[vec]++; - irq_list[vec].handler(vec, irq_list[vec].dev_id, fp); + irq_list[vec].handler(vec, irq_list[vec].dev_id); } else { if (mach_process_int) mach_process_int(vec, fp); diff --git a/include/asm-m68knommu/irq_regs.h b/include/asm-m68knommu/irq_regs.h new file mode 100644 index 0000000..3dd9c0b --- /dev/null +++ b/include/asm-m68knommu/irq_regs.h @@ -0,0 +1 @@ +#include diff --git a/include/asm-m68knommu/irqnode.h b/include/asm-m68knommu/irqnode.h index a2503df..6132a98 100644 --- a/include/asm-m68knommu/irqnode.h +++ b/include/asm-m68knommu/irqnode.h @@ -8,7 +8,7 @@ * interrupt source (if it supports chaining). */ typedef struct irq_node { - irqreturn_t (*handler)(int, void *, struct pt_regs *); + irq_handler_t handler; unsigned long flags; void *dev_id; const char *devname; @@ -18,12 +18,12 @@ typedef struct irq_node { /* * This structure has only 4 elements for speed reasons */ -typedef struct irq_handler { - irqreturn_t (*handler)(int, void *, struct pt_regs *); +struct irq_entry { + irq_handler_t handler; unsigned long flags; void *dev_id; const char *devname; -} irq_handler_t; +}; /* count of spurious interrupts */ extern volatile unsigned int num_spurious; diff --git a/include/asm-m68knommu/machdep.h b/include/asm-m68knommu/machdep.h index 27c90af..6ce28f8 100644 --- a/include/asm-m68knommu/machdep.h +++ b/include/asm-m68knommu/machdep.h @@ -18,7 +18,7 @@ extern int (*mach_kbdrate) (struct kbd_repeat *); extern void (*mach_kbd_leds) (unsigned int); /* machine dependent irq functions */ extern void (*mach_init_IRQ) (void); -extern irqreturn_t (*(*mach_default_handler)[]) (int, void *, struct pt_regs *); +extern irq_handler_t mach_default_handler; extern int (*mach_request_irq) (unsigned int irq, void (*handler)(int, void *, struct pt_regs *), unsigned long flags, const char *devname, void *dev_id); extern void (*mach_free_irq) (unsigned int irq, void *dev_id); -- cgit v0.10.2 From b3438f8266cb1f5010085ac47d7ad6a36a212164 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 20 Nov 2006 11:47:18 -0800 Subject: Add "pure_initcall" for static variable initialization This is a quick hack to overcome the fact that SRCU currently does not allow static initializers, and we need to sometimes initialize those things before any other initializers (even "core" ones) can do so. Currently we don't allow this at all for modules, and the only user that needs is right now is cpufreq. As reported by Thomas Gleixner: "Commit b4dfdbb3c707474a2254c5b4d7e62be31a4b7da9 ("[PATCH] cpufreq: make the transition_notifier chain use SRCU breaks cpu frequency notification users, which register the callback > on core_init level." Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Arjan van de Ven Cc: Andrew Morton , Signed-off-by: Linus Torvalds diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 86e69b7..dd0c262 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -59,7 +59,7 @@ static int __init init_cpufreq_transition_notifier_list(void) srcu_init_notifier_head(&cpufreq_transition_notifier_list); return 0; } -core_initcall(init_cpufreq_transition_notifier_list); +pure_initcall(init_cpufreq_transition_notifier_list); static LIST_HEAD(cpufreq_governor_list); static DEFINE_MUTEX (cpufreq_governor_mutex); diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 9d87316..e60d6f2 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -215,6 +215,8 @@ .notes : { *(.note.*) } :note #define INITCALLS \ + *(.initcall0.init) \ + *(.initcall0s.init) \ *(.initcall1.init) \ *(.initcall1s.init) \ *(.initcall2.init) \ diff --git a/include/linux/init.h b/include/linux/init.h index ff40ea1..5eb5d24 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -93,6 +93,14 @@ extern void setup_arch(char **); static initcall_t __initcall_##fn##id __attribute_used__ \ __attribute__((__section__(".initcall" level ".init"))) = fn +/* + * A "pure" initcall has no dependencies on anything else, and purely + * initializes variables that couldn't be statically initialized. + * + * This only exists for built-in code, not for modules. + */ +#define pure_initcall(fn) __define_initcall("0",fn,1) + #define core_initcall(fn) __define_initcall("1",fn,1) #define core_initcall_sync(fn) __define_initcall("1s",fn,1s) #define postcore_initcall(fn) __define_initcall("2",fn,2) -- cgit v0.10.2 From 3f5a6ca31c334011fd929501a078424c0d3f71be Mon Sep 17 00:00:00 2001 From: Bryan O'Sullivan Date: Mon, 20 Nov 2006 10:54:34 -0800 Subject: IB/ipath: Depend on CONFIG_NET ipath uses skb functions and won't build without CONFIG_NET. Spotted by Randy Dunlap. Signed-off-by: Bryan O'Sullivan Acked-by: Randy Dunlap Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/ipath/Kconfig b/drivers/infiniband/hw/ipath/Kconfig index 5ca471a..90c1454 100644 --- a/drivers/infiniband/hw/ipath/Kconfig +++ b/drivers/infiniband/hw/ipath/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_IPATH tristate "QLogic InfiniPath Driver" - depends on (PCI_MSI || HT_IRQ) && 64BIT && INFINIBAND + depends on (PCI_MSI || HT_IRQ) && 64BIT && INFINIBAND && NET ---help--- This is a driver for QLogic InfiniPath host channel adapters, including InfiniBand verbs support. This driver allows these -- cgit v0.10.2 From d2133717d5f994cca970b5aeb9d4664feeb92ff4 Mon Sep 17 00:00:00 2001 From: Lachlan McIlroy Date: Tue, 21 Nov 2006 18:55:16 +1100 Subject: [XFS] Fix uninitialized br_state and br_startoff in xfs_bmap_add_extent_delay_real() SGI-PV: 957008 SGI-Modid: xfs-linux-melb:xfs-kern:27457a Signed-off-by: Lachlan McIlroy Signed-off-by: Shailendra Tripathi Signed-off-by: Tim Shimmin diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 5b050c0..498ad50d 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -1171,6 +1171,8 @@ xfs_bmap_add_extent_delay_real( xfs_bmap_trace_pre_update(fname, "0", ip, idx, XFS_DATA_FORK); xfs_bmbt_set_blockcount(ep, temp); r[0] = *new; + r[1].br_state = PREV.br_state; + r[1].br_startblock = 0; r[1].br_startoff = new_endoff; temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff; r[1].br_blockcount = temp2; -- cgit v0.10.2 From e5ffd2bb62c3f2c0d9f34e0d16fab6e2c8b056fb Mon Sep 17 00:00:00 2001 From: David Chinner Date: Tue, 21 Nov 2006 18:55:33 +1100 Subject: [XFS] Stale the correct inode when freeing clusters. SGI-PV: 958376 SGI-Modid: xfs-linux-melb:xfs-kern:27503a Signed-off-by: David Chinner Signed-off-by: Tim Shimmin diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index d72c80d..44dfac5 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2258,7 +2258,7 @@ xfs_ifree_cluster( AIL_LOCK(mp,s); iip->ili_flush_lsn = iip->ili_item.li_lsn; AIL_UNLOCK(mp, s); - xfs_iflags_set(ip, XFS_ISTALE); + xfs_iflags_set(iip->ili_inode, XFS_ISTALE); pre_flushed++; } lip = lip->li_bio_list; -- cgit v0.10.2 From 3af9815328bba76e8d11d71d6dbbd6f38beafe58 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 20 Nov 2006 11:29:09 -0500 Subject: [PATCH] x86_64: Align data segment to PAGE_SIZE boundary o Explicitly align data segment to PAGE_SIZE boundary otherwise depending on config options and tool chain it might be placed on a non PAGE_SIZE aligned boundary and vmlinux loaders like kexec fail when they encounter a PT_LOAD type segment which is not aligned to PAGE_SIZE boundary. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index edb24aa..d9534e7 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -60,6 +60,7 @@ SECTIONS } #endif + . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { *(.data) -- cgit v0.10.2 From 6af6e1efb161ffe36e718b1fd58385710879af7c Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Tue, 21 Nov 2006 16:58:59 -0500 Subject: [PATCH] Fix CPU_FREQ_GOV_ONDEMAND=y compile error The ONDEMAND governor needs FREQ_TABLE Signed-off-by: Mattia Dongili Signed-off-by: Dave Jones Signed-off-by: Linus Torvalds diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index 2cc71b6..491779a 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -107,6 +107,7 @@ config CPU_FREQ_GOV_USERSPACE config CPU_FREQ_GOV_ONDEMAND tristate "'ondemand' cpufreq policy governor" + select CPU_FREQ_TABLE help 'ondemand' - This driver adds a dynamic cpufreq policy governor. The governor does a periodic polling and -- cgit v0.10.2 From ea73ee23c43d6eb75f88c4fc9e1230b44673c0d2 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 6 Nov 2006 09:45:44 -0800 Subject: [IPV6] ROUTE: Try to use router which is not known unreachable. Only routers in "FAILED" state should be considered unreachable. Otherwise, we do not try to use speicific routes unless all least specific routers are considered unreachable. Signed-off-by: YOSHIFUJI Hideaki diff --git a/net/ipv6/route.c b/net/ipv6/route.c index c953466..5132821 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -330,6 +330,8 @@ static int inline rt6_check_neigh(struct rt6_info *rt) read_lock_bh(&neigh->lock); if (neigh->nud_state & NUD_VALID) m = 2; + else if (!(neigh->nud_state & NUD_FAILED)) + m = 1; read_unlock_bh(&neigh->lock); } return m; -- cgit v0.10.2 From 557e92efd44878beccd08d5dd54ed343be0e5819 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 6 Nov 2006 09:45:45 -0800 Subject: [IPV6] ROUTE: Prefer reachable nexthop only if the caller requests. Signed-off-by: YOSHIFUJI Hideaki diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 5132821..a972a66 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -349,9 +349,7 @@ static int rt6_score_route(struct rt6_info *rt, int oif, m |= IPV6_DECODE_PREF(IPV6_EXTRACT_PREF(rt->rt6i_flags)) << 2; #endif n = rt6_check_neigh(rt); - if (n > 1) - m |= 16; - else if (!n && strict & RT6_LOOKUP_F_REACHABLE) + if (!n && (strict & RT6_LOOKUP_F_REACHABLE)) return -1; return m; } -- cgit v0.10.2 From ea659e0775281e1c02556e939bf749bb4f55e50c Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 6 Nov 2006 09:45:45 -0800 Subject: [IPV6] ROUTE: Do not enable router reachability probing in router mode. RFC4191 explicitly states that the procedures are applicable to hosts only. We should not have changed behavior of routers. Signed-off-by: YOSHIFUJI Hideaki diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a972a66..b39ae99 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -380,10 +380,11 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif, continue; if (m > mpri) { - rt6_probe(match); + if (strict & RT6_LOOKUP_F_REACHABLE) + rt6_probe(match); match = rt; mpri = m; - } else { + } else if (strict & RT6_LOOKUP_F_REACHABLE) { rt6_probe(rt); } } @@ -636,7 +637,7 @@ static struct rt6_info *ip6_pol_route_input(struct fib6_table *table, int strict = 0; int attempts = 3; int err; - int reachable = RT6_LOOKUP_F_REACHABLE; + int reachable = ipv6_devconf.forwarding ? 0 : RT6_LOOKUP_F_REACHABLE; strict |= flags & RT6_LOOKUP_F_IFACE; @@ -733,7 +734,7 @@ static struct rt6_info *ip6_pol_route_output(struct fib6_table *table, int strict = 0; int attempts = 3; int err; - int reachable = RT6_LOOKUP_F_REACHABLE; + int reachable = ipv6_devconf.forwarding ? 0 : RT6_LOOKUP_F_REACHABLE; strict |= flags & RT6_LOOKUP_F_IFACE; -- cgit v0.10.2 From b3fdd9f115c776d381b30b296849f8e4046bcdaa Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Mon, 6 Nov 2006 10:06:22 -0800 Subject: [IPV6] IP6TUNNEL: Delete all tunnel device when unloading module. Signed-off-by: Yasuyuki Kozakai Signed-off-by: YOSHIFUJI Hideaki diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 84d7ebd..c8d4160 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1149,6 +1149,20 @@ fail: return err; } +static void __exit ip6ip6_destroy_tunnels(void) +{ + int h; + struct ip6_tnl *t; + + for (h = 0; h < HASH_SIZE; h++) { + while ((t = tnls_r_l[h]) != NULL) + unregister_netdevice(t->dev); + } + + t = tnls_wc[0]; + unregister_netdevice(t->dev); +} + /** * ip6_tunnel_cleanup - free resources and unregister protocol **/ @@ -1158,7 +1172,9 @@ static void __exit ip6_tunnel_cleanup(void) if (xfrm6_tunnel_deregister(&ip6ip6_handler)) printk(KERN_INFO "ip6ip6 close: can't deregister tunnel\n"); - unregister_netdev(ip6ip6_fb_tnl_dev); + rtnl_lock(); + ip6ip6_destroy_tunnels(); + rtnl_unlock(); } module_init(ip6_tunnel_init); -- cgit v0.10.2 From 53ab61c6d8f391bda04dbc1e95bac348fe81103b Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Mon, 6 Nov 2006 10:06:23 -0800 Subject: [IPV6] IP6TUNNEL: Add missing nf_reset() on input path. Signed-off-by: Yasuyuki Kozakai Signed-off-by: YOSHIFUJI Hideaki diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index c8d4160..b9f4029 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -542,6 +542,7 @@ ip6ip6_rcv(struct sk_buff *skb) skb->dev = t->dev; dst_release(skb->dst); skb->dst = NULL; + nf_reset(skb); if (t->parms.flags & IP6_TNL_F_RCV_DSCP_COPY) ipv6_copy_dscp(ipv6h, skb->nh.ipv6h); ip6ip6_ecn_decapsulate(ipv6h, skb); -- cgit v0.10.2 From e52726dece1d2e3976c0caf2f4e9e1c452d31282 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sat, 18 Nov 2006 22:14:05 +0100 Subject: [Bluetooth] Attach low-level connections to the Bluetooth bus To receive uvents for the low-level ACL and SCO links, they must be assigned to a subsystem. It is enough to attach them to the already established Bluetooth bus. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 954eb74..3eeeb7a 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -259,7 +259,9 @@ void hci_conn_add_sysfs(struct hci_conn *conn) BT_DBG("conn %p", conn); - conn->dev.parent = &hdev->dev; + conn->dev.bus = &bt_bus; + conn->dev.parent = &hdev->dev; + conn->dev.release = bt_release; snprintf(conn->dev.bus_id, BUS_ID_SIZE, -- cgit v0.10.2 From 6bd57416127e92d35e6798925502c84e14a3a966 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sat, 18 Nov 2006 22:14:22 +0100 Subject: [Bluetooth] Handling pending connect attempts after inquiry After an inquiry completed or got canceled the Bluetooth core should check for any pending connect attempts. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 65f0948..bb94e6d 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -57,6 +57,7 @@ static void hci_cc_link_ctl(struct hci_dev *hdev, __u16 ocf, struct sk_buff *skb) { __u8 status; + struct hci_conn *pend; BT_DBG("%s ocf 0x%x", hdev->name, ocf); @@ -71,6 +72,15 @@ static void hci_cc_link_ctl(struct hci_dev *hdev, __u16 ocf, struct sk_buff *skb clear_bit(HCI_INQUIRY, &hdev->flags); hci_req_complete(hdev, status); } + + hci_dev_lock(hdev); + + pend = hci_conn_hash_lookup_state(hdev, ACL_LINK, BT_CONNECT2); + if (pend) + hci_acl_connect(pend); + + hci_dev_unlock(hdev); + break; default: @@ -565,11 +575,20 @@ static void hci_cs_info_param(struct hci_dev *hdev, __u16 ocf, __u8 status) static inline void hci_inquiry_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) { __u8 status = *((__u8 *) skb->data); + struct hci_conn *pend; BT_DBG("%s status %d", hdev->name, status); clear_bit(HCI_INQUIRY, &hdev->flags); hci_req_complete(hdev, status); + + hci_dev_lock(hdev); + + pend = hci_conn_hash_lookup_state(hdev, ACL_LINK, BT_CONNECT2); + if (pend) + hci_acl_connect(pend); + + hci_dev_unlock(hdev); } /* Inquiry Result */ -- cgit v0.10.2 From ff2d367ac33b9278c9516ac1888207dac11d10b0 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sat, 18 Nov 2006 22:14:42 +0100 Subject: [Bluetooth] Check if RFCOMM session is still attached to the TTY If the RFCOMM session is no longer attached to the TTY device, then it makes no sense to go through with changing the termios settings. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index b8e3a5f..1fb5d42 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -765,7 +765,7 @@ static void rfcomm_tty_set_termios(struct tty_struct *tty, struct termios *old) BT_DBG("tty %p termios %p", tty, old); - if (!dev) + if (!dev || !dev->dlc || !dev->dlc->session) return; /* Handle turning off CRTSCTS */ -- cgit v0.10.2 From 820ae1b865caa05e0614004d0183ca70de2b8665 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sat, 18 Nov 2006 22:15:00 +0100 Subject: [Bluetooth] Always include MTU in L2CAP config responses When sending a positive config response it shall include the actual MTU to be used on this channel. This differs from the Bluetooth 1.1 specification where it was enough to acknowledge the config request. Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index 2b3dcb8..c09d4a2 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -1353,12 +1353,12 @@ static inline int l2cap_conf_output(struct sock *sk, void **ptr) /* Configure output options and let the other side know * which ones we don't like. */ - if (pi->conf_mtu < pi->omtu) { - l2cap_add_conf_opt(ptr, L2CAP_CONF_MTU, 2, pi->omtu); + if (pi->conf_mtu < pi->omtu) result = L2CAP_CONF_UNACCEPT; - } else { + else pi->omtu = pi->conf_mtu; - } + + l2cap_add_conf_opt(ptr, L2CAP_CONF_MTU, 2, pi->omtu); BT_DBG("sk %p result %d", sk, result); return result; -- cgit v0.10.2 From 354f60a9952d0af59e684b91cc6e06b95fbbf5ef Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sat, 18 Nov 2006 22:15:20 +0100 Subject: [Bluetooth] Ignore L2CAP config requests on disconnect Any L2CAP connection in disconnecting state shall not response to any further config requests from the remote side. So in case such a request is received, ignore it. Signed-off-by: Ville Tervo Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index c09d4a2..bbf78e6 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -1533,6 +1533,9 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr if (!(sk = l2cap_get_chan_by_scid(&conn->chan_list, dcid))) return -ENOENT; + if (sk->sk_state == BT_DISCONN) + goto unlock; + l2cap_parse_conf_req(sk, req->data, cmd->len - sizeof(*req)); if (flags & 0x0001) { -- cgit v0.10.2 From fb47ddb2db9c18664bd7b06c201a2398885b64fc Mon Sep 17 00:00:00 2001 From: David L Stevens Date: Sun, 19 Nov 2006 10:38:39 -0800 Subject: [IGMP]: Fix IGMPV3_EXP() normalization bit shift value. The IGMPV3_EXP() macro doesn't correctly shift the normalization bit, so time-out values are longer than they should be. Thanks to Dirk Ooms for finding the problem in IGMPv3 - MLDv2 had a similar problem that was already fixed a year ago. :-( Signed-off-by: David L Stevens Signed-off-by: David S. Miller diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 03f43e2..21dd569 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -191,7 +191,7 @@ struct ip_mc_list #define IGMPV3_MASK(value, nb) ((nb)>=32 ? (value) : ((1<<(nb))-1) & (value)) #define IGMPV3_EXP(thresh, nbmant, nbexp, value) \ ((value) < (thresh) ? (value) : \ - ((IGMPV3_MASK(value, nbmant) | (1<<(nbmant+nbexp))) << \ + ((IGMPV3_MASK(value, nbmant) | (1<<(nbmant))) << \ (IGMPV3_MASK((value) >> (nbmant), nbexp) + (nbexp)))) #define IGMPV3_QQIC(value) IGMPV3_EXP(0x80, 4, 3, value) -- cgit v0.10.2 From 334f3d45d3e0c925eb15003560cdd6f5383c88d0 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 19 Nov 2006 14:53:07 -0800 Subject: [XFRM]: Sub-policies broke policy events XFRM policy events are broken when sub-policy feature is turned on. A simple test to verify this: run ip xfrm mon on one window and add then delete a policy on another window .. Signed-off-by: Jamal Hadi Salim Acked-by: Masahide NAKAMURA Signed-off-by: David S. Miller diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index b43e764..2dae1c1 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2060,6 +2060,9 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event * len += RTA_SPACE(headlen); headlen = sizeof(*id); } +#ifdef CONFIG_XFRM_SUB_POLICY + len += RTA_SPACE(sizeof(struct xfrm_userpolicy_type)); +#endif len += NLMSG_SPACE(headlen); skb = alloc_skb(len, GFP_ATOMIC); -- cgit v0.10.2 From 785fd8b8a597519127c6d9e16e81f48af2ad391e Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 19 Nov 2006 14:55:30 -0800 Subject: [XFRM]: nlmsg length not computed correctly in the presence of subpolicies I actually dont have a test case for these; i just found them by inspection. Refer to patch "[XFRM]: Sub-policies broke policy events" for more info Signed-off-by: Jamal Hadi Salim Acked-by: Masahide NAKAMURA Signed-off-by: David S. Miller diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 2dae1c1..c4cde57 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1927,6 +1927,9 @@ static int xfrm_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *xt, len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); len += NLMSG_SPACE(sizeof(struct xfrm_user_acquire)); len += RTA_SPACE(xfrm_user_sec_ctx_size(xp)); +#ifdef CONFIG_XFRM_SUB_POLICY + len += RTA_SPACE(sizeof(struct xfrm_userpolicy_type)); +#endif skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; @@ -2034,6 +2037,9 @@ static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_eve len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); len += NLMSG_SPACE(sizeof(struct xfrm_user_polexpire)); len += RTA_SPACE(xfrm_user_sec_ctx_size(xp)); +#ifdef CONFIG_XFRM_SUB_POLICY + len += RTA_SPACE(sizeof(struct xfrm_userpolicy_type)); +#endif skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; @@ -2109,10 +2115,12 @@ static int xfrm_notify_policy_flush(struct km_event *c) struct nlmsghdr *nlh; struct sk_buff *skb; unsigned char *b; + int len = 0; #ifdef CONFIG_XFRM_SUB_POLICY struct xfrm_userpolicy_type upt; + len += RTA_SPACE(sizeof(struct xfrm_userpolicy_type)); #endif - int len = NLMSG_LENGTH(0); + len += NLMSG_LENGTH(0); skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) -- cgit v0.10.2 From 4498c80d9c1ebdf42837be6a52ead35a2aa2b819 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 21 Nov 2006 16:17:41 -0800 Subject: [BLUETOOTH]: Fix unaligned access in hci_send_to_sock. The "u16 *" derefs of skb->data need to be wrapped inside of a get_unaligned(). Thanks to Gustavo Zacarias for the bug report. Signed-off-by: David S. Miller diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index f26a9eb..711a085 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -120,10 +120,13 @@ void hci_send_to_sock(struct hci_dev *hdev, struct sk_buff *skb) if (!hci_test_bit(evt, &flt->event_mask)) continue; - if (flt->opcode && ((evt == HCI_EV_CMD_COMPLETE && - flt->opcode != *(__u16 *)(skb->data + 3)) || - (evt == HCI_EV_CMD_STATUS && - flt->opcode != *(__u16 *)(skb->data + 4)))) + if (flt->opcode && + ((evt == HCI_EV_CMD_COMPLETE && + flt->opcode != + get_unaligned((__u16 *)(skb->data + 3))) || + (evt == HCI_EV_CMD_STATUS && + flt->opcode != + get_unaligned((__u16 *)(skb->data + 4))))) continue; } -- cgit v0.10.2 From 6c12c18dfb0ccae21816de3e0c323b5250450d82 Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Tue, 21 Nov 2006 10:31:08 -0600 Subject: [POWERPC] Revert "[POWERPC] Enable generic rtc hook for the MPC8349 mITX" This reverts commit a8ed4f7ec3aa472134d7de6176f823b2667e450b. As advised by David Brownell: http://marc.theaimsgroup.com/?l=linux-kernel&m=116387226902131&w=2 Signed-off-by: Kim Phillips Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/83xx/mpc834x_itx.c b/arch/powerpc/platforms/83xx/mpc834x_itx.c index 5446bab..e2bcaaf 100644 --- a/arch/powerpc/platforms/83xx/mpc834x_itx.c +++ b/arch/powerpc/platforms/83xx/mpc834x_itx.c @@ -108,10 +108,6 @@ static int __init mpc834x_itx_probe(void) return 1; } -#ifdef CONFIG_RTC_CLASS -late_initcall(rtc_class_hookup); -#endif - define_machine(mpc834x_itx) { .name = "MPC834x ITX", .probe = mpc834x_itx_probe, -- cgit v0.10.2 From df9c23095fc8652798c41dd860676d3dafb2f1dc Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Tue, 21 Nov 2006 10:31:14 -0600 Subject: [POWERPC] Revert "[POWERPC] Add powerpc get/set_rtc_time interface to new generic rtc class" This reverts commit 7a69af63e788a324d162201a0b23df41bcf158dd. As advised by David Brownell: http://marc.theaimsgroup.com/?l=linux-kernel&m=116387226902131&w=2 Signed-off-by: Kim Phillips Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index a1b5e4b..46a24de 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -1014,48 +1014,6 @@ void __init time_init(void) set_dec(tb_ticks_per_jiffy); } -#ifdef CONFIG_RTC_CLASS -static int set_rtc_class_time(struct rtc_time *tm) -{ - int err; - struct class_device *class_dev = - rtc_class_open(CONFIG_RTC_HCTOSYS_DEVICE); - - if (class_dev == NULL) - return -ENODEV; - - err = rtc_set_time(class_dev, tm); - - rtc_class_close(class_dev); - - return 0; -} - -static void get_rtc_class_time(struct rtc_time *tm) -{ - int err; - struct class_device *class_dev = - rtc_class_open(CONFIG_RTC_HCTOSYS_DEVICE); - - if (class_dev == NULL) - return; - - err = rtc_read_time(class_dev, tm); - - rtc_class_close(class_dev); - - return; -} - -int __init rtc_class_hookup(void) -{ - ppc_md.get_rtc_time = get_rtc_class_time; - ppc_md.set_rtc_time = set_rtc_class_time; - - return 0; -} -#endif /* CONFIG_RTC_CLASS */ - #define FEBRUARY 2 #define STARTOFTIME 1970 diff --git a/include/asm-powerpc/time.h b/include/asm-powerpc/time.h index a782850..4cff977 100644 --- a/include/asm-powerpc/time.h +++ b/include/asm-powerpc/time.h @@ -39,10 +39,6 @@ extern void generic_calibrate_decr(void); extern void wakeup_decrementer(void); extern void snapshot_timebase(void); -#ifdef CONFIG_RTC_CLASS -extern int __init rtc_class_hookup(void); -#endif - /* Some sane defaults: 125 MHz timebase, 1GHz processor */ extern unsigned long ppc_proc_freq; #define DEFAULT_PROC_FREQ (DEFAULT_TB_FREQ * 8) -- cgit v0.10.2 From 700f9672c9a61c12334651a94d17ec04620e1976 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 21 Nov 2006 17:33:01 -0800 Subject: [IRDA]: Lockdep fix. On Sat, 2006-11-18 at 16:12 +0300, Andrey Borzenkov wrote: > ============================================= > [ INFO: possible recursive locking detected ] > 2.6.19-rc5-2avb #2 > - --------------------------------------------- > pppd/26425 is trying to acquire lock: > (&hashbin->hb_spinlock){....}, at: [] irlmp_slsap_inuse+0x5a/0x170 > [irda] > > but task is already holding lock: > (&hashbin->hb_spinlock){....}, at: [] irlmp_slsap_inuse+0x37/0x170 > [irda] > > other info that might help us debug this: > 1 lock held by pppd/26425: > #0: (&hashbin->hb_spinlock){....}, at: [] > irlmp_slsap_inuse+0x37/0x170 [irda] > > stack backtrace: > [] dump_trace+0x1cc/0x200 > [] show_trace_log_lvl+0x1a/0x30 > [] show_trace+0x12/0x20 > [] dump_stack+0x19/0x20 > [] __lock_acquire+0x8fa/0xc20 > [] lock_acquire+0x5d/0x80 > [] _spin_lock+0x2c/0x40 > [] irlmp_slsap_inuse+0x5a/0x170 [irda] > [] irlmp_open_lsap+0x62/0x180 [irda] > [] irttp_open_tsap+0x181/0x230 [irda] > [] ircomm_open_tsap+0x5d/0xa0 [ircomm] > [] ircomm_open+0xb8/0xd0 [ircomm] > [] ircomm_tty_open+0x4f7/0x570 [ircomm_tty] > [] tty_open+0x174/0x340 > [] chrdev_open+0x89/0x170 > [] __dentry_open+0xa6/0x1d0 > [] nameidata_to_filp+0x35/0x40 > [] do_filp_open+0x49/0x50 > [] do_sys_open+0x47/0xd0 > [] sys_open+0x1c/0x20 > [] sysenter_past_esp+0x56/0x8d > [] 0xb7f86410 > ======================= The comment at the nesting lock says: /* Careful for priority inversions here ! * irlmp->links is never taken while another IrDA * spinlock is held, so we are safe. Jean II */ So, under the assumption the author was right, it just needs a lockdep annotation. Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: David S. Miller diff --git a/net/irda/irlmp.c b/net/irda/irlmp.c index 5073261..fede837 100644 --- a/net/irda/irlmp.c +++ b/net/irda/irlmp.c @@ -1678,7 +1678,8 @@ static int irlmp_slsap_inuse(__u8 slsap_sel) * every IrLAP connection and check every LSAP associated with each * the connection. */ - spin_lock_irqsave(&irlmp->links->hb_spinlock, flags); + spin_lock_irqsave_nested(&irlmp->links->hb_spinlock, flags, + SINGLE_DEPTH_NESTING); lap = (struct lap_cb *) hashbin_get_first(irlmp->links); while (lap != NULL) { IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, goto errlap;); -- cgit v0.10.2 From f2776ff047229c3e7cee2454e2704dd6f98fa32f Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 21 Nov 2006 17:41:56 -0800 Subject: [IPV6]: Fix address/interface handling in UDP and DCCP, according to the scoping architecture. TCP and RAW do not have this issue. Closes Bug #7432. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index eb0ff7a..fc4242c 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -277,7 +277,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, __u64 seq; sk = inet6_lookup(&dccp_hashinfo, &hdr->daddr, dh->dccph_dport, - &hdr->saddr, dh->dccph_sport, skb->dev->ifindex); + &hdr->saddr, dh->dccph_sport, inet6_iif(skb)); if (sk == NULL) { ICMP6_INC_STATS_BH(__in6_dev_get(skb->dev), ICMP6_MIB_INERRORS); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index e0c3934..c83f23e 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -242,14 +242,13 @@ static void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, { struct ipv6_pinfo *np; struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; - struct net_device *dev = skb->dev; struct in6_addr *saddr = &hdr->saddr; struct in6_addr *daddr = &hdr->daddr; struct udphdr *uh = (struct udphdr*)(skb->data+offset); struct sock *sk; int err; - sk = udp_v6_lookup(daddr, uh->dest, saddr, uh->source, dev->ifindex); + sk = udp_v6_lookup(daddr, uh->dest, saddr, uh->source, inet6_iif(skb)); if (sk == NULL) return; @@ -348,7 +347,7 @@ static void udpv6_mcast_deliver(struct udphdr *uh, read_lock(&udp_hash_lock); sk = sk_head(&udp_hash[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); - dif = skb->dev->ifindex; + dif = inet6_iif(skb); sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); if (!sk) { kfree_skb(skb); @@ -429,7 +428,7 @@ static int udpv6_rcv(struct sk_buff **pskb) * check socket cache ... must talk to Alan about his plans * for sock caches... i'll skip this for now. */ - sk = udp_v6_lookup(saddr, uh->source, daddr, uh->dest, dev->ifindex); + sk = udp_v6_lookup(saddr, uh->source, daddr, uh->dest, inet6_iif(skb)); if (sk == NULL) { if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) -- cgit v0.10.2 From 12862086f24d7382b24379bbcbe0dadf12ca5945 Mon Sep 17 00:00:00 2001 From: "Ira W. Snyder" Date: Tue, 21 Nov 2006 17:44:31 -0800 Subject: [TG3]: Add missing unlock in tg3_open() error path. Sparse noticed a locking imbalance in tg3_open(). This patch adds an unlock to one of the error paths, so that tg3_open() always exits without the lock held. Signed-off-by: Ira W. Snyder Signed-off-by: David S. Miller diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 1dbdd6b..c20bb99 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -6979,8 +6979,10 @@ static int tg3_open(struct net_device *dev) tg3_full_lock(tp, 0); err = tg3_set_power_state(tp, PCI_D0); - if (err) + if (err) { + tg3_full_unlock(tp); return err; + } tg3_disable_ints(tp); tp->tg3_flags &= ~TG3_FLAG_INIT_COMPLETE; -- cgit v0.10.2 From 8746ed3dae14e87e9f7ad8e44649b72e22b33274 Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Tue, 21 Nov 2006 18:19:42 -0600 Subject: [POWERPC] Fix ucc_geth of_device discovery on mpc832x mpc832x, as in mpc8360, needs to explicitly find and create the platform device for ucc_geth in 2.6.19. This code will likely be readapted to Benh's new of_ methods for 2.6.20. Signed-off-by: Kim Phillips Signed-off-by: Paul Mackerras diff --git a/arch/powerpc/platforms/83xx/mpc832x_mds.c b/arch/powerpc/platforms/83xx/mpc832x_mds.c index 54dea9d4..a43ac71 100644 --- a/arch/powerpc/platforms/83xx/mpc832x_mds.c +++ b/arch/powerpc/platforms/83xx/mpc832x_mds.c @@ -24,6 +24,7 @@ #include #include +#include #include #include #include @@ -136,6 +137,24 @@ static void __init mpc832x_sys_setup_arch(void) #endif } +static int __init mpc832x_declare_of_platform_devices(void) +{ + struct device_node *np; + + for (np = NULL; (np = of_find_compatible_node(np, "network", + "ucc_geth")) != NULL;) { + int ucc_num; + char bus_id[BUS_ID_SIZE]; + + ucc_num = *((uint *) get_property(np, "device-id", NULL)) - 1; + snprintf(bus_id, BUS_ID_SIZE, "ucc_geth.%u", ucc_num); + of_platform_device_create(np, bus_id, NULL); + } + + return 0; +} +device_initcall(mpc832x_declare_of_platform_devices); + void __init mpc832x_sys_init_IRQ(void) { -- cgit v0.10.2 From f26b90440cd74c78fe10c9bd5160809704a9627c Mon Sep 17 00:00:00 2001 From: David C Somayajulu Date: Wed, 15 Nov 2006 16:41:09 -0800 Subject: [PATCH] qla4xxx: bug fix: driver hardware semaphore needs to be grabbed before soft reset On qla4xxx, the driver needs to grab the drvr semaphore provided by the hardware, prior to issuing a reset. This patches takes care of a couple of places where it was not being done. In addition there is minor clean up. Signed-off-by: David Somayajulu Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla4xxx/ql4_glbl.h b/drivers/scsi/qla4xxx/ql4_glbl.h index 1b221ff..eeefdcb 100644 --- a/drivers/scsi/qla4xxx/ql4_glbl.h +++ b/drivers/scsi/qla4xxx/ql4_glbl.h @@ -8,6 +8,7 @@ #ifndef __QLA4x_GBL_H #define __QLA4x_GBL_H +int ql4xxx_lock_drvr_wait(struct scsi_qla_host *a); int qla4xxx_send_tgts(struct scsi_qla_host *ha, char *ip, uint16_t port); int qla4xxx_send_command_to_isp(struct scsi_qla_host *ha, struct srb * srb); int qla4xxx_initialize_adapter(struct scsi_qla_host * ha, diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c index bb3a1c1..9e81b81 100644 --- a/drivers/scsi/qla4xxx/ql4_init.c +++ b/drivers/scsi/qla4xxx/ql4_init.c @@ -978,7 +978,7 @@ static int qla4xxx_start_firmware_from_flash(struct scsi_qla_host *ha) return status; } -static int ql4xxx_lock_drvr_wait(struct scsi_qla_host *a) +int ql4xxx_lock_drvr_wait(struct scsi_qla_host *a) { #define QL4_LOCK_DRVR_WAIT 300 #define QL4_LOCK_DRVR_SLEEP 100 diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 5b8db61..bab434e 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -919,18 +919,11 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha, if (status == QLA_SUCCESS) { DEBUG2(printk("scsi%ld: %s - Performing soft reset..\n", ha->host_no, __func__)); - status = qla4xxx_soft_reset(ha); - } - /* FIXMEkaren: Do we want to keep interrupts enabled and process - AENs after soft reset */ - - /* If firmware (SOFT) reset failed, or if all outstanding - * commands have not returned, then do a HARD reset. - */ - if (status == QLA_ERROR) { - DEBUG2(printk("scsi%ld: %s - Performing hard reset..\n", - ha->host_no, __func__)); - status = qla4xxx_hard_reset(ha); + qla4xxx_flush_active_srbs(ha); + if (ql4xxx_lock_drvr_wait(ha) == QLA_SUCCESS) + status = qla4xxx_soft_reset(ha); + else + status = QLA_ERROR; } /* Flush any pending ddb changed AENs */ @@ -1016,13 +1009,9 @@ static void qla4xxx_do_dpc(void *data) struct scsi_qla_host *ha = (struct scsi_qla_host *) data; struct ddb_entry *ddb_entry, *dtemp; - DEBUG2(printk("scsi%ld: %s: DPC handler waking up.\n", - ha->host_no, __func__)); - - DEBUG2(printk("scsi%ld: %s: ha->flags = 0x%08lx\n", - ha->host_no, __func__, ha->flags)); - DEBUG2(printk("scsi%ld: %s: ha->dpc_flags = 0x%08lx\n", - ha->host_no, __func__, ha->dpc_flags)); + DEBUG2(printk("scsi%ld: %s: DPC handler waking up." + "flags = 0x%08lx, dpc_flags = 0x%08lx\n", + ha->host_no, __func__, ha->flags, ha->dpc_flags)); /* Initialization not yet finished. Don't do anything yet. */ if (!test_bit(AF_INIT_DONE, &ha->flags)) @@ -1032,16 +1021,8 @@ static void qla4xxx_do_dpc(void *data) test_bit(DPC_RESET_HA, &ha->dpc_flags) || test_bit(DPC_RESET_HA_INTR, &ha->dpc_flags) || test_bit(DPC_RESET_HA_DESTROY_DDB_LIST, &ha->dpc_flags)) { - if (test_bit(DPC_RESET_HA_DESTROY_DDB_LIST, &ha->dpc_flags)) - /* - * dg 09/23 Never initialize ddb list - * once we up and running - * qla4xxx_recover_adapter(ha, - * REBUILD_DDB_LIST); - */ - qla4xxx_recover_adapter(ha, PRESERVE_DDB_LIST); - - if (test_bit(DPC_RESET_HA, &ha->dpc_flags)) + if (test_bit(DPC_RESET_HA_DESTROY_DDB_LIST, &ha->dpc_flags) || + test_bit(DPC_RESET_HA, &ha->dpc_flags)) qla4xxx_recover_adapter(ha, PRESERVE_DDB_LIST); if (test_and_clear_bit(DPC_RESET_HA_INTR, &ha->dpc_flags)) { @@ -1122,7 +1103,8 @@ static void qla4xxx_free_adapter(struct scsi_qla_host *ha) destroy_workqueue(ha->dpc_thread); /* Issue Soft Reset to put firmware in unknown state */ - qla4xxx_soft_reset(ha); + if (ql4xxx_lock_drvr_wait(ha) == QLA_SUCCESS) + qla4xxx_soft_reset(ha); /* Remove timer thread, if present */ if (ha->timer_active) diff --git a/drivers/scsi/qla4xxx/ql4_version.h b/drivers/scsi/qla4xxx/ql4_version.h index b3fe7e6..d05048b 100644 --- a/drivers/scsi/qla4xxx/ql4_version.h +++ b/drivers/scsi/qla4xxx/ql4_version.h @@ -5,9 +5,4 @@ * See LICENSE.qla4xxx for copyright and licensing details. */ -#define QLA4XXX_DRIVER_VERSION "5.00.05b9-k" - -#define QL4_DRIVER_MAJOR_VER 5 -#define QL4_DRIVER_MINOR_VER 0 -#define QL4_DRIVER_PATCH_VER 5 -#define QL4_DRIVER_BETA_VER 9 +#define QLA4XXX_DRIVER_VERSION "5.00.06-k" -- cgit v0.10.2